ARTICLE IN PRESS

Microbiological Research 165 (2010) 21—32

0944-5013/$ - sdoi:10.1016/j.

�CorrespondE-mail addr

www.elsevier.de/micres

Heterotrophic microorganisms in deterioratedmedieval wall paintings in southern Italianchurches

O. Pepea,�, L. Sanninob, S. Palombaa, M. Anastasioc, G. Blaiottaa,F. Villania, G. Moschettid

aDSA Dipartimento di Scienza degli Alimenti, Universita degli Studi di Napoli Federico II, 80055 Portici, ItalybSIB Societa Italiana di Biotecnologie, 80055 Portici, ItalycEURECO European Environmental Company, 81015 Piana di Monte Verna, ItalydS.EN.FI.MI.ZO Dipartimento di Scienze Entomologiche, Fitopatologiche, Microbiologiche e Zootecniche,Universita degli Studi di Palermo, 90128 Palermo, Italy

Received 22 January 2008; received in revised form 18 March 2008; accepted 21 March 2008

KEYWORDSWall paintings;Heterotrophicmicroorganisms;Identification;Deterioration

ee front matter & 2008micres.2008.03.005

ing author. Tel.: +39 81ess: olipepe@unina.it (

SummaryThe Campania region in southern Italy is noted for its large number of churches thatharbour invaluable frescoes, dated from the beginnings of the 4th up to the 13thcentury. The wall paintings represent an integral part of the monuments, and theirdeterioration constitutes a potentially significant loss for the world’s culturalheritage. Heterotrophic microorganisms such as bacteria and mould can grow on thesurface of paintings that contain a wide range of organic and inorganic constituents,and provide different ecological niches that are exploited by a large variety ofmicrobial species. We isolated and identified the heterotrophic microorganismsfound in the biodegraded medieval wall paintings of seven historical churches inCampania. The paintings showed different levels of microbial contamination.Microbiological analysis of different paintings gave an overview of the differentheterotrophic microorganisms. Bacteria and moulds were isolated from 77% of thesampling points analysed, in which the most common type of alteration wasdiscolouration often associated with detachment of the paint layer. Bacterial strainswere identified by 16S rRNA partial sequence analysis. The Bacillus genus wasisolated in all churches, even though the type of species was variable, whereas allactinomycetes strains, isolated in five of the seven churches analysed, could bereferred to the Streptomyces genus. The similarity of the sequences analysed of the42 Bacillus spp., 2 Paenibacillus spp. and reference strains of different speciesshowed that these bacteria differentiated in 14 groups. The most frequently

Elsevier GmbH. All rights reserved.

2539410; fax: +39 81 2539407.O. Pepe).

ARTICLE IN PRESS

O. Pepe et al.22

occurring taxa were most closely related to Bacillus cereus/thurigiensis/anthracisand Bacillus pumilus groups. Thirteen Streptomyces spp. were differentiated inseven groups on the basis of neighbor-joining analysis of 16S rRNA. Fungi belonging tothe genera Penicillium, Aspergillus, Fusarium and Alternaria were also isolated fromdeteriorated wall paintings.& 2008 Elsevier GmbH. All rights reserved.

Introduction

Biological deterioration of works of art is assum-ing ever-greater importance especially in Italy withits disproportionate share (60%) of the world’scultural heritage, only 30% of which is properlyconserved. In particular, the region of Campania insouthern Italy has a wealth of ancient paintingsthat harbour invaluable frescoes dated from thebeginning of the 4th up to the 13th century. Wallpaintings support the growth of microorganismscommonly involved in biodeterioration, contribut-ing to the destruction of paint and its backing; theyare an integral part of the monuments, and theirdeterioration constitutes a loss affecting a signifi-cant part of the world’s cultural heritage. The sub-aerial environment of the churches represents asemi-closed area in which the microbial communityis enclosed in a mesocosm, partially isolated fromthe external environment and relatively stable incomparison to outdoor locations (Gorbushina et al.,2004). Heterotrophic microorganisms such as bac-teria and mould grow on the surface of paintingsthat contain a wide range of organic and inorganicconstituents and provide different ecologicalniches that are exploited by a large variety ofmicrobial species. In particular, wall paintingscontain pigments suspended in water or oil, oftenin the presence of a binder such as casein and milkapplied on the damp lime plaster. Moreover, thespectrum of compounds is further increased bythose that are added at later times duringretouching and restoration (Ciferri, 1999). Theactivity of fungal and bacterial species is supportedby many factors such as relative ambient humidity,temperature fluctuations, light, the nature ofnutrients on the material, its moisture content,physical properties of the surface of the object,moisture adsorption-emission mechanisms in thesupport, pH, dust, oxygen and carbon dioxideconcentration in the atmosphere, and the presenceof microclimates that may induce condensation(Valentın, 2003). However, the environmental con-ditions (humidity, temperature, light and pH) willdevelop specific microbial flora on a fresco andcould exacerbate the damage caused by air pollu-tion, biological attack and natural aging (Ciferri,

1999). The first colonizers are chemolithotrophicbacteria that induce biological corrosion of thebuilding material by the release of acids (Tomaselli,2003; Karpovich-Tate and Rebrikova, 1991). Theoccurrence of chemoorganotrophic bacteria hasalso been investigated for their capacity to produceorganic acids that solubilize the mineral compo-nents of the cultural heritage and affect the colourof the substrate surface (Urzı et al. 1991; Tiano1998; Perito et al. 2000; Perito and Mastromei,2003; Tomaselli 2003). They are commonly found oninorganic substrates containing traces of organiccompounds which settle on the masonry surface(Saiz-Jimenez, 1995, 1997; Zanardini et al., 2000,2002). Heterotrophic bacteria include a variety ofgenera such as Alcaligenes, Arthrobacter, Bacillus,Paenibacillus, Flavobacterium, Pseudomonas,Micrococcus, Staphylococcus, Nocardia, Mycobac-terium and Sarcina, which are the most frequentspecies isolated from wall paintings (Bassi et al.,1986; Saiz Jimenez, 1997; Ciferri, 1999). Recently,the use of a polyphasic approach for the detectionand identification of the bacteria isolated frombiodeteriorated frescoes, as well as the study ofthe microbial community, highlighted several newgenera previously not detected with conventionalmethods, indicating the large biodiversity found insuch inorganic substrates (Rolleke et al., 1998,2000; Gutner et al., 2000; Daffonchio et al., 2000;Saiz-Jimenez and Laiz, 2000. Schabereiter-Gutneret al., 2001, 2002; Heyrman and Swings, 2003).

