www.elsevier.com/locate/vetmic

Veterinary Microbiology 107 (2005) 91–101

Key role of Chlamydophila psittaci on Belgian turkey

farms in association with other respiratory pathogens

M. Van Loock a, T. Geens b, L. De Smit a, H. Nauwynck c, P. Van Empel d,C. Naylor e, H.M. Hafez f, B.M. Goddeeris a,c, D. Vanrompay b,*

a Department of Animal Sciences, Catholic University of Leuven, Belgiumb Department of Molecular Biotechnology, Faculty of Animal Science, Ghent University,

Coupure Links 653, 9000 Ghent, Belgiumc Department of Virology, Parasitology and Immunology, Ghent University, Belgium

d Intervet International, Boxmeer, The Netherlandse Department of Veterinary Pathology, University of Liverpool, UK

f Institute of Poultry Diseases, Free University Berlin, Germany

Received 23 December 2003; received in revised form 3 January 2005; accepted 17 January 2005

Abstract

Two hundred turkey sera from eight Belgian and two French farms were tested for the presence of antibodies against avian

pneumovirus (APV), Ornithobacterium rhinotracheale (ORT), Mycoplasma gallisepticum, Mycoplasma meleagridis and

Chlamydophila psittaci. At slaughter, C. psittaci, APV and ORT antibodies were detected in 94, 34 and 6.5% of the turkeys,

respectively. No antibodies against M. gallisepticum or M. meleagridis were present. Additionally, turkeys on three Belgian

farms were examined from production onset until slaughter using both serology and antigen or gene detection. All farms

experienced two C. psittaci infection waves, at 3–6 and 8–12 weeks of age. Each first infection wave was closely followed by an

ORT infection starting at the age of 6–8 weeks, which was still detectable when the second C. psittaci infection waves started.

Animals on farm A were not vaccinated against APV leading to an APV subtype B outbreak accompanying the first C. psittaci

infection wave. Despite subtype A APV vaccination on farms B and C, the second C. psittaci infection waves were accompanied

(farm B) or followed (farm C) by a subtype B APV infection. On all farms respiratory signs always appeared together with a

proven C. psittaci, APV and/or ORT infection. This study suggests an association between C. psittaci, APV and ORT, and

indicates the multi-factorial aetiology of respiratory infections in commercial turkeys. All three pathogens should be considered

when developing prevention strategies for respiratory disease.

# 2005 Elsevier B.V. All rights reserved.

Keywords: Chlamydophila psittaci; Seroprevalence; Avian pneumovirus; Ornithobacterium rhinotracheale

* Corresponding author. Tel.: +32 09 264 59 72; fax: +32 09 264 62 19.

E-mail address: Daisy.Vanrompay@ugent.be (D. Vanrompay).

0378-1135/$ – see front matter # 2005 Elsevier B.V. All rights reserved.

doi:10.1016/j.vetmic.2005.01.009

M. Van Loock et al. / Veterinary Microbiology 107 (2005) 91–10192

1. Introduction

All European turkey flocks experience one or more

periods of respiratory disease resulting in major

economical problems due to expensive antibiotic

treatments, loss of weight and carcass condemnation

at slaughter (Hall et al., 1975). Several etiological

agents like avian pneumovirus (APV), Escherichia

coli, Ornithobacterium rhinotracheale (ORT), Myco-

plasma spp. and Chlamydophila psittaci have been

described to be involved in respiratory distress

(Vandamme et al., 1994; Vanrompay et al., 1997;

Van de Zande et al., 1997; Dho-Moulin and Fair-

brother, 1999).

Like in most turkey-producing countries, APV

subtypes A and B infections highly prevalent in

Belgium, causing mild to unapparent clinical infec-

tions (Van de Zande et al., 1998). However, this

primary pathogen is economically important as it

allows E. coli, Newcastle disease virus, Bordetella

avium, Mycoplasma gallisepticum and ORT to

colonize the respiratory tract, resulting in severe

clinical signs and mortality (Naylor et al., 1992; Van

Empel et al., 1996; Van de Zande et al., 2001; Turpin

et al., 2002).

ORT, especially serotypes A and B, are highly

prevalent in turkey-producing countries (Van Empel

and Hafez, 1999). Several experimental infections

have demonstrated the potential complicating role of

ORT in respiratory disease (Van Empel et al., 1996;

Droual and Chin, 1997) and recently ORT has been

described as primary pathogen in both broilers and

turkeys (van Veen et al., 2000; Van Empel, personal

communication).

Turkey pathogens M. gallisepticum and M.

meleagridis can cause respiratory disease (Levisohn

and Kleven, 2000). Interactions between M. gallisep-

ticum and Newcastle disease virus, infectious bron-

chitis virus, Haemophilus paragallinarum and

adenovirus in experimentally infected chickens have

been documented (reviewed in Kleven, 1998). Similar

interactions between M. meleagridis and E. coli or M.

synoviae are also reported (Saif et al., 1970).

Chlamydia psittaci, recently reclassified as Chla-

mydophila psittaci, is a primary respiratory pathogen,

although often only regarded as a complicating agent

(Everett et al., 1999). In the past, severe respiratory

outbreaks have stressed the importance of C. psittaci

as turkey pathogen (reviewed in Andersen and

Vanrompay, 2003). In the USA, serovars A–E have

been isolated from turkeys, whereas in Europe only

serovars B and D have been discovered so far

(Andersen, 1997; Vanrompay et al., 1997). Vanrom-

pay et al. (1994) demonstrated differences in virulence

for strains belonging to serovars A, B and D (1994).

Strangely, C. psittaci is nowadays most often

neglected as etiological agent in outbreaks of

respiratory disease in turkeys. This is perhaps not

surprising as diagnosis of infection with this obligate

intracellular bacterium is difficult and handling of this

zoonotic agent requires special biohazard laboratory

conditions.

