Introduction

Visceral leishmaniasis (VL) is considered the second most frequent cause of mortality and the fourth most frequent cause of morbidity within tropical diseases, with 20,000–40,000 deaths per year1 and significant economic impact due to an estimated 2 million disability-adjusted life years lost2. Strategies to eliminate VL in endemic areas rely on rapid detection and treatment of VL to reduce the number of human reservoirs, and vector control using indoor residual spraying3. However, elimination programs in endemic areas have not consistently met the intended milestones, and the need for potent diagnostic, treatment methods and prophylactic vaccines against VL to ensure long term control and prevent reemergence of VL is recognized3. Vaccination against VL is considered feasible since long-term protection is acquired following clinical cure of VL in majority of the cases as evidenced by previous studies in VL endemic areas4,5,6,7. In addition, protection in cutaneus leishmaniasis is observed against infection resulting from the process of leishmanization5. A broad range of vaccines including recombinant antigen vaccines, heat-killed parasites, adeno-viral vectored vaccines have been tested against VL8,9,10,11,12,13. Yet no approved vaccine for human VL exists.

We have previously reported on the safety and immunogenicity characteristics of centrin gene deleted live attenuated Leishmania parasites (LdCen−/−) as prophylactic vaccines in pre-clinical animal models14,15,16,17,18,19,20,21. Gene deleted Leishmania parasites such as LdCen−/− would present a broad repertoire of antigens to generate protective immunity while undergoing limited replication in the immunized host22. Since the LdCen−/− parasite have growth defects in amastigote forms, but not in promastigotes23,24, this deletion prevents cell division and long-term persistence in animals (mice and hamsters) or in human macrophages ex vivo23. The same non-virulent characteristics of Centrin gene deletion were also observed in other species, like L. major, L. mexicana and L. braziliensis25,26,27. In pre-clinical studies, vaccination with LdCen−/− proved to be safe, protective and in mice (BALB/c and SCID), hamsters and dogs, after challenge with virulent parasites, in addition to cross-protection in animals challenged with L. braziliensis, L. infantum and L. mexicana14,15,16,17,18,19,20,21,28. Previous studies analyzing the protective immune response following immunization demonstrated the central role of long-lasting Th1-type response4,9,29,30,31,32,33,34,35,36,37. Thus, the induction of a Th1 type response has been considered a pre-requisite in attempts to identify molecules of the parasite and adjuvants that would induce this phenotypic profile in vivo models9. Since IFN-γ plays an essential role in the activation of macrophages, allowing the elimination of intracellular pathogens and protecting the host cell against infection38, its production is considered one of the main objectives in the immunization process against leishmaniasis. Accordingly, the protective immunity induced by LdCen−/− parasites has been shown to be mediated by Th1 dominant multifunctional CD4+ and CD8+ T cell populations that orchestrate the assembling of granulomas in the spleens followed by the parasiticidal activities mediated by nitric oxide14. Recent studies also showed that in a centrin deletion mutant of L. major, skin resident TRM (Tissue Resident Memory) cells also play an important role in protection39. Towards understanding the immune mechanisms that direct the development of Th1-type response following LdCen−/− immunization, studies investigated early interactions between LdCen−/− parasites and the innate immune cells. These studies revealed reprogramming of M1/M2 macrophages, reconfiguration of the membrane architecture to enhance cholesterol driven fluidity, variable co-stimulatory and co-inhibitory ligands, and chemokine/cytokine signatures and the corresponding regulatory microRNAs enabling the development of Th1 type immune response following LdCen−/− immunization compared to infection with virulent LdWT parasites14,40,41,42. Yet, while the immunological characteristics of the innate and adaptive responses showed significant differences between LdCen−/− and LdWT infections, the parasitic factors that drive these changes remain to be understood.

As the mediators of protection, the characteristics of CD4+ and CD8+ T cell memory subpopulations and their role in prophylaxis or treatment43,44,45,46,47,48,49,50,51 has been studied thoroughly. A study evaluating different types of T cells in visceral leishmaniasis, including memory T cells without identifying subtypes, demonstrated a lower number of memory T cells in patients with VL compared with treated or asymptomatic patients52. Different combinations of effector memory (TEM)/central memory (TCM) T cells have been shown to be important in inducing protection against secondary infections by Leishmania53,54,55,56,57. Thus, successful vaccination must be based on a low dose of antigen, to allow slow replication of effector cells and favor differentiation of memory T cell populations57,58.

Studies in Leishmania and Plasmodium identified a parasite-encoded ortholog of a cytokine macrophage migration inhibition factor (MIF) that has been shown to affect the macrophage apoptosis, activation signals, CD4 T cell apoptosis, and CD4 effector T cell responses59,60,61,62. Evidence indicates that Leishmania encoded MIF cytokine may drive an inflammatory environment that is detrimental to the host response63. Deletion of MIF in Leishmania major parasites showed that antigen presenting, T cell priming and the IFN-γ, IL-7R production by CD4 T cells were significantly affected due to the loss of MIF genes63. Studies in Plasmodium berghei showed that Plasmodium encoded MIF enhanced inflammatory cytokine production and induced antigen experienced CD4+ T cells to develop into short-lived effector cells rather than memory precursor cells. The short-lived effector CD4 T cells were more readily eliminated by Bcl-2-associated apoptosis, resulting in decreased CD4 T-cell recall responses against challenge infections61. Thus, to investigate the role of Leishmania encoded MIF in generating effector and memory T cell populations following LdCen−/− immunization, we created additional deletion of MIF genes (MIF1 and MIF2, tandemly arranged MIF genes) in the vaccine strain and analyzed the long-term immune response. Leishmania MIF has been shown to interact with its receptor CD74 and exhibit an anti-apoptotic activity that may facilitate the intracellular persistence of the parasite in macrophages59. Thus, the MIF deletion would likely increase the long-term immune response that is essential for a successful VL vaccine.

We have evaluated the profile of memory T cells after immunization with the Leishmania donovani double-knock out parasites (LdCen−/−MIF−/−) and single attenuated (LdCen−/−) parasites. Immunized mice with both types of parasites demonstrated a strong immune response, capable of inducing TCM populations following immunization. The changes in the profile of TEM populations after challenge were also observed, resulting in significant cross-protection against infection with virulent L. infantum parasites.