The degrading activity of fungi on materials hasbeen widely investigated and proved, demonstrat-ing that most isolated taxa are common soilinhabitants. (Jeffries, 1986; Sorlini et al., 1987;Bravery, 1988; Sampo and Luppi Mosca, 1989; Garget al., 1995; Ciferri, 1999; Gorbushina et al., 2004).

Specific metabolic activities have been employedto develop new bioremediation methods based onthe use of microbial cells and enzymatic activity toremove organic material (Ranalli et al., 2000, 2005;Beutel et al., 2002; Antonioli et al., 2005) or bio-induce calcite precipitation using specific bacteriafor monumental stone reinforced (Castanier et al.,1999; Tiano et al. 1999; Fernandes 2006). Moreover,bioremediation of artworks is also based on theuse of sulphate-reducing bacteria, which reduce

ARTICLE IN PRESS

Heterotrophic microorganisms in deteriorated medieval wall paintings in southern Italian churches 23

sulphate to gaseous hydrogen sulphide, and nitrate-reducing bacteria, which reduce nitrates to gaseousnitrogen and nitrous oxide (Saiz-Jimenez 1997).

The aim of this study was to detect, isolate andidentify heterotrophic microorganisms from thebiodeteriorated medieval frescoes of seven histor-ical churches in the Campania region.

Materials and method

Site description and sampling zonesSamples for the study were taken from the

medieval churches of Sant’Antonio Abate, SanBiagio in Piedimonte Matese, San Marco in Cellole,Santa Maria in Foro Claudio, Santa Maria a Piazza,San Pietro ad Montes and Sant’Angelo in Formis inCampania. Painted between the 11th and 15thcenturies, the frescoes can be considered as thestarting point of the Romanesque painting in thesouthern Italy. The siting, history and descriptionof sampling places of the churches are shown inTable 1. At least three different frescos per churchwere analyzed by sampling the visually discolouredarea, detached and/or altered by spots, patinas orefflorescence (Table 1). For the sake of conserva-tion, duplicate samples were collected understerile conditions by using sterile nitrocellulosemembranes, following the method reported byPitzurra et al. (1997a, b) and Poletti et al. (1999).In particular, nitrocellulose membranes (SartoriusAG, Gottingen, Germany), 47mm in diameter and17.34 cm2, were pressed on the surface of the wallpaintings for 30 s, using finger tips protected bysterile gloves and immediately transferred toplates containing a specific growth medium. Het-erotrophic bacteria were enumerated onto nutrientagar (NORMAL 1/88), moulds onto malt agar (Oxoid)with chloramphenicol (0.002 g/L) and actinomy-cetes onto starch casein agar (Kuster and Williams,1964). All tests were carried out in duplicate.

Identification of bacteria and actinomycetesRepresentative, morphologically different iso-

lated colonies, obtained from the counting plates,were randomly picked and purified by streaking onthe same isolation medium. The colonies were thenexamined by assessing their morphology (phase-contrast microscopy), and the bacteria werecharacterized by Gram stains and catalase activity.Bacteria and actinomycetes were identified by 16SrDNA sequencing as described below. DNA wasextracted by the InstaGeneTM Matrix (Bio-RadLaboratories, Hercules, CA) according to thesupplier’s recommendations. Supernatant contain-ing DNA was stored at –20 1C. Five microliter of the

mixture was used directly as a template for PCRamplifications. Synthetic oligonucleotide primersdescribed by Weisburg et al. (1991), fD1 (50-AGA-GTTTGATCCTGGCTCAG-30) and rD1 (50-AAGGAGGT-GATCCAGCC-30) (Escherichia coli positions 8–17 and1540-1524, respectively), were used to amplify the16S rDNA. PCR mixture was prepared as previouslyreported (Blaiotta et al., 2002), while PCR condi-tions consisted of 30 cycles (1min at 94 1C, 45 s at54 1C, 2min at 72 1C) plus one additional cycle at72 1C for 7min as a final chain elongation.

The presence of PCR products was verified byagarose (1.5% w/v) gel electrophoresis, at 100 V for2 h, purified by using a QIAquick gel extraction kit(Qiagen S.p.A., Milan) and sequenced by using theprimer fD1 (Weisburg et al., 1991). The DNAsequences were determined by the dideoxy chaintermination method by using the DNA sequencingkit (Perkin-Elmer Cetus, Emeryville, CA) accordingto the manufacturer’s instructions. The sequenceswere analysed by MacDNasis Pro v3.0.7(Hitachi Software Engineering Europe S.A., OlivetCedex, F), and research for DNA similarity wasperformed with the GenBank and EMBL database(http://www.ncbi.nlm.nih.gov/Blast.cgi) (Altschulet al., 1997).

Phylogenetic analysis was performed using MEGAversion 4.0 (Tamura et al., 2007) after multiplealignment of data by ClustalW 1.8 (Thompsonet al., 1994). Distance matrix and neighbor-joiningmethods (Saitou and Nei, 1987) were applied fortree construction.

Identification of fungiFungal isolates were obtained from the cultural

medium used for the counts. The isolates werepurified in the same counting medium and thenexamined by a light microscope using the Johnsonslide culture method of cultivation (Harrigan andMc Cance, 1976). Thus the isolates were identifiedto the level of genus according to their morphologyand sporing structures.