C. psittaci infections are highly prevalent on

Belgian and German turkey farms (Vanrompay

et al., 1997; Hafez et al., 1998a,b). Although C.

psittaci seems to be present on European turkey farms,

its role in the respiratory disease complex is unclear

and needs to be clarified. The present study tries to

contribute to this clarification by examining slaughter-

house sera for the presence of the ‘major’ respiratory

pathogens APV, ORT, M. gallisepticum, M. melea-

gridis and C. psittaci. Additionally, a longitudinal

study was performed on three Belgian turkey farms in

order to elucidate the kinetics of these infections from

production onset until slaughter.

2. Materials and methods

2.1. Farms and management schedule

In the fall of 2001, 200 turkeys from 10 different

farms in Belgium (eight farms) or in northern France

(two farms) were examined at slaughter for the

presence of serum antibodies against APV, ORT, M.

gallisepticum, M. meleagridis and C. psittaci. All

turkeys were vaccinated against Newcastle disease

(NCD) at the age of 1 day using a live-spray vaccine

(Nobilis1 ND LaSota; Intervet International, Boxm-

eer, The Netherlands) and at 3 weeks via drinking

water. In 7 out of the 10 farms animals had also been

vaccinated against APV using live vaccine based on a

subtype A strain (Nobilis1 RTV; Intervet Interna-

tional, Boxmeer, The Netherlands). APV vaccination

was performed at 1 or 2 weeks of age and repeated

once between 3 and 6 weeks of age via drinking water.

M. Van Loock et al. / Veterinary Microbiology 107 (2005) 91–101 93

Farmers provided information about clinical symp-

toms and mortality rates throughout the rearing period.

Additionally, kinetics of APV, ORT, M. gallisepti-

cum, M. meleagridis and C. psittaci infections were

examined on three Belgian (West-Flanders) turkey

farms (A, B and C) from day 1 until slaughter. Farms

A, B and C had experienced respiratory infections in

the past with average mortality rates during former

broods of 8, 7 and 5%, respectively. Farms applied an

all-in all-out management schedule with a sanitary

service period of 2 weeks between slaughter and

restocking during which the houses were cleaned and

disinfected. Each flock consisted of 5000–10,000

BUT/T9/Webster hens and toms, which were raised in

the same house. In the first 2 weeks, toms and hens

were kept separately in 20 groups. Then, two groups

were formed, one of hens and one of toms. Farmers

provided daily information on clinical symptoms and

mortality rates. All three farms tested negative for

Salmonella. All turkeys were vaccinated against NCD

(Nobilis1 ND LaSota) at 1 day and 3 weeks of age, as

mentioned above. Turkeys on farm A received no APV

vaccine, whereas the turkeys on farm B were

vaccinated only once with the Nobilis1 RTV vaccine

at 19 days of age. Turkeys on farm C received the

Nobilis1 RTV vaccine at 10 days of age, followed by a

booster vaccination at 24 days, as mentioned above.

2.2. Samples

Blood samples taken at the slaughterhouse from

hens and toms were stored overnight at room

temperature. Sera were collected after centrifugation

(325 � g, 10 min, 4 8C) and stored at �20 8C until

tested.

At the beginning of the longitudinal study on the

farms A, B and C, 30 hens were tagged with a number

and were allowed to move freely throughout the hens

department. Blood samples for antibody detection and

pharyngeal swabs for antigen detection were collected

from the tagged birds. Samples were weekly collected

from day 1 until week 4 and subsequently every 2 weeks

until slaughter at 14 (farms A and C) or 15 weeks (farm

B) of age. Blood samples were collected by venipunc-

ture of the cutaneous ulnar vein and treated as

mentioned above. Pharyngeal swabs were collected

using cotton-tipped aluminium shafted swabs (Fiers,

Kuurne, Belgium) containing 2 ml incomplete C.

psittaci transport medium, consisting of 0.2 M sucrose

(VWR International, Haasrode, Belgium), 0.015 M

Na2HPO4 (VWR International), 0.01 M NaH2PO4

(VWR International) and 20% inactivated foetal calf

serum (Integro, Leuvenheim, The Netherlands). Swabs

were shaken for 1 h and centrifuged (10 min, 2790 � g,

4 8C). One millilitre of supernatant was mixed with

0.5% streptomycin sulphate (10 mg/ml; Invitrogen),

1% vancomycine (50 mg/ml; Glaxo Smith Kline) and

0.8% fungizone (250 ml/ml; Invitrogen) and subse-

quently used for C. psittaci isolation. The remaining

1 ml was supplemented with 1% tetracycline (1 mg/ml;

Invitrogen) and used for the detection of APVand ORT.

All samples were stored at �80 8C until tested. All sera

were tested for a three to four-fold rise in antibody titre

against APV, ORT, M. gallisepticum, M. meleagridis

and C. psittaci. At the seroconversion time point,

attempts were made to detect and characterize C.

psittaci, APV and ORT from pharyngeal swabs.

2.3. C. psittaci ELISA

The enzyme-linked immunosorbent assay (ELISA)

was performed on turkey sera, which were pretreated

with kaolin to remove background activity (Novak

et al., 1993). Anti-recombinant major outer membrane

protein (rMOMP) antibody titres were determined by

an indirect ELISA using a standard protocol with

microwell plates coated with rMOMP, as previously

described (Vanrompay et al., 1998). Anti-MOMP

immunoglobulin titres were presented as the recipro-

cal of the highest serum dilution that gave an optical

density (OD450) above the cut-off value. The cut-off

value was the mean absorbance of seronegative

turkeys � three times the standard deviation (S.D.).

