FadR functions as the master transcriptional regulator of fatty metabolism in many bacterial taxa (Iram and Cronan, 2005; Fujita et al., 2007; Shi et al., 2015). Exogenous LCFAs are first imported across the outer membrane via the FadL transporter, flipped across the inner membrane, and then esterified in the cytoplasm by the thioesterase FadD into long-chain acyl coenzyme A (CoA) thioesters (Fujita et al., 2007). In this activated form, LCFA-CoAs serve as substrates for beta-oxidation and membrane biosynthesis and can be directly sensed by FadR (Aalten et al., 2000). FadR is comprised of an N-terminal helix-turn-helix (HTH) DNA binding domain linked to a C-terminal effector binding domain that binds the CoA moiety of LCFA-CoAs (Aalten et al., 2000). FadR binds its target DNA sequence in its apo form to activate fatty acid biosynthesis genes (fab) and repress fatty acid degradation (fad) genes (Fujita et al., 2007). Upon binding LCFA-CoA, FadR affinity for DNA binding is decreased, thus alleviating repression of fad genes while diminishing activation of fab genes. Thus, FadR links the regulation of fatty acid metabolism with LCFA-CoA intracellular availability.

More recent studies have established a link between FadR and virulence in two attaching/effacing (A/E) pathogens, enterohemorrhagic E. coli (EHEC) and C. rodentium (Pifer et al., 2018; Ellermann et al., 2021). A/E pathogens utilize type 3 secretion systems (T3SS), needle-and-syringe like structures, to translocate their effector proteins directly into host cells to establish replicative niches at the colonic epithelium (Kaper et al., 2004). The genes that encode the T3SS structural components, effectors, and regulators are almost all located within the locus of enterocyte effacement (LEE) pathogenicity island, which is activated by the transcription factor Ler (Furniss and Clements, 2018). Using a screen to identify novel regulators of the LEE, Pifer et al. identified FadR as a putative transcription factor that regulates ler expression and demonstrated that apo-FadR binds within the LEE1 promoter region (Pifer et al., 2018). Subsequent in vitro studies demonstrated that saturated and unsaturated LCFAs decrease LEE activity in a fadR-dependent manner (Ellermann et al., 2021). Biochemical studies further revealed that acyl-CoAs decrease FadR DNA binding at the LEE1 promoter region, which corresponds with reduced LEE expression (Ellermann et al., 2021). Intestinal expression of the LEE is also decreased in a C. rodentium fadR-deficient mutant, resulting in attenuated disease (Pifer et al., 2018). Together, these findings suggest that FadR functions as an activator of ler ( Figure 1A ). Thus, the LEE pathogenicity island in A/E pathogens somehow became integrated into the FadR regulon, which suggests that intracellular LCFA-CoA sensing is critical for appropriate regulation of this virulence program.

The FadR regulon has also been linked to virulence through other less defined mechanisms. In some V. cholerae strains, genetic inactivation of fadR decreases ToxT-dependent virulence through undefined transcriptional and post-translational mechanisms, one of which is linked to the FadR-mediated activation of unsaturated fatty acid biosynthesis by FabA (Kovacikova et al., 2017). Loss of fadR also decreases virulence gene expression in S. Typhimurium via an undefined mechanism (Golubeva et al., 2016). Deletion of fadD attenuates virulence in EHEC, V. cholerae, and S. Typhimurium (Lucas et al., 2000; Ray et al., 2011; Gastón et al., 2013; Reens et al., 2019; Ellermann et al., 2021). Given its function as an LCFA thioesterase, the lack of functional FadD may increase intracellular non-esterified LCFAs, which also act as cytoplasmic anti-virulence signals as described below. In V. cholerae, decreased virulence gene expression in the fadD mutant is specifically associated with impaired activation of TcpP, a transcriptional inducer of toxT-dependent virulence (Ray et al., 2011). Finally, deletion of fadL in V. cholerae, C. rodentium, and S. Typhimurium attenuates pathogen growth within the mouse intestines, suggesting that import of LCFAs modulates infectivity (Reens et al., 2019; Rivera-Chávez and Mekalanos, 2019; Ellermann et al., 2020). However, loss of fadL in S. Typhimurium has also been reported to enhance its competitive advantage within the murine gut (Golubeva et al., 2016). This discrepancy may be explained by the use of different mouse and S. Typhmurium strains in the two studies (Golubeva et al., 2016; Reens et al., 2019). Taken together, the FadR regulon clearly contributes to the virulence potential of several enteric pathogens through mechanisms that in many cases have not been fully elucidated.