Microclimatic conditionsThe churches examined in this study are situated

a considerable distance inland in the eastern partof Campania (Italy). Microclimatic parameters(temperature and relative humidity (RH)) weremonitored every hour in winter (from 10.00 a.m. to6.00 p.m.) on sampling days. At each site, ambienttemperature (ET, 1C) and RH (%) were measured atindoor locations by using a Microclimatic StationBABUC (LSI, Milan, Italy). All data were elaboratedby Info GAP software (v 2.0.5.; LSI- LASTEM).

ARTICLE IN PRESS

Table 1. Description of sampling places (�) of wall paintings at Sant’Antonio Abate (AA), San Pietro ad Montes (PM),San Biagio in Piedimonte Matese (BPM) and San Marco in Cellole

Churches History Sampling places (�) Description of the samplingplaces

Sant’AntonioAbate (AA)

Church founded in the 15thcentury. The wall paintingsdate from the same century

AAa: heavy discoloration anddetachment on the fresco ofthe ‘‘Virgin’s Crowing’’AAb: heavy discoloration anddetachment on the fresco ofthe ‘‘Annunciation’’AAc: heavy discoloration anddetachment on the fresco ofthe ‘‘Adoration of the Magi’’

San Pietro adMontes (PM)

Benedictine monasteryfounded in the 11th centuryon the ruins of the Romantemple of Jupiter Tifatinus.The Byzantine wall paintingsdate back to the 11thcentury

PMa: heavy discoloration anddetachment on fresco of the‘‘Virgin with Baby andSaints’’PMb: heavy discoloration anddetachment on fresco of‘‘Crucifixion’’PMc: heavy discoloration anddetachment on fresco of the‘‘Virgin with Baby andSaints’’PMd: heavy discoloration anddetachment on illegiblefresco

San Biagio inPiedimonteMatese (BPM)

Late-Gothic church foundedat the end of the 3rd or atthe beginning of the 4thcentury. The wall paintingsare the high expression of4th-century figurative art inCampania

BPMa: heavy discolorationand detachment with whiteefflorescence on frescodepicting a New or OldTestament sceneBPMb: heavy discolorationand detachment with whiteefflorescence on frescodepicting a New or OldTestament sceneBPMc: heavy discolorationand detachment with whiteefflorescence on frescodepicting a New or OldTestament sceneBPMd: heavy discolorationand detachment with whiteefflorescence on frescodepicting a New or OldTestament scene

San Marco inCellole (MC)

Little church founded in the4th–5th century ADdecorated with wallpaintings of Byzantine style

MCa: heavy discoloration anddetachment with whiteefflorescence on ornamentalfresco with two humanfiguresMCb: heavy discoloration anddetachment with whiteefflorescence on ornamentalfresco with human figuresMCc: heavy discoloration anddetachment white

O. Pepe et al.24

ARTICLE IN PRESS

Table 1. (continued )

Churches History Sampling places (�) Description of the samplingplaces

efflorescence on ornamentalfresco with human figuresMCd: heavy discoloration anddetachment with whiteefflorescence on ornamentalfresco with human figuresMCe: heavy discoloration anddetachment on a ‘‘Griffon’’fresco

Santa Mariain ForoClaudio(MFC)

Palaeochristian basilicafounded in the 4th–5thcentury AD. The wallpaintings date back to the11th and 12th century

MFCa: light green patina onthe fresco of ‘‘Virgin withBaby’’MFCb: ochre to brown spotsand heavy discoloration onthe fresco of ‘‘Virgin withSaints’’MFCc: white efflorescence onthe fresco of ‘‘Artisan’sCrafts’’MFCd: heavy discolorationand detachment onornamental fresco withfigures of elephants

Santa Maria aPiazza (MP)

Romanesque-Gothic churchfounded in the 10th–11thcentury. The exactfoundation date is unknown.The wall paintings date backto the 13th–16th century

MPa: diffused discolorationon fresco depicting ‘‘Virginon throne with Baby’’MPb: diffused discolorationon fresco depicting ‘‘IllegibleSaint’’ with book and cordMPc: light discoloration offresco depicting ‘‘St. MichaelArchangel’’MPd: light discoloration of‘‘Crucifixion with Virgin andAngel’’MPe: diffused discolorationof fresco depicting‘‘Crucifixion with Virgin andSaints’’

Sant’Angeloin Formis(AF)

Basilica founded in the 11thcentury by the Lombards onthe ruins of the Romantemple of Diana Tifatina. TheByzantine-style wallpaintings were executedduring the building of thechurch between 1072 and1087

AFa: discoloration anddetachment on right smallapse portraying the ‘‘Virginbetween St. Anastasia andSt. Magdalene’’AFb: discoloration anddetachment on left smallapse portraying ‘‘John theBaptist between St. Peterand St. Paul’’AFc: heavy discoloration anddetachment of the frescosdepicting ‘‘Last Judgmentscene with demons, flames,Lucifer and Judas chained’’

Heterotrophic microorganisms in deteriorated medieval wall paintings in southern Italian churches 25

ARTICLE IN PRESS

O. Pepe et al.26

StatisticsStatistical treatment of data by standard devia-

tion (SD) was performed.

Results

Total counts

In the present study, bacteria, actinomycetes andfungi were first counted to evaluate microbialcontamination on medieval frescos in seven differentchurches of the Campania region. As shown inFigure 1, heterotrophic bacteria prevailed in all wallpaintings monitored, and their concentration rangedfrom 2.8CFU/17.34cm2 in the church of San Biagio inPiedimonte Matese to 23.8CFU/17.34cm2 in thechurch of Sant’Angelo in Formis, although thebacteria count increased to 55.0CFU/17.34cm2 inthe wall paintings of San Marco in Cellole. In this lastchurch, we also detected the highest level ofactinomycetes (13.8CFU/17.34 cm2) compared tothe other churches in which they were absent orlower (from 1.0 to 2.8CFU/17.34 cm2). The mouldswere detected at levels from 2.0 to 8.0CFU/17.34 cm2 on the wall paintings of all churchesanalysed. In general, the level of contaminationvaried with the church and with the sampling pointon the fresco examined, as shown by the SD values

Figure 1. Contamination by heterotrophic bacteria, actinomchurches in Campania region. The SD values were in theactinomycetes from 0.8 to 1.3; and fungi from 0.8 to 2.2.

calculated (heterotrophic bacteria from 1 to 15;actinomycetes from 0.8 to 1.3; fungi from 0.8 to 2.2).

