1

AN INVESTIGATION INTO BIODETERIORATION CAUSED BY MICROBIAL COMMUNITIES COLONISING ARTWORKS

IN THREE MALTESE PALAEO-CHRISTIAN CATACOMBS

G. Zammit*1, 2, P. Psaila1, P. Albertano2 1Heritage Malta, Conservation Division, Bighi, Kalkara, Malta

2University of Rome ‘Tor Vergata’, Department of Biology, Rome, Italy *Corresponding author – e-mail address; gabrielle.zammit@gov.mt, tel – +39 3804338612

ABSTRACT The Maltese islands are rich in late Roman and Byzantine catacombs dating back to the third century A.D. These include pagan, Christian and Jewish burial sites, often decorated by wall paintings, ochre inscriptions, frescoes and reliefs. There are about twelve different subterranean burial sites in the village of Rabat alone. Environmental parameters i.e. a high relative humidity, frequent condensation phenomena, stable temperatures, low light intensities and long photoperiods, provide the ideal conditions for the occurrence of biofilms in Maltese catacomb environments.The main aims of this study were to describe, document and characterise a number of alterations present on wall paintings, ochre inscriptions and reliefs in Maltese Palaeo-Christian Catacombs and to investigate the biodeterioration potential of the microbes causing them using a variety of microscopy techniques. The different types of alterations present were described and catalogued in a visual glossary. Their distribution was evaluated and mapped. This seasonal documentation and monitoring was executed using non-invasive techniques. Predominant biofilms were collected using methods that are non-invasive to the substratum, through the use of adhesive tape strips. Only the superficial microbial film was sampled and not the underlying substratum. Microscopic observation of the most extensive biofilms revealed that the predominant groups of microorganisms present were cyanobacteria and microalgae, often associated to chemoorganotrophic bacterial populations. In cases where microorganisms with a high potential for biodeterioration were isolated from the archaeological surface, low-impact invasive microsamples (2x2 mm) were removed in order to investigate biodegradation processes and interactions of the microbial community with the underlying substratum. INTRODUCTION The surviving wall paintings and decorations in subterranean environments around Malta are rather scarce (Buhagiar, 1986, 1993, 1998). These include those at St. Paul’s Catacombs (Zammit, 1923), St Agatha’s Crypt and Catacombs (Camilleri, 2001), San Catald Catacombs, Taċ-Ċgħaqi, Abbatija tad-Dejr (Buhagiar, 1993), all in Rabat, the cave-sanctuary in Mellieħa (Buhagiar, 1987) and the Ħal Saflieni Hypogeum in Paola (Pace, 2000).

Previous studies conducted on wall paintings and decorations in Maltese Palaeo-Christian Catacombs have involved their manufacturing techniques (Muscat, 2005). However, to date, limited systematic studies have been conducted into the biodeterioration in Maltese Palaeo-Christian Catacombs. The only such studies ever conducted in Malta concerned the biodeterioration of works of art in catacombs have been carried out at St. Agatha’s Crypt and Catacombs in Rabat (Zammit et al, 2008, Psaila, 2006). All subterranean sites may be considered as natural ecosystems which favour the growth of microorganisms due to the prevailing environmental factors. Microorganisms adhere and grow on exposed lithic surfaces including works of art. They form biological patinas or biofilms, which consist of different types of microbes, exocellular polymeric substances (EPS) and other particles (Albertano et al., 2005). Understanding the consequence of the growth of biofilms on works of art in subterranean environments is essential for their proper maintenance and conservation. This study was

9th International Conference on NDT of Art, Jerusalem Israel, 25-30 May 2008For more papers of this publication click: www.ndt.net/search/docs.php3?MainSource=65