Results

LdMIF proteins induce inflammatory cytokines

Leishmania donovani genome contains two tandemly arranged copies of MIF genes that share significant homology (Fig. 1A). The two ORFs corresponding to MIF genes were PCR amplified and used to produce recombinant proteins. Coomassie stained gel showed LdMIF1 and LdMIF2 proteins were purified to a high degree (Fig. 1B). To test whether LdMIF proteins induce inflammatory cytokines similar to other parasitic protozoa, purified recombinant LdMIF1 and LdMIF2 proteins were incubated with murine BMDCs. Culture supernatants showed significant induction of TNF-α in presence of either LdMIF1 or LdMIF2 proteins relative to untreated control (Fig. 1C). Similarly, BMDC cultures incubated with rLdMIF1 and rLdMIF2 (not shown) also showed an induction of IL-12 in a dose dependent manner (Fig. 1D). Thus, we observed that LdMIF1 and LdMIF2 proteins were able to induce a proinflammatory response.

Figure 1
figure 1

Cytokines production after stimulation with LdMIF1 and LdMIF2 proteins. (A) L. donovani sequences of MIF genes. (B) Coomassie stained gel showing LdMIF1 and LdMIF2 proteins purified. (C) TNF-α and (D) IL-12p40 production after stimulation of C57BL/6 murine BMDCs with purified recombinant LdMIF1 and LdMIF2 proteins, compared to positive control (LPS). The experiment was performed four times (n = 3/replication), and cells were pooled and the error bars represent the standard deviation of each group.

Full size image

Deletion of MIF genes does not affect the amastigote proliferation

To test the effect of the LdMIF induced inflammatory response in immunogenicity, we prepared MIF deletion mutants. The two tandemly arranged copies of LdMIF genes and the genomic context are shown (Fig. 2A). The gene replacement construct contained either Blasticidin or Puromycin targeted deletion of both copies of MIF genes including the intergenic region. The MIF genes were deleted in a sequential transfection with Blasticidin and puromycin constructs and selection with respective antibiotics. Southern hybridization with the indicated probes showed the deletion of both MIF genes from LdWT genome (LdMIF−/− lane, Fig. 2B).

Figure 2
figure 2

Deletion of MIF genes. (A)The two tandemly arranged copies of LdMIF genes and the genomic context showing the gene replacement construct containing either Blasticidin (BSD) or Puromycin (PAC) targeted deletion of both copies of MIF genes including the intergenic region. (B) Southern blot of the transfected (autoradiograph). (C) Microscopy of the murine macrophages infected with LdWT and LdMIF−/− parasites to evaluate the effect of fitness and growth as amastigotes. (D) Number of parasites/100 macrophages infected with LdMIF−/− parasites, compared to LdWT parasites during 7 consecutive days. (E) Number of parasites per spleen of BALB/c mice 4 weeks after infection with LdWT and LdMIF−/− parasites. The experiment was performed four times (n = 3/replication), and cells were pooled. The error bars represent the standard deviation of each group.

Full size image

To test whether deletion of MIF genes affects the parasite fitness and growth as amastigotes, murine macrophages were infected with LdWT, LdMIF−/− parasites (Fig. 2C). Results showed that deletion of MIF genes does not affect the growth of amastigotes as their growth paralleled that of LdWT parasites (Fig. 2D). To test if MIF deletion affects the parasite replication in vivo, mice were infected with LdMIF−/− parasites and compared to mice infected with L. donovani WT. After 4 weeks, splenic burden was measured. Results showed that LdMIF−/− infected mice presented parasites in spleen (Fig. 2E). Deletion of MIF in L. donovani parasites does not affect replication. However, we observed less parasites (not statistically significant) in spleen of LdMIF−/− infected mice. It could indicate that the mutant parasites grow slower than wild type strain but further studies are necessary.

MIF deletion results in reduced CD4+ T cell apoptosis

To test whether MIF deficiency results in reduced inflammation and thus diminished T cell apoptosis in vivo, we infected Balb/C mice with LdWT, and LdMIF−/− parasites. T cell apoptosis in the spleens of the infected mice was monitored by flow cytometry as shown (Fig. 3A). Results showed that the CD4+ T cell apoptosis (Annexin-V+) was significantly reduced on days 5 and 9 of the LdMIF−/− infection compared to LdWT infection (Fig. 3B,C, respectively) corresponding to the expansion phase of the T cell responses post-infection. No significant difference was observed in the Annexin+ CD4 T cell population on day14, that corresponds to the post-contraction phase (data not shown). Similar differences in CD8+ T cell populations were also evident to a lesser degree in LdMIF−/− infection (Fig. 3D,E). Correspondingly, IFN-γ production from the splenocytes of LdMIF−/− infected mice showed a significant reduction compared to LdWT infection after 5 (p = 0.01) and 7 (p = 0.001) days (Fig. 3F). Immunohistochemistry of the spleens from LdWT, and LdMIF−/− 9 days post infection showed the presence of TUNEL+ cells in LdWT but much less in LdMIF−/− infection (Fig. 3G) indicating that deletion of MIF promotes the survival of T cells in the infected spleens consistent with our flow cytometric analysis.

Figure 3
figure 3

Apoptosis profile of CD4+ T cells in mice infected with LdMIF−/− parasites. (A) Gate strategy for flow cytometry. Apoptosis evaluation using Annexin-V+ as a marker in (B) and (C) CD4+ and (D) and (E) CD8+ T cells, at 5 (B and D) and 9 (C and E) days post infection. Significant differences are indicated on the graphs (*p < 0.05; **p < 0.01; ***p < 0.001 and ****p < 0.0001). (F) IFN-γ production from splenocytes of LdMIF−/− and LdWT infected mice at days 5 and 7 post injection. 0.25 µg/ml LPS was used in the stimulation. (G) Expression of the anti-apoptotic Bcl2 protein in splenocytes from LdMIF−/− and LdWT infected BALB/c mice. (H) Immunohistochemistry of the spleens from LdWT and LdMIF−/− at 9 days post infection. The cells were labelled with TUNEL+, DAPI and CD3-Alexa 488. The experiment was performed four times (n = 3/replication), and cells were pooled. The error bars represent the standard deviation of each group.

Full size image

Deletion of MIF in LdCen −/− background

Since MIF deletion in L. donovani showed enhanced survival of CD4+ T cells in spleens as was shown in studies with virulent L. major parasites60,63, we performed MIF deletion in the LdCen−/− mutant background to test if the quality of T cell responses can be tweaked further towards inducing strong vaccine induced protection. Using gene targeting strategy described above, we deleted MIF genes from LdCen−/− mutant. Southern hybridization with the indicated probes showed the deletion of both MIF genes from LdCen−/− genome (LdCen−/−MIF−/− lane, Fig. 4A,B). Hybridization with Neo and Hyg probes showed that LdCen−/− genotype remained unperturbed due to MIF deletion (Fig. 4B).

