Published online before print March 17, 2008
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,1,,1,2
* Department of Medicine, University of Texas, Health Science Center, San Antonio, Texas, USA;
Centro Internacional de Entrenamiento e Investigaciones Medicas-CIDEIM, Cali, Colombia; and
Research Service, Department of Veterans Affairs Medical Center, South Texas Veterans Health Care System, San Antonio, Texas, USA
2Correspondence: Department of Medicine, University of Texas, Health Science Center, 7703 Floyd Curl Dr., Mailcode 7881, San Antonio, TX 78229-3900, USA. E-mail: travi{at}uthscsa.edu
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: Leishmania • hamster • estrogen
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
On the other hand, several studies have suggested that heightened innate immunity plays a central role in the maintenance of host defense throughout gestation [1 , 12 ]. Pregnant women have a higher number of activated peripheral blood monocytes and granulocytes than nonpregnant women [13 14 15 ]. Innate immune function in leishmaniasis is of particular importance, as the effector cells, i.e., macrophages and to some extent, neutrophils, serve as host cells for the parasite [16 , 17 ]. However, to our knowledge, there are no studies that address the role of these cells in innate host defense against Leishmania infection in pregnant humans or experimental animals.
In the last decades, an increasing body of evidence has established the relationship between sex hormones and the immune system. Previous studies demonstrated that increased concentrations of estrogen led to enhanced macrophage phagocytic activity [18 ] and that estrogen as well as progesterone and prolactin stimulated cytotoxic mechanisms through the release of reactive oxygen species [19 , 20 ]. The production of NO, which is critical to Leishmania containment in the mouse model [21 ], is also enhanced by the high physiological levels of estrogen and progesterone in pregnant females of different species. In fact, NO is critical for controlling uterine contractility, cervical ripening, and feto-placental blood flow [22 23 24 25 26 ]. In pregnant rats, progesterone seems to act synergistically with NO, playing an important role in embryo implantation [27 , 28 ].
Experimental studies in hamsters infected with L. infantum indicated that females were protected during lactation [29 ]. Furthermore, female mice or hamsters were shown to be more resistant to cutaneous leishmaniasis (caused by Leishmania mexicana and Leishmania (Viannia) panamensis, respectively) than males, and this difference depended, at least in hamsters, on the levels of circulating sex hormones [30 , 31 ]. In this study, we explored the impact of pregnancy and its related hormones on the innate immune response to experimental cutaneous infection with L. panamensis. For this purpose, we used an established hamster model [31 , 32 ] of American cutaneous leishmaniasis and demonstrated through in vitro and in vivo experiments that early parasite control in pregnancy was associated with an estrogen-mediated increase in NO production by macrophages and neutrophils.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-Neomicin- plasmid as described [33
]. Dr. Marc Ouellette (University of Laval, Canada) kindly provided the LUC reporter plasmid. The transfected Leishmania was selected and cultured in complete Schneider’s culture medium supplemented with 10% FCS, 2 mM glutamine, 100 units/mL penicillin, 100 µg/mL streptomycin, and 0.04 mg/mL G418 neomycin. The virulence and pathogenicity of the transfected Leishmania were determined in previous studies [34
]. No significant plasmid loss was observed up to 2 months post-infection (p.i.) in the strain used in the present study.
Parasite burden
Parasite burden in the lesions was determined at 7 days p.i. by luminometry [33
] using a modified method as described by Henao et al. [34
]. Briefly, the whole lesion was excised and immediately homogenized by gentle and thorough scraping of the dermis using a scalpel; subsequently, the tissue was lysed by incubation for 30 min at room temperature with 200 µL 5x lysis buffer (125 mM Tris-HCl, pH 7.8, 10 mM DTT, 50% glycerol, and 5% Triton X-100) and stored at –70°C until processing. After thawing, the lysate was centrifuged at 10,000 rpm, and 20 µL of the supernatant was dispensed in an opaque, white 96-well plate. Fresh assay buffer (80 µL; 25 mM Tris-HCl buffer, pH 7.8, 2.67 mM MgSO4, 0.1 mM EDTA, 0.53 mM ATP, 33.3 mM DTT, 4.7 µM D-luciferin) was added to the plate immediately before reading in a luminometer (Anthos, Austria; 12 wells per reading) using an integration time of 20 s at 37°C and the automatic addition of 100 µL assay buffer. The number of parasites was extrapolated from a linear standard curve generated with LUC-transfected amastigotes [34
].
