invasive group a streptococcal disease...invasive group a streptococcal disease national...

Invasive Group A Streptococcal DiseaseNational Epidemiology and Genetic Analysis

Bart Vlaminckx

Cover: Dadara©Lay-out: Harm Wouter SnippePrinted by: Ponsen & Looijen BV, Wageningen

ISBN-10: 90-393-4319-5ISBN-13: 978-90-393-4319-7

Dit proefschrift werd mede mogelijk gemaakt met financiële steun van Bayer, Becton Dickinson, Minigrip Nederland, Pfizer, Wyeth en de J.E. Jurriaanse Stichting

Invasive Group A Streptococcal DiseaseNational Epidemiology and Genetic Analysis

Invasieve infecties met groep A streptokokkennationale epidemiologie en genetische analyse

(met een samenvatting in het Nederlands)

Proefschriftter verkrijging van de graad van doctor

aan de Universiteit Utrecht op gezag van de rector magnificus,

prof. dr. W.H. Gispen, ingevolge het besluit van het college voor promoties

in het openbaar te verdedigen op vrijdag 15 september des middags te 12.45 uur

door

Bart Johannes Maria Vlaminckx

geboren op 15 april 1975 te Venlo

Promotor: Prof. Dr. J. Verhoef

Co-promotores: Dr. W.T.M. Jansen Dr. J.F.P. Schellekens

Contents

Chapter 1 General introduction

Chapter 2 Epidemiological features of invasive and noninvasive group A streptococcal disease in the Netherlands, 1992–1996

Chapter 3 Epidemiological considerations following long-term surveillance of invasive group A streptococcal disease in The Netherlands, 1992–2003

Chapter 4 Long-term surveillance of invasive group A streptococcal disease in The Netherlands, 1994–2003

Chapter 5 Site-Specific Manifestations of Invasive Group A Streptococcal Disease: Type Distribution and Corresponding Patterns of Virulence Determinants

Chapter 6 Determination of the relationship between group A streptococcal genome content, M-type, and toxic shock syndrome by a mixed-genome microarray

Chapter 7 Dynamics in prophage content of M1 and M28 Streptococcus pyogenes isolates in the Netherlands from 1959 to 1996

Chapter 8 Summary and general discussion

Nederlandse samenvatting

Dankwoord

Curriculum vitae

1

31

53

67

83

107

131

149

169

179

185

Chapter 1General introduction

• 2

Premier Balkenende rea-liseerde zich pas achteraf dat zijn voetinfectie potentieel levensbedrei-gend was. Dat zei hij tijdens een openhartig interview in de actuali-teitenrubriek Netwerk... De dokters ontdekten twee bacterietypen in Balkenendes bloed: sta-fylococcen en strepto-coccen. “Als het alleen

het eerste type was ge-weest, waren antibio-tica waarschijnlijk wel afdoende geweest. Maar streptococcen zijn zeer gevaarlijk, omdat het vleesetende bacteriën zijn. De combinatie van de twee kan fataal zijn”, aldus Balkenende.

Source: NOS (http://www.nos.nl/nieuws/artike-len/2004/10/31/balkenendeoverziekbed.html)

Balkenende openhartig over ziekbed

3 •

Introduction

During the Dutch presidency of the European Union in the second half of 2004, prime minister Balkenende was hospitalized on September 14th. He was operated on several times and unable to chair the European Council for six weeks due to a severe soft tissue infection of his right foot caused by Streptococcus pyogenes. S. pyogenes, or group A streptococcus (GAS) (73) is as-sociated with a wide range of disease manifestations that have afflicted man-kind before understandable accounts were recorded. Before the pathogen was identified, scarlet fever and puerperal sepsis were the most reliably diagnosed manifestations of GAS disease. The history of these diseases provides us with epidemiological clues to understand GAS disease today.

History of group A streptococcal disease

History of scarlet feverAlthough probable case descriptions of scarlet fever have been attributed to Hippocrates (112), Sydenham coined the term febris scarlatina and its description as a separate clinical entity in 1676 (130). In 1778, Johnstone proposed a new classification system for scarlet fever (63). He distinguished three separate clinical entities: uncomplicated scarlet fever (scarlatina sim-plex); scarlet fever with pharyngitis (scarlatina anginosa) and scarlatina with gangrenous inflammation of the oropharynx (scarlatina maligna). Through-out the 18th and early 19th century, scarlet fever was a relatively benign child-hood disease, although some fatal epidemics of scarlatina maligna did occur (68). In the early 19th century the situation changed dramatically. A lethal epidemic in Dublin in 1831 was followed by an equally lethal epidemic in Great Britain with case-fatality rates of more than 15% (49). An Irish clini-cian described the changing picture of scarlet fever as follows: “We now began to hear of cases which proved unexpectedly fatal…still it was not until the year 1834 that the disease spread far and wide, assuming the form of a de-structive epidemic…the contagion seemed to act as a more deadly poison on the individuals of some families than upon those of others” (49). Geographic

• 4

clusters of severe cases were reported throughout the continent as well as in America with case-fatality rates that reached over 30% (141). During 1840-1883, scarlet fever became the most common deadly infectious childhood disease (135). While fatal cyclic epidemics of severe disease were raging in the large cities of Europe and America, epidemics were only rarely seen and of unpredictable severity in rural areas (87).

Scarlet fever continued to be a leading childhood killer until the mid-1880s, when severe cases became less frequent in Europe and America, with case-fatality rates dropping to ~1% or less in the next several decades. For instance, case-fatality rates from the London Fever Hospital decreased from 16% in 1863-1864 to 11% in 1883, 7.5% in 1887, 2.3% in 1890 and 0.9% in 1891 (141). This trend in decreased scarlet fever severity did not begin at the same time nor proceed at the same pace in all countries. In Russia for example, scarlet fever was associated with fatality in 22% of cases from 1892-1915 (68). Overall however, scarlet fever became a milder disease with epidemics gradually ceasing and fatal cases becoming increasingly rare.

History of puerperal sepsisPuerperal sepsis is a pelvic infection following delivery. Whereas postpartum infections have always been a risk (124), epidemics of puerperal fever devel-oped from the mid-17th century onwards when women delivered in urban maternity hospitals rather than in their own homes. The first report of epi-demic puerperal fever dates from 1646 on an epidemic in the Hôtel Dieu, Paris (124). The concept of puerperal fever as a consequence of institutional obstetric care, as well as its contagious nature, was articulated by Alexander Gordon in 1795 (47). He stated “the infection was as readily communicated as that of smallpox or measles, and operated more speedily than any other infection with which I am acquainted…It is a disagreeable declaration for me to mention that I myself was the means of carrying the infection to a great number of women”. Furthermore, he recognized an association between erysipelas and puerperal infections. The medical profession, with very few exceptions, rejected the concept of the contagiousness and microbial origin of the disease.

In 1843, Oliver Wendell Holmes, later dean of Harvard Medical School, read his paper “The contagiousness of Puerperal Fever” before the Boston

5 •

Society for Medical Improvement (60). At about the same time, Thomas Watson, professor of medicine at King’s college in London, voiced his belief that contagious spread of puerperal fever was possible: “…the hand which is relied upon for succor in the painful and perilous hours of childbirth may literally become the innocent cause of the mother’s destruction: innocent no longer, however, if after warning and knowledge of the risk, suitable means are not used to avert a catastrophe so shocking” (138).

Ignaz Semmelweis, a native of Hungary, was very keen to take all measures necessary to “avert a catastrophe so shocking”. He was appointed assistant lecturer at Vienna’s Allgemeines Krankenhaus in 1844, where he observed that the mortality rates in the two different divisions in his clinic differed dramatically (2). The first division, with mortality rates of 16%, was used to instruct medical students. In the second division, where midwives did deliveries, mortality rates were only 2%. Semmelweis also observed the negligible mortality among women who had given birth before reaching the hospital and did not undergo internal examinations during labor. He reasoned that the observed difference in mortality might be related to the medical students’ habit of going to the obstetric ward directly after postmortem examination without hand washing. Midwives did not attend autopsies, practiced on models and examined fewer women in labor. It was clear to Semmelweis that medical students conveyed something to the women in labor which prompted him to institute a routine of hand washing with chlorinated lime in 1847. Within a few months, mortality from puerperal fever dropped from 18% during April and May 1847 to 2.5% during June-November 1847 (137). Nonetheless, Semmelweis met firm opposition by the medical establishment and his appointment in Vienna was not renewed. He returned to Pest, Hungary, where he was appointed obstetrician in Saint Rochus Hospital in 1849. Here too, the institution of strict hand hygiene again led to a great reduction in maternal mortality (118).

Discovery of the pathogenThose who seriously considered the work of Gordon, Holmes and Semmelweis speculated on the nature of the “harmful things” responsible for the transmission of puerperal sepsis. In the 1880s, Louis Pasteur was shifting from his studies on fermentation to infectious diseases. He came to believe that,

• 6

like fermentation, transmissible diseases were due to specific microorganisms infecting the body. In 1874, chain-forming bacteria had been seen in erysipelas lesions by the German surgeon Theodor Billroth who named them Streptococcus (from the Greek streptos=chain and kokkus=berry or seed) (12). In 1879, Pasteur found microbes “in rounded granules arranged in the form of chains or string of beads” in the lochia and blood of women suffering from puerperal sepsis (104). Initially, streptococci were classified according to the source from which they were recovered: Streptococcus erysipelatosus, Streptococcus puerperalis, Streptococcus scarlatinae. In 1884, another German surgeon, Friedrich Julius Rosenbach applied the designation Streptococcus pyogenes to the causative agent of wound infections (113). In 1889, Ferdinand Widal, working at Pasteur Institute Paris, argued that these different diseases were caused by one organism (142). In the subsequent 50 years, the streptococcus became well known as the causative agent of a remarkable range of clinical illnesses. In 1903, Schotmüller established the value of the blood agar plate for growing streptococci and Brown described the use of hemolytic reactions to distinguish the α from β hemolytic streptococci (22,121).

In 1928, Rebecca Lancefield demonstrated that the β hemolytic strep-tococci could be subdivided in groups A-E according to the group specific polysaccharide or, in case of group D streptococci, lipoteichoic acid (LTA) (76). Most strains obtained from human infections belonged to group A. In 1935 she demonstrated that this also holds true for puerperal sepsis patients (79). Lancefield also showed that group A streptococci could be further sub-divided, based on the differences in antigenic nature of their M-protein (77). At about the same time, Griffith differentiated group A streptococci with a slide agglutination test based on the T- rather than M-antigens (51).

Strain typing

T- and M-proteinsThe T- and M-proteins are present at the surface of group A streptococci. In the laboratory, T-typing is performed by an agglutination test (51). Because certain T-types are associated with particular M-proteins, the testing for the M-type can be shortened by knowledge of the T-type (8).

7 •

Like the T-protein, the streptococcal M-protein extends from the cell mem-brane and is used for serotyping. The N- (amino) terminal portion of the protein constitutes the hypervariable region of the protein. Antigenic differences in this hypervariable region form the basis for the Lancefield serological classification scheme (45,78). Serotyping does not always yield unequivocal results, requires high-titered and well-standardized antisera with a limited life-span and does not allow the identification of new M-types (42). Therefore, a molecular biological technique has been developed using the emm gene, that encodes the M-protein, to replace M-protein based serotyping (9). A primer pair is used for the amplifi-cation of the N-terminal nucleotide residues encoding the hypervariable part of the M-protein. The amplicon can subsequently be identified using sequencing (9) or an emm type specific hybridization technique (69). There are currently 124 recognized M-genotypes (41). Functionally, the M-protein inhibits phagocytosis by interfering with opsonization (13). Absence of the emm gene allows rapid phagocytosis of the streptococcus (117). Immunity to the M-protein is protective against reinfection with that M-type and has led to the study of this protein for GAS vaccines (35).

Other typing techniquesOther molecular biological techniques have been applied to allow typing of streptococci: pulsed-field gel electrophoresis (PFGE), ribotyping, random amplified polymorphic DNA analysis (RAPD), multilocus enzyme electro-phoresis (MLEE) and multilocus sequence typing (MLST). PFGE involves digesting the bacterial chromosomal DNA with restriction enzymes that cleave infrequently. The resulting large restriction fragments are separated into a pattern of discrete bands by an alternating high-voltage field (85). Standardized criteria for analyzing the fragment patterns have been defined (133). PFGE patterns correlate with the M-type and are able to differenti-ate between strains of the same M-type (122). Multilocus sequence typing (MLST) is based on nucleotide sequencing the alleles of several housekeeping genes (39). These alleles are given numbers corresponding to specific mutations in the sequenced region. Analogous to a telephone number, the combination of all allele numbers defines the MLST type of a strain. Based on the nucleotide sequences of seven housekeeping genes, an MLST scheme for GAS has been de-veloped that found a strong association between M-type and MLST type (40).

• 8

Pathogenesis

Adherence and invasionBacterial attachment to skin or mucosa is the first step leading to colonization and disease. Small wounds may disrupt the dermal or mucosal barrier and thus establish a portal of entry to the bacterium. The strategies by which GAS adhere, and later invade, the human host are multiple and complex, involving several different adhesins and invasins. Expression of specific adherence fac-tors may differ from strain to strain depending on their genetic endowment and local environmental factors. Adhesion factors may confer a certain site- or tissue-specificity to the bacterium.

Courtney et al. hypothesized a two-step adhesion process (55). The first adhesion step is relatively weak and overcomes electrostatic repulsion. LTA has been proposed to serve as this “first-step” adhesin by hydrophobic interac-tions, bringing the organisms into close contact with host cells. Fibronectin was identified as the epithelial cell receptor binding LTA (101). Second-step adhesion may then involve M-protein (103) and a large number of other ad-hesins that promote high-affinity binding between GAS and host cells. The M-protein is important for attachment to keratinocytes in skin infections where membrane cofactor protein 46 (CD46) serves as the receptor for the M-protein (102). Other adhesion molecules that bind fibronectin, collagen and other extracellular matrix components are important in establishing ad-herence. Fibronectin binding proteins include PrtF1 (protein F1) (52), Sfb1 (streptococcal fibronectin binding protein 1) (131), and related proteins known as PrtF2 (61), PrtF15 (67), Sfb2 (75), Fbp54 (33), Fba (134), and Pfbp (111). Collagen binding proteins include: Cpa (collagen-binding pro-tein) (105), Cpa1 (90), SclA (streptococcal collagen-like surface protein A) (110) and SclB (140).

M-protein and Sfb1 have been implicated in intracellular entry of GAS (34,38). This process of cellular penetration may be the first step in the es-tablishment of tissue invasion (80) but may equally provide an intraepithelial sanctuary where the microorganism is sheltered from immune mechanisms and certain classes of antibiotics (96). Spreading of streptococci through tis-sue is facilitated by the elaboration of a large number of hydrolytic enzymes.

9 •

Extracellular products contributing to virulenceGAS produce several extracellular products that may facilitate the spreading of streptococci through tissue in characteristic streptococcal diseases such as cellulitis, myositis and necrotizing fasciitis. These include enzymes that par-ticipate in the degradation of DNA, so called streptodornases (streptococcal deoxyribonucleases): streptodornase B (SdaB, also known as mitogenic fac-tor (MF)) (86), streptodornase D (SdaD) (106), Spd1 (or MF2) (20), Spd3 (or MF3) (54) and Sda1 (4). Other extracellular streptococcal virulence fac-tors include a hyaluronidase, which degrades hyaluronic acid, an important component in the ground substance of connective tissue (15); streptokinase, which promotes the dissolution of fibrin clots by catalyzing the conversion of plasminogen to plasmin (82); streptococcal pyrogenic exotoxin B (SpeB), a potent protease (56,66); and C5a peptidase, which cleaves the human chemo-taxin C5a (30,43). Streptococcal inhibitor of complement (Sic) is a secreted protein that inhibits complement-mediated lysis of the bacterium (3). Two distinct haemolysins, Streptolysin O (SLO) and Streptolysin S (SLS), dam-age the membranes of polymorphonuclear leucocytes, platelets, and other eukaryotic cells (26,83).

Pyrogenic exotoxins and the streptococcal toxic shock-like syndromeThe streptococcal pyrogenic exotoxins (Spe’s) are a family of proteins with particular structural features that result in the shared ability to bypass the mechanisms of conventional antigen processing (84). Microbial antigens are processed into peptide fragments within antigen-presenting cells (APCs) such as monocytes, B-cells and dendritic cells. These fragments are presented to T-cells in the peptide-binding groove of the MHC class II molecule at the cell surface of the APC. T-cells only respond if they recognize the class II molecule and the specific peptide being presented (81). In contrast, a superantigen can overcome this peptide dependent activation of T-cells by direct binding to the beta chain (Vβ) of a characteristic set of T-cell receptors and to the MHC class II molecule expressed on APCs (Figure 1). Thereby, superantigens can activate up to 25% of an individual’s T-cell repertoire (28). This massive T-cell activation results in the excessive release of proinflammatory cytokines, such as tumour necrosis factor alpha (TNFα), interleukin (IL) 6, interferon gamma (IFNγ), and IL2 (27). The release of proinflammatory cytokines may

• 10

activate the complement, coagulation, and fibrinolytic cascades, resulting in hypotension and multiorgan failure characteristic of many of the clinical fea-tures of toxic shock-like syndrome (TSS) (89).

GAS posses a large number of mitogenic exotoxins, which may function as superantigens. These include: SpeA (139), SpeC (48), SpeF (99), SpeG (107), SpeH (107), SpeJ (107), SpeK (6), SpeL (108), SpeM (108), SSA (91), SMEZ (64), and SMEZ-2 (107). Nucleotide sequencing of the streptococcal pyrogenic exotoxin A gene (speA) revealed four naturally occurring alleles of speA. The speA1, speA2, and speA3 alleles encode toxins that differ in a single amino acid, while speA4 encodes a toxin which is 9% divergent from the other three and has 26 amino acid substitutions (98). The streptococcal su-perantigens are roughly 25 kDa, and they contain distinct T-cell receptor and MHC class II binding sites.

Host susceptibility to TSS may depend on the affinity of GAS exotoxins for different HLA class II haplotypes. Recent research by Kotb et al. suggested that individuals with a certain class II haplotype are more susceptible to de-velopment of TSS (74).

Figure 1 Interaction of antigens (Ag) (left) and superantigens (right) with the antigen pre-senting cell and the T-lymphocyte. Superantigens trigger massiveT-cell activation that results in the excessive release of proinflammatory cytokines (81).

11 •

GAS genome-wide analysis

Genome sequencingCurrently, complete genome sequences from seven S. pyogenes strains have been published: two M1 isolates (SF370, isolated from a patient with a wound infection (44) and MGAS5005, associated with invasive GAS disease (128)); two M3 (MGAS315 (10) and SSI-1 (95), both recovered from patients with TSS); one M6 (MGAS10394, cultured from a child with pharyngitis (7)); one M18 (MGAS8232, associated with acute rheumatic fever (123)) and one M28 (MGAS6180, associated with puerperal sepsis (50)). The GAS genomes range in size from 1.84 to 1.90 Mb and have a G+C content of 38.5-38.7%. They encode for 1697 (SF370) to 1894 (MGAS6180) proteins. Per bacte-rial chromosome, three (MGAS5005) to eight (MGAS10394) prophages (Φ) were identified. With transformation and conjugation appearing to play no significant role, these prophages have been identified as the major source of variation in the gene content between GAS strains (5). For instance, MGAS315 and SF370 share 85.3% of their coding sequences (CDS). Of the remaining 14.7%, 75.3% is phage encoded, whereas prophages constitute less than 13% of the overall genomes. All GAS prophages have a conserved genetic structure between the left (attL) and right attachment sites (attR): lysogeny-DNA replication-transcriptional regulation-DNA packaging-head-joining-tail-tail fiber-lysis modules (Figure 2) (23). The tail fiber protein hy-aluronidase is an enzyme that splits the hyaluronic acid-containing capsule surrounding the bacterial cell. This lytic enzyme allows the phage to reach the cell surface, where it injects its DNA into the bacterial cell. In addition to phage specific proteins, many GAS prophages encode virulence factors such as the pyrogenic exotoxins SpeA, SpeC, SpeH, SpeI, SpeK, SpeL, SpeM and SSA; streptodornases Spd (21), Sda (4) and Sdn (6) and Sla, a streptococcal

Figure 2 Organization and ORF map of the prophage 315.2 present in the genome of strain MGAS315 and encoding the superantigen SSA. Putative ORFs are indicated by arrows that show the direction of transcription. Groups of genes whose protein products are functionally related are color coded (23).

��

��

��

• 12

phospholipase (10). The vast majority of these virulence factors are encoded between the lysis cassette and the right phage attachment site (Figure 2). Even when they encode different virulence factors, some GAS prophages share a large degree of sequence homology. This suggests that phages are prone to recombination processes, which may generate chimeric genomes (18,37).

It has been postulated that the acquisition of multiple prophage-encoded virulence genes might influence the pathogenic potential of S. pyogenes (94). Beres et al. discovered that M3 strains isolated in the 1920s contain Φ315.5 (encoding the SpeA1 variant), whereas M3 strains causing disease in the 1940s have this prophage plus Φ315.2, encoding SSA (Figure 2) (11). In addition, Φ315.5 present in strains from the 1940s onwards, encodes SpeA3, an allelic variant that is more mitogenic (72). Furthermore, the majority of contempo-rary M3 strains contain another prophage: Φ315.4 (encoding SpeK and Sla) that was not present in serotype M3 strains until the mid-1980s (Figure 3) (10).

Figure 3 Early M3 S. pyogenes contained prophage Φ315.5 (encoding SpeA1). This strain then acquired prophage Φ315.2 (encoding SSA, figure 2), and a single nucleotide muta-tion resulted in the SpeA3 variant in the early 1940s. This ssa- and speA3-containing strain acquired prophage Φ315.4, encoding Sla, a phospholipase and SpeK. This new, virulent M3 clone gained predominance in the USA and Europe. Adapted from (10).

13 •

MicroarraysWith the availability of whole genome sequences and the advent of sophis-ticated bioinformatics tools, genome-wide analysis has become available us-ing hybridization techniques on microarrays. For microarray construction, several thousand gene probes that comprise the bacterial DNA are printed on glass slides using robotic arrayers. The probes are either PCR products or synthetic oligonucleotides. Fragmented target DNA to be hybridized to the array is tagged with fluorescent dyes. After hybridization to the array, pres-ence or absence of fluorescence in each microarray spot is used to sensitively detect the genomic composition of the tested strain. The aim of bacterial comparative genomics is to investigate what genetic features determine clini-cal or epidemiological differences between different strains.

Clinical manifestations

GAS infections can give rise to a wide range of clinical syndromes ranging from uncomplicated superficial infections to severe invasive infections.

Superficial infectionsGAS is the most common cause of bacterial tonsillopharyngitis (120). The dis-ease occurs predominantly in children aged 5 to 15 years. Physical findings are redness and oedema of the pharynx, enlarged hyperemic tonsils with a gray-ish-white exudate and fever. In the absence of suppurative complications, fever abates in 3-5 days (14). Incidentally, the infection can spread locally giving rise to otitis media or sinusitis. Antibiotic treatment is indicated to prevent deep suppurative complications (peritonsillar/ retropharyngeal abscess) and acute rheumatic fever (97,120). Streptococci can also cause discrete purulent lesions of the skin known as streptococcal impetigo or pyoderma (14).

Invasive infectionsErysipelas is an acute inflammation of the skin that affects the cutaneous lymphatic vessels. Often a predisposing condition is present such as impaired

• 14

lymphatic drainage or a surgical or traumatic wound (16). The erysipelas skin lesion has a raised border, sharply demarcated from normal skin. This feature allows it to be distinguished from GAS cellulitis: an acute spreading infection of the skin and subcutaneous tissue, often resulting from wounds or surgical incisions (16).

Necrotizing fasciitis is an infection of the deeper subcutaneous tissue and fascia characterized by rapidly spreading necrosis of skin and underlying tis-sues. The main portal of entry is the skin although an initial lesion or primary focus may be inapparent (16,24,114). Necrotizing fasciitis may be difficult to diagnose in its early stages: redness and bullae may be present but sometimes local erythema is the only presenting symptom. Fever, toxicity and severe pain, discrepant to the relatively benign aspect of the skin should prompt surgical inspection of the deep tissues (125). In case of necrotizing fasciitis, extensive tissue necrosis along the fascial planes will be revealed, possibly in combination with myonecrosis. Case-fatality of necrotizing fasciitis is about 30% but increases when TSS develops as a complication (125).

Puerperal sepsis follows delivery or abortion when GAS, colonizing the patient or transmitted from medical personnel, invade the endometrium. GAS pneumonia is frequently associated with preceding viral respiratory infections and chronic pulmonary disease. Empyema often complicates GAS pneumonia (30-40%) and it is accompanied by GAS bacteraemia in 10-15% of cases (17). GAS meningitis presents as other forms of bacterial pyogenic meningitis and is often preceded by GAS infection of the upper respiratory tract (29). Furthermore, GAS are the second cause of non-gonococcal arthritis after Staphylococcus aureus (36,46).

Streptococcal toxic shock-like syndromeIn severe invasive GAS infections, hypotension and multiorgan failure may develop rapidly resulting in the development of TSS. Case definitions for TSS have been developed (Table 1) (1). In the majority of patients, bacteraemia and a focus of infection have been observed. Mortality rates are high (30-80%) and almost half of all TSS survivors have limbs amputated or require major debridements (16,17,36,115,125). Persons of all ages may develop TSS and most are not immunosuppressed, although an association has been observed with advanced age, alcoholism and underlying conditions such as

15 •

diabetes mellitus, malignancies, and chickenpox (24,36,70,114,115,126). M1 and M3 have been particularly implicated in the pathogenesis of TSS (24,57,59,70,93,114,132). T-cell activation by streptococcal pyrogenic exotoxins with subsequent excessive release of proinflammatory cytokines and activa-tion of complement, coagulation, and fibrinolytic cascades is at the heart of TSS pathogenesis (88,89).

Table 1 Case definition for streptococcal toxic shock-like syndrome (TSS)

Isolation of S. pyogenes from a normally sterile site (e.g., blood, cerebrospinal fluid, pleural fluid)a

and

Hypotension (5th percentile of systolic blood pressure for children or less than 90 mm Hg. for adolescents)

and two or more of:

1. Renal impairment (creatinine greater than two times the upper limit for age)

2. Coagulopathy (platelets less than 100 000 ×106 U/L or evidence of disseminated intra-vascular coagulopathy)

3. Liver involvement (transaminases or bilirubin greater than two times upper limit of nor-mal)

4. Adult respiratory distress syndrome (pulmonary infiltrates and hypoxemia without heart failure or generalized edema)

5. Generalized erythematous rash that may desquamate

6. Soft tissue necrosis in the form of necrotizing fasciitis, myositis or gangrene

a If S. pyogenes is isolated from a nonsterile site (e.g., throat, sputum, vagina) but the patient has hypotension and two of the other criteria 1 to 6 above, it is considered a probable case if no other etiology for the illness is found.

• 16

EpidemiologyAs described earlier, the decline in morbidity and mortality due to invasive streptococcal infections began long before the introduction of antibiotics. Apart from the obvious effect of hygienic measures on the prevalence of puerperal sepsis, the reason for the decline in the severity of invasive GAS disease in general has been attributed to changes in host resistance, bacterial virulence and environmental factors, especially crowding in the urban setting. No particular change in either one of these factors has been identified as the sole explanation for the observed decrease in incidence and severity of invasive GAS disease (109). There is no indication to assume that streptococci in general became less prevalent: non-invasive GAS disease such as bacterial pharyngitis continued unabatedly throughout the 20th century (19,109). Nonetheless, there was optimism that serious GAS disease had almost been eradicated and notification for specific GAS diseases such as scarlet fever was abolished in many European countries (114). In the mid-1980s however, reports of a sudden increase of severe invasive GAS infections re-emerged worldwide (14,24,32,58,65,70,92,100,126). This resurgence resembled that of scarlet fever in the early 19th century. Initial reports came from hospital-based series and the first prospective population-based study came from Ontario, Canada, where the estimated incidence of invasive GAS disease was 1.5 per 100 000 person years (36). The overall mortality rate was 15% but much higher (81%) among those with TSS. This, and other population-based surveillance studies, have used microbiological and clinical criteria for the definition of invasive GAS disease (71,100,119,129). Other studies described isolation of GAS from normally sterile body sites as an indication of invasive GAS disease (31,62,116,127,136). Some of these studies relied on the voluntary cooperation of participating hospitals and laboratories whereas other surveillance systems were actively stimulated. In the Netherlands, a prospective population-based national surveillance program on invasive GAS disease was initiated in 1992. It is estimated that there are currently over 660 000 cases of invasive GAS disease worldwide each year, with over 160 000 case-fatalities (25).

17 •

Outline of the thesis

The aim of this thesis was to gain understanding in the epidemiology and genetics of invasive GAS infections in the Netherlands. What are the epide-miological and clinical characteristics of these infections? Are there specific genetic profiles associated with these GAS disease characteristics and why did a resurgence of invasive GAS disease occur? To address these issues we ob-tained population-based data on invasive GAS infections in The Netherlands and investigated the relationship between the genetic composition of the bac-terium and its clinical and epidemiological manifestations.

The first part of the thesis deals with the epidemiology of invasive GAS dis-ease in the Netherlands. In chapter 2, clinical and microbiological character-istics of invasive GAS infections in the Netherlands are described. Further-more, epidemiological differences between invasive and non-invasive GAS disease are evaluated. The national surveillance system that formed the basis for this study initially relied on both clinical and microbiological data. These data were obtained from regional public health laboratories (RPHLs) and individual hospitals on a voluntary basis. After two years, surveillance was formally organized with RPHLs. The effect of actively stimulating the surveil-lance system was evaluated in chapter 3. The dynamics in the incidence of invasive GAS disease and the relative contributions of different M-types over one decade in The Netherlands, as well as patient characteristics, are described in chapter 4.

The second part of the thesis aims at the identification of genetic profiles un-derlying invasive GAS disease. In chapter 5, isolates obtained from patients with clearly defined GAS infections affecting different tissues or organs are evaluated for the presence of superantigens and extracellular matrix binding factors. Thus, a possible relationship between genes encoding these virulence factors and the clinical manifestations of invasive GAS disease is scrutinized. Since GAS virulence is determined by a complex interplay between multi-ple genetic factors, we constructed a mixed-genome microarray, described in chapter 6. The aim was to detect the main genetic differences between GAS strains on a genome-wide level. GAS M1 and M3 types are overrepresented

• 18

in invasive GAS disease and TSS. Therefore, we assessed the main genetic differences between GAS of different M-types and sought after commonali-ties in gene profiles of M1 and M3 as compared to other M-types. Also, the existence of a common gene profile among TSS-associated GAS strains across all M-types was evaluated. Changes in bacterial virulence may have driven the resurgence of severe invasive GAS disease as witnessed since the mid-1980s. In chapter 7, we study temporal alterations in the genetic composition in a collection of GAS strains, spanning a period of more than four decades using the mixed-genome microarray.

19 •

References

1. 1993. Defining the group A streptococcal toxic shock syndrome. Rationale and consensus definition. The Working Group on Severe Streptococcal Infections. JAMA 269:390-391.

2. Adriaanse, A. H., M. Pel, and O. P. Bleker. 2000. Semmelweis: the combat against puer-peral fever. Eur. J Obstet. Gynecol. Reprod. Biol. 90:153-158.

3. Akesson, P., A. G. Sjoholm, and L. Bjorck. 1996. Protein SIC, a novel extracellular pro-tein of Streptococcus pyogenes interfering with complement function. J. Biol. Chem. 271:1081-1088.

4. Aziz, R. K., S. A. Ismail, H. W. Park, and M. Kotb. 2004. Post-proteomic identification of a novel phage-encoded streptodornase, Sda1, in invasive M1T1 Streptococcus pyogenes. Mol. Microbiol. 54:184-197.

5. Banks, D. J., S. B. Beres, and J. M. Musser. 2002. The fundamental contribution of phages to GAS evolution, genome diversification and strain emergence. Trends Microbiol. 10:515-521.

6. Banks, D. J., B. Lei, and J. M. Musser. 2003. Prophage induction and expression of prophage-encoded virulence factors in group A Streptococcus serotype M3 strain MGAS315. Infect. Immun. 71:7079-7086.

7. Banks, D. J., S. F. Porcella, K. D. Barbian, S. B. Beres, L. E. Philips, J. M. Voyich, F. R. DeLeo, J. M. Martin, G. A. Somerville, and J. M. Musser. 2004. Progress toward char-acterization of the group A Streptococcus metagenome: complete genome sequence of a macrolide-resistant serotype M6 strain. J. Infect. Dis. 190:727-738.

8. Beall, B., R. Facklam, T. Hoenes, and B. Schwartz. 1997. Survey of emm gene sequences and T-antigen types from systemic Streptococcus pyogenes infection isolates collected in San Francisco, California; Atlanta, Georgia; and Connecticut in 1994 and 1995. J. Clin. Microbiol. 35:1231-1235.

9. Beall, B., R. Facklam, and T. Thompson. 1996. Sequencing emm-specific PCR products for routine and accurate typing of group A streptococci. J. Clin. Microbiol. 34:953-958.

10. Beres, S. B., G. L. Sylva, K. D. Barbian, B. Lei, J. S. Hoff, N. D. Mammarella, M. Y. Liu, J. C. Smoot, S. F. Porcella, L. D. Parkins, D. S. Campbell, T. M. Smith, J. K. McCormick, D. Y. Leung, P. M. Schlievert, and J. M. Musser. 2002. Genome sequence of a serotype M3 strain of group A Streptococcus: phage- encoded toxins, the high-virulence pheno-type, and clone emergence. Proc. Natl. Acad. Sci. U. S. A. 99:10078-10083.

11. Beres, S. B., G. L. Sylva, D. E. Sturdevant, C. N. Granville, M. Liu, S. M. Ricklefs, A. R. Whitney, L. D. Parkins, N. P. Hoe, G. J. Adams, D. E. Low, F. R. DeLeo, A. McGeer, and J. M. Musser. 2004. Genome-wide molecular dissection of serotype M3 group A Strepto-coccus strains causing two epidemics of invasive infections. Proc. Natl. Acad. Sci. U. S. A. 101:11833-11838.

12. Billroth, T. 1874. Untersuchungen über die Vegetationsformen von Coccobacteria sep-tica und den Antheil, welchen sie and der Enstehung und Verbreitung der accidentellen Wundkrankheiten haben u.s.w., In . G. Reimer, Berlin.

• 20

13. Bisno, A. L. 1979. Alternate complement pathway activation by group A streptococci: role of M-protein. Infect. Immun. 26:1172-1176.

14. Bisno, A. L. 1991. Group A streptococcal infections and acute rheumatic fever. N. Engl. J. Med. 325:783-793.

15. Bisno, A. L., M. O. Brito, and C. M. Collins. 2003. Molecular basis of group A strepto-coccal virulence. Lancet Infect. Dis. 3:191-200.

16. Bisno, A. L. and D. L. Stevens. 1996. Streptococcal infections of skin and soft tissues. N. Engl J Med 334:240-245.

17. Bisno, A. L. and D. L. Stevens. 2000. Streptococcus pyogenes (including Streptocococcal Toxic Shock Syndrome and Necrotizing Fasciitis), p. 2101-2116. In G. L. Mandell, J. E. Bennett, and R. Dolin (ed.), Principles and Practice of Infectious Diseases. Churchill Livingstone, Philadelphia.

18. Botstein, D. 1980. A theory of modular evolution for bacteriophages. Ann. N. Y. Acad. Sci. 354:484-490.

19. Bronze, M. S. and J. B. Dale. 1996. The reemergence of serious group A streptococcal infections and acute rheumatic fever. Am. J Med Sci. 311:41-54.

20. Broudy, T. B., V. Pancholi, and V. A. Fischetti. 2002. The in vitro interaction of Strep-tococcus pyogenes with human pharyngeal cells induces a phage-encoded extracellular DNase. Infect. Immun. 70:2805-2811.

21. Broudy, T. B., V. Pancholi, and V. A. Fischetti. 2002. The in vitro interaction of Strep-tococcus pyogenes with human pharyngeal cells induces a phage-encoded extracellular DNase. Infect. Immun. 70:2805-2811.

22. Brown, J. H. 1919. The use of blood agar for the study of streptococci., In . The Rockefel-ler Institute for Medical Research, New York.

23. Canchaya, C., C. Proux, G. Fournous, A. Bruttin, and H. Brussow. 2003. Prophage ge-nomics. Microbiol. Mol. Biol. Rev. 67:238-76, table.

24. Carapetis, J., R. Robins-Browne, D. Martin, T. Shelby-James, and G. Hogg. 1995. In-creasing severity of invasive group A streptococcal disease in Australia: clinical and mo-lecular epidemiological features and identification of a new virulent M-nontypeable clone. Clin. Infect. Dis. 21:1220-1227.

25. Carapetis, J. R., A. C. Steer, E. K. Mulholland, and M. Weber. 2005. The global burden of group A streptococcal diseases. Lancet Infect. Dis. 5:685-694.

26. Carr, A., D. D. Sledjeski, A. Podbielski, M. D. Boyle, and B. Kreikemeyer. 2001. Similari-ties between complement-mediated and streptolysin S-mediated hemolysis. J Biol. Chem. 276:41790-41796.

27. Chatila, T. and R. S. Geha. 1993. Signal transduction by microbial superantigens via MHC class II molecules. Immunol. Rev. 131:43-59.

28. Choi, Y., J. A. Lafferty, J. R. Clements, J. K. Todd, E. W. Gelfand, J. Kappler, P. Marrack, and B. L. Kotzin. 1990. Selective expansion of T cells expressing V beta 2 in toxic shock syndrome. J Exp. Med 172:981-984.

29. Chow, J. W. and R. R. Muder. 1992. Group A streptococcal meningitis. Clin. Infect. Dis.

21 •

14:418-421.30. Cleary, P. P., U. Prahbu, J. B. Dale, D. E. Wexler, and J. Handley. 1992. Streptococcal C5a

peptidase is a highly specific endopeptidase. Infect. Immun. 60:5219-5223.31. Colman, G., A. Tanna, A. Efstratiou, and E. T. Gaworzewska. 1993. The serotypes of

Streptococcus pyogenes present in Britain during 1980- 1990 and their association with disease. J. Med. Microbiol. 39:165-178.

32. Cone, L. A., D. R. Woodard, P. M. Schlievert, and G. S. Tomory. 1987. Clinical and bacteriologic observations of a toxic shock-like syndrome due to Streptococcus pyogenes. N. Engl. J. Med. 317:146-149.

33. Courtney, H. S., J. B. Dale, and D. I. Hasty. 1996. Differential effects of the streptococcal fibronectin-binding protein, FBP54, on adhesion of group A streptococci to human buc-cal cells and HEp-2 tissue culture cells. Infect. Immun. 64:2415-2419.

34. Cue, D., P. E. Dombek, H. Lam, and P. P. Cleary. 1998. Streptococcus pyogenes sero-type M1 encodes multiple pathways for entry into human epithelial cells. Infect. Immun. 66:4593-4601.

35. Dale, J. B., E. Y. Chiang, and J. W. Lederer. 1993. Recombinant tetravalent group A streptococcal M protein vaccine. J Immunol. 151:2188-2194.

36. Davies, H. D., A. McGeer, B. Schwartz, K. Green, D. Cann, A. E. Simor, and D. E. Low. 1996. Invasive group A streptococcal infections in Ontario, Canada. Ontario Group A Streptococcal Study Group. N. Engl. J. Med. 335:547-554.

37. Desiere, F., W. M. McShan, S. D. van, J. J. Ferretti, and H. Brussow. 2001. Compara-tive genomics reveals close genetic relationships between phages from dairy bacteria and pathogenic Streptococci: evolutionary implications for prophage-host interactions. Virol-ogy 288:325-341.

38. Dombek, P. E., D. Cue, J. Sedgewick, H. Lam, S. Ruschkowski, B. B. Finlay, and P. P. Cleary. 1999. High-frequency intracellular invasion of epithelial cells by serotype M1 group A streptococci: M1 protein-mediated invasion and cytoskeletal rearrangements. Mol. Microbiol. 31:859-870.

39. Enright, M. C. and B. G. Spratt. 1999. Multilocus sequence typing. Trends Microbiol. 7:482-487.

40. Enright, M. C., B. G. Spratt, A. Kalia, J. H. Cross, and D. E. Bessen. 2001. Multilocus sequence typing of Streptococcus pyogenes and the relationships between emm type and clone. Infect. Immun. 69:2416-2427.