Figure 4
figure 4

MIF deletion in the LdCen−/− mutant background and specific antibody production. (A) and (B) Southern blot of MIF deletion in the LdCen−/− mutant background (autoradiograph). (C) Immunization scheme. (D–F) ELISAs using plates coated with SLA from L. infantum were performed to detect production of total IgG (D), IgG1 (E) and IgG2 (F). The antibody OD values are shown on the y-axis, and the error bars indicate the standard deviation. Dotted lines represent the cut-off value. Statistical differences (p < 0.05) are indicated in letters (a: PBS; b: LdCen−/−). The animal experiments were performed in three replicates and the error bars represent the standard deviation of each group (n = 8 mice/group). The BALB/c mice were evaluated individually.

Full size image

To test the immunogenicity of LdCen−/−MIF−/− double deletion mutants, we immunized mice with LdCen−/− or LdCen−/−MIF−/− mutants (Fig. 4C). Anti-Leishmania vaccines induce humoral immune response that are often used as a surrogate of adaptive immunity which is the main driver of protection. The ability of the attenuated parasites to induce IgGTotal, IgG1 and IgG2 antibodies against soluble antigen of Leishmania infantum, after immunization or challenge, was investigated (Fig. 4D–F). The profile of antibodies was measured 4 weeks post-immunization (4wpi), as well as 4, 8 and 12 weeks after the challenge (wpc). Immunization with the double-attenuated strain LdCen−/−MIF−/− increased the secretion of IgGTotal at 8wpc (p < 0.01) (Fig. 4D) and IgG1 at 4wpi and 8wpc (p < 0.01) (Fig. 4E). In addition, both LdCen−/− and LdCen−/−MIF−/− attenuated parasites were able to induce higher levels of IgGTotal (4wpc and 12wpc) (p < 0.01) (Fig. 4D), IgG1 (4wpi, 4 and 12wpc) (p < 0.01, p < 0.001 and p < 0.01, respectively) (Fig. 4E), and IgG2 (4wpi and 4wpc) (Fig. 4F) compared to non-immunized group. The results suggest an induction of cross-reactive antibodies since the animals were immunized with L. donovani attenuated parasites and challenged with a L. infantum strain.

Effect of MIF deletion on T cell proliferation in LdCen −/− MIF −/− immunized mice

In order to assess if vaccination with single dose of LdCen−/− or LdCen−/−MIF−/− parasites (before and after challenge with L. infantum WT) had an effect on T cell function, we first evaluated the capacity of splenocytes T cells to proliferate upon specific SLA stimulation. The profile of proliferation was evaluated at 4wpi, 4wpc and 12wpc. For this, we used flow cytometry to measure the mean intensity fluorescence of Ki67 in CD4+ and CD8+ T cells (Fig. 5). We observed a significant increase of proliferation of CD4+ and CD8+ T cells on groups immunized with LdCen−/− at 4wpi (p < 0.01) (Fig. 5A) and of CD8+ T cells on group immunized with LdCen−/−MIF−/− at 4wpi (p = 0.024) (Fig. 5B). For CD4+ T cell, in vitro stimulation with SLA did not increase proliferation after challenge. In fact, after 4wpi there was a decrease in proliferation of CD4+ T cells under SLA stimulation (p < 0.01) (Fig. 5A). In CD8+ T cell subpopulation, double-attenuated strain LdCen−/−MIF−/− decreased proliferation at 4wpi (Fig. 5A), but after the challenge with L. infantum this group of cells was able to proliferate in vitro under SLA stimulation (p < 0.01) (Fig. 5B).

Figure 5
figure 5

T Cell proliferative responses in animals immunized with attenuated parasites. Proliferation of T CD4+ (A) and T CD8+ (B) cells after pulsing with L. infantum SLA for 72 h. Proliferation responses were expressed in terms of Medium Intensity of Fluorescence (MIF) of Ki67 marker. Significant differences are indicated on the graphs (a: PBS; b: LdCen−/−). The experiments were performed in three replicates and the error bars represent the standard deviation of each group (n = 8 mice/group). The BALB/c mice evaluated individually.

Full size image

Intracellular cytokines expression by T cells in LdCen −/− MIF −/− immunized mice

Having observed that immunization with LdCen−/− or LdCen−/−MIF−/− attenuated parasites leads to proliferation of mostly CD8+ T cells after SLA stimulation, we evaluated the profile of cytokines (IFN-γ, IL-10, IL-4, IL-17, IL-12p40 and TNF-α) expression by CD4+ and CD8+ T cells. The mice immunized with LdCen−/− or LdCen−/−MIF−/− were compared to the PBS group, 4 weeks post immunization (4wpi) and 12 weeks post challenge (12wpc). The analyses strategy can be observed in supplementary Fig. 2. As shown in Fig. 6A, there was an increase of the percentage of IL-10, IL-4, IL-17 and IL-12/IL-23p40 by both CD4+ and CD8+ T cells from LdCen−/−MIF−/− immunized animals at 4wpi, when compared to PBS group (p < 0.01) (Fig. 6A,B). However, there was increase in TNF-α only in CD8+ T cells at 4 wpi in LdCen−/− group (Fig. 6B). At 12wpc, we observed only high expression of IL-12p40 by both CD4+ and CD8+ T cells in LdCen−/−MIF−/− group, compared to PBS group (p < 0.01) (Fig. 6C,D, respectively). We have observed a mixed profile of cytokine expression by T cells after both immunization and challenge.

Figure 6
figure 6

Cytokines production by T cells. Percentage of IFN-γ, IL-12/IL-23p40, TNF-α, IL-17A, IL-10 and IL-4 production was assessed in the stimulated CD4+ (A and C) and CD8+ (B and D) T cells, at 4 weeks post immunization and 12 weeks post challenge. Results from cultures are expressed as ratio (percentage of cultures stimulated with SLA L. infantum/Percentage of unstimulated cultures). Significant differences are indicated on the graphs (a: PBS). The experiments were performed in three replicates and the error bars represent the standard deviation of each group (n = 8 mice/group). The BALB/c mice were evaluated individually.

