Leishmania guyanensis suppressed inducible nitric oxide synthase provoked by its viral endosymbiont

Abstract

Inducible nitric oxide synthase (iNOS) is essential to the production of nitric oxide (NO), an efficient effector molecule against intracellular human pathogens such as Leishmania protozoan parasites. Some strains of Leishmania are known to bear a viral endosymbiont termed Leishmania RNA virus 1 (LRV1). Recognition of LRV1 by the innate immune sensor Toll-like receptor-3 (TLR3) leads to conditions worsening the disease severity in mice. This process is governed by type I interferon (type I IFNs) arising downstream of TLR3 stimulation and favoring the formation of secondary metastatic lesions. The formation of these lesions is mediated by the inflammatory cytokine IL-17A and occurs in the absence, or low level of, protective cytokine IFN-γ. Here, we described that the presence of LRV1 led to the initial expression of iNOS and low production of NO that failed to control infection. We subsequently showed that LRV1-triggered type I IFN was essential but insufficient to induce robust iNOS induction, which requires strong activation of nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB). Leishmania guyanensis carrying LRV1 (LgyLRV1+) parasites mitigated strong iNOS production by limiting NF-kB activation via the induction of tumor necrosis factor-alpha-induced protein 3 (TNFAIP3), also known as A20. Moreover, our data suggested that production of LRV1-induced iNOS could be correlated with parasite dissemination and metastasis via elevated secretion of IL-17A in the draining lymph nodes. Our findings support an additional strategy by which LRV1-bearing Leishmania guyanensis evaded killing by nitric oxide and suggest that low levels of LRV1-induced NO might contribute to parasite metastasis.

Keywords: IL-17A; Leishmania; Leishmania RNA virus 1 (LRV1); inducible nitric oxide synthase (iNOS); metastasis; nuclear factor kappa-light-chain-enhancer of activated B cells (NF-kB); tumor necrosis factor-alpha-induced protein 3 (A20); type I Interferons.

Figure 1

LgyLRV1+ infection induces iNOS expression in murine BMDM. (A) WT BMDMs were infected with LgyLRV1+, LgyLRV1- (MOI 5) or left untreated for 8h. Total RNA were extracted and full RNAseq analysis was performed. Scatter-plot shows all present genes comparing LgyLRV1+ vs non-infected (x-axis) and LgyLRV1+ vs LgyLRV1- (y-axis), (n = 4). (B) WT BMDMs were infected with LgyLRV1+, LgyLRV1- (MOI 10), stimulated with poly (I:C) (2 μg/ml), or left untreated, for 2, 6, 12 or 24h. RT-PCR was used to measure L32-normalized Nos2 expression. (C) WT BMDMs were infected, or treated as in (A). Total cell lysates were analyzed by western blot for iNOS and γ-tubulin expression as loading control at indicated time points. (D) WT BMDMs were infected, or treated as in (A) for 12h. LPS (1μM) and IFN-γ (0.1 U/μl) stimulation were used as controls. RT-PCR was used to measure L32-normalized Nos2 expression. (E) WT BMDMs were infected, or treated as in (C). Total cell lysates were analyzed by western blot for iNOS and γ-tubulin expression as loading control. (F) WT BMDMs were infected, or treated as in (C). Upon 48h supernatants were collected and levels of nitrites were measured using Griess assay. (G) WT or Inos^-/- BMDMs were infected with LgyLRV1+, or LgyLRV1- (MOI 10) for 48h. Levels of parasitaemia were analysed using high content microscopy. Data show mean ± SD from a representative of three independent experiments (A, C, E and F). Representative blots are shown from at least three independent experiments in (C, E) n.i. – non-infected. n.d. - not detected. Data were analyzed using unpaired Student’s t test. Not significant (NS), **p < 0.01, and ****p < 0.0001.

Figure 2

External NF-κB activation boosts LgyLRV1+ triggered iNOS induction and rescues BMDMs from infection. (A) WT BMDMs were infected with LgyLRV1+ or LgyLRV1- (MOI 10), left untreated, or stimulated with poly (I:C) (2 μg/ml) for 12h. Additionally some conditions were treated with FSL-1 (10 ng/ml), or IFN-β (500 U/ml), or left untreated. Total cell lysates were analyzed by western blot for iNOS, P-STAT1(S727), STAT1, P65, P-P65(S536) and γ-tubulin expression. (B) WT BMDMs were infected as in (A). RT-PCR was used to measure L32-normilized Nos2 expression. (C–F) - WT BMDMs were infected as in (A) for 48h. Levels of parasitemia were measured by high content microscopy (F). SNs were collected and levels of nitrites (C), IL-6 (D) and TNF-α (E) were analyzed by Griess assay and ELISA respectively. Data show mean ± SEM (F), or SD (B–E) from a representative of three independent experiments. Representative blot is shown from at least three independent experiments in (A). Data were analyzed using unpaired Student’s t test in F or Sidak’s multiple comparisons (D, E). Not significant (NS), **p < 0.01, and ****p < 0.0001. n.i. – non-infected. n.t. – non-treated. n.d. - not detected.

