Leptospirosis: Aspects of Innate Immunity, Immunopathogenesis and Immune Evasion From the Complement System


L. Isaac, Departamento de Imunologia, Instituto de Ciências Biomédicas, Universidade de São Paulo, Av. Prof. Lineu Prestes 1730, São Paulo, SP 05508-900, Brazil. E-mail: louisaac@icb.usp.br


Leptospirosis is a neglected infectious disease caused by spirochetes from the genus Leptospira. It constitutes a major public health problem in developing countries, with outcomes ranging from subclinical infections to fatal pulmonary haemorrhage and Weil′s syndrome. To successfully establish an infection, leptospires bind to extracellular matrix compounds and host cells. The interaction of leptospires with pathogen recognition receptors is a fundamental issue in leptospiral immunity as well as in immunophatology. Pathogenic but not saprophytic leptospires are able to evade the host complement system, circulate in the blood and spread into tissues. The target organs in human leptospirosis include the kidneys and the lungs. The association of an autoimmune process with these pathologies has been explored and diverse mechanisms that permit leptospires to survive in the kidneys of reservoir animals have been proposed. However, despite the intense research aimed at the development of a leptospirosis vaccine supported by the genome sequencing of Leptospira strains, there have been relatively few studies focused on leptospiral immunity. The knowledge of evasion strategies employed by pathogenic leptospires to subvert the immune system is of extreme importance as they may represent targets for the development of new treatments and prophylactic approaches in leptospirosis.


Leptospirosis is a zoonosis caused by pathogenic bacteria from the genus Leptospira [1]. The disease is an important human and veterinary problem around the world. More than 500,000 cases of severe leptospirosis are reported each year, with mortality rates exceeding 10% [2]. A high incidence of the disease is primarily associated with poor communities lacking proper sanitation facilities and flood-prone regions [3]. Leptospirosis is transmitted by the contact of abraded skin or mucous membranes with water or soil contaminated with urine from reservoir animals, such as rodents [4].

In several affected countries, a major effort has been initiated to find a potent and effective immunotherapy. However, the available vaccines neither induce immunological memory nor confer cross-protection against the leptospiral serovars not included in the preparation. Most of the current research is aimed towards the identification and characterization of Leptospira virulence factors. However, we still lack a complete understanding of how this infection progresses, the main characteristics of innate and acquired immunity and how pathogenic Leptospira species are able to escape from the immune system. The knowledge of immune evasion strategies employed by leptospires is of special interest, as they can be targets for the development of new treatments and prophylactic approaches in leptospirosis. Therefore, to encourage future studies, the aim of this review was to summarize and to discuss some aspects of the leptospiral immunity, immunopathogenesis and immune evasion.

Leptospira and leptospirosis

Leptospira microbiology

The genus Leptospira belongs to the order Spirochaetales that includes the genus Borrelia and Treponema, which are also human pathogens causing Lyme disease and syphilis, respectively. Morphologically unique in their coiled shape, spirochetes emerged early in eubacterial evolution. The Leptospira genus includes pathogenic and saprophytic species, being classified into more than 200 serovars considering the structural heterogeneity of the carbohydrate moiety of lipopolysaccharide (LPS) [4]. Besides this antigenic classification, taxonomy based on genetic similarities is currently being employed [5].

Leptospires are thin and highly motile spirochetes with hooked ends (Fig. 1). They have two periplasmic flagella subterminally attached that confer both translational and non-translational forms of movement [4]. Like other spirochetes, leptospires have a distinctive double-membrane architecture, sharing characteristics of both Gram-positive and Gram-negative bacteria. The leptospiral cytoplasmatic membrane is closely associated with a peptidoglycan cell wall, which is overlaid by an outer membrane [6] (Fig. 2). Phospholipids, outer membrane proteins (OMPs) and LPS are the main components of the outer membrane [7]. Leptospiral LPS has a lower endotoxic potential compared with Gram-negative LPS, possibly related to the unusual features of its Lipid A component [4, 8]. To date, according to mutagenesis data, only four virulence factors were described in leptospires: the proteins Loa22 [9], haem oxygenase [10], FliY (flagellar motor switch protein) [11] and LPS [12].

Figure 1.

 Transmission electron micrographs of Leptospira interrogans serovar Pomona. (A) Leptospires are thin, helically coiled bacteria, classed in the order Spirochaetales [4]. These highly motile spirochetes are normally about 0.1–0.2 μm wide and 6-20 μm long. They have two periplasmic flagella and characteristic hooked ends. Original magnification, ×12,000. (B) Immunoelectron microscopic localization of LipL32, an abundant, surface-exposed, outer membrane protein of pathogenic Leptospira. Original magnification, ×20,000. Courtesy of Dr Aurora Cianciarullo and Dr Enéas de Carvalho, Instituto Butantan, São Paulo, Brazil.

