Biofilm production regulates horizontal gene

Sarah Wilson, Eryn Bernardy & Brian Hammer, PhD


The waterborne bacterial pathogen Vibrio cholerae utilizes a cell-cell communication system called quorum sensing to coordinate group behavior in both a human host and aquatic environments. Virulence genes like the cholera toxin, biofilm genes for sticky secreted attachment factors, and competence genes for DNA uptake are all regulated through this population density-dependent system. In a human host, both virulence and biofilm genes are repressed late in infection presumably to promote transmission upon exhausting the host's resources. However, in the natural environment, regulation is more complex. Namely, at high cell densities, repression of biofilm production is coordinated with activation of competence genes that can promote horizontal gene transfer (HGT). Based on this model, it was proposed that accumulation of biofilm material on cells could hinder the uptake of extracellular DNA in aquatic settings. In support of this hypothesis, significant decreases were seen in DNA uptake with V. cholerae strains engineered to overproduce biofilm. However, reductions were also observed in strains that produced no biofilms. These results suggest that proper timing of biofilm formation plays an important role in the capacity of V. cholerae to engage in HGT, one mechanism thought to allow this pathogen to rapidly evolve in changing environments.


Many bacteria use a form of chemical communication, termed quorum sensing (QS), to regulate gene expression in response to population density. This is accomplished because the bacteria make and secrete chemical signals called autoinducers (AIs). When the population reaches a high enough density, or a quorum, the bacteria respond to the accumulation of AIs and initiate a signal transduction pathway to activate genes controlling numerous behaviors (Ng and Bassler, 2009). This system can be particularly effective in disease-causing bacteria as it enables the cells to collectively activate virulence factors responsible for the onset of disease within a host in a timed manner. In the waterborne, human pathogen Vibrio cholerae, responsible for the fatal cholera diarrheal disease, QS controls numerous genes that are critical for interaction with a human host and survival in a marine environment. However, quorum sensing alone is not sufficient for V. cholerae to regulate certain genes in aquatic settings. Specifically, to express genes to become naturally competent and take up extracellular DNA, one mechanism of horizontal gene transfer, V. cholerae requires both QS AIs and the presence of chitin, which is a polymer of the sugar N-acetylglucosamine and the material that comprises the shells of many aquatic organisms (Sun et al, 2013).

Horizontal gene transfer (HGT) is the process by which bacteria are able to incorporate foreign genetic material onto their chromosome and acquire new traits to better adapt to their environment. The DNA acquired could include genes for antibiotic resistance, virulence factors, or novel metabolic capabilities that are particularly useful for surviving harsh environments (Sun et al, 2013). QS not only controls natural competence in V. cholerae, but also many other phenotypes including virulence and biofilm formation (Zhu et al 2002; Hammer and Bassler, 2003). While numerous genes are activated and repressed by QS in response to cell density (Ng and Bassler, 2009), it remains poorly understood how this timing contributes to many QS-controlled behaviors.

Previous studies (Ng and Bassler 2009) have shown that at low cell density (LCD), when AI levels are low, the QS pathway results in the expression of virulence genes, such as the ctx gene for the cholera toxin, as well as vps genes which synthesize Vibrio polysaccharides crucial in the production of biofilm (Zhu et al 2002; Hammer and Bassler 2003). Conversely, at high cell density (HCD) in the presence of chitin, when autoinducer levels are high and virulence and biofilm genes are repressed, genes coding for factors comprising an apparatus for taking up extracellular DNA are activated. These genes are regulated by HapR, which acts as both an activator and a repressor of gene expression (Bardill et al, 2011). The proposed explanation for why V. cholerae uses quorum sensing to turn off genes when in a group, in contrast to other pathogens like Pseudomonas aeruginosa that turn on genes when at high densities, is that V. cholerae thrives by entering a human host, replicating, and exiting into the aquatic environment to find a new host (Hammer and Bassler, 2003; Srivastava et al, 2011). Therefore, instead of waiting to attack the host until HCD has been reached as in other species, the downregulation of biofilms and virulence at HCD allow for a timely exit by V. cholerae after maximally utilizing the resources within the human host.

