Leptospirosis is a neglected tropical disease with an incidence of a million annual human cases in temperate countries and more than 10 million during epidemic outbreaks in tropical areas (Costa et al., 2015; Roqueplo et al., 2019). About 10-15% patients develop severe leptospirosis with multi organ failure and high mortality rate (Cagliero et al., 2018). However, the exact global impact is unknown due to inadequate surveillance systems in most low-income countries (Guerra, 2013) and to under-reporting in developed countries (Fontes et al., 2015). Furthermore, the ubiquitous presence of pathogenic Leptospira in the environment and in different host species increases the zoonotic transmission of Leptospira (Cordonin et al., 2020). Thus, the disease has a severe socioeconomic impact due to infection of animals of agricultural interest (Schafbauer et al., 2019).

Leptospira pathogenesis depends on the virulence of the Leptospira serovar and how the immune system engages with the spirochete in the host (Cagliero et al., 2018). Despite extensive research to understand the nuances of host-pathogen interactions, several questions remain unanswered regarding the host recognition process. Prior reports have dissected differences in induction of proinflammatory cytokines and chemokines after infection of hamster and guinea pigs (Vernel-Pauillac and Merien, 2006; Lourdault et al., 2009; Fujita et al., 2015). However, studies on how the cellular arm of the immune system distinguishes between pathogenic and non-pathogenic Leptospira during early infection are very limited in mouse models (da Silva et al., 2009; Domingos et al., 2017). The first line of defense to pathogenic versus non-pathogenic Leptospira should be characterized by production of different signatures of attractant chemokines and cytokines that recruit specialized immune cells and drive inflammation (Santecchia et al., 2020) which ultimately leads to early elimination of the pathogen from the host and further modulate pathogenesis and disease progression.

Mice are reservoir hosts of pathogenic Leptospira. As such, this species is generally associated with resistance to infection (Adler and Faine, 1977). A possible factor underlying susceptibility to infection in humans may be a lack of recognition of leptospiral-LPS by human TLR4, whereas resistance in mice has been linked to production of antibody within 48-72h post infection (Adler and Faine, 1976) and the murine TLR4 ability to engage leptospiral-LPS (Werts et al., 2001; Nahori et al., 2005; Santecchia et al., 2020). However, mouse strains that express an impaired TLR4 in their immune cells (C3H-HeJ) are susceptible to lethal and sublethal leptospirosis (Pereira et al., 1998; Nally et al., 2005; Viriyakosol et al., 2006; Richer et al., 2015) and those that have been engineered to not express TLR4 (C57BL/6J-TLR4_ko) succumb to infection (Chassin et al., 2009). We found recently that rather than leptospiral-LPS sensing, the presence of a fully functional TLR4 receptor in mice is necessary to control acute leptospirosis (Nair et al., 2021). In this study, we used the C3H-HeJ mouse model, which recapitulates susceptibility to leptospirosis, to analyze differences in cell-mediated immune markers engaged by pathogenic disseminators as well as those engaged by saprophytic non-disseminators during the earliest phase of infection with Leptospira, at 24h and 72h.

The engagement of the host immune system with pathogenic and non-pathogenic serovars of Leptospira at the point of entry defines the outcome of infection. In this study, we quantified immune cell mediators circulating in serum and genetic expression of chemo/cytokines in whole blood of C3H-HeJ mice infected with pathogenic and saprophytic Leptospira (L. interrogans and L. biflexa) and correlated these circulating signatures with the immune cell populations enriched in the secondary lymphoid organ (spleen) during the earliest phase of infection. We analyzed two timepoints (24h and 72h) within the first week of infection when Leptospira interrogans is known to circulate in blood and there are no weight differences - a clinical score of disease progression - in infected C3H-HeJ mice [ Figure 1 (Richer et al., 2015; Nair et al., 2020)].

We found that infection with L. biflexa was generally characterized by induction of chemoattractant chemokines (IP-10, JE/MCP-1) that engage resident macrophages and dendritic cells in spleen at 24h post infection which was replaced by an increase of Natural Killer (NK) cells and a decrease of neutrophils at 72h post infection ( Figures 2 and 5 ). Although RANTES and BLC chemoattractants were increased there was no evidence of engagement of T cells and B cells, respectively. The marked decrease in neutrophils frequency in spleen from mice infected with L. biflexa could be explained by its rapid recruitment to the spirochete point of entry as it is known that neutrophils phagocytize and destroy saprophytic Leptospira (Isogai et al., 1989; Scharrig et al., 2015; Charo et al., 2019). An efficacious phagocytic response engaged by resident macrophages (Toma et al., 2011), and neutrophils (Scharrig et al., 2015) without an overwhelming increase of pro-inflammatory cytokines contributes to elimination of the spirochete at the point of entry preventing Leptospira dissemination and absence of pathology and disease. However, it is not clear which immune mediators recruited neutrophils given that the neutrophil chemoattractants tested (KC, MIP-2 and IL-1α) were downregulated at 24h and 72h post infection with L. biflexa. Regarding the other two cell types (NK cells and dendritic cells) involved in clearance of L. biflexa, both are induced by JE/MCP-1 which was increased in serum at 24h post-infection. The protective effects of NK cells from vaccinated bovines (Zuerner et al., 2011) and from serum-thymic factor treated gerbils (Yukawa et al., 1994) have been described. Thus, we speculate that NK cells may directly eliminate cells that have phagocytized L. biflexa. Both pathogenic and non-pathogenic Leptospira induced maturation of human dendritic cells and production of pro-inflammatory cytokines (Gaudart et al., 2008; Santecchia et al., 2020). In another study, liposomes prepared from L. biflexa lipids activated bone marrow derived dendritic cells that released pro-inflammatory cytokines and induced strong humoral and cytotoxic T cell responses (Faisal et al., 2011). In our study, we did not detect any pro-inflammatory cytokines in serum or blood cells of C3H-HeJ mice infected with L. biflexa. Thus, the role played by these professional antigen presenting cells is unclear, but it may be related to the recruitment of cytotoxic T cells.