Figure 2.

 Schematic depiction of leptospiral membrane architecture. The inner membrane (IM) is closely associated with the peptidoglycan (PG) cell wall, which is overlaid by the outer membrane (OM) [6]. Surface-exposed lipoproteins (LipL32, LigA, LigB and Loa22), the transmembrane outer membrane protein porin L1 (OmpL1), and lipopolysaccharide are among the main components of the outer membrane. For simplicity, TonB-dependent receptor systems, leptospiral lipoprotein export apparatus and endoflagella were omitted.

Clinical presentation of leptospirosis

Pathogenic leptospires persistently colonize the kidneys from reservoir animals, which do not present clinical symptoms, eliminating the bacteria in the urine. Humans are incidental hosts, being susceptible to the disease [4]. Protean clinical presentation is well documented with patients exhibiting either very mild symptoms or subclinical disease (80–90% of infections) or a more severe illness characterized by jaundice, acute renal failure and bleeding (Weil′s disease) or pulmonary haemorrhage syndrome. Typically, leptospirosis follows a biphasic course. Of sudden onset, the septicemic phase lasts 4–7 days during which leptospires can be found in patient’s blood and cerebrospinal fluid. Non-specific symptoms including fever, headache, chills, muscle aches, abdominal pain and conjunctival suffusion are common during the acute phase. These are followed by a 1- to 3-day period of defervescence. In the second stage of the disease, called the immune phase, fever and earlier symptoms recur, and aseptic meningitis may develop. Anti-Leptospira antibodies begin to be produced and the bacteria are excreted in the urine. A subset (10–15%) of patients develops the severe forms of the disease, which may progress very rapidly, leading to death. In these patients, jaundice is primarily attributed to hepatocellular dysfunction rather than fulminant hepatic necrosis. Acute renal failure occurs in 16–40% of cases (for more comprehensive reviews describing the clinical features of leptospirosis see [1, 4, 13]). Leptospirosis-associated severe pulmonary haemorrhagic syndrome has increasingly become recognized as an important manifestation of leptospiral infection [13–16] and may be considered a relevant prognostic factor associated with fatal outcomes in severe leptospirosis [17]. Interestingly, symptomless infection with Leptospira has been reported in endemic areas [13, 18]. This raises the possibility that protective acquired immunity against severe leptospirosis after recurrent infections may be established in those areas [18].

Leptospiral invasion

To establish and maintain the infection, pathogenic leptospires bind to a wide range of extracellular matrix components and other host ligands, including laminin, collagen types I and IV, cellular and plasma fibronectin, fibrinogen, elastin and proteoglycans [19–22]. In addition, L. interrogans are able to bind endothelial cells, monocytes/macrophages, kidney epithelial cells and fibroblast cell line cultures in vitro [23, 24].

It is important to highlight that virulent leptospires, despite being extracellular pathogens, are highly invasive microorganisms and they are able to adhere to and invade host cells more efficiently than the non-virulent and saprophytic strains [24]. Indeed, pathogenic strains have been shown to translocate through polarized MDCK cell monolayers at a rate significantly greater than that of non-pathogenic Leptospira [25]. Moreover, some authors speculate that Leptospira could have a short intracellular phase [26], probably enabling them to escape from complement and antibodies.

Leptospira immunity

The innate immune system constitutes the first line of host defence, playing a crucial role in early recognition and elimination of leptospires. The activation of the alternative pathway of the complement system is one of the most important effector mechanisms during the first hours of infection [27, 28]. A clear illustration of this fact is the observation that saprophytic L. biflexa is killed within a few minutes in the presence of normal human serum in vitro. By contrast, pathogenic Leptospira species are able to survive, and are more resistant to the action of the complement system, especially if they are virulent [27, 28]. The term pathogenic refers to genotypic properties of leptospires that may be expressed or not whereas the term virulent refers to phenotypic characteristics by which means pathological changes may be affected in the host. For example, L. interrogans serovar Copenhageni causes severe disease in humans when expressing its virulence mechanisms, which can be lost to a variable extent, in culture, thus becoming relatively avirulent, but not non-pathogenic. By contrast, L. biflexa serovar Patoc does not cause disease in any known host, thus receiving the classification of non-pathogenic. The property of virulence may be re-established by passage through a suitable host system [4].