However, distinct from a human host, in aquatic settings where V. cholerae is commonly found, this model for the timing of factors is no longer relevant. V. cholerae communities are typically found on chitinous surfaces such as crab shells (Meibom et al, 2005). HGT can be accomplished by transduction via viral infection, by conjugation where direct contact between two bacterial cells allows for direct passage of genetic material via a sex pilin, and by natural transformation where bacteria acquire "naked" DNA from their environment. Until recently, natural competence by V. cholerae had not been demonstrated, but by studying the natural environment in which V. cholerae commonly resides, it was discovered that chitin itself serves as a signal to induce the natural competence genes of V. cholerae required to take up DNA (Meibom et al, 2005). The presence of chitin induces the expression of the TfoX regulator and as discussed, QS promotes production of HapR. Both of these regulators have been shown to be required for activation of numerous DNA uptake genes including comEA, which encodes a component of a DNA uptake apparatus (Antonova and Hammer, 2011). The impact of biofilms on the process of HGT is not yet understood, though production of biofilm appears deeply intertwined with both pathways leading to competence.

Additionally, bacteria often produce many diguanylate cyclase and phosphodiesterase enzymes which work to produce and degrade c-di-GMP, a secondary signal molecule which promotes the expression of attachment factors and biofilm formation, respectively (Massie et al, 2012; Zhao et al, 2013). HapR accumulation at HCD inhibits c-di-GMP production and the loss of this small molecule prevents the transcriptional activators VpsT and VpsR from inducing expression of the vps genes (Waters et al, 2008; Srivastava et al, 2011). Thus HapR accumultion at HCD leads to repression of biofilm formation concomitant with activation of competence genes for DNA uptake.

The levels of c-di-GMP in the cell can be experimentally controlled by genetic modification to V. cholerae, resulting in alterations in the amount of biofilm produced (Srivastava and Waters, 2012). In this study such manipulations are exploited to test the impact of biofilm accumulation on natural competence for DNA uptake. Innappropriate formation of biofilm was tested to determine whether it can hinder uptake of extracellular DNA and whether the production of biofilm is carefully regulated to maximize the potential for DNA uptake in the natural environment while still allowing for the stability that a biofilm can afford at LCD (Berk et al, 2012). The extent to which biofilm formation is regulated through quorum sensing suggests a deeper involvement of biofilms in the ability of V. cholerae to accomplish HGT, which is thought to play an important role in the survival of bacteria in changing environments. This study contributes to our understanding of how the levels and regulation of biofilms can impact HGT by V. cholerae in marine environments.

Materials and Methods

Figure 1     Model of the QS-dependent regulation of competence, biofilm, and virulence genes in V. cholerae at A) low and B) high cell densities.

This study tested whether inappropriate biofilm formation can alter the capacity of V. cholerae to engage in transformation, and whether upregulation of competence genes with the downregulation of biofilm production by the signaling network (Figure 1) maximizes the potential for DNA uptake. Therefore, unlike wildtype V. cholerae strains that produce biofilms at LCD (Figure 1A) and then stop expressing them at HCD (Figure 1B),V. cholerae mutant strains that either make no biofilms or make biofilms constitutively at both LCD and HCD were engineered. Specifically, V. cholerae mutants were constructed, biofilm formation was monitored, and the frequency of DNA uptake by V. cholerae was calculated under a variety of different physical and time-dependent conditions.

Genetic engineering of mutant V. cholerae strains

Figure 2     Biofilm assay mean optical densities +/- standard deviation of eight replicates of wild type (WT), ?vpsR, and ?hapR strains containing control and inducible diguanylate cyclase (DGC) overexpression vectors. * = significant difference of transformation frequencies from the wild type (WT).

To uncouple upregulation of competence genes and downregulation of biofilm genes by QS (Figure 1), a V. cholerae vps deletion mutant was created that is unable to form biofilms (?vps) using standard "allelic exchange" methods. A constitutive biofilm-producing strain (?hapR) was also engineered that makes biofilms at LCD and HCD, irrespective of QS function. Specifically, plasmid pCMW75 described previously (Waters et al, 2008) was used, which can be induced with IPTG to express a diguanylate cyclase previously shown to activate vps transcription and promote biofilm formation, independent of QS. The pCMW75 plasmid (labeled as "DGC vector" in Figure 2) was introduced into the various V. cholerae strains tested. An empty vector that does not carry the diguanylate cyclase gene for c-di-GMP production was used as a control. Biofilm production was measured by a standard crystal violet staining biofilm assay (Hammer and Bassler, 2003). These strains and methods were used to determine how the different genotypes affect formation of biofilms.

DNA uptake efficiency

The standard assay used to quantify the efficiency of DNA uptake by V. cholerae involves incubating the bacterial cultures in artificial seawater on a chitinous crab shell chip to induce natural competence (Meibom et al, 2005). The bacteria are then incubated for an additional 24 hours with exogenous DNA (eDNA) marked with the gene for resistance to the antibiotic kanamycin (kanR). Finally, V. cholerae are plated onto LB medium containing or lacking antibiotic. The efficiency of DNA uptake is calculated as the number of colonies that grow on the selective plates (LBkan) divided by the number of colonies on permissive plates (LB). This assay quantifies how effectively bacteria take up DNA, and was used to determine whether deficiency or proficiency in biofilm formation alters this process.