Infection of animals
Outbred adult (4 months old), female hamsters were used in the studies. Animals were maintained according to the Guiding Principles for Biomedical Research Involving Animals (Council for International Organizations of Medical Sciences), the Colombian Law 84 of 1989 of the "Estatuto Nacional de Protección de los Animales," and the Resolution #008430 of 1993, which complements Law 84. These studies were reviewed and approved by the Institutional Animal Care and Use Committee of the Centro Internacional de Entrenamiento e Investigaciones Medicas (CIDEIM; Columbia). Pregnant hamsters were obtained after mating with males during two oestral cycles (8 days). Seven days after mating, female hamsters and nonmated control females were inoculated intradermally in the snout with 104 wild-type or LUC-transfected L. panamensis stationary-phase promastigotes suspended in 0.05 mL PBS. Evaluation of the infection during pregnancy was carried out at 7 days p.i. or post-parturition at 30 days p.i. Animals were killed in a CO2 chamber following anesthesia with 50 mg/Kg ketamine clorhydrate (Ketamina®, Holliday-Scott S.A., Argentina) and 0.02 mg/Kg xylazine (Rompun®, Bayer, Colombia). In animals killed prior to parturition, pregnancy was confirmed by visual inspection of the uterus.
Cytokine and inducible NO synthase (iNOS; NOS2) mRNA expression in pregnant hamsters
Cytokine and iNOS mRNA expression was determined ex vivo by RT-PCR using lesion tissue or draining lymph nodes, cultured with leishmanial antigen (1x106 freeze-thaw, inactivated promastigotes) during 20 h at 37°C, 5% CO2, in RPMI and 5% FCS. Evaluations were made at 7 or 30 days p.i. The primer sequences were derived from published sequences [35
, 36
] and validated for use in RT-PCR previously [31
, 32
]. In addition, specific primers for TNF- (forward 5'-CACAATCCTCTTCTGCCTGC-3' and reverse 5'-TGTCTTTGAGAGACATCCCG-3'; expected product, 242 bp) were used. RNA extraction and RT were performed in a final volume of 120 µL as described by Osorio et al. [32
]. The cDNA (10 µL per sample, equivalent to 
200 ng reverse-transcribed RNA) was denatured at 94°C during 2 min and amplified by PCR (94°C 15 s, 55°C 30 s, and 72°C 1 min, with a final 3-min extension at 72°C) for 30 cycles. Less abundant transcripts (IL-4 and IL-12p40) were subjected to 35 amplification cycles. iNOS expression was determined by RT-PCR in lesion tissue and cells from draining lymph nodes stimulated with Leishmania antigen, using primers specific for hamster iNOS: forward (exon 12) 5'-ACCACACAGCCTCCGAGTCC-3' and reverse (exon 13) 5'-CTGCCAGATGTGGGTCTTCC-3'; expected product, 200 bp. The expression of iNOS and cytokine (IL-4, TNF-, TGF-β, IFN-
, IL-10, IL-12p40m) mRNA was analyzed by densitometry using image analysis (Gel Doc®, Bio-Rad, Hercules, CA, USA) and normalized to the expression of the hypoxanthine phosphoribosyltransferase gene (HPRT).
Identification of cell populations by flow cytometry
Lymphocytes, mononuclear cells, and granulocytes were defined by the size and granularity of cells [forward-scatter/side-scatter (FSC/SSC)] on a FACSCalibur flow cytometer (Becton Dickinson, San Jose, CA, USA). CD4+ T lymphocytes and B lymphocytes from peripheral blood were identified using FITC-labeled anti-mouse mAb that cross-react with the corresponding hamster molecules (GK1.5 and 14-4-4S, PharMingen, San Diego, CA, USA, respectively) [37
, 38
]. Neutrophils were obtained from pooled, EDTA-treated hamster peripheral blood following a standard precipitation method with Dextran T-500 (Pharmacia, Uppsala, Sweden) [39
]. Cells were then sorted by size and granularity (SSC and FSC) on a FACSCalibur flow cytometer (Becton Dickinson). All neutrophil preparations had >99% purity confirmed by Giemsa stain and direct microscopy.