2

undertaken in order to provide a first understanding of the factors that influence the growth of biofilms and their deteriorative effects on artworks in Maltese Palaeo-Christian Catacombs. The research focused on works of art embellishing three catacombs in Rabat, located on the outskirts of the old Roman capital of Mdina. The alterations present on decorated surfaces in these sites were documented and catalogued. The term ‘alteration’ refers to any deposit on the surface of the paint layer, intonaco or plaster layer, infilling, mortar, stucco and rock support which might consist of biological patinas and any other deposit. St. Agatha’s Complex was one of the preferred sites since it consists of the hypogean environment embellished by a significant number of wall paintings, some dating back to the fourth century A.D., it is unique and historically important and it is also a popular site being visited by a considerable number of people. The Catacombs of St. Paul’s constitute a complex of interconnected, underground Roman cemeteries that were in use up to the 4th century A.D. An imposing hall and chapel act as the centre from which a number of narrow passages lead into a network of galleries lined by rock-cut tombs which display a wide variety and richness of tomb architecture. There are a few surviving murals and ochre inscriptions from late Roman and early Medieval periods. Decorated tablets and reliefs adorn the graves. Abbatija tad-Dejr consists of a cluster of four hypogea cut into Upper Coralline Limestone and joined together by a rectangular forecourt. In the main hypogeum, Hypogeum 1, the portico leads to a large hall containing sixteen free-standing baldacchino tombs. Various features adorn these tombs, amongst these remains of Siculo-Bizantinesque wall paintings, reliefs, remains of marble slabs, pseudo-Corinthian capitals and voluted scrolls. The aims of this study were the following: • To investigate the different types of alteration and their distribution on works of art, as well

as any seasonal change in their consistency and extent. • To investigate causes for biologically induced alterations. • To monitor environmental factors which might support biological growth. • To investigate interactions between biofilms and the underlying works of art, which might

lead to their deterioration.

MATERIALS AND METHODS In Situ Non-Invasive Investigation The wall paintings, frescoes, ochre inscriptions and decorations at St. Agatha’s Crypt and Catacombs, St. Paul’s Catacombs and Abbatija tad-Dejr Catacombs were periodically observed under diffuse and raking light in order to detect any alteration on their surface. Different types of alteration were documented in terms of their morphology (texture and colour). General photographs of all the art works including the surrounding rock support were taken in order to get a general idea of their conservation state. Close-up photos of the different alterations were taken for documentation and monitoring purposes. Mapping In order to visualise the extent and distribution of the alterations observed, each type found covering an area larger than 4cm2 on the substratum was mapped using graphic symbols.

3

Mapping of alterations was conducted manually on site using digital photos taken previously, and subsequently transferred to AutoCAD 2004 (Fig 1). In order to create a link between the mapping and the actual alterations, a visual glossary was compiled (Fig 2). Each alteration was described and photographed by taking one or more close-ups. This exercise proved to be useful as a first step to understand the types of alteration present on the surface of the works of art. Environmental Monitoring Spot measurements of relative humidity (%), temperature (oC) and photosynthetic photon flux density (PPFD, μmol photons m-2 s-1) were recorded at various points around the catacomb sites on various occasions throughout a one-year period using Hobo® data loggers. Sampling A good estimate of the most extensive alterations present on the art works was made possible by their visual observation and mapping over one year. Samples of the predominant alterations were taken from the original paint layer, intonaco and infillings. These included; 1) samples of biofilms which were non-invasive to the underlying substratum, as only the overlying superficial alteration was sampled, and 2) low-impact invasive microsamples (approx 4 mm2), in which a part of the underlying substratum was included with the biofilm sample. For non-invasive sampling, adhesive tape (Fungi-TapeTM) was pressed firmly over the surface of compact alterations. The tape was affixed onto a sterile glass slide (Urzi & De Leo, 2001), and were stored in the dark at 4oC until observation. A sterile scalpel was used to sample thicker alterations. Care was always taken to remove as little of the valuable substratum as possible. The samples were fixed in 2.5% glutaraldehyde in phosphate buffer (PBS pH 7.2). Samples were stored at 4oC.

4

Figure 1. Mapping of different alterations over and around a wall painting at St Agatha’s Crypt using different graphic symbols to represent different alterations.

Figure 2. A sample page from the visual glossary that was compiled in order to

create a link between the alterations observed on site and the seasonal mapping.

