The msa mutant has a weak biofilm defect
Prior studies have shown that the modulator of SarA gene (msa
) is required for full expression of SarA, which in turn is essential for biofilm formation [4
]. To examine the role of msa
in biofilm formation, we generated an msa
mutant in the methicillin resistant S. aureus
(MRSA) strain COL. S. aureus
COL was chosen for study because it forms a biofilm in vitro
and is virulent in animal models of endocarditis [9
]. We confirmed the msa
mutation, and its effect on sarA
, by measuring transcription levels of msa
in the wild-type COL strain, the msa
mutant, and the complemented msa
mutant by real-time quantitative PCR (RT-qPCR). As expected, the msa
mutant showed no detectable expression of msa
in planktonic cultures or biofilm (Table ). Additionally, transcription of sarA
was reduced at least five-fold in these cultures in both planktonic cultures and biofilm (Table ). These results are consistent with finding from our previous study [8
] and show that msa
is a positive modulator of sarA
during planktonic growth as well as biofilm in strain COL.
Relative expression of msa and sarA in the msa mutant.
The biofilm forming capacity of the wild-type COL strain, the msa mutant, and the complemented msa mutant were examined by microtiter plate assay. Biofilm formation was observed in at 6, 12, and 24 h post-inoculation in the wild-type COL strain and the complemented msa mutant microtiter plates (Fig. ). Conversely, there was no evidence of biofilm formation for the msa mutant at 6 or 12 h post-inoculation. At 24 hours post-inoculation, the msa mutant formed a biofilm that appeared similar to that of the wild-type COL strain (Fig. ). These results were reproduced at least three times and were confirmed in subsequent experiments, suggesting that the msa mutant has a defect in biofilm growth under steady-state conditions.
Figure 1 Biofilm formation in the msa mutant in microtiter plates. The wild type strain COL, the msa mutant and the complemented msa mutant were grown in TSB supplemented with NaCl and glucose. Cultures were incubated for 6, 12 and 24 hours in the wells of microtiter (more ...)
To further examine this phenotype, we used flow cells to test the ability of the msa mutant to form biofilm under shear forces. Three flow cells coated with human plasma were each inoculated with the wild-type COL strain, the msa mutant, and the complemented msa mutant, and then monitored continuously for biofilm formation while media flowed through the chamber. Wild-type COL strain and the complemented msa mutant formed robust and mature biofilms by 12 hours post-inoculation (Fig. ). The msa mutant, however, failed to form a robust and uniform biofilm within the flow cell in the first 12 hours (Fig. ). The gross morphology of the partial biofilm formed by the msa mutant was similar to those formed by the wild type strain and the complemented mutant (Fig. ). However, while the biofilms formed by the wild-type and the complemented msa mutant persisted for up to 36 h before sloughing off, the biofilm formed by the msa mutant rapidly disintegrated (Fig. ). These results were confirmed in three independent experiments, indicating that the msa mutant is defective in its ability to form mature biofilms.
Figure 2 Biofilm formation in the msa mutant in flow cells. The wild type strain COL, the msa mutant and the complemented msa mutant were used to inoculate flow cells. TSB supplemented with NaCl and glucose was provided at a flow rate of 0.5 ml/minute. Biofilm (more ...)
Since expression of sarA is essential to biofilm formation, we wanted to examine the expression of sarA in cells forming a biofilm. We harvested biofilm from flow cells inoculated with msa mutant and wild-type COL at 24 h post-inoculation and measured the expression of sarA by RT-qPCR. We found that sarA expression levels were significantly reduced in the msa mutant, suggesting that msa is required for full expression of sarA in biofilm (Table ). It is possible that the weak biofilm formation phenotype of the msa mutant could be due to a reduction in sarA expression.