Leishmanicidal activity by flow cytometry
To estimate the proportion of cells containing viable parasites, we infect the cells with promastigotes previously labeled with 1 µM CFSE dye (Molecular Probes, Eugene, OR, USA). After infection (20 min, 1 h, 34°C, 5% CO2), viable CFSE Leishmania were detected by gating in the cell population, excluding small FSC/small SSC extracellular parasites, and analyzing the positive region of the fluorescence 1 (FL1) channel. Flow cytometry analysis was accomplished using the CellQuest (BD Biosciences, San Jose, CA, USA) program.
Parasite replication in macrophages
Resident peritoneal macrophages were obtained from pregnant or control hamsters by peritoneal lavage with RPMI-EDTA medium. To quantify parasites replicating within macrophages, 106 resident peritoneal macrophages were distributed in 24-well plates and incubated for 24 h at 37°C, 5% CO2. Adherent macrophages were infected with 5 x 106 LUC-transfected L. panamensis promastigotes and cultured at 34°C, 5% CO2, for 2 h. After washing away extracellular parasites with warm Dulbeco’s PBS, macrophages were cultured for an additional 72 h at 34°C and 5% CO2 to allow Leishmania replication. Infected host cells were carefully detached with the plastic plunger of a tuberculin syringe in cold PBS, counted, adjusted to 100,000 cells, and lysed in 5x LUC lysis buffer. The number of intracellular parasites was determined by luminometry as described above.
NO determination
The fluorescent probe 4-amino-5-methylamino-2',7'-difluorescein (DAF-FM; Molecular Probes) diacetate was also used to detect the proportion of cells producing NO production by flow cytometry. The specificity of NO determinations was established by using the iNOS-specific inhibitor L-N6-(1-iminoethyl) lysine (L-NIL; Cayman Chemical Co., Ann Arbor, MI, USA) dihydrochloride and the peroxynitrite scavenger manganese (III) tetrakis (4-benzoic acid) porphyrin (Calbiochem, San Diego, CA, USA). The percentage of NO-producing cells was determined in peripheral blood neutrophils, resident peritoneal macrophages, and cells derived from the skin at the site of infection (7 days p.i.). Cells from skin were obtained by mechanical dissociation using a Medimachine® apparatus (BD Biosciences) following the instructions of the manufacturer. In brief, cells (500,000 cells in 100 µL PBS) were incubated with 0.5 µM DAF-FM during 30 min at room temperature in dark. The cells were washed with PBS and incubated for 15 min to complete de-esterification of the intracellular diacetates. Fluorescence intensity and percentage of gated cells expressing NO were determined in the FL1 channel (515 nm). DAF-FM-labeled cells from uninfected animals were used to define the threshold for flow cytometry analysis. Total NO production was evaluated in culture supernatants of 2.5 x 106 peritoneal macrophages stimulated with antigen and 1000 nM estrogen (48 h in 250 µl final volume) using the Griess reaction (Cayman Chemical Co.). The specificity of iNOS-derived NO production in the supernatants was determined by the addition of 1 mM L-NIL. A TaqMan assay was used to identify iNOS induced by estrogen. Briefly, Universal Master Mix (Applied Biosystems, Foster City, CA, USA) was mixed with 150 ng reverse-transcribed RNA (SuperScript II, Invitrogen, Carlsbad, CA, USA), 700 nM each primer, and 100 nM TaqMan probe (forward 5'-tga gcc act gag ttc tcc taa gg-3'; reverse 5'-tcc tat ttc aac tcc aag atg ttc tg-3'; probe 5'-6FAM-cgt gga cac ttc ctt tgt ctg tgc tcc-TAMRA-3') in a final volume of 25 µl. The reference gene -actin was used as normalizer and detected in a validated multiplex assay using a VIC-labeled TaqMan probe. The fold increase in iNOS expression was calculated using nonstimulated macrophages as calibrator and the 
comparative threshold method.
In vitro assays using estrogen or progesterone
To evaluate the influence of estrogen or progesterone on NO production, resident peritoneal macrophages were obtained by peritoneal lavage from uninfected hamsters or from hamsters at 7 days p.i. with L. panamensis. For measurement of hormone-induced NO, we used macrophages but not neutrophils, as the latter could only be recovered in limited numbers from peripheral blood. Cells were cultured for 5 min at 37°C, 5% CO2, 95% air, with different concentrations of 17-β estradiol (10–1000 nM) or the progesterone metabolite, 5-pregnane-3-ol-20 (pregnane; 1–100 µM, Sigma Chemical Co., St. Louis, MO, USA). In some experiments, the estrogen receptor inhibitor tamoxifen (1 µM) was used to antagonize the action of 17-β-estradiol as described [24
]. The estrogen and tamoxifen were dissolved in absolute ethanol, and dilutions made in RPMI, 5% FCS, 2 mM glutamine, 100 units/mL penicillin, and 100 µg/mL streptomycin. After treatments, cells were cocultured with L. panamensis at 34°C, and parasite survival and proportion of cells producing NO were determined by flow cytometry. iNOS expression was quantitated by real-time PCR as described above.
