|
|
||||||||
| ABSTRACT |
|
|
|---|
and interleukin-12 (IL-12). When restimulated, these PIV cells also occasionally produced IL-5. Both CD4 and CD8 cells and both HLA class I and II cell activation pathways appeared to play a role in the PIV system, and cell activation was dependent upon the presence of antigen-presenting cells. Moreover, PIV cells generated with L. major showed considerable cross-reactivity with other species of Leishmania. Finally, the PIV cells augmented intracellular killing of L. major when they were co-cultured with macrophages infected with the parasite. | INTRODUCTION |
|
|
|---|
Within the mammalian host, Leishmania resides as an amastigote in phagocytic cells such as macrophages, dendritic cells, and neutrophils. Cell-mediated immunity is essential for resistance to the parasite. Cytokines produced by T cells, natural killer (NK) cells, and antigen-presenting cells (APC) play an integral role in this process.24
Infection of experimental mice with L. major (which causes cutaneous leishmaniasis) is perhaps the best-studied model of a chronic infectious disease. Using this model, it is now well documented that Th1 cells mediate resistance via a type 1 response to the parasite in most strains of mice. The Th1 cells secrete activators of cell-mediated immunity such as interferon (IFN)-
. Interferon-
induces production of nitric oxide (NO) in phagocytic cells that harbor L. major (principally macrophages), which leads to destruction of the parasite.
In contrast, BALB/c mice develop a type 2 response following infection with L. major and the disease progresses inexorably, leading to death of the animals. The Th2 cells produce B cell activation/differentiation factors such as interleukin-4 (IL-4) and IL-5. It has been shown that within minutes to hours after infection with L. major, BALB/c mice produce IL-4 mRNA in response to a single parasite antigen, Leishmania homolog of receptors for activated C kinase (LACK). Interleukin-4 may down-regulate expression of the ß2 subunit of IL-12 receptors on potentially protective Th1 T cells.5 As a result, the cells become unresponsive to IL-12 and production of IFN-
and NO is inhibited. Thus, L. major parasites within macrophages are not killed.24
The immune response of humans to infection with Leishmania is not as well characterized as the response of mice. Much of the research with human leishmaniasis has involved the use of infected or recovered patients. These types of studies have several limitations. For instance, in patients infected with Leishmania, it is often difficult to determine when an individual was initially infected with Leishmania and with what species. Thus, it is not possible to study the early interactions that occur between the host and parasite. There are also confounding factors when investigating patients, particularly from poverty-stricken regions. These issues can include malnutrition, secondary infections, and lack of access to medical care, all of which can influence the outcome of the disease.6
In an effort to gain control over experimental conditions and the timing, species, and dose of infective parasites, some investigators have studied the response of peripheral blood mononuclear cells (PBMC) from Leishmania-naive donors to infection with the parasite in vitro. A similar approach (using spleen or lymph node cells) in a murine system was shown to mimic closely the immune response that occurs to infection with L. major in mice.7,8 These human PBMC studies have suggested a role for T cells9,10 and NK cells1113 in the response of the PBMC to Leishmania in vitro. Indeed, in a report by Kurtzhals and others,9 the investigators showed that Leishmania-naive PBMC produced IFN-
(a type 1 cytokine) and IL-4 (a type 2 cytokine) when stimulated with boiled extracts of L. donovani or L. major. In the report by Russo and others,10 the frozen-and-thawed lysate of L. amazonensis induced a type 1 response by PBMC, and in the presence of IL-12 it primed CD8 cells that lysed autologous, Leishmania-infected macrophages. However, it has been reported that the response of PBMC to dead versus live Leishmania preparations differs significantly.12 Therefore, studies that use dead preparations of Leishmania may not mimic the response of humans to naturally acquired disease. More recently, Bourreau and others14 studied the response of Leishmania-naive PBMC to stimulation with live L. guyanensis. Although the emphasis of the report was on the response to the LACK antigen, they demonstrated the importance of T cells in the PBMC response to living Leishmania. Indeed, in their hands, living L. guyanensis was highly stimulatory, L. guyanensis irradiated with ultraviolet light was moderately stimulatory and soluble antigens of L. guyanensis were poorly, if at all, stimulatory.