Full size image

Effect of MIF deletion on memory T cell populations in LdCen −/− MIF −/− immunized mice

To determine if vaccination with LdCen−/− or LdCen−/−MIF−/− attenuated parasites is able to induce memory T cells, we evaluated the expression of CD62L and CD27 molecules by CD3+CD4+ and CD3+CD8+ T cells using flow cytometry, after specific antigenic stimulation. The profile of the subpopulations of memory cells was assessed as central memory (TCM-CD62L+CD27+), early effector memory (TEM early-CD62L−CD27+) and late effector memory (TEM late-CD62L−CD27−). The analyses strategy can be observed in supplementary Fig. 3. Analyzing the repertoire of CD4+ and CD8+ T cells at 4 weeks post immunization, we observed an increased frequency of TCM cells in LdCen−/−MIF−/− compared to PBS group (p < 0.05) (Fig. 7A,B). After challenge, we demonstrated that immunization with attenuated parasites induced a high frequency of CD4+ and CD8+ TCM cells in 4wpc compared to PBS group (LdCen−/−: p < 0.05 for both) (LdCen−/−MIF−/−: p < 0001 and 0.01, respectively) (Fig. 8A,B). We observed no statistical differences between groups in 12wpc for CD4+ and CD8+ TCM cells (Fig. 8A,B).

Figure 7
figure 7

Subtypes of memory T cells 4 weeks post immunization. Expression of CD62L and CD27 molecules by CD3+CD4+ and CD3+CD8+ T cells, after specific antigenic stimulation. The profile of the subpopulations of memory cells was assessed as central memory (TCM-CD62L+CD27+), early effector memory (TEM early-CD62L−CD27+) and late effector memory (TEM late-CD62L−CD27−). Results from cultures are expressed as ratio (Percentage of cultures stimulated with SLA L. infantum/Percentage of unstimulated cultures-A and B) (Proportion of percentage of cultures stimulated with SLA L. infantum/percentage of unstimulated cultures-7.1 and 7.2). Significant differences are indicated on the graphs (a: PBS). p values can be found in Supplementary Table 1. The experiments were performed in three replicates and the error bars represent the standard deviation of each group (n = 8 mice/group). The BALB/c mice were evaluated individually.

Full size image
Figure 8
figure 8

Subtypes of memory T cells 4- and 12-weeks post challenge. Expression of CD62L and CD27 molecules by CD3+CD4+ and CD3+CD8+ T cells, after specific antigenic stimulation. The profile of the subpopulations of memory cells was assessed as central memory (TCM-CD62L+CD27+) (A and B), early effector memory (TEM early-CD62L−CD27+) (E and D) and late effector memory (TEM late-CD62L−CD27−) (E and F). Results from cultures are expressed as ratio (Percentage of cultures stimulated with SLA L. infantum/Percentage of unstimulated cultures-A and B) (Proportion of percentage of cultures stimulated with SLA L. infantum/percentage of unstimulated cultures—8.1 and 8.2). Significant differences are indicated on the graphs (a: PBS; b: LdCen−/−). p values can be found in Supplementary Table 2. The experiments were performed in three replicates and the error bars represent the standard deviation of each group (n = 8 mice/group). The BALB/c mice were evaluated individually.

Full size image

CD4+ and CD8+ TEM early showed a decreased frequency of cells in LdCen−/−MIF−/− group compared to PBS group after 4 weeks post-immunization (p < 0.05 and 0.01, respectively) (Fig. 7A,B). After challenge, we observed that the frequency of CD4+ and CD8+ TEM early cells significantly decreased in immunized groups compared to PBS group in 4wpc (LdCen−/−: p < 0.01 and 0.05, respectively) (LdCen−/−MIF−/−: p < 0.05 for both), but not in 12wpc (Fig. 8C,D). We also observed that frequency of CD4+ TEM late cells decreased in LdCen−/−MIF−/− group compared to PBS group after 4 weeks post-immunization, but no statistical difference (Fig. 7A), while frequency of CD8+ TEM late cells was similar between groups in 4wpi (Fig. 7B). After challenge, immunization with LdCen−/−MIF−/− followed by 4wpc showed an increase of percentage of CD4+ TEM late cells compared to PBS and LdCen−/− groups (p < 0.05) (Fig. 8E). No differences were observed in frequency of CD8+ TEM late cells between groups post challenge (Fig. 8E,F).

We also evaluated the proportion of memory CD4+ and CD8+ T cells subpopulations 4 weeks post-immunization (Fig. 7.1 and 7.2, Supplementary Table 1) and 4 and 12 weeks after challenge (Figs. 8.1 and 8.2, Supplementary Table 1). The data was represented as parts of whole, to show the fraction of total the subpopulations inside CD3+CD4+ and CD3+CD8+ T cells. Mean values of each subpopulation and timepoint is plotted, and the scientific graphic program considers the sum of the populations as 100%. Post-immunization, we observed a similarity between PBS and LdCen−/− groups, once they demonstrated high proportion of CD4+ (Fig. 7.1A,B) and CD8+ (Fig. 7.2A,B) TEM late cells in relation to TCM and TEM early cells, respectively. The LdCen−/−MIF−/− immunized group demonstrated higher fraction of CD4+ and CD8+ TCM cells compared to TEM cells (early and late) (Fig. 7.1C and 7.2C). After 4 and 12 weeks of challenge, the profile has changed, we observed a predominance proportion of CD4+ and CD8+ T EM late cells (Fig. 8.1C–F) and CD8+ (Fig. 8.2C–F) in LdCen−/−MIF−/− immunized group.

After 12 weeks of challenge, the groups showed a predominance of distinct memory T cells subpopulations. We showed that CD4+ (Fig. 8.1E,F) and CD8+ (Fig. 8.2E,F) TEM late cells proportion was higher in both immunized groups compared to PBS group, where CD4+ and CD8+ TCM cells were majority (Fig. 8.1D and 8.2D).

Cytokines secretion by memory T cells

Previous studies have suggested that the balance of pro-and anti- inflammatory cytokines may be associated with protection against leishmaniasis64,65,66. IFN-γ, IL-12/IL-23p40, TNF-α, IL-17A, IL-10 and IL-4 production was assessed in the subpopulations of stimulated memory CD4+ and CD8+ T cells by flow cytometry, in splenocytes isolated from mice immunized with LdCen−/− and LdCen−/−MIF−/−, at 4wpi and 12 wpc (Fig. 9—9.1-TCM, 9.2-TEM early and 9.3-TEM late). The analyses strategy can be observed in supplementary Fig. 3. p values can be found in supplementary Table 2 (TCM), supplementary Table 3 (TEM Early), and supplementary Table 4 (TEM Late).