0 5 10 15 20cm

5

Microscopy For light and fluorescence microscopy, sections approximately 1x1cm were cut from adhesive tape samples. These were labelled with the following fluorochromes; 3 μg/ml Acridine Orange (AO) for nucleic acids and polysaccharides, 50 μg/ml fluorescein isothiocyanate conjugated to Concanavalin A (FITC–conA) for α-D-mannopyranosyl and α-D-glucopyranosyl residues and 15 μg/ml 4,6-diamidino-2-phenylindole (DAPI) for DNA localisation. Observations were carried out using an Olympus 1X70 Delta Vision System (Applied Precision) equipped with a mercury lamp (103W/2 OSRAM 1XHBO) and 360-440 nm, 490-520 nm and 555-628 nm excitation filters and 457-470 nm, 528-538 nm and 617-673 nm emission filters. A Leica TCR-SP2-AOBS Confocal Laser Scanning Microscope (CLSM) was also used. Wavelengths of the excitation lasers were in the UV (351 nm and 364 nm; Ar) blue (488 nm; Ar) green (514 nm and 543 nm; Ar/HeNe) and red ( 633nm; Ar/HeNe). Images were acquired in the red, blue and green channels simultaneously. For electron microscopy, samples fixed in 2.5% glutaraldehyde were post-fixed in a 1% osmium tetroxide solution, dehydrated in a graded ethanol series, affixed onto clean aluminium stubs using double sided copper tape and gold sputtered. A LEO 1430 Scanning Electron Microscope (SEM), coupled with a secondary electron detector was used to give a detailed and a high magnified view of the surface morphology of gold-sputtered samples. RESULTS AND DISCUSSION Direct visual observation in situ has revealed that the various types of alteration present could be divided into two main types: green and white alterations. The most extensive alterations occurring on works of art in the Maltese Palaeo-Christian Catacombs studied were green and located in proximity to light bulbs which illuminate the decorated areas, frequently throughout the day. Light microscopy observations of the green alterations revealed that they were photoautrtrophic biofilms formed by photoautotrophic cyanobacteria and eukaryotic microalgae, while the white alterations were due to the growth of filamentous bacterial biofilms, i.e. actinomycetes. Fungal hyphae were never observed growing in the most extensive biofilms. As expected, other associated microbial populations, especially of chemoorganotrophic bacteria, were also found within cyanobacterial biofilms (Albertano & Urzì, 1999; Bruno et al., 2006, Zammit et al., 2008). The climatic parameters recorded at the Maltese Palaeo-Christian Catacomb sites investigated showed that relative humidity was always above 90% while temperature ranged seasonally between 17 and 21 oC. During sampling, PPFD available for photosynthesis was lower than 10 µmol photons m-2 s-1. Such stable temperatures and high relative humidities, frequently coupled with the occurrence of water droplets covering the substrata, low light and long photoperiods, selected for a narrow range of phototrophic and chemoorganotrophic taxa in subaerial biofilms occurring in the Maltese Palaeo-Christian catacombs.

6

Figure 3. Alterations sampled from Palaeo-Christian Catacombs in Rabat, Malta (a) Biofilm 1 - a dull green biofilm from the surface of a wall painting dedicated to St Paul at St Agatha’s Crypt (b)Biofilm 2 - close-up

photo of a jade green biofilm growing in close proximity to a light source at St Paul’s Catacombs (c)Biofilm 3 - a 0.5 cm thick patchy patina detaching from the surface of Baldacchino tombs of the Abbatija tad-Dejr

Catacombs. Although the light levels in these Palaeo-Christian Catacomb sites are very low, cyanobacteria exploit this energy to synthesize organic molecules. In fact cyanobacterial biofilms were observed to develop in photic zones under natural light and close to artificial light sources that illuminate the works of art. Cyanobacterial cells are adapted to grow over an extended period at various irradiances and at different wavelengths of light. For example, a larger number of thylakoids have been observed in Leptolyngbya filaments acclimatized to low irradiances (Albertano et al, 2005). Chemoorganotrophic biofilms, on the other hand, are predominant in areas of these hypogean sites which are not illuminated, and which are isolated from sunlight. These organisms might be attracted to organic nutrients available on the rock surface including any debris that collects on the rock surface, remains of past biological attack, as well as active cyanobacterial growth (Groth and Saiz-Jimenez 1999). Low-impact microsamples removed using a fine sterile scalpel gave a better idea of the types of cells present in the biofilm and how these interact with the substratum. On the other hand, adhesive tape sampling, which has the advantage of being non-invasive to the underlying substratum, removes a very thin superficial layer of the biofilm. Thus, it is an ideal mode of sampling for compact films but not for thicker biofilms.

