Crucial role of the central leptin receptor in murine Trypanosoma cruzi infection (original) (raw)

. Author manuscript; available in PMC: 2011 Oct 1.

Published in final edited form as: J Infect Dis. 2010 Oct 1;202(7):1104–1113. doi: 10.1086/656189

Abstract

Mice carrying a defective leptin receptor gene (_db/d_b) are metabolically challenged and upon infection with Trypanosoma cruzi (Brazil strain) suffer high mortality. In a genetically modified db/db mouse, the NSE-Rb db/db, central leptin signaling is reconstituted only in the brain which is sufficient to correct the metabolic defects. NSE-Rb db/db mice were infected with T. cruzi to determine the impact of the lack of leptin signaling on infection in the absence of metabolic dysregulation. Parasitemia, mortality rates and tissue parasitism were significantly increased in infected db/db mice compared with NSE-Rb db/db and FVB-wild-type mice. There was a reduction in fat mass and blood glucose in infected db/db mice. Plasma levels of several cytokines and chemokines were significantly increased in infected db/db mice compared with FVB and NSE-Rb db/db mice. Our findings suggest that leptin resistance in obesity and diabetes may have adverse consequences in _T. cruz_i infection.

Keywords: Trypanosoma cruzi, Chagas disease, diabetes, db/db mice, leptin receptor, leptin, adiponectin

Chagas disease caused by the protozoan parasite, Trypanosoma cruzi is endemic in Mexico, Central and South America [1]. In recent years, there has been an influx of immigrants into North America and Europe from endemic areas. Some of these immigrants are infected with T. cruzi and there has been an increase in the diagnosis of clinical Chagas disease in non-endemic areas [1–3]. The reactivation of T. cruzi infection is observed in individuals undergoing immunosuppressive therapy for malignancies and organ transplantation as well as in individuals with HIV/AIDS [1, 4–6].

Infection with T. cruzi results in an intense inflammatory reaction in many tissues. It has been studied most extensively in the heart where the inflammatory reaction is associated with an upregulation of cytokine and chemokine expression [7–10]. T. cruzi infection of cultured cardiac myocytes also results in increased expression of cytokines and chemokines [11, 12]. More recently, we demonstrated a similar response in adipose tissue and cultured adipocytes [13, 14]. The parasite persists in many organs studied including heart and adipose tissue [9, 10, 13, 15].

Our laboratory has been interested in the consequences of hyperglycemia on T. cruzi infection for many years and the increased incidence of obesity and diabetes in endemic areas with its potential to worsen the course of disease, has rekindled this interest. An association between the risk of T. cruzi infection and obesity or diabetes has been suspected. In recent years, there have been several reports indicating that diabetes and obesity may be more common in this population [16–22]. Although these observations may not have been subjected to rigorous statistical analysis, we believe that an important area of investigation is the consequences of T. cruzi infection in individuals with diabetes. In a study of chemically-induced diabetes with streptozotocin (STZ) in a mouse model of Schistosoma mansoni infection a reduction of granuloma size and an amelioration of hepatosplenic complications were observed [23]. However, we demonstrated that when mice with STZ-induced diabetes were infected with T. cruzi, they had a higher parasitemia and mortality [24]. When insulin was administered, the glucose returned to normal and the parasitemia and mortality were reduced.

In the 1970s and 1980s, there was increasing interest in the role of cell-mediated immunity (CMI) and parasitic disease. C57BLKS db/db mice are obese and hyperglycemic. These mice are a model of Type II diabetes and are known to have a defect in CMI. It was reported that when these mice were infected with S. mansoni, there was an attenuation of the size of the resulting hepatic granulomas and thus a reduction in complications such as portal hypertension [25]. CMI was also known to be important in the pathogenesis of T. cruzi infection and C57BLKS db/db mice infected with T. cruzi displayed increased parasitemia and mortality [24]. The underlying mechanisms of these observations remained unknown and the conventional wisdom was that it was due to defects in CMI and/or hyperglycemia. Subsequently, the molecular deficiency in db/db mice was identified and described as a mutation in the gene encoding the leptin receptor (LEPR) [26,27].

