All living organisms are constantly challenged by microorganisms and a variety of particles (from air pollution and cellular stresses) that represent a health threat. To counteract this burden, the innate immune system needs to react promptly and adequately to eliminate them, while at the same time to preserve tissue normal function. In the last decade there has been an enormous progress in the study of the molecular mechanisms that allow the host to fight against any antigenic stimuli and to keep internal homeostasis. In general, the innate immune host defense includes three essential sequentially events: (1) microbial recognition, (2) activation of signaling pathways, and (3) effector mechanisms.

Hosts are able to recognize distinct PAMPs present in microorganisms through a wide variety of PRRs. To date, a broad range of PRRs have been reported that include TLRs, NLRs, RLRs, and CLRs (for a complete description, see [1]). The different subcellular localization of PRRs and the broad array of PAMPs that can be recognized by them, allows the host to sense a large number of pathogen bacteria and develop an adequate immune response. Upon recognition of PAMPs by PRRs signal transduction pathways are activated that converge on transcription factors, such as NF-κB, AP-1 or IRFs. Activation of these transcription factors regulates the inflammatory and innate immune response through the expression of proinflammatory mediators and antimicrobial effectors. Since some pathogenic bacteria possess PAMPs that can be simultaneously recognized by several PRRs and this leads to the activation of common transcription factors, it is likely that a collaborative response among different signaling molecules may exert regulatory functions after recognition of pathogenic bacteria. Therefore, the aim of this review is to discuss recent findings on the collaborative activity of TLRs and NLRs in the modulation of the inflammatory response induced by virulence factors of pathogenic bacteria.

Animals, including humans, respond to a wide range of antigenic stimuli in order to preserve their homeostatic conditions [2]. Professional (macrophages, neutrophils, and dendritic cells) and nonprofessional (epithelial and endothelial cells) phagocytes express various PRRs that recognize PAMPs as well as other nonbiological stimuli [3, 4]. Among the most important PAMPs are lipoteichoic acid (LTA) and peptidoglycan (PGN) from Gram-positive bacteria, lipopolysaccharide (LPS) from Gram-negative bacteria, lipoarabinomannan (LAM), lipopeptides, lipoglycans and lipomannans from mycobacteria, glycosylphosphatidylinositol (GPI), anchored lipids from Trypanosoma cruzi, zymosan isolated from yeast, profilin from Toxoplasma gondii, and DNA from bacteria and mycobacteria [5–8]. To date, a broad range of PRRs have been reported, such as TLRs, NLRs, RLRs, and CLRs.

TLRs constitute a family of receptors, with different specificities, 10 in humans and 12 in murines. They are able to recognize a structural diversity of PAMPs like glycans, lipids, proteins, lipoproteins, and nucleic acids [9] and are widely known as key players of the inflammatory and innate immune response [10]. TLRs are expressed in various cellular compartments. TLR1, TLR2, TLR4, TLR5, TLR6, and TLR11 are predominantly expressed on the cell surface, although it has been reported that after ligand binding, TLR2 and TLR4 are internalized to phagosomes. TLR3, TLR7, TLR8, TLR9, and TLR13 are expressed in intracellular vesicles such as the endoplasmic reticulum, endosomes, lysosomes, and endolysosomes and mainly recognize nucleic acid [11, 12]. TLRs are expressed in a broad variety of cells such as dendritic cells, macrophages, neutrophils, monocytes, T and B cells, epithelial cells, endothelial cells, fibroblasts, and even neural cells (Table 1) [13–18]; however, each type of cell contains a specific set of TLRs [19–22].

Monocytes, mature macrophages, and dendritic cells virtually express all TLRs, while TLR4 and TLR5 are also expressed in intestinal epithelium. Moreover, it has been reported that plasmacytoid dendritic cells and mast cells express TLR7 and TLR8, respectively [19, 23–25]. Interestingly, TLR12 colocalize with TLR11 in endoplasmic reticulum of macrophages [26, 27]. TLR signaling has been shown to be involved in several functions in gut, such as epithelial cells proliferation, IgA production, maintenance of tight junction, antimicrobial peptide expression and pathogen bacteria recognition [20, 28–32]. Airway epithelial cells express TLR1, TLR2, TLR3, TLR4, TLR5, and TLR6 [33–35], although TLR4 has a constitutive expression and intracellular localization [36]. However, TLR7 and TLR9 are only expressed in primary airway epithelial cells [21, 37].