7

Figure 4. Biofilm1 growing on the pink paint layer of a wall painting (a) autofluorescence from cyanobacterial filaments surrounded by a thick sheath. Scale 10 µm (b) the same cyanobacterial filaments protruding over the

paint layer, almost covering it completely, and observed here to grow inside a pore.

Figure 5. Biofilm 2 from St Paul’s Catacombs. (a) Autofluorescence from Fischerella filaments(wider) found growing in the top layer of this biofilm, surrounded by sheaths of Leptolyngbya, some of them devoid of cells.

Scale 10 µm (b) Leptolyngbya filaments were in close contact to the substratum, observed here to cover it completely. Also present were fine actinobacterial filaments and tiny coccal bacteria.

8

Figure 6. Biofilm 3 from Abbatija tad-Dejr Catacombs. (a) dense growth of microalgal cells and some coccal

cyanobacteria. Scale 10 µm (b) Microalgal filaments grew in close contact with the underlying substratum, and also within pre-existing pores and cracks.

Figure 7. 2-D CLSM reconstruction of Biofilms 2 and 3 (a) Phototrophic biofilm 10 µm thick, formed by Fischerella filaments displaced towards the top and closest to the light source, with Leptolyngbya filaments and

chemoorganotrophic bacteria found growing closer to the substratum. (b) Biofilm 3 consisting principally of microalgal cells surrounded by a thick sheath forming characteristic filaments, some coccal cyanobacteria and masses of chemoorganotrophic bacteria found in close association with the phototrophs. The microorganisms

are surrounded by a gelatinous matrix. Biofilm is approximately 5 mm thick. Evidence of biophysical damage was observed in samples colonized by cyanobacterial and microalgal biofilms. Cyanobacterial filaments were observed to grow inside pores of the paint layer (Fig 4). Microalgae adhered closely to particles of the substratum, and grew inside pre-existing pores and cracks (Fig 5). In such cases, the underlying work of art may be physically

9

affected every time the biofilm experiences physical changes (Sterflinger, 2001), especially in the presence of thick biofilms such in the case of Biofim 3 sampled from the Abbatija tad-Dejr Catacombs. This is the case for wet-dry cycles, when the EPS of the sheath and matrix expands and contracts depending on the amount of water available (Kumar, 1999). This, in turn, causes stresses on the underlying substratum, which may lead to cracking, flaking and detachment of the underlying paint and plaster layers (Karbowska-Berent, 2003; Kumar, 1999; Ortega-Calvo, 1991). Moreover, salt crystals were observed through both light and electron microscopy to occur in close proximity to bacterial cells. The composition of these minerals is presently being investigated further. Cyanobacteria are able to extract and mobilise various nutrients for their nutrition, such as calcium and potassium ions present in paint or intonaco layers of wall paintings (Albertano, 2003). Normally these ions are stored inside the cells, or are precipitated or adsorbed on the cyanobacterial sheath (Albertano et al., 2005). The presence of mineral layers around cyanobacterial filaments was observed in some of the biofilms (Figs. 5, 6). The identity of these mineral layers is presently being confirmed through elemental analysis. These species are potentially the most deteriorating to works of art in Maltese Palaeo-Christian catacombs. ACKNOWLEDGEMENTS This work was supported by the COST Action G8 “Non-destructive Analysis and Testing of Museum Objects” Short Term Scientific Mission COST-STSM-G8-1435, the Italian Government through the Italian Cultural Institute in Malta, a Maltese Government Doctoral Scholarship grant MGSS/2006/026 and Heritage Malta. The authors would like to show their appreciation to Palma Mattioli and Carlo Ramoni for providing kind assistance in the acquisition of fluorescence micrographs. Moreover, the authors would like to express their gratitude to Fr Victor Camilleri MSSP, Curator of St. Agatha’s Crypt and Catacombs, Suzannah Depasquale and David Cardona, Curators of St. Paul’s and Abbatija tad-Dejr Catacombs, for their kind cooperation throughout this study. BIBLIOGRAPHY 1. Albertano, P., 2003. Methodological Approaches to the Study of Stone Alteration caused

by Cyanobacterial Biofilms in Hypogean Environments, in Koestler R.J., Koestler V.R., Charola A.E. and Nieto-Fernandez F.E. (eds.) Art, Biology and Conservation: Biodeterioration of Works of Art, The Metropolitan Museum of Art, New York, 302-315.