In the present study, we demonstrate that the hyperglycemic and obese db/db mice bred on a FVB background lacking LEPR signaling [28] displayed high parasitemia and mortality, whereas infected leaner and normoglycemic NSE-Rb db/db in which LEPR was replaced in the central neurons [29,30] had reduced mortality and low parasitemia, a phenotype comparable to the parental FVB-wild type mouse. Therefore, we conclude that normalization of the metabolic dysfunction in the db/db mouse through restoration of central LEPR signaling but not peripheral leptin signaling restores a normal peripheral response to T. cruzi infection.

MATERIALS AND METHODS

Animals

Mice homozygous for the Lepr (db) mutation are null for the STAT3-signaling competent isoform. These mice are obese, (average weight approximately 60 grams) hyperphagic, cold intolerant, hyperglycemic, insulin resistant, and infertile [29–31]. These mice were bred in our facility on a FVB background. FVB mice weighing approximately 25 grams are normoglycemic and were obtained from Jackson Laboratories (Bar Harbor ME). Mice in which the LEPR-B is expressed only in the brain are referred to as NSE-Rb db/db mice. These mice are devoid of LRPR-B outside the brain [29, 31]. They weigh approximately 40 grams and are relatively normoglycemic. All mice are congenic, having been backcrossed to the FVB strain for 10 generations [28]. These experiments were approved by the Institutional Animal Use Committee of the Albert Einstein College of Medicine.

Parasitology and Pathology

All mice were infected i.p. with 1x 104 trypomastigotes of the Brazil strain of T. cruzi. Parasitemia was assessed by counting in a hemocytometer chamber and mortality was recorded. Tissues were fixed in 10% buffered formalin and stained with hematoxylin and eosin.

Determination of macrophages in adipose tissue

In order to quantitate the number of macrophages in white adipose tissue we used qPCR for F4/80 and normalized using GAPDH as previously described [13]. The primers utilized were: Forward: CCCAGCTTATGCCACCTGCA; Reverse: TCCAGGCCCTGGAACATTGG

Body weight, body composition and food intake

Throughout the experiments, body weight was recorded. For body composition we used magnetic relaxometry employing the Echo Medical System EchoMRI-100 which provides quantitative measures of fat mass and fat-free mass in live mice up to 130 grams without anesthesia or sedation. Daily tuning occurs through automated calibration and testing procedures with self-correcting adjustments via custom software. Saturated and unsaturated fat standards are used to calibrate the MRI immediately prior to each set of MRI scans run in one day, and at the end of studies these are validated by carcass analysis. We measured food intake by weighing the food on a daily basis.

Laboratory blood determinations

Blood glucose was measured with a One Touch Ultra Glucometer (Life Scan Inc., Burnaby, Canada). Adiponectin was determined by mouse RIA adiponectin kit (Millipore Corporation, Saint Charles, MO). Cytokines and chemokines were determined by multiplex analysis analyses (Millipore Corporation) and performed in the Hormone Assay Core of the Einstein Diabetes Research and Training Center.

Statistical analysis

Significance was determined using the student t test and the data are presented as the mean ± standard error of the mean.

RESULTS

Parasitology and Pathology

When wild type-FVB mice were infected, there was a transient parasitemia at 25 to 30 days post infection (dpi), which subsequently disappeared (Fig 1A). The same was true for infected NSE_-_Rb db/db mice. In contrast, infected FVB-db/db mice displayed a significantly higher parasitemia compared with infected NSE-Rb db/db and FVB mice (Fig 1A). Importantly, while there were no deaths in the FVB and NSE-Rb db/db mice, all of the db/db mice were dead by day 35 (Fig 1B). The hearts of infected db/db mice revealed multiple parasite pseudocysts and an intense inflammatory reaction (Fig 2). In contrast, the hearts obtained from infected NSE-Rb db/db mice and FVB mice revealed virtually no pseudocysts and there was a less intense inflammatory infiltrate (Fig 2 ) and at 90 dpi, the hearts of the infected FVB and NSE-Rb db/db mice were normal and displayed no inflammation and fibrosis (not shown).

Figure 1.