Treatment of hamsters with estrogen and progesterone
Hamsters were treated intramuscularly with 2 µg 17-β-estradiol (Sigma Chemical Co.) or 3 µg of the active progesterone metabolite pregnane (Sigma Chemical Co.) every 48 h for 14 days. These treatment schedules were shown previously to increase plasma levels of these hormones [40
]. Protocols of infection and in vitro assays were followed as described for pregnant animals. Hamsters were infected intradermally in the snout with L. panamensis as described before on the 7th day of hormone treatment. At 7 days p.i. (14th day of hormone treatment), the animals were killed, and the parasite burden was determined by luminometry of homogenized tissue harvested from the infection site. A group of animals sham-treated with PBS was included as a control. Leishmanicidal activity and percentage of macrophages, neutrophils, and cells from lesions that were producing NO were determined by flow cytometry as above.
Statistical analysis
Multiple comparisons ANOVA with post-hoc Duncan’s test or Kruskal-Wallis test with Dunn’s procedure were used at a 95% confidence interval using SPSS standard program for Windows. Single comparisons were performed using GraphPad InStat software, Version 3.06, as described in each figure.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (12K): [in a new window] |
Figure 1. Pregnancy enhances the control of early cutaneous infection with L. panamensis. (A) Lesion size. Pregnant female (shaded bars) or nonpregnant female (open bars) hamsters were infected in the snout with 104 L. panamensis promastigotes, and the lesion size (mean±SD) was determined after 15 days p.i. (parturition) and 30 days p.i. (end of lactation). Individuals infected during pregnancy (n=9) had smaller lesions than controls (n=12) at both time-points (P=0.04; unpaired t-test, normality tested by Kolmogorov and Smirnov). (B) Parasite burden. The parasite burden in the snout from pregnant (n=7) and nonpregnant female hamsters (n=11), infected as described above, was determined by luminometry at 7 days p.i. Data are combined from two independent experiments and are presented as number of amastigotes per lesion (mean±SE; P=0.0098; unpaired t-test).
|
|
View this table: [in a new window] |
Table 1. Phenotype of Peripheral Blood Cells in Pregnant Hamsters Infected with L. panamensis (7 Days p.i.) Compared with Infected, Nonpregnant Controls
|
, and TNF- were reduced at the lesion site (IL-12, TNF-) or draining lymph nodes (IFN-) in pregnant compared with nonpregnant animals (P<0.05). The expression of the anti-inflammatory, macrophage-inactivating cytokine TGF-β was also consistently low in the lesion and draining lymph node in pregnant compared with nonpregnant animals (Table 2)
. Thus, there was a generalized pregnancy-associated reduction in the early expression of cytokine mRNA in response to L. panamensis infection. |
View this table: [in a new window] |
Table 2. Cytokine and NO mRNA Expression in the Infection Site and Cells from Draining Lymph Nodes of Pregnant Hamsters Infected with L. panamensis (7 Days p.i.) Compared with Nonpregnant, Infected Controlsa
|
 View larger version (9K): [in a new window] |
Figure 2. Pregnancy enhances the leishmanicidal capacity of macrophages and neutrophils exposed to L. panamensis. (A) Leishmanicidal capacity of macrophages. Resident peritoneal macrophages isolated from infected (7 days p.i.), pregnant and nonpregnant hamsters were infected in vitro with L. panamensis, and the parasite burden was determined after 72 h by luminometry. Shown is the number of amastigotes (mean±SE) obtained in three independent experiments using four to five hamsters per group per experiment (P=0.018; unpaired t-test). (B) Leishmanicidal capacity of neutrophils (N s). Sorted peripheral blood neutrophils were pooled and infected with CFSE-labeled promastigotes. The percentage of neutrophils harboring viable CFSE + L. panamensis was determined by flow cytometry at 20 min–1 h p.i. Shown is the mean ± SD percentage of neutrophils containing viable leishmania observed in one experiment that is representative of two independent experiments. Neutrophils from pregnant hamsters (nine pooled samples) had greater leishmanicidal capacity than neutrophils from the nonpregnant animals (seven pooled samples; P=0.018; unpaired t-test).