41. Facklam, R. F., D. R. Martin, M. Lovgren, D. R. Johnson, A. Efstratiou, T. A. Thompson, S. Gowan, P. Kriz, G. J. Tyrrell, E. Kaplan, and B. Beall. 2002. Extension of the Lancefield classification for group A streptococci by addition of 22 new M protein gene sequence types from clinical isolates: emm103 to emm124. Clin. Infect. Dis. 34:28-38.

42. Facklam, R. R. 1997. Screening for streptococcal pharyngitis: current technology. Infect. Med. 14:891-898.

43. Fernie-King, B. A., D. J. Seilly, C. Willers, R. Wurzner, A. Davies, and P. J. Lachmann. 2001. Streptococcal inhibitor of complement (SIC) inhibits the membrane attack com-

• 22

plex by preventing uptake of C567 onto cell membranes. Immunology 103:390-398.44. Ferretti, J. J., W. M. McShan, D. Ajdic, D. J. Savic, G. Savic, K. Lyon, C. Primeaux, S.

Sezate, A. N. Suvorov, S. Kenton, H. S. Lai, S. P. Lin, Y. Qian, H. G. Jia, F. Z. Najar, Q. Ren, H. Zhu, L. Song, J. White, X. Yuan, S. W. Clifton, B. A. Roe, and R. McLaughlin. 2001. Complete genome sequence of an M1 strain of Streptococcus pyogenes. Proc. Natl. Acad. Sci. U. S. A. 98:4658-4663.

45. Fischetti, V. A., K. F. Jones, S. K. Hollingshead, and J. R. Scott. 1988. Structure, function, and genetics of streptococcal M protein. Rev. Infect. Dis. 10 Suppl 2:S356-S359.

46. Goldenberg, D. L. and J. I. Reed. 1985. Bacterial arthritis. N. Engl J Med 312:764-771.47. Gordon, A. 1795. A treatise on the epidemic puerperal fever of Aberdeen., In . GG and J

Robinson, London.48. Goshorn, S. C., G. A. Bohach, and P. M. Schlievert. 1988. Cloning and characterization

of the gene, speC, for pyrogenic exotoxin type C from Streptococcus pyogenes. Mol. Gen. Genet. 212:66-70.

49. Graves, R. J. 1848. Scarlet fever, p. 421-461. In R.J.Graves and W.W.Gerhard (ed.), A system of clinical medicine. Barrington&Haswell, Philadelphia.

50. Green, N. M., S. Zhang, S. F. Porcella, M. J. Nagiec, K. D. Barbian, S. B. Beres, R. B. Lefebvre, and J. M. Musser. 2005. Genome Sequence of a Serotype M28 Strain of Group A Streptococcus: Potential New Insights into Puerperal Sepsis and Bacterial Disease Spe-cificity. J. Infect. Dis. 192:760-770.

51. Griffith, F. 1934. The serologic classification of streptococcus pyogenes. J. Hyg. Camb 34:542-584.

52. Hanski, E. and M. Caparon. 1992. Protein F, a fibronectin-binding protein, is an ad-hesin of the group A streptococcus Streptococcus pyogenes. Proc. Natl. Acad. Sci. U. S. A. 89:6172-6176.

53. Hasegawa, T., K. Torii, S. Hashikawa, Y. Iinuma, and M. Ohta. 2002. Cloning and char-acterization of the deoxyribonuclease sd alpha gene from Streptococcus pyogenes. Curr. Microbiol. 45:13-17.

54. Hasegawa, T., K. Torii, S. Hashikawa, Y. Iinuma, and M. Ohta. 2002. Cloning and char-acterization of two novel DNases from Streptococcus pyogenes. Arch. Microbiol. 177:451-456.

55. Hasty, D. L., I. Ofek, H. S. Courtney, and R. J. Doyle. 1992. Multiple adhesins of strep-tococci. Infect. Immun. 60:2147-2152.

56. Hauser, A. R. and P. M. Schlievert. 1990. Nucleotide sequence of the streptococcal py-rogenic exotoxin type B gene and relationship between the toxin and the streptococcal proteinase precursor. J Bacteriol. 172:4536-4542.

57. Hauser, A. R., D. L. Stevens, E. L. Kaplan, and P. M. Schlievert. 1991. Molecular analysis of pyrogenic exotoxins from Streptococcus pyogenes isolates associated with toxic shock-like syndrome. J. Clin. Microbiol. 29:1562-1567.

58. Holm, S. E. 1996. Invasive group A streptococcal infections. N. Engl J Med 335:590-591.

23 •

59. Holm, S. E., A. Norrby, A. M. Bergholm, and M. Norgren. 1992. Aspects of pathogenesis of serious group A streptococcal infections in Sweden, 1988-1989. J. Infect. Dis. 166:31-37.

60. Holmes, O. W. 1843. The contagiousness of puerperal fever. New England Quarterly Jour-nal of Medicine and Surgery 503-530.

61. Jaffe, J., S. Natanson-Yaron, M. G. Caparon, and E. Hanski. 1996. Protein F2, a novel fi-bronectin-binding protein from Streptococcus pyogenes, possesses two binding domains. Mol. Microbiol. 21:373-384.

62. Johnson, D. R., D. L. Stevens, and E. L. Kaplan. 1992. Epidemiologic analysis of group A streptococcal serotypes associated with severe systemic infections, rheumatic fever, or uncomplicated pharyngitis. J. Infect. Dis. 166:374-382.

63. Johnstone, J. 1778. Remarks on the angina and scarlet fever of 1778. Memoirs of the Medical Society of London 3:335-367.

64. Kamezawa, Y., T. Nakahara, S. Nakano, Y. Abe, J. Nozaki-Renard, and T. Isono. 1997. Streptococcal mitogenic exotoxin Z, a novel acidic superantigenic toxin produced by a T1 strain of Streptococcus pyogenes. Infect. Immun. 65:3828-3833.

65. Kaplan, E. L. 1991. The resurgence of group A streptococcal infections and their sequelae. Eur. J Clin. Microbiol. Infect. Dis. 10:55-57.

66. Kapur, V., M. W. Majesky, L. L. Li, R. A. Black, and J. M. Musser. 1993. Cleavage of interleukin 1 beta (IL-1 beta) precursor to produce active IL-1 beta by a conserved ex-tracellular cysteine protease from Streptococcus pyogenes. Proc. Natl. Acad. Sci. U. S. A. 90:7676-7680.

67. Katerov, V., A. Andreev, C. Schalen, and A. A. Totolian. 1998. Protein F, a fibronectin-binding protein of Streptococcus pyogenes, also binds human fibrinogen: isolation of the protein and mapping of the binding region. Microbiology 144 ( Pt 1):119-126.

68. Katz, A. R. and D. M. Morens. 1992. Severe streptococcal infections in historical perspec-tive. Clin. Infect. Dis. 14:298-307.

69. Kaufhold, A., A. Podbielski, G. Baumgarten, M. Blokpoel, J. Top, and L. Schouls. 1994. Rapid typing of group A streptococci by the use of DNA amplification and non-radioac-tive allele-specific oligonucleotide probes. FEMS Microbiol. Lett. 119:19-25.

70. Kaul, R., A. McGeer, D. E. Low, K. Green, and B. Schwartz. 1997. Population-based surveillance for group A streptococcal necrotizing fasciitis: Clinical features, prognostic indicators, and microbiologic analysis of seventy-seven cases. Ontario Group A Strepto-coccal Study. Am. J. Med. 103:18-24.

71. Kiska, D. L., B. Thiede, J. Caracciolo, M. Jordan, D. Johnson, E. L. Kaplan, R. P. Grun-inger, J. A. Lohr, P. H. Gilligan, and F. W. Denny, Jr. 1997. Invasive group A streptococcal infections in North Carolina: epidemiology, clinical features, and genetic and serotype analysis of causative organisms. J. Infect. Dis. 176:992-1000.

72. Kline, J. B. and C. M. Collins. 1996. Analysis of the superantigenic activity of mutant and allelic forms of streptococcal pyrogenic exotoxin A. Infect. Immun. 64:861-869.

73. Koneman, E. W., S. D. Allen, W. M. Janda, P. C. Schreckenberger, and W. C. Winn.

• 24

1997. The Gram-positive cocci: Part II: streptococci, enterococci, and the “streptococ-cus-like” bacteria, p. 539-576. In Color Atlas and Textbook of Diagnostic Microbiology. Lippincott Williams & Wilkins, Philadelphia.

74. Kotb, M., A. Norrby-Teglund, A. McGeer, H. El Sherbini, M. T. Dorak, A. Khurshid, K. Green, J. Peeples, J. Wade, G. Thomson, B. Schwartz, and D. E. Low. 2002. An immuno-genetic and molecular basis for differences in outcomes of invasive group A streptococcal infections. Nat. Med. 8:1398-1404.

75. Kreikemeyer, B., S. R. Talay, and G. S. Chhatwal. 1995. Characterization of a novel fi-bronectin-binding surface protein in group A streptococci. Mol. Microbiol. 17:137-145.

76. Lancefield, R. C. 1933. A serological differentiation of human and other groups of hemo-lytic streptococci. Journal of Experimental Medicine 57:571-595.

77. Lancefield, R. C. 1928. The antigenic complex of Streptococcus haemolyticus. I. Demon-stration of a type-specific substance in extracts of Streptococcus haemolyticus. J Exp. Med 47:91-103.

78. Lancefield, R. C. 1962. Current knowledge of type-specific M antigens of group A strep-tococci. J. Immunol. 89:307-313.

79. Lancefield, R. C. and R. Hare. 1935. The serological differentiation of pathogenic and non-pathogenic streptococci from parturient women. Journal of Experimental Medicine 61:335-349.

80. LaPenta, D., C. Rubens, E. Chi, and P. P. Cleary. 1994. Group A streptococci efficiently in-vade human respiratory epithelial cells. Proc. Natl. Acad. Sci. U. S. A. 91:12115-12119.

81. Llewelyn, M. and J. Cohen. 2002. Superantigens: microbial agents that corrupt immu-nity. Lancet Infect. Dis. 2:156-162.

82. Lottenberg, R., D. Minning-Wenz, and M. D. Boyle. 1994. Capturing host plasmin(ogen): a common mechanism for invasive pathogens? Trends Microbiol. 2:20-24.

83. Madden, J. C., N. Ruiz, and M. Caparon. 2001. Cytolysin-mediated translocation (CMT): a functional equivalent of type III secretion in gram-positive bacteria. Cell 104:143-152.

84. Marrack, P. and J. Kappler. 1990. The staphylococcal enterotoxins and their relatives. Sci-ence 248:705-711.

85. Maslow, J. N., A. M. Slutsky, and R. D. Arbeit. 1993. Application of pulsed field gel electrophoresis to molecular epidemiology, p. 563-572. In D. H. Persing, T. F. Smith, F. C. Tenover, and T. J. White (ed.), Diagnostic molecular microbiology: principles and ap-plications. American Society for Microbiology, Washington, D.C.

86. Matsumoto, M., N. Ishikawa, M. Saito, K. Shibayama, T. Horii, K. Sato, and M. Ohta. 1999. Streptococcal pyrogenic exotoxin F (SpeF) causes permeabilization of lung blood vessels. Infect. Immun. 67:4307-4311.

87. McCollom, J. H. 1907. Modern medicine. Its theory and practice., p. 334-372. In W.Osler and T.McCrae (ed.), Infectious diseases. Lea Brothers, Philadelphia.

88. McCormick, J. K., J. M. Yarwood, and P. M. Schlievert. 2001. Toxic shock syndrome and bacterial superantigens: an update. Annu. Rev. Microbiol. 55:77-104.

89. Miethke, T., C. Wahl, K. Heeg, B. Echtenacher, P. H. Krammer, and H. Wagner. 1992.

25 •

T cell-mediated lethal shock triggered in mice by the superantigen staphylococcal entero-toxin B: critical role of tumor necrosis factor. J Exp. Med 175:91-98.

90. Miyoshi-Akiyama, T., N. Wakisaka, J. Zhao, and T. Uchiyama. 2003. Characterizaition of collagen binding protein (Cpa) of M1-type group A streptococcus. unpublished.

91. Mollick, J. A., G. G. Miller, J. M. Musser, R. G. Cook, D. Grossman, and R. R. Rich. 1993. A novel superantigen isolated from pathogenic strains of Streptococcus pyogenes with aminoterminal homology to staphylococcal enterotoxins B and C. J. Clin. Invest 92:710-719.

92. Musher, D. M., R. J. Hamill, C. E. Wright, J. E. Clarridge, and C. M. Ashton. 1996. Trends in bacteremic infection due to Streptococcus pyogenes (group A streptococcus), 1986-1995. Emerg. Infect. Dis. 2:54-56.

93. Musser, J. M., A. R. Hauser, M. H. Kim, P. M. Schlievert, K. Nelson, and R. K. Selander. 1991. Streptococcus pyogenes causing toxic-shock-like syndrome and other invasive dis-eases: clonal diversity and pyrogenic exotoxin expression. Proc. Natl. Acad. Sci. U. S. A. 88:2668-2672.

94. Musser, J. M., V. Kapur, S. Kanjilal, U. Shah, D. M. Musher, N. L. Barg, K. H. Johnston, P. M. Schlievert, J. Henrichsen, D. Gerlach, and . 1993. Geographic and temporal distri-bution and molecular characterization of two highly pathogenic clones of Streptococcus pyogenes expressing allelic variants of pyrogenic exotoxin A (Scarlet fever toxin). J Infect. Dis. 167:337-346.

95. Nakagawa, I., K. Kurokawa, A. Yamashita, M. Nakata, Y. Tomiyasu, N. Okahashi, S. Kawabata, K. Yamazaki, T. Shiba, T. Yasunaga, H. Hayashi, M. Hattori, and S. Hamada. 2003. Genome sequence of an M3 strain of Streptococcus pyogenes reveals a large-scale genomic rearrangement in invasive strains and new insights into phage evolution. Genome Res. 13:1042-1055.

96. Neeman, R., N. Keller, A. Barzilai, Z. Korenman, and S. Sela. 1998. Prevalence of inter-nalisation-associated gene, prtF1, among persisting group-A streptococcus strains isolated from asymptomatic carriers. Lancet 352:1974-1977.

97. Nelson, J. D. 1984. The effect of penicillin therapy on the symptoms and signs of strepto-coccal pharyngitis. Pediatr. Infect. Dis. 3:10-13.

98. Nelson, K., P. M. Schlievert, R. K. Selander, and J. M. Musser. 1991. Characterization and clonal distribution of four alleles of the speA gene encoding pyrogenic exotoxin A (scarlet fever toxin) in Streptococcus pyogenes. J. Exp. Med. 174:1271-1274.

99. Norrby-Teglund, A., D. Newton, M. Kotb, S. E. Holm, and M. Norgren. 1994. Super-antigenic properties of the group A streptococcal exotoxin SpeF (MF). Infect. Immun. 62:5227-5233.

100. O’Brien, K. L., B. Beall, N. L. Barrett, P. R. Cieslak, A. Reingold, M. M. Farley, R. Danila, E. R. Zell, R. Facklam, B. Schwartz, and A. Schuchat. 2002. Epidemiology of invasive group a streptococcus disease in the United States, 1995-1999. Clin. Infect. Dis. 35:268-276.

101. Ofek, I., W. A. Simpson, and E. H. Beachey. 1982. Formation of molecular complexes between a structurally defined M protein and acylated or deacylated lipoteichoic acid of

• 26

Streptococcus pyogenes. J Bacteriol. 149:426-433.102. Okada, N., M. K. Liszewski, J. P. Atkinson, and M. Caparon. 1995. Membrane cofactor

protein (CD46) is a keratinocyte receptor for the M protein of the group A streptococcus. Proc. Natl. Acad. Sci. U. S. A. 92:2489-2493.

103. Okada, N., A. P. Pentland, P. Falk, and M. G. Caparon. 1994. M protein and protein F act as important determinants of cell-specific tropism of Streptococcus pyogenes in skin tissue. J. Clin. Invest 94:965-977.

104. Pasteur, L. 1879. Septicémie puerpérale. Bulletin de l’academie de medecine (Paris) 2me série:8256-8260.

105. Podbielski, A., M. Woischnik, B. A. Leonard, and K. H. Schmidt. 1999. Characterization of nra, a global negative regulator gene in group A streptococci. Mol. Microbiol. 31:1051-1064.

106. Podbielski, A., I. Zarges, A. Flosdorff, and J. Weber-Heynemann. 1996. Molecular char-acterization of a major serotype M49 group A streptococcal DNase gene (sdaD). Infect. Immun. 64:5349-5356.

107. Proft, T., S. L. Moffatt, C. J. Berkahn, and J. D. Fraser. 1999. Identification and charac-terization of novel superantigens from Streptococcus pyogenes. J. Exp. Med. 189:89-102.

108. Proft, T., P. D. Webb, V. Handley, and J. D. Fraser. 2003. Two novel superantigens found in both group A and group C Streptococcus. Infect. Immun. 71:1361-1369.

109. Quinn, R. W. 1982. Epidemiology of group A streptococcal infections--their changing frequency and severity. Yale J Biol. Med 55:265-270.

110. Rasmussen, M., A. Eden, and L. Bjorck. 2000. SclA, a novel collagen-like surface protein of Streptococcus pyogenes. Infect. Immun. 68:6370-6377.

111. Rocha, C. L. and V. A. Fischetti. 1999. Identification and characterization of a novel fi-bronectin-binding protein on the surface of group A streptococci. Infect. Immun. 67:2720-2728.

112. Rolleston, J. D. 1937. Scarlet fever, p. 47-75. In J.D Rolleston (ed.), The history of acute exanthema. William Heinemann, London.

113. Rosenbach, A. J. F. 1884. Mikroorganismen bei Wundinfektionskrankheiten des Men-schen, In, Wiesbaden.

114. Schellekens, J. F., L. M. Schouls, W. van Pelt, M. Esveld, and W. J. van Leeuwen. 1998. Group A streptococci: a change in virulence? Neth. J. Med. 52:209-217.

115. Schellekens, J. F., L. M. Schouls, A. van Silfhout, K. Elzenaar, H. Brunings, H. ten Broek, J. Top, and W. J. van Leeuwen. 1995. The resurgence of Group A streptococcal disease. Nederlands Tijdschrift voor Medische Microbiologie 4:78-83.

116. Schwartz, B., R. R. Facklam, and R. F. Breiman. 1990. Changing epidemiology of group A streptococcal infection in the USA. Lancet 336:1167-1171.

117. Scott, J. R., W. M. Pulliam, S. K. Hollingshead, and V. A. Fischetti. 1985. Relationship of M protein genes in group A streptococci. Proc. Natl. Acad. Sci. U. S. A. 82:1822-1826.

118. Semmelweis, I. P. 1861. Die Aetiologie, der Begriff und die Prophylaxis des Kindbettfiebers,

27 •

In, C.A. Hartleben, Leipzig.119. Sharkawy, A., D. E. Low, R. Saginur, D. Gregson, B. Schwartz, P. Jessamine, K. Green,

and A. McGeer. 2002. Severe group a streptococcal soft-tissue infections in Ontario: 1992-1996. Clin. Infect. Dis. 34:454-460.

120. Shoemaker, M., R. M. Lampe, and M. R. Weir. 1986. Peritonsillitis: abscess or cellulitis? Pediatr. Infect. Dis. 5:435-439.

121. Shottmüller, H. 1903. Die Artunterscheidung der für den menschen Pathogenen Strep-tokokken durch Blutagar. Munch. Med. Wochenschr. 50:849-853.

122. Single, L. A. and D. R. Martin. 1992. Clonal differences within M-types of the group A Streptococcus revealed by pulsed field gel electrophoresis. FEMS Microbiol. Lett. 70:85-89.

123. Smoot, L. M., J. C. Smoot, M. R. Graham, G. A. Somerville, D. E. Sturdevant, C. A. Migliaccio, G. L. Sylva, and J. M. Musser. 2001. Global differential gene expression in response to growth temperature alteration in group A Streptococcus. Proc. Natl. Acad. Sci. U. S. A. 98:10416-10421.

124. Speert, H. 1973. Iconographia gyniatrica. A pictorial history of gynaecology and obstet-rics. F.A. Davis Company, Philadelphia.

125. Stevens, D. L. 1992. Invasive group A streptococcus infections. Clin. Infect. Dis. 14:2-11.

126. Stevens, D. L., M. H. Tanner, J. Winship, R. Swarts, K. M. Ries, P. M. Schlievert, and E. Kaplan. 1989. Severe group A streptococcal infections associated with a toxic shock-like syndrome and scarlet fever toxin A. N. Engl. J. Med. 321:1-7.

127. Stromberg, A., V. Romanus, and L. G. Burman. 1991. Outbreak of group A streptococcal bacteremia in Sweden: an epidemiologic and clinical study. J. Infect. Dis. 164:595-598.

128. Sumby, P., S. F. Porcella, A. G. Madrigal, K. D. Barbian, K. Virtaneva, S. M. Ricklefs, D. E. Sturdevant, M. R. Graham, J. Vuopio-Varkila, N. P. Hoe, and J. M. Musser. 2005. Evolutionary origin and emergence of a highly successful clone of serotype M1 group a Streptococcus involved multiple horizontal gene transfer events. J Infect. Dis. 192:771-782.

129. Svensson, N., S. Oberg, B. Henriques, S. Holm, G. Kallenius, V. Romanus, and J. Giesecke. 2000. Invasive group A streptococcal infections in Sweden in 1994 and 1995: epidemiology and clinical spectrum. Scand. J Infect. Dis. 32:609-614.

130. Sydenham, T. 1676. Febris scarlatina, p. 387-389. In T.Sydenham (ed.), Observationes medicae circa morborum acutorum historiam et curationem. G. Kettilby, London.

131. Talay, S. R., P. Valentin-Weigand, K. N. Timmis, and G. S. Chhatwal. 1994. Domain structure and conserved epitopes of Sfb protein, the fibronectin- binding adhesin of Strep-tococcus pyogenes. Mol. Microbiol. 13:531-539.

132. Talkington, D. F., B. Schwartz, C. M. Black, J. K. Todd, J. Elliott, R. F. Breiman, and R. R. Facklam. 1993. Association of phenotypic and genotypic characteristics of invasive Streptococcus pyogenes isolates with clinical components of streptococcal toxic shock syn-drome. Infect. Immun. 61:3369-3374.

• 28

133. Tenover, F. C., R. D. Arbeit, R. V. Goering, P. A. Mickelsen, B. E. Murray, D. H. Pers-ing, and B. Swaminathan. 1995. Interpreting chromosomal DNA restriction patterns produced by pulsed- field gel electrophoresis: criteria for bacterial strain typing. J. Clin. Microbiol. 33:2233-2239.

134. Terao, Y., S. Kawabata, E. Kunitomo, J. Murakami, I. Nakagawa, and S. Hamada. 2001. Fba, a novel fibronectin-binding protein from Streptococcus pyogenes, promotes bacterial entry into epithelial cells, and the fba gene is positively transcribed under the Mga regula-tor. Mol. Microbiol. 42:75-86.

135. Tweedie, A. 1847. Scarlatina, p. 67-82. In J.Forbes, A.Tweedie, and J.Conolly (ed.), The cyclopedia of practical medicine, vol. 4. Lea and Blanchard, Philadelphia.

136. Tyrrell, G. J., M. Lovgren, B. Forwick, N. P. Hoe, J. M. Musser, and J. A. Talbot. 2002. M types of group a streptococcal isolates submitted to the National Centre for Streptococcus (Canada) from 1993 to 1999. J. Clin. Microbiol. 40:4466-4471.

137. von Hebra, F. 1847. Höchst wichtige Erfahrungen über die Aetiologie der in Gebäran-stalten epidemischen Puerperalfieber. Zeitschrift der kaiserlich-königlichen Gesellschaft der Ärzte zu Wien 4:242-244.

138. Watson, T. 1843. Lectures on the principles and practice of physic delivered at King’s Col-lege, London., In . J.W. Parker, London.

139. Weeks, C. R. and J. J. Ferretti. 1986. Nucleotide sequence of the type A streptococcal exo-toxin (erythrogenic toxin) gene from Streptococcus pyogenes bacteriophage T12. Infect. Immun. 52:144-150.

140. Whatmore, A. M. 2001. Streptococcus pyogenes sclB encodes a putative hypervariable surface protein with a collagen-like repetitive structure. Microbiology 147:419-429.

141. Whitelegge, B. A. 1893. The Milroy lectures on changes of type of epidemic disease. Lec-ture II. Diseases of more stable type. BMJ 1:449-451.

142. Wilson, L. G. 1987. The early recognition of streptococci as causes of disease. Med Hist 31:403-414.

29 •

• 30

Chapter 2

Epidemiological features of invasive and noninvasive group A streptococcal disease in the Netherlands, 1992–1996

B. Vlaminckx • W. van Pelt • L. Schouls • A. van Silfhout • C. Elzenaar • E. Mascini • J. Verhoef • J. Schellekens

European Journal of Clinical Microbiology & Infectious Diseases (2004) 23: 434–444

• 32

33 •

Abstract

A prospective, nationwide, laboratory-based surveillance of invasive group A streptococcal (GAS) infections was conducted in the Netherlands from 1992 through 1996. Clinical and demographic data were obtained and all isolates were T/M typed. All noninvasive GAS isolates were registered from 1994 through 1996. A total of 880 patients with invasive streptococcal disease were identified. The annual incidence was found to be 2.2 per 100 000. Predominant M-types were M1 (21%), M3 (11%), M6 (5%), M12 (5%), and M28 (8%). Particular age and M-type distributions were observed in different clinical enti-ties. The case-fatality rate was 18% overall, but it reached 59% among cases of toxic shock-like syndrome. Older age, necrotizing fasciitis, sepsis without focus, pneumonia, infection with type M1 or M3 strains, and underlying cardiopul-monary disease were associated with fatality. A total of 10 105 patients with noninvasive GAS disease were registered. These patients differed significantly from patients with invasive disease with regard to age distribution and primary foci of infection.

• 34

Introduction

Infections with group A streptococci (GAS) range from mild and superficial to very severe and lethal invasive disease. Postinfectious sequelae, such as rheu-matic fever or glomerulonephritis, may follow superficial infections (1). All manifestations of invasive GAS disease can be complicated by the development of toxic shock-like syndrome (TSS) (2, 3).

In the northern part of the world’s Western Hemisphere, a high endemic level of GAS infection and subsequent sequelae in the 19th century was fol-lowed by a decline, thought to be attributable to improvements in nutrition and hygienic standards. This declining trend continued unabated with the ad-vent of antibiotics (1).

A resurgence of fulminant invasive GAS infections was noted in the mid-1980s. Although frequently associated with comorbid conditions, cases oc-curred in previously healthy individuals as well (4–7). Often, unimpressive superficial lesions of skin or mucosa preceded invasive GAS disease. Historical explanations for a high level of endemicity of GAS infections involved poor socioeconomic conditions and the absence of antibiotics (8). Since these fac-tors cannot explain the reemergence of severe invasive streptococcal infections in the developed world, recent studies have focussed on bacterial virulence and host factors. A large number of studies have reported enhanced invasiveness and toxicity in some M-types of GAS, particularly M1 and M3 (5, 7, 9–15). Using appropriate noninvasive controls, a recent study questioned the postu-lated enhanced invasiveness of these M-types, suggesting that their association with invasive disease might reflect their predominance among noninvasive GAS strains (16).

The aim of this study was to obtain population-based data on the microbio-logical and clinical characteristics of invasive GAS infections in the Netherlands and to compare epidemiological features of invasive and noninvasive GAS dis-ease.

35 •

Materials and methods

Surveillance of invasive group A streptococcal disease A nationwide laboratory-based surveillance system for invasive GAS infections was conducted in the National Institute of Public Health (Dutch acronym: RIVM) from March 1992 to December 1996 (17–19) in the Netherlands, where the average population for the period 1992–1996 was 15 412 913. All Dutch clinical microbiology laboratories participated in this surveillance sys-tem on a voluntary basis. Invasive streptococcal disease was defined as the isola-tion of Streptococcus pyogenes from a normally sterile body site (definite invasive GAS disease) or from a normally nonsterile body site in conjunction with a clinical picture consistent with invasive GAS disease (probable invasive GAS disease) (20). In May 1994, the surveillance system was adapted for part of the laboratories, i.e. for the Regional Public Health Laboratories (RPHLs), which together have a national coverage of 50%. The submission to RIVM of all inva-sive GAS isolates by RPHLs was organized formally, including reimbursement of costs. This was done in anticipation of an ongoing laboratory-based sentinel surveillance of invasive GAS infections in the Netherlands (chapter 4).

Upon receipt of an isolate, a questionnaire was sent to the treating physician and the medical microbiologist to obtain extensive clinical and demographic data. Data obtained involved the following: patient age and sex, the presence of comorbid conditions, clinical symptoms and signs upon presentation, the primary site of infection, clinical manifestations, complications, and outcome. Soft tissue infections included necrotizing fasciitis and/or myositis, traumatic or surgical wound infection, cellulitis, lymphadenitis/ lymphangitis, bursitis (without positive synovial cultures), and deep abscesses. Patients were catego-rized as having GAS arthritis when clinical symptoms and signs of arthritis were found in conjunction with positive synovial or blood cultures. GAS meningitis was defined clinically in conjunction with positive CSF or blood cultures. Di-agnosis of GAS pneumonia relied on clinical interpretation and/or radiological signs of pulmonary infiltrates in combination with GAS isolated from spu-tum, bronchial lavage, or pulmonary aspirates. Puerperal sepsis was defined as postpartum sepsis with GAS isolated from both blood and the cervix uteri or vagina. Patients who had a systemic inflammatory response syndrome and posi-tive blood cultures but for whom no focus could be established were categorized

• 36

as having sepsis without a focus. Patients with more than one clinical manifesta-tion of invasive GAS disease were categorized under the most severe syndrome, as determined by its case-fatality ratio. Cases were considered nosocomial if symptoms arose more than two days after hospital admission. The criteria used to define TSS were those given by the Working Group on Severe Streptococcal Infections (20).

Streptococci enter the body through the skin or mucosal membranes. For in-vasive isolates, the primary focus of infection was established by culture of GAS from a particular body site together with a compatible clinical picture. Portals of entry were categorized as the skin, the respiratory tract, the perineo-genital area, or unknown, based on clinical and microbiological data. For noninvasive GAS isolates, the focus of infection was based solely on the source of isolation and was also categorized as the skin, the respiratory tract, the perineo-genital area, or unknown.

Surveillance of noninvasive group A streptococcal disease As part of the above-mentioned contract surveillance system, RPHLs registered the foci of all noninvasive GAS infections as well as the age and sex of patients involved from May 1994 to December 1996. The focus of infection was de-duced from the source of isolation.

Descriptive epidemiology Descriptive analyses were based on all clinically evaluated cases of invasive in-fection from March 1992 to December 1996. For incidence calculations, data on clinically evaluated (correcting for nonresponse) invasive isolates obtained in the period January 1993– December 1996 were used. For incidence calcula-tions of noninvasive GAS infections, data from RPHLs (correcting for their na-tional coverage) obtained in the period May 1994–December 1996 were used.

Laboratory methodsIdentification of isolates was confirmed using bacitracin susceptibility testing and latex agglutination (Streptex; Wellcome, Dartford, UK). All isolates sub-mitted were subjected to T-serotyping (21) and M-genotyping (22).

37 •

StatisticsDifferences in group proportions were assessed by the chi-square test or Fish-er’s exact test and differences in median values by the Wilcoxon rank-sum test. Graphical illustration of differences in age distribution for different clinical manifestations and M-types, compared to the total number of invasive iso-lates, was done using bootstrap analysis. Within each age category, the 95% confidence interval of expected frequencies was plotted, taking 10,000 random subsamples from all invasive isolates, with a sample size identical to the sample size of the clinical entity or M-type tested. Observed frequencies above the 95% boundary were considered significantly different from expected. Risk factors for death were evaluated by means of univariate and multivariate analysis. The vari-ables considered in univariate analysis were age, sex, clinical manifestation, M-type of the isolate, and the presence of comorbid conditions. Variables considered for inclusion into multivariate models were those potentially associated with death in univariate analysis (P<0.25) (23). Analyses were done using SPPS 11.0 software.

Results

Rates of diseaseFrom March 1992 through December 1996, 2437 GAS isolates were sent to the RIVM with corresponding patient data on age, sex, and source of isolation. Complete clinical and demographic data were obtained for 1312 cases (response rate, 54%). These cases did not differ in any respect (age, sex, source, M-type distribution) from those for which no complete questionnaire was obtained. Upon revision of clinical data, 880 cases of invasive GAS disease (66% definite, 34% probable invasive GAS disease) were identified, whereas the remaining 432 were classified as noninvasive. These were excluded from further analysis.

The estimated average annual incidence of invasive GAS disease was 2.2 cases per 100 000 persons (Figure 1). The median age of patients with invasive disease was 36 years (25%, 26 years; 75%, 65 years). In the 10,105 patients with non-invasive GAS disease, the median age was significantly lower: 20 years (25%, 5 years; 75%, 37 years; P<0.001). The incidence of invasive GAS disease was high in the age groups 0–4, 30–39, and >65 years.

The peak incidence of invasive GAS disease in the >65 years age group is

• 38

Tabl

e 1

Diff

eren

t por

tals

of e

ntry

in d

iffer

ent a

ge c

ateg

orie

s: in

vasiv

e di

seas

e vs

non

inva

sive

dise

ase

� ��

� ��

��

��

��

� ��

��

� � ��

� � ��

��

� ��

��

� ��

��

��

� ��

��

��

� � ��

��

��

� ��

� ��

��

��

��

��

��

��

� ��

��

��

� ��

��

��

��

� ��

��

��

� � ��

� � ��

��

��

��

� ��

��

��

� ��

� ��

��

��

��

� ��

��

� � ��

��

� ��

��

��

��

��

��

��

�

� ��

��

� ��

��

��

� ��

��

��

��

��

�

��

��

��

��

��

� ��

��

��

��

� ��

� � ��

��

��

��

� ��

� � � ��

��

� � ��

��

��

��

completely absent in patients with noninva-sive disease (Figure 1). Figure 2, which shows absolute numbers of all clinically and bacte-riologically documented cases of invasive GAS disease in the period March 1992–December 1996, illustrates that the peak in the middle age group is not solely attributable to cases of puerperal sepsis.

Acquisition of group A streptococci, and portal of entry

For clinically evaluated cases of invasive disease, a primary focus of infection (skin, respiratory tract, or perineo-genital area) could be defined on the basis of clinical and microbiological find-ings in 778 of 880 (88%) cases. In noninvasive cases of infection, a focus of infection could be defined in 9960 of 10 105 (99%) on the basis of the source of isolation. Most noninvasive iso-lates were of respiratory origin, whereas those that gave rise to invasive infections had entered predominantly through the skin (Table 1). On the basis of age-specific incidence of invasive GAS disease (Figure 1), all patients were cat-egorised into three age groups: 0–20, 21–55, and >56 years. For most GAS disease manifes-tations, a portal of entry was evident (e.g. the respiratory tract in GAS pneumonia or the perineo-genital area in GAS puerperal sepsis). In the meningitis cases in which a primary fo-cus could be established (n=20), the respiratory tract predominated in comparison to all invasive infections (18/20 vs 202/778, P<0.001). The es-tablished portals of entry in TSS cases were simi-larly distributed compared to non-TSS cases.

39 •

Figure 1 Age-specific incidence of invasive (left scale) and noninvasive GAS infection (right scale) in the Netherlands. Solid line, invasive isolates; dotted line, noninvasive isolates

�

�

�

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

��

��

��

��

Figure 2 Age distribution of clinically and bacteriologically documented invasive GAS infections in the Netherlands March 1992-December 1996 (n=880). Grey bars, puerperal sepsis cases without TSS. Solid black bars, TSS cases associated with puerperal sepsis. White bars, other cases of invasive GAS disease without TSS. Dotted bars, TSS cases associated with other invasive GAS diseases

• 40

Clinical presentationManifestations of invasive GAS disease are presented in Table 2. Seven percent of all invasive infections were acquired nosocomially. In 288 of 880 cases, the patient had an underlying disease (Table 2, for differentiation see Table 3). TSS developed in 215 (24%) patients, and 155 (18%) patients died from GAS in-fection. Mortality among the 665 patients with invasive GAS infection without TSS was 4% (n=28), whereas the case-fatality rate among the 215 patients with TSS was 59% (n=127) (P<0.005). The median time to death from the onset of symptoms was five days (range, 0–85 days). Men and women were equally distributed among all TSS cases. Signs and symptoms of systemic toxicity in TSS patients had a remarkable age distribution: diarrhoea or generalized rash was seen frequently in TSS patients who were 21–50 years of age: 31% and 31%, respectively, vs 15% and 7% in patients >65 years and 8% and 23% in those <20 years. As shown in figure 3, particular age distribution patterns were observed for different clinical entities: meningitis and arthritis were overrepre-sented among the younger age groups (<10 years), necrotizing fasciitis/myositis among the middle age group (21–55 years), and sepsis without a focus, and pneumonia among the elderly (>65 years). Table 3 lists patient characteristics and clinical and microbiological findings found to be associated with TSS and

Table 2 Patient characteristics and clinical features of 880 invasive group A streptococcal infections in relation to underlying condition and outcome

��

��

��

��

��

��

Total 880 (100) 36 288 (33) 215 (24) 155 (18) Diagnosis Traumatic wound infection 135 (15) 41 48 (36) 16 (12)a 16 (12) Necrotizing fasciitis/ myositis 112 (13) 42b 31 (28) 62 (55)a 30 (27)b Cellulitis 98 (11) 41 42 (43)b 16 (16) 11 (11) Pneumonia 88 (10) 60a 53 (60) a 38 (43)a 41 (47) a Arthritis 73 (8) 29 17 (23) 10 (14)b 4 (5) a Surgical wound infection 71 (8) 43 b 23 (32) 10 (14)b 8 (11) Puerperal sepsis 64 (7) 30 a 1 (2)a 6 (9)a 0 (0) a Sepsis without focus 63 (7) 53 a 34 (54)a 27 (43)a 23 (37) a Meningitis 36 (4) 30 b 12 (33) 6 (17) 6 (17) Peritonitis 22 (3) 36 4 (18) 7 (32) 3 (14)

Otherc 118 (13) 27 23 (19) 17 (14) 13 (11)

41 •

Tabl

e 3

Patie

nt a

nd m

icro

biol

ogic

al c

hara

cter

istic

s ass

ocia

ted

with

TSS

(n=2

15) a

nd fa

tal o

utco

me

(n=1

55).

For d

iffer

ent a

ge c

ateg

orie

s, th

e 0-

20 y

ear a

ge g

roup

serv

ed a

s ref

eren

ce. F

or a

ll ot

her v

aria

bles

, the

com

plem

enta

ry g

roup

was

use

d as

refe

renc

e (e

.g. n

on-n

ecro

tizin

g fa

sciit

is, n

=768

; num

ber o

f TSS

case

s, n=

153;

and

num

ber o

f fat

al ca

ses,

n=12

5) fo

r odd

s rat

io (O

R) c

alcu

latio

ns. C

hara

cter

istic

s sho

wn

are

thos

e in

clud

ed in

mul

tivar

iate

ana

lysis

, i.e

. tho

se w

ith P

val

ues <

0.25

in u

niva

riate

ana

lysis

. Fac

tors

con

sider

ed in

uni

varia

te a

naly

sis w

ere

age,

sex,

clin

ical

dise

ase

man

ifesta

tion,

M-ty

pe a

nd th

e pr

esen

ce o

f com

orbi

d co

nditi

ons.

n.s.,

non

signi

fican

t

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

� ��

��

��

��

��

��

� ��

��

��

��

��

��

� ��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

� ��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

� ��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

� ��

� ��

��

��

� ��

��

��

� � ��

��

��

��

��

��

��

� ��

� � ��

��

��

��

� ��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

� ��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

� ��

��

��

��

��

��

��

� ��

��

��

��

� ��

� ��

� ��

� ��

��

� ��

��

��

� ��

��

��

��

��

� ��

��

� ��

��

� �

• 42

Figure 3 Age distribution of traumatic wound infection (n=135), necrotizing fasciitis and/or myositis (n=112), cellulitis (n=98), pneumonia (n=88), arthritis (n=73), surgical wound infec-tion (n=71), puerperal sepsis (n=64), sepsis without focus (n=63), meningitis (n=36), peritoni-tis (n=22), and other invasive GAS disease manifestations (osteomyelitis, bursitis, endometritis, salpingitis, lymphadenitis, abscesses, n=118). Drawn line represents the age distribution of all invasive GAS infections. Thin lines correspond to the 95% confidence intervals in each age category, depending on sample size of the different clinical entities. X-axis: age (years); Y-axis: number of isolates.