TLRs also play a major role in cutaneous host defense against microorganisms [22, 38]. Normal keratinocytes of the epidermis constitutively express TLR1, TLR2, and TLR5, while TLR3 and TLR4 are barely detectable [38, 39]. However, other work showed evidence that keratinocytes express TLR4 at the mRNA and protein level [40, 41]. Latter it has been shown that keratinocytes respond to double-strand RNA (dsRNA) and express a functional TLR3 [42, 43]. Interestingly, TLR6 and TLR9, but not TLR7 and TLR8 [44], are also expressed by keratinocytes [45].

NLRs constitute a family of intracellular receptors that detects PAMPs and endogenous molecules. This family contains 16 members that have been categorized into five structurally different subfamilies: NLRA, with an acidic transactivation domain; NLRB, with a baculovirus inhibitor of apoptosis protein repeat; NLRC, with a CARD domain; NLRP, with a Pyrin domain; and NLRX that contains an uncharacterized domain [46]. Probably, the NLRC receptors NOD1 and NOD2 that recognize intracellular bacterial products, as well as NLRPs that respond to multiple stimuli to form a multiprotein complex termed the NALP-inflammasome, are the best characterized so far [47–49] and will be discussed below.

The expression of NLRs has been described in a variety of cellular types (Table 1), for example, NOD1 is ubiquitously expressed in various cell types such as macrophages, human mononuclear cells, intestinal epithelial cells, and dendritic cells, while NOD2 is expressed at higher levels in phagocytic cells and Paneth cells of the small intestine [50–52]. NOD1 and NOD2 have emerged as key pathogen recognition molecules of the innate immune responses [53, 54]. Since the first report of NOD1 as a receptor of invasive Shigella flexneri [55, 56], other works have shown that NOD1 is the receptor involved in the cytosolic recognition of invasive Gram-negative bacteria or PGN delivered into the epithelial cells through outer membrane vesicles (OMV) derived from these types of bacteria or injected through type three secretion systems (TTSS) [57–60]. The participation of NOD1 in Gram-negative bacteria sensing might represent a selective advantage to the host because these types of bacteria are a common threat of the epithelial cells lining the intestinal mucosa [61]. NOD1 and NOD2 have been shown to detect enteric bacteria such as Shigella, Salmonella, Listeria, Yersinia, pathogenic Escherichia coli strains, and Mycobacterium species [62]. The expression of NOD2 has also been associated with the chronic intestinal inflammation in Crohn's disease, where stimulation with muramyl dipeptide (MDP) seems to play an important role [63].

In contrast to NOD1 that is expressed in a wide range of cells and tissues, the expression of NOD2 seems to be restricted to macrophages, neutrophils, dendritic cells, and lung epithelium [64, 65]. Specifically in the lung, several reports have shown that NOD1 is expressed in epithelial cells, endothelial cells, human airway smooth muscle cells, and leukocytes [66–69] and responds to pathogens such as Chlamydophila pneumoniae, Legionella pneumophila, Klebsiella pneumoniae, Haemophilus influenzae, and Pseudomonas aeruginosa [58, 70–73]. NOD2 has been found mainly in macrophages, neutrophils, and bronchial cells [70, 74–76] and senses Streptococcus pneumoniae, Staphylococcus aureus, E. coli, C. pneumoniae, and M. tuberculosis [77–79].

NLRPs are a subgroup of NLRs constituted by proteins such as NLRP1, NLRP3, NLRP4, NLRP6, NLRP7, and NLRP12 that are involved in the formation of multiprotein complexes termed inflammasomes [80]. These complexes consist of one or two NLR proteins, the adapter molecule apoptosis associated speck-like containing a CARD domain (ASC) and pro-caspase-1 [81]. These inflammasomes might sense several microbial products and a variety of stress and damage associated endogenous signals. Probably the best characterized inflammasome is the one formed by the NLRP3 scaffold, the ASC adaptor and caspase-1 [82], and its expression is induced by inflammatory cytokines and TLR agonists in myeloid cells and human bronchial epithelial cells [82]. As the other inflammasomes, the NLRP3 inflammasome mediates the caspase-1-dependent conversion of pro-IL-1β and pro-IL-18 to IL-1β and IL-18 and are involved in a form of cell death termed pyroptosis [83].