Negative control sera were obtained from 1-week-old

specific pathogen-free (SPF) turkeys (CNEVA; Plou-

fragan, France). Positive control sera originated from

experimentally infected SPF turkeys (Vanrompay

et al., 1999).

2.4. C. psittaci isolation and molecular

characterization

Pharyngeal swabs were examined for the presence

of viable C. psittaci by isolation in Buffalo Green

Monkey (BGM) cells and immunofluorescence stain-

ing, as previously described (IMAGENTM Chlamydia

M. Van Loock et al. / Veterinary Microbiology 107 (2005) 91–10194

immunofluorescence staining, Dakocytomation, Den-

mark, Vanrompay et al., 1992).

C. psittaci isolates were characterized by ompA

sequence analysis. The ompA gene was PCR-amplified

from genomic DNA and PCR products were purified

using Qiagen spin columns and cloned into pGemT

(Promega), as previously described (Vanrompay et al.,

1998). Transformation of E. coli DH5a was achieved

by heat-shock. Transformants were grown on LB-plates

in the presence of the appropriate antibiotic (Carbeni-

cillin, 30 mg/ml, Duchefa). Up to five selected clones

were picked and subsequently grown in liquid LB

medium for plasmid purification (Qiagen Tip 100).

OmpA sequence analysis was performed by the VIB

Genetic Service Facility and by the Laboratory of

Physiology and Immunology of Domestic Animals

(K.U. Leuven) using vector associated T7 and SP6

priming sites. All DNA sequence data were double-

stranded. Sequences of ompAwere aligned with related

sequences identified by BLAST (http://www.ncbi.nlm.-

nih.gov). Multiple alignments were done with CLUS-

TAL X software (default settings).

2.5. APV seroneutralisation assay

Sera were examined for the presence of APV

antibodies using a seroneutralisation (SN) test (Van de

Zande et al., 1998). The SN-titres were presented as

the reciprocal of the highest serum dilutions that

inhibited cytopathic effect. Based on Van de Zande

et al. (2002), SN titres �log2 6 were considered

indicative for a natural APV infection.

2.6. APV molecular characterization

APV was identified using a nested RT-PCR based

on the detection of the G protein gene (Naylor and

Shaw, 1997). This allowed the differentiation between

subtypes A and B, generating 300 and 400 bp PCR

products, respectively.

3. ORT ELISA

The ORT ELISA was performed using boiled,

extracted ORT serotype A antigens, as described by

Van Empel et al. (1997). Sera were considered positive

if titres were �log2 10, negative if titres were <log2 8

and suspicious when titres were between the latter two

values.

3.1. ORT detection using PCR

ORT 16S rDNAwas detected from pharyngeal swabs

using PCR primers described by Hung and Alvarado

(2001), which amplified a fragment of 78 bp. The DNA

was extracted from the pharyngeal swab suspensions

using the QIAamp DNA1 Mini Kit (Qiagen, Hilden,

Germany), following the manufacture’s recommenda-

tion. DNA amplification was carried out using ‘‘Ready

To Go- PCR’’ Beads (Amersham Bioscience, Freibug,

Germany) in a total volume of 25 ml containing 50 ng

DNA and 50 pmol of each primer. The PCR conditions

were as follows: after 5 min initial denaturation at

94 8C, 45 cycles were performed, each with 30 s

denaturation at 94 8C, 1 min annealing at 52 8C and 90 s

elongationat 72 8C. A final extension was performedfor

7 min at 72 8C. PCR products were separated on a 1%

SeaKem agarose gel (Cambrex Bio Science, USA)

together with a 100 bp ladder (Biolabs, Germany) and

stained with ethidium bromide.

3.2. M. meleagridis and M. gallisepticum ELISA

M. meleagridis and M. gallisepticum antibodies

were detected by use of Biocheck (The Netherlands)

and Svanova (Sweden) ELISA, respectively. Both

ELISA’s were performed according to the procedures

described by the manufacturers. For the Biocheck

ELISA, samples were considered positive when the

ratio of the optical densities of sample to positive

control was �0.5, negative if the ratio was <0.350 and

suspicious when the ratio was in between the latter two

values. For the Svanova Blocking ELISA samples were

considered positive if the inhibition percentage was

greater than 40%, negative if the inhibition percentage

was lower than 30% and suspicious when the inhibition

percentage was in between the latter two values.

4. Results

4.1. Farm history

All 10 farms, participating in the seroprevalence

study had experienced one or more periods of

M. Van Loock et al. / Veterinary Microbiology 107 (2005) 91–101 95

Table 1

Clinical symptoms and treatments in all three farms during the follow-up study

Age Symptoms Treatment

Farm A

Days 12–17 (weeks 2–3) Nasal discharge Soludox 500 g/1000 l water

Days 34–38 (weeks 5–6) Nasal discharge Flumiquil 400 g/1000 l water

Days 40–42 (week 6) Nasal discharge Flumiquil 400 g/1000 l water

Days 43–51 (weeks 7–8) Nasal discharge Soludox 500 g + Tylan 100 g/1000 l water

Days 62–67 (weeks 9–10) Nasal discharge Dicural 0.5 l

Farm B

Days 14–16 (week 3) Huddle together None

Days 21–28 (week 4) Coccidiosis + poor general condition Baycox 1 l/100 l water

Days 67–80 (weeks 10–12) Nasal discharge + diarrhoea Flumiquil 400 g/1000 l water

Days 75–79 (weeks 11–12) Bad feed uptake None

Days 88–92 (weeks 13–14) Diarrhoea Tylan 100 g/1000 l water

Farm C

Days 1–3 (week 1) Overall weakness Glucose

Days 27–30 (weeks 4–5) Huddle together Baycox 1 l/100 l water

Days 30–34 (week 5) Nasal discharge Soludox + Tylan

Days 63 (week 9) Diarrhoea Tylan 100 g/1000 l water

Days 75–83 (week 11) Nasal discharge + diarrhoea Tylan 100 g/1000 l water

respiratory distress from 10 days of age until

production week 12. Etiological diagnoses were not

performed, but after antibiotic treatment respiratory

signs disappeared; doxycycline (Soludox1) appeared

to be the most effective antibiotic. The mean mortality

rate on the farms was 5.95%. The mean carcass

condemnation rate at slaughter was 0.07%.