�

�

��

��

��

��

��

�

�

��

��

��

��

��

�

�

��

��

��

��

��

��

�

�

��

��

��

��

��

��

�

�

��

��

��

��

��

�

�

�

�

�

��

��

��

��

��

��

��

��

�

�

�

�

�

�

�

�

��

��

�

�

��

��

��

��

��

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

43 •

�

�

��

��

��

��

��

��

��

�

�

��

��

��

��

�

�

�

�

�

��

��

�

�

�

�

�

��

��

��

��

�

�

�

�

�

��

��

�

��

��

��

��

��

��

��

��

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

��

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

��

�

Figure 4 Age distribution of M1 (n=186), M3 (n=101), M6 (n=47), M12 (n=46), M28 (n=71), M89 (n=43), and other M-types (n=386). Drawn line represents the age distribution for all invasive GAS infections. Thin lines correspond to the 95% confidence interval in each age category, depending on sample size of the different M-types. X-axis: age (years); Y-axis: number of isolates

• 44

death. Age >56 years, the presence of underlying cardiac or pulmonary disease, and the use of immunosuppressive drugs were patient characteristics associated with increased mortality. Furthermore, patients who presented with necrotizing fasciitis/myositis, pneumonia, or sepsis without a focus were more likely to have a fatal outcome, as were those infected with M1 or M3 types. When looking at TSS, age within 21–55 years (middle age group) was associated with TSS as well. The significance of most comorbid conditions, except for asplenism, dis-appeared in multivariate analysis.

Microbiology and M-typesAmong the 880 invasive isolates, 22 different M-types were found; 14% were M-nontypeable (Table 4). All M-nontypeable isolates yielded an emm ampli-con but did not hybridize with any of the M-type specific probes. The most common types, which accounted for 56% of all isolates, were M1 (21%), M3 (11%), M28 (8%), M6 (5%), M12 (5%), and M89 (5%). M-types were not uniformly distributed among different age groups (Figure 4): M12 was found predominantly among the younger age groups (<4 years), whereas M28 had a preference for the middle age groups. M1 and M3 were more evenly distributed among the younger, middle aged, and elderly patients. Of the 880 invasive isolates evaluated, the majority were obtained from blood (39%). Other sites of isolation of invasive strains were the skin (aspirates of abscesses, intraoperative or postmortem swabs or tissue) (34%), the respiratory tract (10%), the peri-neo-genital area (6%), synovial fluid (4%), cerebrospinal fluid (4%), and other (4%).

Association between M-types and clinical manifestationsAmong different GAS diseases, different M-types were not homogeneously dis-tributed (Table 5). M3 was strongly associated with necrotizing fasciitis/myosi-tis and was underrepresented among cases of traumatic wound infection and puerperal sepsis. M6 predominated in meningitis cases and M28 in puerperal sepsis cases. M1 and M3 constituted 16% (110/675) and 9% (62/675) of all invasive infections without TSS, respectively, and 35% (76/215, P<0.005) and 18% (39/215, P<0.005) of all TSS cases, respectively.

45 •

Table 5 Relative contribution of predominant M-types in different clinical manifestations and in TSS and fatal cases

Table 4 Contribution of different M-types and associated T-types in invasive isolates of group A streptococci

��

��

��

��

��

��

��

��

��

• 46

Discussion

In this study, we found an average annual incidence of invasive GAS disease in the Netherlands of 2.2 per 100 000 persons. In agreement with our findings, the estimated incidence from 1992 to 1995 was 2.4 per 100 000 persons in Canada (24). In a recent report from the USA, it was 3.5 per 100 000 in the period from 1997 to 1999 (25). Sweden has reported incidences from 1.5 to 3.5 per 100 000 between 1987 and 1996 (26), and similar figures were reported from Denmark and Norway (27, 28).

Soft tissue infections accounted for about half of all invasive GAS infections. TSS was seen as a complication in 24% of all cases and in 55% of necrotizing fasciitis/ myositis cases. Although no other clinical manifestation of GAS disease carried a higher propensity to be complicated by TSS, necrotizing fasciitis/my-ositis did not carry the highest case-fatality rate. Remarkably, patients suffering from GAS sepsis without a focus or from pneumonia were at highest risk to die. Cryptogenic bacteraemia has been associated with a relatively high case-fatal-ity (29, 30), and a recent report from Canada underscores the high mortality and rapid progressive nature of GAS pneumonia (31). These diseases affected primarily the elderly with more comorbidity. Most comorbid conditions were not associated with TSS development but did enter the equation when assessing risk factors for death. Cardiovascular and pulmonary disease facilitate progres-sion to fatal outcome in response to the stress of an invasive infection but do not appear to affect the bacterium’s intrinsic propensity to induce TSS.

Predominant M-types were M1, M3, M6, M12, and M28, similar to reports from other countries (7, 24, 32– 34). Site-specific manifestations of invasive disease such as puerperal sepsis (35), pneumonia (36), and perianal infection (37, 38) have been described in epidemics involving one particular M-type. Furthermore, clear correlations between the presence of certain protein binding factors and M-types have been described (39– 41). Protein binding factors con-fer selective tissue binding potential to the bacterium (42), allowing it to inter-act with specific organs or tissues. The association of M28 with gynaecological infections has been described (5, 9), but to our knowledge the propensity of M6 to cause meningitis has not been reported before. There was no indication of an epidemiological link between the different meningitis cases. M3 was over-represented among necrotizing fasciitis/ myositis cases but underrepresented among other soft tissue infections such as traumatic and surgical wound infec-

47 •

tions. Since necrotizing fasciitis/ myositis and surgical wound infections involve similar tissue, intrinsic tissue preference cannot provide an explanation for this discrepancy.

The age distribution for invasive GAS disease has a remarkable shape, with a high incidence in the younger, the middle aged, and the elderly. Some epide-miological studies on invasive GAS disease did not find a higher incidence in the middle age group (2, 3, 24, 43), whereas some of those that did attributed this phenomenon to puerperal sepsis (26). However, even in the age category 30–34 years, where the relative contribution of puerperal sepsis to all invasive GAS cases was highest, it contributed only 20% (28/135) to all cases. Frequent encounters with GAS result in the development of immunity, including the development of opsonizing antibodies against the M-antigen, which can be detected in patients’ sera decades after an infection (44). Likewise (neutralizing) antibodies to streptococcal exotoxins are acquired in the first decade of life (17, 45). The very low incidence in the 10–20 year age group may be a reflection of acquired immunity. Is then the relatively high incidence in the 30–39 year age group a reflection of waning immunity and/or of increased household exposure to GAS through children? Our data show that patients in the middle age group were, in most cases, previously healthy and that signs and symptoms of systemic toxicity (vomiting, diarrhoea, or rash) were more frequent than in the other age groups. These data may support the hypothesis of waning immunity, which has been suggested by investigators from Sweden (46). The high incidence of invasive streptococcal disease in the middle age group is unique when compared to the incidence of other invasive bacterial diseases such as infection due to Neisseria meningitidis or Streptococcus pneumoniae or sepsis in general. Infection due to N. meningitidis or S. pneumoniae typically shows the age distribution of the immune-compromised host with the burden of disease concentrated at the extremes of age (6, 47), whereas infection due to N. meningitidis has a second-ary peak in adolescents (48).

Remarkably, when the age distribution of invasive GAS disease is compared to that of noninvasive disease, the high incidence in the elderly is completely absent in noninvasive cases. The higher-than-expected values for the younger and mid-dle-aged patients are present in both types of GAS disease, but not in a fixed pro-portion. The extent to which patients are affected by invasive disease is not mir-rored by noninvasive GAS infections. This provides epidemiological support to the notion that invasive disease is not simply a reflection of noninvasive disease.

• 48

For most of our invasive isolates, a primary focus, or portal of entry, could be defined on the basis of clinical and microbiological findings. If invasive cases were a reflection of noninvasive cases, one would expect their portals of entry to be similar. This, however, was not the case: one-third of all noninvasive isolates were of dermal origin, whereas 62% of invasive isolates had entered through the skin. A predominance of invasive isolates of respiratory tract origin among young patients and of skin origin in the elderly has been described before in an M1 epidemic (26) and in a hospital-based setting (49).

Johnson et al. (16) compared a large number of invasive isolates to control isolates and concluded that M-type distribution patterns over time in invasive disease are to be interpreted as reflecting changes in noninvasive isolates. An-other recent study (50) showed the contrary: M1 isolates were overrepresented among invasive isolates. This corroborates the notion of increased intrinsic in-vasiveness of some M-types, as has been advocated by a large number of other studies (5, 7, 9–15). In our study, M1 and M3 were independently associated with fatality and TSS in multivariate analysis, indicating enhanced virulence among these M-types in invasive disease. Another study, conducted during the same time period in the Netherlands, evaluated the frequency of occurrence of different M-types in pharyngeal isolates. The relative frequency of type M1 among all pharyngeal GAS isolates was 4.5% (51). In contrast, we found type M1 in 21% of all invasive isolates in this study, supporting the notion of a higher invasive potential for type M1.

The findings of this study are limited in some respects. No clinical informa-tion was obtained regarding noninvasive isolates. Therefore, a primary focus of infection had to be deduced from their source of isolation. Although noninva-sive isolates can be obtained from various body sites, they are limited to one an-atomical location, thus validating this approach. Moreover, we did not receive the noninvasive isolates to assess the prevalence of the different M-types. Ongoing studies of GAS disease and the distribution of M-type patterns for isolates that cause GAS disease are needed to track trends in disease incidence, monitor M-type distribution, and evaluate the corresponding clinical manifes-tation associated with each M-type. Ideally, such data should be paralleled by data on superficial infections to identify newly emerging invasive clones.

49 •

References

1. Bisno AL (1990) Nonsuppurative poststreptococcal sequelae: rheumatic fever and glomer-ulonephritis. In: Mandell GL, Bennett JE, Dolin R (eds) Principles and practice of infec-tious diseases. Churchill Livingstone, New York, pp 1799–1810

2. O’Brien KL, Beall B, Barrett NL, Cieslak PR, Reingold A, Farley MM, Danila R, Zell ER, Facklam R, Schwartz B, Schuchat A (2002) Epidemiology of invasive group A streptococ-cus disease in the United States, 1995–1999. Clin Infect Dis 35:268–276

3. Sharkawy A, Low DE, Saginur R, Gregson D, Schwartz B, Jessamine P, Green K, McGeer A (2002) Severe group A streptococcal soft-tissue infections in Ontario: 1992–1996. Clin Infect Dis 34:454–460

4. Cone LA, Woodard DR, Schlievert PM, Tomory GS (1987) Clinical and bacteriologic observa-tions of a toxic shock-like syndrome due to Streptococcus pyogenes. N Engl J Med 317:146–149

5. Gaworzewska E, Colman G (1988) Changes in the pattern of infection caused by Strepto-coccus pyogenes. Epidemiol Infect 100:257–269

6. Martin GS, Mannino DM, Eaton S, Moss M (2003) The epidemiology of sepsis in the United States from 1979 through 2000. N Engl J Med 348:1546–1554

7. Schwartz B, Facklam RR, Breiman RF (1990) Changing epidemiology of group A strep-tococcal infection in the USA. Lancet 336:1167–1171

8. Bisno AL (1991) Group A streptococcal infections and acute rheumatic fever. N Engl J Med 325:783–793

9. Colman G, Tanna A, Efstratiou A, Gaworzewska ET (1993) The serotypes of Streptococcus pyogenes present in Britain during 1980–1990 and their association with disease. J Med Microbiol 39:165–178

10. Ferrieri P (1991) Microbiological features of current virulent strains of group A strepto-cocci. Pediatr Infect Dis J 10:S20–S24

11. Hauser AR, Stevens DL, Kaplan EL, Schlievert PM (1991) Molecular analysis of pyogenic exotoxins from Streptococcus pyogenes isolates associated with toxic shock-like syndrome. J Clin Microbiol 29:1562–1567

12. Johnson DR, Stevens DL, Kaplan EL (1992) Epidemiologic analysis of group A strepto-coccal serotypes associated with severe systemic infections, rheumatic fever, or uncompli-cated pharyngitis. J Infect Dis 166:374–382

13. Musser JM, Hauser AR, Kim MH, Schlievert PM, Nelson K, Selander RK (1991) Strep-tococcus pyogenes causing toxicshock- like syndrome and other invasive diseases: clonal diversity and pyrogenic exotoxin expression. Proc Natl Acad Sci USA 88:2668–2672

14. Reichardt W, Muller-Alouf H, Alouf JE, Kohler W (1992) Erythrogenic toxins A, B and C: occurrence of the genes and exotoxin formation from clinical Streptococcus pyogenes strains associated with streptococcal toxic shock-like syndrome. FEMS Microbiol Lett 79:313–322

15. Stromberg A, Romanus V, Burman LG (1991) Outbreak of group A streptococcal bacter-emia in Sweden: an epidemiologic and clinical study. J Infect Dis 164:595–598

16. Johnson DR, Wotton JT, Shet A, Kaplan EL (2002) A comparison of group A streptococci

• 50

from invasive and uncomplicated infections: are virulent clones responsible for serious streptococcal infections? J Infect Dis 185:1586–1595

17. Mascini EM, Jansze M, Schellekens JF, Musser JM, Faber JA, Verhoef-Verhage LA, Schouls L, van Leeuwen WJ, Verhoef J, van Dijk H (2000) Invasive group A streptococcal disease in the Nether-lands: evidence for a protective role of anti-exotoxin A antibodies. J Infect Dis 181:631–638

18. Schellekens JF, Schouls L, van Silfhout A, Elzenaar K, Brunings H, Blokpoel M, Top J, van Leeuwen WJ (1994) Invasieve infecties door beta-haemolytische Streptokokken Lancefield Groep A (Streptococcus pyogenes) in Nederland, 1992–1993. Report No. 253631001 of the Na-tional Institute of Public Health and Environmental Protection (RIVM). RIVM, Bilthoven

19. Schellekens JF, Schouls LM, van Pelt W, Esveld M, van Leeuwen WJ (1998) Group A streptococci: a change in virulence? Neth J Med 52:209

20. The Working Group on Severe Streptococcal Infections (1993) Defining the group A streptococ-cal toxic shock syndrome. Rationale and consensus definition. J Am Med Assoc 269:390–391

21. Griffith F (1934) The serologic classification of Streptococcus pyogenes. J Hyg Camb 34:542–58422. Kaufhold A, Podbielski A, Baumgarten G, Blokpoel M, Top J, Schouls L (1994) Rapid

typing of group A streptococci by the use of DNA amplification and non-radioactive al-lele-specific oligonucleotide probes. FEMS Microbiol Lett 119:19–25

23. Hosmer DW, Lemeshow S (1989) Applied logistic regression. Wiley, New York, pp 82–13424. Davies HD, McGeer A, Schwartz B, Green K, Cann D, Simor AE, Low DE (1996) Inva-

sive group A streptococcal infections in Ontario, Canada. Ontario Group A Streptococcal Study Group. N Engl J Med 335:547–554

25. Centers for Disease Control and Prevention (2002) Prevention of invasive group A strepto-coccal disease among household contacts of case patients and among postpartum and post-surgical patients: recommendations from the Centers for Disease Control and Prevention. Clin Infect Dis 35:950–959

26. Svensson N, Oberg S, Henriques B, Holm S, Kallenius G, Romanus V, Giesecke J (2000) Invasive group A streptococcal infections in Sweden in 1994 and 1995: epidemiology and clinical spectrum. Scand J Infect Dis 32:609–614

27. Andersen MM, Ronne T (1995) Group A streptococcal bacteraemias in Denmark 1987–1989. J Infect 31:33–37

28. Martin PR, Hoiby EA (1990) Streptococcal serogroup A epidemic in Norway 1987–1988. Scand J Infect Dis 22:421– 429

29. Bibler MR, Rouan GW (1986) Cryptogenic group A streptococcal bacteremia: experience at an urban general hospital and review of the literature. Rev Infect Dis 8:941–951

30. Ispahani P, Donald FE, Aveline AJ (1988) Streptococcus pyogenes bacteraemia: an old en-emy subdued, but not defeated. J Infect 16:37–46

31. Muller MP, Low DE, Green KA, Simor AE, Loeb M, Gregson D, McGeer A (2003) Clini-cal and epidemiologic features of group A streptococcal pneumonia in Ontario, Canada. Arch Intern Med 163:467–472

32. Beall B, Facklam R, Hoenes T, Schwartz B (1997) Survey of emm gene sequences and T-anti-gen types from systemic Streptococcus pyogenes infection isolates collected in San Francisco, Cali-

51 •

fornia; Atlanta, Georgia; and Connecticut in 1994 and 1995. J Clin Microbiol 35:1231–123533. Hoge CW, Schwartz B, Talkington DF, Breiman RF, MacNeill EM, Englender SJ (1993)

The changing epidemiology of invasive group A streptococcal infections and the emer-gence of streptococcal toxic shock-like syndrome. A retrospective population- based study. J Am Med Assoc 269:384–389

34. Kiska DL, Thiede B, Caracciolo J, Jordan M, Johnson D, Kaplan EL, Gruninger RP, Lohr JA, Gilligan PH, Denny FW Jr (1997) Invasive group A streptococcal infections in North Carolina: epidemiology, clinical features, and genetic and serotype analysis of causative organisms. J Infect Dis 176:992– 1000

35. Lehtonen OP, Kero P, Ruuskanen O, Gaworzewska ET, Hollo O, Erkkola R, Salmi T (1987) A nursery outbreak of group A streptococcal infection. J Infect 14:263–270

36. Centers for Disease Control and Prevention (2003) Outbreak of group A streptococcal pneumonia among Marine Corps recruits, California, November 1–December 20, 2002. MMWR Morb Mortal Wkly Rep 52:106–109

37. Muotiala A, Saxen H, Vuopio-Varkila J (1997) Group A streptococcal outbreak of perianal infection in a day-care centre. Adv Exp Med Biol 418:211–215

38. Petersen JP, Kaltoft MS, Misfeldt JC, Schumacher H, Schonheyder HC (2003) Community out-break of perianal group A streptococcal infection in Denmark. Pediatr Infect Dis J 22:105–109

39. Jaffe J, Natanson-Yaron S, Caparon MG, Hanski E (1996) Protein F2, a novel fibronectin-binding protein from Streptococcus pyogenes, possesses two binding domains. Mol Microbiol 21:373–384

40. Natanson S, Sela S, Moses AE, Musser JM, Caparon MG, Hanski E (1995) Distribution of fibronectin-binding proteins among group A streptococci of different M types. J Infect Dis 171:871–878

41. Rocha CL, Fischetti VA (1999) Identification and characterization of a novel fibronectin-binding protein on the surface of group A streptococci. Infect Immun 67:2720–2728

42. Valentin-Weigand P, Talay SR, Kaufhold A, Timmis KN, Chhatwal GS (1994) The fibronec-tin binding domain of the Sfb protein adhesin of Streptococcus pyogenes occurs in many group A streptococci and does not cross-react with heart myosin. Microb Pathog 17:111–120

43. Eriksson BK, Andersson J, Holm SE, Norgren M (1998) Epidemiological and clinical aspects of invasive group A streptococcal infections and the streptococcal toxic shock syn-drome. Clin Infect Dis 27:1428–1436

44. Lancefield RC (1959) Persistence of type-specific antibodies in man following infection with group A streptococci. J Exp Med 110:271–292

45. Basma H, Norrby-Teglund A, Guedez Y, McGeer A, Low DE, El Ahmedy O, Schwartz B, Kotb M (1999) Risk factors in the pathogenesis of invasive group A streptococcal infec-tions: role of protective humoral immunity. Infect Immun 67:1871–1877

46. Norrby-Teglund A, Pauksens K, Holm SE, Norgren M (1994) Relation between low ca-pacity of human sera to inhibit streptococcal mitogens and serious manifestation of dis-ease. J Infect Dis 170:585–591

47. Davidson M, Parkinson AJ, Bulkow LR, Fitzgerald MA, Peters HV, Parks DJ (1994) The epidemiology of invasive pneumococcal disease in Alaska, 1986–1990, ethnic differences

• 52

and opportunities for prevention. J Infect Dis 170:368–37648. Jones DM, Mallard RH (1993) Age incidence of meningococcal infection in England and

Wales, 1984–1991. J Infect 27:83– 8849. Bucher A, Martin PR, Hoiby EA, Halstensen A, Odegaard A, Hellum KB, Westlie L, Hal-

lan S (1992) Spectrum of disease in bacteraemic patients during a Streptococcus pyogenes serotype M-1 epidemic in Norway in 1988. Eur J Clin Microbiol Infect Dis 11:416–426

50. Ho PL, Johnson DR, Yue AW, Tsang DN, Que TL, Beall B, Kaplan EL (2003) Epidemio-logic analysis of invasive and noninvasive group A streptococcal isolates in Hong Kong. J Clin Microbiol 41:937–942

51. Zwart S, Ruijs GJ, Sachs AP, van Leeuwen WJ, Gubbels JW, de Melker RA (2000) Beta-haemolytic streptococci isolated from acute sore-throat patients: cause or coincidence? A case-control study in general practice. Scand J Infect Dis 32:377–384

Chapter 3

Epidemiological considerations following long-term surveillance of invasive group A streptococcal disease in The Netherlands, 1992–2003

B. Vlaminckx • W. van Pelt • J. Schellekens

Clinical Microbiology & Infection (2005) 11: 564–568

• 54

55 •

Abstract

A nationwide laboratory-based surveillance system for invasive group A strepto-coccal (GAS) infections was conducted in The Netherlands from March 1992 until December 2003. Until 1996, all isolates submitted were evaluated clinically. During this period there was a transition from passive to active surveillance for some of the participating laboratories, corresponding to a national coverage of 50%. During active surveillance, participating laboratories submitted twice as many isolates from invasive GAS disease, whereas the relative submission of isolates representing very severe manifestations (toxic shock-like syndrome, fa-tality) did not increase. From 1997 onwards, invasiveness was defined solely on the basis of source of isolation (without clinical evaluation). During the period of microbiological and clinical evaluation, microbiological evaluation alone was found to be specific (> 99%), but had limited sensitivity (66%). Estimation of the true rate of invasive GAS disease should be based on an active surveillance system with inclusion of both microbiological and clinical data.

• 56

Introduction

After a steady decrease in morbidity and mortality resulting from severe group A streptococcal (GAS) infections, the last two decades have witnessed a resurgence of invasive GAS disease. As a result, a number of prospective and population-based studies have reported rates of incidence and mortality. Some of these studies have relied on both clinical and microbiological findings for their definition of invasive GAS disease (1–6). According to the Working Group on Severe Streptococcal Infections, invasive GAS disease consists of ‘definite’ and ‘probable’ cases (7). Definite invasive GAS disease is defined as the isolation of Streptococcus pyogenes from a normally sterile site in conjunction with clinical symptoms of invasive bacterial disease. Isolation of S. pyogenes from a normally non-sterile site in association with invasive bacterial disease is defined as probable invasive GAS disease. Since the isolation of group A streptococci from a normally sterile body site constitutes invasive GAS disease, other studies have relied only on microbiological data, including isolates obtained from normally sterile sources such as blood, biopsy material, cerebrospinal fluid and synovial fluid (8–12).

For some population-based studies, it remains unclear whether surveillance was active or passive, which potentially affects the efficiency of reporting (13). In the present study, the effect of actively stimulating a GAS surveillance system was evaluated. Also, the additional value of integrating clinical data in this sur-veillance system, together with the implications for incidence calculations, was assessed. This report focuses on these epidemiological considerations.


Surveillance of invasive group A streptococcal diseaseNationwide laboratory-based surveillance for invasive GAS infections was con-ducted at the National Institute of Public Health (RIVM), Bilthoven, from March 1992 until December 2003 (average population during this period: 15 655 447) (14–17). Initially, all Dutch clinical microbiology laboratories, comprising Regional Public Health Laboratories (RPHLs) and other laborato-ries, participated in this surveillance system on a voluntary basis. GAS isolates, known or suspected to be associated with invasive disease, were sent to the

57 •

RIVM, together with information regarding patient age, gender and source of isolation. From March 1992 to December 1996, a questionnaire was returned to the medical microbiologist and treating physician to obtain clinical and de-mographic data for the patients involved. Invasive streptococcal disease was defined as the isolation of S. pyogenes from a normally sterile body site (definite invasive GAS disease), or from a normally non-sterile body site in conjunction with clinical symptoms consistent with invasive GAS disease (probable invasive GAS disease) (7). From May 1994, the surveillance system was organised for-mally for RPHLs (estimated national coverage of 50%), including reimburse-ment of costs for every invasive GAS isolate submitted. RPHLs were required to submit weekly reports on the number or absence of GAS isolates obtained from normally sterile sites. This active surveillance system with RPHLs contin-ued until December 2003. In December 1996, GAS surveillance was continued with RPHLs only (no other laboratories were involved), and clinical evaluation of reported and submitted isolates was discontinued.

Descriptive epidemiology To evaluate the effect of switching to active surveillance, relative increases or decreases in the numbers of isolates submitted by RPHLs after May 1994 were compared to data from RPHLs before May 1994, and from other laborato-ries before and after May 1994 as a reference. The sensitivity and specificity of microbiologically defined invasiveness was calculated from the standard of both microbiological and clinical evaluation (1992–1996). After December 1996, when clinical evaluation ceased, this sensitivity figure was used to estimate the annual incidence of invasive GAS disease from RPHL data.

Laboratory methodsIdentification of isolates was confirmed using bacitracin susceptibility testing and latex agglutination tests (Streptex; Wellcome, Dartford, UK). All isolates were M-genotyped by hybridisation of the denatured emm amplicon with a panel of 26 emm type specific probes in a reverse line blotting system (18).

StatisticsDifferences in group proportions were assessed by the chi-square test.

• 58

Results

During the surveillance period with clinical evaluation (March 1992 to De-cember 1996), RPHLs and other laboratories submitted 2437 GAS isolates to the RIVM (Figure 1). A complete questionnaire with demographic and clinical data was obtained for 1312 (54%) of 2437 patients. All isolates were M-geno-typed, and the age and gender of most patients (98% and 86%, respectively) were known. The data summarised in Table 1 show that there were no signifi-cant differences in patient (age distribution and gender) and isolate (M-type distribution) characteristics between the response and non-response groups.

During the period of passive surveillance (March 1992 to May 1994), RPHLs and other laboratories submitted similar numbers of isolates from clini-cally invasive GAS disease (RPHLs, 77 definite and 35 probable cases; other laboratories, 66 definite and 42 probable cases; figure 1). This confirmed that the national coverage of RPHLs for invasive GAS disease was circa 50%. Thus, incidence estimates for invasive GAS disease were the same as calculated from RPHLs and other laboratories in the period before May 1994. Furthermore, trends in time, estimated mortality and the relative contribution of predomi-nant M-types were very similar (data not shown).

As the surveillance system was not altered for the non-RPHL laboratories, their data served as a reference denominator for evaluating the effects of the switch to active surveillance for RPHLs after May 1994 (Table 2). After May 1994, RPHLs submitted significantly more isolates, both microbiologically in-vasive as well as microbiologically non-invasive or unknown (Figure 1 and Ta-ble 2). With integration of clinical data, the submission from RPHLs of isolates representing invasive (probable and definite) GAS disease increased two-fold, but submission of isolates associated with severe manifestations of invasive GAS disease, as measured by toxic shock-like syndrome (TSS) and fatality, did not increase (Table 2).

Following assessment of clinical and microbiological data, 880 of 1312 cases evaluated were characterised as either definite (n = 582) or probably invasive (n = 298) GAS disease (Figure 1). Of these evaluated cases, 585 isolates were considered to represent definite invasive GAS disease based on microbiological criteria alone (Figure 1). Three (< 1%) of these isolates were clinically nonin-vasive when the clinical evaluation was taken into account. When compared to

59 •

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

Figure 1 Numbers of group A streptococcal (GAS) isolates submitted by Regional Public Health Laboratories (RPHLs) and other laboratories in March 1992 – April 1994 and May 1994 – December 1996. Cases of definite invasive GAS disease are shown in solid squares. Cases of probable invasive GAS disease are shown in dashed squares. TSS, toxic shock-like syndrome.

Table 1 Evaluation of potential reporting bias of group A streptococci by comparison of patient (age and gender) and microbial (M-genotype) characteristics between the response and non-response groups

��

��

��

��

��

��

• 60

the reference standard of clinical and microbiological data (Table 3), defining invasive GAS disease solely on the basis of the source of the isolate had a spe-cificity for the presence of clinically invasive disease of 99% (429 ⁄ 432) and a sensitivity of 66% (582 ⁄ 880).

Since almost 100% specificity was obtained for defining invasive GAS dis-ease on the basis of microbiological criteria alone, active surveillance was con-tinued with RPHLs without clinical evaluation from January 1997 to Decem-ber 2003.

Discussion

In response to reports of a worldwide increase in the incidence of severe invasive GAS infections, a surveillance system that relied on the voluntary cooperation of microbiological laboratories was initiated in The Netherlands in 1992. How-ever, from May 1994, RPHLs were required to report weekly on the number of invasive GAS isolates, with financial compensation for each isolate reported and submitted. During the first two years of passive notification, both RPHLs and other laboratories generated very similar epidemiological data. Thus, the unaltered passive surveillance system for non-RPHL laboratories, which con-tinued until December 1996, was a valid reference denominator against which to evaluate the effect of a transition from a passive to an active surveillance system on different epidemiological parameters.

After the initiation of active surveillance, the estimated incidence of all inva-sive GAS infections based on RPHL data increased two-fold. Similarly, reports of GAS bacteraemias more than doubled between 2002 and 2003 in England, Wales and Northern Ireland as a result of the enhanced Strep- EURO surveil-lance scheme that was initiated in January 2003 (20). It was noted that, despite the overall increase in estimated incidence, the number of fatal cases or cases involving TSS notified by RPHLs did not increase after the institution of ac-tive surveillance. Thus, in a system involving passive notification, very severe cases appear to be detected almost as well as in a system of active stimulation. Possibly because of their severe effects, there appears to be high compliance in submitting these isolates, irrespective of the nature of the surveillance system.

61 •

Table 3 Comparison of invasiveness of group A streptococcal isolates defined on microbio-logical criteria alone with invasiveness defined on the basis of both clinical and microbiological criteria

Table 2 Relative change in numbers of group A streptococcal isolates submitted by Regional Public Health Laboratories (RPHLs) after May 1994, using data from other laboratories before and after May 1994 and RPHLs before May 1994 as a reference (e.g. factor increase for all isolates is calculated as (1289/579)/(226/343)= 3.4; P<0.0001, see figure 1)

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

• 62

The clinical symptoms of severe invasive GAS disease in The Netherlands during 1992–1996 have been reported in chapter 3 (17). Few cases of severe invasive GAS disease appear to have been missed during this period. How-ever, an actively stimulated surveillance system is much more sensitive for ‘milder’ (i.e., non-fatal and non- TSS) cases of invasive GAS disease, and can therefore be expected to give a more accurate estimate of the incidence of all cases of invasive GAS disease. Interestingly, the relative contribution of TSS to all cases of invasive GAS disease has been reported to be higher in a passive surveillance study compared to active surveillance studies; Svensson et al. (5) reported TSS in 24% of all cases, whereas active population- based studies reported TSS in 13% and 6% of cases, respectively (1,4). These reported dif-ferences may be the result of different study designs rather than geographical variations.

An accurate estimate of the incidence of invasive GAS disease must also rely on clinical evaluation (including the use of questionnaires) to detect probable cases. No active follow-up of questionnaires was undertaken, and thus report-ing bias might have been introduced. The overall response rate in the present study was 54%, similar to other reports regarding nationwide surveillance stud-ies of bacterial infectious diseases (21,22). Since age and gender of the patient were registered routinely following submission of an isolate, these data were known for both evaluated and non-evaluated cases. Furthermore, all isolates were subjected to M-genotyping. An overrepresentation of M-types M1 and M3 among severe cases of GAS disease has been found previously, as well as different age distribution patterns for particular manifestations of invasive GAS disease (14,17). No differences were found in these parameters between the evaluated and non-evaluated cases, implying that there was no reporting bias and that the isolates evaluated were a true representation of all cases of invasive GAS disease.

The present study found that a large number of the isolates evaluated did not meet the criteria for invasive GAS disease. In particular, the submission of isolates from RPHLs that did not meet the criteria for invasive GAS disease increased after May 1994. It is impossible to determine whether these isolates were truly suspected to be associated with invasive disease, or whether there were other reasons for this increased rate of submission. Of 1312 cases evalu-ated, 432 (33%) proved to be non-invasive, thereby allowing the specificity of

63 •

defining invasiveness solely on microbiological criteria to be assessed. This ap-proach was found to be > 99% specific; that is, isolation of S. pyogenes from a normally sterile site implied invasive disease. However, not all isolates that gave rise to an invasive syndrome were obtained from a sterile site, thereby limit-ing the sensitivity of this approach. Among all cases of invasive GAS disease, only 66% were detected on microbiological criteria alone, with the remaining isolates being identified only following the inclusion of clinical data. This phe-nomenon must be taken into account when comparing incidence figures from studies based only on microbiological criteria with those from studies that also include clinical evaluation. Indeed, a significant proportion of the reported geo-graphical variation in the incidence and serotype distribution of invasive GAS disease may be attributable to differences in the selection of cases, which are intrinsic to different surveillance systems.

Clinical evaluation adds significantly to the costs of a surveillance system. Therefore, GAS surveillance based on microbiological criteria alone was con-tinued after December 1996 (19), but was corrected to take account of the limited sensitivity of this approach, which was quantifiable after both clinical and microbiological evaluation for almost five years. An estimate of the true rate of invasive GAS disease in a country should be based on an active, popula-tion-based surveillance system which includes both microbiological and clini-cal data. However, invasiveness defined on the basis of microbiological criteria alone is very specific and can be used for monitoring incidence trends.

• 64

References

1. Davies HD, McGeer A, Schwartz B et al. Invasive group A streptococcal infections in Ontario, Canada. Ontario Group A Streptococcal Study Group. N Engl J Med 1996; 335: 547–554.

2. Kiska DL, Thiede B, Caracciolo J et al. Invasive group A streptococcal infections in North Carolina: epidemiology, clinical features, and genetic and serotype analysis of causative organisms. J Infect Dis 1997; 176: 992–1000.

3. Sharkawy A, Low DE, Saginur R et al. Severe group A streptococcal soft-tissue infections in Ontario: 1992–1996. Clin Infect Dis 2002; 34: 454–460.

4. O’Brien KL, Beall B, Barrett NL et al. Epidemiology of invasive group A streptococcus disease in the United States, 1995–1999. Clin Infect Dis 2002; 35: 268–276.

5. Svensson N, Oberg S, Henriques B et al. Invasive group A streptococcal infections in Sweden in 1994 and 1995: epidemiology and clinical spectrum. Scand J Infect Dis 2000; 32: 609–614.

6. Moses AE, Goldberg S, Korenman Z, Ravins M, Hanski E, Shapiro M. Invasive group A streptococcal infections, Israel. Emerg Infect Dis 2002; 8: 421–426.

7. The Working Group on Severe Streptococcal Infections. Defining the group A streptococcal toxic shock syndrome. Rationale and consensus definition. JAMA 1993; 269: 390– 391.

8. Colman G, Tanna A, Efstratiou A, Gaworzewska ET. The serotypes of Streptococcus pyogenes present in Britain during 1980–1990 and their association with disease. J Med Microbiol 1993; 39: 165–178.

9. Stromberg A, Romanus V, Burman LG. Outbreak of group A streptococcal bacteremia in Sweden: an epidemiologic and clinical study. J Infect Dis 1991; 164: 595–598.

10. Schwartz B, Facklam RR, Breiman RF. Changing epidemiology of group A streptococcal infection in the USA. Lancet 1990; 336: 1167–1171.

11. Johnson DR, Stevens DL, Kaplan EL. Epidemiologic analysis of group A streptococcal serotypes associated with severe systemic infections, rheumatic fever, or uncomplicated pharyngitis. J Infect Dis 1992; 166: 374–382.

12. Tyrrell GJ, Lovgren M, Forwick B, Hoe NP, Musser JM, Talbot JA. M types of group A streptococcal isolates, submitted to the National Centre for Streptococcus (Canada) from 1993 to 1999. J Clin Microbiol 2002; 40: 4466–4471.

13. Kriz P, Motlova J. Analysis of active surveillance and passive notification of streptococcal diseases in the Czech Republic. Adv Exp Med Biol 1997; 418: 217–219.

14. Schellekens JF, Schouls LM, van Pelt W, Esveld M, van Leeuwen WJ. Group A strepto-cocci: a change in virulence? Neth J Med 1998; 52: 209–217.

15. Schellekens JF, Schouls LM, van Silfhout A et al. The resurgence of group A streptococcal disease. Nederlands Tijdschrift Med Microbiol 1995; 4: 78–83.

16. Mascini EM, Jansze M, Schellekens JF et al. Invasive group A streptococcal disease in the Netherlands: evidence for a protective role of anti-exotoxin A antibodies. J Infect Dis 2000; 181: 631–638.

65 •

17. Vlaminckx B, Van Pelt W, Schouls L et al. Epidemiological features of invasive and nonin-vasive group A streptococcal disease in the Netherlands, 1992–1996. Eur J Clin Microbiol Infect Dis 2004; 23: 434–444.

18. Kaufhold A, Podbielski A, Baumgarten G, Blokpoel M, Top J, Schouls L. Rapid typing of group A streptococci by the use of DNA amplification and non-radioactive allelespecific oligonucleotide probes. FEMS Microbiol Lett 1994; 119: 19–25.

19. Vlaminckx B, Van Pelt W, Schouls L et al. Long-term surveillance of invasive group A streptococcal disease in The Netherlands, 1994–2003. Clin Microbiol Infect 2005; 11: 226– 231.

20. Health Protection Agency. Pyogenic and non-pyogenic streptococcal bacteraemias, Eng-land, Wales, and Northern Ireland: 2003. Commun Dis Rep Weekly 2004; 14 [serial on-line].

21. Abudu L, Blair I, Fraise A, Cheng KK. Methicillin-resistant Staphylococcus aureus (MRSA): a community-based prevalence survey. Epidemiol Infect 2001; 126: 351–356.

22. Schwartz RE, Lowe DA, Stayer SA, Pasquariello CA, Schlichting CM. Bacterial endo-carditis prophylaxis: what is recommended and what is practiced? J Clin Anesth 1994; 6: 5–9.

• 66

Chapter 4

Long-term surveillance of invasive group A streptococcal disease in The Netherlands, 1994–2003

B. Vlaminckx • W. van Pelt • L. Schouls • A. van Silfhout • E. Mascini • C. Elzenaar • T. Fernandes • A. Bosman • J. Schellekens

Clinical Microbiology & Infection (2005) 11: 226–231

• 68

69 •

Abstract

An active, nationwide laboratory-based surveillance study of invasive group A streptococcal (GAS) infections was conducted in The Netherlands from May 1994 until December 2003 (average population during this period was 15 729 704). Microbiologically invasive isolates were obtained from 1504 pa-tients, with most (70%) isolates cultured from blood. There was a clear seasonal pattern in invasive streptococcal infections, with an estimated annual incidence that peaked in 1996 (4.0 cases ⁄ 100 000 individuals ⁄ year) and was at its lowest in 1999 (2.0 cases ⁄ 100 000 individuals ⁄ year). Twenty-eight different M-types were identified, of which the most frequent were M1 (339 ⁄ 1504, 23%), M3 (187 ⁄ 1504, 12%), M89 (174 ⁄ 1504, 12%), M28 (164 ⁄ 1504, 11%), M12 (109 ⁄ 1504, 7%) and M6 (55 ⁄ 1504, 4%). There was a high degree of variation in the relative annual contributions of the predominant M-types, but varia-tions in M1 and M3 combined correlated with overall changes in the annual incidence. The contribution of the patient group aged ≥ 56 years to all cases of invasive GAS disease increased during the study period, whereas that of the group aged 0–20 years decreased. A peak in the incidence of invasive GAS dis-ease among the patient group aged 30–34 years did not vary during the study period, indicating that the high incidence of invasive GAS disease in this age group was age-specific rather than cohort-related.