To effectively infect and then be expelled from its human hosts into aquatic environments, V. cholerae utilizes QS to coordinate its virulence genes (Figure 1A), but QS also plays an important role in marine systems, which aid in transmission back into humans. All of the described techniques above were utilized to determine whether the repression of biofilms at the same time as the expression of competence genes is an intentionally coupled and timed switch to maximize DNA uptake in the natural environment of V. cholerae.


Biofilm Assays

Biofilm assays were performed to quantify the average amount of biofilm growth in each strain (Figure 2). Lab wildtype strain C6706 (WT) produced reasonable biofilms as described prior (Hammer and Bassler, 2003), and the addition of IPTG did not effect biofilm levels (Figure 2, bars 1-2). There was a decrease in biofilms for the WT strain with the control vector, but this is likely an effect of the added antibiotic required to maintain the vector, and addition of IPTG had no effect (Figure 2, bars 3-4). The WT strain carrying the DGC vector (pCMW75) was similar to WT when not induced with IPTG, but showed a significant increase in biofilm when IPTG was added (p<0.0001) (Figure 2, bars 5-6). These results confirm that biofilm formation can be enhanced by induction of a DGC enzyme that produces c-di-GMP.

Similar tested were performed with a V. cholerae strain that has a deletion of the vps genes and is unable to produce biofilms (Hammer and Bassler, 2003) Biofilm levels were minimal in the V. cholerae ?vpsR mutant, and were unaltered when carrying vectors or by addition of IPTG (Fig. 2, bars 7-12). Biofilm could not be recovered for ?vpsR strains induced by IPTG as no polysaccharides are able to be produced. These control strains served as an additional measure to be sure there were no other contributors to biofilm growth other than the vps genes.

Finally, a V. cholerae ?hapR strain with a deletion in the gene for the QS regulator HapR was tested that is locked at LCD and constitutively produces biofilms due to the inability to repress VpsT and VpsR (Hammer and Bassler, 2003; Srivastava and Waters, 2012). As expected, biofilm levels were significantly higher in the ?hapR strain than in WT (p<0.0001) and remained at these levels with the addition of IPTG (Figure 2, bars 13-14). There was a decrease in biofilms seen with the addition of the control vector, but again this is likely an effect of the needed antibiotic (Figure 2, bars 15-16). The ?hapR strain with the DGC vector also had significantly higher biofilm levels than WT, but even with the addition of IPTG these levels were no higher than the ?hapR strains without the vector (Figure 2, bars 17-18). These results confirm that the constitutive biofilm producers always had significantly higher biofilm growth regardless of whether they had the overexpression vector or not. This test ensured that the vectors behaved how they were expected to and also did not inhibit cell growth or viability in the process.

DNA Uptake Assays

Figure 2     Chitin assay mean transformation frequencies +/- standard deviation of three replicates of wild type (WT) and ?luxO strains containing control and inducible diguanylate cyclase (DGC) overexpression vectors all grown with IPTG. *= a significant difference of transformation frequencies between strains with the control vector (pEVS141) and strains with the DGC vector (pCMW75).

DNA uptake assays on chitin were performed with triplicate samples of each strain tested, using established methods (Meibom et al, 2005). The WT C6706 V. cholerae strain had a transformation frequency of ~ 1e-5, consistent with previous studies (Antonova and Hammer 2011) (Figure 3, bar 1). The control vector did not significantly change the transformation frequency of the WT strain as expected (Figure 3, bar 2). The WT strain carrying the DGC vector and induced with IPTG had a transformation frequency that was significantly decreased relative to the WT strain (Figure 3, bar 3, p<0.0001). The same pattern was seen in the V. cholerae ?luxO strain, which has a mutation that results in constitutive expression of HCD genes (Figure3, bar 6-8). Because a DluxO strain is effectively constitutive for QS, it displays a higher transformation frequency than WT (Figure 3, bar 6), as shown prior (Meibom et al, 2005). However, like WT, the DNA uptake by the ?luxO mutant was not changed by the control vector (Figure 3, bar 7), and was significantly lowered when the DGC expression vector was present (Figure 3, bar 8, p = 0.0002).