Figure 9
figure 9

Cytokines production by subtypes of memory T cells. Percentage of IFN-γ, IL-12/IL-23p40, TNF-α, IL-17A, IL-10 and IL-4 production was assessed at 4 weeks post immunization and 12 weeks post challenge in the: (9.1) Central Memory T cell subpopulation (TCM) of stimulated memory CD4+ (A and C) and CD8+ (B and D) T cells. (9.2) Early Effector Memory T cell subpopulation (TCM) of stimulated memory CD4+ (A and C) and CD8+ (B and D) T cells. (9.3) Late Effector Memory T cell subpopulation (TCM) of stimulated memory CD4+ (A and C) and CD8+ (B and D) T cells. Results from cultures are expressed as ratio (percentage of cultures stimulated with SLA L. infantum/percentage of unstimulated cultures). Significant differences are indicated on the graphs (a: PBS; b: LdCen−/−). p values can be found in Supplementary Table 3. The experiments were performed in three replicates and the error bars represent the standard deviation of each group (n = 8 mice/group). The BALB/c mice were evaluated individually.

Full size image

In TCM cells (Fig. 9.1), in general, we observed the same profile of cytokine expression in CD4+ and CD8+ cell compartment. At 4wpi LdCen−/− and LdCen−/−MIF−/− groups showed an increase of cytokine production (IFN-γ, TNF-α, IL-10 and IL-4) compared to PBS group (Fig. 9.1A,B). Also, LdCen−/−MIF−/− immunized mice showed an increase of IL-12/IL-23p40 and TNF-α producing CD4+ T cells compared to LdCen−/− (Fig. 9.1A). LdCen−/−MIF−/− group presented an increase of CD8+ T cells producing most of cytokines but IL-12 (Fig. 9.1B).

Interestingly, at 12wpc the cytokine profile for TCM cells was different. In CD4+ TCM cells subpopulation, LdCen−/−MIF−/− group showed a higher percentage of IL-12p40 expressed by CD4+ T cells, compared to LdCen−/− group (Fig. 9.1C). Meanwhile, CD8+ TCM cells showed a higher percentage of IFN-γ, IL-12p40, IL-10 and IL-4, compared to PBS group (Fig. 9.1D), indicating that the manipulated Leishmania can address differential activation in different T cell subtypes.

Regarding the expression of cytokines by TEM early (Fig. 9.2), at 4wpi, LdCen−/−MIF−/− group showed an increase in all evaluated cytokines produced by T cells compared to PBS and LdCen−/− groups (Fig. 9.2A,B). In CD8+ compartment such difference was observed in IFN-g, TNF-a, IL-17A and IL-10 at 4wpi (Fig. 9.2B). Interestingly, at 12wpc significant differences between immunized groups were only seem against PBS group for all cytokines, but IL-10 by CD4+ T cells (Fig. 9.2C,D).

Evaluating the TEM late population (Fig. 9.3) at 4wpi, only IL-4 producing cells showed differences between LdCen−/− and LdCen−/−MIF−/− groups for both CD4+ (Fig. 9.3A) and CD8+ (Fig. 9.3B). On the other hand, at 12wpc, LdCen−/−MIF−/− group showed increased percentages of CD4+ (Fig. 9.3C) for all analyzed cytokines, but IFN-γ, compared to LdCen−/− immunized mice. These phenomena were not kept for CD8+ TEM late. LdCen−/−MIF−/− group presented an increase of Il-12p40, IL-17A IL-10 and IL-4, when compared to PBS group (Fig. 9.3D).

Parasite load

Having observed induction of antibodies production, T cell activation and proliferation, long term memory and cytokines secretion, we evaluated parasitic load in mouse spleen (Fig. 10A) and liver (Fig. 10B) at times 4wpc, 8wpc and 12wpc by serial dilution. At 4wpc, very few L. infantum parasites were observed in spleen and liver by LdCen−/−MIF−/− immunized animals, decreasing the number of parasites observed at 8wpc and 12wpc indicating that the immunization with LdCen−/−MIF−/− was efficient as a vaccination protocol (Fig. 10). p values can be found in supplementary Table 5.

Figure 10
figure 10

Parasite burden after challenge. Mice (n = 8) were challenged intravenously in the tail with virulent L. infantum (PP75) and presence of parasites was individually detected 4, 8 and 12 weeks after challenge. The experiments were performed in three replicates and the error bars represent the standard deviation of each group. Spleen and liver from BALB/c mice were used in limiting dilution assay and data expressed as number of parasites/organ. Statistical differences (p < 0.05) are indicated in letters (a: PBS; b: LdCen−/−). p values can be found in Supplementary Table 4.