Because of this, we developed a primary in vitro (PIV) system in which PBMC from Leishmania-naive donors were stimulated with living L. major promastigotes. This is the form of the parasite that is injected by sand flies (as opposed to killed forms of Leishmania used by others) when humans are naturally infected with the parasite. This approach allowed us to test the hypothesis of whether the response of humans resembled the much-studied type 1/type 2 response of mice to infection with L. major. In addition, this approach allowed us to monitor the early human PBMC response to the parasite as well as to restimulate responding PIV cells with fresh autologous APC to examine secondary responses to Leishmania. Our goal was to broadly characterize this human response. That is, rather than focusing strictly on cytokine responses, we also measured the dependence of the PIV response on APCs, as well as the 1) surface phenotype, 2) HLA restriction, 3) degree of proliferation, 4) specificity, and 5) ability of PIV cells to activate L. major-infected macrophages to destroy the parasite in vitro. All of these characteristics and functions would presumably be important for humans to overcome initial infection with Leishmania and to develop resistance to reinfection.
| MATERIALS AND METHODS |
|
|
|---|
In vitro system.
Human blood was obtained from healthy individuals at the Hartshorn Health Center at Colorado State University. All of the donors reported that they had never traveled to areas where leishmaniasis exists, nor had they ever had the disease; thus, we considered this population of donors to be immunologically Leishmania nave. Blood was drawn after informed consent was obtained from each donor. The Human Research Committee at Colorado State University reviewed and approved all of the procedures. One hundred milliliters of blood was collected during each blood draw. The PBMC were isolated from heparinized venous blood by passage over a Ficoll-Hypaque gradient (Pharmacia, Uppsala, Sweden).18 the PBMC were washed three times and resus-pended at a concentration of 2.5 x 106 cells/mL in complete medium consisting of RPMI 1640 medium supplemented with 2 mM L-glutamine, penicillin (100 units/mL), gentamicin (100 µg/mL), and 10% heat-inactivated human AB serum (Pel-Freeze, Brown Deer, WI). The cells were plated in 24-well tissue culture plates (Costar, Corning, NY) in a volume of 1 mL/well. Leishmania major promastigotes (2.5 x 105 parasites/mL) were added to stimulate the cells (Figure 1A
).
|
Proliferation assay. The PIV cells were harvested and rested in medium alone at a concentration of 2 x 106/mL for two days so that the cells exited the growth cycle and became quiescent. Triplicate wells were prepared on 96-well microtiter plates (Costar) with 5 x 104 rested PIV cells, 106 irradiated autologous APCs, and 106 Leishmania/mL or 106 L. major equivalents/mL for the lysate preparation. Proliferation was measured by the addition of 1 µCi of 3H-thymidine 48 hours after plating followed by scintillation counting 24 hours later. All experiments were repeated two times.
Reagents. The following reagents were used in cell cultures: monoclonal antibody to human HLA-DR, DP, DQ (clone TÜ39) and monoclonal antibody to human HLA-A, B, C (clone G46-2.6) both at a concentration of 10 µg/mL. The following reagents were used for flow cytometry: fluorescein isithiocyanate (FITC)-labeled anti-human CD25 (clone M-A251), phycoerythrin (PE)-labeled anti-human CD4 (clone RPA-T4), cychrome (CY)-labeled anti-human CD8 (clone RPA-T8), (CY)-labeled anti-human CD19 (clone HIB19), FITC-labeled anti-human HLA-DR, DP, DQ (clone TÜ39), and PE-labeled anti-human CD56 (clone B159). All antibodies were obtained from Pharmingen Beckton Dickenson (San Diego, CA) and appropriate irrelevant isotype-matched antibodies from the same supplier were used as controls. Phytohemagglutinin was obtained from Sigma.
Cytokine assays.
Concentrations of IFN-
, IL-12, and IL-5 in culture supernatants were determined by an enzyme-linked immunosorbent assay (ELISA) using commercial ELISA kits (Genzyme, Cambridge, MA). The IL-12 kit was specific for both IL-12p40 and p70 subunits. The limits of detection for these assays were 10 pg/mL for IFN-
and IL-12 and 2 pg/mL for IL-5. Supernatants were collected at 24, 48, and 72 hours after initial stimulation or at 48 hours after restimulation. The supernatants were stored at 80°C until they were assessed for their content of cytokines. It should be noted that we also attempted to measure IL-4 in the supernatants, but we were unable to detect the cytokine in any experimental setting.