Strains of V. cholerae (tfoX*) can be genetically engineered to express the TfoX regulator irrespective of chitin, by expressing the tfoX gene from a constitutive promoter. It has already been documented that tfoX* strains have higher transformation frequencies than WT (Meibom et al, 2005) and similar results were observed here in an otherwise WT (Figure 3, compare bars 4 and 1) or DluxO (Figure 3, compare bars 9 and 6) background. Production of c-di-GMP from the DGC vector by ITPG induction in these two strains led to a modest (Figure 3, bar 5) to negligible (Figure 3, bar 10) reduction in transformation frequency. These results showed that the overexpression of biofilm production decreases the transformation frequency of three different strains of V. cholerae.

Figure 4     Chitin assay mean transformation frequencies +/- standard deviation of wild type (WT) and ?luxO strains with and without the ?vpsR mutation. * = significant difference of transformation frequencies from the wild type (WT). ** = significant difference from the ?luxO strain.

In the DNA uptake assay in which a biofilm-defective V. cholerae ?vpsR strain was tested alongside the WT C6706 strain, the presence of a constitutive tfoX* allele significantly enhanced transformation frequencies in all backgrounds (Figure 4, bars 2, 4, 6, and 8). However, in the wild type strain lacking tfoX* a significant decrease in transformation frequency was seen when the vpsR gene was deleted (Figure 4, bars 1 and 3, p<0.0001). However, the transformation frequency seen in either the tfoX*, ?vpsR or the ?luxO, ?vpsR double mutants was unaffected by deletion of vpsR (Figure 4, compare bars 4 and 2, and bars 8 and 6). These results indicate that the absence of a biofilm lowers transformation in strains where tfoX expression is chitin-induced, but not in strains in which tfoX is overexpressed irrespective of chitin.

Discussion and Future Work

In the well-studied pathogen Pseudomonas aeruginosa, quorum sensing activates virulence factors and biofilm only when they have reached a high cell density presumably allowing this chronic pathogen to more effectively invade the host cells (Davies et al, 1998). However, in V. cholerae, the opposite is true as the bacteria express virulence factors and biofilm at low cell density. The proposed role of QS in human hosts is that AI accumulation allows V. cholerae to transit out of the host upon effectively reaching a HCD, allowing the pathogen to re-enter the aquatic environment to find a new host (Hammer and Bassler, 2003). This model is based on observations that V. cholerae strains "locked" at HCD make no biofilms and are avirulent in mouse models, while strains "locked" at LCD produce robust biofilms and are virulent (Zhu et al, 2002; Zhu and Mekalanos, 2003). Thus it has been proposed that virulence genes and biofilm producing genes are expressed at LCD when ingested bacteria are presumably entering the small intestine, and then are repressed after replicating to HCD. This concordant regulation of multiple traits through a single QS pathway is thought to provide an elegant mechanism of synchronizing the needed phenotypic changes for a timely exit from the infected host. However, V. cholerae spends the majority of its life in marine environments and the need for this precisely regulated system was unclear in this environment. This study revealed a similarly balanced relationship as the one described above between biofilm production and DNA uptake.

Overproduction of biofilm through the use of an inducible plasmid significantly decreases the transformation frequency of V. cholerae (Figure 3) . However, additional work needs to be done to determine whether this result is due to the accumulation of biofilm material on the outside of the cells impeding DNA uptake or other issues with overexpression of a diguanylate cyclase. For example, it is possible that accumulation of c-di-GMP inside the cells effects DNA uptake in a manner that is unrelated to biofilm production. It is known that c-di-GMP inside V. cholerae cells not only binds to VpsT and VpsR to alter biofilms, but in other bacteria also binds numerous other factors (Hengee, 2009). Even if the effects were due to the accumulation of c-di-GMP, it would be interesting to see if c-di-GMP alters DNA uptake as this has not been documented to date.

It was also seen that V. cholerae mutants (DvpsR) that produce no biofilms have a reduced transformation frequency compared to WT C6706 (Figure 4). Therefore, it seems likely that not only does an overproduction of biofilm affect the ability of V. cholerae to take up extracellular DNA, but minimal biofilm also impairs competence. Hence, it may be beneficial for V. cholerae to tightly regulate the balance of biofilm formation and DNA uptake through the quorum sensing pathway. In future works, directly testing the expression levels of biofilm genes in response to the overexpression plasmids will ensure that the seen effects are not due to other effects of the vector inside the cell. There should also be more work done in co-culture assays where extracellular DNA is provided by donor cells in the assay, to test the effects of biofilm formation when V. cholerae is interacting with other bacteria as in nature, as opposed to artificially adding extracellular DNA. There are still many unanswered questions on how biofilm is able to affect the horizontal gene transfer of V. cholerae. This study is a step towards fully understanding the evolutionary advantages of maintaining the quorum sensing system in nature which, when introduced into human hosts, regulates the virulence of V. cholerae.


