|Home | About | Journals | Submit | Contact Us | Français|
Brown adipose tissue (BAT) and white adipose tissue (WAT) and adipocytes are targets of Trypanosoma cruzi infection. Adipose tissue obtained from CD-1 mice 15 days after infection, an early stage of infection revealed a high parasite load. There was a significant increase in macrophages in infected adipose tissue and a reduction in lipid accumulation, adipocyte size, and fat mass and increased expression of lipolytic enzymes. Infection increased levels of Toll-like receptor (TLR) 4 and TLR9 and in the expression of components of the mitogen-activated protein kinase pathway. Protein and messenger RNA (mRNA) levels of peroxisome proliferator-activated receptor γ were increased in WAT, whereas protein and mRNA levels of adiponectin were significantly reduced in BAT and WAT. The mRNA levels of cytokines, chemokines, and their receptors were increased. Nuclear Factor Kappa B levels were increased in BAT, whereas Iκκ-γ levels increased in WAT. Adipose tissue is an early target of T. cruzi infection.
Adipose tissue, the largest endocrine organ in the body, consists of adipocytes, fibroblasts, macrophages, and endothelium . Adipose tissue is not only a storage site for triglycerides but a major contributor to a variety of physiologic functions, including energy homeostasis and immune responses [2–7]. Thus, any insult that affects adipose tissue will likely have an effect on the entire systemic metabolism. The role of adipose tissue and adipocytes in infection has recently been reviewed . An important finding linking adipose tissue and response to infection is the observation that injection of lipopolysaccharide into mice that were rendered fatless did not result in immediate death of mice, as was seen in control mice with normal adipose tissue. This observation underscores the importance of the adipocyte as a factor in the innate immune response to infection [9, 10].
Chagas disease, caused by the hemoflagellate protozoan Trypanosoma cruzi, is a major cause of morbidity and mortality in endemic areas of Latin America and a major cause of heart disease [11, 12]. There has been an increase in the immigration of individuals from these areas to nonendemic areas, where the disease is now recognized with increased frequency . Chagas disease can present as an opportunistic infection in immunosuppressed individuals, especially organ transplant recipients and those with human immunodeficiency virus (HIV)/AIDS [14–16].
Acute T. cruzi infection causes an intense systemic inflammatory response. Although many organs may be affected, T. cruzi infection has been most extensively examined in the heart, where there is a strong inflammatory response and an upregulation of inflammatory mediators [17, 18]. As this inflammatory reaction subsides, there is extensive fibrosis and remodeling of the cardiovascular system. Eventually, 15%–30% of infected individuals display myocardial dysfunction .
Shoemaker et al [19, 20] observed that T. cruzi parasitized adipose tissues, especially brown adipose tissue (BAT). Andrade and Silva  demonstrated that T. cruzi parasitized adipocytes in vivo. In recent years, our laboratory has confirmed these findings and also examined the consequences of these observations on the pathogenesis of T. cruzi infection in a mouse model . In that regard, electron microscopy clearly demonstrated that adipose tissue and adipocytes are readily invaded by T. cruzi and that parasites persist in adipose tissue long after the acute phase has subsided [22, 23]. Here, we examined the response in adipose tissue to T. cruzi infection 15 days after infection in CD-1 mice. We found that at this early or prepatent stage of infection, during which there was no overt evidence of infection, there was already an increased expression of inflammatory mediators and a concomitant reduction in triglyceride content of adipose tissue. In addition, this is the first report to identify differential responses of BAT and white adipose tissue (WAT) in response to T. cruzi infection.
Male CD-1 mice (Jackson Laboratories) were infected intraperitoneally at 8–10 weeks of age with 5 × 104 trypomastigotes of T. cruzi (Brazil strain). Fifteen days after infection, mice were sacrificed, and BAT and WAT from various depots were harvested for analysis. At this time, there was no peripheral parasitemia, and mice moved around the cage and had no ruffled fur.
