Environmental stress perception activates structural remodeling of extant Streptococcus mutans biofilms

Abstract

Transcription regulators from the LexA-like Protein Superfamily control a highly diverse assortment of genetic pathways in response to environmental stress. All characterized members of this family modulate their functionality and stability via a strict coordination with the coprotease function of RecA. Using the LexA-like protein IrvR from Streptococcus mutans, we demonstrate an exception to the RecA paradigm and illustrate how this evolutionary innovation has been coopted to diversify the stress responsiveness of S. mutans biofilms. Using a combination of genetics and biophysical measurements, we demonstrate how non-SOS stresses and SOS stresses each trigger separate regulatory mechanisms that stimulate production of a surface lectin responsible for remodeling the viscoelastic properties of extant biofilms during episodes of environmental stress. These studies demonstrate how changes in the external environment or even anti-biofilm therapeutic agents can activate biofilm-specific adaptive mechanisms responsible for bolstering the integrity of established biofilm communities. Such changes in biofilm community structure are likely to play central roles in the notorious recalcitrance of biofilm infections.

Fig. 1. IrvR is RecA-independent and autocleaves constitutively.

a DDAG assay of the ldh-irvR strain and its derivatives performed in the absence of environmental stress. The strains from left to right are: parent ldh-irvR (WT), recA deletion mutant (ΔrecA), irvA deletion mutant (ΔirvA), and recA/irvA double deletion mutant (ΔrecA ΔirvA). b The same strains were assayed in the presence of xylitol stress. c Western blot analysis of the ldh-irvR strain and its derivatives performed in the absence of environmental stress. The ldh-irvR strain was engineered to express both a C-terminal FLAG tag as well as an internal FLAG tag, which supports the detection of the full-length protein (Full), N-terminal autocleavage fragment (NTD), and C-terminal autocleavage fragment (CTD). The strains from left to right are: parent ldh-irvR (WT), recA deletion mutant (ΔrecA), irvA deletion mutant (ΔirvA), recA/irvA double deletion mutant (ΔrecA ΔirvA), and ldh-irvR strain lacking the FLAG epitope (–ctrl). The bottom panel is an HA-tagged lactate dehydrogenase loading control. d The same strains were assayed in the presence of xylitol stress. See also Supplementary Fig. 1.

Fig. 2. IrvR has the characteristic features of a LexA-like protein.

a Schematic map of the IrvR protein. The numbered red hash marks indicate the corresponding amino acid positions in the protein. The sequence of the ClpXP C-degron is written in blue font and the position of the autocleavage site is indicated by an adjacent dashed black line. The location of the DNA binding domain (DBD) is indicated in orange, while the peptidase S24 domain is shown in green with the positions of its catalytic residues indicated underneath. b DDAG assay of the ldh-irvR strain and its derivatives performed in the absence of environmental stress. The strains from left to right are: parent ldh-irvR (WT), S224A mutant IrvR (S224A), K260A mutant IrvR (K260A), C-degron/autocleavage site mutant IrvR (VRR), and C-degron/autocleavage site + K260A double mutant IrvR (VRR K260A). c The same strains were assayed in the presence of xylitol stress. d Western blot analysis of the ldh-irvR strain and its derivatives performed in the absence of environmental stress. The strains from left to right are: parent ldh-irvR (WT), S224A mutant IrvR (S224A), K260A mutant IrvR (K260A), C-degron/autocleavage site mutant IrvR (VRR), and C-degron/autocleavage site + K260A double mutant IrvR (VRR K260A). The bottom panel is an HA-tagged lactate dehydrogenase loading control. e The same strains were assayed in the presence of xylitol stress. See also Supplementary Fig. 2.