Observed clinical symptoms and treatments on

farms A, B and C participating in the longitudinal

study are presented in Table 1. Mortality rates on

farms A, B and C were 21.9, 28.9 and 14.7%,

respectively. On farm A, respiratory disease was the

primary cause of mortality. After an initial mild

respiratory infection with nasal discharge at week 2, a

more severe respiratory infection with nasal discharge

and dyspnoea was observed from week 4 onwards.

Flumiquil treatment was started at week 5. However,

after initial treatment, clinical symptoms almost

immediately reoccurred and only after treatment with

Soludox1 (Eurovet) followed by Dicural1 (Fort

Dodge) during weeks 7 till 9 turkeys became clinically

healthy again.

An extremely aggressive coccidiosis outbreak

during production week 4 was responsible for

28.9% mortality on farm B and could be successfully

treated with Baycox1 (Bayer HealthCare). From

week 10 onwards, birds suffered from nasal discharge

and diarrhoea, notwithstanding treatment with Flu-

miquil1 (CEVA Sante Animale, France) and Tylan1

(Elanco Animal Health, USA). However, treatment

was unsuccessful and disease symptoms were still

present at slaughter.

On farm C, weekly mortality rates were rather low

(<1% moralities/week) except during weeks 1 and 14.

During the first week 5% of the hens had experienced

overall weaknesses of an unknown cause, resulting in

dehydration and starvation. During the last week 7%

of the hens died accidentally by suffocation. Respira-

tory signs, characterized by nasal discharge, were only

observed at weeks 5 and 11 and were successfully

treated with Soludox1.

4.2. Seroprevalence on 10 turkey farms

Two hundred turkeys from 10 different farms

were examined at slaughter for the presence of

antibodies against APV, ORT, M. gallisepticum, M.

meleagridis and C. psittaci (Table 2). None of the

turkeys showed antibodies against M. gallisepticum or

M. meleagridis. C. psittaci antibodies were present in

188 of the 200 sera (94%). In 6 out of 10 farms all

tested sera were positive, whereas a minimum of 80%

of the examined samples were positive on the four

remaining farms.

M. Van Loock et al. / Veterinary Microbiology 107 (2005) 91–10196

Table 2

Seroprevalence of C. psittaci, APV and ORT in turkeys

Farm Country APV vaccine N C. psittaci-positive APV-positive O. rhinotracheale

Suspicious Positive

I Belgium Ya 20 18 5 4 4

II Belgium N 20 20 0 3 1

III Northern France Y 20 17 0 0 0

IV Belgium N 20 20 5 1 0

V Northern France Y 20 20 0 2 0

VI Belgium Y 20 18 2 0 1

VII Belgium Y 20 15 0 0 0

VIII Belgium Y 20 20 16 2 0

IX Belgium N 20 20 20 1 1

X Belgium Y 20 20 20 5 6

Total (%) 200 188 (94%) 68 (34%) 18 (9%) 13 (6.5%)

a Y = yes and N = no.

Although, 7 out of the 10 examined farms

vaccinated against APV, 34% of the sera at slaughter

had infection-correlated APVantibody titres (>6 log2)

(Table 2). Based on serum titres APV infection did

occur on four out of the seven vaccinated farms. Not

surprisingly, two out of three non-vaccinated farms

showed infection-correlated APV antibody titres at

slaughter. The third non-vaccinated farm was ser-

onegative.

In addition, all sera were screened for the presence

of ORT antibodies. At slaughter, only 13 of the 200

samples (6.5%) were positive and 18 samples (9%)

were considered as suspicious. At slaughter, the

overall percentage of positives was rather low, but still

8 out of 10 farms were found positive or suspected,

with one farm having an infection rate up to 30%.

4.3. Longitudinal study on three Belgian farms

At all time, all turkeys were seronegative for the

presence of M. gallisepticum and M. meleagridis

antibodies.

4.3.1. Chlamydophila psittaci

On all three farms, turkeys showed high maternal

antibody titres against C. psittaci (Fig. 1 I), as no C.

psittaci antigen or gene could be detected at that time

(data no shown). Maternal antibody titres declined and

almost disappeared at 4 weeks of age. C. psittaci

outbreaks were serologically noticed on all three

farms even before maternal antibodies completely

disappeared. Those first outbreaks were characterized

by an antibody peak at the age of 6 weeks. Turkeys

with antibody increases on farms A and C showed

nasal discharge while on farm B nasal discharge,

dyspnoea and a poor general condition was present.

During those first infection waves, C. psittaci

genotypes A, A plus B and E were detected on farms

A, B and C, respectively (Fig. 1 IV–VI). On farm A, a

slow and steady decline in C. psittaci antibody titre

was observed during the remaining production period,

indicative for continued stimulation of the immune

system, whereas on farms B and C serological

evidence of a second C. psittaci infection was present

at 8–10 and 10–12 weeks of age, respectively.

Infections on farms B and C were confirmed by

isolation and molecular characterization of different

C. psittaci strains. At 12 weeks of age a mixed

infection with genotypes A and F strains could be

demonstrated on farm B. Analyses of pharyngeal

swabs from farm C, taken on week 12, resulted in the

identification of a C. psittaci genotype E strain.

Because of the slow decrease in antibody titres on

farm A, pharyngeal swabs were examined at week 12.

