Recombinant Leishmania major secreting biologically active granulocyte-macrophage colony-stimulating factor survives poorly in macrophages in vitro and delays disease development in mice

Abstract

Leishmania is an intracellular pathogen that replicates inside macrophages. Activated macrophages produce a specific subset of cytokines that play an important role in the control of Leishmania infections. As part of our interest in developing suicide parasites that produce abortive infections for the purposes of vaccination, we engineered recombinant Leishmania major strains producing biologically active granulocyte-macrophage colony-stimulating factor (GM-CSF). We showed that GM-CSF is being produced in the phagosomes of infected macrophages and that it can be detected in the culture supernatants of both infected macrophages and extracellular parasites. Our data support the notion that GM-CSF secreted by both developmental forms of recombinant L. major can activate macrophages to produce high levels of proinflammatory cytokines such as interleukin-1beta (IL-1beta), IL-6, and IL-18 and various chemokines including RANTES/CCL5, MIP-1alpha/CCL3, MIP-1beta/CCL4, MIP-2/CXCL2, and MCP-1/CCL2, which enhance parasite killing. Indeed, GM-CSF-expressing parasites survive poorly in macrophages in vitro and produce delayed lesion development in susceptible BALB/c mice in vivo. Selective killing of intracellular Leishmania expressing cytokine genes capable of activating cellular responses may constitute a promising strategy to control and/or prevent parasitic infections.

FIG. 1.

Recombinant L. major expressing GM-CSF as part of an episomal vector. (A) Expression vectors pNEO-mGM-CSF and pNEO-hGM-CSF were made by inserting the murine mGM-CSF and hGM-CSF genes, respectively, downstream of the neomycin phosphotransferase (NEO) expression cassette as described in Materials and Methods. Arrows indicate the orientation of GM-CSF transcript maturation. In pNEO-hGM-CSF/R, the hGM-CSF gene is cloned in the opposite orientation relative to mRNA maturation signals. Bg, BglII. (B) Southern blot of L. major genomic DNA digested with BglII and hybridized to the murine and human GM-CSF-specific probes. Lanes: 1, L. major pNEO control; 2, L. major expressing either the mGM-CSF or the hGM-CSF gene. (C) GM-CSF expression in L. major GM-CSF recombinant cells. A Northern blot of total Leishmania RNA hybridized to the same probes as above. Lanes 1 and 2 are as in panel B. Lane 3, L. Major expressing the hGM-CSF gene.

FIG. 2.

Stable integration of the GM-CSF gene into the L. major LV39 genome. (A) Schematic representation of the L. major rRNA locus with the RNA pol I promoter indicated by a flag. The GM-CSF expression cassette was inserted downstream of the rRNA promoter for stable and high expression. The EcoRV (EV) and BamHI (B) restriction sites and the size of the restriction fragments are indicated. (B) Southern blot hybridization of the L. major LV39 wild-type strain (lane 1) and of L. major mGM-CSF recombinant strain made by the integration of the mGM-CSF gene into the ribosomal locus (lane 2) with probes specific to the rRNA promoter sequence and the mGM-CSF gene, respectively. The EcoRV restriction fragments confirming the correct integration of the GM-CSF gene into the parasite genome are indicated by arrows. (C) Clamped homogeneous electric field electrophoresis and Southern blot hybridization of L. major chromosomes (lanes and probes are as indicated in panel B). The ribosomal locus is part of chromosome 27. (D) Northern blot hybridization of L. major total RNA (lanes are as indicated in panel B) with the GM-CSF gene probe. RNA loading was monitored by ethidium bromide staining (lower panel).

FIG. 3.

GM-CSF release by intracellular Leishmania expressing the GM-CSF gene. (A) CM-CSF production and release within macrophages at different time points following infection of J774 macrophages with L. major expressing the episomal form of the mGM-CSF gene (□) or with L. major wild type (•). The amount of GM-CSF in lyzed macrophages was measured by ELISA. (B) CM-CSF release in the extracellular milieu of macrophages infected with L. major expressing the episomal form of the mGM-CSF gene (▴) or with L. major wild type (○). The amount of GM-CSF in the culture supernatants of infected macrophages was measured by ELISA. Dilutions of both the capture and biotinylated antibodies were at 1/250.

FIG. 4.