V. cholerae initially establishes infection within the small intestines. Following epithelial attachment that is in part facilitated by its toxin co-regulated pilus (Tcp), V. cholera secretes cholera toxin (CT), which causes the characteristic watery diarrhea of cholera (Pierce et al., 1971; Thelin and Taylor, 1996; Kumar et al., 2020). The master virulence regulator ToxT activates CT and Tcp expression by binding toxbox sequences within the promoter regions of the ctxAB and tcp operons, respectively (Prouty et al., 2005; Withey and DiRita, 2006; Childers et al., 2011). The structure of ToxT is comprised of two HTH DNA binding motifs within the C-terminal that is connected by a linker sequence to the regulatory and dimerization domains within the N-terminal (Lowden et al., 2010; Cruite et al., 2019). ToxT dimerization is required for inducing CT and Tcp expression and for full V. cholerae virulence (Prouty et al., 2005; Shakhnovich et al., 2007; Childers et al., 2011).

Several in vitro studies initially linked long chain unsaturated fatty acids (LCUFAs) with the decreased expression of ToxT-regulated genes and consequent virulence activity (Chatterjee et al., 2007; Lowden et al., 2010; Childers et al., 2011; Plecha and Withey, 2015; Withey et al., 2015). In contrast, saturated LCFAs have been reported to either have no effect or decrease CT expression (Chatterjee et al., 2007; Lowden et al., 2010; Childers et al., 2011). Electrophoretic mobility shift assays (EMSAs) later demonstrated that LCUFAs specifically decrease the binding affinity of ToxT to DNA (Lowden et al., 2010; Plecha and Withey, 2015; Withey et al., 2015). Interestingly, the first reported crystal structure of ToxT captured the transcription factor as a monomer bound to the LCUFA cis-palmitoleate (Lowden et al., 2010). A hydrophobic ligand binding pocket was revealed within the N-terminal domain that interfaces the dimerization domain, where the carboxylate head of the LCUFA interacts with residues supplied by the two domains (Lowden et al., 2010). A later crystal structure captured apo-ToxT in a different conformation that favors dimerization (Cruite et al., 2019). These findings, together with supporting ToxT mutational and functional studies, revealed that LCUFAs repress virulence by inhibiting ToxT dimerization, which in turn precludes DNA binding ( Figure 1B ) (Childers et al., 2011; Cruite et al., 2019). Notably, this mechanism of action resembles that of the well-characterized ToxT inhibitor virstatin ( Shakhnovich et al., 2007 ). Taken together, ToxT functions as a cytoplasmic sensor that directly links intracellular LCUFAs with virulence repression in V. cholerae.

Recent studies have reported the rational design of potent LCUFA mimics that inhibit DNA binding by ToxT (Woodbrey et al., 2017; Woodbrey et al., 2018; Markham et al., 2021). Notably, one of these putative anti-virulence inhibitors also decreases V. cholerae colonization in a murine infection model (Woodbrey et al., 2018). However, the evolution of resistance remains a concern especially considering that virstatin-resistant V. cholerae isolates have been reported (Shakhnovich et al., 2007). Some of these isolates encode toxT alleles with altered dimerization domains that confer resistance to virstatin (Shakhnovich et al., 2007). Interestingly, some virstatin-resistant isolates are also impervious to the anti-virulence effects of bile (Prouty et al., 2005). However, it remains to be tested whether these isolates are specifically resistant to LCUFAs present in the bile and/or to LCUFA mimics. More broadly, these studies also raise interesting questions regarding the selective pressures that drove the emergence of virstatin-resistant ToxT alleles in the first place – and whether such pressures may also result in strain-specific responses to LCUFAs.

S. Typhimurium establishes infection within the small intestines and invades enterocytes by using a T3SS secretion encoded within the Salmonella pathogenicity island 1 (SPI-1) (Wallis and Galyov, 2000; Zhou and Galán, 2001; Yip et al., 2005; Ellermeier and Slauch, 2007). SPI-1 genes are directly activated by the transcriptional regulator HilA (Bajaj et al., 1995). The expression of hilA is positively regulated by interactions between the transcriptional regulators HilD, HilC, and RtsA (Ellermeier et al., 2005; Koh-Eun et al., 2020). HilD, which is structurally similar to ToxT, is considered the principal regulator that links environmental conditions to SPI-1 activity, while HilC and RtsA primarily function as signal amplifiers (Ellermeier et al., 2005; Lin et al., 2008; Golubeva et al., 2012; Koh-Eun et al., 2020; Chowdhury et al., 2021).