2. Albertano, P., Bruno, L., Bellezza, S., 2005. New strategy for the monitoring and control of cyanobacterial films on valuable lithic faces. Plant Biosystems 139, 311-322.

3. Albertano, P., Urzì, C., 1999. Structural interactions among epilithic cyanobacteria and heterotrophic microorganisms in Roman hypogea. Microbial Ecology 38, 244-252.

4. Bruno, L., Billi, D., Urzì, C., Albertano, P., 2006. Genetic characterisation of epilithic cyanobacteria and their associated bacteria. Geomicrobiology Journal 23, 293-299.

5. Buhagiar, M., 1986. Late Roman and Byzantine Catacombs and Related Burial Places in the Maltese Islands, Great Britain.

6. Buhagiar, M., 1987. The Iconography of the Maltese Islands 1400-1900 Painting, Valletta.

7. Buhagiar, M., 1993. The Maltese Paleochristian Hypogea – A Reassessment of the Archaeological, Iconographic and Epigraphic Source Material, Malta.

8. Buhagiar, M., 1998. The Iconography of the Maltese Rock-Tombs Punico-Hellenistic, Paleochristian and Byzantine Vol. XII No. 3, in Melita Historica Vol. 12, Qormi.

10

Back to Top

9. Camilleri, V. J., 2001. Saint Agatha An Archaeological Study of the Ancient Monuments at St. Agatha’s Building Complex: Crypt, Catacombs, Church and Museum, Marsa.

10. Groth, I. and Saiz-Jimenez, C. 1999. Actinomycetes In Hypogean Environments. Geomicrobiology Journal 16, 1-8.

11. Karbowska-Berent, J., 2003. Microbiodeterioration of Mural Paintings: A Review, in Koestler R.J., Koestler V.R., Charola A.E. and Nieto-Fernandez F.E. (eds.) Art, Biology and Conservation: Biodeterioration of Works of Art, The Metropolitan Museum of Art, New York, 266-299.

12. Kumar, R., Kumar, A. V., 1999. Biodeterioration of Stone in Tropical Environments – An Overview, The Getty Conservation Institute, USA.

13. Muscat, F., 2005. The Wall Paintings at St. Agatha’s Crypt: A Study of the Manufacturing Techniques and the Past Interventions. Unpublished dissertation, B. Cons. (Hons), Malta.

14. Ortega-Calvo, J. J., Hernandez-Marine, M., Saiz-Jimenez, C., 1991. Mechanical deterioration of building stones by cyanobacteria and algae, in Biodeterioration and Biodegradation 8, 392-394.

15. Pace, A., 2000. The Prehistoric Hypogeum at Ħal Saflieni, in The Ħal Saflieni 4000 BC – 2000 AD, Malta, 5-21.

16. Psaila, R., 2006. An Investigation into the Biodeterioration of the Wall Paintings at St. Agatha’s Crypt and Catacombs, Rabat. Unpublished dissertation, B. Cons. (Hons), Malta.

17. Sterflinger, K., 2001. Monument as Microbial Environment, in Science and technologies of the materials and of the environment for the protection of stained glass and stone monuments, European Communities, Luxemburg, 147-154.

18. Zammit, T., 1923. The St. Paul’s Catacombs and Other Rock-Cut Tombs in Malta, Valletta.

19. Zammit, G., De Leo, F., Urzì, C., Albertano, P., 2008. A non-invasive approach to the polyphasic study of biodeteriogenic biofilms at St Agatha Crypt and Catacombs at Rabat, Malta in Science and Cultural Heritage in the Mediterranean Area – Diagnostics and Conservation Experiences and Proposals for a Risk Map, Conference Proceedings, Palermo, Italy, 18-21 October 2007, in press.