Biofilm formation is a complex process that generally involves three stages: (1) primary adhesion to surfaces, (2) accumulation of multilayered clusters of cells, and (3) detachment. We carried out experiments to determine which stage of biofilm formation is disrupted in the msa mutant. Using two different adherence assays, we measured the ability of cells to attach to surfaces in the presence of host proteins by coating with human plasma and in their absence by not coating with plasma (Fig. ). We found that the msa mutant had no defect in initial adherence to surfaces. In fact, the msa mutant adhered to surfaces significantly better than the wild-type COL strain especially in the catheter assay where no plasma was used (Fig. ). The complemented msa mutant showed a level of adhesion to surfaces that was similar to wild-type (Fig. ). These results indicated that primary binding to surfaces was not responsible for the biofilm formation defect in the msa mutant. It was important to test initial adherence to surfaces with or without plasma coating because binding of host proteins is a major contributor to primary adhesion. In this case, however, we found that it does not play a role in the biofilm phenotype of the msa mutant.
Figure 3 Initial adherence assays. A. Catheter adherence assay. Standardized overnight cultures of the wild type strain COL, the msa mutant and the complemented msa mutant were incubated with catheters at 37°C for 30 minutes. Results represent the mean (more ...)
A recent study by O'Neill et al. [11
] showed that in addition to primary adherence, fibronectin binding contributes to intercellular accumulation in biofilm. We examined the ability of the msa
mutant to bind the immobilized ligands by coating microtitre wells with fibronectin or fibrinogen and compared the capacity of the wild type COL strain, the msa
mutant and the complemented mutant to bind these host proteins. We found that the wild type strain and the complemented mutant bind both fibronectin and fibrinogen, however, the msa
mutant binds fibronectin but not fibrinogen (data not shown). In an effort to explain these results, we examined the expression of fibronectin-binding protein A (fnbA
) and clumping factor (clfA
) in biofilm (Table ). Consistent with the binding assays, there was no significant difference in expression levels of fnbA
between the three strains, while the expression of clfA
in the msa
mutant was significantly reduced in biofilm. This indicated that the lack of fibrinogen binding by the msa
mutant is primarily due to lack of expression of ClfA.
Relative expression of biofilm-related genes.
There are many other genes known to be involved in biofilm formation. The major autolysin, atl
, has been shown to promote adherence of bacterial cells to solid surfaces [12
]. The Atl homolog of S. epidermidis
, AtlE, has also been shown to play an important role in primary attachment to polystyrene surfaces [14
]. Gene expression studies using RT-qPCR in the msa
mutant revealed that atl
levels were significantly reduced in biofilm (Table ). Therefore, our studies with S. aureus
strain COL msa
mutant have demonstrated that primary adhesion to surfaces with or without a plasma coat does not require full expression of the major autolysin Atl. This is consistent with previous findings in which enhanced biofilm formation occurred in the absence of the major autolysin Atl [15
]. Collectively, these results suggest that the strain COL msa
mutant is not defective in primary adhesion to surfaces, but that the defect manifests in the accumulation stage of biofilm formation.
msa mutant is defective in the accumulation stage of biofilm formation
We investigated the possibility that mutation of the msa
gene causes a growth defect that could explain a weak biofilm. We measured growth rates of the wild-type strain, the msa
mutant, and the complemented msa
mutant in planktonic cultures in TSB and found no significant difference between the msa
mutant and the wild-type strain (data not shown). This was important to verify in order to eliminate the possibility of a growth defect caused by mutation of msa
. We then measured the rate of accumulation of cells within the flow cell system for wild-type and msa
mutant strains. In order to monitor cell deposition, we introduced into our strains plasmid pSB2019 (a kind gift from Dr. Phillip J. Hill; [16
], which carries the constitutively expressed Gfp3), and used confocal microscopy to monitor biofilm formation at regular intervals. Consistent with our previous observations, there was no significant difference in initial adherence to the surface or formation of microcolonies between the msa
mutant and the wild-type. However, the wild-type strain formed a thick biofilm as early as four hours after inoculation, while the msa
mutant failed to develop a multilayered biofilm in the first six hours (Fig. ). We observed a specific absence of biofilm "towers" in the msa
mutant biofilms, as compared to the wild-type biofilms, further suggesting that the defect in biofilm formation in the msa
mutant occurs in the accumulation stage.
Figure 4 Confocal microscopy images of biofilm. The msa mutant and the wild type strain COL were imaged 6 hours post inoculation of flow cells. The panels on the left are an overlay of multiple slices, and the side frames of the panels on the right show the z-stack (more ...)