Identification of heterotrophicmicroorganisms

The heterotrophic bacteria were isolated from77% of the sampling points analysed in which themost common type of alteration was discoloura-tion, often associated with detachment of the paintlayer (Table 1). A total of 50 bacterial strains wereidentified by the 16S rRNA partial sequenceanalysis. Forty-two were related to Bacillus genus,three to Staphylococcus, two to Paenibacillus, twoto Brevundimonas and one to Brevibacterium. TheBacillus genus was isolated in all churches, eventhough the type of species was variable. The resultsof the neighbor-joining analysis on the basis of the50 end of 16S rRNA sequences of the 42 Bacillusspp., 2 Paenibacillus spp. and 20 reference strainsof different species are shown in the dendrogramdepicted in Figure 2. On the basis of similarity ofthe sequences analysed, the strains were differ-entiated into 14 groups. The lowest number ofspecies was detected on wall paintings of San Pietroad Montes, in which only strains most closelyrelated to Bacillus megaterium (group IX) andBacillus muralis/macroides (group XI) specie wereisolated. By contrast, the highest variability withseven different Bacillus groups, closely related to

ycetes and fungi detected on wall paintings in medievalfollowing ranges: heterotrophic bacteria from 1 to 15;

ARTICLE IN PRESS

Figure 2. Neighbor-joining tree based on comparison of the 50 end of 16S rRNA gene sequences showing thephylogenetic relationships between Bacillus (B.) spp. and Paenibacillus (P.) spp. strains. Bootstrap values (expressed aspercentages of 1000 replications) greater than 50% are given at the nodes. The scale bar estimates the number ofsubstitutions per site. The accession number of 16S rRNA genes of reference strains is reported in square brackets.*Origin: AA, Sant’Antonio Abate; PM, San Pietro ad Montes; BPM, San Biagio in Piedimonte Matese; MP, Santa Maria aPiazza; AF, Sant’Angelo in Formis; MC, San Marco in Cellole; MFC, Santa Maria in Foro Claudio. The minor-casealphabetic letter indicates the sampling point.

Heterotrophic microorganisms in deteriorated medieval wall paintings in southern Italian churches 27

ARTICLE IN PRESS

Groups AA PM BPM MP AF MC MFCb

ac

b

cdc

aada

IIIIVVVI

a

b

Origin*

I

II

VII

Strain A46Strain A72Strain A54Strain A31Str. griseus subsp. griseus NBRC 15744 [AB184699]Strain A13Strain A48Strain A44Strain A71Strain A43Strain A49Strain A42Str. cavourensis subsp. washingtonensis NRRL B-8030 [AY999909]Str. halstedii NRRL B-1238 [EF178695]

Str. cyaneofuscatus NBRC 13190 [AB184860]Str. anulatus DSM 40128 [AY999838]Str. griseus ATCC 51928T [AF112160]

Str. virginiae JCM 4974 [D85117]Str. cavourensis subsp. cavourensis NRRL 2740 [DQ445791]Str. roseochromogenus NBRC 13080 [AB184297]

Streptomyces lavendulae subsp.lavendulae IFO 14028 [D85114]Str. fungicidicus NBRC 13848 [AB184529]Str. roseochromogenus subsp. albocyclini NBRC 13828 [AB184512]

Str.albus subsp. albus DSM 40313T [AJ621602]Str. saprophyticus NBRC 13440 [AB184404]strain 57B

Str. krainskii NBRC 13053 [AB184278]strain 59A32

32100

8078

4966

38

6845

97

63

20

0.005

Figure 3. Neighbor-joining tree based on comparison of 50 end of 16S rRNA gene sequences showing the phylogeneticrelationships between Streptomyces (Str.) spp. strains. Bootstrap values (expressed as percentages of 1000replications) greater than 50% are given at the nodes. The scale bar estimates the number of substitutions per site.The accession number of 16S rRNA genes of reference strains in reported in square brackets. *Origin: AA, Sant’AntonioAbate; PM, San Pietro ad Montes; BPM, San Biagio in Piedimonte Matese; MP, Santa Maria a Piazza; AF, Sant’Angelo inFormis; MC, San Marco in Cellole; MFC, Santa Maria in Foro Claudio. The minor-case alphabetic letter indicates thesampling point.

O. Pepe et al.28

the species Bacillus polyfermenticus/velezensis/subtilis/amyloliquefaciens/atrophaeus (group I),was detected on frescos of Santa Maria in ForoClaudio. In this last church, other bacteria belong-ing to Staphylococcus and Brevundimonas sp. werealso detected (data not shown). Moreover, at thesampling sites MFCb of Santa Maria in Foro Claudio,we detected the presence of ochre to brown spots,in which we isolated Bacillus, while no actinomy-cetes and fungi were detected (Figures 2 and 3).The most frequently occurring taxa were mostclosely related to Bacillus cereus/thurigiensis/anthracis (group IV; 10 sampling points/sixchurches) and Bacillus pumilus (group III; eightsampling points/four churches). Two strains belong-ing to the groups VII and VIII, both isolated from thesampling point MFCa, were genetically relativelyfar from the type strains (o98%), indicating thepossibility of the presence of unknown typical wallpainting species.