This resulted in the isolation and molecular identifica-

tion of a C. psittaci genotype B strain. Thus, all three

farms had experienced two C. psittaci infection waves

from production onset until slaughter. The first

infection wave started at about 3–6 weeks of age

when maternal antibody titres had declined signifi-

cantly and the second one started at the age of 8–12

weeks.

M. Van Loock et al. / Veterinary Microbiology 107 (2005) 91–101 97

Fig. 1. Mean serological response of the turkeys against C. psittaci (&), APV (~) and ORT (*) on farms A (—), B ( � � � ) and C (. . .) throughout

the growing period. Results are grouped for each pathogen separately: C. psittaci (I), APV (II) and ORT (III), or for each farm separately: farms A

(IV), B (V) and C (VI). Arrows indicate time points of identification of C. psittaci strains (genotypes A, B, E and F) and avian pneumovirus

(subtype B).

M. Van Loock et al. / Veterinary Microbiology 107 (2005) 91–10198

4.3.2. Avian pneumovirus

Maternal antibodies were present on all farms but

were no longer detectable at the age of 3 weeks (Fig. 1

II). At 4 weeks of age seroconversion occurred in hens

on farms A and C. Hens on farm A had not been

vaccinated against APV and seroconversion was

unambiguously related to APV infection as demon-

strated by RT-PCR identification of a subtype B strain

in pharyngeal swabs taken at 6 weeks of age. In

contrast, hens on farm C received an APV vaccine at

the age of 10 and 24 days. Antibody titres augmented 1

week after the booster vaccination and were therefore

most probably due to vaccination rather than due to

infection. This was supported by the fact that we were

unable to demonstrate APV in pharyngeal swabs by

RT-PCR. On farm B, antibody titres augmented from 4

weeks of age on, about 1 week following APV

vaccination. Similar to farm C, this was probably due

to vaccination as APV could not be detected in

pharyngeal swabs. Following APV infection on

unvaccinated farm A, antibody titres slowly decreased

and APV was never redetected on this farm. In

contrast, hens of the vaccinated farms B and C showed

serological evidence of an APV infection at weeks 14

and 10, respectively. At that time, a subtype B strain

could be detected in pharyngeal swabs of farm B.

However, despite the significant antibody titre

increase on farm C, APV could not be detected in

pharyngeal swabs taken at 8, 10 and 12 weeks of age.

Nonetheless, antibody titres were as high as those

observed during a proven natural infection on the other

farms. Thus, the seroconversion at 10–12 weeks of age

was most probably due to an APV infection.

4.3.3. Ornithobacterium rhinotracheale

Considering a titre threshold of log2 5 as lower

detection limit, none of the turkeys possessed maternal

antibodies against ORT (Fig. 1 III). On all three farms,

ORT antibody titres started to appear at 6–10 weeks of

age. ORT ELISA titres of �log2 9.5 are considered

indicative for a positive flock (Van Empel, personal

communication). The flock was considered to be

negative if the titre was <log2 7.5 and suspicious if the

titre was between the latter two values. Although only

farm A appeared positive (�log2 9.5), antibody

increase on farms B and C suggested an ORT

infection. This was confirmed by PCR analysis

performed on 15 pharyngeal swabs of each farm:

50, 100 and 81.25% of the examined swabs of farms

A, B and C were positive, respectively.

5. Discussion

Despite the availability of commercial vaccines

against NCD and APV and the intensive viral

vaccination schemes, respiratory disease still remains

one of the most important economical problems in

turkey production. Probably, management failures as

well as a possible pathogen interplay between viral

and bacterial pathogens are involved in this multi-

factorial disease complex. This encouraged us to study

the possible pathogenic interactions between APV,

ORT, M. gallisepticum, M. meleagridis and C. psittaci

in the field. Initially, the present study examined the

prevalence of these pathogens in 10 different turkey

farms, which had experienced respiratory problems in

the past. Next, the co-occurrence of these infections

during one brood was determined on three Belgian

turkey farms during a follow-up study.

At slaughter, only seroneutralising antibody titres

against APV �log2 6 were considered indicative for

natural APV infections, since former vaccinations

with attenuated subtypes A or B strains (Poulvac1,

Fort Dodge and Aviffa RTI1, Merial, respectively)

always yielded titres lower than log2 6 (Van de Zande

et al., 2002). Based on this criterion, 7 out of the 10

farms still had infection-correlated antibodies at

slaughter with an average of 34% seropositive turkeys

on all farms. Interestingly, APV vaccination did not

induce full protection, as both vaccinated and non-

vaccinated flocks experienced APV infections.

Recently however, a new vaccination strategy pro-

posed by Nauwynck (personal communication) was

introduced in Belgium. Vaccination at day 1 by spray

and at 3 weeks in drinking water seemed to diminish

the APV outbreaks (Dr. P. Zwaenepoel, personal

communication, 2003). In Germany, an even higher

percentage (54.3%) of serologically positive turkeys

had been reported in 1993 (Redmann et al., 1993).

Also in the USA a significant APV seroprevalence rate

of 36.3%, ranging from 14.2 to 64.8%, was observed

between August 1998 and July 2002 in 2500

Minnesota turkey flocks (Goyal et al., 2003).

Although only 6.5% of the 200 examined sera were

ORT-positive, this still meant that 5 out of 10 farms

M. Van Loock et al. / Veterinary Microbiology 107 (2005) 91–101 99

had experienced an ORT infection. Moreover, Van

Empel and Hafez (1999) reported that ORT antibody

titres might decrease quite rapidly after infection

suggesting false negative diagnosis. Former German

serological surveys revealed 55 and 96.6% ORT-

positive turkey flocks (Ryll et al., 1997). In the USA,

Roepke et al. (1998) found 43% of the flocks between

production weeks 5 and 7 positive for ORT by

isolation from tracheal swabs. All this suggests the

limited value of the serodiagnosis for epidemiological

surveys, when performed at random times and

certainly at slaughter. Currently, no commercial

vaccine against ORT is available, although recently

the European Committee for Veterinary Medical

Products has granted EU marketing authorization

for the product Nobilis OR inac, an inactivated oil

emulsion vaccine containing an ORT serotype A strain

(Intervet International, Boxmeer, The Netherlands).