As leptospires are extracellular pathogens, the acquired immune response depends on the production of antibodies and the activation of the classical pathway of the complement system. In leptospirosis, most of the specific antibodies produced are against the LPS. As a consequence, passive immunization with polyclonal or monoclonal anti-LPS antibodies is able to confer protection against the disease [29]. In many experimental models, it became evident that phagocytosis of Leptospira by neutrophils and macrophages is only effective if this pathogen is opsonized by specific IgG [30–32]. Besides opsonization, these antibodies may agglutinate leptospires [34] and activate the classical pathway of the complement system. Opsonins such as iC3b generated after complement activation may also be important to enhance phagocytosis as anti-complement receptor type 3 antibodies block the adhesion of different strains of Leptospira to polymorphonuclear leucocytes [33].

Humoral-mediated immunity is of utmost importance to confer protection against leptospirosis in humans, dogs, pigs, guinea pigs and hamsters [4]. However, the role of cell-mediated immunity remains poorly understood. In cattle, the development of cellular immune responses has been associated with protection against leptospirosis. Indeed, vaccination of cattle with a killed L. borgpetersenii preparation results in INF-γ production and proliferation of both CD4+αβ and WC1+γδ T cells following peripheral blood mononuclear cell (PBMC) stimulation with Leptospira [35, 36]. In humans, when PBMC from healthy donors or patients recovered from leptospirosis are stimulated with Leptospira, an expansion of both αβ and γδ T lymphocytes occurs [37]. However, in patients with acute leptospirosis, an increased number of peripheral blood γδ T cells have been observed [38]. Therefore, it seems that not only αβ but also γδ T lymphocytes are important in the leptospiral immune response.

Interestingly, although pathogenic Leptospira is not considered a typical intracellular pathogen, it was reported that L. interrogans may be able to escape from the phagolysossome to the cytosol of a human macrophage cell line (THP-1) [39]. Therefore, leptospiral peptides could be complexed with MHC class I molecules, and be later presented to CD8+ T lymphocytes. Indeed, CD8+ T lymphocytes specific to peptides derived from leptospiral immunoglobulin-like (Lig) A protein were identified in human patients [40].

Additional studies are required to explore the phenotypic and functional characteristics of T cell populations in leptospirosis. Such studies may point towards the improvement of leptospirosis therapies and vaccine formulations.

Leptospires and pathogen recognition receptors

Following invasion, leptospires can be detected by the immune system through the pathogen recognition receptors (PRRs), which recognize molecular motifs known as pathogen-associated molecular patterns. PRRs can be divided into endocytic, signalling and secreted receptors. Two main families of signalling PRRs play a decisive role in pathogens’ recognition, the Toll-like receptors (TLRs) and the Nod-like receptors (NLRs) [41].

Among TLRs, TLR2 and TLR4 are the most studied in leptospirosis until now. LPS from Gram-negative bacteria is able to activate TLR4, resulting in a pro-inflammatory cytokine and chemokine dependent response [42]. However, leptospiral LPS, which is less endotoxic than Gram-negative LPS, activates human macrophages through TLR2 instead of TLR4 [43]. This different recognition is attributed to the unusual composition of the leptospiral Lipid A moiety [8], and could be a strategy that pathogenic Leptospira may use to avoid adequate activation of immune cells, contributing to the establishment of the disease in humans.

In mice, which are resistant to leptospirosis, the LPS is recognized by TLR2 and TLR4 [44]. Indeed, both TLR4 and TLR2 stimulation are crucial to control leptospirosis in mice. Double TLR2/TLR4 knockout mice infected with L. interrogans rapidly died from hepatic and renal failure. The severe inflammation observed in the liver and the kidneys of these animals appears to be independent of MyD88, the main adaptor of TLRs [45]. Another spirochete, Borrelia burgdorferi, which causes Lyme disease, also promotes inflammation independent of MyD88 [46]. Therefore, innate immune receptors from the Nod-like family could be responsible for triggering inflammation in response to invasive leptospires in absence of TLR stimulation [45].

At this point, regarding the differences in recognition of leptospires by TLRs in human and murine macrophages, it is relevant to raise some considerations about the mechanisms possibly involved in mice resistance and human susceptibility to leptospirosis. In a human macrophage cell line (THP-1), L. interrogans were visualized in the cytosol. The leptospiral replication inside the human macrophage was followed by induction of apoptosis and bacterial release. By contrast, in a murine macrophage cell line (J774A.1), L. interrogans were localized in the phagolysossome, being efficiently killed by the cell machinery [39]. Moreover, LPS from L. interrogans activates the TLR4-MD2 pathway in murine macrophages but not in humans [44]. The lack of TLR4 activation together with the cytosolic localization of virulent leptospires and macrophage apoptosis induction may contribute to human susceptibility to leptospirosis in contrast to mouse resistance.