A recent study demonstrated that luminal LCFAs extracted from murine colonic contents potently inhibit hilA promoter activity (Chowdhury et al., 2021). The diffusible signal factor (DSF) cis-2-hexadecenic acid (c2-HDA), a C16:1 derivative, was among the most abundant molecules recovered from these colonic LCFAs (Chowdhury et al., 2021). Further studies revealed that purified c2-HDA is a strong inhibitor of hilA promoter activity by interfering with HilD binding of DNA (Bosire et al., 2020; Chowdhury et al., 2021). Similarly, common dietary LCFAs such as oleic acid inhibit HilD binding to DNA, thus repressing SPI-1 virulence genes (Golubeva et al., 2016). These findings correspond with the enhanced intestinal fitness of a fadL-deficient mutant in a fadR and beta-oxidation independent manner, which supports the idea that LCFA derivatives may function as anti-virulence signals in the gut via their interactions with HilD (Golubeva et al., 2016; Bosire et al., 2020). In silico modeling and HilD mutational studies identified common residues in HilD that likely interact with the carboxylic heads of LCFAs and DSFs (Chowdhury et al., 2021). Distinct residues in HilD were also identified that uniquely interact with LCFAs or with DSFs (Chowdhury et al., 2021). Notably, unlike dietary LCFAs with a cis-2 saturation, c2-HDA also promotes the proteolytic degradation of HilD and decreases the DNA binding activities of the other two SPI-1 regulators RtsA and HilC (Golubeva et al., 2016; Bosire et al., 2020; Chowdhury et al., 2021; Chowdhury et al., 2021). Finally, c2-HDA also inhibits the promoter activity of the ToxT-regulated ctxAB operon, suggesting that this DSF may also repress virulence in V. cholerae by targeting ToxT activity (Bosire et al., 2020).

DSFs like c2-HDA are produced by several Gammaproteobacteria including opportunistic respiratory pathogens such as Pseudomonas aeruginosa and Burkholderia cenocepacia (Boon et al., 2008; Twomey et al., 2012). While DSFs have been detected in the murine gut, their producers remain to be identified (Chowdhury et al., 2021). Given their anti-virulence effects, DSFs may represent a mechanism of colonization resistance that is imparted by some members of the gut microbiome against closely related pathogens. It is also possible that Salmonella and V. cholerae utilize DSFs as signals that inform them of their luminal localization, where the maximal expression of virulence factors that facilitate epithelial colonization is suboptimal. Thus, it will be interesting to determine whether DSFs also modulate the virulence of other enteric pathogens. More broadly, given that different classes of LCFAs inhibit the DNA binding activities of two AraC-like virulence activators – HilD and ToxT – it will be interesting to determine whether LCFAs also inhibit similar virulence regulators in other pathogens to impart their anti-virulence effects.

L. monocytogenes initiates infection in the small intestines by first invading epithelial cells, and then escaping into the cytoplasm to replicate and disseminate into adjacent cells via actin-based motility (Dussurget et al., 2004). PrfA is the master virulence regulator that activates genes required for intracellular replication (Scortti et al., 2007). PrfA is comprised of a C-terminal domain that contains the HTH DNA-binding motif and an N-terminal domain that contains the dimerization domain and a ligand binding pocket (Scortti et al., 2007; Tran et al., 2022). PrfA binds its target DNA sequences called PrfA boxes as a homodimer in a hierarchical fashion (Sheehan et al., 1995). Early virulence genes including hly that are required for pathogen escape into the cytoplasm contain PrfA boxes with high affinity for apo-PrfA within their promoters (Scortti et al., 2007). Virulence genes that are activated later in the intracellular replication cycle contain low affinity PrfA boxes within their promoters (Scortti et al., 2007). Thus, PrfA requires allosteric co-activation by glutathione to bind these low affinity DNA sequences (Reniere et al., 2015; Hall et al., 2016).