|
L. panamensis-triggered NO production by macrophages or neutrophils was determined in a fluorescence-based (DAF-FM) flow cytometry assay. The specificity of the NO detection was established by using the NO inhibitor L-NIL dihydrochloride (Cayman Chemical Co.). No NO production by L. panamensis promastigotes was detected using DAF-FM. Hamster peritoneal macrophages infected in vitro with L. panamensis promastigotes and cultured for 24 h with 10–1000 µM L-NIL showed a dose-dependent decrease in NO production (linear regression, P=0.01); NO production by these macrophages was abolished at 1000 µM L-NIL (data not shown).
There was a trend toward a higher proportion of resident peritoneal macrophages isolated from infected, pregnant hamsters to produce NO compared with macrophages from infected, nonpregnant animals (mean±SE of five experiments: pregnant, 44.8±2.8; control, 34.6±5.1; P=0.07). Additionally, the percentage of NO-producing neutrophils was 6.8-fold greater in L. panamensis-infected, pregnant animals compared with infected, nonpregnant animals (mean±SE of two different experiments: pregnant, 72.8±8.3; control, 10.7±1.5; P=0.0001). Pregnancy also enhanced the percentage of neutrophils producing NO in uninfected animals (P=0.006). Collectively, these data indicate that neutrophils from pregnant or infected animals show increased NO production over controls and that pregnancy amplifies NO production induced by infection.
Estrogen and pregnane enhance macrophage NO
The efficient parasite control observed in pregnant animals during the early stages of infection, which was associated with high iNOS expression, suggested that reproductive hormones could play a role in Leishmania containment. Therefore, we determined the influence of the pregnancy-associated hormone estrogen (17-β estradiol) and pregnane (a progesterone metabolite) on NO production by macrophages upon in vitro infection with L. panamensis. Resident peritoneal macrophages isolated from hamsters infected with L. panamensis (7 days p.i.) and restimulated with Leishmania antigens showed a dose-dependent increase in production of NO after ex vivo exposure to increasing doses of 17-β estradiol (P=0.01; Fig. 3A
). Addition of 1 µM tamoxifen (which competitively blocks the estrogen receptor) to the culture medium before supplementation with estrogen impaired the hormone-mediated increase in NO production (P=0.056; Fig. 3A
). Peritoneal macrophages from L. panamensis-infected hamsters exposed to this hormone for 6 h showed a 35-fold increase in iNOS expression over nonstimulated macrophages (P=0.01; Fig. 3B
). Tamoxifen inhibited iNOS expression induced by estrogen treatment, suggesting iNOS expression by a classic, receptor-mediated pathway (P=0.01; Fig. 3B
).
 View larger version (26K): [in a new window] |
Figure 3. Estrogen and progesterone modulate NO production and iNOS expression in resident peritoneal macrophages from hamsters infected with L. panamensis. Resident peritoneal macrophages were isolated from hamsters infected with L. panamensis and exposed in vitro for 5 min to estrogen (17-β estradiol; 0–2000 nM), with or without tamoxifen (1 µM) or pregnane (5-pregnant-3-ol-20; 1–100 µM). The estrogen receptor inhibitor tamoxifen was included 30 min before estrogen treatment and L. panamensis infection. (A) Effect of estrogen (E2) on the proportion of NO-producing cells determined by flow cytometry. Estrogen was added to the macrophage cultures that were stimulated with L. panamensis promastigotes. The data shown (mean±SD) are from a single experiment representative of two independent experiments with four hamsters each; the threshold in the experiment was established using uninfected hamsters (Uninf). The frequency of NO-producing cells increased with estrogen treatment (P=0.01) and decreased by blocking the estrogen receptor with tamoxifen (P=0.056; Kruskal-Wallis test; nonparametric ANOVA). (B) Estrogen-induced iNOS expression (real-time PCR). iNOS mRNA expression in peritoneal macrophages from hamsters infected with L. panamensis 6 h after treatment with 1000 nM estrogen or estrogen + 1 µM tamoxifen (TX). Data of four replicates from pooled samples of five animals are shown as fold increase with respect to cells cultured with medium alone. Estrogen treatment induced iNOS mRNA expression (P=0.01) and was blocked with tamoxifen (P=0.01; unpaired t-test). (C) Effect of estrogen on iNOS-derived NO production. Resident peritoneal macrophages (2.5x106) isolated from L. panamensis-infected female hamsters (n=5) were stimulated for 48 h with L. panamensis (L.p.) antigen (100,000 frozen-thawed L. panamensis promastigotes) in the presence or absence of estrogen (1000 nM), with or without iNOS inhibitor L-NIL (1 mM). NO (mean±SD) released in the supernatants after 48 h of in vitro culture, determined by Griess reaction, showed that estrogen treatment induced NO production (P=0.05), which was blocked with the addition of tamoxifen (P=0.02; unpaired t-test). (D) Effect of progesterone on NO production. Resident peritoneal macrophages from hamsters infected with L. panamensis were exposed in vitro to 1–100 µM pregnane, and following 1 h of in vitro coculture with L. panamensis, the NO-producing cells were identified by reaction with DAF-FM diacetate and flow cytometry. The frequency of NO-producing cells was increased with pregnane treatment (P=0.001; Mann-Whitney test).