GM-CSF detection by immunofluorescence within macrophages infected with GM-CSF-producing parasites. After overnight infection of J774 murine macrophages with either L. major LV39 expressing the integrated mGM-CSF gene or L. major expressing the episomal GM-CSF vector, macrophages were washed, fixed, and then incubated with the monoclonal anti-mouse GM-CSF antibody and Alexa Fluor 546, as indicated in Materials and Methods. Coverslips were then mounted and sealed on the microscope slide for confocal observation. Color images were created with Fluoview 300 version 3.3 software, and color contrasts were adjusted using Photoshop version 6.0 software. (A) macrophages infected with L. major wild type (dilutions are at 1/250 for the first antibody and 1/1,000 for the second antibody), (B) macrophages infected with L. major expressing the episomal form of the mGM-CSF gene (dilutions are at 1/250 for the first antibody and 1/1,000 for the second antibody), (C) uninfected macrophages (dilutions are at 1/1,000 for the first antibody and 1/2,000 for the second antibody), and (D) macrophages infected with L. major LV39 expressing the integrated form of the mGM-CSF gene (dilutions are at 1/1,000 for the first antibody and 1/2,000 for the second antibody).

FIG. 5.

Intracellular killing of L. major amastigotes expressing the mGM-CSF and/or the hGM-CSF gene. J774 murine macrophages and/or human monocytes differentiated to macrophages (5 × 10⁴ cells/well) were incubated for 6, 48, or 72 h with late-stationary-phase L. major promastigotes (parasite-to-cell ratio, 20:1) expressing either the mGM-CSF or the hGM-CSF gene as described in Materials and Methods. At these fixed time points, cell cultures were dried and stained with Diff Quik to determine the level of infection. (A) Infection of murine macrophages with the L. major pNEO control, L. major mGM-CSF expressing the episomal mGM-CSF gene, and L. major mGM-CSF incubated with an anti-GM-CSF neutralizing antibody prior to macrophage infection. (B) Infection of human monocytes with L. major pNEO, L. major hGM-CSF, and L. major hGM-CSF/R (with the human GM-CSF gene cloned in the reverse orientation). The results shown here are the mean of three independent experiments with duplicate samples and the standard errors of the mean.

FIG. 6.

Outcome of cutaneous disease in susceptible BALB/c mice infected with late-stationary-phase L. major promastigotes expressing the murine GM-CSF gene. The infection was monitored for up to 7 weeks postinoculation as described in Materials and Methods. Results obtained with mice infected with the L. major NEO control, mice infected with the L. major LV39 recombinant strain expressing the mGM-CSF gene integrated into the 18S ribosomal locus under the control of the rRNA promoter, and mice infected with L. major GM-CSF expressing the episomal form of the mGM-CSF gene are shown. The result shown here is the average mean of three independent experiments.

FIG. 7.

A time course pattern of cytokine expression by J774 macrophages infected with L. major LV39 expressing the integrated mGM-CSF gene. Graphs represent densitometric quantification of bands from cytokine mRNA (IL-1Ra, IL-1β, IL-18, and IL-6) normalized to the GAPDH signal. LPS was used as a control. The relative increase in the levels of the various cytokines is in comparison to macrophages infected with the wild-type parasites. The data presented here are the mean of three independent experiments. Error bars, ± 0.1 to 0.5.

FIG. 8.

Patterns of chemokine expression by J774 macrophages infected with L. major LV39 expressing the integrated mGM-CSF gene. Graphs represent densitometric quantification of bands from chemokine mRNA (RANTES, MCP-1, MIP-2, MIP-1α, and MIP-1β) normalized to the GAPDH signal. The relative increase in the levels of the various chemokines is in comparison to macrophages infected with the wild-type parasites is shown. The data presented here are the mean of three independent experiments. Error bars, ± 0.03 to 0.4.

FIG. 9.

No inhibition in mice infected with GM-CSF-expressing recombinant parasites. BALB/c mice infected either with L. major LV39 expressing the integrated mGM-CSF gene (L.m-GM-CSF) or with the L. major NEO strain (L.m-NEO) were treated with aminoguanidine, a selective inhibitor of inducible iNOS synthesis, and the outcome of the cutaneous infection in BALB/c mice was monitored for up to 4 weeks and compared to that for the untreated animals. The data presented here are the mean of two independent experiments.