Flow cytometry. The PBMC exposed to L. major were characterized by flow cytometry to determine if there were any changes in the cell surface phenotype of the cells. The PIV cells (1 x 106/sample) were prepared for analysis as previously described.19 Briefly, cells were resuspended in phosphate-buffered saline, 1% bovine serum albumin, 0.05% sodium azide (PAB) and blocked with mouse IgG (20 µg/mL) and 10% fetal bovine serum for 30 minutes on ice. Cells were then incubated with labeled antibodies or corresponding isotype controls for an additional 30 minutes. Cells were fixed with 1% paraformaldehyde in PAB and analyzed on an EPICS XL flow cytometer (Coulter Corp., Miami, FL). As determined by forward light scatter, a gate was drawn around live cells (excluding debris and doublets) during data collection. Fluorescence data were obtained on 10,000 cells. Analyses of flow cytometry results were performed using FlowJo software (TreeStar, Palo Alto, CA).
Macrophage leishmanicidal assay. To determine if macrophages could kill live L. major in our assay system, macrophage killing assays15,20 were performed. Tissue culture-treated glass coverslips were placed at the bottom of 24-well plates and PBMC were plated at a concentration of 5 x 106/mL in each well. Monocytes were allowed to adhere to the coverslips overnight and then non-adherent cells were washed away. The remaining cells were incubated for eight days to allow the monocytes to mature into macrophages.
Leishmania major was then added to the cultures at a ratio of 10 parasites to one macrophage. Unphagocytized parasites were washed away after eight hours. When used, autologous PIV cells were added to the macrophage cultures at a ratio of 10 PIV cells to one macrophage. Some cultures were further treated with conditioned medium (CM) or neutralizing antibodies. Conditioned medium was prepared by collecting the supernatant from eight-day PIV cultures (Figure 1A
). Anti-human IFN-
antibody (clone 25718.111; R&D Systems, Minneapolis, MN) or the appropriate isotype control was added at a concentration of 10 µg/mL to designated cultures. This was the concentration of antibody recommended by the manufacturer. Moreover, this concentration of anti-IFN-
was capable of blocking IFN-
activity in studies published elsewhere21 that analyzed the assay system reported here.
The coverslips were removed 18 hours later and stained with Diff-Quick (Scientific Products, McGraw, IL). The macrophages on the coverslips were then examined by light microscopy to enumerate intracellular parasites. The results are expressed as number of intracellular parasites per 100 macrophages counted.
Statistical analyses. Statistical analyses were performed using Sigma Stat (SPSS, Chicago, IL) and InStat (Graph Pad Software, San Diego, CA) software. Since the data were not distributed normally, non-parametric analyses were necessary. For early time point studies, Friedmans nonparametric repeated measures test was used to analyze the data, followed by Dunns Multiple Comparison Test when P < 0.05. For restimulation studies, data were compared with Wilcoxon matched pairs tests. For proliferation studies and macrophage leishmanicidal assays, Kruskal-Wallis one-way analysis of variance was performed. When P < 0.05, Dunns Multiple Comparisons Test was used to determine the significance of the difference between samples. Overall, results were considered significantly different when the P value was < 0.05. All experiments were repeated at least two times.
| RESULTS |
|
|
|---|
and IL-12 and the type 2 cytokines IL-4 and IL-5 was determined by ELISA after PBMC had been exposed to L. major for 24, 48, or 72 hours. No IL-4 or IL-5 production was detected at any time point. In contrast, both IL-12 and IFN-
were detected. Although stimulation with L. major tended to increase IL-12 secretion, (Figure 2A
early (24 hours) and the difference in production when the cells were exposed to L. major compared with the unstimulated controls was significant (P < 0.05) at 48 and 72 hours (Figure 2B
|
and IL-12 as representatives of type 1 cytokines and IL-4 and IL-5 as type 2 cytokines. Upon restimulation, all of the donors were able to produce IL-5 (Figure 3A
|
Most striking, however, when PIV cells were restimulated with the parasite, all donors secreted large amounts of IFN-
(Figure 3C
; P < 0.001). The results shown in Figures 2B
and 3C
suggest that of the cytokines we examined, IFN-
is the principal cytokine produced by human PBMC both after primary stimulation (nearly 5,000 pg of the cytokine per milliliter of culture, Figure 2
) and restimulation (as much as 50,000 pg/mL, Figure 3
) with L. major. This was true even with PBMC from donors that also produced the highest levels of IL-5; these two donors produced > 15,000 pg IFN/mL.