Full size image

Discussion

The development of immunity against Leishmania induced by vaccination is widely discussed due to the complexity and antigenic variability of the parasites. Among the models used to study the effect of novel vaccines candidates against Leishmania infection, BALB/c and C57BL/6 strains are widely used on those studies67,68. The reasons to choose between both strains are: BALB/c mice show an intermediate pattern of activation that favors parasite persistence and chronicity of the disease, while C57BL/6 mice show a classical pattern of activation associated with resolution of the disease69. Vaccination using attenuated forms of parasites allow the immune system to interact with a wide repertoire of antigens69, inducing a more robust and complete response when compared to recombinant antigen vaccines70 because of the presentation of complete array of parasite antigens to the host immune system. In this study, we evaluated a vaccine against visceral leishmaniasis consisting of a double genetically deficient Leishmania donovani for the Centrin 1 gene and macrophage migration inhibitory factor (LdCen−/−MIF−/−). The centrin1 gene is associated with cell division, specifically affects the proliferative capacity of amastigote forms that replicate within macrophages but does not affect the replication of promastigote forms. The safety of infection with LdCen−/− has been previously demonstrated in mice, hamsters and dogs14,15,16,17,18,19,20,21. The MIF gene encodes a lymphokine involved in cell-mediated immunity, inhibiting the proliferation of memory cells71. It is demonstrated that parasites express MIF-like genes and could interact functionally with the MIF receptor (CD74) acting as evasion mechanism for infection success59,72,73,74,75,76,77. We hypothesized that the use of a parasite knockout for MIF ortholog could induce a long-term memory response, activation and proliferation of B and T cells, inducing cytokines production, and resulting in a cross-protective immunity against L. infantum. Studies of MIF deletion in Leishmania major showed that mice infected with such parasites presented a reduced ability of the parasite to activate antigen-presenting cells, and consequently a reduction in T-cell priming60. Also, mice infected with MIF deleted parasites presented a reduction in generation of inflammation and effector CD4+ T-cell. Effector CD4+ T cells from MIF deleted parasites from infected mice showed a profile of decreased apoptosis, and increased expression of IFN-γ and IL-7R, suggesting that the expression of the orthologue MIF promotes parasite persistence by manipulating the host response to increase the exhaustion and depletion of protective CD4+ T cells60. However, while low dose infection with wild-type L. major parasite is known to result in the acquisition of long-term protection, a process known as leishmanization, after resolution of cutaneous lesions, a persistent infection with L. major is established in the host. Thus, this practice has been discontinued due to safety concerns, mainly related to the pathogenicity eventually resulting from uncured lesions, persistence of the parasite in the lesion and reduced vaccine effect (immunosuppression) in patients immunized with vaccine against diphtheria, Bordetella pertussis and tetanus4,5,11. Based on this concept, Zhang et al.25, using a CRISPR genome edited L. major strain (LmCen−/−), demonstrated that wildtype L. major infected/healed (leishmanization) and LmCen−/− immunized mice presented high percentage of T CD4+ memory cells producing IFN-γ. The low levels of persistent antigens may be important for maintaining long term protection profile, as the generation of IFN-γ producing CD4+ T effector populations25,78, despite the difference between the survival profile of parasites used in leishmanization and immunization with attenuated parasites. Independently TCM and skin resident TRM memory T cells were also shown to play a role in protection in L. major mouse models39,79. Thus, it would be hard to discriminate the role of memory cell populations in a L. major infection model due to the presence of memory and effector populations and the kinetics of their actions following a challenge infection. Therefore, LdCen−/−MIF−/− parasites provide an ideal vector to test the role of memory cells in protection considering all characteristics before described here.

The role of the anti-leishmanial antibody response seen in VL is unclear. Some authors have suggested that the presence of anti-leishmanial antibodies could be predictive of disease80,81,82. In the other hand, it has been demonstrated that antibodies against Leishmania persist for a long time (> 15 years) after cure and immunity to VL83. Moreover, it has a high prevalence of seropositive healthy individuals in areas endemic for VL84. Here, we demonstrated that immunization with the double-attenuated strain LdCen−/−MIF−/− increased the secretion of IgG, IgG1 and IgG2 (regulated by IL-12 induced production of IFN-γ85,86), being able to activate B cells as late as 12 wpc.

The protective immune response against Leishmania is mainly mediated by T cells87, and the proliferation of these cells is an important indicator of vaccine immunogenicity tested in mice and dogs88,88,90. Our previous work has shown that immunization with LdCen−/−, in dogs and mice, induced T cell proliferation upon stimulation with Leishmania antigens15,16,28. Consistent with our previous reports, our new study presented here shows that it also happens in mice for LdCen−/−MIF−/− immunization.

Naïve CD4 T cells post-activation undergo programming for inducible production of cytokines leading to generation of memory cells with various functions. The importance of the Th1/Th2 balance in the outcome of leishmaniasis has been demonstrated by many studies91,91,92,93,95. Peine and collaborators96 described that hybrid Th1/2 cells arise naturally during parasite infections and that the two opposing differentiation programs can stably co-exist in resting memory for months, demonstrating a cell-intrinsic self-limiting mechanism that can prevent excessive inflammation. These facts corroborate our findings with an increased mixed cytokines production by T cells after immunization and a persistent of IL-12 production by double-mutant parasites after challenge leaded to a protective profile seem in this study.

One of the most crucial aspects of vaccination is the understanding of the T cell memory profile required to obtain effective vaccines against parasites and virus. T cell memory is the ability of a population to respond to a challenge, recognizing an antigen by its receptors (TcRs). T cell response occurs after the initial exposure to the antigen, by proliferating and/or expressing molecules capable of mediating an effector reaction. A pertinent memory is associated with protection against infection and/or disease when challenged with a pathogen in experimental model97. Briefly, naïve T cells (TN) respond to antigenic peptides complexed to major histocompatibility complex (MHC) molecules on antigen-presenting dendritic cells (DCs) after an initial antigenic exposure and priming. After priming, part of those cells proliferates and become effector-memory T cells (TEM), losing the molecules and CD45RA in the process, and being held back in secondary lymphoid tissues (SLT)98. Prompt protection is conferred by tissue resident or circulating effector memory T cells (TEM) that survey frontline barriers and affected tissues for incoming pathogens and exhibit immediate effector role upon antigen recognition. Another portion of primed cells become central memory T cells (TCM), responsible for recalling responses and to patrol the T cell areas of secondary lymphoid tissues, where they can quickly proliferate in response to antigens98. Those circulating quiscent cells are a provision that can respond to the re-encounter with an antigen within SLT by proliferating and differentiating into TEM and TEff cells over the course of few days. Central memory T cells (TCM) are considered better applicable to protect against pathogens with longer incubation periods, such as protozoans97. Studies testing single or polyproteins recombinant proteins from Leishmania showed that immunization successfully generate antigen-specific cells that exhibit characteristics of TCM, cytokine production upon antigen re-exposure and increased Th1 response upon challenge compared to nonimmunized animals43,99,99,100,102. In addition, it has been demonstrated that parasites can use MIF ortholog to actively modulate the host immune response, preventing the development of effective memory CD4+ T cells61. Our data demonstrated that immunization with double-attenuated parasites induced TCM cells, while PBS control and LdCen−/− groups presented high percentage of TEM cells after immunization. Interestingly, the percentage of TCM cells of LdCen−/−MIF−/− group after challenge decreased while TEM Late cells increased. It suggests a conversion of the memory subtype using the double-deletion mutant parasites. Thus, the deletion of MIF gene in LdCen−/− attenuated parasites could yield a long-lasting immune response, suggested by the increase of TCM cells after immunization.

