Exosome Secretion by the Parasitic Protozoan Leishmania within the Sand Fly Midgut

Abstract

Despite several studies describing the secretion of exosomes by Leishmania in vitro, observation of their formation and release in vivo has remained a major challenge. Herein, we show that Leishmania constitutively secretes exosomes within the lumen of the sand fly midgut through a mechanism homologous to the mammalian pathway. Through egestion experiments, we demonstrate that Leishmania exosomes are part of the sand fly inoculum and are co-egested with the parasite during the insect's bite, possibly influencing the host infectious process. Indeed, co-inoculation of mice footpads with L. major plus midgut-isolated or in-vitro-isolated L. major exosomes resulted in a significant increase in footpad swelling. Notably, co-injections produced exacerbated lesions through overinduction of inflammatory cytokines, in particular IL-17a. Our data indicate that Leishmania exosomes are an integral part of the parasite's infectious life cycle, and we propose to add these vesicles to the repertoire of virulence factors associated with vector-transmitted infections.

Figure 1. Leishmania constitutively releases vesicles in the lumen of the sand fly midgut

L. infantum- and L. major-infected midguts were processed for TEM. (A, E, F) Promastigotes secreting single membrane vesicles (red arrows, insets). (B, F) Promastigotes secreting several membrane vesicles (red arrows, inset). (C, G) MVBs fused to the parasite’s membrane, releasing vesicles (red arrows) into the midgut. (D, H) Flagellar pockets containing vesicles (green arrows). (D) Vesicles released into the sand fly midgut (red arrows). The images are representative of 3 independent experiments. K, kinetoplast, N, nucleus, F, flagellum, FP, flagellar pocket, MV, microvillar structures, MVB, multivesicular bodies. See also Figures S1, S2 and S3.

Figure 2. The proteome of in vivo-secreted Leishmania vesicles

Midgut lavages were separated from the infected pellet (IMP) and the uninfected pellet (UMP). Vesicles were isolated from the supernatants (infected midgut vesicles, IMV; uninfected midgut vesicles, UMV). (A) Midgut-isolated L. infantum vesicles. (B) Midgut-isolated L. major vesicles. (C) In vitro-isolated L. major exosomes. (D) Distribution of the proteins from midgut-isolated L. infantum vesicles identified by databases (db). (E) Peptide fold-change for proteins belonging to IMV (LIME) compared to IMP. (F) The LIME interaction network. Clusters: I. Antioxidant and metabolic proteins. II. Chaperones and proteasome proteins. III. Vesicular trafficking and microtubule-associated proteins. IV. Translation proteins. (G) Western blots of: UMP and L. infantum (L.i.) IMP; uninfected and L.i-infected midgut lavages before (UML, IML) or after vesicle purification (UMV, IMV); in vitro-purified L. major (L.m.) exosomes (Exo) and L.m. whole cell lysates (Pro); L.m. IMVs (2 experiments). The results are representative of 3 independent experiments. See also Figures S4, S5 and S6.

Figure 3. The sand fly inoculum contains Leishmania exosomes

L. infantum- and L. major-infected sand flies where fed in vitro through a chicken-skin membrane. Sand fly egests where collected and analyzed for exosomes by TEM and western blot. (A) Exosome vesicles isolated from L. infantum and L. major inocula. (B) Western blots of: vesicles isolated from L. infantum (Li) and L. major (Lm)-infected sand fly egests (4 experiments each) compared to L. infantum and L. major egested parasites (LiP and LmP), in vitro-isolated L. major exosomes (LmE) and uninfected sand fly inocula (Un). The results are representative of 8 independent experiments (4 experiments for each Leishmania species). See also Figure S7.

Figure 4. Injection of Leishmania alongside in vivo- or in vitro-isolated exosomes significantly enhances mice footpad lesions

BALB/c mice were infected with 10⁵ L. major metacyclics (META) or with 5 × 10⁶ L. major stationary promastigotes (STAT) with or without 1 µg of midgut-isolated L. major exosomes (Gut) or 1 µg (A) and 10 µg (B) of in vitro-isolated L. major exosomes (Exo). (A, B) Lesions were monitored up to 10 weeks post-infection. (C–E) STAT and META groups from (B) on the 6^th and 10^th week post-infection, respectively. (C) Footpad images. (D) Lesion volumes. (E) Footpad parasite loads. Each data point represents the average + SEM of 6 to 8 mice per group and statistical significance at + p≤0.05, * p≤0.01, ** p≤0.001, *** p≤0.0001, ns, non-significant. FP, footpad. The results are representative of 1(A) or 2 (B–E) independent experiments.

Figure 5. Exosome integrity is important for the enhancing effect on lesion development

BALB/c mice were infected with 5 × 10⁶ L. major stationary promastigotes (STAT) with or without 10 µg of in vitro-isolated L. major exosomes, sonication-disrupted exosomes (D), disrupted and boiled exosomes (D/B) or with exosomes alone (Exo). Exosome preparations were prepared for TEM or submitted to SDS-PAGE before (total) and after ultracentrifugation (pellet). Their protein content was compared to whole parasite lysates (Pro). (A) Intact and disrupted exosome structures. (B) Top, total protein silver staining. Bottom, HSP83 and GP63 western blots. (C) Lesions were monitored up to 7 weeks post-infection. (B) Footpad parasite loads at week 7. (C) Footpad images at week 7. Each data point represents the average + SEM of 6 to 8 mice per group and statistical significance at + p≤0.05, * p≤0.01. FP, footpad. The results are representative of 2 independent experiments.

Figure 6. Leishmania exosomes exacerbate disease outcome by enhancing cytokine expression

(A) BALB/c mice were infected with 5 × 10⁶ L. major stationary promastigotes (STAT) with or without in vitro-isolated L. major exosomes (0.1, 1 or 10 µg) and lesions were monitored. (B) Footpad parasite loads at week 6. (C) Images of footpads co-injected or not with 1 µg of exosomes 5 weeks post-infection. (D) Cytokine mRNA levels in draining lymph nodes 5 weeks post-infection. Symbols on top of each bar represent p values compared to the STAT group. Each data point represents the average + SEM of 4 to 8 mice per group and statistical significance at + p≤0.05, * p≤0.01, ** p≤0.001, *** p≤0.0001. FP, footpads, NI, non-infected, L.m., Leishmania major. The results are representative of 2 independent experiments. See also Figure S7.