Although LCFAs can exert bactericidal effects against L. monocytogenes, some LCFAs also function as anti-virulence signals at subinhibitory concentrations (Lillebæk et al., 2017; Santos et al., 2020). Similar to ToxT in V. cholerae, PrfA-regulated genes are repressed following treatment with different unsaturated, but not saturated, LCFAs (Lillebæk et al., 2017; Santos et al., 2020; Chen et al., 2021). A follow-up study demonstrated that LCUFAs inhibit DNA binding at the hly promoter region by a PrfA allele with enhanced DNA affinity (Santos et al., 2020). The precise mechanism by which LCUFAs interact with PrfA to modulate its DNA binding activity remains to be elucidated ( Figure 1C ). Notably, small molecule inhibitors have been identified that induce a conformational shift that precludes PrfA binding to DNA (Good et al., 2016; Kuleín et al., 2018; Tran et al., 2022). Thus, LCUFAs may also act through a similar mechanism. Collectively, these findings demonstrate that LCFAs attenuate virulence in Gram-positive organisms, in a similar manner to Gram-negative enteric pathogens, by allosterically inhibiting the activities of pro-virulence transcriptional activators.

QseC is a HK receptor that autophosphorylates following interactions with the host neurotransmitters epinephrine and norepinephrine or with the quorum sensing signal autoinducer-3 (Clarke et al., 2006; Hughes et al., 2009). In EHEC and C. rodentium, QseC activation stimulates complex intracellular signaling cascades that ultimately induce ler expression to activate the LEE (Hughes et al., 2009; Moreira et al., 2016). A recent study identified the endocannabinoid 2-AG, an arachidonic acid derivative, as a host-derived signal that attenuates virulence in several enteric pathogens (Ellermann et al., 2020). In EHEC and C. rodentium, 2-AG inhibits the activation of LEE-encoded genes and decreases T3SS activity in a qseC-dependent manner (Ellermann et al., 2020). Autophosphorylation assays further revealed that 2-AG specifically inhibits QseC activation to impart its anti-virulence effects ( Figure 2A ). Supporting these findings, mice with elevated 2-AG levels in the colon were resistant to C. rodentium infection, a protective effect that was lost in mice challenged with a qseC-deficient mutant (Ellermann et al., 2020). 2-AG also attenuates SPI-1 and SPI-2 associated functions in S. Typhimurium through an undefined mechanism (Ellermann et al., 2020). Notably, virulence regulation in S. Typhimurium has also been linked to QseC activity (Moreira et al., 2010; Moreira and Sperandio, 2012). Given the identification of small molecule inhibitors of QseC that attenuate virulence in various pathogens (Rasko et al., 2008; Curtis et al., 2014), it will be interesting to determine whether 2-AG and other host-derived arachidonic acid derivatives act as general anti-virulence signals in pathogens in which QseC activity modulates virulence.

PhoPQ is a TCS that regulates a complex network of genes involved in virulence, cationic antimicrobial peptide (CAMP) resistance, magnesium homeostasis, and outer membrane physiology (Groisman et al., 2021). Accordingly, the HK receptor PhoQ becomes autophosphorylated in response to various environmental signals including moderate acid stress, osmotic stress, low Mg²⁺ availability, and CAMPs (Fields et al., 1986; Fields et al., 1989). More recently, a screen of potential plant-derived inhibitors of PhoQ identified LCUFAs, and not saturated LCFAs, as exogenous signals that inhibit autokinase activity in PhoQ and repress PhoP regulated genes (Gastón et al., 2013; Carabajal et al., 2020). Interestingly, this included conjugated linoleic acids (CLAs), which are common products of bacterial biotransformation reactions that occur in the gut. However, the administration of a CLA supplement failed to attenuate S. Typhimurium gut infection or systemic dissemination in the streptomycin infection model (Carabajal et al., 2020). Subsequent studies demonstrated that the periplasmic sensor domain in PhoQ likely interacts with various LCUFAs to inhibit its autophosphorylation (Carabajal et al., 2020). ( Figure 2B ) Finally, in addition to its phosphorylation by PhoQ, recent work has demonstrated that PhoP DNA binding activity is further modulated by the acetylation of lysine residues within its DNA binding domain (Ren et al., 2016; Li et al., 2021). PhoP acetylation diminishes its DNA binding activity, resulting in attenuated virulence in S. Typhimurium. One mechanism of PhoP acetylation involves the enzymatic transfer of an acetyl moiety from an acetyl-CoA donor (Ren et al., 2016). Notably, acetyl-CoA is a product of LCFA beta-oxidation, thus introducing a further putative mechanism by which LCFA metabolism may modulate bacterial virulence. Taken together, as with QseC, the PhoPQ TCS is expressed in various Gram-negative enteric bacteria and is stimulated by many of the same activating signals (Groisman et al., 2021). Therefore, it will be interesting to investigate whether LCUFAs and post-translational modifications such as acetylation also inhibit the PhoPQ TCS in diverse enteric bacteria, and if so, to delineate the resulting transcriptional and functional effects of such interactions.