|
Naïve macrophages also showed increased iNOS expression and NO production upon estrogen treatment. The quantification of iNOS using a TaqMan assay showed that peritoneal macrophages from uninfected hamsters increased iNOS expression fivefold after 16 h exposure to 1000 nM estrogen (nontreated=onefold±0.45; estrogen=5.3-fold±1.86; P=0.018; unpaired t-test). On the other hand, flow cytometry evaluations showed that the percentage of macrophages producing NO increased sixfold after estrogen treatment (nontreated=3.4%±1.1; estrogen-treated=20.7%±12).
Treatment of peritoneal macrophages with concentrations of pregnane ranging from 1 to 100 µM significantly increased the percent of NO-producing cells (sixfold) compared with cells from infected or uninfected animals sham-treated with medium alone (P=0.001; Fig. 3D ). An increase in the percentage of NO-positive cells was not observed at lower doses of pregnane (1–500 nM).
In vivo administration of estrogen and pregnane modulates host NO production and parasite control
To determine the in vivo relevance of the previous ex vivo and in vitro observations, different groups of hamsters were treated with the aforementioned hormones. Estrogen or pregnane was administered to hamsters every 48 h for 14 days, and the animals were infected on Day 7 of hormone treatment with LUC-transfected L. panamensis to determine NO-producing resident peritoneal macrophages and peripheral blood neutrophils; parasite survival after in vitro infection of resident peritoneal macrophages and peripheral blood neutrophils; and parasite burden and NO producer cells at the infection site by luminometry or flow cytometry, respectively.
Seven days after infection of hamsters that had been treated with estrogen or pregnane, we found that NO production by resident peritoneal macrophages was greater in the hormone-treated compared with untreated animals. Resident macrophages collected from infected hamsters treated with estrogen produced eightfold more NO than those treated with PBS (P=0.001; Table 3 ). Also, there was a tendency of macrophages from pregnane-treated hamsters to secrete higher levels of NO (threefold; not significant) than controls. Macrophages from estrogen-treated hamsters demonstrated significantly greater leishmanicidal capacity than controls; this enhancement in Leishmania killing ability could not be detected in macrophages derived from animals treated with pregnane (Table 3) . A significant inverse correlation was found in the percentage of estrogen-induced, NO-producing cells and parasite survival in peritoneal macrophages (Spearman’s correlation; P=0.041).
|
View this table: [in a new window] |
Table 3. Percentage of Cells Producing NO and Percentage of Viable Parasites in Peritoneal Macrophages or Peripheral Blood Neutrophils from Hamsters Treated with Estrogen (17-β-Estradiol) or Pregnane (5-Pregnant-3-ol-20)a
|
At 7 days p.i., prior to the detection of a measurable cutaneous lesion, the parasite burden in the inoculation site was significantly lower in estrogen-treated hamsters than in controls (Fig. 4A ; P=0.036). A similar but not significant trend was observed in animals treated with pregnane (Fig. 4A) . Consistent with the in vitro observations, flow cytometry analysis showed that NO levels in the infected skin of hamsters treated with estrogen were higher than in the corresponding controls (P=0.001; Fig. 4B ). Also, a negative correlation between the proportion of NO-producer cells and parasite burden was found in the dermis of hamsters treated with estrogen or pregnane (Spearman’s correlation; P=0.05).