Figure 3

LgyLRV1+ triggered type I IFNs are necessary but not sufficient in order to induce iNOS in an auto- and paracrine manner. (A) WT, Tlr3^-/- or Ifnar^-/- BMDMs were infected with LgyLRV1+ or LgyLRV1- (MOI 10), stimulated with poly (I:C) (2 μg/ml), FSL-1 (10 ng/ml), IFN-β (500 U/ml), alone or in combination, or left untreated for 12h. Total cell lysates were analyzed by western blot for iNOS and γ-tubulin expression. (B, C) WT, Tlr3^-/- or Ifnar^-/- BMDMs were treated as in (A) for 48h. SNs were collected and levels of nitrites were measured using Griess assay. (D) WT BMDMs were infected with LgyLRV1- (MOI 10), treated with FSL-1 (10 ng/ml) alone or in combination. Except one, all conditions were treated with poly (I:C), with one of eight two-fold serial dilution concentrations ranging from 2 to 0.015 μg/ml. Upon 48h SNs were collected and levels of nitrites were measured using Griess assay. (E) scheme depicting experiments performed in (F, G). (F, G) WT BMDMs were infected with LgyLRV1+ or LgyLRV1- (MOI 10), stimulated with poly (I:C) (2 μg/ml), or left untreated. Upon 6h, SNs were collected, filtered with 22 um filter, supplemented with FSL-1 (10 ng/ml) and added to the fresh set of untreated WT and Ifnar^-/- BMDMs. Upon 24h, SNs from treated BMDMs were collected and levels of nitrites were measured using Griess assay (G). Total cell lysates were analysed by western blot for iNOS and γ-tubulin expression (F). Data show mean ± SD from a representative of three independent experiments (B, C, F). Representative blots are shown from at least three independent experiments in (A, F). the following: ****p < 0.0001, n.i. – non-infected. n.t. – non-treated. n.d. - not detected.

Figure 4

Type I IFNs mediated iNOS does not control LgyLRV1+ infection in vivo. Hind footpads of WT, Inos^-/- , Ifng^-/- and Ifng^-/-Inos^-/- DKO mice were infected with 3 × 10⁶ LgyLRV1+. (A, C) – graphs display weekly measurements of footpad swelling. (B, D) Parasite burden was determined after infection by bioluminescence imaging. Data show mean ± SEM from representative experiments (n = 4–5 mice) of at least three independent experiments. Each data point on (B, D) represents one footpad. Statistical significance is calculated using two-way ANOVA analysis with Bonferonni’s test (A). Not significant (NS), *p < 0.05 and ***p < 0.001.

Figure 5

Lgy infection minimizes iNOS expression by suppressing NF-κB via A20 induction. (A) WT or A20^fl/flLySM^Cre/wt (KO) BMDMs were infected with LgyLRV1+, or LgyLRV1- (MOI 10), stimulated with poly (I:C) (2 μg/ml), or left untreated for 12h. Total cell lysates were analysed by western blot for iNOS, A20, P-P65(S536), P65 and γ-tubulin expression. (B–E) - WT or (KO) BMDMs were treated as in (A) for 48h. Levels of parasitemia were measured by high content microscopy (C). SNs were collected and levels of nitrites (B), IL-6 (D) and TNF-α (E) were analyzed by Griess assay and ELISA, respectively. Data show mean ± SD from a representative of three independent experiments (B–E). Representative blot is shown from at least three independent experiments in (A). Statistical significance is calculated using unpaired Student’s t test in (C) and Tukey’s multiple comparison test in (D) and (E). Not significant (NS), ***p < 0.001, ****p < 0.0001. n.i. – non-infected. n.t. – non-treated. n.d. - not detected.

Figure 6

Ifng^-/-Inos^-/- mice showed delayed metastasis formation compared to Ifng^-/- upon LgyLRV1+ infection. Hind footpads of Ifng^-/- and Ifng^-/-Inos^-/- DKO mice were infected with 3 × 10⁶ stationary phase promastigotes of LgyLRV1+. (A) Photographic images of representative mice showing metastatic lesions on the tail at week 8 post infection. Arrow heads indicate the metastatic nodules. bone destruction. (B) X-ray images of representative mice showing metastatic lesions on the tail at indicated time points. Arrow heads and enlarged image indicate bone destruction. (C) Number of metastatic lesions per mouse. (D–G) Ifng^-/- and Ifng^-/-Inos^-/- DKO mice were infected into hind foot pads with 3 × 10⁶ stationary phase promastigotes of LgyLRV1+ or LgyLRV1- as indicated. At 6- and 8-weeks post infection, lymphocytes were extracted from popliteal LNs draining the primary lesion and then re-stimulated ex vivo with UV-irradiated promastigotes of LgyLRV1+ or LgyLRV1- respectively. Upon 72h, IL-17A secretion was quantified by ELISA in cell-free supernatants. Data show mean ± SEM from three independent experiments with at least 3 mice per group for LgyLRV1+ infection and from two independent experiments with at least 3 mice per group for LgyLRV1- infection. Statistical significance is calculated using 2-way Anova test in (C) and Student’s t test in (D–G). Not significant (NS), **p < 0.01, and ***p < 0.001.