All actinomycetes strains, isolated in five of theseven churches analysed, could be referred to theStreptomyces genus. The results of the neighbor-joining analysis on the basis of the 50 end of 16SrRNA sequences of the 13 Streptomyces spp. and

some reference strains are shown in the dendro-gram depicted in Figure 3. On the basis of thesimilarity of the sequences analysed, the strainswere differentiated into seven groups. Elevenstrains were closely related among them andto more reference strains of species Str. griseus,Str. cavourensis, Str. halstedii, Str. cyaneofuscatusand Str. anulatus (group I, Figure 3). The remainingtwo strains (57B and 59A), both isolated from SanBiagio in Piedimonte Matese church (BMPa andBMPb), were related to Str. saprophyticus andStr. krainskii (group VII, Figure 3). More than 50%of Streptomyces strains were isolated from evidentwhite efflorescence developed in different frescosof San Biagio in Piedimonte Matese (BPMa, b and c),Santa Maria in Foro Claudio (MFCc) and San Marco inCellole (MCa, b and c), in which heavy discoloura-tion and detachment were also observed (Table 1;Figure 3).

The specific composition of the fungal isolateshowed that one or both the genera, Aspergillusand Penicillium, were isolated in all wall paintingsanalysed. Indeed, the genera Alternaria andFusarium were found only in Sant’Angelo in Formis(AFb and AFc) and Santa Maria a Piazza (MPd),

ARTICLE IN PRESS

Heterotrophic microorganisms in deteriorated medieval wall paintings in southern Italian churches 29

respectively. In particular, in two churches ana-lysed, Aspergillus niger and Aspergillus flavus wereisolated.

Only one wall painting analysed (MFCa) showed alight green patina from which we isolated Asper-gillus sp., Bacillus species and Brevundimonas sp.Moreover, strains closely related to the B. pumilusgroup, detected in different sampling pointsanalyzed, was always associated with fungi belong-ing to the genera Aspergillus and Penicillium (datanot shown).

Microclimatic conditions

In the churches examined, the daily averagetemperatures ranged between 9.30 and 13.80 1C.San Marco in Cellole showed the highest value of RH(80.60%), whereas the RH (%) of all other sitesranged between 44.46 and 73.86 (data not shown).

Discussion

Microbial contamination was evaluated by enu-meration of bacteria, actinomycetes and fungipresent on ancient wall paintings of medievalchurches of the Campania region. According toPoletti et al. (1999), the use of nitrocellulosemembrane allows the study of the quantity andquality of heterotrophic microbial population of thepainted surface. This method is a well-defined,standardized and non-destructive sampling techni-que. Culture-dependent methods, technique ofmicrobial identification, were used to obtainmicrobial strains to employ in laboratory experi-ments (Ranalli et al., 2005; Alakomi et al., 2006),and that can be characterized using differenttechniques. Heterotrophic bacteria were prevalentin all wall paintings monitored, whereas actinomy-cetes and fungi were detected at lower levels. Thiscould explain the diffusion of discolouration thatwas detected in all wall paintings analysed. Muralpaintings contain a wide range of organic andinorganic constituents and provide different ecolo-gical niches that may be exploited by a largevariety of microbial flora that colonize thesesurfaces and may cause aesthetic and structuraldamage (Ciferri, 1999). Bacteria seem to be thefirst colonizers of moist frescos and buildingmaterials, and thus represented the first supply oforganic matter (Garg et al., 1995; Karpovich-Tateand Rebrikova, 1991). As previously reported(Sorlini et al. 1987; Tomaselli 2003), a large varietyof heterotrophic bacteria are commonly found oninorganic substrata containing traces of organicmaterial but do not exhibit pigmented cells (Tiano

and Tomaselli 1989). They excrete organic acidsthat have biocorrosive action, thus contributing tothe discolouration of the painted surface.

According to Laiz et al. (2003), the dominantbacterial genera detected by standard spreadplating on nutrient agar were Bacillus and Paeni-bacillus. Comparable results were also obtained byGorbushina et al. (2004) who isolated bacteriabelonging to Bacillus and Bacillus-related generafrom mural paintings of the St. Martin’s church atGreene-Kreiensen in Germany. These results con-firm that a culture-dependent method could yieldmainly spore-forming bacteria (Laiz et al., 2003),although Saarela et al. (2004), in their study on theheterotrophic microorganisms in biofilm samplesfrom Roman catacombs, isolated overall aerobic,anaerobic and facultatively anaerobic bacteria inwhich Bacillus spp. were sporadic. The bacterialgroups obtained by phylogenetic relationshipsbetween the Bacillus strains isolated in this workindicated the presence of species previously iso-lated in other studies from mural paintings andmonuments (Gorbushina et al., 2004; Radaelliet al., 2004; Saarela et al., 2004). This reflect theadaptation possibilities for a subaerial system inindoor environment of these isolates that haveconquered such specific niche. Strains belonging tothe Bacillus muralis group, a novel species of thegenus Bacillus previously isolated from muralpaintings and monuments (Gorbushina et al.,2004; Heyrman et al., 2005; Suihko et al., 2007)and to potential slime-forming Bacillus group (Pepeet al., 2003) isolated from ochre-to-brown spots,were detected on frescos in Santa Maria in ForoClaudio. Species belonging to the genus Staphylo-coccus, isolated from the churches of Sant’AntonioAbate, Santa Maria in Foro Claudio and San Biagio inPiedimonte Matese, which is generally associatedwith human skin, could represent mere contami-nants (Gorbushina et al., 2004). In this study weused culture-dependent methods, which allowed usto obtain microbial strains to use in futurelaboratory experiments and that can be character-ized using different techniques. Moreover, thecultivation of microorganisms could determinewhether the DNA derives from living or deadmicroorganisms and, more importantly, could allowus to distinguish between microorganisms poten-tially responsible for the observed damage andthose that do not contribute to it (Ciferri, 1999;Gorbushina et al., 2004; Fernandes, 2006).