• 70

Introduction

Group A streptococci continue to represent a major cause of infectious dis-ease-linked morbidity and mortality worldwide (1). Although most group A streptococcal (GAS) infections have a mild clinical course, a certain proportion progress to severe invasive disease, which can be complicated by the develop-ment of streptococcal toxic shock-like syndrome (TSS) (2,3). A remarkable fea-ture of invasive GAS disease is its epidemiology. After more than half a century of decreasing morbidity and mortality, a resurgence and persistence of severe invasive GAS infections has been noted since the mid-1980s (1–6). The cause for this resurgence of streptococcal infections remains enigmatic, and many hy-potheses have emerged, focusing either on impaired host defences or an increase in the virulence of the bacterium itself (7–9).

Previous studies have shown that immunity to proteins and toxins that at-tach to the cell wall may play a role in protection against severe invasive disease (10). The lack of protective anti-toxin antibodies is known to be age-dependent (11). Furthermore, a number of laboratory-based studies have suggested that the recent appearance of serotypes with increased virulence may provide an ex-planation for the observed changes in GAS epidemiology (2,12–17). Increased intrinsic virulence of some GAS types, particularly types M1 and M3, has been reported (16,18–23). These M-types frequently produce streptococcal pyrogen-ic exotoxins, which have the capacity to non-specifically activate a large subset of T-cells, resulting in a massive release of cytokines (24). This mechanism has been implicated in the pathogenesis of severe complications such as TSS (25).

Only a few population-based prospective epidemiological studies (4,16,18,20,26–28) have addressed the long-term dynamics of invasive GAS disease and the contribution of ‘virulent’ M-types over time. The present study analysed data obtained prospectively from a laboratory-based nationwide sur-veillance system for invasive GAS infections that has been ongoing for a period of ten years in The Netherlands. Changes in the incidence and the relative con-tributions of different M-types are addressed, as well as the temporal dynamics of the population affected by invasive GAS disease.

71 •


SurveillanceA nationwide laboratory-based surveillance system for invasive GAS infections in The Netherlands was formally organised with all regional public health labo-ratories (RPHLs) from May 1994. There are 16 RPHLs in The Netherlands, covering 50% of the population. Financial support was provided to RPHLs to enable submission of every GAS isolate from a normally sterile body site, to-gether with accompanying data on the source of the isolate and the age, sex and place of residence of the patient. Weekly reports on the number (or absence) of GAS isolates obtained from normally sterile sites were also submitted. This active surveillance system for invasive GAS disease in The Netherlands (average population during the study period was 15 729 704) remained in place until December 2003.

Invasive GAS disease includes definite and probable cases (29). Definite cas-es were defined as the isolation of GAS from a normally sterile site in conjunc-tion with clinical symptoms of invasive bacterial disease. GAS isolation from a normally nonsterile site in clinical association with invasive bacterial disease was considered to be probable invasive GAS disease, but was not included in this analysis. The study design of only counting GAS isolates cultured from a nor-mally sterile site results in an underestimation of the incidence of invasive GAS disease. The degree of underestimation was estimated using data from a nation-wide surveillance of invasive GAS disease, which included clinical evaluation, that was conducted in The Netherlands in the early 1990s (30). In this previous study, 34% of patients with clinically invasive GAS disease had GAS isolated from a non-sterile, unknown or indeterminate site. These probable cases of in-vasive GAS disease represented the limited sensitivity of the ‘microbiologically invasive’ criterion, which had a specificity of >99%, but a sensitivity of only 66% (30). The previous study received 50% of isolates from RPHLs, in accord-ance with the estimated coverage of the Dutch population by RPHLs in the present study. Therefore, calculation of the incidence of invasive GAS disease in The Netherlands from the present sentinel laboratory surveillance included a correction for the limited sensitivity (× 100 ⁄ 66), as well as a correction for the incomplete coverage of the population by RPHLs (× 100 ⁄ 50).

• 72

Definition of invasivenessInvasiveness was defined solely on microbiological criteria, i.e., the source of the isolate. GAS isolates obtained from blood, cerebrospinal, pleural, peritoneal or joint fluid, and aspirates or biopsy materials from deep, normally sterile tis-sues (e.g., lung, kidney, bone marrow, uterus) were included. National cover-age of the RPHLs over the study period was constant, and it was assumed that the sensitivity of the microbiological definition of invasiveness (based on the period 1992–1996, when all cases were also evaluated clinically) was constant throughout the study period.

Identification and typingIsolates sent to the National Institute of Public Health (RIVM) were confirmed as GAS following bacitracin susceptibility and latex agglutination tests (Strep-tex; Wellcome, Dartford, UK). All GAS isolates were subjected to M-genotyp-ing (31), thereby allowing a direct comparison with the internationally applied M-serotyping system.

Statistical analysisDifferences in group proportions were assessed by the chi-square or Fisher’s exact test, with P values < 0.05 considered significant.

Results

IncidenceFrom May 1994 to December 2003, the RIVM received microbiologically invasive isolates of GAS from 1504 patients. Most invasive isolates were from blood (1047 ⁄ 1504, 70%), with other isolation sites or specimens including aspirates of abscesses, intra-operative or post-mortem swabs or tissue (118 ⁄ 1504, 8%), respiratory tract (114 ⁄ 1504, 8%), synovial fluid (69 ⁄ 1504, 5%), perineal or genital tract (47 ⁄ 1504, 3%), cerebrospinal fluid (38 ⁄ 1504, 2%) and others (71 ⁄ 1504, 5%). The absolute number of cases obtained ⁄ month during the period May 1994 to December 2003 is shown in figure 1. There was a clear

73 •

Figure 1 Monthly distribution (observed absolute number of cases ⁄ month) of invasive group A streptococcal disease in The Netherlands (May 1994 to December 2003)

Figure 2 Annual incidence (cases ⁄ 100 000 individuals ⁄ year) of invasive group A streptococ-cal disease and the proportion of M1, M3, M89, M28, M12, M6, non-typeable and other M-types in The Netherlands (1995–2003)

�

�

��

��

��

��

��

��

��

��

��

��

� ��

��

��

� ��

� ��

��

� ��

��

��

� ��

� ��

��

� ��

��

��

� ��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

• 74

Figure 3 Contribution of different age groups (0–20 years, 21–55 years and ≥ 56 years) to all cases of invasive group A streptococcal disease (1995–2003)

Figure 4 Age distribution of cases of invasive group A streptococcal disease in the periods 1995–1997, 1998–2000 and 2001–2003

�

�

�

�

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

75 •

seasonal pattern in invasive streptococcal infections, with a low incidence at the end of each year, and with most cases occurring in late winter and spring. The estimated annual incidence from 1995 until 2003 is shown in figure 2. The estimated incidence was highest in 1996 (4.0 ⁄ 100 000 individuals ⁄ year), and also relatively high in 2002 (3.1 ⁄ 100 000 individuals ⁄ year), with a lowest estimated incidence of 2.0 ⁄ 100 000 individuals ⁄ year in 1999.

M-typesTwenty-eight different M-types were identified, of which the most frequent were M1 (339 ⁄ 1504, 23%), M3 (187 ⁄ 1504, 12%), M89 (174 ⁄ 1504, 12%), M28 (164 ⁄ 1504, 11%), M12 (109 ⁄ 1504, 7%) and M6 (55 ⁄ 1504, 4%). Other M-types constituted 19% (292 ⁄ 1504) of the isolates, but although all isolates yielded an emm amplicon, a further 184 ⁄ 1504 (12%) did not hybridise with one of the M-type specific probes (i.e., they were M-nontypeable). There was considerable variation in the relative prevalence of the predominant M-types during the study period. Thus, the proportion of M1 invasive isolates varied (Figure 2) from 11% to 12% in 1995 (23 ⁄ 197) and 1999 (11 ⁄ 100) to 40% in 2001 (62 ⁄ 154). M3 isolates were rare in 2000, but comprised almost 20% of all invasive isolates in 1995 (35 ⁄ 197) and 2002 (27 ⁄ 160). The relative pro-portions of M1 and M3 combined correlated with the overall annual incidence (Figure 2).

Age ⁄ sex distribution of invasive GAS isolatesThe date of birth was known for all cases in the present study. On the basis of a previous report (30) that described three age-groups in relation to the risk of contracting invasive GAS infections, figure 3 shows the relative proportions of the age groups 0–20, 21–55 or ≥ 56 years. The group aged ≥ 56 years consti-tuted 39% of all invasive cases in 1995 (76 ⁄ 197), increasing to > 50% in 2002 (82 ⁄ 160) and 2003 (80 ⁄ 157). The relative contribution of the youngest group (aged < 20 years) decreased from 19% (38 ⁄ 197) in 1995 to 8% (13 ⁄ 157) in 2003. Figure 4 indicates that the peak in incidence for patients aged 30–34 years did not vary significantly during the 10-year surveillance period. This was the case for both men and women (results not shown).

• 76

Discussion

The data from this study, obtained during a decade of nationwide laboratory-based surveillance of invasive GAS disease, indicated that the incidence of dis-ease was highest in 1995 and 1996 (4.0 ⁄ 100 000 individuals ⁄ year), with a similar level almost being reached again in 2002. The lowest incidence was in 1999 (2.0 ⁄ 100 000 individuals ⁄ year). Sharkawy et al. (26) reported a nearly two-fold increase in the incidence of invasive GAS infections during a 5-year surveillance period in Ontario, Canada. The highest incidence in 1995 was similar to that reported from the USA and Israel (27,32), while the lowest inci-dence in the present study was similar to that reported from Ontario (18) and Sweden (13).

For incidence calculations, it was necessary to make corrections for the national coverage of participating RPHLs, and for the fact that only isolates obtained from normally sterile sites were included. No significant changes oc-curred in the national coverage of RPHLs during the study period, and a previ-ous study from 1992 to 1996 found that a constant proportion of 66% of all cases of invasive GAS disease were associated with isolation of the bacterium from a sterile site (30). This sensitivity level was extrapolated for the present study, but it is not absolutely certain that it has remained constant over time.

Some studies, either historical (33) or hospital-based (34), have described cyclical peaks in the incidence of invasive GAS disease every three to four years. This was not observed in the present study, but in accordance with other reports (13,18,20,21), a clear seasonal pattern was noticed, with a sharp increase in incidence in late winter and a low incidence in summer, during the ten years of the study.

It has been reported previously that changes in the incidence of infections associated with M1 isolates is the main explanation for changes in the overall in-cidence of invasive GAS disease (13,35). The present study found that the contri-bution of M1 isolates, and of M1 and M3 isolates combined, correlated with the overall annual incidence. These two M-types are associated particularly with TSS and fatality (36). Therefore, cases of GAS disease may have a more severe course during periods with a high overall incidence of invasive GAS disease.

The overall prevalence of M1 isolates was 23%. Other M-types identified frequently were M3 (12%), M89 (12%), M28 (11%), M12 (7%) and M6

77 •

(4%). In total, 12% of all isolates were M-nontypeable, which is consistent with other reports (37,38). The high prevalence of M1 isolates was expected fol-lowing the results of other studies. However, in a study conducted in the 1980s by the UK Central Public Health Laboratory, the prevalence of M1 isolates was much lower (10.4%) (38), although almost half of all isolates originated from throat swabs. Davies et al. (18) reported prevalences of 24% for M1, 7.4% for M12, 6.5% for M4, 6.2% for M28, and 5.8% for M3, while O’Brien et al. (20) found that these M-types accounted for 20.8%, 7.6%, 4.1%, 9.2% and 7.1%, respectively, of invasive isolates. Tyrell et al. (37) reported that M1 isolates accounted for 28.2% of all blood culture isolates, and argued that this much higher proportion, compared to the UK, might be caused either by geo-graphical reasons or by the fact that the UK study included a large number of non-invasive isolates. In a previous study in The Netherlands (19), it was found that M1 and M3 isolates were overrepresented among isolates associated with TSS and mortality, supporting the hypothesis that these serotypes have enhanced virulence (19). Furthermore, a separate nationwide study conducted in The Netherlands in 1994–1995 (39), investigating pharyngeal carriage of GAS, showed a much lower proportion (4.5%) of M1 isolates, thereby provid-ing evidence for an association of this serotype with invasiveness. Therefore, the inclusion of noninvasive isolates in some studies might provide an explanation for a lower incidence of M1 isolates. However, the dynamics of M1 isolates over time should be emphasised. In the present study, involving a clearly defined geographical region, the yearly proportion of M1 isolates among all invasive isolates varied from 10% to 40%, with no evidence for the occurrence of local outbreaks. This hampers any comparison of the proportions of M1 isolates in different geographical regions over different time periods.

The proportion of any particular M-type among all invasive isolates re-flects the prevalence of that serotype in the community, its invasiveness and the M-specific immunity in the population (20). However, contemporaneous data regarding all noninvasive isolates circulating in the community is lacking. Therefore, it is not clear whether the observed fluctuations in the incidence of invasive infections with M1 and M3 isolates reflect the dynamics in the com-munity or relative alterations in invasiveness or population immunity. Of the predominant M-types, M1, M3 and M6 isolates showed the most conspicuous fluctuations. Notably, these are the M-types that were found previously to be highly clonal in nature (36).

• 78

A high incidence of invasive infections in the elderly and youngest age groups is well-recognised (4,18,20,21,26,27,35,37,40). In the present study, both men and women in the middle-aged category also had an increased incidence of invasive GAS disease. Whether this is associated with (selective) decreased im-munity in the middle-aged group (41) or re-exposure through household con-tacts is unclear. However, it does not appear to be a cohort phenomenon, as the peak in the middle-aged group did not shift correspondingly during a decade of surveillance. Previous studies have shown that immunity to proteins and toxins that attach to the cell wall may play a role in protection against severe invasive disease (10). Furthermore, a lack of protective anti-toxin antibodies has been found to be agedependent (11,29). The present study provides epidemiologi-cal support that this is specific for an age group rather than a cohort. However, there was a change in the relative contributions of the different age groups to all invasive GAS infections, in that the group aged 0–20 years contributed progres-sively less, while the group aged ≥ 56 years gradually contributed more. This change was not paralleled by any significant alterations in the population age distribution during the decade of the study.

In summary, the epidemiology of invasive GAS disease and the contribu-tion of individual M-types evolve over time. Considering the overall dynamics of the various M-types, and the possibility of the appearance of new serotypes (2,12–17), continued monitoring of the distribution of different serotypes is necessary. New M-types, or virulent clones within a serotype, might appear and, with little specific immunity in the population, gain predominance. Pop-ulationbased surveillance provides data for monitoring incidence trends and identifying the importance of particular virulent strains, thus providing a ra-tional basis for disease control measures and future vaccine development.

79 •

References

1. Efstratiou A. Group A streptococci in the 1990s. J Antimicrob Chemother 2000; 45(suppl): 3–12.

2. Holm SE, Norrby A, Bergholm AM, Norgren M. Aspects of pathogenesis of serious group A streptococcal infections in Sweden, 1988–1989. J Infect Dis 1992; 166: 31–37.

3. Carapetis J, Robins-Browne R, Martin D, Shelby-James T, Hogg G. Increasing severity of invasive group A streptococcal disease in Australia: clinical and molecular epidemiological features and identification of a new virulent M-nontypeable clone. Clin Infect Dis 1995; 21: 1220–1227.

4. Kiska DL, Thiede B, Caracciolo J et al. Invasive group A streptococcal infections in North Carolina: epidemiology, clinical features, and genetic and serotype analysis of causative organisms. J Infect Dis 1997; 176: 992–1000.

5. Bisno AL. Group A streptococcal infections and acute rheumatic fever. N Engl J Med 1991; 325: 783–793.

6. Kaul R, McGeer A, Low DE, Green K, Schwartz B. Population- based surveillance for group A streptococcal necrotizing fasciitis: clinical features, prognostic indicators, and microbiologic analysis of seventy-seven cases. Ontario Group A Streptococcal Study. Am J Med 1997; 103: 18–24.

7. Norrby-Teglund A, Chatellier S, Low DE, McGeer A, Green K, Kotb M. Host variation in cytokine responses to superantigens determine the severity of invasive group A streptococ-cal infection. Eur J Immunol 2000; 30: 3247–3255.

8. Norrby-Teglund A, Kotb M. Host–microbe interactions in the pathogenesis of invasive group A streptococcal infections. J Med Microbiol 2000; 49: 849–852.

9. Kotb M, Norrby-Teglund A, McGeer A et al. An immunogenetic and molecular basis for differences in outcomes of invasive group A streptococcal infections. Nat Med 2002; 8: 1398–1404.

10. Akesson P, Rasmussen M, Mascini E et al. Low antibody levels against cell wall-attached proteins of Streptococcus pyogenes predispose for severe invasive disease. J Infect Dis 2004; 189: 797–804.

11. Mascini EM, Jansze M, Schellekens JF et al. Invasive group A streptococcal disease in the Netherlands: evidence for a protective role of anti-exotoxin A antibodies. J Infect Dis 2000; 181: 631–638.

12. Colman G, Tanna A, Efstratiou A, Gaworzewska ET. The serotypes of Streptococcus py-ogenes present in Britain during 1980–1990 and their association with disease. J Med Microbiol 1993; 39: 165–178.

13. Stromberg A, Romanus V, Burman LG. Outbreak of group A streptococcal bacteremia in Sweden: an epidemiologic and clinical study. J Infect Dis 1991; 164: 595–598.

14. Schwartz B, Facklam RR, Breiman RF. Changing epidemiology of group A streptococcal infection in the USA. Lancet 1990; 336: 1167–1171.

15. Johnson DR, Stevens DL, Kaplan EL. Epidemiologic analysis of group A streptococcal

• 80

serotypes associated with severe systemic infections, rheumatic fever, or uncomplicated pharyngitis. J Infect Dis 1992; 166: 374–382.

16. Hoge CW, Schwartz B, Talkington DF, Breiman RF, MacNeill EM, Englender SJ. The changing epidemiology of invasive group A streptococcal infections and the emergence of streptococcal toxic shock-like syndrome. A retrospective population-based study. JAMA 1993; 269: 384–389.

17. Musser JM, Kapur V, Kanjilal S et al. Geographic and temporal distribution and molecu-lar characterization of two highly pathogenic clones of Streptococcus pyogenes express-ing allelic variants of pyrogenic exotoxin A (Scarlet fever toxin). J Infect Dis 1993; 167: 337–346.

18. Davies HD, McGeer A, Schwartz B et al. Invasive group A streptococcal infections in Ontario, Canada. Ontario Group A Streptococcal Study Group. N Engl J Med 1996; 335: 547–554.

19. Schellekens JF, Schouls LM, van Pelt W, Esveld M, van Leeuwen WJ. Group A strepto-cocci: a change in virulence? Neth J Med 1998; 52: 209–217.

20. O’Brien KL, Beall B, Barrett NL et al. Epidemiology of invasive group A streptococcus disease in the United States, 1995–1999. Clin Infect Dis 2002; 35: 268–276.

21. Svensson N, Oberg S, Henriques B et al. Invasive group A streptococcal infections in Sweden in 1994 and 1995: epidemiology and clinical spectrum. Scand J Infect Dis 2000; 32: 609–614.

22. Cone LA, Woodard DR, Schlievert PM, Tomory GS. Clinical and bacteriologic observa-tions of a toxic shocklike syndrome due to Streptococcus pyogenes. N Engl J Med 1987; 317: 146–149.

23. Stevens DL, Tanner MH, Winship J et al. Severe group A streptococcal infections associ-ated with a toxic shock-like syndrome and scarlet fever toxin A. N Engl J Med 1989; 321: 1–7.

24. Abe J, Forrester J, Nakahara T, Lafferty JA, Kotzin BL, Leung DY. Selective stimulation of human T cells with streptococcal erythrogenic toxins A and B. J Immunol 1991; 146: 3747–3750.

25. McCormick JK, Yarwood JM, Schlievert PM. Toxic shock syndrome and bacterial super-antigens: an update. Ann Rev Microbiol 2001; 55: 77–104.

26. Sharkawy A, Low DE, Saginur R et al. Severe group A streptococcal soft-tissue infections in Ontario: 1992–1996. Clin Infect Dis 2002; 34: 454–460.

27. Moses AE, Goldberg S, Korenman Z, Ravins M, Hanski E, Shapiro M. Invasive group A streptococcal infections, Israel. Emerg Infect Dis 2002; 8: 421–426.

28. Ikebe T, Murai N, Endo M et al. Changing prevalent T serotypes and emm genotypes of Streptococcus pyogenes isolates from streptococcal toxic shock-like syndrome (TSLS) patients in Japan. Epidemiol Infect 2003; 130: 569–572.

29. The Working Group on Severe Streptococcal Infections. Defining the group A strep-tococcal toxic shock syndrome. Rationale and consensus definition. JAMA 1993; 269: 390– 391.

81 •

30. Vlaminckx B, Van Pelt W, Schouls L et al. Epidemiological features of invasive and nonin-vasive group A streptococcal disease in the Netherlands, 1992–1996. Eur J Clin Microbiol Infect Dis 2004; 23: 434–444.

31. Kaufhold A, Podbielski A, Baumgarten G, Blokpoel M, Top J, Schouls L. Rapid typing of group A streptococci by the use of DNA amplification and non-radioactive allelespecific oligonucleotide probes. FEMS Microbiol Lett 1994; 119: 19–25.

32. Anonymous. Prevention of invasive group A streptococcal disease among household con-tacts of case patients and among postpartum and postsurgical patients: recommendations from the Centers for Disease Control and Prevention. Clin Infect Dis 2002; 35: 950–959.

33. Krause RM. Dynamics of emergence. J Infect Dis 1994; 170: 265–271.34. Samuels N, Yinnon AM, Rudensky B. Cyclical peaking of streptococcal infections: recent

and historical trends. J Infect Dis 1995; 172: 1165.35. Eriksson BK, Andersson J, Holm SE, Norgren M. Epidemiological and clinical aspects of

invasive group A streptococcal infections and the streptococcal toxic shock syndrome. Clin Infect Dis 1998; 27: 1428–1436.

36. Vlaminckx BJ, Mascini EM, Schellekens J et al. Site-specific manifestations of invasive group A streptococcal disease: type distribution and corresponding patterns of virulence determinants. J Clin Microbiol 2003; 41: 4941–4949.

37. Tyrrell GJ, Lovgren M, Forwick B, Hoe NP, Musser JM, Talbot JA. M types of group A streptococcal isolates, submitted to the National Centre for Streptococcus (Canada) from 1993 to 1999. J Clin Microbiol 2002; 40: 4466–4471.

38. Gaworzewska E, Colman G. Changes in the pattern of infection caused by Streptococcus pyogenes. Epidemiol Infect 1988; 100: 257–269.

39. Zwart S, Ruijs GJ, Sachs AP, van Leeuwen WJ, Gubbels JW, de Melker RA. Beta-haemo-lytic streptococci isolated from acute sore-throat patients: cause or coincidence? A casec-ontrol study in general practice. Scand J Infect Dis 2000; 32: 377–384.

40. Schellekens JF, Schouls LM, van Silfhout A et al. The resurgence of group A streptococcal disease. Nederlands Tijdschrift Med Microbiol 1995; 4: 78–83.

41. Norrby-Teglund A, Pauksens K, Holm SE, Norgren M. Relation between low capacity of human sera to inhibit streptococcal mitogens and serious manifestation of disease. J Infect Dis 1994; 170: 585–591.

• 82

Chapter 5

Site-Specific Manifestations of Invasive Group A Streptococcal Disease: Type Distribution and Corresponding Patterns of Virulence Determinants

B. Vlaminckx • E. Mascini • J. Schellekens • L. Schouls •A. Paauw • A. Fluit • R. Novak • J. Verhoef • F. Schmitz

Journal of Clinical Microbiology (2003) 41: 4941–4949

• 84

85 •

Abstract

As part of a national surveillance program on invasive group A streptococci (GAS), isolates that caused specific manifestations of invasive GAS disease in The Netherlands were collected between 1992 and 1996. These site-specific GAS infections involved meningitis, arthritis, necrotizing fasciitis, and puer-peral sepsis. An evaluation was performed to determine whether GAS virulence factors correlate with these different disease manifestations. PCRs were devel-oped to detect 9 genes encoding exotoxins and 12 genes encoding fibronectin binding proteins. The genetic backgrounds of all isolates were determined by M-genotyping and pulsed-field gel electrophoresis (PFGE) analysis. The pre-dominant M-types included M1, M2, M3, M4, M6, M9, M12, and M28. Most M-types were associated with all manifestations of GAS disease. However, M2 was found exclusively in patients with puerperal sepsis, M6 predominated in patients with meningitis, and M12 predominated in patients with GAS ar-thritis. While characteristic gene profiles were detected in most M-types, the resolution of detection of different gene profiles within M-genotypes was en-hanced by PFGE analysis, which clearly demonstrated the existence of some clonal lineages among invasive GAS isolates in The Netherlands. M1 isolates comprised a single clone carrying highly mitogenic toxin genes (speA, smeZ) and were associated with toxic shock-like syndrome. Toxin profiles were highly conserved among the most virulent strains, such as M1 and M3.

• 86

Introduction

Since the mid-1980s, a conspicuous increase in the incidence of invasive infec-tions caused by group A streptococci (GAS) has been observed worldwide (3, 9, 10, 18, 27, 37, 38, 53, 70). Invasive GAS diseases include various clinical syndromes such as bacteremia, arthritis, pneumonia, puerperal sepsis, menin-gitis, and necrotizing fasciitis. These infections are often of a fulminant nature and can be complicated by streptococcal toxic shock-like syndrome (TSS). This clinical entity is particularly associated with necrotizing fasciitis but can occur as a complication of any GAS infection (14, 45, 53, 67). The cause of the resur-gence of severe GAS infections remains enigmatic (14, 49). It has been shown that the immunogenetic background of the host influences the outcome of in-vasive streptococcal infection (4, 39, 48). In addition, predisposing factors such as trauma, diabetes, alcoholism, or varicella have been documented. Nonethe-less, otherwise healthy subjects can be affected as well (15, 38, 41, 62). This ob-servation suggests a relevant role for changes in bacterial virulence, which was the focus of this study. The pathogenicity of GAS disease has been associated with the expression of a plethora of gene products that comprise extracellular and cell-associated proteins.

Extracellular proteins that reportedly play a role in the pathogenesis of GAS are represented by streptococcal exotoxins. They include SPE-A, SPE-B, SPE-C, SPE-F, SPE-G, SPE-H, SPE-J, SSA, and SMEZ, most of which have super-antigenic properties that allow them to activate large subsets of T cells, with massive cytokine release as a consequence (1, 23, 33, 43, 47, 50, 56, 57). They are associated with pyrogenicity and enhancement of endotoxic shock. Recent-ly, more streptococcal exotoxins have been described (6), and SMEZ has been identified as the major immunologically active agent (76).

Bacterial attachment to host tissue is the first step leading to colonization and subsequent development of invasive disease. Binding of fibronectin pro-motes adherence to epithelial cells. Several studies have suggested an association between fibronectin binding and the pathogenic potential of GAS (22, 25, 42, 46, 61, 74, 77). Invasion of streptococci into deeper tissues may be facilitated by specialized adhesion mechanisms. In addition, a particular adhesion factor might confer a certain site or tissue specificity to the GAS (54). Although many adhesins have been identified, an association of these with clinically defined

87 •

syndromes has not been investigated. The aim of the present study was to further investigate possible relationships

between genes encoding M proteins, exotoxins, and protein binding factors and the clinical manifestations of invasive Streptococcus pyogenes disease. Between 1992 and 1996, all Dutch patients with microbiologically documented invasive GAS disease were registered in a national surveillance study. We selected all GAS isolates from Dutch patients with necrotizing fasciitis, puerperal sepsis, septic arthritis, and meningitis; determined their M-genotypes; and evaluated them for the presence of 9 exotoxin and 12 fibronectin binding protein genes. This is the first report on an extensive gene analysis of all clinically well-docu-mented GAS isolates obtained in one country over a limited time span.


SurveillanceFrom 1992 to 1996, Dutch patients with microbiologically defined invasive GAS disease were registered in a nationwide laboratory-based surveillance sys-tem (62, 63). Isolates were considered pathogenic and invasive when they were obtained from normally sterile body sites. Upon receipt of an invasive strain, extensive clinical and demographic data for the patients involved were obtained by mailing a standardized questionnaire to the treating physician; the response rate was 57%. For this study, we selected all invasive GAS isolates that had given rise to meningitis, septic arthritis, necrotizing fasciitis, or puerperal sep-sis in patients for whom corresponding clinical and demographic data were known. GAS meningitis was defined on the basis of GAS-positive cultures of either cerebrospinal fluid or blood, together with clinically defined meningitis. GAS arthritis was defined on the basis of GAS-positive cultures of either syno-vial fluid or blood, combined with clinically defined arthritis. Puerperal sepsis was defined as postpartum endometritis in combination with positive blood cultures. The occurrence of clinical complications was registered as TSS and was defined as the presence of hypotension in combination with at least two of the following: acute renal failure, coagulation or liver abnormalities, rash, or necrotizing fasciitis (2).

• 88

TypingAll GAS isolates were subjected to M-genotyping. The M-genotype was de-termined by hybridization of the denatured emm amplicon with a panel of 26 emm type-specific probes in a reverse line blotting system, as described previ-ously (36).

Bacterial strains and culture conditionsStrains were grown on blood agar plates at 5% CO2 and 37°C overnight. Ge-nomic DNA as the target for PCR assays was extracted by heating bacterial suspensions for 10 min at 95°C (64).

Identification of toxin and fibronecting binding protein genesSequences specific for speA (32), speB (34), speC (24), speF (28), speG (56), speH (56), speJ (56), smeZ (21), ssa (59), cpa (55), cpa-1 (T. Miyoshi-Akiyama, N. Wakisaka, J. Zhao, and T. Uchiyama, unpublished data), fba (75), fbp-54 (Miyoshi-Akiyama et al., unpublished), pfbp (61), prtf-1 (66), prtf-2 (29), prtf-15 (35), sciA (58), sciB (78), sfb (72), and sfb-II (40) were detected by PCR on a model 9600 thermocycler (Perkin-Elmer, Gouda, The Netherlands) with the primers listed in Table 1. Amplification of all genes was carried out under the following conditions: an initial 5-min denaturation step at 96°C, followed by 30 cycles of denaturation at 96°C for 55 s, 65 s of annealing at the appropriate temperature for each gene specified in Table 1, and 70 s of extension at 72°C, with a final extension step at 72°C for 5 min. All PCR products were subjected to electrophoresis on 1.5% agarose gels and then stained with ethidium bro-mide and visualized under UV light. In another study (64), this PCR protocol had a sensitivity of 96%. To minimize the risk of possible sequence variation affecting the primer binding site, for each gene we aligned all published se-quences and selected a conserved region for primer design. Furthermore, partial sequencing of the first amplification products yielded no nonspecific sequences for any of the 21 genes under investigation, indicating the complete specificities of the primers used.

89 •

Typing by PFGEFor pulsed-field gel electrophoresis (PFGE), the bacteria were suspended to a density of a McFarland no. 3 standard, and bacterial plugs were made in 0.75% pulsed-field-certified agarose (Bio-Rad, Hercules, Calif.) with Tris-HCl and NaCl (pH 8.0) at final concentrations of 10 mM and 1 M, respectively. After solidification, the plugs were treated with lysis buffer (6 mM Tris-HCl, 1MNaCl, 100 mM EDTA [pH 8.0], 0.2% sodium desoxycholate, 0.5% so-dium laurylsarcosine, 0.5% Brij 58) containing freshly added lysozyme (2.88 g/liter; Sigma, St. Louis, Mo.) for 48 h at 37°C. After 48 h of incubation at 56°C in 0.5 M EDTA (pH 9.0)–1.0 g of proteinase K per liter, the plugs were washed in TE (Tris-EDTA) buffer and stored at 4°C until use. Digestion of the bacterial DNA was performed with restriction endonuclease SmaI (Roche,

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

Table 1 Oligonucleotide primers and reference strains used for gene detection

• 90

Almere, The Netherlands) according to the guidelines of the manufacturer. Restriction fragments were then separated by PFGE in a 1% pulsed-field-certi-fied agarose gel (Bio-Rad) with 1 ml of thiourea per liter with a CHEF-DRII drive module (Bio-Rad). The initial pulse time of 1.0 s was increased linearly to 50.0 s over 24 h at a voltage of 6 V/cm and a temperature of 11.3°C. The gels were then stained with ethidium bromide for 15 min, and the restriction frag-ments were visualized under UV light. The images of the gels were imported into Gelcompar software (version 4.1; Applied Maths, Kortrijk, Belgium) by using band-based clustering, and the banding patterns were evaluated for each individual M-type. By use of the dendrogram, isolates with a genetic related-ness of >80% were considered to represent the same PFGE type. Confirmation of genetic similarity or difference was performed by visual interpretation of the gels.

Statistical analysisChi-square tests were performed to compare data from patients with different site-specific infections. P values <0.05 were considered significant. To show more complex relationships between some virulence genes and site-specific disease independent of the M-genotype, logistic regression was performed by using SPSS (version 10) software.

Results

SurveillanceIn our nationwide surveillance system, a total of 170 isolates were obtained between 1992 and 1996 from patients with microbiologically defined cases of invasive GAS disease causing meningitis (27 of 170), arthritis (41 of 170), necrotizing fasciitis (61 of 170), and puerperal sepsis (41 of 170). Twenty-eight percent (47 of 170) of these cases of invasive GAS disease were complicated by the occurrence of TSS. The isolates were derived from 170 different patients and represented 26% (170 of 650) of a collection of all isolates from patients infected with invasive GAS isolates that were sent to the national public health laboratory during the time period.

91 •

TypingIn order to establish whether there is a relationship between genetic profile and disease type, all isolates were subjected to M-genotyping and PFGE analysis. Table 2 shows the results of M-genotyping of the isolates involved in the four different site- or tissue-specific infections. A total of 20 different M-types were detected; however, 30 isolates could not be typed because of a lack of hybridi-zation with any of the oligonucleotides used. Predominant M-types were de-fined as those that contained five or more isolates that could be typed. These predominant M-types were M1 (25% of isolates), M2 (3%), M3 (14%), M4 (6%), M6 (6%), M9 (4%), M12 (6%), and M28 (7%). They comprised a to-tal of 122 (72%) of all M-types involved and were nonuniformly distributed among the patients with different site- or tissue-specific infections. Remarkably, M1 predominated in TSS cases but was not as prevalent in non-TSS cases (21 of 47 and 22 of 123, respectively [P < 0.001]). TSS was seen as a complication in all groups with invasive GAS diseases but significantly less often in patients with puerperal sepsis (3 of 41 versus 44 of 129 [P < 0.001]). M3 also tended to be overrepresented among TSS cases, but this was not supported by statistical significance.

PFGE and identification of toxin and fibronectin binding protein genesFigure 1 shows the PFGE patterns and their relation to the M-types. Among the eight predominant M-types (i.e., M1, M2, M3, M4, M6, M9, M12, and M28), we found 23 different patterns with differences of more than three bands (Figure 1 and Table 3). Clonal lineages were demonstrated among M1, M3, M6, and M12 isolates, whereas M2 isolates were genetically very different.

The relationship between the distributions of the 21 genes for which the

Table 2 Association of M-genotypes with GAS disease and TSS

��

��

��

��

��

��

� ��

Meningitis 27 6 (22) 8 (30) 0 (0) 4 (15) 2 (7) 5b (19) 1 (4) 1 (4) 0 (0) 6 (22) Arthritis 41 8 (20) 7 (17) 0 (0) 6 (15) 3 (7) 2 (5) 1 (2) 6c (15) 4 (10) 12 (29) Necrotizing fasciitis 61 30 (49) 19 (31) 0 (0) 12 (20) 3 (5) 1 (2) 4 (7) 3 (5) 4 (7) 15 (25) Puerperal sepsis 41 3d (7) 9 (22) 5d (12) 1c (2) 3 (7) 2 (5) 0 (0) 1 (2) 5 (12) 15 (35)

Non-TSS 123 22 (13) 4 (2) 14 (8) 9 (5) 10 (6) 5 (4) 9 (5) 12 (7) 38 (22) TSS 47 21d (45) 1 (2) 9 (19) 2 (4) 0 (0) 1 (2) 2 (4) 1 (2) 10 (21)

• 92

Tabl

e 3

Dist

ribut

ion

of v

irule

nce

gene

s am

ong

pred

omin

ant M

- and

PFG

E ty

pes

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

� � ��

��

��

��

� ��

� ��

��

� ��

��

��

��

��

��

� ��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

� ��

��

� ��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

� ��

��

��

��

� ��

��

��

� ��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

� ��

��

��

��

��

� ��

��

��

��

� ��

��

��

��

��

��

��

��

� ��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

� ��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

� ��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

� ��

��

� ��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

� ��

��

��

��

��

��

��

��

��

��

��

� ��

��

��

��

��

��

��

� ��

��

��

��

� ��

��

��

��

��

��

��

��

��

� ��

��

��

��

� ��

��

��

��

��

��

� ��

� ��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

� ��

� ��

� ��

��

��

��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

93 •

Figure 1 PFGE patterns of SmaI-digested chromosomal DNA and their association with M-type and TSS. The numbers before the multiplication signs indicate the numbers of iso-lates of the indicated M-type. nt, nontypeable

�

��

�

��

�

�

�

��

�

�

�

�

� ��

�

�

�

�

�

��

��

�

�

�

�

�

��

�

�

�

�

��

��

��

�

��

�

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

• 94

isolates were tested and the genetic backgrounds of the isolates, as defined by M-genotyping and PFGE analysis, is shown in Table 3. M1 and M3 strains shared a number of striking similarities. They represented 39% of all isolates tested, and all of these isolates except for one M3 isolate carried an speA allele and were lacking the speC gene. cpa-1 and fbp-54 were found exclusively in all M1 strains. Furthermore, the smeZ gene was found in all M1 isolates, a feature shared only with all M4 strains. In M3 strains, the existence of the ssa allele stood out. The M2 group of strains showed a large degree of genetic variability in terms of both their PFGE types and their gene profiles. M4 strains were characterized by the absence of speG and the presence of ssa, smeZ, and fba. Size variations were observed for the fba gene, as well as for the prtf and sfb genes. M6 isolates were found to be positive for speC, prtf-1, and, in most cases, speA. In general, PFGE allowed a better distinction of the different gene profiles than M-genotyping, although some slight variations were observed among isolates indistinguishable by PFGE (i.e., pulsotypes 3A, 4A, 4B, 6A, 12A, 12B, 28A, and 28B) as well as identical patterns among different PFGE types (i.e., pulso-types 4B and 4C and pulsotypes 12A and 12B).

To uncover potential associations of virulence genes and types of GAS dis-ease, the rate of occurrence of these genes in isolates that had given rise to the different manifestations of invasive streptococcal disease was analyzed. Some toxin and matrix binding protein genes were nonuniformly distributed among the GAS isolates causing different types of disease (Table 4). However, most of these correlations reflect the M-type distributions among the various site-specific infections (Table 2) and the association of these M-types with certain genes (Table 3). For instance, within the group of isolates causing meningitis, speA and prtf-1 predominated, a representation of the genetic makeup of men-ingitis-specific genotype M6. In isolates that had given rise to GAS arthritis, speH and pfbp were predominant (M12), as was the case for speC, prtf-15, and

Table 4 Association of gene profile and type of invasive GAS disease

��

��

��

� � ��

��

��

95 •

sfb among isolates causing puerperal sepsis, in which M2 was exclusively found. The overrepresentation of the fba gene among the isolates causing puerperal sepsis, however, could not be attributed to an association of one M-type with puerperal sepsis. In this case, different strains shared one gene that gave rise to a similar clinical syndrome. Remarkably, no specific toxin genes or genes encod-ing matrix binding proteins were associated with necrotizing fasciitis.

The invasive isolates were scrutinized for the existence of a toxin gene pro-file associated with severe disease, as defined by TSS. The distribution of genes significantly associated with a severe clinical outcome is illustrated in Table 5. The toxin genes significantly overrepresented among isolates causing TSS were smeZ and speA, whereas the presence of speC appeared to be associated with a more favorable course of invasive GAS disease. This is again a reflection of the genetic makeup of those M-genotypes associated with the development of TSS (i.e., M1 and M3).

Discussion

The reemergence of severe invasive GAS infections provided impetus to scien-tists and epidemiologists to define the factors responsible for disease manifesta-tions and clinical outcomes. Traditionally, M1 and M3 serotypes that contain the speA and speB genes have been particularly implicated in the pathogenesis of invasive GAS disease (10, 26, 27, 37, 45, 52, 62, 73). More recently, the importance of other GAS virulence factors has been underscored in several studies (11, 47, 50, 51, 76). In this study we have focused on GAS strain char-acteristics and their correlation to clinical syndromes. To this end we analyzed all microbiologically defined cases of meningitis, arthritis, necrotizing fasciitis, and puerperal sepsis in a national surveillance study conducted in The Nether-

Table 5 Association of gene profile and development of TSS

��

��

��

��

• 96

lands between 1992 and 1996. Twenty-eight percent of these infections were complicated by the occurrence of TSS. To address the question of the intrinsic virulence in GAS, the strains were characterized for their M-types, PFGE types, and the presence of genes encoding toxins and matrix binding proteins.

