Air–liquid biofilm formation is largely dependent on Flo11p and seems related to cell lipid content and composition. Here, it is shown that in the presence of cerulenin, a known inhibitor of the fatty acid synthase complex, biofilm formation is inhibited together with FLO11 transcription in a flor strain of Saccharomyces cerevisiae, while the administration of saturated fatty acids to cerulenin-containing medium restores biofilm formation and FLO11 transcription. It is also shown that, in biofilm cells, the FLO11 transcription is accompanied by the transcription of ACC1, ACS1 and INO1 key genes in lipid biosynthesis and that biofilm formation is affected by the lack of inositol in flor medium. These results are compatible with the hypothesis that the air–liquid biofilm formation depends on FLO11 transcription levels as well as on fatty acids biosynthesis.
Air–liquid biofilm formation in Saccharomyces cerevisiae flor strains involves cell rising, buoyancy and hydrophobicity. Cell rising and hydrophobicity are largely dependent on Flo11p, a glycosyl-phosphatidylinositol (GPI)-anchored protein (Zara et al., 2005; Ishigami et al., 2006). In particular, FLO11 gene length and transcriptional levels influence biofilm formation in S. cerevisiae flor strains (Zara et al., 2009). Cell buoyancy seems to depend on lipid content and composition, as flor strains display both chain length of fatty acid residues and unsaturation levels that are higher than those shown by non-film-forming strains of S. cerevisiae (Farris et al., 1993). However, in spite of these evidences, the involvement of fatty acid biosynthesis in the development of the air–liquid biofilm still needs to be investigated.
For this purpose, biofilm formation and the transcription levels of FLO11 were analysed in the flor strain M25 growing in flor medium (FM), in FM added with cerulenin (FMC) and in FMC supplemented with fatty acid (FMCFA) (Fig. 1a). Cerulenin is an antibiotic that by inhibiting de-novo fatty acid biosynthesis (Nomura et al., 1972) may provide further information regarding the impact of fatty acid biosynthesis on air–liquid biofilm formation. Cerulenin addition to FM strongly inhibited biofilm formation and resulted in a dramatic reduction in FLO11 transcription levels (Fig. 1a and b). Fatty acid supplementation to FMC (FMCFA), while restoring biofilm formation and FLO11 transcription levels, was not able to fully restore cell viability of M25 (Fig. 1a and b). Thus, the inability of strain M25 to form a biofilm in FMC, and the recovery of biofilm formation in FMCFA, is not related to the observed decreases in cell viability but supports a role for fatty acids biosynthesis in biofilm formation in flor strains.
These results are in agreement with the involvement of the fatty acid synthase complex in the formation of the biofilm observed in other yeast species (Chayakulkeeree et al., 2007; Nguyen et al., 2009). In addition, the decrease in FLO11 transcription levels in FMC and their increase in FMCFA support the hypothesis of a link between fatty acids biosynthesis and the transcriptional regulation of FLO11.
To test this hypothesis, M25 was inoculated in FM and transcription levels of FLO11 and of ACS1 and ACC1, key genes in lipid biosynthesis, were analysed in both cells growing at the air–liquid interface (biofilm cells), and cells growing at the bottom of the beaker (bottom cells). Acs1p produces acetyl-CoA (van den Berg et al., 1996), while Acc1p is the enzyme that converts acetyl-CoA to malonyl CoA, the first step in the biosynthesis of fatty acids (Hasslacher et al., 1993). Results obtained showed that FLO11, ACS1 and ACC1 were up-regulated in biofilm cells in agreement with the hypothesis that FLO11 overexpression is accompanied by the overexpression of genes involved in fatty acid biosynthesis.
Indeed, the transcription of ACC1 in biofilm cells may indirectly depend on oxygen availability, but also on the availability of inositol through the Ino2p-Ino4p/Opi1p regulon. This regulon is involved in the activation/repression of genes that present the inositol/choline-responsive elements in their promoter (ICRE motifs, also known as UASINO) (Wagner et al., 1999). The decrease in the intracellular concentration of inositol activates the expression of ICRE regulated genes, such as ACC1 and INO1, which are involved in lipid biosynthesis. As FLO11 transcription is also subject to Opi1p regulation (Reynolds, 2006), a role for inositol in air–liquid biofilm formation was hypothesized. In particular, we postulated that in biofilm cells overexpressing FLO11, inositol is used as a precursor of phosphatidylinositol, which is required for the assembly of the GPI-anchor of Flo11p (Chen et al., 2008; Bethea et al., 2010). To assess this hypothesis, the transcription levels of three key genes in inositol biosynthesis and uptake, namely INO1, ITR1 and PIS1, were evaluated in biofilm and bottom cells. INO1 encodes the inositol-3-phosphate synthase necessary for inositol biosynthesis (Majumder et al., 1997). ITR1 encodes a myo-inositol transporter that influence uptake and intracellular concentration of inositol (Lai & McGraw, 1994), and PIS1 is required for the biosynthesis of phosphatidylinositol (Nikawa & Yamashita, 1997).
The data showed that the transcriptional levels of INO1 were higher in biofilm than in bottom cells (Fig. 1c). This increase in INO1 transcription levels suggests a decrease in intracellular availability of inositol in biofilm cells, possibly due to its more intense use for the production of phosphatidylinositol. Accordingly also ITR1 and PIS1 showed higher transcription levels in biofilm than in bottom cells.
Thus, to confirm the importance of inositol in biofilm formation, M25 cells were grown in FM lacking inositol (FM0I) and FM with added inositol (FM2I: FM0I with added 2 mg L−1 inositol that is the amount contained in YNB). It resulted that a complete lack of inositol in the FM (FM0I), although not affecting cell viability (data not shown), resulted in a significant decrease (P < 0.05) in biofilm dry weight as compared to FM2I (0.341 ± 0.093 and 1.127 ± 0.291 mg biofilm dry weight in FM0I and FM2I, respectively).
In summary, the data presented here support the hypothesis that air–liquid biofilm formation depends on FLO11 transcription levels as well as on fatty acids biosynthesis. According to this model, inositol availability affects biofilm formation possibly due to its key role in the assembly of GPI-anchor of Flo11p.
G.Z. received a fellowship from Fondazione Banco di Sardegna. This work was partially supported by MURST PRIN Anno 2003 – Prot. N 2003077174. The authors wish to thank Mauro Fanari for valuable help in collecting the data, and Chris Berrie for critical appraisal of the manuscript.