Other receptors may mediate the binding of pathogens to host cells before their ingestion by phagocytic cells. Dendritic cells (DCs) express diverse PRRs, including C-type lectins as DC-specific intercellular adhesion molecule-3-Grabbing Non-integrin (DC-SIGN), an endocytic receptor that recognizes high-mannose glycans and fucose-containing antigens on pathogens’ surfaces [47]. L. interrogans serovar Autumnalis isolated from leptospirosis patients and both virulent and avirulent strains of L. interrogans serovar Pyrogenes isolated from rat, contained high mannose components on their surfaces. These L. interrogans serovars were able to bind DC-SIGN in DCs, inducing their maturation and production of IL-12p70 and TNF-α, but reduced secretion of IL-10. Secreted antigens from leptospires also interacted with DC-SIGN, being possibly involved in modulation of bystander DCs. Furthermore, the avirulent serovar Pyrogenes presented a higher DC binding and DC-SIGN affinity than the virulent strain [48]. The lower interaction of virulent strains with DCs could result in a reduced bacterial uptake and antigen presentation to T cells, favouring disease development. Besides endocytosis, DC-SIGN recognition can activate intercellular pathways that cross-talk with TLRs signalling, influencing DCs activation [49]. Considering that immature DCs express TLR2 and that leptospires are able to interact with this receptor in human macrophages [47], it is particularly relevant to investigate the cross-talk between DC-SIGN and TLR pathways in DCs [48]. Therefore, further investigations on DCs activation and antigen presentation to T cells are required to elucidate the establishment of adaptive immune responses in leptospirosis.

Leptospiral phagocytosis and microbicidal activity of phagocytic cells

As polymorphonuclear neutrophils (PMNs) constitute the largest population of intravascular phagocytes, they are expected to play an important role in leptospiral clearance. Relatively few studies have deeply explored the contribution of these leucocytes to eliminate non-pathogenic or pathogenic leptospires during the early phase of infection. The slow growth rate of leptospires in vitro and requirement of opsonic antibodies for an adequate killing of pathogenic strains by neutrophils [30] hampered the scientific community’s perception of the real importance of PMNs in the early control of leptospirosis. Besides these limitations, it was reported that PMNs are able to kill both non-pathogenic and pathogenic strains of Leptospira by oxygen dependent and independent mechanisms. L. biflexa and L. interrogans are both killed by hydrogen peroxide (H2O2) and by primary granules of PMNs in vitro, although pathogenic leptospires are more resistant than non-pathogenic [50]. This could be associated with the expression of catalase (KatE), an enzyme involved in resistance to oxidative killing that is produced only by pathogenic strains [51, 52].

Antimicrobial peptides (AMPs) are small cationic peptides active against bacteria, viruses, fungi and certain parasites. In humans, the most studied AMPs are cathelicidins and defensins, which are produced mainly by neutrophils and epithelial cells. To exert their microbicidal potential, the cationic peptides bind negatively charged molecules on the pathogens’ surface, such as LPS, promoting membrane permeabilization and endotoxin neutralization [53]. L. interrogans are efficiently killed by cathelicidins from sheep and cattle, but the human peptide LL-37 presents a reduced activity [54]. However, human defensins (HNP1, HNP2 and HNP3) were able to successfully kill L. interrogans [55]. In addition, the spirochetes Borrelia and Treponema, which lack LPS, were more resistant to cathelicidins than Leptospira [54]. Future investigations should aim to better characterize the role of AMPs in leptospirosis. It would be interesting to evaluate their interaction with leptospiral outer membrane components, as well as their action on different leptospiral strains (saprophytic, pathogenic virulent and culture attenuated). Physiological concentrations of these peptides in epithelial cells and neutrophils should also be considered in those experiments.

Cytokines and pentraxins in leptospirosis

As there is a diverse array of clinical manifestations in leptospirosis, it would be helpful to determine parameters that could predict the disease outcome. The clinical hallmarks of severe leptospirosis can resemble septic shock, with multi-organ failure, hypotension and death [4]. Therefore, the expression of molecules usually evaluated in the context of sepsis, such as cytokines and pentraxins, has been investigated in leptospirosis.

The pro-inflammatory cytokine TNF-α, which has a prognostic value in sepsis, has been associated with disease severity and poor outcome in leptospirosis patients [56]. However, as TNF-α is an early phase cytokine, its levels are reduced or undetectable with disease progression. At this point, elevated plasma levels of IL-6 and IL-8 have been correlated with patients’ mortality [57].

Pentraxins are a family of proteins usually characterized by a pentameric structure. C-reactive protein (CRP) and serum amyloid P component (SAP) are short pentraxins, and Pentraxin 3 (PTX-3) is a long pentraxin. Both CRP and PTX-3 are acute phase proteins and have been studied as serological markers in sepsis [58]. However, in leptospirosis patients, increasing serum levels of PTX-3 but not CRP were associated with mortality and disease severity [59].