NLRPs respond to a broad variety of bacteria and it has been shown that NLRP3 is activated by the lung pathogenic microorganisms K. pneumoniae, Listeria monocytogenes [84, 85], S. pneumoniae, S. aureus [86], C. pneumoniae [87], M. tuberculosis [88], L. pneumophila [89], influenza virus [90, 91], Porphyromonas gingivalis [92], Aspergillus fumigatus [93], and Aeromonas veronii [94]. NLRP3 seems to be involved in the host defense against the enteric pathogens Citrobacter rodentium and Clostridium difficile in mice [62]; however, this response is far from being fully characterized.

Although NLRP1 was the first NLR described as a part of an inflammasome, its mechanism of activation is not well studied. It is abundantly expressed in lymphocytes, respiratory epithelial cells, and myeloid cells [95, 96]. The best-characterized activator of NLRP1 is the lethal toxin (LT) from Bacillus anthracis [97]; LT activates caspase-1 and induces rapid cell death via NLRP1 [81]. A recent work showed that NLPR1 inflammasome is activated by T. gondii in mice and rats infection models [98]. NLRP7 is only present in human peripheral blood mononuclear cells after LPS and IL-1β stimulation [99]. Despite its function in bacterial infections the experimental evidence indicates that NLRP7 is activated in macrophages by bacterial lipopeptides and Mycoplasma as well as S. aureus infection, leading to formation of an inflammasome [100].

NLRPs also negatively control the inflammatory response by lowering the NF-κB activation and IFNβ production [101, 102]. They regulate autophagy during group A streptococcal infection by interacting with the autophagy regulator Beclin-1 [103]. On the other hand, NLRP6 inhibits NF-κB signaling downstream of TLRs in macrophages in vitro and mouse in vivo [104], which seems to be important to regulate the immune response against components of the gut microflora [105]. Also it has been described that ablation of Nlrp6 confers resistance to L. monocytogenes and Salmonella typhimurium infections [104]. Although the lack of Nlrp6 gene could be beneficial to control infection caused by these pathogens, it must be studied how its deficiency might affect the gut homeostasis. Another member of the NLRs family NLRP12 functions as a negative regulator of inflammation. It is expressed in myeloid cells and its expression is reduced by TNFα and TLR stimulation [106, 107]. However, a recent report has demonstrated that NLRP12 does not significantly contribute to the in vivo host innate immune response to LPS stimulation, K. pneumoniae infection, or M. tuberculosis [108].

Early experiments revealed that flagellin delivered into the cytosol is an important bacterial component for the activation of the NLRC4 inflammasome independently of TLR5 activation [109]. Besides NLRC4 regulates of host defense by activating caspase-1 and IL-1β/IL-18 secretion in macrophages infected with Salmonella enterica serovar Typhimurium, L. pneumophila, and P. aeruginosa. [110, 111]. S. flexneri, a pathogen bacterium lacking flagellum, also induces the activation of the NLRC4 inflammasome through PrgJ, a protein that forms the basal body rod of the type three secretion system [112–114]. Furthermore, NLRC4 protects the gut from chemically induced acute colitis as well as the mortality caused by dissemination of Salmonella beyond the gut [115].

On the other hand, the absent in melanoma-2 (AIM2) protein is a member of the IFI20X/IFI16 (PYHIN) protein family that binds DNA from virus and bacterial pathogens. Upon DNA sensing, AIM2 triggers the assembly of the inflammasome, leading to caspase-1 activation, IL-1β maturation and pyroptotic cell death [116, 117]. Several studies have shown that AIM2 inflammasome is important in the recognition of DNA from pathogen bacteria, such as Francisella tularensis in macrophages [118], Francisella novicida in dendritic cells [119], L. monocytogenes in macrophages [84, 120–122], Mycobacterium sp. in macrophages [123, 124], S. pneumoniae in macrophages [125], and P. gingivalis in gingival epithelial cells [92]. Even AIM2 inflammasome is a critical molecular platform for regulating IL-1β release and survival during acute central nervous system (CNS) S. aureus infection [126].

Initially, sensing of pathogenic bacteria by host activate signaling pathways that turn on mechanisms to kill the microorganism [2]. However, when the infection and inflammatory response have been resolved different mechanisms are launched to repair any tissue damage and return to the basal state [127]. This means that initiation, control, and termination of the inflammatory response and infection must be highly regulated. The inflammatory response is under the control of the NF-κB, AP-1 or IRFs transcription factors, which driving the expression of genes that mediate several processes such as cell proliferation and release of antimicrobial molecules and cytokines that regulate the immune response [128]. In the following sections, the signaling mechanisms activated by TLRs, NLRs, and the collaborative action of both are discussed.