Interestingly, no M. gallisepticum and M. melea-

gridis infections were detected. Indeed, all sera tested

negative, while positive controls from the test kit as

well as our own positive controls from former

experimental infections (kindly obtained from P.

Van Empel, Intervet International) were positive.

Consequently, M. gallisepticum and M. meleagridis do

not seem to play a major role in the multi-factorial

respiratory disease complex on Belgian farms, which

is similar with results of Hafez et al. (1998a,b), who

found that all five examined German turkey flocks

were free from M. gallisepticum, M. meleagridis and

even M. synoviae.

All 10 examined farms were seropositive for C.

psittaci in our recombinant MOMP ELISA, with an

overall percentage of 94% positive turkeys. This is in

accordance with our previous study, 6 years ago,

where all examined farms reacted positive in Western

blot; at the end of a summer and winter brood,

respectively, 90 and 70% of the examined turkeys

were positive (Vanrompay et al., 1997). The presently

examined farms as well as those in the past suffered

from respiratory disease, but could be treated

successfully with tetracycline, the drugs of choice

against chlamydiosis. Unfortunately, respiratory dis-

tress often reappeared. Belgium is apparently not the

only European country with high prevalence rates for

C. psittaci. Studies in Germany, a far more important

turkey-producing country, revealed prevalence rates of

64% positives in 18 turkey flocks in 1987, 81.3%

positives in 25 flocks examined in 1997 and 60%

positives in five flocks examined in 1998 (Hafez and

Sting, 1997; Hafez et al., 1998a,b). Reports from other

European countries are absent as C. psittaci is often

neglected as a possible etiological agent in respiratory

disease.

During the longitudinal study, turkeys on all three

farms showed low and high maternal antibody titres

against APV and C. psittaci, respectively. APV

maternal antibody titres originated probably from

vaccination of breeder flocks. However, as a C. psittaci

vaccine is non-existent, maternal antibody titres have

to come from natural infection of breeder flocks. The

latter implicates that C. psittaci infections are present

in France wherefrom all our turkey eggs are imported.

The BUT/T9/Webster breeder flocks were apparently

not vaccinated against ORT and had no natural ORT

infection during egg production as antibodies could

not be detected in examined commercial turkey flocks

until production weeks 8–10.

Interestingly, all farms experienced two C. psittaci

infection waves, the first one starting at the age of 3–6

weeks when maternal antibody titres had declined

significantly and the second one starting at the age of

8–12 weeks. Respiratory signs were observed during

those two infection waves. Each first infection wave

was closely followed by an ORT infection starting at

the age of 6–8 weeks. Animals on farm A were not

vaccinated against APV leading to an APV subtype B

outbreak accompanying the first C. psittaci infection

wave. Similarly, in two German commercial turkey

flocks surveyed between 1989 and 1990, a significant

increase in C. psittaci antibody levels accompanied or

followed an increase in APV antibodies was detected

(Hafez et al., 1998a).

The C. psittaci first infection wave on farm A was

identified as genotype A, a genotype known to be very

pathogenic for turkeys (Vanrompay et al., 1994). After

primary infection antibody titres decreased very

slowly, indicative for continued stimulation of the

immune system. This was confirmed by isolation and

identification of a C. psittaci genotype B strain. Since

genotype B strains are only mildly pathogenic for

turkeys, infection pressure was probably not severe

enough to increase antibody responses (Vanrompay

et al., 1994) or could be partially neutralised by C.

psittaci antibodies derived from the first infection

wave. On farm B, the diversity of C. psittaci strains

M. Van Loock et al. / Veterinary Microbiology 107 (2005) 91–101100

was even greater: strains belonging to genotypes A, B

and F were identified. Since the first report of a

genotype F strain isolation from a parakeet, this is only

the third time a genotype F strain was detected and the

first time in turkeys (Andersen, 1997; Sudler et al.,

2004). On farm C, a single C. psittaci genotype E

strain was responsible for infection and re-infection of

the turkeys, despite treatment with Soludox1, a

chlamydiostatic antibiotic (Table 1). In conclusion,

more attention should be paid to chlamydiosis and its

prevention, including the development of a vaccine.

Despite vaccination with a subtype A APV vaccine,

farms B and C underwent a subtype B infection. These

results are in accordance with earlier studies, which

reported breakthroughs of subtypes A and B APV

infections after vaccinations (Naylor and Shaw, 1997;

Van de Zande et al., 2000). Interestingly, in the present

study, the APV infection occurred during (farm B) or

after (farm C) a C. psittaci infection.

C. psittaci seems to play a major role in the

respiratory disease complex in Belgian commercial

turkey farms. Evidence was provided that C. psittaci

can occur at early age, without a predisposing APV or

ORT infection, in contrast to Hafez et al. (1998b), who

observed that in four out of five examined turkey

flocks with respiratory manifestation, an increase in

the number of positive sera to C. psittaci was only

detected after a significant increase in the number of

positive sera to APV and/or ORT. Reservoirs for C.

psittaci are not known. However, wild birds and blood-

sucking ectoparasites are described as potential

reservoirs (Brand, 1989). Furthermore, transmission

from wild birds to domestic turkeys has been reported

(Page, 1976; Grimes, 1978). Vertical transmission

cannot be ruled out since it can occur in chickens, sea

gulls, ducks and psittacine birds (reviewed in Shewen,

1980). C. psittaci infections could weaken the health

of the turkeys making them more susceptible to APV

and ORT infections. This in turn could act as a

predisposing factor for the occurrence of additional C.

psittaci and APV infections at the end of the brood.