The usefulness of the cytokines and PTX-3 in monitoring severe leptospirosis should be evaluated in large prospective studies to better characterize the prognostic value of these markers. In addition, regarding the high PTX-3 levels in leptospirosis patients [59], it is relevant to raise some considerations about this long pentraxin. Apart from being an acute phase protein, PTX3 can also act as an opsonin and participate in the activation and regulation of the complement system, via binding to C1q, Factor H and L-ficolin [58]. Therefore, the roles of PTX3 in leptospirosis should be explored to better characterize the mechanisms involved in leptospiral immunity.

Genetic polymorphisms and susceptibility to leptospirosis

The severity of leptospirosis depends on the combination of diverse factors including the virulence of the infecting serovar as well as the bacterial load. The role of genetic polymorphisms in the susceptibly to leptospirosis has also been explored. The highly polymorphic human leucocyte-like antigen DQ-6 (HLA-DQ6) was associated with an increased risk of leptospirosis among triathletes who ingested contaminated water. Considering that this association was not allele specific, a possible role of leptospiral superantigens was suggested, as these molecules could bind HLA outside the peptide-binding groove, resulting in a widespread polyclonal T cell activation [60]. Additionally, alleles from HLA-A and HLA-B, as well as single nucleotide polymorphisms in the interleukin (IL)-4 and IL-4Rα genes had significantly higher frequencies in patients with a history of leptospirosis from Terceira Island in the Azores archipelago [61]. The associations encountered between genetic factors and leptospirosis represent an important initial step towards larger cohort studies with application of more stringent statistical tests. Indeed, the identification of genetic susceptibility factors to infectious diseases is quite relevant as they can be used as markers to identify risk populations that could benefit from preventive measures, such as a future vaccine.

Immunopathology in leptospirosis

Immunological aspects of renal disease

The kidneys are target organs in human leptospirosis pathology. The acute phase of the renal disease is characterized by tubule-interstitial nephritis that may progress to fibrosis in chronic stages [62]. Renal epithelial cells express TLRs, including TLR2 and TLR4 [63]. Therefore, the involvement of TLR in leptospirosis renal disease has been explored. It was suggested that pathogenic leptospires induce tubule-interstitial nephritis through the recognition of outer membrane proteins by TLR2 in renal tubule cells [64]. This recognition is followed by activation of nuclear transcription factor κB (NF-κB) and mitogen-activated protein kinase (MAPK) signalling pathways, leading to early inflammation and leucocyte recruitment, respectively [65]. In the chronic stages of the disease, leptospires are able to induce renal fibrosis. Indeed, leptospiral outer membrane proteins induce the accumulation of extracellular matrix components in tubular cells through the activation of transforming growth factor-β/Smad-dependent pathways, contributing to the development of fibrosis in the kidneys [66].

Pulmonary haemorrhage and autoimmune process in leptospirosis

Pulmonary haemorrhage is a serious life threatening disorder and is the major cause of death due to leptospirosis in Brazil [17]. The mechanisms proposed to explain pulmonary pathology are the direct action of toxins and an autoimmune process. The relatively low number of leptospires recovered from lung tissue suggests a possible role of circulating bacterial toxins produced at distant sites, such as the liver [67]. Although specific toxins remain unidentified, a wide range of predicted proteases and haemolysins that could be involved in lung damage are encoded in the genomes of pathogenic Leptospira [52, 68, 69]. Apart from the toxins, the deposition of IgM, IgA, IgG and C3 in the alveolus of patients with pulmonary haemorrhage syndrome and in guinea pigs experimentally infected with clinical isolates has been reported [70, 71]. This deposition of immunoglobulins and complement observed in the alveolus fits with the pattern of Goodpasture’s syndrome. However, it is still not clear whether an autoimmune process occurs in the kidneys, as sera from leptospirosis patients did not recognize the human glomerular basement membrane [72]. Therefore, the development of renal autoimmunity associated with leptospirosis requires more detailed analysis.

Uveitis is an inflammatory process in the eye that can be triggered by autoimmune reactions. Uveitis is frequently observed in Leptospira infected horses and represents a late complication in 40% of human patients [73, 74]. A significant concentration of serovar-specific LPS was observed in the aqueous humour of patients, suggesting an endotoxin-mediated process [74]. However, antibodies against the leptospiral lipoproteins LruA and LruB were detected in sera from patients who had leptospiral uveitis [73]. Indeed, antibodies against LruA and LruB presented cross-reactivity with eye proteins of horses, reinforcing the hypothesis of autoimmune uveitis in leptospirosis [75].

Guillain-Barré syndrome, an autoimmune disease that affects the nervous system, was described in a child infected with L. interrogans, probably as a result of re-infection [76]. In addition, anticardiolipin IgG antibodies were raised in patients with severe leptospirosis, which could be related to vascular endothelial injury observed in the disease [77].