 View larger version (12K): [in a new window] |
Figure 4. Hormone treatment modulates parasite burden and NO production at the site of infection. Female hamsters (n=6 per group) were treated with 2 µg 17-β-estradiol (Estrogen) in PBS, 3 µg 5-pregnane-3-ol-20 (Pregnane) in PBS, or PBS alone every 48 h for 14 days. The animals were infected with L. panamensis in the snout on Day 7 of treatment. The number of animals in each group is shown on the horizontal axis. (A) Parasite burden at the site of inoculation. The parasite burden in the snout was determined 7 days p.i. by luminometry. The data (amastigote numbers) are shown as a box plot (median plus first–third quartile) with whiskers showing the smallest and largest values. Significant differences were identified between PBS and estrogen treatment (P=0.036) but not between PBS and pregnane (Kruskal-Wallis test and Dunn procedure). (B) NO-producing cells at the site of inoculation. NO production by cells in the infected snout tissue was evaluated by flow cytometry of DAF-FM-labeled cells. Significant differences were identified between PBS and pregnane treatment (P=0.001) and PBS and estrogen (P=0.026; Kruskal-Wallis test; nonparametric ANOVA).
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The activation of innate immunity during pregnancy, as we observed in this study in hamsters, has been described in humans [12 , 14 , 15 ]. Furthermore, female hamsters that were infected with L. infantum 24 h after parturition were also found to have less evidence of visceral leishmaniasis and a reduced visceral parasite burden [29 ]. The influence of elevated prolactin was considered to be the cause of the increased resistance; however, our data suggest that other residual, pregnancy-related hormonal changes could have also contributed to the enhanced resistance in this early challenge model. Our findings are distinct from what was reported for pregnant C57BL/6 mice infected with L. major, in which the latter were more susceptible than nonpregnant females [7 ]. This apparent discrepancy could be in part a result of differences in the Leishmania strain used and/or host species between the studies. The study of Krishnan et al. [7 ] focused principally on the development of the adaptive, T cell-mediated immune response in a chronic infection model rather than in the early events of innate immunity as studied here. Nevertheless, our studies do not exclude the possibility that females infected during pregnancy could ultimately develop a Th2 response that worsens chronic infections.
Parasite internalization by resident peritoneal macrophages was found to be unaltered during pregnancy, but macrophages from pregnant, infected hamsters had enhanced capacity to restrict Leishmania replication at 72 h p.i. The increased leismanicidal activity of hamster macrophages and neutrophils during pregnancy was accompanied by a general decrease in expression of cytokines (IFN-, IL-12, and TNF-) typically associated with classical macrophage microbicidal activity. This cytokine down-regulation concurs with the classic paradigm of suppression of T cell responses (without a Th2 cytokine bias) during human pregnancy [1
2
3
, 13
14
15
] but differs from a previous report of an enhanced Th2 (IL-4) response in pregnant mice infected with L. major [7
]. The relatively greater leishmanicidal activity of macrophages from pregnant females, despite the reduced expression of classical macrophage activators, may be related to the reduced expression of TGF-β and to a lesser extent, IL-10, cytokines known to suppress macrophage effector function and promote Leishmania infection [41
]. Alternatively, macrophages of pregnant females might have enhanced responsiveness to IFN- through up-regulation of expression of IFN- receptors, as has been described in myometrial macrophages of pregnant mice [42
].
Our results indicate that during the first days of infection, not only macrophages but also neutrophils from pregnant females generated more NO and killed Leishmania more effectively than those from nonpregnant animals. The leishmanicidal capacity of neutrophils has also been demonstrated in human cells infected ex vivo with L. major [43 ] in resistant mouse strains [44 ] and in hamsters infected with L. panamensis (unpublished), suggesting that they constitute an early, primary barrier to Leishmania infection. However, other reports have suggested that neutrophils could also act as a "vector" for intracellular Leishmania via a parasite-induced delay in apoptosis [16 , 43 , 45 ] or through the phagocytosis of apoptotic, amastigote-laden neutrophils by human macrophages. Our data indicate that during pregnancy, the neutrophils were highly activated and had enhanced parasite killing, so were probably not acting as permissive Leishmania host cells.