Respiratory signs on the farms always appeared

together with a proven C. psittaci, APV and/or ORT

infection. The present study clearly indicates the

multi-factorial aetiology of respiratory infections in

commercial turkey, APV and ORT as well as C.

psittaci should be taken into account when developing

prevention strategies for respiratory disease.

Acknowledgements

R. Pensaert, A. Doop, Ph. Deloddere and D.

Gilliaert are acknowledged for their assistance in the

slaughterhouse (Volys-Star, Lendelede, Belgium). P.

Zwaenepoel (Versele-Laga, Deinze, Belgium) is

acknowledged for providing general background

information on the Belgian turkey industry. M.

Vorstenbosch-Van der Ploeg, C. Boone and Dr. Dorte

Luschow are acknowledged for performing the ORT

ELISA, the APV serum neutralisation test and the

ORT-PCR, respectively. C. Savage is acknowledged

for performing APV RT-PCR. C. Borgers is acknowl-

edged for overall technical assistance. We also wish to

thank A. Wouters, S. Lambin, S. De Keersmaeker, E.

Huyck and Thi Quynh Tran Hoang for help in

sampling the turkeys at the farms. Thanks to C.

Ververken and F. Vandemaele for help in sequencing.

Additionally, we would like to thank all turkey farmers

involved for their voluntary participation. The Belgian

Ministry of Public Health (project S6037-Section 2)

and Intervet International N.V. (Boxmeer, The

Netherlands) are acknowledged for financial support.

References

Andersen, A.A., 1997. Two new serovars of Chlamydia psittaci from

North American birds. J. Vet. Diagn. Invest. 9, 159–164.

Andersen, A.A., Vanrompay, D., 2003. Chlamydiosis. In: Saif, Y.M.

(Ed.), Diseases of Poultry. 11th ed. Iowa State University Press,

Iowa, USA, pp. 863–879.

Brand, C.J., 1989. Chlamydial infections in free-living birds. J. Am.

Vet. Med. Assoc. 195, 1531–1535.

Dho-Moulin, M., Fairbrother, J.M., 1999. Avian pathogenic Escher-

ichia coli (APEC). Vet. Res. 30, 299–316.

Droual, R., Chin, R., 1997. Interaction of Ornithobacterium rhino-

tracheale and Escherichia coli O78 H9 when inoculated in the

air sac in turkey poults. In: Proceedings of the 46th Western

Poultry Disease Conference, Sacramento. p. 11.

Everett, K.D.E., Bush, R.M., Andersen, A.A., 1999. Emended

description of the order Chlamydiales, proposal of Parachlamy-

diaceae fam. nov. and Simkaniaceae fam. nov., each containing

one monotypic genus, revised taxanomy of the family Chlamy-

diaceae with descriptions of five new species, and standards for

the identification of organisms. Int. J. Syst. Bacteriol. 49,

415–440.

Goyal, S.M., Lauer, D., Friendshuh, K., Halvorson, D.A., 2003.

Seroprevalence of avian pneumovirus in Minnesota turkeys.

Avian Dis. 47, 700–706.

Grimes, J.E., 1978. Transmission of Chlamydiae from grackles to

turkeys. Avian Dis. 22, 308–312.

M. Van Loock et al. / Veterinary Microbiology 107 (2005) 91–101 101

Hafez, H.M., Sting, R., 1997. Uber das Vorkommen von Chlamy-

dien-Infektionen beim Mastgeflugel. Tierarztl. Umschau. 52,

281–285.

Hafez, H.M., Sting, R., Jodas, S., Stadler, A., 1998a. In: Kaleta,

F.E., Heffels-Redmann, U. (Eds.), Investigations on the Inter-

action between Chlamydia psittaci und Avian Pneumovirus

Infections, Proceedings of the International Symposium on

Infectious Bronchitis and Pneumovirus Infections in Poultry,

pp. 126–135.

Hafez, H.M., Sting, R., Jodas, S., Stadler, A., 1998b. In: Hafez,

H.M., Mazaheri, A. (Eds.), Chlamydia psittaci Infections in

Meat Turkey: Investigations on the Interaction with Other Avian

Infectious Agents, Proceedings of the 1st International Sympo-

sium on Turkey Diseases, pp. 208–217.

Hall, C.F., Glass, S.E., Grimes, J.E., Moore, R.W., 1975. An

epidemic of ornithosis in Texas turkeys in 1974. Southwest

Vet. 28, 19–21.

Hung, A.L., Alvarado, A., 2001. Phenotypic and molecular char-

acterization of isolates of Ornithobacterium rhinotracheale

from Peru. Avian Dis. 45, 999–1005.

Kleven, S.H., 1998. Mycoplasmas in the etiology of multifactorial

respiratory disease. Poultry Sci. 77, 1146–1149.

Levisohn, S., Kleven, S.H., 2000. Avian mycoplasmosis (Myco-

plasma gallisepticum). Rev. Sci. Tech. 19, 425–442.

Naylor, C., Shaw, K., 1997. Appearance of type B avian pneumo-

virus in Great Britain. Avian Pathol. 26, 327–338.

Naylor, C., Al-Ankari, A., Al-Afaleq, A., Bradbury, J., Jones, R.,

1992. Exacerbation of Mycoplasma gallisepticum infection in

turkeys by rhinotracheitis virus. Avian Pathol. 21, 295–305.

Novak, M., Moldoveanu, Z., Schafer, D.P., Mestecky, J., Compans,

R.W., 1993. Murine model for evaluation of protective immunity

to influenza virus. Vaccine 11, 55–60.