Fig. 3. IrvR contains an internal homodimerization domain.

a S. mutans was engineered to express two copies of irvR, one encoding an N-terminal FLAG tag (R-FLAG) and the other encoding an N-terminal HA tag (R-HA). Homodimerization between the FLAG and HA tagged IrvR proteins was assessed via coimmunoprecipitation. Figure columns labeled FLAG IP and HA IP indicate the antibody-conjugated resins used for immunopurification, while the rows labeled α-FLAG and α-HA indicate the antibodies used for western blot detection. b N-terminal epitope tagged IrvR proteins were assessed for homodimerization via coimmunoprecipitation. The full-length FLAG tagged IrvR (R-FLAG) and Δ66–122 internal deletion FLAG tagged IrvR (Δ66–122-FLAG) were compared for their ability to dimerize with the full-length HA tagged IrvR (R-HA). c Western blot was used to compare the autocleavage abilities of the parent ldh-irvR strain (R) and its Δ66–122 internal deletion mutant (Δ66–122) derivative. The bottom panel is a lactate dehydrogenase loading control. d The ldh-irvR strain encoding an N-terminally FLAG tagged full-length IrvR as well as its internal deletion mutant derivatives were tested via coimmunoprecipitation for their homodimerization abilities with a full-length N-terminally HA tagged IrvR. Strains from top to bottom are: full-length FLAG tagged IrvR (R-FLAG), Δ66–80 internal deletion FLAG tagged IrvR (Δ66–80), Δ81–94 internal deletion FLAG tagged IrvR (Δ81–94), Δ95–108 internal deletion FLAG tagged IrvR (Δ95–108), Δ109–122 internal deletion FLAG tagged IrvR (Δ109–122), Δ66–122 internal deletion FLAG tagged IrvR (Δ66–122), and full-length HA tagged IrvR (R-HA). See also Supplementary Fig. 3.

Fig. 4. GbpC production is controlled by two parallel stress-dependent pathways.

a The DDAG phenotypes of the ldh-irvR strain (WT) and its ΔgbpC derivative (gbpC) were compared in the presence of xylitol stress, caffeine stress, and mitomycin C (MMC) stress. b The DDAG phenotypes of the ldh-irvR strain and its mutant derivatives were compared in the absence of added stress. Strains from left to right are: ldh-irvR parent strain (WT), ΔirvR deletion mutant (irvR), ΔgbpC deletion mutant (gbpC), P80D/D171R mutant RecA (RecA^#), P80D/D171R mutant RecA + ΔirvA deletion mutation (RecA^# irvA), and P80D/D171R mutant RecA + ΔgbpC deletion mutation (RecA^# gbpC). c The expression of the LexA paralog hdiR was assessed via qRT-PCR in both normal and stress growth conditions. Expression values (±s.d.) and statistical comparisons are presented relative to the ldh-irvR parent strain cultured in normal growth conditions, which was arbitrarily assigned an expression value of 1. Bars colored in red indicate samples cultured in normal conditions, whereas green indicates xylitol stress, blue indicates caffeine stress, and purple indicates mitomycin C (MMC) stress. Strains from left to right are: ldh-irvR parent strain (WT), P80D/D171R mutant RecA (RecA^#), ldh-irvR parent strain (WT), ldh-irvR parent strain (WT), ldh-irvR parent strain (WT), ΔrecA deletion mutant (recA), and complemented ΔrecA deletion mutant (Comp). d The ldh-irvR strain was cultured in normal and stress growth conditions and then assayed for irvR transcript abundance via qRT-PCR. Expression values (±s.d.) and statistical comparisons are presented relative to the ldh-irvR parent strain cultured in normal growth conditions, which was arbitrarily assigned an expression value of 1. e The ldh-irvR strain was cultured in normal and stress growth conditions and then assayed for irvA transcript abundance via qRT-PCR. Expression values (±s.d.) and statistical comparisons are presented relative to the ldh-irvR parent strain cultured in normal growth conditions, which was arbitrarily assigned an expression value of 1. f The ldh-irvR strain was assayed via western blot to compare IrvR abundance in normal and stress growth conditions. The bottom panel is a lactate dehydrogenase loading control. Error bars represent the standard deviations from at least three independent experiments. ***P < 0.001 and **P < 0.01, Unpaired two-tailed Student’s t test. See also Supplementary Fig. 4.