Culture requirements for restimulation of human PIV cells and specificity of the cells.
To characterize our system, we first determined the culture requirements for restimulating PIV cells. As shown in Figure 4A
, for maximal proliferation, PIV cells required a source of APC, as well as the parasite (far right column in Figure 4
). This group responded to a greater extent than any other group tested (an average of 22,600 counts per minute [cpm] versus a range of 3143,510 cpm in the control groups; P < 0.05).
|
Phenotypic characterization of human PIV cells.
Next we phenotypically characterized PIV cells to determine what cells were responding to the parasite. We measured the expression of the following cell surface markers: CD4, CD8, CD56 (an NK cell marker), CD19 (a B cell marker), CD25 (an IL-2 receptor that indicates activation of T cells), and HLA class II after the PIV cells had been exposed to the parasite for eight days. Few differences were found when we compared cells that had been in culture for eight days and either exposed or not exposed to the parasite (Table 1
), except for increased expression of HLA class II and CD25 by the cells cultured with the parasite. The median percentage of control cells expressing HLA class II was 12.9% (range = 6.319.1%) and when the cells were exposed to the parasite, the percent of cells increased to 20% (range = 6.629.5%). Similarly, the percent of CD25-expressing cells was 4.0% (range = 1.47.2%) in the L. major-negative population and 10.3% (range = 2.217.4%) in the parasite-positive population. The increase in expression of both CD25 and HLA class II was significant compared with control cultures not exposed to L. major.
|
Role of HLA class I and II in the human PIV response.
It has been reported in a murine PIV system that CD4 cells respond to L. major and that this response is major histocompatibility complex (MHC) class II-restricted.7 Therefore, we examined the role of HLA class I and II in cell activation/proliferation and cytokine (IFN-
) production in our human PIV response.
To determine the role of HLA class I and II in cell activation/proliferation, PBMC were stimulated with L. major for eight days and the PIV cells were harvested. The PIV cells were then restimulated in the presence or absence of blocking antibodies against either HLA class I or II, syngeneic APC, and L. major. The difference in proliferation between PIV cells cultured with APC and L. major and PIV cells blocked with anti-HLA class II antibody was not statistically significant, although there was reduced proliferation (Figure 5
). However, PIV cells exposed to anti-HLA class I antibody proliferated significantly less (P < 0.01) than PIV cells cultured with APC and L. major (Figure 5
).
|
production, L. major-naive PBMC were treated with either blocking anti HLA class I or II antibody and the cells were stimulated with L. major for seven days. Results showed that compared with control cultures in which PBMC were exposed to the parasite only, there was a significant decrease (P < 0.01 or P < 0.001) in the production of IFN-
whether the cells had been exposed to either anti-class I or II antibodies and L. major (Figure 6
|
production, we performed control experiments to ensure that the effects of the antibodies were not due to non-specific toxic effects on target cells. The PIV cells were stimulated in a non-HLA-dependent manner with phytohemagglutinin (2 µg/mL) and the anti-HLA antibodies were added to the cultures. Anti-HLA-class II antibodies had no effect on this system; however, anti-class I antibodies inhibited IFN-
production by 15.9%. Data presented that used anti-HLA class I antibody have been adjusted for this non-specific inhibition.
Influence of PIV cells on the ability of Macrophages to kill L. major.
Finally, it has been shown that macrophages are very effective at engulfing Leishmania and that activation of the cells via IFN-
leads to efficient killing of the parasite.15,22,23 Since there are high levels of IFN-
in our PIV cultures (Figure 2B
), we determined whether killing of L. major was occurring in infected macrophages. We first tested whether supernatants from eight-day PIV cultures (CM) could induce macrophages to kill L. major. The CM was collected from PIV cultures in which there was known to be high levels of IFN-
present (> 15,000 pg/mL). The CM reduced the number of intracellular L. major within macrophages, but this reduction was not significant (Figure 7
). However, adding autologous PIV cells to the cultures (in the presence or absence of CM) led to a significant reduction of intracellular parasites (P < 0.05). Adding a neutralizing anti-IFN-
anti- body to the cultures did not inhibit the killing of parasites (Figure 7
), suggesting that the soluble IFN-
present in the supernatants of the cultures was not solely responsible for the activation of infected macrophages.