In accordance with other studies (5, 15, 38), M1, M3, M12, and M28 iso-lates were the most frequently present. Interestingly, the different manifesta-tions of invasive GAS disease (meningitis, arthritis, necrotizing fasciitis, and puerperal sepsis) appeared to be associated with certain M-types (Table 2). M2 was found exclusively in puerperal sepsis cases, whereas M3 was a rare causa-tive agent of this disease. M6 was associated with meningitis, and M12 was associated with arthritis. M1 was distributed equally among cases of these four different manifestations of invasive GAS disease, while no particular M-type was associated with necrotizing fasciitis. A preferential distribution of different M-genotypes in skin and throat isolates has been well described (7, 8), but this is the first report of a nonrandom distribution of M-genotypes among cases of the different clinical manifestations of invasive GAS disease.

Among all invasive isolates, the development of TSS was caused primarily by M1 types. The overrepresentation of M1 isolates among isolates responsible for TSS implies an increased intrinsic virulence for isolates of this M-type. In the past, a large number of different studies that corroborate this notion have been published (13, 19, 20, 26, 30, 45, 60, 65, 71). A recent publication by John-son et al. (31) could not identify a virulent clone among isolates of serotypes M1, M3, and M28 but suggested that the prevalence of the “virulent” clones reflected their normal prevalence in the population. Nonetheless, those inves-tigators “still favor the hypothesis that GAS strains of M-types 1 and 3 have an increased association with invasive infections.” On the basis of an increased propensity of the M1 and M3 types among all invasive GAS isolates in our study to cause TSS, we agree with the notion that these strains have increased intrinsic virulence.

All strains with the same PFGE pattern belonged to the same M-type. As expected, PFGE proved to have a discriminatory power superior to that of M-genotyping (68). Clonality was observed among isolates of the M1, M3, M6 and M12 types (Figure 1 and Table 3). In contrast, a larger degree of heteroge-neity was found among M4, M9, M12, and M28 isolates; and no homology was observed among the M2 types. The M1 isolates constituted the largest

97 •

cluster of strains (n = 43). In order to study the strain characteristics contributing to aggressive clinical

behavior, the isolates were evaluated for the presence of different toxin genes and other presumptive virulence factors encoded by chromosomal as well as plasmidlocated genes. We showed that the patterns of the protein binding fac-tor and toxin genes were exclusively associated with particular M-types. Simi-larly, Natanson et al. (46) reported on the detection of the prtf protein in M6 and M28 strains and found that it was absent from M1, M3, and M18 isolates. Also, the presence of pfbp in M12 strains and its association with prtf-2 have been reported previously (29, 61). We noted size variations in the sfb, prtf, and fba genes. This has been reported to be the result of intragenic repeats (46, 72, 74). The size of the fba gene in M1 isolates was unique to this genotype.

Some of the observed associations between M-types and toxin genes have been reported by others. Murakami et al. (44) detected the ssa gene in the ma-jority of M2 and M4 strains. Furthermore, they detected speC, speG, and speH in all M12 isolates as well as the absence of the speG gene in M4 isolates. These findings, with the exception of the presence of ssa in M2 strains, are completely in accordance with those of our study. Descheemaeker et al. (17) described the exotoxin gene profile of their M1 genotype as speA, speB, and speF positive and speC and ssa negative. To this profile we would like to add the speG-, speJ-, and smeZ-positive and speH-negative genotype as well as the exclusive presence of cpa-1 and fbp-54. In M1, the exclusive combination of speA and smeZ, which both encode proteins known to be major immunoactive agents, might contrib-ute to increased intrinsic virulence. Although it is speculative, the link between M-types and the presence of virulence genes appears to be best conserved for the most virulent M-types, such as M1 and M3. Their specific set of virulence genes might render them more successful as pathogens.

Since we do not know if particular virulence factors are involved in different types of invasive GAS disease, we searched for a particular site-specific profile. Different repertoires of adhesins and the interactions of different adhesins with each other may determine tissue tropism and the pathogenicity of S. pyogenes (16). Therefore, isolates with differences in tropism (i.e., isolates causing men-ingitis, necrotizing fasciitis, arthritis, and puerperal sepsis) were differentiated as non-TSS and TSS isolates and were analyzed for the presence of microbial surface adhesins as well as superantigens. We sought to define a “pathogenic”

• 98

toxin gene profile that resulted in the development of severe complications such as TSS. Although the presence of some factors such as speA or smeZ, as opposed to speC, appears to be positively correlated with the development of severe com-plications, the development of severe complications in patients infected with strains with these genes is a reflection of the association of these genes with the M1 and M3 types. The same applies to genes found significantly more often in association with GAS strains responsible for infections at the different sites. Most of these associations mirror the differences in M-type distributions seen in different disease manifestations and the correlation of these M-types to various virulence genes. However, the overrepresentation of the fba gene among isolates causing puerperal sepsis could not be attributed to the association of one M-type with puerperal sepsis. This was the only gene shared by different strains that gave rise to a similar clinical syndrome.

In conclusion, the profiles of exotoxin and protein binding factor genes were closely associated with M-types. Each M-type was characterized by one or two dominant gene profiles that were exclusive for a given type, whether or not these profiles also included a variable number of less frequently occurring pat-terns. How can the linkage between M-types and gene profile be explained? Most genes analyzed in this study are located on bacteriophages. A phenotypic correlation between resistance to bacteriophage infection and M-protein sur-face expression has been described (12, 69). Thus, the nonrandom association between M-types and gene distribution that was observed could suggest a di-rect or indirect biological interaction between M-protein surface structures and bacteriophages. In our study, toxin and protein binding factor gene profiling revealed for each M-type a number of less frequent profiles, in addition to one or two dominant toxin profiles. Remarkably, all gene profiles were exclusive for their respective M-types. Although this illustrates the fundamental nature of the lines of division drawn by M-typing, the different genotypes were distin-guished at a higher resolution by PFGE.

Toxin profiles are highly conserved among the most virulent M-types (M1 and M3). These strains do not seem to easily incorporate or release any viru-lence genes, suggesting that an optimal composition for their pathogenetic po-tential has been achieved. In this respect, it is noteworthy that in a study in Japan (44) the genotype profiles of M1 isolates changed from the lack of speA in the early 1980s to a speA-positive genotype in the 1990s. Furthermore, most

99 •

associations between the clinical manifestations of invasive GAS disease and virulence genes could be attributed to the nonrandom distribution patterns of M-types among the isolates that cause these clinical manifestations. However, this reasoning could well be reversed: the M-type associated with the presence of genes encoding exotoxins and extracellular matrix binding proteins, which enable the streptococcus to selectively interact with certain extracellular matrix components of different tissues and organs, might explain the observed tropism of particular M-types for certain organs. This view is further supported by the observation that one matrix binding gene (fba) was more strongly correlated with a particular GAS disease manifestation than one particular M-type. None-theless, in this geographically and temporally narrowly defined study, a diversity of clones displaying specific genetic patterns proved to have the capacity to cause similar types of invasive disease. This underscores the complexity of the interplay between bacterial virulence factors, bacterial gene regulation, and host factors.

• 100

References

1. Abe, J., J. Forrester, T. Nakahara, J. A. Lafferty, B. L. Kotzin, and D. Y. Leung. 1991. Selective stimulation of human T cells with streptococcal erythrogenic toxins A and B. J. Immunol. 146:3747–3750.

2. Anonymous. 1993. Defining the group A streptococcal toxic shock syndrome. Rationale and consensus definition. JAMA 269:390–391.

3. Banks, D. J., S. B. Beres, and J. M. Musser. 2002. The fundamental contribution of phages to GAS evolution, genome diversification and strain emergence. Trends Microbiol. 10:515–521.

4. Basma, H., A. Norrby-Teglund, Y. Guedez, A. McGeer, D. E. Low, O. El Ahmedy, B. Schwartz, and M. Kotb. 1999. Risk factors in the pathogenesis of invasive group A strep-tococcal infections: role of protective humoral immunity. Infect. Immun. 67:1871–1877.

5. Beall, B., R. Facklam, T. Hoenes, and B. Schwartz. 1997. Survey of emm gene sequences and T-antigen types from systemic Streptococcus pyogenes infection isolates collected in San Francisco, California; Atlanta, Georgia; and Connecticut in 1994 and 1995. J. Clin. Microbiol. 35:1231–1235.

6. Beres, S. B., G. L. Sylva, K. D. Barbian, B. Lei, J. S. Hoff, N. D. Mammarella, M. Y. Liu, J. C. Smoot, S. F. Porcella, L. D. Parkins, D. S. Campbell, T. M. Smith, J. K. McCormick, D. Y. Leung, P. M. Schlievert, and J. M. Musser. 2002. Genome sequence of a serotype M3 strain of group A streptococcus: phage-encoded toxins, the high-virulence phenotype, and clone emergence. Proc. Natl. Acad. Sci. USA 99:10078–10083.

7. Bessen, D. E., T. R. Fiorentino, and S. K. Hollingshead. 1997. Molecular markers for throat and skin isolates of group A streptococci. Adv. Exp. Med. Biol. 418:537–543.

8. Bessen, D. E., M. W. Izzo, T. R. Fiorentino, R. M. Caringal, S. K. Hollingshead, and B. Beall. 1999. Genetic linkage of exotoxin alleles and emm gene markers for tissue tropism in group A streptococci. J. Infect. Dis. 179:627– 636.

9. Bisno, A. L. 1991. Group A streptococcal infections and acute rheumatic fever. N. Engl. J. Med. 325:783–793.

10. Carapetis, J., R. Robins-Browne, D. Martin, T. Shelby-James, and G. Hogg. 1995. In-creasing severity of invasive group A streptococcal disease in Australia: clinical and mo-lecular epidemiological features and identification of a new virulent M-nontypeable clone. Clin. Infect. Dis. 21:1220–1227.

11. Chatellier, S., N. Ihendyane, R. G. Kansal, F. Khambaty, H. Basma, A. Norrby-Teglund, D. E. Low, A. McGeer, and M. Kotb. 2000. Genetic relatedness and superantigen expres-sion in group A streptococcus serotype M1 isolates from patients with severe and nonse-vere invasive diseases. Infect. Immun. 68:3523–3534.

12. Cleary, P. P., and Z. Johnson. 1977. Possible dual function of M protein: resistance to bacteriophage A25 and resistance to phagocytosis by human leukocytes. Infect. Immun. 16:280–292.

13. Colman, G., A. Tanna, A. Efstratiou, and E. T. Gaworzewska. 1993. The serotypes of

101 •

Streptococcus pyogenes present in Britain during 1980–1990 and their association with disease. J. Med. Microbiol. 39:165–178.

14. Cunningham, M. W. 2000. Pathogenesis of group A streptococcal infections. Clin. Micro-biol. Rev. 13:470–511.

15. Davies, H. D., A. McGeer, B. Schwartz, K. Green, D. Cann, A. E. Simor, D. E. Low, et al. 1996. Invasive group A streptococcal infections in Ontario, Canada. N. Engl. J. Med. 335:547–554.

16. Delvecchio, A., B. J. Currie, J. D. McArthur, M. J. Walker, and K. S. Sriprakash. 2002. Streptococcus pyogenes prtFII, but not sfbI, sfbII or fbp54, is represented more frequently among invasive-disease isolates of tropical Australia. Epidemiol. Infect. 128:391–396.

17. Descheemaeker, P., F. Van Loock, M. Hauchecorne, P. Vandamme, and H. Goossens. 2000. Molecular characterisation of group A streptococci from invasive and non-invasive disease episodes in Belgium during 1993–1994. J. Med. Microbiol. 49:467–471.

18. Efstratiou, A. 2000. Group A streptococci in the 1990s. J. Antimicrob. Chemother. 45(Sup-pl.):3–12.

19. Ferrieri, P. 1991. Microbiological features of current virulent strains of group A strepto-cocci. Pediatr. Infect. Dis. J. 10:S20–S24.

20. Gaworzewska, E., and G. Colman. 1988. Changes in the pattern of infection caused by Streptococcus pyogenes. Epidemiol. Infect. 100:257–269.

21. Gerlach, D., B. Fleischer, M. Wagner, K. Schmidt, S. Vettermann, and W. Reichardt. 2000. Purification and biochemical characterization of a basic superantigen (SPEX/SMEZ3) from Streptococcus pyogenes. FEMS Microbiol. Lett. 188:153–163.

22. Goodfellow, A. M., M. Hibble, S. R. Talay, B. Kreikemeyer, B. J. Currie, K. S. Sriprakash, and G. S. Chhatwal. 2000. Distribution and antigenicity of fibronectin binding proteins (SfbI and SfbII) of Streptococcus pyogenes clinical isolates from the Northern Territory, Australia. J. Clin. Microbiol. 38: 389–392.

23. Goshorn, S. C., G. A. Bohach, and P. M. Schlievert. 1988. Cloning and characterization of the gene, speC, for pyrogenic exotoxin type C from Streptococcus pyogenes. Mol. Gen. Genet. 212:66–70.

24. Goshorn, S. C., and P. M. Schlievert. 1988. Nucleotide sequence of streptococcal pyro-genic exotoxin type C. Infect. Immun. 56:2518–2520.

25. Hanski, E., and M. Caparon. 1992. Protein F, a fibronectin-binding protein, is an ad-hesin of the group A streptococcus Streptococcus pyogenes. Proc. Natl. Acad. Sci. USA 89:6172–6176.

26. Hauser, A. R., D. L. Stevens, E. L. Kaplan, and P. M. Schlievert. 1991. Molecular analysis of pyrogenic exotoxins from Streptococcus pyogenes isolates associated with toxic shock-like syndrome. J. Clin. Microbiol. 29:1562– 1567.

27. Holm, S. E., A. Norrby, A. M. Bergholm, and M. Norgren. 1992. Aspects of pathogen-esis of serious group A streptococcal infections in Sweden, 1988– 1989. J. Infect. Dis. 166:31–37.

28. Iwasaki, M., H. Igarashi, Y. Hinuma, and T. Yutsudo. 1993. Cloning, characterization and

• 102

overexpression of a Streptococcus pyogenes gene encoding a new type of mitogenic factor. FEBS Lett. 331:187–192.

29. Jaffe, J., S. Natanson-Yaron, M. G. Caparon, and E. Hanski. 1996. Protein F2, a novel fi-bronectin-binding protein from Streptococcus pyogenes, possesses two binding domains. Mol. Microbiol. 21:373–384.

30. Johnson, D. R., D. L. Stevens, and E. L. Kaplan. 1992. Epidemiologic analysis of group A streptococcal serotypes associated with severe systemic infections, rheumatic fever, or uncomplicated pharyngitis. J. Infect. Dis. 166:374–382.

31. Johnson, D. R., J. T. Wotton, A. Shet, and E. L. Kaplan. 2002. A comparison of group A streptococci from invasive and uncomplicated infections: are virulent clones responsible for serious streptococcal infections? J. Infect. Dis. 185:1586–1595.

32. Johnson, L. P., J. J. L’Italien, and P. M. Schlievert. 1986. Streptococcal pyrogenic exotoxin type A (scarlet fever toxin) is related to Staphylococcus aureus enterotoxin B. Mol. Gen. Genet. 203:354–356.

33. Kamezawa, Y., T. Nakahara, S. Nakano, Y. Abe, J. Nozaki-Renard, and T. Isono. 1997. Streptococcal mitogenic exotoxin Z, a novel acidic superantigenic toxin produced by a T1 strain of Streptococcus pyogenes. Infect. Immun. 65:3828–3833.

34. Kapur, V., S. Topouzis, M. W. Majesky, L. L. Li, M. R. Hamrick, R. J. Hamill, J. M. Patti, and J. M. Musser. 1993. A conserved Streptococcus pyogenes extracellular cysteine pro-tease cleaves human fibronectin and degrades vitronectin. Microb. Pathog. 15:327–346.

35. Katerov, V., A. Andreev, C. Schalen, and A. A. Totolian. 1998. Protein F, a fibronectin-binding protein of Streptococcus pyogenes, also binds human fibrinogen: isolation of the protein and mapping of the binding region. Microbiology 144(Pt 1):119–126.

36. Kaufhold, A., A. Podbielski, G. Baumgarten, M. Blokpoel, J. Top, and L. Schouls. 1994. Rapid typing of group A streptococci by the use of DNA amplification and non-radioac-tive allele-specific oligonucleotide probes. FEMS Microbiol. Lett. 119:19–25.

37. Kaul, R., A. McGeer, D. E. Low, K. Green, B. Schwartz, et al. 1997. Population- based surveillance for group A streptococcal necrotizing fasciitis: clinical features, prognostic indicators, and microbiologic analysis of seventyseven cases. Am. J. Med. 103:18–24.

38. Kiska, D. L., B. Thiede, J. Caracciolo, M. Jordan, D. Johnson, E. L. Kaplan, R. P. Grun-inger, J. A. Lohr, P. H. Gilligan, and F. W. Denny, Jr. 1997. Invasive group A streptococcal infections in North Carolina: epidemiology, clinical features, and genetic and serotype analysis of causative organisms. J. Infect. Dis. 176:992–1000.

39. Kotb, M., A. Norrby-Teglund, A. McGeer, H. El Sherbini, M. T. Dorak, A. Khurshid, K. Green, J. Peeples, J. Wade, G. Thomson, B. Schwartz, and D. E. Low. 2002. An immuno-genetic and molecular basis for differences in outcomes of invasive group A streptococcal infections. Nat. Med. 8:1398– 1404.

40. Kreikemeyer, B., S. R. Talay, and G. S. Chhatwal. 1995. Characterization of a novel fi-bronectin-binding surface protein in group A streptococci. Mol. Microbiol. 17:137–145.

41. Mascini, E. M., M. Jansze, J. F. Schellekens, J. M. Musser, J. A. Faber, L. A. Verhoef-Ver-hage, L. Schouls, W. J. van Leeuwen, J. Verhoef, and H. van Dijk. 2000. Invasive group A

103 •

streptococcal disease in The Netherlands: evidence for a protective role of anti-exotoxin A antibodies. J. Infect. Dis. 181:631–638.

42. Molinari, G., S. R. Talay, P. Valentin-Weigand, M. Rohde, and G. S. Chhatwal. 1997. The fibronectin-binding protein of Streptococcus pyogenes, SfbI, is involved in the internaliza-tion of group A streptococci by epithelial cells. Infect. Immun. 65:1357–1363.

43. Mollick, J. A., G. G. Miller, J. M. Musser, R. G. Cook, D. Grossman, and R. R. Rich. 1993. A novel superantigen isolated from pathogenic strains of Streptococcus pyogenes with aminoterminal homology to staphylococcal enterotoxins B and C. J. Clin. Investig. 92:710–719.

44. Murakami, J., S. Kawabata, Y. Terao, K. Kikuchi, K. Totsuka, A. Tamaru, C. Katsukawa, K. Moriya, I. Nakagawa, I. Morisaki, and S. Hamada. 2002. Distribution of emm geno-types and superantigen genes of Streptococcus pyogenes isolated in Japan, 1994–9. Epide-miol. Infect. 128:397–404.

45. Musser, J. M., A. R. Hauser, M. H. Kim, P. M. Schlievert, K. Nelson, and R. K. Seland-er. 1991. Streptococcus pyogenes causing toxic-shock-like syndrome and other invasive diseases: clonal diversity and pyrogenic exotoxin expression. Proc. Natl. Acad. Sci. USA 88:2668–2672.

46. Natanson, S., S. Sela, A. E. Moses, J. M. Musser, M. G. Caparon, and E. Hanski. 1995. Distribution of fibronectin-binding proteins among group A streptococci of different M types. J. Infect. Dis. 171:871–878.

47. Newton, D., A. Norrby-Teglund, A. McGeer, D. E. Low, P. M. Schlievert, and M. Kotb. 1997. Novel superantigens from streptococcal toxic shock syndrome Streptococcus pyo-genes isolates. Adv. Exp. Med. Biol. 418:525–529.

48. Norrby-Teglund, A., S. Chatellier, D. E. Low, A. McGeer, K. Green, and M. Kotb. 2000. Host variation in cytokine responses to superantigens determine the severity of invasive group A streptococcal infection. Eur. J. Immunol. 30:3247–3255.

49. Norrby-Teglund, A., and M. Kotb. 2000. Host-microbe interactions in the pathogenesis of invasive group A streptococcal infections. J. Med. Microbiol. 49:849–852.

50. Norrby-Teglund, A., D. Newton, M. Kotb, S. E. Holm, and M. Norgren. 1994. Super-antigenic properties of the group A streptococcal exotoxin SpeF (MF). Infect. Immun. 62:5227–5233.

51. Norrby-Teglund, A., P. Thulin, B. S. Gan, M. Kotb, A. McGeer, J. Andersson, and D. E. Low. 2001. Evidence for superantigen involvement in severe group A streptococcal tissue infections. J. Infect. Dis. 184:853–860.

52. Nunthapisud, P., S. Sirilertpanrana, S. Reinprayoon, and A. Tanna. 1997. Detection of the erythrogenic toxin A, B, and C genes in group A streptococci isolated from clinical specimens. Adv. Exp. Med. Biol. 418:729–731.

53. O’Brien, K. L., B. Beall, N. L. Barrett, P. R. Cieslak, A. Reingold, M. M. Farley, R. Danila, E. R. Zell, R. Facklam, B. Schwartz, and A. Schuchat. 2002. Epidemiology of invasive group a streptococcus disease in the United States, 1995–1999. Clin. Infect. Dis. 35:268–276.

• 104

54. Okada, N., A. P. Pentland, P. Falk, and M. G. Caparon. 1994. M protein and protein F act as important determinants of cell-specific tropism of Streptococcus pyogenes in skin tissue. J. Clin. Investig. 94:965–977.

55. Podbielski, A., M. Woischnik, B. A. Leonard, and K. H. Schmidt. 1999. Characteriza-tion of nra, a global negative regulator gene in group A streptococci. Mol. Microbiol. 31:1051–1064.

56. Proft, T., S. L. Moffatt, C. J. Berkahn, and J. D. Fraser. 1999. Identification and character-ization of novel superantigens from Streptococcus pyogenes. J. Exp. Med. 189:89–102.

57. Proft, T., S. L. Moffatt, K. D. Weller, A. Paterson, D. Martin, and J. D. Fraser. 2000. The streptococcal superantigen SMEZ exhibits wide allelic variation, mosaic structure, and significant antigenic variation. J. Exp. Med. 191:1765–1776.

58. Rasmussen, M., A. Eden, and L. Bjorck. 2000. SclA, a novel collagen-like surface protein of Streptococcus pyogenes. Infect. Immun. 68:6370–6377.

59. Reda, K. B., V. Kapur, J. A. Mollick, J. G. Lamphear, J. M. Musser, and R. R. Rich. 1994. Molecular characterization and phylogenetic distribution of the streptococcal superanti-gen gene (ssa) from Streptococcus pyogenes. Infect. Immun. 62:1867–1874.

60. Reichardt, W., H. Muller-Alouf, J. E. Alouf, and W. Kohler. 1992. Erythrogenic toxins A, B and C: occurrence of the genes and exotoxin formation from clinical Streptococcus pyogenes strains associated with streptococcal toxic shock-like syndrome. FEMS Micro-biol. Lett. 79:313–322.

61. Rocha, C. L., and V. A. Fischetti. 1999. Identification and characterization of a nov-el fibronectin-binding protein on the surface of group A streptococci. Infect. Immun. 67:2720–2728.

62. Schellekens, J. F., L. M. Schouls, W. van Pelt, M. Esveld, and W. J. van Leeuwen. 1998. Group A streptococci: a change in virulence? Neth. J. Med. 52:209–217.

63. Schellekens, J. F., L. M. Schouls, A. van Silfhout, K. Elzenaar, H. Brunings, H. ten Broek, J. Top, and W. J. van Leeuwen. 1995. The resurgence of group A streptococcal disease. Ned. Tijdschr. Med. Microbiol. 4:78–83.

64. Schmitz, F. J., A. Beyer, E. Charpentier, B. H. Normark, M. Schade, A. C. Fluit, D. Haf-ner, and R. Novak. Toxin gene profile heterogeneity among endemic invasive European group A streptococcal isolates. J. Infect. Dis., in press.

65. Schwartz, B., R. R. Facklam, and R. F. Breiman. 1990. Changing epidemiology of group A streptococcal infection in the USA. Lancet 336:1167–1171.

66. Sela, S., A. Aviv, A. Tovi, I. Burstein, M. G. Caparon, and E. Hanski. 1993. Protein F: an adhesin of Streptococcus pyogenes binds fibronectin via two distinct domains. Mol. Microbiol. 10:1049–1055.

67. Sharkawy, A., D. E. Low, R. Saginur, D. Gregson, B. Schwartz, P. Jessamine, K. Green, and A. McGeer. 2002. Severe group A streptococcal soft-tissue infections in Ontario: 1992–1996. Clin. Infect. Dis. 34:454–460.

68. Single, L. A., and D. R. Martin. 1992. Clonal differences within M-types of the group A Streptococcus revealed by pulsed field gel electrophoresis. FEMS Microbiol. Lett. 70:85–89.

105 •

69. Spanier, J. G., and P. P. Cleary. 1980. Bacteriophage control of antiphagocytic determi-nants in group A streptococci. J. Exp. Med. 152:1393–1406.

70. Stevens, D. L. 1994. Invasive group A streptococcal infections: the past, present and fu-ture. Pediatr. Infect. Dis. J. 13:561–566.

71. Stromberg, A., V. Romanus, and L. G. Burman. 1991. Outbreak of group A streptococcal bacteremia in Sweden: an epidemiologic and clinical study. J. Infect. Dis. 164:595–598.

72. Talay, S. R., P. Valentin-Weigand, K. N. Timmis, and G. S. Chhatwal. 1994. Domain structure and conserved epitopes of Sfb protein, the fibronectinbinding adhesin of Strep-tococcus pyogenes. Mol. Microbiol. 13:531–539.

73. Talkington, D. F., B. Schwartz, C. M. Black, J. K. Todd, J. Elliott, R. F. Breiman, and R. R. Facklam. 1993. Association of phenotypic and genotypic characteristics of invasive Streptococcus pyogenes isolates with clinical components of streptococcal toxic shock syn-drome. Infect. Immun. 61:3369– 3374.

74. Terao, Y., S. Kawabata, E. Kunitomo, J. Murakami, I. Nakagawa, and S. Hamada. 2001. Fba, a novel fibronectin-binding protein from Streptococcus pyogenes, promotes bacterial entry into epithelial cells, and the fba gene is positively transcribed under the Mga regula-tor. Mol. Microbiol. 42:75–86.

75. Terao, Y., S. Kawabata, M. Nakata, I. Nakagawa, and S. Hamada. 2002. Molecular char-acterization of a novel fibronectin-binding protein of Streptococcus pyogenes strains iso-lated from toxic shock-like syndrome patients. J. Biol. Chem. 277:47428–47435.

76. Unnikrishnan, M., D. M. Altmann, T. Proft, F. Wahid, J. Cohen, J. D. Fraser, and S. Sriskandan. 2002. The bacterial superantigen streptococcal mitogenic exotoxin Z is the major immunoactive agent of Streptococcus pyogenes. J. Immunol. 169:2561–2569.

77. Valentin-Weigand, P., S. R. Talay, A. Kaufhold, K. N. Timmis, and G. S. Chhatwal. 1994. The fibronectin binding domain of the Sfb protein adhesin of Streptococcus pyogenes occurs in many group A streptococci and does not cross-react with heart myosin. Microb. Pathog. 17:111–120.

78. Whatmore, A. M. 2001. Streptococcus pyogenes sclB encodes a putative hypervariable surface protein with a collagen-like repetitive structure. Microbiology 147:419–429.

• 106

Chapter 6

Determination of the relationship between group A streptococcal genome content, M-type, and toxic shock syndrome by a mixed-genome microarray

B. Vlaminckx • F. Schuren • M. Caspers • A. Fluit • W. Wannet • L. Schouls • J. Verhoef • W. Jansen

• 108

109 •

Abstract

Group A streptococci (GAS) or S. pyogenes are associated with a remarkable variety of diseases, ranging from superficial infections to life threatening diseases such as toxic shock-like syndrome (TSS). GAS strains belonging to M-types M1 and M3 are associated with TSS. This study aims to obtain insight in the gene profiles underlying different M-types and disease manifestations. Genomic differences between 76 clinically well-characterized GAS strains collected in The Netherlands were examined using a mixed-genome microarray. Inter M-type genomic differences clearly outweighed intra M-type genome variation. Phages were major contributors to observed genome diversifications. We identified four novel genes, including two fibronectin-binding like proteins, which are highly specific to a subset of M-types and thus may contribute to M-type associated disease manifestations. M12 was characterized by the unique absence of the citrate lyase complex. Furthermore, six virulence factors including the complement inhibiting protein sic, exotoxin speA, iron III binding factor, collagen binding factor cpa and fibrinogen binding factor prtF2-like were unique to M1 and/ or M3 strains. These virulence factors may contribute to the potential of these strains to cause TSS. Finally, in contrast to M-type specific virulence profiles, we did not identify a common virulence profile among strains associated with TSS irrespective of their M-type.

• 110

Introduction

Group A streptococci (GAS) or S. pyogenes are human pathogens able to cause diseases ranging from superficial lesions to rapidly progressive and often fatal conditions such as necrotizing fasciitis and toxic shock-like syndrome (TSS). Despite persisting sensitivity to penicillin and other antibiotics, a resurgence of GAS infections has been reported since the mid 1980s (13,20,35,45).

GAS are genotyped on the basis of the emm gene encoding the M-protein. Although all M-types can give rise to severe manifestations of GAS disease, M1 and M3 types are overrepresented in the severest complication of invasive GAS disease, i.e. TSS. In our Dutch surveillance system on invasive GAS disease, M1 and M3 isolates were associated with TSS in 41% and 39% of cases respectively whereas other M-types gave rise to TSS in 23% of cases (54). Other specific M-types are associated with pharyngitis and acute rheumatic fever, or skin infections and glomerulonefritis (14,54). The M-type bias in GAS disease manifestations is not absolute: within a given M-type, strain-specific virulence characteristics also seem to contribute (26).

Current studies assessing the relationship between GAS gene profile and virulence are largely limited to a single M-type or virulence factors that are already known. Although microarray analyses do provide information on the whole genome, their design is restricted by the availability of the sequenced GAS genomes, which are predominantly from North-American strains (10,48). Our aim was to study GAS isolates from The Netherlands, where M12 is an important cause of S. pyogenes disease. This serotype has not been fully sequenced yet.

To circumvent the limitations mentioned above, we have developed a novel approach. In this approach random DNA fragments obtained from different well-characterized GAS strains are used to produce a mixed-genome DNA microarray. A well-documented GAS strain collection from The Netherlands was analyzed using this novel mixed-genome microarray (54,56). The method does not require prior genome sequence information, allows genomic screening of a large strain collection, and enables the identification of new genes.

First, the main genetic differences between the GAS strains of different M-types were determined. Subsequently, the genetic basis for the M-type bias in TSS was explored. Commonalities in gene profiles of ‘TSS’ M-types M1 and M3 were compared to other M-types. Finally, we sought for a common GAS gene profile that correlates with TSS, independent of the M-type. Understanding

111 •

the gene profiles underlying different M-types and TSS may deepen our understanding of GAS pathogenesis.

Materials and Methods

Bacterial strainsA total of 76 clinically well-documented isolates representing the predominant M-types M1 (n=15), M3 (n=14), M12 (n=14), M28 (n=14), M4 (n=5), M6 (n=5), M11 (n=5), and M89 (n=4) were selected from a Dutch surveillance program on GAS disease between 1992 and 2003 (54,56). These M-types are responsible for 74% of invasive GAS infections in the Netherlands (56). Per M-type, isolates associated with TSS and those associated with non-invasive superficial GAS disease were equally represented. For simplicity, we use the term ‘TSS strains’ to describe strains isolated from normally sterile sites of patients with a severe GAS infection that met the clinical criteria for TSS (1). Similarly, ‘superficial or non-TSS’ strains implicate strains isolated from patients with a non-severe superficial GAS infection. Per M-type the distribution of TSS versus superficial GAS disease was as follows: M1: eight versus seven isolates; M3: seven versus seven; M12: six versus eight; M28: seven versus seven; M4: three versus two; M6: two versus three; M11: three versus two; M89: two versus two cases. The M-genotype was determined by hybridization of the denatured emm amplicon with a panel of 26 emm type-specific probes in a reverse line blotting system (30). The M-genotype was confirmed with ‘conventional’ emm-typing for a subset of 16 strains (two random strains per M-type, data not shown) (7). Strains were grown on blood agar plates at 5% CO2 and 37°C overnight.

Typing by PFGEA subset of 25 representative strains was subjected to pulsed-field gel electrophoresis (PFGE) as previously described (55). Briefly, the bacteria were suspended and bacterial plugs were treated with lysis buffer for 48 h. Digestion of the bacterial DNA was performed with restriction endonuclease SmaI. Restriction fragments were separated by PFGE for 24 h, stained and visualized under UV light.

• 112

Microarray constructionThe mixed-genome array was constructed using genomic DNA from eight different strains of GAS representing M1 (n=3; one TSS associated isolate, one pharyngitis isolate, one isolate from 1950s), M3 (n=2; one TSS associated isolate, one pharyngitis isolate), M6 (TSS), M12 (TSS) and M28 (TSS). Per M-type, equimolar amounts of genomic DNA were mixed and 10 µg of DNA mix was ultrasonically sheared (Branson 250/450 Sonifier, 6 mm microtip, output intensity 1). Fragments of 1-1.9 kb were separated on gel and extracted (Quiagen). DNA fragments were cloned into pSMART-HC-Kan vectors (Clone-SMART, Lucigen). Ligation products were transformed into electro-competent E. coli cells (ElectroMAX DH10B, Invitrogen) and plated on kanamycin containing (30 µg/ml) tryptone yeast plates. A total of 3840 recombinant clones were arrayed into 96-well plates. Clone inserts were amplified by PCR using SMART primers (Lucigen) with 5’-C6 aminolinkers to facilitate crosslinking to the aldehyde coated glass slides. Sequencing of amplicons of randomly selected clones confirmed the presence of GAS specific fragments. Furthermore, probes specific for 34 known virulence genes were obtained by PCR with primers and templates listed in table 1. PCR products were purified and correct size was evaluated on agarose gels for all virulence probes and random DNA inserts. PCR products were dissolved in 3xSSC (SSC: 150 mM sodium chloride, 17 mM sodium citrate, pH 7.2) and transferred to 384-wells flat bottom plates (NUNC) for array printing. Amplicons were arrayed in a controlled atmosphere on CSS-100 silylated aldehyde glass slides (Telechem, USA), with quill pins (Telechem SMP3) in a SDDC 2 Eurogridder (ESI, Canada).

HybridizationGenomic DNA (0.5 µg) of the 76 strains to be tested was Cy5-dUTP (final concentration: 0.06 mM, Amersham) labeled using the BioPrime DNA labeling system (Invitrogen). Reference DNA (0.5 µg from mix used for micro-array construction) was Cy3-dUTP labeled. Arrays were prehybridized (1% BSA, 5x SSC and 0.1% SDS) for 45 minutes at 42°C and washed in miliQ water. Hybridization was performed overnight at 42°C.

Image analysis and data processingFor scanning, a ScanArray TM Express (Packard BioScience) was used. Quantification of hybridization signals was done using ImaGene version 5.6

113 •

Table 1 Oligonucleotide primers, reference strains and conditions used for virulence gene amplifications. Underlined genes were used for microarray validation by PCR. Two different primer combinations were used for smeZ, cpa, pfbp, prtf-1 and prtf-2 due to polymorphism in these genes in the published genomes

Gene Reference strain Annealing temp (°C) Amplicon size (bp) Primer sequence (5’-3’)

SuperantigensspeA 04.309 44 248 TAA GAA CCA AGA GAT GG ATT CTT GAG CAG TTA CC speC 04.304 42 584 GAT TTC TAC TTA TTT CAC C AAA TAT CTG ATC TAG TCC C speF 04.309 42 782 TAC TTG GAT CAA GAC G GTA ATT AAT GGT GTA GCC speG 04.309 42 155 AGA AAC TTA TTT GCC C TAG TAG CAA GGA AAA GG speH 01A091 42 416 AGA TTG GAT ATC ACA GG CTA TTC TCT CGT TAT TGG speJ 04.309 44 535 ATC TTT CAT GGG TAC G TTT CAT GTT TAT TGC C speK 04.306 50 718 TGT GTG TGT ATC GCT TG GAA CAT CAA AGT GAC TAA CTsmeZ 04.309 44 391 TAA CTC CTG AAA AGA GGC T CAT TGG TTC TTC TTG ATA AG smeZ 04.309 55 392 CCG AAA TGG ACG AAT ATG CAG CC CTC TAT ATC CTG TAT AGA AAA GAT Cssa 04.306 44 706 AAG AAT ACT CGT TGT AGC CTC ACT GTC TTA TTA TCG

(Biodiscovery) software. For all spots, signal intensity was measured for Cy5 (test strains) and Cy3 (reference) and local background signals were subtracted. Ratio calculations were normalized by correcting for the overall signal intensities in the respective Cy5 and Cy3 channels. Estimated Probability of Presence (EPP) for genomotyping analysis was done to define cut-off values (31). Briefly, log 2 transformed ratios are fitted to a normal distribution curve to define data sets representing absent and present genes as well as a weighed distribution for markers that cannot be certainly ascribed to one of these two groups. From a total of 3874 spotted DNA fragments (3840 recombinant clones and 34 virulence genes), 2704 spots (70%, including all virulence genes) yielded a significant Cy3 signal in all strains and were included for further analysis. The remaining 30% of spots induced a rather weak Cy3 signal and did not yield a significant signal in at least one of the strains. A differentiating biomarker was defined as a spot that was absent in at least one of the 76 analyzed strains after EPP transformation (i.e. no significant Cy5 signal).

• 114

Adhesinscpa 04.306 48 689 AAG ATT ATC CGT GGT ATG G GTT TAG AGT AGT CAC CTT CCcpa 04.306 55 521 TAA GGA AAT CGC AGA GCC ATA T GGT GAT CAA CTC ATG AAC TAC AGCcpa-1 04.309 48 1.000 TGT GAA CTT CCA TTT TTA TT AGA GTA GCA CAC GAT TTA AG fba 04.309 46 587 GGT GAT TCA ACA TCA GTT AC CGT TTT GTG ACT AAA AGA CT fbp-54 04.309 48 795 CTT CAG AAT CTG TTT CTT TG AGT TCA CAG GTT GTC TAT TG grab 04.309 50 486 AAC TAA TCT TCT TGG CAA TG TGA ACT TAC AAC TAA CAC AC Imb 04.309 50 610 GTT TAG TAA TGA TAG CAG GG GAA TCG TTT AGC CAG ATA AGpfbp 04.300 48 597 CTG AAT ATG CTG CTT TTA CT TTA TCC TTC GTT ACT TCT TG pfbp 04.300 55 732 TGA CGG CAA AGA GTT AGC TGG TG CTA ATA AGT GAG CAA GAA GTA ACCprtf-1 04.304 46 806 CCT TTG TAG ATT ATG CTC AC TTC TGT CTC AAC CAT ATT TC prtf-1 04.304 55 778 GAA CCT ACA TTG CCG CCA GTG AT AAC ACT AAC TTC GGA CGC GTA TCT prtf-2 04.300 47 767 AAA GCA ATT ATA TTA CTA ATG TTT TGT TTC ATA CAG GTC prtf-2 04.300 55 548 TGA CGG CAA AGA GTT AGC TGG TG CCA GAA TCC CCG TGC ATG Cprtf-15 04.298 49 399 TGG GAG TAC AGA AAC TTT TA ATC AGG TAC ATA TTC AGC AC sciA 04.309 47 267 TGA CAT CAA AGG AGA GAC AA CAC GAG CAC CAG CTT TAC sciB 04.309 46 592 TGA CAA ACA AAC AAA CTC ACT ATA AAC TGC AAA ATC CCA AA sfb 04.298 42 1.001 CAT ATC AGG CTT ATT GTT TT TTC TGT CTC AAC CAT ATT TC Hemolysins/ proteolytic enzymes sagA/B 04.309 50 415 TGG CAT AGA GGT GTT AGA ATT AGC AAC GGC AGA ATC scpA 04.309 49 490 AGT GGG TCA ATG ATA AGG GCT GTA AGA AGC AAC TG sda 04.306 50 456 ACA AAT CCC GAA CCA AG TAA ATC CAT CAG CGG TG sic 04.309 49 630 TGG AGT TGG TTT GTC TC ATC AAT ACC TCT ACC TGA G slo 04.309 50 497 ATC AAC ACT ACA CCA GTC GGT CAC CGA TTT ATC AAA C Immunoreactive antigens isp 04.309 50 864 CCT ACA CTC ACA ACC TG CCC ATA ACA TTT GAC TCT G isp2 04.309 49 590 TCC TAT TCA GTC GTT GTC TCA ATC CAC TTA TCC AGG Capsule hasA 04.309 49 738 AAT GAA GAT GCC GAG TC CAA CGA TGG GAT TAG AAA G Regulatory elements luxS 04.306 50 329 CCC TTA TGT TCG TCT TAT TTC TTC CCA CAG GAT TCA AG mga 04.309 50 583 ATG AAG GAA GTG GGA GG AAC AAG CGT GAA AGG TC nra 04.306 51 538 TAT CGG CTA CGG GAA TC ATG GGC TCT AAC AAT AAC C rofA 04.309 49 612 TTT CCC TGG TAG TCT GT GTC CAA ATC TCC TGC TG

Table 1 (continued)

115 •

Data analysisPrincipal Component Analysis (PCA) was used as an unsupervised multivariate method to reduce the multidimensional space of data to its principal components (PC) (34). PCA concentrates strongly correlating variables (biomarkers) that vary in a similar way in all experiments, in a new variable, a principle component. The PC computation, done with MATLAB software (Natick, Massachusetts, USA) is displayed as a 2D scatter plot where the position along the axes shows the PCA score of a strain.