In our research we observed that discolourationwas sometimes associated with detachment of thepaint layer and/or to the development of efflores-cence or a patina. We isolated from these samplingpoints some heterotrophic filamentous microbes like

ARTICLE IN PRESS

O. Pepe et al.30

fungi, belonging to the genera Aspergillus andPenicillium and actinomycetes, belonging to thegenera Streptomyces, which contribute to mechan-ical destruction of wall paintings due to myceliaproduction. The hyphae penetrate the painted layer,degrading some of its components (especially gluesand binders), which results in a reduction in itscohesion, thus giving rise to exfoliations, crackingand loss of the paint (Rolleke et al., 1996; Ciferri,1999). According to Suihko et al. (2007), identifica-tion of Streptomyces at the species level is quitedifficult, as was also demonstrated in this study inwhich several sequences of the isolates were highlyrelated to more than one type of strain. Speciesbelonging to the genus Streptomyces have beenreported as the first predominant colonizers whenthe frescos were excavated and exposed to the air.They can exist for extended periods as restingarthrospores that germinate in the occasionalpresence of nutrients (Suihko et al., 2007).

According to Gorbushina et al. (2004), the fungalgenera isolated from the seven medieval churcheswere frequently mentioned as being typical formural paintings (Garg et al., 1995; Berner et al.,1996). Different species of the genera Aspergillus,Penicillium and Fusarium inhabit indoor environ-ments (Berner et al., 1996; Gorbushina andPetersen, 2000; Gorbushina et al., 2004). Inparticular, we found the species Aspergillus sydo-wii, a ubiquitous microorganism, reported byprevious studies (Sampo and Luppi Mosca, 1989;Berner et al., 1996), to utilize materials employedin restoration as a source of nutriment and toinduce colourations on the painting under suitableenvironmental conditions. Aspergillus niger, whichis one of the causes of biodarkening of stonesurfaces (Tomaselli, 2003), and Aspergillus flavuswere also found. According to Ciferri (1999), weisolated from the same sampling points, fungibelonging to the genera Aspergillus and Penicilliumassociated with different species of Bacillus. Ciferri(1999) in his investigations on painted canvasesreported that cellulolytic and proteolytic activitiesof A. niger stimulated the survival and growth on‘‘mock paintings’’ of B. pumilus. Thus, a trophicinterrelationship plays the main role within micro-bial communities on frescos, plaster and brick, and,in such cases, the death and lyses of such bacteriawould promote the growth of fungi (Karpovich-Tateand Rebrikova, 1991; Garg et al., 1995).

Conclusion

In conclusion, the wall paintings of the sevenmedieval churches in question had different levels

of heterotrophic microbial contamination. Theculture-dependent identification method alloweddetection of spore-forming bacteria and moulds inall the churches analysed. This microbial presenceis an index of potential fresco decay since bacilliand fungi are able to produce a large number ofspores, structures that guarantee their survival fora long time, and restore their growth rapidly whenthe ambient conditions are once again favourable.

Since it is generally difficult to demonstrate theinvolvement of bacteria in biodeterioration, theircontribution to biodeterioration is yet to beestablished. Further studies are necessary to obtainan accurate diagnosis of the biodeteriorigens thatcolonize a specific work of art, in such a way asto understand their precise biodeteriorationpotential.

Acknowledgements

We acknowledge the financial support of theItalian Biotechnology Society (SIB) and Paolo Gallo,through the project ‘‘Development of rapid expertdiagnostics to prevent deterioration of culturalheritage in external environments’’. We thank theHeritage Offices of Caserta and Benevento pro-vinces for allowing sampling in the churches. Theauthors are grateful to Prof. S. Coppola for readingthe manuscript and providing us with helpfulsuggestions.

References

Alakomi H-L, Paananen A, Suihko M-L, Helander IM,Saarela M. Weakening effect of cell permeabilizers ongram-negative bacteria causing biodeterioration.Appl. Environ. Microbiol. 2006;72:4695–703.

Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z,Miller W, et al. Gapped BLAST and PSI-BLAST: a newgeneration of protein database search programs.Nucleic Acids Res. 1997;25:3389–402.

Antonioli P, Zapparoli G, Abbruscato P, Sorlini C, RanalliG, Rigetti PG. Art-loving bugs: the resurrection ofSpinello Aretino from Pisa’s cemetery. Proteomics2005;5:2453–9.

Bassi M, Ferrari A, Realini M, Sorlini C. Red stains on theCertosa of Pavia a case of biodeterioration. Int.Biodeterior. Biodegr. 1986;22:201–5.

Berner M, Wanner G, Lubitz W. A comparative study ofthe fungal flora present in the medieval wall paintingsin the Chapel of the Castle Herberstein and in theParish Church of St Georgen in Styria, Austria. Int.Biodeterior. Biodegr. 1996;40:53–61.

Beutel S, Klein K, Knobbe G, Konigfeld P, Petersen K,Ulber R, et al. Controlled enzymatic removal of

ARTICLE IN PRESS

Heterotrophic microorganisms in deteriorated medieval wall paintings in southern Italian churches 31

damaged casein layers on medieval wall paintings.Biotechnol. Bioeng. 2002;80:13–21.

Blaiotta G, Pepe O, Mauriello G, Villani F, Andolfi R,Moschetti G. 16S–23S rDNA intergenic spacer regionpolymorphism of Lactococcus garvieae, Lactococcusraffinolactis and Lactococcus lactis as revealed by PCRand nucleotide sequence analysis. Syst. Appl. Micro-biol. 2002;25:520–7.

Bravery AF. Biodeterioration of paint -a -state-of-the-artcomment. In: Houghton DR, Smith RN, Eggings HOW,editors. Biodeterioration, vol. 7. New York, NY: Else-vier Applied Science; 1988. p. 466–85.

Castanier S, Le Metayer-Levrel G, Perthuisot J-P. Carbo-nates precipitation and limestone genesis – themicrobiogeologist point of view. Sediment Geol.1999;126:9–23.