|
. Interferon-
is the principal cytokine produced by PIV cells upon initial stimulation and upon restimulation. Cell activation is APC and parasite dependent (but not parasite specific). Both HLA class I and II presentation pathways appear to be involved in the PIV response, and PIV cells can induce marked destruction of intracellular L. major in macrophages. However, this ability of PIV cells to induce parasite destruction appears not to be due solely to secretion of IFN-
by the cells. | DISCUSSION |
|
|
|---|
We hypothesized that our PIV system would result in L. major-specific, CD4 and HLA-class II-restricted T cells that would produce IFN-
. Our results indicate that human PBMC reacted to the parasite differently in several ways compared with results reported in a murine PIV system.7,8 These differences include cross-reactivity with other Leishmania species (Figure 4B
) and the apparent ability of not only CD4 T cells, but also CD8 T cells, to recognize parasite antigens (Figure 5
and 6
). In addition, while different inbred strains of mice can produce either a type 1 or 2 response in a PIV system, human PBMC respond with predominantly a type 1 response (Figures 2
and 3
). Finally, both HLA class I and class II antigen presentation appear to play a role in the response to L. major (Figures 5
and 6
), whereas in the murine PIV system, the response was MHC class II-restricted.
First, we characterized the response of Leishmania-naive PBMC to the parasite early after exposure (within the first three days). It has been demonstrated in the mouse model that early interactions between the parasite, host cells and their resultant cytokine production leads to a Th1 or Th2 commitment and that once committed, this response is difficult to reverse.24,25 We found that we could detect only type 1 cytokines (e.g., IFN-
) in the first three days after exposure to the parasite (Figure 2B
), since neither IL-4 nor IL-5 was detected. Since IFN-
activates macrophages to kill Leishma-nia,26 our PIV results would predict that most individuals would resolve an infection with L. major without the need for therapeutic intervention, which indeed is the case.27,28 Although our data suggest that both CD4 and CD8 T cells played a role in the production of IFN-
in our human PIV system (Figure 6
), we cannot rule out a role for NK cells, especially since these cells are an early source of IFN-
and since others have reported that NK cells are important in the early production of this cytokine in humans.13 We also could detect IL-12 in our PIV system (Figure 2A
, although significant amounts of the cytokine were produced only when PIV cells were restimulated, Figure 3B
), which suggests that IL-12 promotes a type 1 response to L. major in humans as it does in mice.
When we restimulated PIV cells, the cells of all donors produced IFN-
, IL-12, and IL-5 (but still no IL-4, Figure 3
). When we measured IL-5 levels, two donors produced much higher levels of IL-5 compared with the other donors. However, these donors also produced high IFN-
levels and moderate concentrations of IL-12. We are classifying these two donors (of the 15 studied) as Th1/Th2 mixed responders, donors whose PIV cells produce both type 1 (Th1) and type 2 (Th2) cytokines. The remainder of the donors would then be classified as Th1 responders. A similar phenomenon was reported by Russo and others29 when naive human T cells were stimulated with Leishmania-infected macrophages. In human leishmaniasis, the presence of type 1 cytokines correlates with control of the disease and healing,3032 while the presence of type 2 cytokines is generally associated with active or chronic infections.33,34 Therefore, we could speculate that donors that were Th1 responders would resist infection with Leishmania, while the Th1/Th2 mixed responders might develop a more chronic disease.
We found that PIV cells induced with L. major had significant cross-reactivity with other leishmanial species and responded significantly less only when restimulated with L. amazonensis (Figure 4B
). This finding is consistent with limited studies using human subjects, indicating that cross-protection (and thus presumably cross-reactivity) can occur between different species of Leishmania.35,36 Individuals "vaccinated" with parasites from cutaneous lesions were protected in some cases from diffuse cutaneous or visceral leishmaniasis. Other studies have shown that various parasite antigens, such as parasite surface antigen 2 or whole parasites are capable of cross-species protection in mouse models.28,3739 There is increasing evidence in murine models of leishmaniasis that there are marked differences in L. major and L. amazonensis infections and immune responses to these parasites,4042 and these results are compatible with the lack of cross-reactivity reported here when human L. major-derived PIV cells were restimulated with L. amazonensis (Figure 4B
). Thus again, the results in our PIV system are consistent with findings with humans and with experimental animals infected with Leishmania.