In contrast to L. biflexa, infection with L. interrogans was characterized by a substantial induction of attractant chemokines whose function is to recruit myeloid phagocytic cells and pro- and anti-inflammatory cytokines whose main function is to engage lymphoid cells ( Figures 2 – 5 ). The frequency of macrophages and dendritic cells was likely driven by MIG, IP-10 and JE/MCP-1 chemokines increased in blood at 24h post-infection. These cells were replaced by a subsequent increase in monocytes and Natural Killer cells and a decrease of dendritic cells driven by IP-10, I-309, JE/MCP-1, MIP-1β and MCP-5 at 72h post-infection. T cell recruitment may have been driven by IP-10, I-309, RANTES, MCP-5 and TARC at 72h post infection. As observed for L. biflexa, BLC which is involved in B cell recruitment was increased in serum of L. interrogans infected mice but differences in B cell frequency in spleen were not significant at this timepoint. This suggests engagement of the adaptive arm of the immune system as early as 24h-72h post infection with L. interrogans as described by others (Adler and Faine, 1976). Chemokine proteome arrays in L. interrogans infected mice have been done by other groups (Domingos et al., 2017; Silva et al., 2020). Differences observed in the panel of chemokines detected can be explained by differences in methodology in that we quantified circulating chemokines (serum and blood cells) and they quantified chemokines in spleen/lung (Domingos et al., 2017) and in liver/kidney (Silva et al., 2020). Nevertheless, MIG/CXCL9, IP-10/CXCL10, I-TAC/CXCL11, BLC/CXCL13 and JE/MCP-5/CCL2 detected in kidney of C3H-HeJ mice at 24h post infection (Silva et al., 2020) was also detected in serum/blood cells in our study.

Neutrophil chemoattractants, KC and MIP-2, were increased in the serum proteome and in whole blood gene transcripts of L. interrogans infected mice. However, neutrophils were not increased or decreased in spleen. Earlier studies have shown that neutrophils do not phagocytize pathogenic Leptospira (Isogai et al., 1989; Scharrig et al., 2015; Charo et al., 2019; Santecchia et al., 2020). Others have shown that macrophages, but not neutrophils, infiltrate organs of C3H-HeJ mice infected with L. interrogans (Chen et al., 2017). This may explain the lack of engagement of neutrophils in L. interrogans infected C3H-HeJ mice in our study.

Discrepant results between serum protein and genetic expression in blood may reflect the limited number of cell types circulating in blood at the time of blood collection, whereas serum receives proteins from all immune cell types from different organs and tissues. This may explain why we detected a more comprehensive panel of immune markers in serum than in transcripts from whole blood. Nevertheless, it is interesting to note that the immune markers detected by genetic expression represent the overall engagement of all arms of the immune cell response: MIP-2 (neutrophils), MIG and TNF-α (macrophages/NK cells), IP-10 (monocytes, macrophages, dendritic cells, NK cells and T cells), RANTES (T cells, eosinophils, basophils), IL-1β (immune cell proliferation, differentiation and apoptosis) and IL-6 (pro-inflammatory and regulatory-T cell suppression).

As expected, and in contrast to L. biflexa infection, L. interrogans induced a strong adaptive immune response with production of proinflammatory cytokines TNF-α (innate immune alarm), IL-1ra (inhibitor of IL-1), IL-3, IL-4 (Th2), IL-6 (pro- and anti-inflammatory, Treg suppression), IL-10 (anti-inflammatory), IL-12p70 (NK and Th1 cell function), IL-13 (Th2, IgE antibody), IL-16 (activated T cells, dendritic cells and eosinophils), and IL-23 (Th17 expansion, induces IL-6). Furthermore, a considerable number of chemokines that recruit eosinophils and basophils (RANTES, Eotaxin, MCP-5, G-CSF, GM-CSF) were also increased at 72h post infection with L. interrogans. Although neutrophils do not appear to be engaged by L. interrogans, these other granulocytes may contribute to inflammation.

The stark differences in expression of chemoattractant chemokines and pro- and anti-inflammatory cytokines can be correlated with pathogenesis. These data suggest that prevention of dissemination of L. biflexa is associated with an early engagement of the innate immune system characterized by secretion of a few chemoattractants that results in an efficacious phagocytic response by resident macrophages and neutrophils, which are helped by dendritic and Natural Killer cells without an overwhelming increase of pro-inflammatory cytokines. However, when macrophages fail to clear a pathogenic serovar such as L. interrogans, monocytes, NK cells, T cells and possibly other granulocytes (eosinophils and basophils) may be recruited to help out. The resulting chemo-cytokine storm mediates a robust but non-protective inflammatory response to pathogenic Leptospira (Cagliero et al., 2018) which results in dissemination, kidney colonization, pathology and disease.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

The animal study was reviewed and approved by University of Tennessee Health Science Center Institutional Animal Care and Use Committee.

AS and SK contributed equally. MGS is the corresponding author who conceptualized the project, procured funding, designed experiments, data analysis, prepared figures and wrote the manuscript. AS and SK designed and performed the experiments, assisted in writing the manuscript, data analysis and figure preparation. All authors contributed to the article and approved the submitted version.

This work was supported by Public Health Service grants R43AI136551, R21AI142129 and R43AI155211 to MGS from the National Institutes of Health, National Institute of Allergy and Infectious Diseases.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.