Whereas PCA tries to provide a low dimensional summary of the data, partial least squares-discriminant analysis (PLS-DA) is a supervised multivariate classification method that can be applied to search for a set of biomarkers that distinguishes between two defined classes within the total data set. PLS-DA was used to extract a differential gene profile that could discriminate isolates associated with TSS from those that had given rise to superficial manifestations of GAS disease only. Once a PLS-DA model is calculated it can be validated by prediction of the class membership of samples not used for model construction (18).

Hierarchical clustering of differentiating biomarkers from all strains was done with TIGR software (42) (available at http://www.tigr.org/software/tm4) using complete linkage and Pearson correlation as distance matrix. PFGE results were imported into Bionumerics software (Applied Maths, Kortrijk, Belgium) and successive hierarchical clustering (complete linkage) was done using Pearson’s correlation. Comparison of the genetic resolution of PFGE and the mixed genome microarray was performed by visual comparison of the dendrograms. Biomarkers that were present or absent in one or two M-types (uniquely or commonly present or absent) were subjected to sequencing. To determine the function of a given sequence and integrate it into an appropriate pathway, ERGOTM bioinformatics was used (http://ergo.integratedgenomics.com/ERGO/) as well as BLAST searches in GenBank.

Results

Genetic clustering and differential genome content of Dutch GAS isolatesTo examine the most pronounced genetic differences among a collection of 76 GAS isolates belonging to M-types M1, M3, M12, M28, M4, M6, M11 and M89, this collection was hybridized on a mixed whole-genome microarray. The

• 116

resulting hybridization patterns were compared using PCA (Figure 1). From a total of 2704 biomarkers included for analysis, 366 differentiating biomarkers were found (14%). Of the 366 differentiating biomarkers, 18 biomarkers were unique to a single M-type. The PCA analysis clustered the different strains primarily according to their M-type. This indicates that inter M-type differences outweigh intra M-type genomic differences. Nevertheless, substantial genetic variation was observed within certain M-types such as M12 and M28. 2-D hierarchical clustering of the 366 differentiating markers with the 76 GAS strains is depicted in figure 2. In accordance with PCA analysis, many of the differentiating biomarkers segregate into separate (M-specific) groups.

Genetic dissimilarities between M-typesTo obtain insight in the genetic differences underlying different M-types, M-type specific biomarker profiles were determined. Hereto, biomarkers were identified that were uniquely present or absent in one or two M-types. Thus,

Figure 1 PCA scorings of 76 GAS strains along the first two principal component axes. Dif-ferent M-types are represented by different colours. Strains with highly overlapping PCA scores are shown within ovals, in which “n” denotes the number of strains

117 •

Figure 2 2-D Hierarchical clustering of 76 S. pyogenes strains. The dendrogram on the x-axis shows clustering of the strains. The dendrogram on the y-axis shows the clustering of the 366 differentiating biomarkers. Red represents presence and green absence of a biomarker. The bars on the right depict genes unique to one or two M-types (solid bars) as well as genes that are uniquely absent (empty bars) in one or two M-types (Table 2). Twenty-five representative strains, marked by arrows in the top of the figure were subjected to PFGE (Figure 3)

M12 M1 M11 M89 M28 M3 M4 M6

� � � � � � � � ��

��

��

��

��

��

��

��

��

��

��

��

• 118

Tabl

e 2

Biom

arke

rs u

niqu

ely

or c

omm

only

pre

sent

(+)

ver

sus

abse

nt (

-) in

the

M-ty

pes

incl

uded

in t

his

study

. Unk

now

n bi

omar

kers

w

ere

subj

ecte

d to

sequ

ence

ana

lysis

and

the

clos

est p

rote

in m

atch

was

iden

tified

usin

g ER

GO

softw

are

as w

ell a

s BLA

ST se

arch

es a

gain

st G

enBa

nk se

quen

ces.

Und

erlin

ed b

iom

arke

rs w

ere

used

for m

icro

arra

y va

lidat

ion

by P

CR

.

Pres

ent

mar

ker

M1

M3

M4

M6

M11

M12

M28

M89

func

tion

nam

eM

-pre

senc

e*co

ntig

(nuc

leot

ide)

begi

nen

dst

rand

rem

arks

1+

Fib

rone

ctin

-bin

ding

pro

tein

prtf2-like

SF3

70-M

1N

C_0

0273

711

6923

1190

64+

2+

Iron

(III

)-bin

ding

pro

tein

SPy1063

SF3

70-M

1N

C_0

0273

787

2055

8730

23+

Unk

now

nSPy1064

8731

2387

3467

+

O-a

cety

ltran

sfer

ase

(cel

l wal

l bio

synt

hesis

)SPy1065

8738

0587

4368

+

3+

Hyd

rola

se (H

AD

supe

rfam

ily)

SPy1066

SF3

70-M

1N

C_0

0273

787

4375

8750

73+

2*

4+

Com

plem

ent i

nhib

itor p

rote

insic

SF3

70-M

1N

C_0

0273

716

8359

816

8267

5-

5+

Pha

ge p

rote

inSpyM3_0723

pha

ge 3

15.1

N

C_0

0458

428

492

2925

9+

Pha

ge e

ndop

eptid

ase

SpyM3_0724

2925

931

307

+

6+

Pha

ge p

rote

inSpyM3_1451

pha

ge 3

15.6

N

C_0

0458

9 41

2642

93+

Pha

ge p

rote

inSpyM3_1450

4289

4459

+

Pha

ge p

rote

inSpyM3_1449

4493

4873

+

7+

Pha

ge p

rote

inSpyM3_1445

pha

ge 3

15.6

N

C_0

0458

9 56

1267

66+

8+

Neg

ativ

e tra

nscr

iptio

nal r

egul

ator

nra

M3

SSI-

1 N

C_0

0460

611

1314

1097

82-

9+

Col

lage

n-bi

ndin

g pr

otei

ncpa

M3

SSI-

1 N

C_0

0460

611

1745

1139

76+

10+

Pha

ge p

rote

in_

M6

MG

AS1

0394

N

C_0

0608

610

1671

310

1739

3+

11+

Unk

now

nSPs1552

M3

SSI-

1 N

C_0

0460

615

4797

415

4753

4-

mat

ch:1

18/6

13

12+

Pha

ge D

NA

/RN

A h

elic

ase

NP_607398.1

M18

MG

AS8

232

NC

_003

485

1078

848

1077

484

-

Com

mon

pre

sent

M1

M3

M4

M6

M11

M12

M28

M89

13+

+ P

yrog

enic

exo

toxi

nspeA

M3

SSI-

1 N

C_0

0460

658

3201

5824

49-

14+

+ P

hage

-ass

ocia

ted

cell

wal

l hyd

rola

seSPy1438

SF3

70-M

1 N

C_0

0273

7 11

9251

411

9131

2-

15+

+ U

nkno

wn

abiR

SF3

70-M

1 N

C_0

0273

7 49

621

5126

1+

16+

+ P

hage

pro

tein

SpyM3_1302

pha

ge 3

15.5

N

C_0

0458

836

180

3683

6+

mat

ch 3

37/9

54

17+

+ P

yrog

enic

exo

toxi

nspeJ

SF3

70-M

1 N

C_0

0273

7 36

2130

3614

35-

18+

+ P

hage

end

opep

tidas

eSpyM3_0724

pha

ge 3

15.1

N

C_0

0458

429

259

3130

7+

19+

+ H

yalu

rono

gluc

osam

inid

ase

hylP.1

pha

ge 3

15.1

N

C_0

0458

431

307

3231

1+

2*

20+

+ P

hage

infe

ctio

n pr

otei

nSpyM3_0726

pha

ge 3

15.1

N

C_0

0458

432

324

3422

5+

3*

21+

+ U

nkno

wn

SPs0590

M3

SSI-

1 N

C_0

0460

661

9250

6184

74-

SPs0591

6196

7261

9484

-

22+

+ T

rans

posa

se_

M6

MG

AS1

0394

N

C_0

0608

640

8121

4091

16+

23+

+ P

yrog

enic

exo

toxi

nspeH

M6

MG

AS1

0394

N

C_0

0608

674

4646

7450

82+

24+

+ F

ibro

nect

in-b

indi

ng p

rote

inprt f-15-l

ike

M6

MG

AS1

0394

N

C_0

0608

615

9255

1611

35+

mat

ch 1

233/

2801

119 •

Abs

ent

M1

M3

M4

M6

M11

M12

M28

M89

25-

Tra

nscr

iptio

nal r

egul

ator

Rof

ArofA

SF3

70-M

1 N

C_0

0273

711

6531

1150

41-

26-

DN

A p

olym

eras

e IV

dinP

SF3

70-M

1 N

C_0

0273

715

3258

315

3149

2-

27-

RN

AtRNA-Lys

SF3

70-M

1 N

C_0

0273

716

6372

516

6379

7+

Ace

tyltr

ansf

eras

eSPy1994

1663

936

1664

373

+

28-

Pyr

ogen

ic e

xoto

xin

speG

SF3

70-M

1 N

C_0

0273

719

0526

1912

27+

29-

Dau

noru

bici

n re

sista

nce

ATP

-bin

ding

pro

tein

drr

ASPy0518

SF3

70-M

1 N

C_0

0273

741

9436

4204

25+

Dau

noru

bici

n re

sista

nce

trans

mem

bran

e pr

otei

nSPy0519

4204

3042

1245

+

30-

Hya

luro

nan

synt

hase

hasA

M3

SSI-

1 N

C_0

0460

618

7435

518

7553

9+

31-

Pho

spho

hydr

olas

eSPs1835

M3

SSI-

1 N

C_0

0460

618

6211

018

6179

6-

32-

Sho

rt-ch

ain

fatty

aci

ds tr

ansp

orte

ratoE

SF3

70-M

1 N

C_0

0273

7 12

6461

1278

34+

33-

Ace

tate

CoA

-tran

sfer

ase

alph

a su

buni

tatoD.2

SF3

70-M

1 N

C_0

0273

7 13

0132

1307

88+

34-

Ser

ine

(thre

onin

e) d

ehyd

rata

se (l

antib

iotic

bio

synt

hesis

)salB

SF3

70-M

1 N

C_0

0273

7 15

9826

215

9664

0-

35-

Rib

osom

al sm

all s

ubun

it ps

eudo

urid

ine

synt

hase

ASPs0799

M3

SSI-

1 N

C_0

0460

678

8120

7888

33+

36-

Bio

tin c

arbo

xyl c

arrie

r pro

tein

of o

xalo

acet

ate

deca

rbox

ylas

eSP

y117

6 S

F370

-M1

NC

_002

737

9662

6896

6615

+

37-

Mg2

+/c

itrat

e co

mpl

ex se

cond

ary

trans

porte

rSPy1180

SF3

70-M

1 N

C_0

0273

7 97

1094

9696

91-

2*

38-

Citr

ate

lyas

e be

ta c

hain

citE

SF3

70-M

1 N

C_0

0273

7 97

3897

9747

81+

3*

39-

Citr

ate

lyas

e al

pha

chai

ncitF

SF3

70-M

1 N

C_0

0273

7 97

4787

9763

16+

3*

40-

Apo

-citr

ate

lyas

e ph

osph

orib

osyl

-dep

hosp

ho-C

oA tr

ansf

eras

ecitX

SF3

70-M

1 N

C_0

0273

7 97

6312

9768

87+

2*

41-

Oxa

loac

etat

e de

carb

oxyl

ase

alph

a ch

ain

oadA

SF3

70-M

1 N

C_0

0273

7 97

6890

9782

96+

2*

42-

no

feat

ures

kno

wn

_ M

3 SS

I-1

NC

_004

606

9196

1491

9730

43-

Sen

sor k

inas

e dp

iBSPy1107

SF3

70-M

1 N

C_0

0273

7 90

7372

9058

34-

Mal

ate-

sodi

um sy

mpo

rtmalP

9075

3390

8861

+

44-

NA

D-d

epen

dent

mal

ic e

nzym

eSPy1110

SF3

70-M

1 N

C_0

0273

7 90

8895

9100

58+

Com

mon

abs

ent

M1

M3

M4

M6

M11

M12

M28

M89

45-

- F

ibro

nect

in-b

indi

ng p

rote

insf b- l

ike

M6

MG

AS1

0394

N

C_0

0608

615

9255

1611

35+

mat

ch 4

30/6

90

46-

- F

ibro

nect

in-b

indi

ng p

rote

inp rtf -1-

like

M6

MG

AS1

0394

N

C_0

0608

615

9255

1611

35+

47-

- H

ypot

hetic

al c

ytos

olic

pro

tein

SPy1046

SF3

70-M

1 N

C_0

0273

7 85

4757

8588

60+

48-

- P

hage

pro

tein

SpyM3_1331

pha

ge 3

15.5

N

C_0

0458

813

230

1373

0+

2*

49-

- P

hage

enc

oded

tran

scrip

tiona

l reg

ulat

or, A

rpU

fam

ilySpyM3_1330

pha

ge 3

15.5

N

C_0

0458

814

184

1461

5+

2*

50-

- P

hage

pro

tein

SpyM3_1329

pha

ge 3

15.5

N

C_0

0458

815

252

1560

5+

51-

- H

ypot

hetic

al m

embr

ane

asso

ciat

ed p

rote

inSPy2174

SF3

70-M

1 N

C_0

0273

7 18

0692

618

0800

2+

Tabl

e 2

(con

tinue

d)

• 120

sixty-four biomarkers were identified (marked by filled and empty boxes in figure 2). These biomarkers were sequenced and their function identified using ERGO bioinformatics (Table 2). After function identification, biomarkers representing the same ORF were discarded and the remaining 51 biomarkers were further analyzed. Six of 51 markers (12%) spanned more than one open reading frame (ORF). Twelve of 51 (24%) markers were phage-related and 18 (35%) represented a (putative) virulence factor. Forty-seven of the biomarkers showed >95% homology to genes present in at least one of the seven completed GAS genome sequences (6,9,19,22,37,48,52). Four markers (11, 16, 24 and 45 in table 2) showed less than 65% homology to the closest matching gene in ERGO. These four biomarkers include one hypothetical

TSS (n (n=38) TSS (n (n=38)

M1 (n=15) 47 ± 1,5% 42 ± 10,6% 68 ± 0,9% 64 ± 6,3%

M3 (n=14) 43 ± 0% 43 ± 1,8% 84 ± 3,1% 82 ± 3,7%

M12 (n=14) 49 ± 4,5% 47 ± 1,5% 66 ± 5,8% 62 ± 2%

M28 (n=14) 57 ± 0% 56 ± 1,6% 54 ± 6,1% 52 ± 5%

M4 (n=5) 33 ± 3,1% 30 ± 4,3% 58 ± 1,2% 58 ± 1,8%

M6 (n=5) 41 ± 3,1% 46 ± 5% 62 ± 1,4% 63 ± 1,6%

M11 (n=5) 39 ± 0% 39 ± 0% 42 ± 0,4% 42 ± 0,2%

M89 (n=4) 48 ± 0% 46 ± 3,1% 43 ± 4,1% 42 ± 6,2%

total (n=76) 46 ± 7,7% 46 ± 7,2% 64 ± 13,2% 61 ± 12,3%

differentiating biomarkers (n=366)a,b

virulence factors (n

Table 3 Presence of differentiating biomarkers per M-type in TSS versus non-TSS associated strains.

aValues are percentages ± standard deviation.bAll 366 differentiating biomarkers are subdivided in known virulence factors (n=23) and other differentiating biomarkers (n=343).

121 •

protein, one phage protein and two putative fibronectin-binding proteins based on protein characteristics as analyzed by ERGO software. M3 and M6 shared three different biomarkers that encompass a 5 kb fragment with >95% homology to phage 315.1 sequences. M11 and M89 strains uniquely lacked three biomarkers covering a 2.4 kb genetic fragment with >95% homology to phage 315.5 sequences. A unique property of all M12 strains was the absence of a functional citrate lyase operon. Based on microarray data, the citE, citF and citX genes were absent in M12 strains. This finding was confirmed with citE, citF and citX specific PCRs.

Genetic dissimilarities between TSS and non-TSS associated M-typesM1 and M3 have been traditionally associated with severe manifestations of GAS disease (13,15,25,39,43,51). The presence of M1 and M3 specific biomarker

Figure 3 PFGE dendrogram. Twenty-five strains (marked in figure 2), representing all differ-ent M-types included in this study were subjected to PFGE. Numbers corresponds to number-ing in figure 2.

�

�

�

�

�

�

�

�

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

• 122

profiles may provide an explanation for this overrepresentation in severe invasive GAS disease. A set of ten biomarkers (with an attributable function by ERGO) was identified that uniquely belonged to either or both of these M-types. These biomarkers include three fragments representing phage proteins, the collagen and fibronectin binding proteins cpa and prtF2-like protein (11), a hydrolase of the HAD superfamily (29), streptococcal inhibitor of complement (sic, (19)), the negative regulator of virulence genes (nra, (40)), and streptococcal pyrogenic exotoxin A (speA, (27)). Finally, a biomarker was identified that comprised three gene fragments matching iron III binding factor, a hypothetical protein and O-acetyl transferase.

Common genetic profiles among TSS associated GAS strainsWe examined whether the gene content of TSS strains differed from that of non-TSS strains, excluding M-type related differences. Since TSS strains may have acquired additional (virulence) genes as compared to noninvasive strains, both groups were first quantitatively compared for each M-type. The average number of differential biomarkers and putative virulence factors present in each disease category did not differ significantly (Table 3). To examine possible qualitative differences in gene combinations present in each category, we applied multivariate biostatistics. A total of 52 strains representing 50% TSS and 50% non-TSS strains equally distributed among the different M-types was selected at random (the ‘training set’). These strains were used to construct a PLS-DA model to separate TSS isolates from non-TSS isolates. Although PLS-DA showed a marginal difference in gene pattern between TSS and non-TSS isolates in the training set, this gene pattern had no predictive value in categorizing the remaining 24 samples (the ‘test set’, data not shown). It was therefore concluded that PLS-DA failed to detect a ‘TSS specific’ gene profile.

Validation of microarray experimentsReproducibility of hybridization experiments was controlled using duplicate screening for one random representative of each of the eight M-types. All 366 differentiating biomarkers showed 100% reproducible patterns in all the duplicate experiments (data not shown). Twenty-five representative strains (marked by arrows in figure 2) were subjected to PFGE analysis (Figure 3). In agreement with the microarray PCA and Pearson correlation clustering, PFGE dendrogram patterns showed M-type specific clusters, and some genetic variation within certain M-types.

123 •

The presence of 19 virulence factors (Table 1) among the 76 GAS strains was determined by PCR (55). For five of these 19 genes, namely smeZ, cpa, pfbp, prtf-1 and prtf-2, polymorphic sequences are found in the different published GAS genomes. Two different primer combinations per gene (see table 1) were used to PCR amplify these five genes. Using the primers listed in table 1, PCR findings were completely in accordance with microarray data for all the strains and all the virulence factors (data not shown). Likewise, the distribution of 13 randomly chosen markers (ten of these genes are marked in table 2) was completely confirmed with gene specific PCRs in all the strains (data not shown).

Discussion

This study provides insight in the relationship between GAS genome profile, M-type and TSS. The strain collection consisted of 76 strains belonging to eight M-types that together cover 74% of invasive GAS disease in the Netherlands (56). Microarray data were highly reproducible and completely in line with PCR control experiments. Using this method, gene profiles of GAS strains were examined on three different levels. First, differentiating biomarkers among the whole strain collection were identified. Second, M-type specific gene profiles were determined. Third, the presence of a common gene profile associated with TSS was explored.

Differentiating biomarkers among the whole strain collection were identified and visualized in figure 2. PCA analysis showed a strong clonality for M1, M3, M6 and M11 and a larger degree of genetic diversity for M4, M12, M28 and M89. These findings are in agreement with earlier genotyping studies (12,16,17,32,36,46,50). As expected, genetic variation within the strain collection was largely attributable to phages or phage-like elements (5,9,37). The identification of phage elements commonly shared between different M-types is suggestive for horizontal transfer of these elements, as has been recently documented (4). The presence of 366 differential biomarkers among a total of 2704, suggests that roughly 86% of the GAS genome is conserved. This is in agreement with Banks et al., who observed an average genome conservation of 90.6% (5). Dutch M1 and M3 isolates are associated with the severest complication of invasive GAS disease, i.e. TSS (54). We therefore determined which genetic profiles were unique to these M-types. Ten different biomarkers were found to be uniquely present among M1 and M3 (Table 2). Of these ten

• 124

biomarkers, three biomarkers represented phage proteins uniquely present in M3 strains. This underscores the importance of phages and phage elements in the diversification and virulence characteristics of different M-types. The seven remaining biomarkers included six virulence factors. The GAS surface proteins Cpa and PrtF2-like are microbial surface components recognizing adhesive matrix molecules (MSCRAMMs), allowing the bacterium to selectively interact with host tissue (11). Iron III binding factor is described exclusively in the M1 genome. Mechanisms for iron acquisition are very common among bacterial pathogens and important for full virulence (57). In a recent publication an important role for iron binding in the modulation of superantigen expression in GAS was found (28). However, as an O-acetyl transferase gene co-resides on one biomarker fragment together with iron III binding factor, we cannot exclude a hitchhiker effect for this gene. GAS hydrolases are considered to be major virulence factors, playing a role in the release of bacterial surface proteins as well as in the degradation of host tissue (41). Sic has been described predominantly in M1, but also in M57 (3). This inhibitor of the membrane attack complex enhances bacterial survival during infection (23). Streptococcal pyrogenic exotoxins (Spe’s) have superantigenic properties (2,27) and the speA gene has been suggested to play a causative role in the development of severe GAS disease including TSS (44). Whereas these six are well-known virulence factors (21,24,28,38,40,41), it is unclear whether nra and O-acetyl transferase are associated with the pathogenic potential of M1 and M3 strains. Nra negatively regulates the expression of cpa (40) and other virulence factors. This factor is not specific for M3, as it has been described in other M-types (i.e M4, M18, M49) not included in this study (40). In addition to M1 and M3 specific virulence factors, we identified four novel biomarkers, including two fibronectin binding-like proteins, which were unique to one or two M-types out of the eight M-types included in the study. In addition, this is the first report showing the unique absence of the citrate lyase operon in M12 strains. Citrate lyase is a key enzyme that allows the microorganism to enter the citric acid cycle in the reductive mode. This metabolic “switch” facilitates survival of the pathogen during environmental transitions encountered in the infective process (49). Preliminary data indeed show that, compared to the other M-types in this study, M12 strains have a substantially reduced growth under nutritionally deprived and acidic conditions. Further research is required to establish whether the other identified biomarkers contribute to specific M-type characteristics and thereby to the M-type bias in GAS disease.

125 •

Finally, a PLS-DA model was used to explore a possible common gene-profile among TSS isolates, irrespective of their M-type. PLS-DA is the most common method used in comparative genomics and transcriptomics to assess the differences between two different groups within a large data set (18). For all M-types, TSS isolates were similar to non-TSS isolates both in the total number of differentiating biomarker genes and putative virulence factors present in the strains. In addition, PLS-DA analysis did not yield a specific, predictive set of biomarkers that distinguished TSS from non-TSS isolates. The negative PLS-DA results suggest that there is no common, differential gene pattern present among TSS isolates as compared to non-TSS isolates. This might be relevant to the hotly debated issue about prophylaxis to close contacts of cases with severe invasive GAS disease (47). Other factors such as differences in expression of virulence genes (8,53) or host related factors (33) might be more decisive in the outcome of a GAS infection.

In conclusion, Dutch GAS strains appear to be characterized by unique combinations of commonly shared genes and phage elements. These unique gene combinations contribute to the M-type characteristics and possibly the M-type bias in GAS disease. An example is the unique absence of the citrate lyase cluster in M12 strains, which may render these strains less fit under nutrient-deprived, hypoxic conditions. In addition, we identified four novel M-type specific genes, which would not have been identified with ‘conventional’ microarray strategies using previously sequenced fragments only. Furthermore, ten biomarkers including six virulence factors were unique to M1 and M3 strains in our collection, which may contribute to the extraordinary pathogenic potential of these M-types. Finally, we did not find indications for the presence of a common gene profile among strains associated with TSS.

• 126

References

1. 1993. Defining the group A streptococcal toxic shock syndrome. Rationale and consensus definition. The Working Group on Severe Streptococcal Infections. JAMA 269:390-391.

2. Abe, J., J. Forrester, T. Nakahara, J. A. Lafferty, B. L. Kotzin, and D. Y. Leung. 1991. Selective stimulation of human T cells with streptococcal erythrogenic toxins A and B. J. Immunol. 146:3747-3750.

3. Akesson, P., A. G. Sjoholm, and L. Bjorck. 1996. Protein SIC, a novel extracellular pro-tein of Streptococcus pyogenes interfering with complement function. J. Biol. Chem. 271:1081-1088.

4. Aziz, R. K., R. A. Edwards, W. W. Taylor, D. E. Low, A. McGeer, and M. Kotb. 2005. Mosaic prophages with horizontally acquired genes account for the emergence and diver-sification of the globally disseminated M1T1 clone of Streptococcus pyogenes. J. Bacteriol. 187:3311-3318.

5. Banks, D. J., S. B. Beres, and J. M. Musser. 2002. The fundamental contribution of phages to GAS evolution, genome diversification and strain emergence. Trends Microbiol. 10:515-521.

6. Banks, D. J., S. F. Porcella, K. D. Barbian, S. B. Beres, L. E. Philips, J. M. Voyich, F. R. DeLeo, J. M. Martin, G. A. Somerville, and J. M. Musser. 2004. Progress toward char-acterization of the group A Streptococcus metagenome: complete genome sequence of a macrolide-resistant serotype M6 strain. J. Infect. Dis. 190:727-738.

7. Beall, B., R. Facklam, and T. Thompson. 1996. Sequencing emm-specific PCR products for routine and accurate typing of group A streptococci. J. Clin. Microbiol. 34:953-958.

8. Beres, S. B., E. W. Richter, M. J. Nagiec, P. Sumby, S. F. Porcella, F. R. DeLeo, and J. M. Musser. 2006. Molecular genetic anatomy of inter- and intraserotype variation in the hu-man bacterial pathogen group A Streptococcus. Proc. Natl. Acad. Sci. U. S. A. 103:7059-7064.


10. Beres, S. B., G. L. Sylva, D. E. Sturdevant, C. N. Granville, M. Liu, S. M. Ricklefs, A. R. Whitney, L. D. Parkins, N. P. Hoe, G. J. Adams, D. E. Low, F. R. DeLeo, A. McGeer, and J. M. Musser. 2004. Genome-wide molecular dissection of serotype M3 group A Strepto-coccus strains causing two epidemics of invasive infections. Proc. Natl. Acad. Sci. U. S. A. 101:11833-11838.

11. Bessen, D. E. and A. Kalia. 2002. Genomic localization of a T serotype locus to a recom-binatorial zone encoding extracellular matrix-binding proteins in Streptococcus pyogenes. Infect. Immun. 70:1159-1167.

12. Cockerill, F. R., III, R. L. Thompson, J. M. Musser, P. M. Schlievert, J. Talbot, K. E. Hol-ley, W. S. Harmsen, D. M. Ilstrup, P. C. Kohner, M. H. Kim, B. Frankfort, J. M. Mana-

127 •

han, J. M. Steckelberg, F. Roberson, and W. R. Wilson. 1998. Molecular, serological, and clinical features of 16 consecutive cases of invasive streptococcal disease. Southeastern Minnesota Streptococcal Working Group. Clin. Infect. Dis. 26:1448-1458.

13. Cone, L. A., D. R. Woodard, P. M. Schlievert, and G. S. Tomory. 1987. Clinical and bacteriologic observations of a toxic shock-like syndrome due to Streptococcus pyogenes. N. Engl. J. Med. 317:146-149.

14. Cunningham, M. W. 2000. Pathogenesis of group A streptococcal infections. Clin. Micro-biol. Rev. 13:470-511.


16. Descheemaeker, P., F. Van Loock, M. Hauchecorne, P. Vandamme, and H. Goossens. 2000. Molecular characterisation of group A streptococci from invasive and non-invasive disease episodes in Belgium during 1993-1994. J. Med. Microbiol. 49:467-471.

17. Enright, M. C., B. G. Spratt, A. Kalia, J. H. Cross, and D. E. Bessen. 2001. Multilocus sequence typing of Streptococcus pyogenes and the relationships between emm type and clone. Infect. Immun. 69:2416-2427.

18. Eriksson, L., E. Johansson, N. Kettanah-Wold, and S. Wold. 1999. Introduction to Multi and Megavariate Data Analysis Using Project Methods (PCA and PLS-DA) (Umetrics, Malmo, Sweden).

19. Ferretti, J. J., W. M. McShan, D. Ajdic, D. J. Savic, G. Savic, K. Lyon, C. Primeaux, S. Sezate, A. N. Suvorov, S. Kenton, H. S. Lai, S. P. Lin, Y. Qian, H. G. Jia, F. Z. Najar, Q. Ren, H. Zhu, L. Song, J. White, X. Yuan, S. W. Clifton, B. A. Roe, and R. McLaughlin. 2001. Complete genome sequence of an M1 strain of Streptococcus pyogenes. Proc. Natl. Acad. Sci. U. S. A. 98:4658-4663.

20. Gaworzewska, E. and G. Colman. 1988. Changes in the pattern of infection caused by Streptococcus pyogenes. Epidemiol. Infect. 100:257-269.

21. Gorton, D., R. Norton, R. Layton, H. Smith, and N. Ketheesan. 2005. Presence of fi-bronectin-binding protein gene prtF2 in invasive group A streptococci in tropical Aus-tralia is associated with increased internalisation efficiency. Microbes. Infect. 7:421-426.


23. Hoe, N. P., R. M. Ireland, F. R. DeLeo, B. B. Gowen, D. W. Dorward, J. M. Voyich, M. Liu, E. H. Burns, Jr., D. M. Culnan, A. Bretscher, and J. M. Musser. 2002. Insight into the molecular basis of pathogen abundance: group A Streptococcus inhibitor of comple-ment inhibits bacterial adherence and internalization into human cells. Proc. Natl. Acad. Sci. U. S. A. 99:7646-7651.

24. Hoe, N. P., J. Vuopio-Varkila, M. Vaara, D. Grigsby, L. D. De, Y. X. Fu, S. J. Dou, X. Pan, K. Nakashima, and J. M. Musser. 2001. Distribution of streptococcal inhibitor of complement variants in pharyngitis and invasive isolates in an epidemic of serotype M1

• 128

group A Streptococcus infection. J. Infect. Dis. 183:633-639.25. Hoge, C. W., B. Schwartz, D. F. Talkington, R. F. Breiman, E. M. MacNeill, and S. J.

Englender. 1993. The changing epidemiology of invasive group A streptococcal infections and the emergence of streptococcal toxic shock-like syndrome. A retrospective popula-tion-based study. JAMA 269:384-389.

26. Johnson, D. R., D. L. Stevens, and E. L. Kaplan. 1992. Epidemiologic analysis of group A streptococcal serotypes associated with severe systemic infections, rheumatic fever, or uncomplicated pharyngitis. J. Infect. Dis. 166:374-382.

27. Johnson, L. P., J. J. L’Italien, and P. M. Schlievert. 1986. Streptococcal pyrogenic exotoxin type A (scarlet fever toxin) is related to Staphylococcus aureus enterotoxin B. Mol. Gen. Genet. 203:354-356.

28. Kansal, R. G., R. K. Aziz, and M. Kotb. 2005. Modulation of expression of superantigens by human transferrin and lactoferrin: a novel mechanism in host-Streptococcus interac-tions. J. Infect. Dis. 191:2121-2129.

29. Kapatral, V., I. Anderson, N. Ivanova, G. Reznik, T. Los, A. Lykidis, A. Bhattacharyya, A. Bartman, W. Gardner, G. Grechkin, L. Zhu, O. Vasieva, L. Chu, Y. Kogan, O. Chaga, E. Goltsman, A. Bernal, N. Larsen, M. D’Souza, T. Walunas, G. Pusch, R. Haselkorn, M. Fonstein, N. Kyrpides, and R. Overbeek. 2002. Genome sequence and analysis of the oral bacterium Fusobacterium nucleatum strain ATCC 25586. J. Bacteriol. 184:2005-2018.

30. Kaufhold, A., A. Podbielski, G. Baumgarten, M. Blokpoel, J. Top, and L. Schouls. 1994. Rapid typing of group A streptococci by the use of DNA amplification and non-radioac-tive allele-specific oligonucleotide probes. FEMS Microbiol. Lett. 119:19-25.

31. Kim, C. C., E. A. Joyce, K. Chan, and S. Falkow. 2002. Improved analytical methods for microarray-based genome-composition analysis. Genome Biol. 3:RESEARCH0065.



34. Krzanowski, W. J. 2000. Principles of multivariate analysis :a user’s perspective. Oxford University Press, Oxford.

35. Martin, G. S., D. M. Mannino, S. Eaton, and M. Moss. 2003. The epidemiology of sepsis in the United States from 1979 through 2000. N. Engl. J. Med. 348:1546-1554.

36. Musser, J. M., V. Kapur, J. Szeto, X. Pan, D. S. Swanson, and D. R. Martin. 1995. Genetic diversity and relationships among Streptococcus pyogenes strains expressing serotype M1 protein: recent intercontinental spread of a subclone causing episodes of invasive disease. Infect. Immun. 63:994-1003.

37. Nakagawa, I., K. Kurokawa, A. Yamashita, M. Nakata, Y. Tomiyasu, N. Okahashi, S. Kawabata, K. Yamazaki, T. Shiba, T. Yasunaga, H. Hayashi, M. Hattori, and S. Hamada.

129 •

2003. Genome sequence of an M3 strain of Streptococcus pyogenes reveals a large-scale genomic rearrangement in invasive strains and new insights into phage evolution. Genome Res. 13:1042-1055.

38. Norrby-Teglund, A. and M. Kotb. 2000. Host-microbe interactions in the pathogenesis of invasive group A streptococcal infections. J. Med. Microbiol. 49:849-852.

39. O’Brien, K. L., B. Beall, N. L. Barrett, P. R. Cieslak, A. Reingold, M. M. Farley, R. Danila, E. R. Zell, R. Facklam, B. Schwartz, and A. Schuchat. 2002. Epidemiology of invasive group a streptococcus disease in the United States, 1995-1999. Clin. Infect. Dis. 35:268-276.

40. Podbielski, A., M. Woischnik, B. A. Leonard, and K. H. Schmidt. 1999. Characterization of nra, a global negative regulator gene in group A streptococci. Mol. Microbiol. 31:1051-1064.

41. Rasmussen, M. and L. Bjorck. 2002. Proteolysis and its regulation at the surface of Strep-tococcus pyogenes. Mol. Microbiol. 43:537-544.

42. Saeed, A. I., V. Sharov, J. White, J. Li, W. Liang, N. Bhagabati, J. Braisted, M. Klapa, T. Currier, M. Thiagarajan, A. Sturn, M. Snuffin, A. Rezantsev, D. Popov, A. Ryltsov, E. Ko-stukovich, I. Borisovsky, Z. Liu, A. Vinsavich, V. Trush, and J. Quackenbush. 2003. TM4: a free, open-source system for microarray data management and analysis. Biotechniques 34:374-378.

43. Schellekens, J. F., L. M. Schouls, W. van Pelt, M. Esveld, and W. J. van Leeuwen. 1998. Group A streptococci: a change in virulence? Neth. J. Med. 52:209-217.

44. Schlievert, P. M. 1993. Role of superantigens in human disease. J. Infect. Dis. 167:997-1002.


46. Single, L. A. and D. R. Martin. 1992. Clonal differences within M-types of the group A Streptococcus revealed by pulsed field gel electrophoresis. FEMS Microbiol. Lett. 70:85-89.

47. Smith, A., T. L. Lamagni, I. Oliver, A. Efstratiou, R. C. George, and J. M. Stuart. 2005. Invasive group A streptococcal disease: should close contacts routinely receive antibiotic prophylaxis? Lancet Infect. Dis. 5:494-500.

48. Smoot, J. C., K. D. Barbian, J. J. Van Gompel, L. M. Smoot, M. S. Chaussee, G. L. Sylva, D. E. Sturdevant, S. M. Ricklefs, S. F. Porcella, L. D. Parkins, S. B. Beres, D. S. Campbell, T. M. Smith, Q. Zhang, V. Kapur, J. A. Daly, L. G. Veasy, and J. M. Musser. 2002. Genome sequence and comparative microarray analysis of serotype M18 group A Streptococcus strains associated with acute rheumatic fever outbreaks. Proc. Natl. Acad. Sci. U. S. A. 99:4668-4673.

49. Srinivasan, V. and H. J. Morowitz. 2006. Ancient genes in contemporary persistent mi-crobial pathogens. Biol. Bull. 210:1-9.

50. Stanley, J., M. Desai, J. Xerry, A. Tanna, A. Efstratiou, and R. George. 1996. High-resolu-tion genotyping elucidates the epidemiology of group A streptococcus outbreaks. J. Infect. Dis. 174:500-506.

• 130


52. Sumby, P., S. F. Porcella, A. G. Madrigal, K. D. Barbian, K. Virtaneva, S. M. Ricklefs, D. E. Sturdevant, M. R. Graham, J. Vuopio-Varkila, N. P. Hoe, and J. M. Musser. 2005. Evo-lutionary Origin and Emergence of a Highly Successful Clone of Serotype M1 Group A Streptococcus Involved Multiple Horizontal Gene Transfer Events. J. Infect. Dis. 192:771-782.

53. Sumby, P., A. R. Whitney, E. A. Graviss, F. R. DeLeo, and J. M. Musser. 2006. Genome-wide analysis of group a streptococci reveals a mutation that modulates global phenotype and disease specificity. PLoS. Pathog. 2:e5.

54. Vlaminckx, B., P. W. van, L. Schouls, S. A. van, C. Elzenaar, E. Mascini, J. Verhoef, and J. Schellekens. 2004. Epidemiological features of invasive and noninvasive group A strepto-coccal disease in the Netherlands, 1992-1996. Eur. J. Clin. Microbiol. Infect. Dis. 23:434-444.

55. Vlaminckx, B. J., E. M. Mascini, J. Schellekens, L. M. Schouls, A. Paauw, A. C. Fluit, R. Novak, J. Verhoef, and F. J. Schmitz. 2003. Site-specific manifestations of invasive group a streptococcal disease: type distribution and corresponding patterns of virulence determi-nants. J. Clin. Microbiol. 41:4941-4949.

56. Vlaminckx, B. J., P. W. van, L. M. Schouls, S. A. van, E. M. Mascini, C. P. Elzenaar, T. Fernandes, A. Bosman, and J. F. Schellekens. 2005. Long-term surveillance of invasive group A streptococcal disease in The Netherlands, 1994-2003. Clin. Microbiol. Infect. 11:226-231.