The finding that PIV cells responded poorly when restimulated with L. major lysate (Figure 4B
) is identical with results obtained in a murine PIV system7 and in a human priming system.14 The reasons for these observations are unknown. However, others have reported that living L. major has adjuvant qualities since it can act as such when co-injected with peptides.43
Of the cell surface markers surveyed, only HLA class II and CD25 expression was increased on PIV cells after stimulation with L. major (Table 1
). Since our PIV cultures contained high levels of IFN-
and this cytokine has been shown to cause increased expression of HLA class II,44 it is likely that IFN-
was involved in the increased expression of HLA class II. It is interesting that CD25 expression was also increased on PIV cells. CD25 expression is increased in lesions of patients with cutaneous leishmaniasis45,46 and CD25 plays an important role in mice infected with L. major.47 Thus, again, our PIV system seems to mimic many aspects of infection with Leishmania in humans or experimental animals.
Finally, since PIV cells produced high levels of IFN-
(Figures 2B
and 3C
), we determined whether supernatants of these cultures or the PIV cells themselves would activate L. major-infected macrophages to destroy the parasite. We found that efficient killing of L. major occurred only when PIV cells were co-cultured with the infected macrophages (Figure 7
). One explanation for these observations is that cell-cell contact may be required between PIV cells and infected macrophages for activation of the macrophages to occur. A similar finding was reported in a mouse assay system.48
In conclusion, we have analyzed a human PIV system that closely mimics the response of humans to L. major by stimulating with living L. major. Analysis showed several differences between this human system and the murine system we studied previously.7,8 Perhaps the greatest differences between the two systems are that while mice can mount either a Th1 or Th2 response and the principal responding cell is a CD4 cell, humans respond with CD4 and CD8 cells, which mount principally, if not exclusively, a type 1 response (IFN-
). Moreover, the human PIV response to L. major is cross-reactive with other species of Leishmania. Since this PIV system mimics the response of patients to infection, it should be useful in dissecting the immune response to Leishmania, especially in the critical first days of infection when the nature of the subsequent immune response to the parasite is determined. In fact, analysis of the in vitro response of individuals to Leishmania can predict the effectiveness of a vaccine against the parasite.49 This same PIV approach should prove useful in analyzing vector (sand fly) effects on human immune cells and vector-based vaccines for human leishmaniasis, since recent evidence suggests that vector-based vaccines will be efficacious50,51 and that human immune cells are markedly affected by vector salivary molecules.52
Received March 22, 2004. Accepted for publication June 11, 2004.
Acknowledgments: We thank Drs. Gregory DeKrey, Lamine Mbow, Claudia Brodskyn, and Dean Gillespie for their insightful discussions and assistance. We are also indebted to Monica Estay, Julie Bleyenberg, Jeremy Jones, and Leanna Nosbisch for their excellent technical assistance.
Financial support: This work was supported by National Institutes of Health grants AI-27511 and AI-29955, the Colorado Institute for Research in Biotechnology, and the International Foundation for Ethical Research.
Authors addresses: Kathleen A. Rogers, Division of Geographic Medicine and Infectious Diseases, Tufts-New England Medical Center, Boston, MA 02111, Telephone: 617-636-8437, Fax: 617-636-5292, E-mail: KRogers{at}tufts-nemc.org. Richard G. Titus, Department of Microbiology, Immunology and Pathology, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, CO 80523-1619, Telephone: 970-491-4964, Fax: 970-491-0603, E-mail: rtitus{at}colostate.edu.
| REFERENCES |
|
|
|---|
. J Infect Dis 175: 737741.[ISI][Medline]
This article has been cited by other articles:
![]() |
N. GALINDO-SEVILLA, N. SOTO, J. MANCILLA, A. CERBULO, E. ZAMBRANO, R. CHAVIRA, and J. HUERTO LOW SERUM LEVELS OF DEHYDROEPIANDROSTERONE AND CORTISOL IN HUMAN DIFFUSE CUTANEOUS LEISHMANIASIS BY LEISHMANIA MEXICANA Am J Trop Med Hyg, March 1, 2007; 76(3): 566 - 572. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||