ABSTRACT: Investigation of whole genome gene expression levels of P. gingivalis W83, F. nucleatum DSMZ 25586, S. sanguinis SK36, A. actinomycetemcomitans HK1651, S. mutans UA159 in an 24 h old culture. Additionally, whole genome gene expression level changes of S. mutans UA159 biofilm cells after co-cultivation with S. mitis ATCC 11843 were compared to its single species biofilm growth after 24 h. Aim: Demonstration of the usefulness of a five-species gene expression array. Multiple probes per gene enabled identification of single inter-species cross-hybridizing probes. The deletion of such probes lead almost not to the deletion of the whole gene. This was investigated and confirmed by a two-species biofilm expression analysis: The here described array was used for the identification of genes of S. mutans influenced by the presence of S. mitis. Materials and Methods: P. gingivalis W83, F. nucleatum DSMZ 25586, S. sanguinis SK36, A. actinomycetemcomitans HK1651,and S. mutans UA159 were grown in CDM/succrose or artificial saliva/galactose in a single-species culture for 24 h anaerobically resulting in biofilm structures or monolayers. Total RNA was isolated and used for microarray analysis. Probes were analysed for the presence of biological false positive signals caused by cross-hybridizing probes of one of the other species presented on the chip. Further, a simple procedure was developed for automatical identification and deletion of false positive signals caused by washing artefacts, resulting in a more reliable outcome. In the case of the S. mutans/S. mitis mixed-species biofilm, both species were cultured together for 24 h like previously described. The found gene regulations were verified by RT-PCR. Results: Experiments with cDNA from 24 h old single-species cultures allowed the identification of cross-species hybridizing probes on the array, which can be eliminated in mixed-species experimental settings without the need to exclude the whole genes from the analysis. Between 69 % and almost 100 % represented genomes on this array were found actively transcribed under the mono-species monolayer and biofilm conditions used here. S. mutans / S. mitis co-culture: Physiological investigations revealed an increase in S. mutans biofilm mass with a decrease in pH-value under the influence of S. mitis, thereby confirming previously published data. A stringent fold change cut-off of 2 (p<0.05) identified 19 S. mutans transcripts with increased abundance, and 11 with decreased abundance compared to a S. mutans mono-species biofilm. Many of the genes have previously been found differentially regulated under general and acid stress, thereby confirming the value of this array. Conclusions: Taken together, this new array allows transcriptome studies on multi-species oral biofilm interactions and could become an important asset in future oral biofilm and inhibitor/therapy studies. The chip study used pooled total RNA recovered from three biologically independent mono-species biofilms or adherent cells/monolayers of P. gingivalis W83, F. nucleatum DSMZ 25586, S. sanguinis SK36, A. actinomycetemcomitans HK1651, and S. mutans UA159. In the case of gene expression analysis of S. mutans/S.mitis biofilm structures compared to the single species biofilm of S. mutans three separate single and three separate two-species biofilm cultures were analysed. Each chip measured the expression level of all together 10186 genes (1883 genes of P. gingivalis W83, 1964 genes of F. nucleatum DSMZ 25586, 2244 genes of S. sanguinis SK36, 2168 genes of A. actinomycetemcomitans HK1651, 1927 genes of S. mutans UA159) with up to thirteen 60-mer probes per gene and with a three-fold technical redundancy.