57. Wandersman, C. and I. Stojiljkovic. 2000. Bacterial heme sources: the role of heme, he-moprotein receptors and hemophores. Curr. Opin. Microbiol. 3:215-220.

Chapter 7

Dynamics in prophage content of M1 and M28 Streptococcus pyogenes isolates in the Netherlands from 1959 to 1996

B. Vlaminckx • F. Schuren • M. Caspers • M. Beitsma • W. Wannet • L. Schouls • J. Verhoef • W. Jansen

• 132

133 •

Introduction

Group A streptococcus (GAS) or S. pyogenes is associated with a wide range of different disease manifestations (6). Throughout the 19th century it was a cause of high endemic levels of erysipelas, puerperal sepsis and other manifestations of invasive bacterial disease. Until today, penicillin remains the drug of choice and with its introduction in the late 1940s, it was believed that GAS would be soon eliminated as a significant pathogen. However, a remarkable resurgence of invasive GAS disease, often complicated by the development of toxic shock-like syndrome (TSS), was noted in the late 1980s throughout the Western world (4,5,10,11).

GAS can be categorized on the basis of antigenic differences in the M-pro-tein, an important virulence factor that confers antiphagocytic properties to the bacterium. As the reappearance of invasive GAS disease involved multiple M-types, its resurgence cannot be solely explained by the expansion of one clone. Likewise, alterations in host susceptibility are unlikely to explain such a dramatic change in GAS epidemiology in a relatively short period of time. Several studies suggest a direct relation between the genetic content of GAS M-types and their clinical importance in terms of frequency and severity of disease (9,13,15,16). Thus, alterations in the bacterial genome composition over time may explain the resurgence of GAS disease in the Western world. To iden-tify the genomic changes underlying the resurgence of invasive GAS disease in the Netherlands, we compared the genetic composition of M1 and M28 GAS strains isolated before the mid-1980s with recent isolates using a mixed-whole genome microarray. M1 is highly virulent and represents the most predomi-nant M-type in The Netherlands (20). A recent M1 strain (MGAS 5005) as well as an older non-invasive M1 isolate (SF370) have been completely se-quenced (7,19). Serotype M28 is particularly associated with puerperal sepsis and the complete genome of a puerperal sepsis isolate has recently been pub-lished (MGAS6180) (8). This is the first microarray study on temporal genetic alterations in a clinically well-documented collection of GAS strains of different M-types, spanning a period of four decades. Recent GAS M1 and M28 strains were clearly enriched in prophages encoding superantigens or streptodornases. As these virulence factors are associated with invasive disease, their acquisition underlies -and possibly explains- the resurgence of invasive GAS disease in the Netherlands.

• 134

Materials and Methods

Bacterial strainsGAS strains were obtained from the Dutch National Institute of Public Health (Dutch acronym: RIVM). Recent GAS M1 and M28 isolates representative of invasive and non-invasive GAS disease in the 1990s were obtained from a national surveillance program conducted from 1992 to 1996 (20). Historical isolates were collected by the RIVM in the period 1959-1983. Per M-type and decade, half of isolates were obtained from normally sterile body sites, repre-senting invasive GAS disease, whereas the other half were obtained from su-perficial cultures. Twenty-two M1 isolates (nine historical and 13 obtained in 1992-1996), and 19 M28 isolates (five historical and 14 obtained in 1990s) were included in this study. Nine historical M1 isolates were obtained in 1959, 1960, 1966, 1968, 1971, 1982 (n=3), and 1983. Historical M28 isolates were collected in 1962, 1966, 1982, 1983 (n=2). The emm genotype was determined by sequencing the emm amplicon as described on the CDC website (www.cdc.gov/ncidod/biotech/strep/emmtypes.htm). Strains were grown on blood agar plates at 5% CO2 and 37°C overnight.

Microarray constructionWe have developed a mixed-genome microarray approach (see chapter 6). In short, random DNA fragments obtained from eight different well-documented GAS strains were used to produce a mixed-genome DNA microarray. Random DNA fragments of 1-1.9 kb from these strains were cloned into Escherichia coli. The inserted random DNA fragments were amplified by PCR and spot-ted on the microarray. Furthermore, probes specific for 34 known virulence genes (superantigens, adhesins, hemolysins/ proteolytic enzymes, immunoreac-tive antigens and regulatory elements) were obtained by PCR and spotted on the microarray. The microarray was used to analyze the genome of GAS isolates from The Netherlands by hybridization. The main genomic differences were identified by differentially hybridizing fragments on the microarray. A differen-tiating biomarker was defined as a spot that was absent in at least one of the 41 analyzed strains. The inserts in the plasmid DNA of the corresponding E. coli transformants were sequenced. This method does not require prior genome se-quence information, allows genomic screening of a large strain collection of dif-

135 •

ferent M-types, and enables the identification of new genes. Furthermore, most GAS genomes published thus far are from North-American origin, whereas the current approach provides a dedicated microarray to study well documented GAS strains from the Netherlands.

Data analysisHierarchical clustering of differentiating biomarkers from all strains was done with TIGR software (14) (available at http://www.tigr.org/software/tm4). Clus-ters of biomarkers that emerged or disappeared in one of the M-types in the course of time were sequenced. To determine the function of a given sequence ERGOTM bioinformatics was used (http://ergo.integratedgenomics.com/ERGO/) as well as BLAST searches in GenBank. For the sake of simplicity we use the terms ‘acquisition’ and ‘loss’ to describe genetic differences between historical and recent strains.

Validation of microarray experimentsValidation of microarray experiments is described in chapter 6. Briefly, all dif-ferentiating biomarkers showed 100% reproducible patterns in all duplicate experiments and the distribution of 13 randomly chosen biomarkers and 19 virulence factors was completely confirmed with PCRs throughout all isolates tested (data not shown). Additional PCRs were performed to distinguish be-tween different GAS prophages by amplification of phage specific virulence factors using primers and annealing temperatures as listed in table 1.

Bacteriophage induction GAS strains were grown overnight at 37ºC in Todd-Hewitt medium plus yeast extract (THY). The overnight GAS culture was diluted 1:100 with prewarmed THY and grown to an OD660 of 0.2. Mitomycin C was added to the cultures to final concentrations of 0.2 µg/ ml. Cultures were incubated for an additional 3 h at 37ºC.

Lytic assayBacteria with induced bacteriophages (see above) were centrifuged at 4 000 x g

• 136

for 15 minutes, and the supernatant was sterilized with a 0.22 µm-pore-size fil-ter (Millipore). GAS strains were grown to an OD660 of 1.0, taken up in THY soft agar and plated on THY agar plates. Plates were allowed to solidify and dry before 10 µl of phage mixture was spotted onto lawns. Plates were incubated overnight at 37ºC and lysis was scored as a visible clear area under the point of phage application. Experiments were performed in threefold.

Phage restriction analysisBacteria with induced bacteriophages (see above) were centrifuged at 4 000 x g for 15 minutes. The supernatant was centrifuged at 141,000 x g for 4 h at 10°C, and the pellet was suspended in 400 µl of phage suspension buffer (0.15 M NaCl, 10 mM Tris HCl (pH 7.5), 5 mM MgCl2, and 1 mM CaCl2). The phage particles were lysed with 0.5% SDS, 10 mM EDTA, and 400 µg of proteinase K/ ml for 1 h at 37°C. Phage DNA was extracted with phenol-chloroform-isoamyl alcohol (25:24:1), followed by chloroform-isoamyl alcohol (24:1) and precipitated with a 20% volume of 3M NaAc (pH 4.2) and a twofold volume of ethanol at -70°C overnight. Finally, it was washed with 70% ethanol, and suspended in 40 µl distilled H2O (2). Digestion of the phage DNA was per-formed with restriction endonuclease EcoR1 (Roche) and PinA1 (Gibco BRL) according to the guidelines of the manufacturer. Restriction endonucleases were selected on their ability to distinguish between the different GAS prophages (8) using Webcutter 2.0 software (available at: http://rna.lundberg.gu.se/cutter2/. Phage restriction fragments were analyzed on 1.0% agarose gels.

Table 1 Oligonucleotide primers used for detection of phage-specific virulence factors.

�� ′��′��

�� Φ�� Φ�� Φ�� Φ�� Φ�� Φ��

137 •

Results

We analyzed differences in the genetic composition of 41 GAS isolates belong-ing to M-types M1 and M28 isolated in the Netherlands from 1960s to 1990s using mixed-genome DNA microarray analysis. From a total of 2704 biomark-ers included for analysis, 203 biomarkers were differentially present (8%). Hi-erarchical clustering of the 203 differentiating markers is depicted in figure 1. Differential hybridisation patterns were largely related to the M-type. Nonethe-less, clear genetic differences were observed within each M-type when compar-ing older isolates with recent isolates: six clusters, composed of 114 biomarkers, were ‘acquired’ by either M1 or M28, or both (Figure 1). These biomarkers were sequenced and analysed using ERGO bioinformatics and BLAST searches in GenBank (Table 1). All biomarkers showed >95% homology to genes present in at least one of the completed GAS genome sequences. After their identi-fication, biomarkers representing the same open reading frame (ORF) were discarded and the remaining 38 biomarkers were further analyzed. All markers were phage-related. Using the BLAST algorithm, five of the six clusters each corresponded to one particular prophage in GenBank (Φ6180.1 (8), Φ6180.2 (8), Φ315.6 (3), Φ5005.1 (18) or Φ5005.3 (18)). One cluster (#4 in figure 1) matched with both Φ315.6 and Φ5005.3. This latter cluster consisted of biomarkers that span a 33 kb genetic fragment which is identical in these two prophages (1). M1 strains had comparable microarray hybridization profiles and the emm1.0 genotype. All recent M1 GAS strains possessed prophage clus-ter Φ5005.1 and Φ5005.3. In contrast, two strains from 1959 and 1960 lacked cluster Φ5005.3, whereas one of these strains also lacked cluster Φ5005.1. M28 strains showed a higher degree of variability. The 19 M28 isolates all belonged to emm28.0. However, within this cluster two different emm gene profiles were observed. Fourteen out of 19 isolates had sequences that were identical to the emm28.0 reference sequence (http://www.cdc.gov/ncidod/biotech/strep/emm-data.htm#emm28). The remaining four isolates showed seven basepair muta-tions in the emm sequence outside the emm typing region. These seven basepair mutations corresponded to three altered amino acids. These four M28 isolates were therefore designated emm28.0” and are indicated with an arrow in figure 1.

M28 strains from the 1990s possessed the Φ6180.2 cluster, the Φ315.6 cluster, or the Φ5005.3 and Φ5005.1 cluster. These clusters were all absent in

• 138

�

�

�

�

�

�

Table 2 Biomarkers obtained by M1 and M28 and their putative function. Large numbers correspond to clusters of biomarkers in figure 1. All markers were phage-related and represented one phage per cluster, except for cluster 4. This cluster consisted of biomarkers that span a 33 kb genetic fragment which is identical in prophages 315.6 and 5005.3.

139 •

M1 M28

1 (Φ��

(Φ5005.3)

2 (Φ6180.1)

3

5

6 (Φ5005.1)

4

(Φ315.6)

* * ** * *** * * ** * **- - - - - - - - - - - - - - - - - - - -* * * * * *

1 2 3 4

1

2

(

* * ** * *** * * ** * **

Figure 1 2-D Hierarchical clustering, illustrating acquisition of biomarkers over time by M-types M1 and M28. The GAS strains are ordered along the x-axis on M-type and year of isola-tion (except isolates from 1990s). Isolates associated with invasive GAS disease are marked with asterisks. The dendrogram on the y-axis shows the clustering of 203 differentiating biomarkers. Red represents presence and green absence of a biomarker. The upper bars illustrate year of isolation per M-type. Pink corresponds to isolates obtained in the 1950s, brown to the 1960s, yellow to the 1970s, green to the 1980s and blue to the 1990s. All 22 M1 isolates were emm1.0 and all 19 M28 isolates belonged to emm28.0. However, four of these had an identical change in nucleotide sequence of the emm gene and are marked by arrows. Prophages obtained from numbered isolates were subjected to DNA restriction profiling (Figure 2). Clusters of biomark-ers of interest are marked by brackets. These biomarkers were amplified, sequenced and their functions assigned (Table 2). Numbering of the clusters corresponds to the numbering in table 2.

• 140

strains isolated between the 1960s and 1980s. All three GAS emm28.0” from the 1990s contained the Φ6180.1 cluster, in contrast to a GAS emm28.0” from the 1950s. Interestingly, Φ6180.2 clusters were present in recent emm28.0 and emm28.0” strains, whereas both of these emm variants lacked this cluster in the 1960s (figure 1).

To confirm that the different prophage clusters correspond to the prophages described in GenBank, additional PCRs, phage lytic assays and phage restric-tion analyses were performed. For all strains, PCRs on prophage specific vir-ulence factors were completely in line with the predicted prophage content (Table 1). All strains containing the prophage 5005.1 cluster were positive on speA2, as determined by sequencing of the speA amplicon. Strains containing the Φ5005.3 cluster tested positive in an sdaD2 specific PCR, whereas Φ315.6 cluster containing strains tested positive on sdn. Likewise, only strains that con-tained the Φ6180.1 cluster were positive on speC, whereas cluster Φ6180.2 strains were positive in the speK and sla PCR.

All strains were challenged with mitomycin C to examine the mobility and host range of phages in the M1 and M28 strain collection. Mitomycin induced phages were tested on lysis of the different isolates in a bacterial lysis assay. Only from induced M1 strains that possessed the Φ5005.3 cluster and the one M28 isolate that had the same prophage composition, phages were collected that demonstrated lytic activity. These phages exclusively lysed M1 and M28 strains that lacked the Φ5005.3 and the Φ315.6 cluster (data not shown).

Phage DNA could be isolated from Φ5005.3 containing M1 and M28 strains, from the M1 isolate containing only the Φ5005.1 cluster, and from M28 isolates containing the Φ6180.2 cluster. Restriction profiling of these phages essentially confirmed their identification as based on microarray, PCRs and lysis experiments (Figure 2).

141 •

Figure 2 EcoR1 and PinA1 DNA restriction profiles of prophages isolated from M1 and M28 strains. For each strain/ prophage combination, a representative restriction profile is depicted in the figure. The predicted restriction patterns corresponded with the observed restriction pat-terns, except for a few fragments indicated with an asterisk. These aberrant fragments may be explained by mutations in the corresponding restriction site.a Numbering identifies isolates used for phage DNA restriction profiling in figure 1.b Predicted DNA fragment sizes for the restriction profile of prophage 5005.1, 5005.3 and 6108.2 sequences deposited in GenBank, using Webcutter 2.0

� � ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

�

�

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

�

�

�

• 142

Discussion

This study determined alterations in the genetic composition of M1 and M28 strains in the Netherlands over a period of more than four decades. Microarray analysis showed that recent M1 and M28 strains have an enriched genome as compared to the older isolates. All temporal genetic differences were prophage related. Prophage ”acquisition” was similar between GAS isolates associated with invasive and non-invasive disease (Figure 1).

Prophages were characterized by microarray analysis, PCR, lysis profile and phage restriction patterns. Each cluster of prophage biomarkers uniquely matched one prophage genome deposited in GenBank, namely phage 6180.1, 6180.2, 315.6, 5005.1 or 5005.3. Prophage specific PCRs were completely in accordance with the predicted prophage content of the strains. The induction and lysis profile of prophages was consistent with the presence or absence of specific prophage clusters in the different strains. The lysis experiments strongly suggest that prophage 5005.3 is mobile in both M1 and M28. Only from strains that harbored the 5005.3 prophage cluster, phages could be mobilized which only induced lysis of strains in which this cluster was absent. Finally, phage restriction profiles from phages that could be mobilized (Φ6180.2, Φ5005.1 and Φ5005.3) essentially confirmed their identification. Thus, the observed prophage clusters are identical, or at least very similar to the corresponding prophages in GenBank.

Our results strongly suggest that M1 and M28 have acquired several prophag-es, as summarized in figure 3. M1 isolates may have acquired prophages 5005.1 and 5005.3 on separate occasions around 1960. These findings are in line with the observed evolution of virulent M1 strains in the USA (19). In contrast to the findings of Sumby and coworkers (19), the acquisition of phages 5005.1 and 5005.3 may already have occurred in the 1950s. The oldest M1 strain in our collection, isolated in 1959, appears to represent an “intermediate” M1 iso-late that had already obtained Φ5005.1 but was still lacking Φ5005.3. Since all subsequent M1 isolates possessed both Φ5005.1 and Φ5005.3, these phages, or the M1 clone harbouring them, rapidly gained predominance among M1 isolates in both invasive as well as non-invasive M1 GAS disease.

The phage induction and lysis experiments suggest that Φ5005.3 remained highly mobile in all strains harboring this phage. Its predominance among M1

143 •

Figure 3 A hypothetical model of evolutionary events to accommodate the data presented in this paper. M1 isolates acquired prophages 5005.1, encoding SpeA2, and Φ5005.3, encod-ing SdaD2, on separate occasions around 1960. A similar event took place in emm28.0 in the 1990s. In these years, other emm28.0 isolates obtained an M3 prophage (Φ315.6) containing the sdn gene. M28 isolates that did not obtain these M1 and M3 prophages obtained the M28 phages 6180.1 and 6180.2 with corresponding virulence factors. These events involved both types of the allelic variants of emm28.0.

��

Φ��

Φ��

� ��

� ��

��

� ��

� ��

� ��

� ��

Φ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

� ��

Φ��

Φ��

� ��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

��

Φ

Φ

Φ

Φ

Φ

Φ

Φ

Φ

Φ

Φ

Φ

Φ

Φ

Φ

ΦΦ

• 144

strains in combination with its mobility, suggest that Φ5005.3 confers a se-lective advantage to the bacterium. Indeed, Φ5005.3 is likely to enhance the virulence of its host: Aziz and colleagues showed that phages 5005.1 (sphinx) and 5005.3 (phyramid) constituted the main difference between SF370, a se-quenced strain that is infrequently associated with invasive infections, and a pandemic virulent M1 strain (1).

Prophages 5005.1 and 5005.3 were also present in an emm28.0 isolate from the 1990s (Figure 1). To our knowledge, this is the first report on the acquisi-tion of M1 phages by another serotype. Analogous to M1, acquisition of these phages may have increased the virulence of M28 strains in the Netherlands. Given the early acquisition and predominance of prophages 5005.1 and 5005.3 in M1 isolates, these phages may have been introduced first in M1 and subse-quently been transferred to M28. Indeed, lysis experiments suggest that phage 5005.3 can be induced from M1 strains and is able to infect and lyse M28 iso-lates that lack this phage. Other emm28.0 isolates are likely to have obtained an M3 prophage (Φ315.6) containing the streptodornase-encoding sdn gene (2). The microarray hybridization patterns suggest that several prophages in M28 may be mutually exclusive, such as phages 315.6 and 5005.3. Lysis experiments were in agreement with this observation, since induced phages from Φ5005.3 containing strains showed lysis of all strains that lacked prophage 5005.3, ex-cept for the Φ315.6 containing M 28 strains. Blast analysis revealed that both prophages are highly homologous and share the same phage repressor module, which explains their mutual exclusiveness.

Phage 6180.2 may have been acquired by both allelic variants of emm28.0 (emm28.0 and emm28.0”), on separate occasions (Figure 3). Alternatively, phage 6180.2 may have been obtained by a single emm28.0 variant which, upon ac-quisition of the prophages, diverged in the two emm28.0 variants. Since both emm subtypes -without this phage- were already present among 1960s isolates and all emm28.0” strains had an identical mutation profile, the latter explana-tion is unlikely. Phages 6180.1 and 6180.2 have recently been described as part of a completed M28 genome sequence (8). However, this is the first report describing the acquisition of these phages in M28 over time, based on the no-tion that the oldest M28 isolate in this study did not contain either one of these phages and Φ6180.1 was only identified in M28 strains isolated in the 1990s.

Although the prophages themselves have been described in literature before,

145 •

the apparent acquisition of M1 and M3 phages by M28 and the early presence of 5005.1 and 5005.3 phages in M1 is novel. In contrast to M28 GAS strains, the genome composition of M1 strains remained relatively stable after the mid-1960s. The clonality of this latter M-type over time may reflect its continued success as an established virulent M-type.

Taken together, our results suggest that over a period of four decades, M1 and especially M28 strains underwent substantial genome enrichment due to phage acquisition. As depicted in figure 3, this may have happened at least eight times. All these prophages carried superantigens (speA2, speC, speK), a phospholipase (sla) or streptodornases (sdaD2, sdn). These virulence factors are clearly associated with invasive GAS disease (12,17). Given the fact that the prophage enrichment was similar in GAS strains associated with invasive dis-ease and non-invasive disease, our results suggest an overall increase in virulence of M1 and M28 strains over the last four decades. This increased overall viru-lence potential may have enhanced the frequency of invasive GAS disease and hence explain the reemergence of invasive GAS disease in the Netherlands.

• 146

References

1. Aziz, R. K., R. A. Edwards, W. W. Taylor, D. E. Low, A. McGeer, and M. Kotb. 2005. Mosaic prophages with horizontally acquired genes account for the emergence and diver-sification of the globally disseminated M1T1 clone of Streptococcus pyogenes. J. Bacteriol. 187:3311-3318.

2. Banks, D. J., B. Lei, and J. M. Musser. 2003. Prophage induction and expression of prophage-encoded virulence factors in group A Streptococcus serotype M3 strain MGAS315. Infect. Immun. 71:7079-7086.


4. Bisno, A. L. 1991. Group A streptococcal infections and acute rheumatic fever. N. Engl. J. Med. 325:783-793.

5. Carapetis, J., R. Robins-Browne, D. Martin, T. Shelby-James, and G. Hogg. 1995. In-creasing severity of invasive group A streptococcal disease in Australia: clinical and mo-lecular epidemiological features and identification of a new virulent M-nontypeable clone. Clin. Infect. Dis. 21:1220-1227.

6. Cunningham, M. W. 2000. Pathogenesis of group A streptococcal infections. Clin. Microbiol. Rev. 13:470-511.

7. Ferretti, J. J., W. M. McShan, D. Ajdic, D. J. Savic, G. Savic, K. Lyon, C. Primeaux, S. Sezate, A. N. Suvorov, S. Kenton, H. S. Lai, S. P. Lin, Y. Qian, H. G. Jia, F. Z. Najar, Q. Ren, H. Zhu, L. Song, J. White, X. Yuan, S. W. Clifton, B. A. Roe, and R. McLaughlin. 2001. Complete genome sequence of an M1 strain of Streptococcus pyogenes. Proc. Natl. Acad. Sci. U. S. A. 98:4658-4663.



10. Kaul, R., A. McGeer, D. E. Low, K. Green, and B. Schwartz. 1997. Population-based surveillance for group A streptococcal necrotizing fasciitis: Clinical features, prognostic indicators, and microbiologic analysis of seventy-seven cases. Ontario Group A Strepto-coccal Study. Am. J. Med. 103:18-24.

11. Kiska, D. L., B. Thiede, J. Caracciolo, M. Jordan, D. Johnson, E. L. Kaplan, R. P. Grun-inger, J. A. Lohr, P. H. Gilligan, and F. W. Denny, Jr. 1997. Invasive group A streptococcal

147 •

infections in North Carolina: epidemiology, clinical features, and genetic and serotype analysis of causative organisms. J. Infect. Dis. 176:992-1000.


13. Musser, J. M., A. R. Hauser, M. H. Kim, P. M. Schlievert, K. Nelson, and R. K. Selander. 1991. Streptococcus pyogenes causing toxic-shock-like syndrome and other invasive dis-eases: clonal diversity and pyrogenic exotoxin expression. Proc. Natl. Acad. Sci. U. S. A. 88:2668-2672.

14. Saeed, A. I., V. Sharov, J. White, J. Li, W. Liang, N. Bhagabati, J. Braisted, M. Klapa, T. Currier, M. Thiagarajan, A. Sturn, M. Snuffin, A. Rezantsev, D. Popov, A. Ryltsov, E. Ko-stukovich, I. Borisovsky, Z. Liu, A. Vinsavich, V. Trush, and J. Quackenbush. 2003. TM4: a free, open-source system for microarray data management and analysis. Biotechniques 34:374-378.



17. Sumby, P., K. D. Barbian, D. J. Gardner, A. R. Whitney, D. M. Welty, R. D. Long, J. R. Bailey, M. J. Parnell, N. P. Hoe, G. G. Adams, F. R. DeLeo, and J. M. Musser. 2005. Ex-tracellular deoxyribonuclease made by group A Streptococcus assists pathogenesis by en-hancing evasion of the innate immune response. Proc. Natl. Acad. Sci. U. S. A. 102:1679-1684.


19. Sumby, P., S. F. Porcella, A. G. Madrigal, K. D. Barbian, K. Virtaneva, S. M. Ricklefs, D. E. Sturdevant, M. R. Graham, J. Vuopio-Varkila, N. P. Hoe, and J. M. Musser. 2005. Evo-lutionary Origin and Emergence of a Highly Successful Clone of Serotype M1 Group A Streptococcus Involved Multiple Horizontal Gene Transfer Events. J. Infect. Dis. 192:771-782.

20. Vlaminckx, B., P. W. van, L. Schouls, S. A. van, C. Elzenaar, E. Mascini, J. Verhoef, and J. Schellekens. 2004. Epidemiological features of invasive and noninvasive group A strepto-coccal disease in the Netherlands, 1992-1996. Eur. J. Clin. Microbiol. Infect. Dis. 23:434-444.

• 148

Chapter 8Summary and general discussion

• 150

151 •

The aim of this thesis was to gain understanding in the epidemiology of invasive group A streptococcal (GAS) infections in the Netherlands and the role of un-derlying genetic profiles. To address these issues we obtained population-based data and investigated the relationship between the genetic profile of GAS and its clinical and epidemiological manifestations.

Epidemiology of invasive GAS disease

Surveillance systemsA worldwide increase in the incidence of severe invasive GAS infections was observed in the late 1980s and early 1990s. In response to these reports, a nationwide laboratory-based surveillance system for invasive GAS infections was conducted in The Netherlands by the National Institute of Public Health (RIVM) from March 1992 to December 2003. This surveillance system formed the basis for the studies conducted in this thesis. Initially, it relied on the vol-untary cooperation of all Dutch microbiological laboratories. Although there was no active stimulation to submit all isolates, the awareness among clinicians and microbiologists of a sudden resurgence of invasive GAS disease probably led to high degree of compliance. The system therefore has been described as “passionately passive”.

After the initiation of a formalized and active surveillance for regional public health laboratories (RPHLs) in May 1994, the reported incidence of all invasive GAS infections increased two-fold. Interestingly, the reported number of fatal and TSS cases did not increase (Chapter 3). Here, an interesting psychological mechanism may be at work: because of their impressive nature, compliance in submitting isolates associated with severe or fatal cases is high, regardless of the nature of surveillance. However, an actively stimulated surveillance system is more sensitive for detecting ‘milder’ (i.e., non-fatal and non-TSS) cases of invasive GAS disease as well. Therefore, an active surveillance system can be expected to provide more accurate estimates of the overall incidence of invasive GAS disease.

For proper detection of all cases, microbiological criteria alone do not suffice. Although the microbiological definition of invasive GAS disease (i.e. isolation

• 152

form a normally sterile body site) is very specific, it will lead to an underestima-tion of the true incidence: not all isolates that give rise to an invasive syndrome are obtained from a sterile site. This limits the sensitivity of the microbiologi-cal approach. In our study, only 66% of all cases of invasive GAS disease were detected on microbiological criteria alone. The remaining one-third could only be identified with the inclusion of clinical data (Chapter 3).

These two phenomena must be taken into account when comparing inci-dence figures from different countries: a significant proportion of the reported geographical variation in GAS incidence may be attributable to differences in surveillance systems.

The most accurate rate of invasive GAS disease in a country is likely to be obtained from an active surveillance system that includes clinical data. This sur-veillance approach, however, is costly. On the other hand, surveillance cannot be expected to be carried out passionately indefinitely without active (financial) stimulation. If the primary goal of invasive GAS disease surveillance is to moni-tor incidence trends and identify the most dominant strains, active surveillance should be done with a representative subset of microbiological laboratories without collecting clinical data (Chapter 4).

Clinical manifestationsAll invasive GAS cases collected in the surveillance program from March 1992 to December 1996 were clinically evaluated, as described in chapter 2. Half of all GAS infections consisted of soft tissue infections, the severest form of which is necrotizing fasciitis. No other clinical manifestation of GAS disease was found to carry a higher propensity to be complicated by toxic shock-like syndrome (TSS). Nonetheless, it did not have the highest case-fatality rate. Instead, patients suf-fering from either GAS sepsis without a focus or from GAS pneumonia were at highest risk to die. These diseases affected primarily the elderly. Comorbid condi-tions in the elderly facilitate progression to fatal outcome. However, comorbidity was not found to be associated with TSS development. Instead, a bacterial com-ponent was strongly associated with TSS: particularly M1 and M3 strains have the capacity to induce this state of severe, systemic dysregulation.

The age distribution of invasive GAS disease has a remarkable, trident-like, shape: it not only affects the very young and the elderly, but also the middle-aged (Chapter 2). This age distribution is unique when compared to other in-

153 •

vasive bacterial diseases or sepsis in general. These typically show the age distri-bution of the relatively immunocompromised host, with the burden of disease concentrated at the extremes of age only. Along with invasive GAS infections, meningococcal infections also have a secondary peak. This occurs in adoles-cents, generally attributed to crowding in this age group (18). Some studies have attributed the secondary peak of GAS infections in the middle-aged to pu-erperal sepsis. In The Netherlands this is not the case as the increased incidence was seen in both men and women. Furthermore, puerperal sepsis had a limited contribution to all cases of invasive GAS disease in women at fertile age.

Interestingly, invasive and noninvasive GAS diseases have different age dis-tribution patterns (Chapter 2). The high incidence of invasive GAS disease in the elderly is completely absent in noninvasive cases. Furthermore, the propor-tion of invasive to noninvasive cases in the middle-aged patients is higher than that in the younger age-groups. Thus, there appears to be a positive correlation between age and the chance of progressing to invasive disease.

By far the highest burden of GAS disease consists of noninvasive GAS phar-yngitis in children and adolescents (4). Such encounters with GAS result in the development of opsonizing antibodies. These antibodies are mostly acquired in the first decade of life and provide protection against invasive manifestations of GAS disease (1,23). This acquired immunity has been shown to wane in people between 21 and 40 years (23,29). Therefore, the increase in the proportion of invasive to noninvasive GAS disease with progression of age may be an epide-miological reflection of waning immunity.

Waning protection against invasive GAS disease in combination with in-creased household exposure to GAS through children may explain the rela-tively high incidence in the middle-aged group. Children are most likely to introduce and spread GAS into the household (6), and exposure to children with sore throats has been associated with an increased risk for invasive GAS disease in adults (9). Unfortunately, in our study household contacts and par-enthood status of patients were not registered. In the ongoing sentinel surveil-lance, we found the increased incidence of invasive disease in the middle-aged to be specific to that age group rather than a cohort phenomenon, as the peak did not shift during one decade (Chapter 4). In the elderly, failing immunity and crowding in elderly homes (9) have been associated with the development of invasive GAS disease. Here too, exposure to grandchildren may play a role.

• 154

M-type distributionsThe predominance of M1, M3, M6, M12 and M28 in invasive infections in the Netherlands (Chapters 2 and 4) is in agreement with other reports (7,15,20). It has been argued that invasive disease is associated with the prevalence rather than the virulence of a given M-type (17). Others have shown the contrary: M1 and M3 isolates were overrepresented among invasive isolates as compared to noninvasive isolates. This corroborates the notion of increased intrinsic inva-siveness of some M-types, as has been advocated by a large number of studies (10,11,13,26). Our strain collection does not permit a comparison between the prevalence of different M-types among invasive versus noninvasive GAS disease. Nevertheless, our data show a clear association between GAS M-type and invasive disease progression and outcome: M1 and M3 were independently associated with TSS and fatal outcome of invasive GAS disease in a multivariate analysis. This indicates enhanced virulence among these serotypes in invasive disease. Furthermore, Zwart and co-workers conducted a study during the same period in The Netherlands, showing a much lower relative frequency of M1 among pharyngitis isolates (4.5% in pharyngitis (39) versus 21% in invasive disease). Thus, we subscribe to the notion of increased virulence among some M-types, particularly M1 and M3.

Long-term surveillanceFormal surveillance by the RPHLs was instituted from May 1994 until December 2003. The incidence of invasive GAS disease was lowest in 1999 (2.0 ⁄ 100 000 individuals ⁄ year) and twice as high in 1995 and 1996 (Chapter 4). The highest incidence was similar to reports from the USA and Israel (25), while the lowest incidence was conform reports from Canada (7) and Sweden (34). Furthermore, in the Netherlands, a clear seasonal pattern, with a sharp increase in incidence in late winter and a low incidence in summer, was observed during the ten years of surveillance. Seasonal and annual fluctuations in the incidence of invasive GAS infections have been reported in other countries as well. Varicella infection is a well-documented risk factor in the development of invasive GAS disease among children (22). Recently, influenza A virus infection was shown to facilitate progression to invasive GAS disease in a mouse model (31). Both of these viral infections have a similar seasonality as invasive GAS disease. If the synchronicity in incidence is explained by GAS superinfection

155 •

upon respiratory viral infections, a corresponding seasonality in types of GAS disease affecting the respiratory tract (e.g. pneumonia, otitis, and sinusitis) would be expected. This was not seen in our study. Alternatively, increased viral infections in this season may have a more general systemic effect, enhancing the susceptibility to invasive GAS infections. Another possibility is that the same physiological conditions that enhance viral infections (temperature, dryness, crowding of people) may favour invasive GAS infections

In addition to annual fluctuations in invasive disease, the proportional con-tribution of M1 isolates as the cause of invasive GAS disease also varied per year, ranging between 10% and 40%. The same holds true for M3 isolates. The contribution of a given M-type to invasive GAS disease depends on its prevalence, its virulence and M-specific immunity in the population. Thus, fluctuations in M1 and M3 in invasive infections may correspond to fluctua-tions in their general prevalence in the community. An alternative explanation is that these fluctuations are a result of altered virulence traits and/ or host im-munity. The latter explanation, however, is unlikely as overall attack rates of GAS disease appear too low to provide a mechanism for annual fluctuations in population immunity. Although acquisition of certain virulence factors such as sic have been shown to be associated with epidemic waves of the corresponding clone (14), such changes do not occur annually. In fact, using a mixed genome microarray we observed that M1 and M3 were highly genetically stable over the 1990s. Therefore, annual M-type fluctuations in invasive GAS disease probably mirrored fluctuations in their general prevalence.

To obtain more insight in the complex relationship between GAS M-type, viru-lence profiles and disease manifestations, the second part of the thesis aimed at the identification of genetic profiles underlying particular tissue preference (“tropism”) and the development of severe complications such as TSS. Also, alterations in the genetic composition as a potential explanation for the GAS resurgence were investigated.

• 156

Genetic analysis of invasive GAS disease

The relation between M-type and tissue tropismThe notion of an ecological preference among particular M-types in noninvasive GAS disease is supported by decades of epidemiological research: some are over-represented among skin infections and impetigo cases whereas other M-types are mainly found in association with pharyngitis (19). For invasive GAS dis-ease, M-type dependent tissue tropism is less well defined. We, however, found clear associations between certain M-types and particular disease manifestations of invasive GAS infection: M28 was associated with gynaecological infections and puerperal sepsis, M6 with meningitis and M3 isolates were overrepresented among necrotizing fasciitis (Chapter 2). M-type associated genetic differences between GAS strains may account for M-type determined tissue tropism. Using multilocus sequence typing (MLST), Kalia et al. found that variations in house-keeping alleles are randomly distributed in M-types corresponding to phar-yngitis or skin infection. Thus, despite their anatomical and temporal (peak incidence for streptococcal pharyngitis and impetigo varies) separation, there was no evidence for core genome divergence between throat and skin isolates. Maintenance of an association between M-type and tissue preference in the face of underlying recombination suggests that the emm gene- or closely linked gene products may have a direct role in tissue tropism (19). Indeed, the M-protein itself is a strong determinant for tropism as it mediates interaction with keratinocytes or pharynx epithelium (30). In addition, we found a clear genetic linkage between M-type and adhesin repertoire (Chapter 5).

Establishment of invasive GAS infection starts with adherence to and in-ternalization into the skin or mucosal surfaces. GAS may express more than a dozen different microbial surface components recognizing adhesive matrix molecules (MSCRAMMs) to adhere to different host tissues. The availability of these adhesins in a particular strain is likely to affect its tissue specificity. Since GAS of these different M-types posses a unique adhesion factor repertoire, this adhesin profile may be responsible for the observed tissue tropism, possibly in combination with a direct role for the M-protein itself.

In this respect, recent findings of Green and colleagues are relevant. They sequenced the complete genome of an M28 isolate obtained from a puerperal sepsis case. Comparative microarray analysis revealed M28 specific genes en-

157 •

coding novel extracellular proteins, including putative adhesion factors, as well as a 37.4 kb foreign DNA element that is shared with group B streptococci (GBS). GBS are frequent colonizers of the female genital tract and the most abundant cause of neonatal sepsis (12). These genetic elements shared between GBS and M28 GAS may have helped to create a disease specialist variant of GAS. Thus in addition to M-protein and adhesin profile, also mobile elements may contribute to tissue tropism.

The relation between M-type and toxic shock-like syndromeUsing PCRs for well-recognized and putative adhesins and superantigens, we sought to define a “pathogenic” gene profile associated with the development of severe complications such as TSS (Chapter 5). The presence of some virulence factors such as speA or smeZ was positively correlated with the development of TSS. SpeA and smeZ are both powerful superantigens (13,26). Thus, these virulence factors may render GAS more virulent although we cannot exclude that these factors are simply overrepresented in TSS because of their association with M1 and M3.

Since GAS virulence appears to be determined by a complex interplay be-tween multiple virulence factors rather than a limited number of genes only, we constructed a mixed-genome microarray. The aim was to detect the main genetic differences between GAS strains on a genome level and to explore a pos-sible common gene-profile among TSS isolates.

Microarray analysis revealed that inter M-type differences clearly outweighed intra M-type differences on a whole genomic level. Differential biomarkers be-tween GAS strains of different M-types clustered in large segments. These clus-ters were usually present in more than one (but not all) M-types. As a result, the genetic profiles showed a ‘checker-board motive’ with an uneven distribution of gene clusters among the different M-types (Chapter 6). This motive illustrates that the genetic variation between M-types arises mainly through unique com-binations of common M-type gene clusters, rather than through the presence of M-type unique genes. Genetic variation between and within M-types was largely attributable to phages or phage-like elements. Although the GAS surface receptor for phage infection is not yet identified, a phenotypic correlation be-tween resistance to bacteriophage infection and M-protein has been described (5,33). Thus, the nonrandom association between M-types and gene distribu-

• 158

tion could suggest a direct or indirect biological interaction between M-protein surface structures and bacteriophages.

Using the mixed-genome microarray technique we could identify biomark-ers that were uniquely present among M1 and M3 strains. These biomarkers represented phage proteins as well as some well-known virulence factors. Fur-ther research is required to establish whether these biomarkers contribute to specific M1 and M3 characteristics, such as their TSS inducing potential and association with necrotizing fasciitis (M3).

For all M-types, including M1 and M3, the bulk of GAS disease consists of pharyngitis episodes (4). The questions as to why some superficial GAS infec-tions progress to a rapidly progressing severe and often fatal invasive disease remains enigmatic. A large variety of genetic techniques have been used to es-tablish differences between invasive and noninvasive GAS strains. Although particular M-types are more prone to progress to invasive disease, all predomi-nant M-types are capable of causing TSS and fatal GAS disease. We therefore constructed a PLS-DA model to explore a possible common gene-profile among TSS isolates, irrespective of their M-type. PLS-DA analysis did not yield a spe-cific, predictive set of biomarkers that distinguished TSS from non-TSS isolates. The negative PLS-DA results suggest that there is no common, differential gene pattern present among TSS isolates as compared to non-TSS isolates. Other factors such as differences in expression of virulence genes or host related factors (21) might be more decisive in the outcome of a GAS infection.

The importance of differences in expression of virulence genes is illustrated in a recent study by Sumby et al. (37). By assaying total gene expression (tran-scriptomics) the authors found that M1 GAS isolates from invasive and nonin-vasive pharyngeal disease had distinct virulence gene expression patterns during growth in standard laboratory media. Transition from pharynx to invasive gene expression patterns required a seven base-pair deletion of the two-component signal transduction system CovRS only. Similarly, whole genome sequence analysis of 12 contemporary M3 strains revealed a unique frameshift mutation among noninvasive M3 strains. This mutation truncates MtsR, a transcription-al regulator controlling expression iron-acquisition proteins (2).