It is clear that the development of autoimmune processes as a result of leptospiral infection is a relevant topic of research. Additional studies are needed to evaluate other autoimmune reactions in leptospirosis, such as IgA nephropathy and systemic lupus erythematosus.

Immune evasion in leptospirosis

Leptospiral complement evasion

An antileptospiral activity in normal serum mediated by the complement system was first reported by Johnson and Muschel in the mid-1960s [78]. The authors observed that serum from a variety of mammals exerted a bactericidal effect against different Leptospira serotypes. According to their studies, non-pathogenic forms could be distinguished from pathogenic ones by a greater susceptibility to normal serum. It became clear that virulence correlates with the ability to survive in mammalian hosts, which can be greatly attributed to the capacity of resisting complement-mediated killing [79]. Further studies on the bactericidal effect of lower-vertebrate sera towards Leptospira revealed a different pattern of killing compared with that of mammalian sera [80]. Curiously, normal serum from turtles (Chrysemys picta and Chelydra serpentina) and from the frog Rana pipiens displayed bactericidal activity against both pathogenic and saprophytic Leptospira strains. One plausible interpretation for these findings would be the presence of either specific or cross-reacting antibodies in the serum of these lower-vertebrates, enabling antibody-dependent complement-mediated lysis [80].

Although leptospiral serum resistance against host complement was described many decades ago, the mechanisms underlying this resistance started to be unraveled only recently. Pathogens have evolved sophisticated strategies to circumvent the immune defence systems of a variety of hosts, notably mechanisms to escape complement activation and/or lytic complement attack. Among these mechanisms are the acquisition of host fluid-phase complement regulators, notably Factor H and C4b-binding protein (C4BP), the secretion of proteases that inactivate key complement components, and the expression of proteins in the pathogens’ surface that may inhibit or modulate complement activation (reviewed in [81]). The first study reporting a complement evasion strategy in pathogenic Leptospira was carried out by Meri et al. [27]. The authors demonstrated that serum-resistant and serum-intermediate strains of Leptospira were able to bind Factor H and Factor H-related protein 1 (FHR-1α and FHR-1β) from human serum. Bound Factor H remained functionally active, acting as a cofactor in the cleavage of C3b by Factor I [27] (Fig. 3). To date, only two leptospiral ligands for human Factor H have been described, LenA and LenB. LenA (leptospiral endostatin-like protein A), formerly called LfhA (for leptospiral Factor H-binding protein A) [82] and Lsa24 (for leptospiral surface adhesin, 24 kDa) [19], was first identified through the screening of a lambda phage expression library of L. interrogans serovar Pomona [82]. Sequence analyses of genes from L. interrogans allowed identification of five additional lenA paralogs, designated lenB, lenC, lenD, lenE and lenF, which encode domains with putative structural and functional similarities with mammalian endostatins [83]. All Len proteins have been shown to interact with the extracellular matrix components laminin and fibronectin (LenA is able to interact only with laminin), but binding specificities for human Factor H are displayed solely by LenA and LenB [83]. Besides interacting with Factor H and laminin, LenA is a receptor for human plasminogen [84]. A number of surface proteins of pathogenic microorganisms involved in complement escape may also bind other host molecules, such as plasminogen, fibrinogen, trombin, IgA, IgG and extracellular matrix components (reviewed by [81]), thus contributing to tissue degradation and adhesion to host cells as well.

Figure 3.

 Complement evasion strategies in Leptospira. Saprophytic Leptospira are susceptible to complement-mediated killing because they do not bind the host complement regulatory proteins Factor H (FH) [27] and C4b Binding Protein (C4BP) [28]. By contrast, pathogenic Leptospira evade complement attack by acquiring these soluble proteins of the alternative and classical pathways on their surfaces. Factor H, a 150-kDa plasma protein, inhibits the alternative pathway of complement by preventing binding of Factor B to C3b, accelerating decay of the C3-convertase C3bBb and acting as a cofactor for the cleavage of C3b by Factor I. LenA and LenB are leptospiral ligands for human FH [82, 83]. C4BP is a 570-kDa plasma glycoprotein that inhibits the classical pathway of complement by interfering with the assembly and decay of the C3-convertase C4bC2a and acts as a cofactor for Factor I in the proteolytic inactivation of C4b. LcpA is a leptospiral outer membrane protein which interacts with human C4BP [85]. As a consequence of the acquisition of those fluid-phase regulators on the surface of a given pathogen, complement activation is down-regulated preventing opsonization and the formation of the lytic membrane attack complex on its surface. In the schematic representation of FH and C4BP, each circle represents one SCR domain. Open circles indicate the three SCR domains of the C4BP β-chain.