Figure 1

A. Parasitemia in mice infected with T. cruzi (Brazil strain) (n=5 for each group). 1B: Mortality in mice infected with T. cruzi (Brazil strain). Beginning with 18 days post infection there was a significant (* p<0.05) increase in peripheral parasitemia in infected db/db mice and a 100% mortality compared with NSE-RB db/db and FVB mice (top line).

Figure 2.

Figure 2

Representative myocardial histopathology in mice infected with T. cruzi (Brazil strain). Top panels are all low power (10X) and bottom panels are all high power (20x). db/db (LEPR-B null mice), NSE-Rb db/db and FVB mice. Note the increase parasite load and inflammation in the infected db/db mice. Arrows point to pseudocysts which contain T. cruzi amastigotes.

Body weight, body composition and food intake

Over the period of observation there was a trend to a reduction in body weight in the infected db/db mice which did not achieve statistical significance. The infected NSE-Rb db/db mice exhibited no significant weight change, but the infected FVB displayed a significant weight gain. In all three infected groups there was a significant loss in fat mass and a significant increase in fat-free mass (table 1). Fat-free mass is a measure of water and the volume of visceral organs. It is well-established that organomegaly is associated with this infection. The hyperphagic uninfected db/db mice ingested 15 to 20 grams of food daily. Infected db/db mice ingested 15 to 20 grams daily until 13 dpi thereafter they ingested 5 to 10 grams of food per day until death. There were five mice in each group (p< 0.05). Infected and uninfected NSE-Rb db/db mice consumed between 7.5 and 8.5 grams of food per day during a 26 days period of observation post infection. This difference was not significant.

Table 1.

Body weight and composition of three _T. cruzi_-infected mouse groups

Days Post-infection Genotype
db/db NSE-Rb db/db FVB
Body Weight (g) 0 60.36 ± 2.26 41.75 ± 0.40 21.60 ± 0.87
14 60.32 ± 2.34 40.73 ± 1.09 24.72 ± 0.72
26 49.63 ± 2.72 39.50 ± 1.16 26.34 ± 0.82*
Fat mass (g) 0 31.50 ± 1.78 16.29 ± 1.89 3.92 ± 0.58
14 27.51 ± 1.60 11.85 ± 1.51 2.70 ± 0.44
26 17.27 ± 2.03* 5.95 ± 0.68* 1.67 ± 0.23*
Fat-free mass (g) 0 27.92 ± 0.60 24.10 ± 1.26 16.46 ± 0.40
14 32.21 ± 0.74 28.14 ± 0.70 21.10 ± 0.36
26 30.86 ± 1.00 * 32.25 ± 1.27* 21.55 ± 0.77*

Blood glucose determinations

Post-absorptive blood glucose concentrations were determined at baseline prior to infection and at various time-points after infection. Infection caused a decrease in plasma glucose across all genotypes, but the fold-reduction was significantly greater in the db_/db_ mice (P<0.05) (Fig 3A). The decrease in glucose levels coincided with the reduction in food intake.

Figure 3.

Figure 3

A: Blood glucose determinations in mice infected with T. cruzi (Brazil strain) (n=5 for each group) over 27 days post-infection. There was a significant reduction (p<0.05) in blood glucose concentration in infected db/db mice beginning at day 15 post infection. In the FVB and NSE-RB db/db mice the reduction in blood glucose was significant at day 22 post-infection compared to day zero**. B:** Adiponectin levels in mice infected with T. cruzi (Brazil strain) (n=5 for each group). There was a significant reduction (p<0.05) in adiponectin levels in all 3 groups compared to day zero. Note that in the db/db mice, there was a significant reduction in adiponectin prior to infection.

Blood Adiponectin levels

Consistent with our previous studies, [13] there was a reduction in serum adiponectin levels in all infected mice (P<0.05). As expected, the pre-infection adiponectin levels in db/db mice were significantly lower compared with the other groups (P<0.05) (Fig 3B). This is consistent with the fact that these mice are obese and have many of the characteristics of human Type II diabetes and the metabolic syndrome (hyperglycemia, obesity, low adiponectin).