To further characterize the biofilm formation defect in cells lacking msa, we used RT-qPCR to analyze the expression of icaA (ica operon), arcA (arginine deiminase), tcaR (transcription regulator), atlA (autolysin), alsS (alpha-acetolactate synthase), spxA (transcriptional regulator), genes that are reported to be involved in biofilm development (Table ). Some of these selected loci are regulated by sarA (alsS, atlA, and icaA); the others are not directly associated with the sarA regulon (arcA, spx, and tcaR). Expression levels of these genes were analyzed in biofilm using the wild-type strain COL, the msa mutant, and the complemented msa mutant (Table ). A change in expression level of twofold or higher was considered significant (Table ).
Mutation of msa
resulted in a significant increase in alsS
expression in biofilm. The alsSD
operon encodes acetolactate synthase and an acetolactate decarboxylase. Previous studies have reported that mutation of the alsSD
operon in S. aureus
resulted in a biofilm defect [17
]. The biofilm defect of the alsSD
mutant was attributed to the role of this operon in the production of acetoin from pyruvate [18
]. Acetoin production is necessary for acid tolerance within biofilms [5
]. The effect of over-expression of the alsSD
operon as observed in the msa S. aureus
mutant on biofilm formation is not clear. One might speculate that premature build up acetoin in the medium could signal exhaustion of glucose and lead to detachment of cells from biofilm. Further studies are needed to explore this possibility.
Mutation of msa
resulted in a significant decrease in expression of arcA
in biofilm (Table ). The arcA
gene encodes arginine deiminase which is a member of the arginine deaminase (ADI) pathway. This pathway is used to generate energy using arginine under anaerobic conditions, [21
]. The results of several studies point to the importance of the ADI pathway in biofilm formation and pathogenesis. Some oral bacteria have been shown generate ammonia via the ADI pathway to maintain pH homeostasis in biofilms [24
]. Other studies have shown that the ADI pathway was induced during biofilm formation [5
]. Additionally, bacteria in biofilm selectively utilize six amino acids, including arginine [25
], further demonstrating the importance of the ADI pathway in biofilm formation and pathogenesis and may explain the msa
mutant biofilm phenotype. However, when arcD
was disrupted in the S. aureus
strain UAMS-1, the mutant formed effective biofilms and were as virulent as wild-type in a catheter infection mouse model despite the fact that PIA was significantly reduced [25
]. These discrepant results may be due to differences in strains, growth conditions, or infection model but they clearly indicated the need for more studies in the role of arginine metabolism in biofilm development and pathogenesis.
Biofilm accumulation relies on cell-cell adhesion mediated by the polysaccharide intercellular adhesin (PIA), which is produced by the icaADBC
operon, and was shown to play a major role in biofilm accumulation [26
]. Recent studies, however, have indicated that the icaADBC
operon is not essential for biofilm formation in some strains [5
]. We studied the expression levels of genes encoded by this operon and found that the msa
mutation reduced the expression of icaA
in the post-exponential growth phase of planktonic cultures only (data not shown). In biofilm, however, the expression of icaA
in the mutant was not significantly different from wild-type (Table ). The ica
-dependent pathway is primarily regulated by the icaR
]. When IcaR becomes activated by Spx, PIA levels are reduced [30
]. We found that there was no significant difference in expression of icaR
in the msa
mutant compared to the wild-type (data not shown). This is consistent with findings by Tu Quoc et al. [32
] that some biofilm-defective mutants did not show altered PIA levels. Additionally, O'Neill et al
], recently showed that glucose-induced biofilm formation in MRSA strains is ica
-independent. This is relevant to our data, since the COL strain is a MRSA strain and glucose was added to the culture media to induce biofilm in this study suggesting that msa
is involved in biofilm formation using an ica
-independent mechanism as was previously described in some strains [5
]. Another important locus, sasG
, which is similar to accumulation- associated protein in S. epidermidis
has been shown to play a role in cell-cell adhesion and accumulation of biofilm [35
], however, there was no change in expression of sasG
between the msa
mutant and wild type grown in planktonic cultures as determined by DNA microarray experiments in our lab (data not shown).