Besides recruiting Factor H to its surface, Leptospira also have the ability to bind human C4BP, a key fluid-phase inhibitor of the classical and lectin pathways [28]. Pathogenic leptospiral strains efficiently acquire C4BP from human serum, whereas the non-pathogenic strain Patoc I binds negligible amounts of this complement regulator to its surface. Surface-bound C4BP retains cofactor activity, indicating that acquisition of this complement regulator may contribute to leptospiral serum resistance [28] (Fig. 3). We recently identified a 20-kDa Leptospira outer membrane protein named LcpA which interacts with this complement regulator [85]. LcpA is surface-exposed, binds both purified and soluble C4BP from human serum, and is expressed by leptospiral strains that are at least partially able to resist complement-mediated killing. Moreover, C4BP remains functionally active when bound to LcpA, acting as a cofactor for Factor I in the cleavage of C4b [85]. Considering that pathogenic leptospires are able to widely spread in the host organism, it is reasonable to suppose that these pathogens may express many different surface receptors for complement and other host molecules. The identification and characterization of these proteins is of great relevance, as they may represent interesting targets for immune intervention.

One interesting observation is that, while pathogenic leptospires evade the complement system, purified peptidoglycan (PG) is able to activate complement [86]. At this point, it is important to consider that PG is not exposed in leptospires, being overlaid by the outer membrane. As a result of mechanical shear stress or bacterial killing, the PG could be exposed and activate the complement system. However, in pathogenic leptospires this activation would be counter-balanced by the inhibitory effects of bound Factor H and C4BP.

Recently, mutations affecting L. interrogans LPS demonstrated that it plays a crucial role in leptospiral virulence. Although previous studies have demonstrated that an intact LPS is essential for complement resistance, the above-mentioned LPS mutants did not present a higher susceptibility to complement-mediated killing [12].

Immune evasion strategies involved in renal colonization

Leptospires colonize the proximal renal tubules of reservoir animals, where they are able to replicate and persist, being constantly eliminated in the urine. A wide range of mechanisms possibly involved in the ability of leptospires to survive in the kidneys has been suggested [87].

Despite the recent finding that the absence of LPS did not reduce the complement resistance of leptospires in sera [12], it is possible that the mechanisms underlying persistence in the renal tubules are different. Leptospiral LPS recovered from rat kidneys presents a higher content of the O antigen compared with the LPS of leptospires isolated from guinea pig liver with acute infection [88]. This increased content of LPS O antigen in chronically infected kidneys could constitute an immune evasion strategy. Indeed, the O antigen expression was associated with complement resistance in Francisella tularensis, a facultative intracellular Gram-negative pathogen [89]. Therefore, the role of the LPS O antigen in the leptospiral immune evasion should be evaluated.

Besides increased LPS O antigen content, proteomic analysis revealed a reduced expression of antigenic proteins in leptospires from rat kidneys in contrast to in vitro cultured bacteria [90]. This antigenic reduction could also reflect a means of escaping from host immune responses.

Saprophytic and pathogenic leptospires are able to form biofilms, helping them to survive in environmental habitats and to colonize the hosts [91]. Indeed, biofilms can constitute a barrier against the immune effector cells and molecules, including antibodies and complement, and represent one of the major mechanisms of Pseudomonas aeruginosa persistence in chronic infections [92]. Investigations on biofilm formation by leptospires in renal tubule cells from resistant and susceptible hosts could certainly contribute to our understanding of immune evasion strategies and disease pathology.

Experimental animal models in leptospirosis

Experimental animal models have been used to test vaccine candidates as well as to elucidate host immune responses and leptospirosis pathology. Hamsters, guinea pigs and gerbils are suitable for reproducing acute lethal infection [4, 29, 93]. Mice and rats, which are resistant to leptospirosis, are experimental models to study persistent chronic infection [4]. Moreover, mice with specific mutations, such as in tlr4 gene, have been used to reproduce acute infection [94]. The pathogenesis of severe leptospirosis with a particular focus on pulmonary haemorrhage has also been studied in animal models. For this purpose, not only guinea pigs and rodents, but also non-human primates, such as marmoset monkeys, are used [67, 70, 95].

The Golden Syrian hamster is a well characterized model of leptospirosis. The hamster’s infection has the same pattern of that observed in accidental hosts, and so represents an appropriate model of the severe form of human leptospirosis [4, 93]. Indeed, the hamster model has been used to determine leptospiral infectivity, to restore the virulence of culture-attenuated strains and in challenge assays, where the protective capacity of vaccine candidates is evaluated [93, 96, 97]. The advantages of this model include the reproducibility of the results and susceptibility to a broad range of pathogenic strains [96]. However, there are relatively few hamster-specific reagents commercially available, which constitute a major difficulty to study the immunologic and cytokine response to infection. This problem has been partially overcome by the use of real time PCR for cytokines [98, 99].