Blood Cytokines and chemokines levels

Plasma levels of IL-6, TNF-α, were significantly increased (p<0.05) in infected db/db mice compared with infected FVB and NSE-Rb db/db mice (Fig 4). IFN-γ levels were elevated in all three infected groups but only significant in db/db mice compared with other two groups at 27 dpi. There were no clear differences among the three groups regarding levels of IL-1β (data not shown). The levels of CCL2, CCL3, and CCL5 were significantly elevated in infected db/db mice over the two other infected groups (Fig 5). Levels of CXCL-10 were elevated in infected db/db and NSE-Rb db/db mice.

Figure 4.

Figure 4

Plasma cytokine levels in mice infected with T. cruzi (Brazil strain) (n=5 for each group) over 27 days post-infection. There was a significant (*p<0.05) increase in IL-6 and TNF-α in infected db/db mice compared with FVB and NSE-RB db/db mice. The levels of IFN-γ were increased in all three groups but were significantly increased (*p<0.05) in db/db mice compared with FVB and transgenic only at day 24 post infection.

Figure 5.

Figure 5

Plasma chemokine levels in mice infected with T. cruzi (Brazil strain) (n=5 for each group) over 27 days post-infection. Levels of CCL-2, CCL-3 and CCl-5 were significantly (*p<0.05) increased in infected db/db mice. CXCL-10 levels were significantly (*p<0.05) increased in infected db/db and NSE-RB db/db mice.

Determination of macrophages in adipose tissue

By using qPCR analysis we determined the F4/80 expression in white adipose tissue which is a measurement of macrophage recruitment. Baseline values for F4/80 in db/db mice were elevated compared with FVB and NSE-Rb db/db mice reflecting the increased adiposity of db/db mice (Fig 6). After 15 dpi, the values did not change in db/db mice. However, significant increases in F4/80 expression were seen in infected FVB and NSE-Rb db/db mice

Figure 6.

Figure 6

Determination of macrophages by qPCR analysis of F4/80 in white adipose tissue obtained from mice infected with T. cruzi (Brazil strain) 15 days post infection. The baseline values for db/db mice are elevated and infection does not results in an increase over base-line. (n=5 for each group). There was a significant increase (*p<0.05) in levels of F480 in infected NSE-RB db/db and FVB mice.

DISCUSSION

In the present report, FVB db/db mice, which are obese, hyperglycemic and lack the B isoform of the leptin receptor (LEPR-B), exhibit high blood parasitemia, intense tissue parasitism in the heart and early death in response to T. cruzi infection. These observations were associated with a reduction in blood glucose and adiponectin. T. cruzi infection also resulted in a significant increase in levels of some serum cytokines and chemokines in infected db/db mice. Infected FVB mice and NSE-Rb db/db mice had no mortality, minimal blood parasitemia and little cardiac pathology. These mice also displayed a reduction in blood glucose and adiponectin, but the fold-reduction was much less than in infected db/db mice. Body composition analysis revealed a significant loss in fat mass in all infected groups and an increase in fat-free mass. The loss of fat mass may be due to a reduction in food intake or lipolysis, a consideration currently being examined in our laboratory. The increase in fat-free mass is likely due to infection-induced increase in water and visceral organ volume. In the present report, infected FVB and NSE-Rb db/db mice had a modest reduction in blood glucose whereas, in infected db/db mice, the reduction in blood glucose was more profound. Thus, the restoration of leptin signaling in the brain led to a reversal of the metabolic dysfunction and resistance to T. cruzi infection. The precise mechanism(s) for this observation may be the reversal of the hyperglycemia or that central leptin signaling has an anti-inflammatory effect in the periphery.

The recognition that db/db mice had a defect in the LEPR demonstrated in 1995 [26, 27] well after our initial publication in 1988 [24]. In the present study**,** we employed the FVB-db congenic line which has several advantages over the C57BLKS db line used in our prior studies [24]. While BLKS-db/db mice lose weight due to severe insulinopenia from loss of pancreatic β-cells, the FVB-db/db mice continue to produce insulin and gain weight over time. This feature of the FVB-db/db mice helps in ascertaining the causes of morbidity and mortality of T. cruzi infection in diabetes. Additionally, the FVB-db/db mice survive well beyond 9 months of age, unlike BLKS-db/db mice that suffer early mortality between 3 to 5 months of age. The near-normal life span of FVB-_db/d_b mice permits longitudinal studies of chronic T. cruzi infection. However, in our case, T. cruzi infection of these mice resulted in early death.