The evaluation of parasite load allows us to visualize not only number of parasites, but also viability of the parasite. The level of parasite burden in spleen and liver observed in both LdCen−/− and LdCen−/−MIF−/− immunized groups is decreased compared to the positive control (animals immunized with PBS and challenged) group at 12wpc, suggesting a robust degree of protection. Therefore, the protection obtained in the present study confirms the ability of LdCen−/−MIF−/− and LdCen−/− vaccines to limit parasite replication and prevent severe disease after challenge. Okwor and colleagues103, evaluating the differences in the immune responses to live and killed L. major in experimental model, have demonstrated that both are qualitatively different. The data demonstrated that live attenuated parasite induced strong and durable protection against virulent secondary challenges103, indicating that is a good way to achieve protective immunity against Leishmania infection by vaccination. In our previous work, vaccination with the attenuated parasite proved to be safe, protective and persistent in mice (BALB/c and SCID), hamsters and dogs, after challenge with wild forms, in addition to cross-protection in animals challenged with L. braziliensis, L. infantum14,15,16,17,18,27,40. In addition, LdCen−/−MIF−/− immunized mice appear to clear infection in spleen and liver sooner than those immunized with LdCen−/− parasites.

Overall, the results indicate that the combination of deletion of Centrin and MIF genes produced a prominent immunological response, inducing central memory T cells (long-term immune response) after immunization, T and B cell activation, balanced cytokine production and protection against challenge with wild type strain. Despite the deletion only for MIF gene does not seem to affect growth or replication of the parasite, it does seem to affect the virulence factor. These finding points to the fact that the induction of the profile of memory cells are necessary for a protective response and provide novel insights into developing vaccines against pathogens. The results indicate that LdCen−/−MIF−/− attenuated parasites are potential candidates for the development of an attenuated vaccine against leishmaniasis.

Methods

Expression of MIF proteins in E. coli

Leishmania donovani MIF1 and MIF2 ORFs were amplified by PCR and ligated into pCR T7/CT-Topo (Invitrogen). The recombinant proteins were expressed from E. coli and purified in native conditions through Ni-agarose column chromatography.

MIF induced TNF-α and IL-12 production by BMDCs

BMDCs were incubated in presence of purified recombinant LdMIF1 and LdMIF2 proteins (10 ng/ml) and LPS (0.25 µg/mL) for 24 h and the culture supernatants were used for measuring TNF-α by ELISA. To test the production of IL-12 by BMDCs in presence of recombinant LdMIF1 protein, cells were incubated in presence of an increasing concentrations of rLdMIF1 (0–100 ng/ml) for 24 h and IL-12 production in the culture supernatants was measured by ELISA. The experiment was performed four times (n = 3/replication), and cells were pooled.

Deletion of MIF genes

Leishmania infantum genome contains two homologs of MIF gene on chromosome 33 (LINF_330025900 and LINF_330026000) each encoding a 342 bp ORF. The drug resistance markers Blasticidin and Puromycin were used to obtain LdCen−/−MIF−/−. The PvuII restriction site in the Blasticidin ORF was altered by PCR prior to incorporating in the targeting construct. To generate the targeting construct, a 462 bp fragment from the 5′ region and an 834 bp fragment from the 3′ region flanking the L. infantum MIF open reading frames were amplified by PCR using L. infantum genomic DNA. The primers used to amplify 5′ flanking fragment included restriction sites HindIII and BamHI. Similarly, the primers added SpeI and XbaI sites to the 3′ flanking fragment. The drug resistance markers blasticidin (BSD) and puromycin (PAC) were amplified with primers that add BamHI and SpeI to the open reading frame. These DNA fragments were subcloned into the pCR2.1-Topo vector and the nucleotide sequence was determined to ensure fidelity. The plasmid containing the 5′ flanking fragment was digested with HindIII/BamHI, gel purified and ligated into a similarly digested plasmid containing either blasticidin or puromycin. The resultant plasmids, containing both the 5’flanking region and the drug resistance markers were digested with SpeI/XbaI and the 3’flanking fragment isolated by SpeI/XbaI digestion was ligated into these sites. The authenticity of the final plasmid was confirmed by DNA sequencing. For the purpose of transfection, the targeting construct was prepared by digestion with HindIII/XbaI, which releases a linear fragment containing the MIF 5′ flanking sequence, the blasticidin/puromycin encoding DNA fragments and the MIF 3′ flanking sequence. The fragment was gel purified and used in transfection. SalI digested genomic DNA from LdWT, LdCen−/−, LdCen−/−MIF−/− parasite clones selected on Nobel agar plates was resolved on agarose gels. The blots from these gels were probed with 32P labeled probes corresponding to Centrin, Neomycin, Hygromycin, MIF, Puromycin and Blasticidin ORFs. The 168 bp MIF probe selected corresponds to c’end of the ORF and is common to both MIF1 and MIF2 genes. The visualization was made by autoradiograph. Figures 2B, 4A,B were visualized in different days, but visualized by autoradiograph with no exposure differences.

IFN-γ expression

Balb/C mice were intravenously infected with 3 × 106 stationary phase LdWT or LdMIF−/− parasites and spleens were collected on 5- and 7-days post-infection. The cells were stimulated with 0.25 µg/mL of lipopolysaccharide (LPS). Expression of IFN-γ was measured by ELISA from the culture supernatants. The experiment was performed four times (n = 3/replication), and cells were pooled.

Macrophage infection

C57BL/6 murine bone marrow derived macrophages were cultured in RPMI medium containing 10% FBS macrophage colony-stimulating factor (20 ng/ml, ProSpec, Israel), plated in 0.5 ml on eight-chamber Lab-Tek tissue-culture slides (Miles Laboratories). The differentiated macrophages were infected with LdWT, LdCen−/−, LdMIF−/− and LdCen−/−MIF−/− stationary phase promastigote cultures (10:1 parasite-to-macrophage ratio). Free extra cellular parasites were aspirated after 6 h incubation at 37 °C in 5% CO2 and the cultures were incubated in macrophage culture medium for 7 days. The parasite counts were measured by using Diff-Quick Stain (Baxter Healthcare Corporation). A minimum of 300 macrophages were counted. The results are expressed as the number of amastigotes per 100 macrophages. The experiment was performed four times (n = 3/replication), and cells were pooled.

T cell apoptosis

Balb/C mice were infected intravenously with 3 × 106 stationary phase LdWT, LdMIF−/− or LdCen−/−MIF−/− parasites. Spleens were collected on 5-, 9-, and 14-days post-infection. Splenocytes were stained with 7AAD, CD3-Alexaflour700, CD4-Pacific blue, CD8-BV650 and Annexin-V-PE antibodies and analyzed on BD LSR-Fortessa.

