Biofilms on murals: a habitat of novel bacterial species

It is widely acknowledged that micro-organisms can be responsible for the destruction of paintings. For this reason, restoration efforts do not always achieve the expected result and sometimes even accelerate the deterioration process. Ideally, restoration projects should thus include an inventory of the micro-organisms associated with the damage of art. However, such data are not obtained by simple diagnostic methods. To characterise bacteria reliably, a polyphasic study, including genomic techniques, should be conducted. In the past, most studies on biodeterioration have focused on understanding the deterioration potential of the microbial community, so the characterisation of the bacterial community associated with damage has often been unsatisfactory as it yielded data that were not comparable with other studies. There was thus a clear

need for characterisation studies of a large group of isolates. Such a study was performed at the Laboratory of Microbiology at Ghent University in Belgium with material from 3 severely deteriorated mural painting sites: The necropolis of Carmona in Spain with Roman mural paintings (1st-2nd century), the Chapel of Saint Catherine in Herberstein Castle in Austria with medieval mural paintings (14th century) and the Church of Greene-Kreiensen in Germany with late renaissance mural paintings (16th century). Eleven samples were taken to study the bacterial community and 429 strains were isolated. These isolates were studied by fatty acid methyl ester (FAME) gas chromatography (1). The identification of the main clusters at the generic level (Bacillus, Paenibacillus, Micrococcus and Arthrobacter) was in accordance with previous studies. However, in general, FAME analysis did not result in clear-cut identifications at the species level and 46 representative strains were chosen for further characterisation by 16S rDNA sequencing (2). Almost half of the sequences showed less than 97% similarity with the species in the EMBL, thus excluding their allocation to any of the species covered. 16S rDNA sequencing identified the majority of the strains at the genus level as belonging to Bacillus and related genera (Paenibacillus, Virgibacillus), Arthrobacter, Brachybacterium, Brevibacterium, Halomonas, Methylobacterium, Micrococcus, Nocardioides, Sphingomonas, Staphylococcus and Streptomyces. Additional isolates were further studied by rep-PCR genomic fingerprinting, partial 16S rDNA sequencing and DNA:DNA reassociation experiments. The results of this study demonstrated the existence of novel genospecies and further chemotaxonomic and phenotypic analysis allowed description of seven novel species: Brachybacterium sacelli and Brachybacterium fresconis (3), Halomonas muralis (4), Bacillus decolorationis (5), and Virgibacillus carmonensis, Virgibacillus necropolis and Virgibacillus picturae (6). Type strains and additional representative strains are deposited at the BCCM™/LMG Bacteria Collection. The finding and description of novel bacterial species associated with damage on mural paintings is clear evidence that research on biodeterioration of art needs the input of taxonomists in order to improve treatment of these art objects.