  • Antonova, Elena S., & Hammer, Brian K. (2011). Quorum-sensing autoinducer molecules produced by members of a multispecies biofilm promote horizontal gene transfer to Vibrio cholerae. FEMS Microbiology Letters, 322(1), 68-76.
  • Bardill, J. P., X. Zhao & B. K. Hammer, (2011) The Vibrio cholerae quorum sensing response is mediated by Hfq-dependent sRNA/mRNA base pairing interactions. Molecular Microbiology 80, 1381-1394.
  • Berk, Veysel, Fong, Jiunn C. N., Dempsey, Graham T., Develioglu, Omer N., Zhuang, Xiaowei, Liphardt, Jan, Yildiz, Fitnat H., & Chu, Steven. (2012). Molecular Architecture and Assembly Principles of Vibrio cholerae Biofilms. Science, 337(6091), 236-239.
  • Davies, David G., Parsek, Matthew R., Pearson, James P., Iglewski, Barbara H., Costerton, J. W., & Greenberg, E. P. (1998). The Involvement of Cell-to-Cell Signals in the Development of a Bacterial Biofilm. Science, 280(5361), 295-298.
  • Hammer, Brian K., & Bassler, Bonnie L. (2009). Distinct Sensory Pathways in Vibrio cholerae El Tor and Classical Biotypes Modulate Cyclic Dimeric GMP Levels To Control Biofilm Formation. Journal of Bacteriology, 191(1), 169-177.
  • Hammer, Brian K., & Bassler, Bonnie L. (2003). Quorum sensing controls biofilm formation in Vibrio cholerae. Molecular Microbiology, 51(5), 1521-1521.
  • Hengge, Regine. (2009). Principles of c-di-GMP signaling in bacteria. Nat Rev Micro, 7(4), 263-273.
  • Massie, Jonathan P., Reynolds, Evan L., Koestler, Benjamin J., Cong, Jian-Ping, Agostoni, Marco, & Waters, Christopher M. (2012). Quantification of high-specificity cyclic diguanylate signaling. Proceedings of the National Academy of Sciences.
  • Marvig, Rasmus, & Blokesch, Melanie. (2010). Natural transformation of Vibrio cholerae as a tool - Optimizing the procedure. BMC Microbiology, 10(1), 155.
  • Meibom, Karin L., Blokesch, Melanie, Dolganov, Nadia A., Wu, Cheng-Yen, & Schoolnik, Gary K. (2005). Chitin Induces Natural Competence in Vibrio cholerae. Science, 310(5755), 1824-1827.
  • Ng, Wai-Leung, & Bassler, Bonnie L. (2009). Bacterial Quorum-Sensing Network Architectures. Annual Review of Genetics, 43(1), 197-222.
  • Srivastava, D., Harris, R. C., & Waters, C. M. (2011). Integration of cyclic di-GMP and quorum sensing in the control of vpsT and aphA in Vibrio cholerae. J Bacteriol, 193(22), 6331-6341.
  • Srivastava, D., & Waters, C. M. (2012). A tangled web: regulatory connections between quorum sensing and cyclic Di-GMP. J Bacteriol, 194(17), 4485-4493.
  • Sun, Yan, Bernardy, Eryn E., Hammer, Brian K., & Miyashiro, Tim. (2013). Competence and natural transformation in vibrios. Molecular Microbiology, 89(4), 583-595.
  • Waters, C. M., Lu, W., Rabinowitz, J. D., & Bassler, B. L. (2008). Quorum sensing controls biofilm formation in Vibrio cholerae through modulation of cyclic di-GMP levels and repression of vpsT. Journal of bacteriology, 190(7), 2527-2536.
  • Zhao, Xiaonan, Koestler, Benjamin J., Waters, Christopher M., & Hammer, Brian K. (2013). Post-transcriptional activation of a diguanylate cyclase by quorum sensing small RNAs promotes biofilm formation in Vibrio cholerae. Molecular Microbiology, 89(5), 989-1002.
  • Zhu, Jun, & Mekalanos, John J. (2003). Quorum Sensing-Dependent Biofilms Enhance Colonization in Vibrio cholerae. Developmental Cell, 5(4), 647-656.
  • Zhu, Jun, Miller, Melissa B., Vance, Russell E., Dziejman, Michelle, Bassler, Bonnie L., & Mekalanos, John J. (2002). Quorum-sensing regulators control virulence gene expression in Vibrio cholerae. Proceedings of the National Academy of Sciences, 99(5), 3129-3134.