Parasites and soluble antigen (SLA) preparation

The L. donovani centrin1-deleted (LdCen−/−) and centrin1 and MIF-deleted (LdCen−/−MIF−/−) parasites were used for immunization and maintained as previously described23. L. infantum promastigote forms (MHOM/BR/1972/PP75) were grown as described104. For preparation of SLA, L. infantum stationary-phase promastigotes were harvested, washed three times in PBS and ruptured using a cell disruptor (Sonifier Cell Disruptor, Branson Sonic Power Co., Danbury, CT, USA). The ruptured parasite suspension was centrifuged at 18,500 rpm for 90 min at 4 °C. The supernatant was dialyzed against PBS for 24 h and sterilized by filtration through 0.22 µm syringe filters and stored at − 80 °C. Protein quantification was performed using Pierce® BCA Protein Assay Kit (Thermo Scientific, USA) as described by the manufacturer.

Animals and vaccination protocol

Animal studies were carried out in accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and the guidelines set by the Brazilian Animal Experimental College (COBEA). For studies in Brazil, the protocol was reviewed and approved by the Ethical Committee for the Use of Experimental Animals of the Oswaldo Cruz Foundation (CEUA/FIOCRUZ Protocol LW35/14). The animal protocol for the studies in USA has been approved by the Institutional Animal Care and Use Committee at the Center for Biologics Evaluation and Research, US Food and Drug Administration (FDA) (ASP 1995#26). In addition, the animal protocol is in full accordance with “The guide for the care and use of animals as described in the US Public Health Service policy on Humane Care and Use of Laboratory Animals 2015.” This study is reported in accordance with ARRIVE guidelines. All animals were certified free of endo/ectoparasites by animal house logistics. Female 6- to 8-week-old C57Bl/6 (Jackson labs) mice were immunized or infected, intravenously with 3 × 106 total stationary phase promastigotes of LdMIF−/− or LdWT parasites in 10 μl PBS. The mice, with 5–7 weeks of age were divided into three groups (8 animals per group/time). LdCen−/− and LdCen−/−MIF−/− groups received intravenously 3 × 106 LdCen−/− or LdCen−/−MIF−/− promastigotes at stationary phase, respectively. Control group received PBS alone. Four weeks after immunization, all animals (including PBS group) were challenged with 3 × 106 L. infantum parasites. The immunological parameters were measured 4 weeks post-immunization (4wpi), 4, 8 and 12 weeks (wpc) after the challenge with 3 × 106 of stationary phase promastigotes of L. infantum intravenously, as demonstrated in Supplementary Fig. 1. Regarding antibody response, T cell proliferation, intracellular cytokine measurement, subtype of memory and parasite load experiments, we performed the experiments three times. The BALB/c mice (n = 8/group/time point, three replicates). were individually assessed, using the same animals for all those experiments mentioned before.

Antibody responses

Antigen-specific IgGTotal, IgG1 and IgG2 levels were measured by indirect ELISA64. Briefly, 96 wells micro titer plates (Nalgen Intl., USA) were coated overnight with 5 µg/mL of SLA. For IgGTotal, IgG1 and IgG2 analysis, sera were added at a 1:100 dilution. Peroxidase-conjugated rabbit anti-mouse IgGTotal (1:3000), IgG1 (1:2000) or IgG2 (1:1000) antibodies were added for 1 h. The reaction was developed using TMB (Sigma, USA) and H2SO4 stop solution was used. Absorbance was measured on VersaMax 340PC microplate reader (Molecular Devices, USA) at 450 nm.

Flow cytometric analysis of phenotypic profile and intracytoplasmic cytokine production

Spleens were sterilely removed, and single cell suspensions prepared. Mononuclear cells were enumerated using a Countess Automated Cell Counter (Thermo Fischer, Invitrogen, MA). Cells were cultured at 2 × 105 cells per well in duplicate in a 96 wells plate (Corning Incorporated, Corning, NY) in RPMI-1640 supplemented with 10% heat-inactivated FBS, 50,000 Units penicillin/streptomycin (Invitrogen) and 1% l-glutamine (Gibco). The cells were incubated in the presence or absence of 25 µg/mL SLA at 37 °C in 5% CO2 for 72 h. On the last 4 h, cultures received brefeldin A (10 µg/mL). Cells were stained for the surface markers, all purchased from Biolegend (CA,USA): (a) CD3 FITC (clone 145–2C11), (b) CD3 AlexaFluor700 (clone 17A2), (c) CD4 PerCP-Cy5.5 (clone RM4-4), (d) CD8 BV421 (clone 53–6.7), (e) CD62L BV605 (clone MEL-14), (f) CD27 PE-Cy7 (clone LG.3A10), (g) CD25 BV510 (clone PC61), (h) GATA3 AlexaFluor647 (clone 16E10A23), (i) T-bet PE-Cy7 (clone 4B10). Cells were then fixed, permeabilized and stained for the following cytokines: (a) IL-4 (clone 11B11), (b) IL-12p40 (clone C15.6), (c) IFN-γ (clone XMG1.2), (d) TNF-α (clone MP6-XT22), (e) IL-5 (clone TRFK5), (f) IL-17A (clone TC11-18H10.1) and (g) IL-10 (clone JES5–16E3), all PE. For each sample, at least 100,000 cells were analyzed. The data were analyzed using FlowJo software and a FACSFortessa flow cytometer (both from Becton Dickinson, San Jose, CA).

In vitro proliferative response of lymphocytes

Splenocytes were isolated as described above. After 72 h incubation, cell proliferation analysis was performed on splenocytes labeled with Ki-67 APC (clone16A8, Biolegend, CA) essentially as described above, and analyzed the stimulation index using monoclonal antibodies CD3, CD4 and CD8. In this sense, proliferation responses were expressed in terms of stimulation ratio that was calculated as: mean proliferation response of cultures stimulated SLA L. infantum/mean proliferation response of unstimulated cultures as described previously15.

Determination of parasite burden

At 4 weeks post immunization, and 4-, 8- and 12-weeks post challenge, the parasite load was measured in the spleen and liver by the limiting dilution assay as previously described105.

Statistical analysis

Statistical analysis was performed using GraphPad Prism 6.0 software (GraphPad Software Inc, USA). Non-parametric Kruskal–Wallis test followed by Dunns test was used to compare data from all three groups (LdCen−/−, LdCen−/−MIF−/− and PBS). Differences were considered significant when a p value ≤ 0.05 was obtained.