FVB db/db mice are obese, hyperglycemic and lack the LEPR which is important in a variety of situations in health and disease [32, 33]. The NSE-Rb db/db and FVB mice are lean and are normoglycemic. Infection of all mice resulted in a reduction of blood glucose. However, the absolute decrease in blood glucose in the infected db/db mice was greater compared to the other two groups. At the time of death, the glucose levels in the infected db/db mice had dropped from a baseline value of 500–600 mg/dl to ~110 mg/dl. The drop in blood glucose can be attributed to the reduction in food intake. Insulin levels in the FVB-db/db are elevated compared with BLKS-db/db mice suggesting that insulin levels do not correlate with parasitemia [24].

We analyzed cytokine and chemokine levels in the blood after infection in FVB, NSE-Rb db/db and db/db mice. Several cytokines and chemokines were significantly elevated in the infected db/db group. These data are consistent with observations made by our group and others that in hearts and adipose tissue obtained from infected mice as well as in infected cultured cardiac myocytes and adipocytes that there are increases in inflammatory mediators [7,8,10,11,14,19]. Importantly, the _T. cruzi_-associated increases in cytokines and chemokines observed in the db/db mice were restored to the level of FVB mice in the NSE-RB db/db mice.

Individuals with diabetes have an increase in bacterial foot infections. This was recently studied in the diabetic, LEPR null mouse (C57BL/6J-Lepr_db_ mouse), infected in the paw with the bacterium Staphylococcus aureus. Whereas non-diabetic mice resolved the infection within 10 days, in LEPR null mice the infection was prolonged with a more robust inflammatory response which the authors attributed to the increased synthesis of chemokines CCL1 and CCL2 [34]. The same may be true in our model of T. cruzi infection in db/db mice where the increased inflammation in the heart may be due, in part, to the increased expression of several cytokines and chemokines such as TNF-α, IL-6, IFN-γ, CCL-2, CCL-3 and CCL-5. In contrast, IL-1β levels were significantly increased in all infected groups and there was no clear difference between them.

Adiponectin is a polypeptide which circulates as high and low molecular weight isoforms. It has many beneficial qualities when normally expressed and detrimental when expression is reduced [35–39]. Paradoxically, even though adiponectin is synthesized by adipose tissue, it is reduced in human obesity and in obese rodents. Visceral adipose pads (intra-abdominal/mesenteric) are the prominent sources of systemic adiponectin in the lean state and its production is significantly reduced in the obese state. Adiponectin contributes towards protection against cardiovascular disease [35–39]. Previously, we reported a reduction in adiponectin levels in serum and adipose tissue in T. cruzi infected mice [13]. Infection of cultured adipocytes also resulted in a reduction in adiponectin levels and an increase in the levels of chemokines and cytokines [14]. The plasma levels of adiponectin were reduced in all infected mice over a 26 day period of observation. As expected, in the obese and diabetic db/db mice the baseline levels of adiponectin was significantly lower compared with infected FVB and NSE-RB db/db mice. It has been hypothesized that adiponectin has inflammatory-modulating activities [35–39] and some of the proven anti-atheromatous effects of adiponectin may be mediated by anti-inflammatory activities acting directly on the vasculature. Okamoto et al [40] reported that adiponectin inhibited the production of CXCR-3 chemokine ligands in macrophages and causes a reduction in T-lymphocyte recruitment. Adiponectin production by adipocytes in culture is inhibited by inflammatory cytokines such as TNF-α. This inhibition may be mediated by one of its signaling targets, the AMP-activated protein kinase (AMPK) [38].