As it was explained earlier TLRs and NLRs regulate the cytokine and chemokine expression in response to bacterial ligands through their respective signaling pathways [72, 168, 169]. It is likely that TLRs and NLRs act in a collaborative/synergistic, complementary, or compensable manner, with the aim to increase the sensitivity to detect and efficiently eliminate pathogenic bacteria. A number of reports that analyze the interaction between TLRs and NLRs have been published (Table 2). In human monocytic THP-1 cells a marked synergistic secretion of IL-8 was induced by synthetic agonists of NOD1/2 and TLR2/4/9. This enhanced IL-8 mRNA expression and NF-κB activation was suppressed in NOD1 and NOD2 genes silenced monocytes [170]. This synergism between NOD1/2 and TLRs was also observed in the production of antimicrobial peptides PGN recognition proteins (PGRPs) and β-defensin 2 in human oral epithelial cells via NF-κB [171]. Interestingly, costimulation of NOD1/2 and TLRs did not have any effect on IL-8 production, which suggests a cell-type specific inflammatory response. Likewise, in human monocytes and dendritic cells the NOD1/2 ligands, DAP and MDP, respectively, exert a synergetic activity with LPS in the expression of proinflammatory cytokines TNF-α, IL-1β, IL-6, and IL-8 [172]. The synergistic effect of MDP and TLR2/3/4 ligands on IL-6 and IL-12p40 expression was also observed in wild type and NOD2^−/− macrophages, via NF-κB, p38 and ERK signaling pathways [173]. Additionally, treatment of human dendritic cell (DC) with MDP and the NOD1 agonist FK565 along with TLR3/4/9 agonists synergistically induced IL-12p70 and INF-γ production [174]. Another NOD1 ligand M-TriDAP markedly increased the response induced by LPS for multiple cytokines such as IL-1α, IL-1β, IL-4, IL-6, GM-CSF, IL-10, and TNF-α. In the same study carried out by van Heel et al. [175] a strong synergistic increase in IL-1β production was observed for TLR1/2, TLR2/6, TLR4, TLR5, and TLR7/8 ligands (Pam₃CysLys₄; MALP2; LPS; flagellin from S. typhimurium and R-848, resp.) combined with M-TriDAP. Assays performed with homozygotic macrophages for the 3020insC mutation and/or TLR2^−/− mice demonstrated that the NOD2 ligand MDP has a synergistic effect on the induction of TNF-α, IL-1β, and IL-10 upon costimulation with specific TLR2 agonists Pam₃Cys and MALP2 [176].

In a more recent work, it was shown that NOD1 and TLR2 cooperate to enhance human CD8 T cells proliferation and expansion, and this cooperating action caused an enhanced secretion of IL-2, IFN-γ, and TNF-α that was related to increased activation of NF-κB, JNK, and p38 signaling pathways [177]. Simultaneous stimulation of monocytes-derived DC with NOD1 and NOD2 ligands combined with TLR7/8 or TLR4 agonists results in highly increased production of IL-1β and IL-23 and expression of the inhibitor suppressor of cytokine signaling 2 (SOCS2), where the NOD1/TLR agonists combination was more relevant for the synergistic activity observed [178]. Altogether, these results clarify the existence of a cooperative action of TLRs and NLRs, which become relevant in the context of infection by bacteria that can be recognized by extra- and intracytosolic receptors. Ferwerda et al. [179] demonstrate that TLR2 and NOD2 are two nonredundant receptors that sense M. tuberculosis. They found a synergistic action between these two receptors for cytokine expression that was lost in cells from individuals homozygous for NOD2 3020insC mutation or macrophages harvested from TLR2^−/− mice (Table 3). The same authors also found that macrophages from TLR2 and TLR4 knockout mice and NOD2 mutant had a decreased production of TNF-α, IL-1β, IL-6, and IL-10, which suggest that collaborative activity of these PRRs is important to balance the amount of pro- and anti-inflammatory response in M. paratuberculosis infection [179]. Evidence on the collaborative activity between TLR2 and NOD2 was obtained by measuring IL-10 secreted from macrophages challenged with S. pneumoniae cell-wall (PnCW) fragments [180]. Interestingly, in this IL-10 production participated the protein adaptors RIP-2 and MyD88, reflecting that both canonical signaling pathways were involved. Infection of mice mesothelial cells (MC) with L. monocytogenes caused an increased production of CXCL1 and CCL2 chemokines, which was notably decreased in MC deficient in NOD1 and RICK [160].