Page, L.A., 1976. Observations on the involvement of wildlife in an

epornitic chlamydiosis in domestic turkeys. J. Am. Vet. Med.

Assoc. 169, 932–935.

Redmann, T., Kamphausen, L., Neumann, U., Failing, K., 1993.

Seroprevalence of turkey rhinotracheitis in Germany. Prev. Vet.

Med. 16, 263–269.

Roepke, D., Back, A., Shaw, P., Nagaraja, K., Sprenger, S., Hal-

vorson, D., 1998. Case report: isolation and identification of

Ornithobacterium rhinotracheale from commercial turkey

flocks in the upper Midwest. Avian Dis. 42, 219–221.

Ryll, M., Hinz, K-H., Neumann, U., Behr, K.-P., Kruse, W., Lohren,

U., Sudbeck, M., 1997. Zu Pathogenitat und Verbreitung der

Ornithobacterium rhinotracheale-infection bei der Pute. In:

Proceedings of the 52nd Meeting of the Fachgruppe ’Geflugelk-

rankheiten’ der Deutsche Veterinarmedizinische Gesellschaft,

Hannover. pp. 39–60.

Saif, Y.M., Moorhead, P.D., Bohl, E.H., 1970. Mycoplasma melea-

gridis and Escherichia coli infections in germfree and specific-

pathogen-free turkey poults: production of complicated airsac-

culitis. Am. J. Vet. Res. 31, 1637–1643.

Shewen, P.E., 1980. Chlamydial infection in animals: a review. Can.

Vet. J. 21, 2–11.

Sudler, C., Hoelzle, L.E., Schiller, I., Hoop, R.K., 2004. Molecular

characterisation of chlamydial isolates from birds. Vet. Micro-

biol. 98, 235–241.

Turpin, E.A., Perkins, L.E., Swayne, D.E., 2002. Experimental

infection of turkeys with avian pneumovirus and either New-

castle disease virus or Escherichia coli. Avian Dis. 46, 412–422.

Vandamme, P., Segers, P., Vancanneyt, M., van Hove, K., Mutters,

R., Hommez, J., Dewhirst, F., Paster, B., Kersters, K., Falsen, E.,

1994. Ornithobacterium rhinotracheale gen. nov., sp. nov.,

isolated from the avian respiratory tract.. Int. J. Syst. Bacteriol.

44, 24–37.

Van de Zande, S., Nauwynck, H., Hendrickx, W., Pensaert, M.,

1997. Prevalentie van respiratoire and lymfotrope virussen bij

kalkoenen in Belgie. Vlaams Diergeneeskd. Tijdschr. 66, 223–

228.

Van de Zande, S., Nauwynck, H., Cavanagh, D., Pensaert, M., 1998.

Infections and reinfections with avian pneumovirus subtypes A

and B on Belgian turkey farms and relation to respiratory

problems. J. Vet. Med. B 45, 621–626.

Van de Zande, S., Nauwynck, H., Naylor, C., Pensaert, M., 2000.

Duration of cross-protection between subtypes A and B avian

pneumovirus in turkeys. Vet. Rec. 147, 132–134.

Van de Zande, S., Nauwynck, H., Pensaert, M., 2001. The clinical,

pathological and microbiological outcome of an Escherichia

coli O2:K1 infection in avian pneumovirus infected turkeys. Vet.

Microbiol. 81, 353–365.

Van de Zande, S., Nauwynck, H., Pensaert, M., 2002. Efficacy of

avian pneumovirus vaccines against avian pneumovirus/Escher-

ichia coli dual infection in turkeys. Vet. Rec. 150, 340–343.

Van Empel, P.C.M., Hafez, H.M., 1999. Ornithobacterium rhino-

tracheale: a review. Avian Pathol. 28, 217–227.

Van Empel, P.C.M., van den Bosch, H., Goovaerts, D., Storm, P.,

1996. Experimental infection in turkeys and chickens with

Ornithobacterium rhinotracheale. Avian Dis. 40, 858–864.

Van Empel, P.C.M., van den Bosch, H., Loeffen, P., Storm, P., 1997.

Identification and serotyping of Ornithobacterium rhinotra-

cheale. J. Clin. Microbiol. 35, 418–421.

Vanrompay, D., Ducatelle, R., Haesebrouck, F., 1992. Diagnosis of

avian chlamydiosis: specificity of the modified Gimenez stain-

ing on smears and comparison of the sensitivity of isolation in

eggs and three different cell cultures. J. Vet. Med. B 39, 105–

112.

Vanrompay, D., Ducatelle, R., Haesebrouck, F., 1994. Pathogenicity

for turkeys of Chlamydia psittaci strains belonging to the avian

serovars A, B and D. Avian Pathol. 23, 247–262.

Vanrompay, D., Butaye, P., Van Nerom, A., Ducatelle, R., Haeseb-

rouck, F., 1997. The prevalence of Chlamydia psittaci infections

in Belgian commercial turkey poults. Vet. Microbiol. 54,

85–93.

Vanrompay, D., Cox, E., Mast, J., Goddeeris, B., Volckaert, G.,

1998. High-level expression of Chlamydia psittaci major outer

membrane protein in COS cells and in skeletal muscles of

turkeys. Infect. Immunol. 66, 5494–6000.

Vanrompay, D., Cox, E., Vandenbussche, F., Volckaert, G., God-

deeris, B., 1999. Protection of turkeys against Chlamydia psit-

taci challenge by gene gun-based DNA immunisations. Vaccine

17, 2628–2635.

van Veen, L., Van Empel, P., Fabria, R., 2000. Ornithobacterium

rhinotracheale, a primary pathogen in broilers. Avian Dis. 44,

896–900.