Ciferri O. Microbial degradation of painting. Appl.Environ. Microbiol. 1999;65:879–85.

Daffonchio D, Borin S, Zanardini E, Abbruscato P, RealiniM, Urzı C, et al. Molecular tools applied to the study ofdeteriorated artworks. In: Ciferri O, Tiano P, Mastro-mei G, editors. Of Microbes and Art. The Role ofMicrobial Communities in the Degradation and Protec-tion of Cultural Heritage. New York: Kluwer academic/Plenum Publishers; 2000. p. 21–38.

Fernandes F. Applied microbiology and biotechnology inthe conservation of stone cultural heritage materials.Appl. Microbiol. Biotechnol. 2006;73:291–6.

Garg KL, Kamal KJ, Mishra AK. Role of fungi in thedeterioration of wall paintings. Sci. Total Environ.1995;167:255–71.

Gorbushina AA, Petersen K. Distribution of micro-organ-isms on ancient wall paintings as related to associatedfaunal elements. Int. Biodetererior. Biodegrad. 2000;46:277–84.

Gorbushina AA, Heyrman J, Dornieden T, Gonzales-Delvalle M, Krumbein WE, Laiz L, et al. Bacterial andfungal diversity and biodeterioration problems inmural painting environments of St. Martins church(Greene-Kreisen, Germany). Int. Biodetererior. Biode-grad. 2004;53:13–24.

Gutner C, Heyrman J, Pinar G, Lubitz W, Swings J, RollekeS. Comparative analyses of the bacterial diversity ontwo different biodeteriorated wall paintings by DGGEand 16S rDNA sequence analysis. Int. Biodetererior.Biodegrad. 2000;46:229–39.

Harrigan WF, Mc Cance ME. Laboratory Methods in Foodand Dairy Microbiology. London: Academic Press; 1976.

Heyrman J, Swings J. Modern diagnostic techniques onisolates. COALITION a concerted action from theEuropean Commission (EVK4-CT-1999-2001) on mole-cular microbiology as an innovative conservationstrategy for indoor and outdoor cultural assets News-letters 6, 2–5, 2003, online at /http://www.geomic.uni-oldenburg.de/projekte/coalition/S.

Heyrman J, Logan NA, Rodrıguez-Dıaz M, Scheldeman P,Lebbe L, Swings J, et al. Study of mural paintingisolates, leading to the transfer of Bacillus marocca-nus and Bacillus carotarum to Bacillus simplex,emended description of Bacillus simplex, re-examina-

tion of the strains previously attributed to Bacillusmacroides and description of Bacillus muralis sp. nov.Int. J. Syst. Bacteriol. 2005;55:119–31.

Jeffries P. Growth of Beauveria alba on mural paintings inCanterbury Cathedral. Int. Biodeterior. Biodegrad.1986;22:11–3.

Karpovich-Tate N, Rebrikova NL. Microbial communitieson damaged frescos and building materials in theCathedral of the Nativity of the Virgin in the Pafnutii-Borovskii Monastery, Russia. Int. Biodeterior. Bio-degrad. 1991;27:281–96.

Kuster E, Williams ST. Selection of media for isolation ofStreptomycetes. Nature 1964;202:928–9.

Laiz L, Pinar G, Lubitz W, Saiz-Jimenez C. Monitoring thecolonization of monuments by bacteria: cultivationversus molecular methods. Environ. Microbiol.2003;5:72–4.

NORMAL 1/88. Microflora autotrofa ed eterotrofa: tec-niche di isolamento in coltura.

Pepe O, Blaiotta G, Moschetti G, Greco T, Villani F. Rope-producing strains of Bacillus spp. from wheat breadand strategy for their control by lactic acid bacteria.Appl. Environ. Microbiol. 2003;69:2321–9.

Perito B, Mastromei G. Conservation of monumentalstones by bacterial biomineralization. Microbiol. Today2003;30:113–4.

Perito B, Biagiotti L, Daly S, Galizzi A, Tiano P, MastromeiG. Bacterial genes involved in calcite crystal pre-cipitation. In: Ciferri O, Tiano P, Mastromei G, editors.Of Microbes and Art. The Role of Microbial Commu-nities in the Degradation and Protection of CulturalHeritage. New York: Kluwer academic/Plenum Pub-lishers; 2000. p. 219–30.

Pitzurra M, Savino A, Pasquarella C. Il monitoraggioambientale microbico. Ann. Igiene 1997a;9:439–54.

Pitzurra M, Savino A, Pasquarella C, Poletti L. A newmethod to test the microbial contamination ofsurfaces. Hyg. Med. 1997b;22:77–92.

Poletti L, Pasquarella C, Pitzurra M, Savino A. Compara-tive efficiency of nitrocellulose membranes versusRODAC plates in microbial sampling on surfaces.J. Hosp. Infect. 1999;41:195–201.

Radaelli A, Paganini M, Basavecchia V, Elli V, Neri M,Zanotto C, et al. Identification, molecular biotypingand ultrastructural studies of bacterial communitiesisolated from two damaged frescoes of St Damian’sMonastery in Assisi. Lett. Appl. Microbiol. 2004;38:447–53.

Ranalli G, Matteini M, Tosini I, Zanardini E, Sorlini C.Bioremediation of cultural heritage: removal ofsulphates, nitrates and organic substances. In: CiferriO, Tiano P, Mastromei G, editors. Of Microbes and Art –

The role of Microbial Communities in the Degradationand Protection of Cultural Heritage. New York: KluwerAcademic/Plenum Publisher; 2000. p. 231–45.

Ranalli G, Alfano G, Belli C, Lustrato G, Colombini MP,Bonaduce I, et al. Biotechnology applied to culturalheritage: biorestoration of frescoes using viablebacterial cells and enzymes. J. Appl. Microbiol. 2005;98:73–83.