These studies show the limitations of our microarray approach. Only tran-scriptome analysis in combination with whole-genome sequencing allowed the identification of the abovementioned mechanisms. Although we could not

159 •

identify a common genetic profile among TSS isolates, more subtle genetic dif-ferences such as found in CovRS and MtsR may dramatically influence expres-sion patterns and thereby the invasive potential of the strain. These observations support the idea that specific genetic characteristics underlie the invasiveness of a particular GAS strain. This is relevant to a currently hotly debated issue in invasive GAS disease, i.e. the value of secondary prophylaxis to close contacts. Davies and colleagues showed a higher secondary attack rate in close contacts of index cases with invasive GAS disease (7). In Canada, and also in The Neth-erlands, this study served as a rationale for antibiotic prophylaxis. On the other hand, vigilant observance but no prophylaxis is standard of care in most other countries in the world. Apart from the relatively low secondary attack rates another argument to this policy is the assumption that there are no genetic dif-ferences between invasive and noninvasive GAS isolates (32). New insights into this latter assumption provide a sound molecular basis to this discussion.

Genetic events underlying the resurgence of invasive GAS diseaseIn the late1980s, the frequency and severity of invasive GAS infections increased throughout the western world. In analogy to previous studies (3,28,36), we hy-pothesized that acquisition of novel genetic elements could have created unusu-ally successful clones. Mixed-genome DNA-DNA microarray analysis showed that contemporary M1 and M28 strains in The Netherlands have evolved from ancestral strains through a series of transduction events: the core genome of all strains was identical and variation in prophage content accounted for the dif-ferences in genome composition over time.

Although we could not include large numbers of historical GAS isolates, our data indicate that sequential acquisition of phages 5005.1 and 5005.3 in M1 occurred in the early 1960s and the resulting strain rapidly replaced all M1 strains lacking these prophages. Remarkably, this appears to have been the case for isolates associated with invasive as well as noninvasive GAS disease.

Despite their ubiquity in M1 over a long period of time, these phages could still be mobilized. Nevertheless, all recent M1 isolates harbored these phages. In general, the genetic equilibrium between bacterium and phage is a dynamic one: prophage acquisition confers an extra metabolic burden to the bacterium by the need to replicate extra DNA. Thus, the mobile and ubiquitous M1 phag-es have to encode functions advantageous to the bacterium. Prophages 5005.1

• 160

and 5005.3 carry the speA2 and sdaD2 genes, which encode a potent super-antigen and a DNase, respectively. SpeA is a well-characterized superantigen that plays a pivotal role in the pathogenesis of TSS (24) and sdaD2 has recently been established as the most important DNase in M1 that contributes to GAS virulence in a mouse infection model (35).

In M28 isolates, the picture appears to be more diverse: different phages, some mutually exclusive, have been acquired by different M28 strains over the last decades. Interestingly, we could identify one M28 isolate that has acquired the abovementioned “M1” phages 5005.1 and 5005.3. The identification of these phages in another M-type is a novel finding. Similar to their predom-inance among M1 isolates, these prophages may become ubiquitous among M28 strains. Analysis of recent isolates may reveal whether this is the case. In the 1990s however, no single phage, or phage combination, was predominant in M28 strains. Other phages acquired by some M28 strains included Φ6180.1 and Φ6180.2, recently described in the completed M28 genome sequence (12), as well as Φ315.6. Virulence factors encoded by these GAS prophages include the superantigens SpeC (Φ6180.1) and SpeK (Φ6180.2), a phospholipase Sla (Φ6180.1) and another DNase (Spd, Φ315.6). The ability, offered by these phage combinations, to produce multiple antigenically distinct superantigens and DNases could preserve the capacity of some of these factors to function when the bacterium infects a host with enzyme-inhibiting antibodies to some of these antigenic variants. Alternatively, the production of multiple DNases and superantigens with different specificities or characteristics could provide a survival advantage by enhancing the range of conditions across which the en-zymatic activity functions. Thus the apparent redundancy in virulence factors offered by the different phage combinations could have created disease special-ists, conferring selective advantages to the M28 bacterium in different hosts and types of infection.

Acquisition of virulence factors is likely to have contributed to an overall increase in GAS virulence over time. Since acquisition of prophages was seen to a similar extent in isolates associated with invasive and noninvasive manifesta-tions of GAS disease, we hypothesize that the overall proportion of streptococci associated with invasive disease has increased over time.

Nonetheless, acquisition of virulence factors does not necessarily equal a direct increase in invasiveness. Expression of newly acquired virulence traits has

161 •

to be well-balanced: unbridled superantigen activity with massive immune ac-tivation, and possibly death to the host is not advantageous to the bacterium or phage. Thus, acquisition of new virulence genes has to be paralleled by a well-orchestrated expression. In addition, prophage encoded factors may mainly contribute to colonization and persistence capacity. Successful colonizers may simply have a better chance to disseminate and cause (invasive) infection.

Finally, some phage encoded putative superantigens may also have other properties. For instance, SpeC has a high degree of structural similarity to a well-characterized protein in S. aureus that specifically impairs the response of neutrophils and monocytes to chemoattractants (chemotaxis inhibitory protein of S. aureus, CHIPS) (8). Based on its homology to CHIPS, SpeC may have a similar immunomodulating effect in GAS infection (J.A. van Strijp, personal communication). Ninety percent of human nasal S. aureus isolates carry the CHIPS encoding prophage. This supports the notion of phage encoded factors contributing to colonization and persistence rather than to virulence alone.

Similar to our findings, Sumby et al. showed that the modern M1 clone has evolved from an ancestral strain through a complex series of gene transfer events. These events included transduction by prophages encoding SpeA and DNase as well as a reciprocal recombination of a 36kb segment probably ob-tained from an M12 strain (36). Beres and coworkers have described the evolu-tion of a virulent clone of serotype M3 as the acquisition of prophage encoded virulence factors at different time-points (3). Based on the temporal association between these gene transfer events and the reemergence of severe GAS disease, a direct causal relationship between phage acquisition and the establishment of invasive GAS disease was hypothesized.

Similar to the situation in North America, invasive GAS disease did not re-emerge in The Netherlands until the late 1980s. However, the observation that phages rapidly gained predominance in the most important M-type in both invasive as well as noninvasive M1 isolates in a period preceding the resurgence of invasive GAS disease by at least 20 years, casts doubt on phage acquisition as the sole explanation for severe invasive GAS disease resurgence in the late 1980s in The Netherlands (3,27). In this respect, the abovementioned findings by Sumby and Beres may be relevant (2,38): identical GAS isolates require al-terations in different transcriptional regulators for the transition to the invasive phenotype. Apart from the (phage determined) genetic make-up for invasive-

• 162

ness, the ability of GAS to remodel the expression of these virulence traits ap-pears essential in the establishment of an invasive phenotype.

Nonetheless, the key role for prophage encoded virulence factors in GAS dis-ease is evident. Our finding that phages can be shared among different M-types raises the possibility of horizontal spread of important virulence factors across different M-types with corresponding effects on GAS virulence and epidemiol-ogy. These phenomena would go undetected in conventional GAS surveillance systems and questions the validity of the current, M-based, classification sys-tem. Therefore, the integration of genomic tools to chart phage epidemiology would be a useful addendum to surveillance systems of invasive GAS disease.

163 •

References

1. Basma, H., A. Norrby-Teglund, Y. Guedez, A. McGeer, D. E. Low, O. El Ahmedy, B. Schwartz, and M. Kotb. 1999. Risk factors in the pathogenesis of invasive group A strep-tococcal infections: role of protective humoral immunity. Infect. Immun. 67:1871-1877.

2. Beres, S. B., E. W. Richter, M. J. Nagiec, P. Sumby, S. F. Porcella, F. R. DeLeo, and J. M. Musser. 2006. Molecular genetic anatomy of inter- and intraserotype variation in the hu-man bacterial pathogen group A Streptococcus. Proc. Natl. Acad. Sci. U. S. A. 103:7059-7064.


4. Bisno, A. L. 1996. Acute pharyngitis: etiology and diagnosis. Pediatrics 97:949-954.5. Cleary, P. P. and Z. Johnson. 1977. Possible dual function of M protein: resistance to

bacteriophage A25 and resistance to phagocytosis by human leukocytes. Infect. Immun. 16:280-292.

6. Cockerill, F. R., III, K. L. MacDonald, R. L. Thompson, F. Roberson, P. C. Kohner, J. Besser-Wiek, J. M. Manahan, J. M. Musser, P. M. Schlievert, J. Talbot, B. Frankfort, J. M. Steckelberg, W. R. Wilson, and M. T. Osterholm. 1997. An outbreak of invasive group A streptococcal disease associated with high carriage rates of the invasive clone among school-aged children. JAMA 277:38-43.


8. de Haas, C. J., K. E. Veldkamp, A. Peschel, F. Weerkamp, W. J. Van Wamel, E. C. Heezius, M. J. Poppelier, K. P. Van Kessel, and J. A. van Strijp. 2004. Chemotaxis inhibitory pro-tein of Staphylococcus aureus, a bacterial antiinflammatory agent. J Exp. Med 199:687-695.

9. Factor, S. H., O. S. Levine, L. H. Harrison, M. M. Farley, A. McGeer, T. Skoff, C. Wright, B. Schwartz, and A. Schuchat. 2005. Risk factors for pediatric invasive group A strepto-coccal disease. Emerg. Infect. Dis. 11:1062-1066.

10. Ferrieri, P. 1991. Microbiological features of current virulent strains of group A strepto-cocci. Pediatr. Infect. Dis. J. 10:S20-S24.

11. Gaworzewska, E. and G. Colman. 1988. Changes in the pattern of infection caused by Streptococcus pyogenes. Epidemiol. Infect. 100:257-269.


• 164

13. Hauser, A. R., D. L. Stevens, E. L. Kaplan, and P. M. Schlievert. 1991. Molecular analysis of pyrogenic exotoxins from Streptococcus pyogenes isolates associated with toxic shock-like syndrome. J. Clin. Microbiol. 29:1562-1567.


15. Hoge, C. W., B. Schwartz, D. F. Talkington, R. F. Breiman, E. M. MacNeill, and S. J. Englender. 1993. The changing epidemiology of invasive group A streptococcal infections and the emergence of streptococcal toxic shock-like syndrome. A retrospective popula-tion-based study. JAMA 269:384-389.

16. Ikebe, T., A. Wada, Y. Inagaki, K. Sugama, R. Suzuki, D. Tanaka, A. Tamaru, Y. Fujinaga, Y. Abe, Y. Shimizu, and H. Watanabe. 2002. Dissemination of the phage-associated novel superantigen gene speL in recent invasive and noninvasive Streptococcus pyogenes M3/T3 isolates in Japan. Infect. Immun. 70:3227-3233.

17. Johnson, D. R., J. T. Wotton, A. Shet, and E. L. Kaplan. 2002. A comparison of group a streptococci from invasive and uncomplicated infections: are virulent clones responsible for serious streptococcal infections? J. Infect. Dis. 185:1586-1595.

18. Jones, D. M. and R. H. Mallard. 1993. Age incidence of meningococcal infection Eng-land and Wales, 1984-1991. J Infect. 27:83-88.

19. Kalia, A., B. G. Spratt, M. C. Enright, and D. E. Bessen. 2002. Influence of recombina-tion and niche separation on the population genetic structure of the pathogen Streptococ-cus pyogenes. Infect. Immun. 70:1971-1983.



22. Laupland, K. B., H. D. Davies, D. E. Low, B. Schwartz, K. Green, and A. McGeer. 2000. Invasive group A streptococcal disease in children and association with varicella-zoster virus infection. Ontario Group A Streptococcal Study Group. Pediatrics 105:E60.

23. Mascini, E. M., M. Jansze, J. F. Schellekens, J. M. Musser, J. A. Faber, L. A. Verhoef-Ver-hage, L. Schouls, W. J. van Leeuwen, J. Verhoef, and H. van Dijk. 2000. Invasive group A streptococcal disease in the Netherlands: evidence for a protective role of anti-exotoxin A antibodies. J. Infect. Dis. 181:631-638.


25. Moses, A. E., S. Goldberg, Z. Korenman, M. Ravins, E. Hanski, and M. Shapiro. 2002.

165 •

Invasive group a streptococcal infections, Israel. Emerg. Infect. Dis. 8:421-426.26. Musser, J. M., A. R. Hauser, M. H. Kim, P. M. Schlievert, K. Nelson, and R. K. Selander.

1991. Streptococcus pyogenes causing toxic-shock-like syndrome and other invasive dis-eases: clonal diversity and pyrogenic exotoxin expression. Proc. Natl. Acad. Sci. U. S. A. 88:2668-2672.

27. Musser, J. M., V. Kapur, S. Kanjilal, U. Shah, D. M. Musher, N. L. Barg, K. H. Johnston, P. M. Schlievert, J. Henrichsen, D. Gerlach, and . 1993. Geographic and temporal distri-bution and molecular characterization of two highly pathogenic clones of Streptococcus pyogenes expressing allelic variants of pyrogenic exotoxin A (Scarlet fever toxin). J Infect. Dis. 167:337-346.

28. Nakagawa, I., K. Kurokawa, A. Yamashita, M. Nakata, Y. Tomiyasu, N. Okahashi, S. Kawabata, K. Yamazaki, T. Shiba, T. Yasunaga, H. Hayashi, M. Hattori, and S. Hamada. 2003. Genome sequence of an M3 strain of Streptococcus pyogenes reveals a large-scale genomic rearrangement in invasive strains and new insights into phage evolution. Genome Res. 13:1042-1055.

29. Norrby-Teglund, A., K. Pauksens, S. E. Holm, and M. Norgren. 1994. Relation between low capacity of human sera to inhibit streptococcal mitogens and serious manifestation of disease. J Infect. Dis. 170:585-591.

30. Okada, N., M. K. Liszewski, J. P. Atkinson, and M. Caparon. 1995. Membrane cofactor protein (CD46) is a keratinocyte receptor for the M protein of the group A streptococcus. Proc. Natl. Acad. Sci. U. S. A. 92:2489-2493.

31. Okamoto, S., S. Kawabata, I. Nakagawa, Y. Okuno, T. Goto, K. Sano, and S. Hamada. 2003. Influenza A virus-infected hosts boost an invasive type of Streptococcus pyogenes infection in mice. J Virol. 77:4104-4112.

32. Smith, A., T. L. Lamagni, I. Oliver, A. Efstratiou, R. C. George, and J. M. Stuart. 2005. Invasive group A streptococcal disease: should close contacts routinely receive antibiotic prophylaxis? Lancet Infect. Dis. 5:494-500.

33. Spanier, J. G. and P. P. Cleary. 1980. Bacteriophage control of antiphagocytic determi-nants in group A streptococci. J. Exp. Med. 152:1393-1406.

34. Stromberg, A., V. Romanus, and L. G. Burman. 1991. Outbreak of group A streptococcal bacteremia in Sweden: an epidemiologic and clinical study. J. Infect. Dis. 164:595-598.

35. Sumby, P., K. D. Barbian, D. J. Gardner, A. R. Whitney, D. M. Welty, R. D. Long, J. R. Bailey, M. J. Parnell, N. P. Hoe, G. G. Adams, F. R. DeLeo, and J. M. Musser. 2005. Ex-tracellular deoxyribonuclease made by group A Streptococcus assists pathogenesis by en-hancing evasion of the innate immune response. Proc. Natl. Acad. Sci. U. S. A. 102:1679-1684.


37. Sumby, P., A. R. Whitney, E. A. Graviss, F. R. DeLeo, and J. M. Musser. 2006. Genome-

• 166

wide analysis of group a streptococci reveals a mutation that modulates global phenotype and disease specificity. PLoS. Pathog. 2:e5.

38. Sumby, P., A. R. Whitney, E. A. Graviss, F. R. DeLeo, and J. M. Musser. 2006. Genome-wide analysis of group a streptococci reveals a mutation that modulates global phenotype and disease specificity. PLoS. Pathog. 2:e5.

39. Zwart, S., G. J. Ruijs, A. P. Sachs, J. F. Schellekens, and R. A. de Melker. 2001. Poten-tially virulent strains and high colony counts of group A beta-haemolytic streptococci in pharyngitis patients having a delayed recovery or a complication. J Antimicrob. Chemother. 47:689-691.

167 •

• 168

Nederlandse samenvatting

• 170

171 •

De Streptococcus pyogenes, ofwel de groep A streptokok (GAS), is een bacterie die geassocieerd is met een zeer breed scala aan infectieziekten. Het merendeel van de infecties met de S. pyogenes verloopt mild en geeft aanleiding tot een keelont-steking of een oppervlakkige huidinfectie. Wanneer de infectie zich niet langer aan de oppervlakte afspeelt en de bacterie dus te isoleren is uit lichaamscompar-timenten die normaal steriel zijn (zoals bloed, hersenvloeistof, gewrichtsvloei-stof, spieren, diepe luchtwegen) is er sprake van een invasieve GAS ziekte. Zeer ernstige, en vaak fatale, invasieve GAS ziektebeelden zijn bloedvergiftiging (sep-sis), infectie van de bindweefselbladen die de spieren omgeven (necrotiserende fasciitis) en het zogenaamde toxic shock-like syndroom (TSS). Van dit laatste ziektebeeld is sprake wanneer de vitale systemen van het menselijk lichaam zoals de nier- en leverfunctie, de bloedsomloop en de ademhaling ernstig verstoord raken door eiwitten die de bacterie vrijmaakt gedurende het infectieproces. Deze eiwitten worden streptokokken pyrogene exotoxinen (Spe’s) genoemd en de verschillende varianten ervan worden aangeduid met SpeA, SpeB etcetera.

Niet alle typen S. pyogenes zijn in gelijke mate vertegenwoordigd onder in-vasieve GAS infecties. De bacterie kan op basis van de samenstelling van het M-eiwit, dat zich aan de buitenkant van het microorganisme bevindt, worden onderverdeeld. Inmiddels zijn er meer dan 130 verschillende M-typen bekend maar het zijn vooral M1 en M3 GAS die aanleiding geven tot ernstig inva-sieve infecties. Een andere belangrijke groep eiwitten bij infectie zijn adhesinen. Deze eiwitten bevinden zich ook aan de buitenkant en stellen de bacterie in staat om selectieve interacties aan te gaan met bepaalde weefselbestanddelen van de gastheer.

De genetische informatie (het “genoom”), coderend voor het M-eiwit, de su-perantigenen en alle andere bacteriële eiwitten is gelegen op een enkel circulair chromosoom dat bestaat uit ongeveer twee miljoen basenparen (ter vergelijking: het menselijk genoom bestaat uit 46 chromosomen die in totaal drie miljard basenparen bevatten). Momenteel is de genomische informatie van een zevental verschillende GAS stammen volledig bekend. Per bacterieel genoom blijken drie tot acht prophagen aanwezig. Een prophaag is een virus dat ingebouwd is in het bacteriële genoom. Naast de structurele phaag eiwitten codeert een prop-haag vaak ook voor eiwitten die bijdragen aan het ziekmakend vermogen van de bacterie. Veel pyrogene exotoxinen bijvoorbeeld, zijn prophaag-gecodeerd.

Met het beschikbaar komen van complete genomische informatie heeft

• 172

genetisch onderzoek een enorme vlucht kunnen nemen. Het meest bekende voorbeeld hiervan zijn de zogenaamde microarrays. Bij de microarray techniek worden kleine fragmenten van het complete genoom naast elkaar op een glaasje (de “microarray”) aangebracht. Enkele duizenden fragmenten van een beperkte grootte representeren op deze wijze het complete genoom. Daar het gefragmen-teerde DNA van een te analyseren stam alleen zal binden aan identieke frag-menten op de microarray kan worden nagegaan welke stukken van het genoom de te testen stam gemeenschappelijk heeft met de stam, die de basis vormde voor de microarray.

Reeds lang voor de ontdekking van de bacterie in 1874 was men bekend met typische GAS ziektebeelden zoals roodvonk en kraamvrouwenkoorts. Het aantal slachtoffers tengevolge van dit soort infecties nam sinds het einde van de 19e eeuw enorm af. Nog steeds speelt resistentie ontwikkeling in de strep-tokok een zeer beperkte rol en met de introductie van penicilline, halverwege de 20e eeuw, hoopte men dat invasieve GAS infecties definitief tot het verleden zouden behoren. In de jaren tachtig van de 20e eeuw kwam aan dit optimisme abrupt een einde toen vanuit veel westerse landen melding werd gemaakt van een plotse stijging in het aantal ernstige invasieve infecties met S. pyogenes. Zeer indrukwekkend hierbij was dat snel progressieve, en vaak fatale, infecties zich ook bij voorheen kerngezonde jongvolwassenen voordeden. Hierom werd de S. pyogenes in de volksmond ook wel de vleesetende bacterie genoemd.

Alhoewel niet-invasieve groep A streptokokken ziektebeelden, zoals keelont-stekingen, gedurende de gehele 20e eeuw onverminderd waren voorgekomen, kwam de plotse toename in ernstige invasieve infecties met deze bacterie ge-heel onverwacht. Voor de daling in de incidentie ongeveer een eeuw eerder werden betere hygiëne, volkshuisvesting en voeding als verklarende factoren aangevoerd. Eind jaren tachtig van de 20e eeuw hadden zich echter geen grote veranderingen op deze terreinen voorgedaan. Was er dan wellicht iets aan de bacterie veranderd waardoor zijn invasieve vermogen was toegenomen?

Naar aanleiding van deze toename zijn rond 1990 in veel landen, waaronder Nederland, surveillance systemen opgestart om de omvang en aard van het pro-bleem van invasieve GAS infecties in kaart te brengen. In maart 1992 werd in Nederland begonnen met een vanuit het Rijksinstituut voor Volksgezondheid en Milieu (RIVM) gecoördineerd surveillance systeem. Alle medisch microbio-logische laboratoria in Nederland werden verzocht om alle S. pyogenes stammen,

173 •

die aanleiding hadden geven tot een invasief GAS ziektebeeld, in te sturen naar het RIVM. Na ontvangst van zo’n “invasief GAS isolaat” stuurde het RIVM een vragenlijst terug naar de arts-microbioloog en behandelaar om klinische infor-matie over de patiënt te verkrijgen. Daar iedereen zich zeer bewust was van de plotse toename in ernstig invasieve infecties was er een grote mate van bereid-heid om aan het landelijk surveillance netwerk deel te nemen. Na twee jaar, in mei 1994, werd besloten om dit “gepassioneerd passieve” surveillance systeem te formaliseren en “actief ” te maken, middels een financiële compensatie voor ieder ingestuurd invasief GAS isolaat. Deze vorm van contractsurveillance werd alleen ingesteld voor de streeklaboratoria. Voor de overige laboratoria bleef de aard van het surveillance systeem ongewijzigd. Streeklaboratoria hebben, ten opzichte van de overige medisch microbiologische laboratoria in Nederland, een dekkingsgraad van 50%. Vanaf januari 1997 werden invasieve GAS isola-ten alleen nog verzameld via de streeklaboratoria en werd de klinische evaluatie ervan gestopt. In afwezigheid van klinische informatie werd een GAS isolaat gedefinieerd als “invasief ” wanneer de stam verkregen was uit een normaal ste-riel lichaamscompartiment.

Het doel van dit proefschrift was om enerzijds inzicht te verkrijgen in de epidemiologie van invasieve groep A streptokokken ziektebeelden in Nederland (hoofdstuk 2, 3 en 4) en anderzijds om de rol van onderliggende genetische profielen te bestuderen (hoofdstuk 5, 6 en 7).

Hoofdstuk 2. Van maart 1992 tot januari 1997 werd gepoogd om alle invasieve GAS isolaten klinisch te evalueren. Ongeveer de helft van alle GAS infecties bestond uit weke-delen infecties (wondinfecties, cellulitis, myositis, necroti-serende fasciitis). TSS, de meest ernstige systemische ontregeling tengevolge van een invasieve GAS infectie, ontstond het meest frequent in aansluiting op een necrotiserende fasciitis. Desalniettemin was de sterfte tengevolge van GAS pneumonie en GAS sepsis zonder focus hoger dan bij necrotiserende fasciitis. Deze aandoeningen komen relatief vaker voor bij ouderen met meer onderlig-gend medisch lijden. Naast necrotiserende fasciitis was een infectie met M1 of M3 S. pyogenes een onafhankelijke risicofactor voor de ontwikkeling van TSS in multivariate analyse. Dit vormt een epidemiologische onderbouwing voor de intrinsieke virulentie van deze M-typen. Verder waren bepaalde M-typen oververtegenwoordigd onder bepaalde ziektebeelden. Zo was M28 geassocieerd

• 174

met kraamvrouwenkoorts en M3 met necrotiserende fasciitis.De leeftijdsverdeling van invasieve GAS infecties is uitzonderlijk ten op-

zichte van het merendeel van de overige invasieve bacteriële ziektebeelden: naast de zeer jonge- en oudere leeftijdsgroepen is de incidentie ook sterk verhoogd bij jongvolwassenen met een piekincidentie tussen de 30 en 34 jaar. Sommige studies wijten dit verschijnsel aan kraamvrouwenkoorts. In Nederland worden echter zowel mannen als vrouwen in deze leeftijdscategorie vaker getroffen en maakt kraamvrouwenkoorts slechts een gering deel uit van alle invasieve GAS infecties.

Hoofdstuk 3. Het surveillance systeem was van maart 1992 tot mei 1994 geheel afhankelijk van de “gepassioneerd passieve” deelname van alle medisch micro-biologische laboratoria. Daarna werden de streeklaboratoria actief gestimuleerd middels een financiële compensatie voor ieder invasief GAS isolaat. Omdat het systeem voor de overige laboratoria ongewijzigd bleef vormde deze groep een goede referentie om het effect van actieve stimulatie te meten. Vanaf mei 1994 verdubbelde de geschatte incidentie van invasieve GAS infecties op basis van de streeklaboratoria data (ten opzichte van de data afkomstig van de overige laboratoria). Het aantal door de streeklaboratoria gerapporteerde TSS- en fatale gevallen bleef echter volkomen gelijk. Hier lijkt een interessant psychologisch mechanisme werkzaam: ongeacht de aard van het surveillance systeem worden de zeer ernstige gevallen altijd gemeld maar in een actief surveillance systeem worden de minder ernstige gevallen, die wel voldoen aan de definitie van inva-sieve GAS infecties, beter gerapporteerd.

Om alle gevallen van invasieve GAS ziekten te detecteren kan men niet vol-staan met alleen een microbiologische definitie (d.w.z. isolatie van S. pyogenes uit een normaal steriel lichaamscompartiment). Bij een vrij groot aantal inva-sieve GAS infecties zal namelijk alleen een oppervlakkige kweek verkregen wor-den. Alleen met de integratie van klinische gegevens kan aannemelijk worden gemaakt dat het om een invasieve GAS infectie gaat. Omdat zowel microbiolo-gische als klinische gegevens bekend waren uit de studieperiode van maart 1992 tot december 1996 kon de gevoeligheid en specificiteit van de microbiologi-sche definitie bepaald worden: de specificiteit bedroeg nagenoeg 100% (m.a.w. wanneer een S. pyogenes uit een normaal steriel compartiment wordt gekweekt past dat bij een invasieve GAS infectie), maar de sensitiviteit was beperkt: 66%

175 •

(m.a.w. in een op de drie gevallen van een klinisch invasieve GAS infectie wordt geen isolaat verkregen uit een steriel compartiment).

Deze fenomenen moeten in ogenschouw worden genomen wanneer de inci-denties van invasieve GAS infecties uit verschillende landen met elkaar worden vergeleken: een aanzienlijk deel van de gemelde variatie zou verklaard kunnen worden door de aard van de surveillance systemen gehanteerd in verschillende landen.

Hoofdstuk 4. Daar klinische evaluatie kostbaar is werd hiermee per 1 januari 1997 gestopt. Actieve surveillance voor, microbiologisch gedefinieerde, inva-sieve GAS infecties werd gecontinueerd met de streeklaboratoria tot december 2003. Corrigerend voor de dekkingsgraad van de streeklaboratoria en de speci-ficiteit van een zuiver microbiologische definitie van invasiviteit (66%), kon de jaarlijkse incidentie geschat worden. De incidentie van invasieve GAS infecties varieerde van 2.0 per 100 000 inwoners in 1999 tot 4.0 per 100 000 in 1995 en 1996. Ook was er een opmerkelijke seizoensinvloed: ieder jaar was de inciden-tie eind winter het hoogst. Van 1994 tot 2003 bleef de leeftijdsspecifieke inci-dentie gelijk. De sterk verhoogde incidentie bij de jongvolwassenen bleef zijn piek behouden tussen de 30 en 34 jaar. De verhoogde bevattelijkheid voor een invasieve GAS infectie in deze categorie is dus een leeftijdsspecifiek fenomeen en niet specifiek voor een cohort.

Het overgrote deel van alle GAS infecties wordt gevormd door keelontste-kingen bij kinderen. Deze niet-invasieve GAS infecties leiden tot specifieke antistoffen waarvan het beschermend effect in de loop der jaren echter weer afneemt. Hernieuwde expositie aan S. pyogenes via kinderen in jonge gezinnen, in combinatie met afnemende immuniteit, lijkt een aannemelijke verklaring voor de tweede piek in de leeftijdsspecifieke incidentie curve van invasieve GAS ziekten. Helaas is in deze studie de gezinssituatie van patiënten niet in kaart gebracht.

Hoofdstuk 5. In S. pyogenes zijn vele verschillende pyrogene exotoxinen en ad-hesinen bekend. Pyrogene exotoxinen, ook wel superantigenen genoemd, spe-len een centrale rol in de ontwikkeling van TSS. Verschillende adhesinen stellen de S. pyogenes in staat om te binden aan verschillende weefselcomponenten van de gastheer. Om te evalueren in welke mate het aanwezige repertoire van deze

• 176

virulentiefactoren het klinisch beeld bepaalt hebben we in hoofdstuk 5 gekeken naar de verdeling van 21 superantigenen en adhesinen in 170 verschillende S. pyogenes stammen die allemaal aanleiding hadden gegeven tot een GAS infectie van een specifiek weefsel of orgaan zoals meningitis (infectie van de hersenvlie-zen), artritis (infectie van de gewrichten), necrotiserende fasciitis (infectie van de bindweefselbladen rondom de spieren) en kraamvrouwenkoorts (infectie van de baarmoeder na bevalling). Wij vonden een zeer sterke relatie tussen het repertoire van virulentiefactoren en het M-type. Het M-type specifieke patroon van adhesinen en superantigenen zou derhalve een verklaring kunnen vormen voor de oververtegenwoordiging van bepaalde M-typen onder bepaalde ziekte-beelden.

Hoofdstuk 6. Omdat een infectie een complexe interactie tussen gastheer en bacterie vertegenwoordigt, waarbij aan de bacteriële zijde een groot aantal ge-nen een rol speelt, hebben we een microarray gemaakt waarin een groot deel van het GAS genoom vertegenwoordigd is. Doel van deze microarray studie was om genetische verschillen tussen verschillende GAS stammen in kaart te brengen en de aanwezigheid van een genetisch profiel te identificeren dat TSS veroorzakende stammen gemeenschappelijk hebben. Microarray analyse van 76 verschillende GAS stammen (acht verschillende M-typen) liet zien dat het M-eiwit een zeer sterke indicator is voor de genomische compositie van de bacterie: genomische verschillen tussen stammen van een verschillend M-type waren vele malen groter dan de verschillen binnen een M-type. De genetische verschillen tussen de M-typen werd met name bepaald door de specifieke aanwezigheid van bepaalde phagen. Deze observatie suggereert een directe of indirecte rol voor het M-eiwit bij phaag infecties van S. pyogenes.

Met behulp van de microarray konden we een aantal bekende en onbekende genen identificeren die uniek zijn voor M1 en M3 stammen. Verder onderzoek van deze nieuwe factoren zal moeten uitwijzen of zij bijdragen aan het buiten-gewoon invasieve karakter van deze M-typen. Desalniettemin geldt ook voor M1 en M3 S. pyogenes dat het merendeel hiervan aanleiding geeft tot niet-inva-sieve ziektebeelden. Anderzijds kunnen alle M-typen een TSS beeld veroorza-ken. Wij konden geen M-type onafhankelijk genetisch profiel te identificeren dat TSS stammen van niet-invasieve groep A streptokokken onderscheidt. An-dere factoren, zoals verschillende expressie van virulentiefactoren en gastheer-

177 •

gerelateerde verschillen lijken derhalve meer bepalend voor het beloop van een GAS infectie.

Hoofdstuk 7. Daar er geen grote veranderingen waren aan de zijde van de gast-heer ten tijde van de plotse toename van ernstig invasieve infecties rond 1990 lijkt een toegenomen invasiviteit van de bacterie rond die tijd de meest waar-schijnlijke verklaring. Om te onderzoeken of hiervoor een genetische basis is, vergeleken wij oude GAS isolaten uit de jaren zestig tot en met tachtig met M-type identieke “moderne” isolaten uit de jaren negentig. Microarray analyse van oude en moderne M1 en M28 stammen liet zien dat in M1 en M28 stammen een sequentiële acquisitie van phagen heeft plaatsgevonden. Deze phagen co-deren voor superantigenen en andere bekende virulentiefactoren zoals DNA en celmembraan-afbrekende enzymen. De acquisitie van deze virulentiefactoren in de streptokok lijkt een sleutelrol gespeeld te hebben. Desalniettemin valt voor M1, het meest voorkomende M-type, het verkrijgen van deze factoren (in de jaren zestig) niet samen met de toename van ernstige GAS infecties (eind jaren tachtig). Ook andere studies wijzen erop dat het verkrijgen van virulentiefacto-ren een voorwaarde is voor een toegenomen invasiviteit maar dat regulatie van de expressie van deze verkregen virulentiefactoren van groot belang is voor het tot stand brengen van een “invasief phenotype”. Gelet op de prominente rol van phagen is het van belang om de verspreiding ervan te vervolgen in het kader van de surveillance van invasieve GAS infecties.

• 178

179 •

Dankwoord

• 180

181 •

Graag wil ik op deze plek iedereen bedanken die heeft bijgedragen aan dit proefschrift.

Professor Verhoef, promotor en opleider: ik ben u zeer erkentelijk voor de oplei-ding tot arts-microbioloog en de mogelijkheid om deze specialisatie te combi-neren met promotieonderzoek. Ik ben vereerd dat ik de 100e promovendus ben die u mag bedanken voor uw inspirerende en coördinerende rol en hoop dat de inhoud van dit proefschrift voldoet aan uw verwachtingen vooraf.

Ellen Mascini: door jou wist ik vanaf het begin dat ik met S. pyogenes onderzoek goed zat: je was gezellig, kritisch en inspirerend. Hartelijk dank voor je grote bijdrage aan dit onderzoek.

Wouter Jansen, co-promotor: de wetenschappelijke diepgang in de tweede helft van dit proefschrift is voor een groot deel jouw verdienste. Naast je enthousi-asme in tijden van voorspoed en je steun als het even tegenzat ben ik je vooral dankbaar voor je vriendschap.

Joop Schellekens, co-promotor: het RIVM surveillance systeem, en daarmee jij, vormde de basis voor dit onderzoek. Je deur stond altijd open om de analyse van de surveillance gegevens en de opzet van onze manuscripten te bespreken. Jouw enthousiasme en creativiteit daarbij maakten passief meeroken ineens tot geen enkel probleem. Wilfrid van Pelt dank ik voor zijn uitstekende begeleiding bij de epidemiologi-sche hoofdstukken. De bijdrage van Leo Schouls en Wim Wannet aan vrijwel ieder hoofdstuk van dit proefschrift mag niet onvermeld blijven. Albert van Silfhout en Cees Elzenaar ben ik dankbaar voor het invoeren van de vragen-lijsten en het typeren van alle stammen. In dit kader ben ik ook veel dank ver-schuldigd aan alle artsen-microbioloog, infectiologen en andere inzenders van stammen en vragenlijsten in het kader van de landelijke surveillance.

Franz-Josef Schmitz: my sincere thanks for your scientific guidance at the initial

• 182

phase of this thesis. You are as friendly as you are inspiring and I am grateful that you have facilitated my stay in Vienna.

Rodger Novak and Emmanuelle Charpentier: my three months in Vienna were very rewarding in every respect, thanks to you. To be involved in such a sophis-ticated research project with such friendly people was a true honor.

Microarrays zijn high-tech en kostbaar maar dankzij Frank Schuren, Roy Mon-tijn en Rik Thijssen heb ik toch met deze technologie mogen werken. Ik dank jullie zeer hartelijk voor een fantastische tijd bij TNO. Martien Caspers: met de jou kenmerkende mengeling van vriendelijkheid, snelheid en intelligentie heb je mij geleerd hoe een microarray te maken en gebruiken. Bedankt!

Ad Fluit: stimulerende discussies op de gang en jouw kritische blik op de ma-nuscripten waren van grote waarde voor een heldere presentatie van onze data.

Mabel Beitsma: het “phaag werk” was buitengewoon concreet en ik ben je zeer dankbaar voor jouw bijdrage aan het laatste hoofdstuk. Armand Paauw en Adri-enne Box hebben me wegwijs gemaakt in het laboratorium van het Eijkman-Winkler instituut. Zonder Armand zou er geen PFGE-figuur in dit proefschrift staan.

Mijn collega arts-assistenten dank ik voor hun interesse en gezelligheid. Mede dankzij jullie collegialiteit en morele steun was het mogelijk om dit proefschrift tot een goed einde te brengen. In dit kader wil ik ook alle stafleden medische microbiologie en infectiologie noemen voor hun bijdrage aan onze opleiding en hun interesse voor mijn onderzoek.

Harm Wouter Snippe: dankzij jou is en het een “mooi boekje” geworden. Ik ben blij dat je, ondanks je drukke werkzaamheden in je nieuwe baan, tijd hebt gevonden voor de lay-out.

Behalve aan bovengenoemden ben ik veel dank verschuldigd aan de mensen in

183 •

mijn naaste omgeving. Mijn ouders en broer dank ik voor een liefdevolle jeugd. Als jonge vader realiseer ik me nu pas wat de onvoorwaardelijke liefde voor een kind betekent.

Heren paranimfen, beste Thijs en Mathew: iemand als paranimf vragen is een kleine uiting van grote genegenheid.

Oscar en Noortje: wat een rijkdom dat jullie in ons leven zijn. Tenslotte het erelid van het vierkoppige “Vlaminckx-Seebregts Dream Team”: Leonie, voor mijn gevoel ben ik nooit meer van de Domtoren afgekomen. Dank voor al je liefde en steun.

• 184

185 •

Curriculum vitae

Opleiding

1987-1993 Gymnasium β, Sint Thomas College, Venlo1993-1994 Propedeuse Geneeskunde (cum laude), Universiteit van Utrecht1994-1998 Doctoraal Geneeskunde (cum laude), Universiteit van Utrecht1998 United States Medical Licensing Examination (USMLE) Step 1

(cum laude)1998-2000 Co-assistentschappen2000-2001 History of Science (cum laude), Faculty of Modern History,

University of Oxford, UK2002-2007 Opleiding Medische Microbiologie, UMC Utrecht2007- Werkzaam als arts-microbioloog in het Sint Anthonius

Ziekenhuis, Nieuwegein

Opleiding overig

1998 Talma Eykman-prijs, Medisch Faculteitsfonds Talma Eykman2000 Talentenprijs, Ministerie van Onderwijs, Cultuur en Wetenschap

Wetenschappelijk onderzoek

1997-1998 Diffuse Reflectance Spectroscopy, Massachusetts General Hospital, Boston, MA, USA

2000 Prevalentie van integronen, Medische Microbiologie, UMC Utrecht2001 The use of graphic material in the early Royal Society 1662-

1686, University of Oxford, UK2002-2006 Virulentiefactoren en epidemiologie van Streptococcus pyogenes,

Medische Microbiologie, UMC Utrecht en RIVM, Bilthoven2003 Regulatie van virulentie-expressie door RNA in S. pyogenes, Max

F. Perutz Laboratory, Wenen, Oostenrijk2004 Microarray analyse van S. pyogenes, TNO, Zeist

Samenwonend met Leonie Seebregts. Zij hebben een zoon: Oscar (2005) en een dochter: Noortje (2006).

• 186

invasive group a streptococcal disease...invasive group a streptococcal disease national...

Documents