C. pneumoniae, which is a common pathogen that causes pneumonia in humans can be recognized by TLR2 and TLR4 [181], as well as by NOD1/2 [70]. Sensing of C. pneumoniae by these receptors induced a reduction in the expression of IL-6, IL-12p40, and IFN-γ in RIP-2^−/− mice at day 3 after infection compared with wild-type mice. However, at days 5 and 14 after infection, the production of these cytokines was significantly increased in wild-type mice, indicating an initial impaired and delayed kinetics of cytokine production in C. pneumonia-infected RIP-2^−/− mice. Thus, the collaborative activity of both NLRs and TLRs is fundamental for efficient pathogen bacterial clearance. This idea has been supported by later experiments in which NLRC4 was necessary to eliminate L. pneumophila, while TLR5 was necessary to recruit neutrophils [182]. Furthermore, it has been demonstrated that activation of RIP2-, MyD88-, and Naip5/NLRC4-dependent signaling pathways triggers a coordinated and synergistic response that protects the host against lethal infection by L. pneumophila [183]. However, contrasting results were observed in P. aeruginosa macrophage infection, since NLRC4 and caspase-1 activation attenuated the autophagy activated by TLR5 and reduced the type I interferon production [184], besides NLRC4 dampens a beneficial IL-17-mediated antimicrobial host response through IL-18 secretion [185].

On the other hand, in DCs infected with Helicobacter pylori, the cooperative interaction between TLR2 and NOD2 showed to be important for IL-1β production and NLRP3 activation. This cooperative interaction in H. pylori infection was confirmed in IL-1β- and IL-1 receptor-deficient mice in which the clearance of bacteria from the stomach was impaired compared with wild-type mice [186]. Costimulation of BALB/c mice with NOD1 and TLR5 ligands showed to be important for efficient S. enterica clearance and improved mice survival. This effect was accompanied with an increase in IL-5, IL-6, IL-13, IL-21, IL-22, TNF-α, and β-defensin 3 in small intestine [187]. More evidence of NOD2 and TLR2 cooperative action was obtained in DC stimulated with PGN from S. aureus [188]. Analysis of IL-6 and IL-1β production, revealed an additive effect of both receptors in keratinocytes from murine oral epithelium, since in TLR2- or NOD2-deficient keratinocytes the cytokine release was decreased by approximately 50% compared to wild-type cells [189]. Infection of mouse macrophages with Vibrio vulnificus and Vibrio cholerae revealed the activation of caspase-1 via the NLRP3 inflammasome [190]. In this work, experiments made with mice doubly deficient in MyD88 and TRIF (Myd ^−/− Trif ^−/−), demonstrated that NLRP3 activation required NF-κB-dependent TLR stimulation. Altogether, these results indicate that several sensors are necessary to fight against pathogenic bacteria. Moreover, the specific combination of PRRs seems to be coordinated according to the bacteria or PAMPs involved, which subsequently affect the host response by driving collaborative/synergistic activity.

Most studies have focused on the characterization of the inflammatory response triggered by several virulence factors alone. However, it is important to take into account that in physiological conditions the participation of several PRRs that respond to different PAMPs could be more effective for the host to combat infections. Regarding this issue, the experimental evidence accumulated so far has pointed out that the host response against pathogenic bacteria may be the sum of several pathways induced by the recognition of different PAMPs by different PRRs, which in turn trigger and shape the subsequent innate and adaptive immune responses. Although synergistic activity among NLRs and TLRs has been demonstrated, the subjacent mechanisms are not clear. Both receptors are able to activate NF-κB, MAPKs, AP-1, or IRFs signaling pathways; however, the molecules that are involved in the synergistic activation of these pathways have not been identified. Then, for a better understanding of the molecular mechanisms by which NLRs and TLRs collaborate, this synergistic activity will have to be further analyzed. Since TRLs and NLRs play a fundamental role in the eradication of invading pathogen microorganisms through the induction of inflammatory and antimicrobial peptides, this collaborative activity could be exploited to modulate or improve the host response against pathogenic bacteria that causes an exacerbated inflammatory response.

Finally, there is a growing interest in targeting these PRRs for the treatment of sepsis, but also to fight against inflammatory diseases such as cancer, rheumatoid arthritis, inflammatory bowel disease, and systemic lupus erythematosus [191, 192]. Several approaches, like ligand mimetics to activate PRRs and antibodies or molecules to inhibit them, have been used to identify therapeutics targets [193]. However, something that should be taken into account in the development of these strategies is the collaborative activity among different PRRs.