Leptin, a hormone derived from adipocytes, like adiponectin links metabolism, nutritional status and the immune response. The actions of leptin are mediated through the LEPR [41,42]. The leptin receptors in the brain responsible for the metabolic regulation are predominantly found in the hypothalamic area. However, leptin receptors show widespread expression in many regions of the brain, but the functional relevance of leptin receptors outside the hypothalamic region remains poorly understood [42].

The role of leptin in infection is not completely understood. Leptin upregulates mouse macrophage phagocytic function by virtue of its link to the activation of phospholipase [43]. In macrophages, leptin upregulates the synthesis of several pro-inflammatory cytokines such as TNF-α, IL-6, IFN-γ and IL-12 [44, 45] and leptin signaling deficiency causes a mild T-cell mediated immune defect. Plasma levels of leptin are elevated during bacterial sepsis, but it is unclear if it directly mediates the anorexia observed during sepsis [38]. Interestingly, leptin levels are also elevated in T. cruzi infected CD-1 mice [13] and may account, in part, for the loss of adipose tissue during infection.

The roles of glucose and leptin have also been examined in the pathogenesis of bacterial and Mycobacterial infections. For example, leptin has been demonstrated to have a role in the early immune response in mice during pulmonary infection caused by Mycobacterium tuberculosis [46]. This is thought to mediate an effective IFN-γ-driven Th1 response and granuloma formation and Martena et al [47] demonstrated that in mice rendered diabetic by the administration of STZ, there was an increased burden of Mycobacterium tuberculosis in chronically but not acutely infected mice. In leptin deficient (ob/ob) and LEPR-deficient (db/db) mice, there was an impairment of host resistance to the intracellular bacterium, Listeria monocytogenes [48]. Similarly, in the present report there was a significantly higher burden of parasite in the hearts of infected db/db mice compared with FVB or NSE_-_Rb db/db, suggesting that LEPR contributes to resistance to infection in this model.

Recently, cerebral malaria was studied in a mouse model of obese leptin deficient ob/ob mice and ob/+ mice [49]. When infected with Plasmodium berghei ANKA the leptin deficient mice were resistant to the development of cerebral malaria whereas the ob/+ mice, which are not deficient in leptin, developed signs of cerebral involvement [49]. While the mechanism of these observations is not known, these observations indicate that the potential role of leptin signaling is not the same for every parasite.

Our observations strongly suggest that direct leptin action on peripheral T-cells is not responsible for the protective effect against T. cruzi infection. This is the first time that this type of experiment could be performed. We cannot exclude an indirect effect of central leptin action on T-cell function. The LEPR transgene expression in neurons reduced the hyperglycemia and adiposity of db/db mice, suggesting that either the improvements in glycemia or adiposity per se are responsible for improved resistance to T. cruzi infection. As noted above, insulin itself is not a major factor in the control of parasite proliferation. Thus, the current study differs from many of the studies cited, in that, as a control, we utilized the NSE-Rb db/db or transgenic mouse model in which the LEPR gene was present only in the central neurons [29,30]. Our study is unique because we demonstrate that the restoration of the metabolic dysfunction through central mechanisms is an important determinant in susceptibility to infection and mortality.

To our knowledge, this is the first report that central LEPR expression, either through regulation of metabolic parameters or immune functions, contributes to resistance to T. cruzi infection in a mouse model. The present study does raise interesting questions regarding the consequences of T. cruzi infection in an individual with Type II diabetes and the metabolic syndrome as well as in individuals with obesity where leptin resistance is often observed. Additionally, our data argues against an involvement of leptin in the periphery via direct interactions on immune cells in this infection. Leptin signaling is an area of research that requires increased attention as it relates to the pathogenesis of infections in humans.

Acknowledgments

Supported in part by NIH Grants R01-AI-076248, R01-HL-73732, R21-AI-06538, (HBT), R01-DK55758, R01-CA112023 and RC1 DK086629 (PES), P60-DK020541 (SCC, GJS). DK047208 (GJS), NIH/FIRCA (1 R03-TW006857 (MMT,HBT), Conselho Nacional de Desenvolvimento Científico e Tecnólogico (CNPq) (FSM and MMT), PRPq-UFMG (FSM) 3.

Footnotes

Authors have no commercial or financial interest or conflict of interest

References