Hamsters, like humans, express iNOS at levels substantially lower than mice [36
, 46
]. We demonstrated previously that in L. donovani-infected hamsters and in IFN-/LPS-stimulated hamster macrophages, iNOS mRNA expression and NO production could not be detected by Northern blot and Greiss reaction, respectively [36
, 46
]. The data presented here are not incongruent with these previous findings for several reasons. First, NO-producing macrophages and neutrophils were detected in this hamster model of cutaneous L. panamensis infection using a much more sensitive, fluorescence-based flow cytometry assay, and similarly low levels of NO production and iNOS were measured by Griess reaction and real-time PCR. Second, parasite-specific differences in the induction of iNOS expression are evident; the relatively controlled, cutaneous L. panamensis infection is accompanied by iNOS expression, whereas the progressive, fatal L. donovani infection actually is coincident with suppressed iNOS expression (our unpublished observations). That the generation of reactive nitrogen species is stimulus-specific is further supported by the demonstration of iNOS expression and NO production in hamsters immunized against L. donovani [47
] and in hamsters infected with Entamoeba histolytica [48
].
This work adds support to a growing number of reports, in experimental animals and humans, that estrogen is important for iNOS expression and NO production by different cell types, including monocytes and macrophages [24 , 26 , 49 ]. The ex vivo and in vitro studies presented here identify a direct effect of estrogen on iNOS in macrophages, which was mediated through the estrogen receptor (our data and refs. [25 , 26 ]).
We confirmed that the estrogen-enhanced NO production and iNOS expression found in cultured phagocytes were also evident in vivo following administration of estrogen to nonpregnant hamsters. Most notably, the estrogen-enhanced phagocyte leishmanicidal activity was associated with improved control of the early phase of cutaneous infection. In addition to the direct effect on macrophages, at the site of infection in vivo, the estrogen effect might be mediated by T cell-dependent signals [50 , 51 ].
Physiological studies have also associated the increased progesterone levels in pregnancy with the biosynthesis of NO [27 , 28 , 40 ]. The administration of pregnane to hamsters increased the number of NO-producing cells at the site of cutaneous infection and by ex vivo-cultured macrophages and neutrophils. However, there was no pregnane-mediated increase in parasite killing in the lesion or by ex vivo-cultured macrophages or neutrophils. This suggests that a hormone-mediated increase in NO production is not sufficient in and of itself to affect parasite killing and that there must be other estrogen-induced effector mechanisms that act additively or synergistically with the generation of NO.
This experimental study demonstrates that the innate immune response, characterized by phagocyte NO production and control of L. panamensis infection, is enhanced during pregnancy. Although ex vivo studies demonstrated that macrophages and neutrophils could be a source of the increased NO production, and both are found at the site of infection [32 ], further studies are needed to define more clearly the relative contribution of these cells to the pregnancy-enhanced NO production and parasite killing in vivo. To our knowledge, this is the first study in which a clear association between estrogen-induced iNOS expression/NO production and Leishmania killing has been found. Significantly, this pregnancy- and hormone-enhanced innate host defense was demonstrated in an animal model whose macrophages show relatively low iNOS expression and NO production in response to inflammatory stimuli, a feature similar to humans but distinct from the murine rodents. These results underscore the importance of hormones in the regulation of the immune response and support the concept that physiological changes that are associated with varying levels of circulating hormones can impact the balance between host and pathogens such as Leishmania. Additional studies will need to explore the consequences of pregnancy on the evolution of the adaptive immune response and chronic disease. It will also be important to understand how infection before or during pregnancy in humans could affect the natural history of disease in the mother and the child. The knowledge of mechanisms involved in the estrogen-induced production of NO is of potential interest for the control of intracellular pathogens.
Received February 26, 2007; revised January 11, 2008; accepted January 22, 2008.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
response and increased production of T helper 2 cytokines J. Immunol. 156,644-652[Abstract] production by female but not male DBA/2 mice Immunol. Cell Biol. 76,159-166[CrossRef][Medline]-pregnan-3 -ol-20-one on nitric oxide biosynthesis and plasma volume in rats Am. J. Physiol. Regul. Integr. Comp. Physiol. 280,R1902-R1905 Endocrinology 147,662-671-inducing transcription factor, T-bet is upregulated by estrogen in murine splenocytes: role of IL-27 but not IL-12 Mol. Immunol. 44,1808-1814[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  R. Dai, R. A. Phillips, Y. Zhang, D. Khan, O. Crasta, and S. A. Ahmed Suppression of LPS-induced Interferon-{gamma} and nitric oxide in splenic lymphocytes by select estrogen-regulated microRNAs: a novel mechanism of immune modulation Blood, December 1, 2008; 112(12): 4591 - 4597. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