Considering that mice are not susceptible to leptospirosis, transgenic and mutant murine models have been employed to study the disease. Severe combined immunodeficiency SCID/C3H, TLR4 deficient C3H/HeJ [94] and double-knockout (TLR2 and TLR4) C57BL/6 mice [45] are highly susceptible to lethal leptospirosis, and represent additional models. Indeed, young C3H/HeJ mice, which have a mutation in the tlr4 gene, were used to evaluate recombinant protein vaccine candidates [100].

Some differences regarding resistant mouse strains have been reported in protocols using sub-lethal infection with L. interrogans. The mouse strains CBA and C57BL/6, which frequently develop inflammatory lesions, are suitable for studying interstitial nephritis [101]. The mouse strain A may be the choice for studies aiming to recover large amounts of leptospires from the kidneys. Among the mouse strains tested, BALB/c proved to be the most resistant to leptospirosis [101], and was used as a resistance model [102]. The genus Rattus, which is the most important transmission source in human leptospitosis in urban settings, also represents a model of persistent infection and renal colonization [103].

Overall, diverse animal models have been employed to study leptospirosis. The choice of the experimental model depends on the aim of the study, as well as the availability of suitable reagents to address specific pathological and immunological issues.

Vaccine development in leptospirosis

In the veterinary field, leptospirosis causes significant economic impact. Abortion and stillbirths are frequent in infected cattle, requiring broad vaccination measures. The available vaccines are inactivated leptospires or outer membrane fractions. These vaccines do not confer a T cell dependent response, requiring annual revaccinations [4]. Furthermore, there is no cross-protection against the leptospiral serovars not included in the preparation, demanding the production of a new vaccine when different serovars emerge. The side effects associated with whole-cell leptospiral vaccines limits their feasibility for human use [4]. Nevertheless, licensed human vaccines are currently administered in a few countries such as Cuba and China [104, 105]. Indeed, the search for a conserved immunogenic protein that could subvert all these limitations is an actual challenge for the scientific community.

Recently, the genomes of two saprophytic and four pathogenic Leptospira strains were sequenced (Table 1) [51, 52, 68, 69]. The genome sequencing contributed in a decisive manner to the vaccine and therapeutic field, allowing the identification of potential targets for immune interference. Indeed, several vaccine candidates identified by genome bioinformatics analyses have been tested. The majority of these studies focus on subunit vaccines composed of recombinant outer membrane proteins.

Table 1.   Saprophytic and pathogenic Leptospira species whose genomes have been sequenced.
Leptospira speciesSerovarStrainTypical reservoir [13]Reference
 L. biflexaPatocPatoc1 (strains Paris and Ames)Do not survive in hosts[51]
 L. interrogansLai56601Rat[68]
 L. interrogansCopenhageniFiocruz L1-130Rat[52]
 L. borgpeterseniiHardjoL550
Cattle and sheep[69]

Leptospiral immunoglobulin-like (Lig) proteins belong to a family of surface-exposed determinants that have Ig-like repeat domains found in virulence factors such as intimin and invasin. The Lig proteins are expressed during host infection and culture attenuation of the pathogenic strains results in lost of its expression [106]. Lig proteins can bind to a variety of extracellular matrix components, thereby mediating adhesion to host cells [20]. Considering the relevant role of Lig proteins in leptospirosis, they have been tested as vaccine and diagnosis candidates. In fact, to date the most promising candidate is the C-terminal portion of leptospiral immunoglobulin-like protein A (LigA), which although not being able to completely eliminate leptospires from kidneys, confers protection in a hamster model of leptospirosis [97].

Concluding remarks

This worldwide zoonosis is an important human and veterinary health problem mainly in impoverished populations lacking proper sanitation facilities and flood-prone regions. Besides the need for more economical investment in sanitary improvements and better education, it is essential to investigate vaccine candidates that induce long lasting immune protection. To improve the success of this immunotherapy, we have to fully understand:

  • 1 the biological requirements for Leptospira survival;
  • 2 the main molecular differences between non-pathogenic and pathogenic species that define leptospiral persistence and its evasion from the host immune system;
  • 3 virulence factors and bacterial ligands implicated in immune evasion;
  • 4 gene polymorphisms that could favour the severity of leptospiral infection.


We thank Dr Aurora Cianciarullo and Dr Enéas de Carvalho from Instituto Butantan, São Paulo, for the electronic micrographs of Leptospira. We also thank Dr Shaker Chuck Farah from Instituto de Química and Dr Marilis do Valle Marques from Instituto de Ciências Biomédicas, Universidade de São Paulo, for critical reading of the manuscript.