ARTICLE IN PRESS

O. Pepe et al.32

Rolleke S, Muyzer G, Wawer C, Wanner G, Lubitz W.Identification of bacteria in a biodegraded wallpainting by denaturing gradient gel electrophoresisof PCR-amplified gene fragments coding for 16S rRNA.Appl. Environ. Microbiol. 1996;62:2059–65.

Rolleke S, Witte A, Wanner G, Lubitz W. Medieval wallpaintings a habitat for archaea: identification ofarchaea by denaturing gradient gel electrophoresis(DGGE) of PCR amplified gene fragments coding for16s rRNA in a medieval wall painting. Int. Biodeterior.Biodegrad. 1998;41:85–92.

Rolleke S, Gutner C, Pinar G, Lubitz W. Molecularapproaches for the assessment of microbial deteriora-tion of objects of art. In: Ciferri O, Tiano P, MastromeiG, editors. Of Microbes and Art. The Role of MicrobialCommunities in the Degradation and Protection ofCultural Heritage. New York: Kluwer Academic/Plenum Publishers; 2000. p. 39–48.

Saarela M, Alakomi H-L, Suihko M-L, Maunuksela L,Raaska L, Mattila-Sandholm T. Heterotrophic micro-organisms in air and biofilm from Roman catacombs,with special emphasis on actinobacteria and fungi.Int. Biodeterior. Biodegrad. 2004;54:27–37.

Saitou N, Nei M. The neighbor-joining method: a newmethod for reconstructing phylogenetic trees. Mol.Biol. Evol. 1987;4:406–25.

Saiz-Jimenez C. Deposition of anthropogenic compoundson monuments and their effect on airborne micro-organisms. Aerobiology 1995;11:161–75.

Saiz-Jimenez C. Biodeterioration vs. biodegradation: therole of microorganisms in the removal of pollutantsdeposited on historic buildings. Int. Biodeterior.Biodegrad. 1997;24:225–32.

Saiz-Jimenez C, Laiz L. Occurrence of halotolerant/halophilic bacterial communities in deterioratedmonuments. Int. Biodeterior. 2000;46:319–26.

Sampo S, Luppi Mosca AM. A study of the fungi occurringon 15th century frescos in Florence, Italy. Int.Biodeterior. Biodegrad. 1989;25:343–53.

Schabereiter-Gutner C, Pinar Dietmaret Vybiral G, LubitzW, Rolleke S. Rubrobacter-related bacteria associatedwith rosy discolouration of mansory and lime wallpaintings. Arch. Microbiol. 2001;176:347–54.

Schabereiter-Gutner C, Pinar Dietmaret Vybiral G, Lubitz W,Rolleke S. Phylogenetic 16S rRNA analysis reveals thepresence of complex and partly unknown bacterialcommunities in Tito Bustilio cave, Spain, and on itsPaleolithic paintings. Environ. Microbiol. 2002;4:392–400.

Sorlini C, Sacchi M, Ferrari A. Microbiological deteriora-tion of Gambara’s frescos exposed to open air inBrescia, Italy. Int. Biodeterior. 1987;23:167–79.

Suihko M-L, Alakomi HL, Gorbushina A, Fortune I,Marquardt J, Saarela M. Characterization of aerobic

bacterial and fungal microbiota on surfaces of historicScottish monuments. Syst. Appl. Microbiol. 2007;30:494–508.

Tamura K, Dudley J, Nei M, Kumar S. MEGA4: MolecularEvolutionary Genetics Analysis (MEGA) software ver-sion 4.0. Mol. Biol. Evol. 2007;24:1596–9.

Thompson JD, Higgins DG, Gibson TJ. Clustal W: improv-ing the sensitivity of progressive multiple sequencealignment through sequence weighting, position-specific gap penalties and weight matrix choice.Nucleic Acids Res. 1994;22:4673–80.

Tiano P. Biodeterioration of monumental rocks: decaymechanisms and control methods. Sci. Technol. Cult.Heritage 1998;7:13–32.

Tiano P, Tomaselli L. Un caso di biodeterioramento delmarmo. Arkos 1989;6:12–8.

Tiano P, Biagiotti L, Mastromei G. Bacterial bio-mediatedcalcite precipitation for monumental stones conserva-tion: methods of evaluation. J. Microbiol. Methods1999;36:139–45.

Tomaselli L. Biodeterioration process on inorganic sub-strata. COALITION: a concerted action from theEuropean Commission (EVK4-CT-1999-2001) on mole-cular microbiology as an innovative conservationstrategy for indoor and outdoor cultural assets News-letters 6, 5–9, 2003. online at /http://www.geomic.uni-oldenburg.de/projekte/coalition/S.

Urzı C, Cristo G, Pernice A. Adhesion to and degradationof marble by a Micrococcus strains isolated from it.Geomicrobiol. J. 1991;9:81–90.

Valentın N. Microbial contamination and insect infesta-tion in organic materials. COALITION: a concertedaction from the European Commission (EVK4-CT-1999-2001) on molecular microbiology as an innovativeconservation strategy for indoor and outdoor culturalassets Newsletters 6, 2–5, 2003. Online at /http://www.geomic.uni-oldenburg.de/projekte/coalition/S.

Weisburg WG, Barns SM, Pelletier DA, Lane DJ. 16Sribosomal DNA amplification for phylogenetic study.J. Bacteriol. 1991;173:697–703.

Zanardini E, Abbruscato P, Ghedini N, Realini M, Sorlini C.Influence of atmospheric pollutants on the biodeter-ioration of stone. Int. Biodeterior. Biodegrad. 2000;45:35–42.

Zanardini E, Abbruscato P, Realini M, Brusetti L, Sorlini C.Molecular techniques applied to the study of microbialcommunities colonising different lithotypes. COALI-TION: a concerted action from the European Commis-sion (EVK4-CT-1999–2001) on molecular microbiologyas an innovative conservation strategy for indoor andoutdoor cultural assets Newsletters 6, 8–10, 2002.Online at /http://www.geomic.uni-oldenburg.de/projekte/coalition/S.