Fig. 5. IrvR- and RecA-dependent stresses trigger gbpC expression within flow cell biofilms.

a An S. mutans gbpC-gfp transcription fusion reporter strain and its derivatives were cultured in a drip flow biofilm reactor for 16 h to develop biofilms. The biofilms were subsequently cultured for an additional 4 h ± the indicated environmental stresses before dispersing the cells via sonication and then imaging via differential interference contrast microscopy and epifluorescence microscopy. Fluorescent images were captured with identical camera settings and a 300 ms. exposure time. Strains from top to bottom are: parent gbpC-gfp reporter strain (WT), IrvR K260A mutant (K260A), C-degron/autocleavage site mutant IrvR (VRR), and ΔirvA deletion mutant (irvA). b An identical experimental set up was used to assess the effect of mitomycin C (MMC) stress. Strains from top to bottom are: parent gbpC-gfp reporter strain (WT), ΔrecA deletion mutant (recA), and complemented ΔrecA deletion mutant (Comp). Scalebars indicate 1 µm. See also Supplementary Fig. 5.

Fig. 6. Environmental stress alters the biophysical properties of extant biofilms.

a The ldh-irvR strain and its mutant derivatives were cultured on hydroxyapatite (HA) disks for 16 h in a drip flow biofilm reactor to develop biofilms. The biofilms were subsequently cultured for an additional 4 h ± the indicated environmental stresses before dispersing the cells using identical sonication conditions and then imaging via differential interference contrast microscopy. Red arrows indicate biofilm fragments resistant to sonication. The strains from top to bottom are ldh-irvR parent strain (WT), IrvR K260A mutant (K260A), C-degron/autocleavage site mutant IrvR (VRR), and ΔirvA deletion mutant (irvA). Scalebar indicates 1 µm. b The ldh-irvR parent strain and its mutant derivatives were cultured in static growth conditions for 16 h to develop biofilms directly onto detachable rheometer plates. The biofilms were subsequently cultured for an additional 4 h ± the indicated environmental stresses before attaching the biofilm-coated rheometer plates onto a rheometer and then measuring the storage modulus of the resulting biofilms. Results are presented as the average shear storage modulus values ± s.d. Strains from left to right are: ldh-irvR parent strain (WT), K260A IrvR mutant (K260A), C-degron/autocleavage site mutant IrvR (VRR), ΔirvA deletion mutant (irvA), and ΔrecA deletion mutant (recA). c An identical experimental set up was used to assess the effect of mitomycin C (MMC) stress. Strains from left to right are: ldh-irvR parent strain (WT), ΔrecA deletion mutant (recA), and complemented ΔrecA deletion mutant (Comp). All statistical comparisons were performed relative to the parent ldh-irvR strain cultured in normal growth conditions. **P < 0.01 and *P < 0.05, Unpaired two-tailed Student’s t test. See also Supplementary Fig. 6.

Fig. 7. Model of IrvR interactions and stress-induced biofilm remodeling.

a Schematic map of IrvR. The protein-protein interaction (PPI) domain required for IrvR homodimerization is shown in red, while the autocleavage site is indicated by a dashed black line. b NTD dimerization facilitates DNA binding and the repression of irvA gene expression. In addition, NTD dimerization is essential for autocleavage, presumably supporting a trans cleavage reaction of one or both IrvR monomers. Autocleavage exposes a terminal C-degron (represented by a red asterisk), which targets the IrvR NTD for ClpXP proteolysis. c Illustration of the two parallel pathways controlling GbpC production during environmental stress. In a favorable growth environment, IrvR represses irvA gene expression, whereas RecA exists in its basal nonactivated state. Due to the unusually short half-life of gbpC mRNA, only a modest amount of cell wall anchored GbpC is constitutively produced (represented by red lines extending from the cells). This creates a biofilm with a minimal network of GbpC lectin interactions with the glucan polymers present in the biofilm matrix. The resulting biofilm exhibits a flexible architecture that is readily dispersible. Under conditions of environmental stress, GbpC production is stimulated in situ within biofilms. Depending upon the source of stress, either IrvR production could be inhibited to trigger the IrvR-dependent pathway or RecA could be activated to trigger the RecA-dependent pathway. For the IrvR-dependent pathway, the derepression of irvA results in a substantial stabilization of gbpC mRNA and a concomitant increase in GbpC production. As GbpC concentration increases, biofilm mechanical properties are modified due to the formation of a highly rigid structure supported by an increasingly complex network of GbpC lectin interactions with glucan polymers in the biofilm matrix. Biofilms exposed to environmental stress become progressively more difficult to disperse proportionally with GbpC abundance. Since GbpC production is also proportional to the severity of environmental stress, the internal structure of S. mutans biofilms is able to maintain synchronization with the current environmental conditions.