We used magnetic relaxometry (magnetic resonance sepctroscopy, MRS) employing the Echo Medical System EchoMRI-100, which provides quantitative measures of fat mass and fat-free mass in live mice up to 130 g 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 EchoMRI immediately prior to each set of MRS scans run in 1 day. In our experiments, 5 mice were used for each group (infected and uninfected).
Anti–peroxisome proliferator-activated receptor γ (PPAR-γ) antibody was obtained from Santa Cruz Biotechnology, and antibodies to components of the mitogen-activated protein kinase (MAPK) pathway (total and phosphorylated extracellular signal-regulated kinase (ERK), p38 MAPK, and c-Jun-N-terminal kinase [JNK]) and lipases (adipose triglyceride lipase [ATGL], hormone sensitive lipase [HSL], and lipoprotein lipase [LPL]) were obtained from ABCAM. Adiponectin antibody was produced in the laboratory of Dr Scherer as described previously . Fifteen days after infection, mice were sacrificed, and BAT and WAT were collected and frozen at –80°C until further use. For immunoblot analysis, protein lysates (30–40 μg) were prepared from frozen tissues, resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis on 12% acrylamide gels, and transferred to BA85 nitrocellulose (Schleicher & Schuell). Blots were probed with antibodies as indicated. Primary antibodies (1:1000) and secondary antibodies (mouse 1:2000; rabbit 1:5000) were diluted in phosphate-buffered saline with 0.1% Tween-20 and 1% bovine serum albumin. Horseradish peroxidase–conjugated goat anti-rabbit and goat anti-mouse antibodies were obtained from Biomeda Corp. Bound antibodies were detected by enhanced chemiluminescence according to the manufacturer’s instructions (GE Healthcare). Immunoblots from a minimum of 3 independent experiments were scanned, and the background-corrected signal from each band was quantified by densitometry using an Alpha Innotech Multi-image Light Cabinet with Chemi-imager 4400 software (Alpha Innotech). The signal for each sample lane was normalized to the signals obtained for guanosine diphosphate dissociation inhibitor (GDI). The normalized relative levels for each experimental group are represented as the mean plus or minus the standard error of the mean (SEM).
Freshly isolated adipose tissue was fixed in 10% normal buffered formalin overnight and embedded in paraffin wax. Five-micrometer-thick sections were used. The sections were deparaffinized, then boiled at 95°C for 20 minutes in sodium citrate solution (DAKO) for antigen retrieval. To assess for macrophage activity, immunohistochemical analysis was performed using rabbit antibody to ionized calcium-binding adaptor molecule 1 (Iba1) (Wako Chemicals), a polypeptide that is selectively expressed in cells of monocytic origin. The sections were incubated overnight at 4°C with Iba1 at a dilution of 1:300. A standard avidin-biotin complex method (Vector Laboratories) was used for the secondary antibody (anti-rabbit), using a 1:200 dilution and a 1-hour incubation. Slides were developed using a peroxidase detection kit (Vector Laboratories) counterstained with hematoxylin (Sigma-Aldrich) after immunolabeling.
Freshly isolated tissues were fixed for 28 hours in 4% paraformaldehyde at 4°C, soaked in 30% sucrose (4°C) over 2 days, then frozen in optimum cooling temperature and stored at –80°C. The tissues were sectioned into 30-micrometer-thick slices and mounted onto glass slides. Tissues were stained with Oil Red O (Sigma-Aldrich) for 3 hours. The sections were counterstained with hematoxylin for 25 seconds. The lipid content of frozen sections was quantified using Oil Red O staining. Frozen sections stained with Oil Red O were washed 3 times in isopropyl alcohol for 3 minutes each. Bound Oil Red O was eluted by incubating in isopropyl alcohol (5 mL) for 16 hours; the eluted lipid content stained with Oil Red O was then measured at 540 nm in a spectrophotometer (Shimadzu uv-1201).
Adipose tissue (BAT and WAT) was obtained from mice 15 days after infection, fixed in 10% buffered formalin, embedded in paraffin, and stained with hematoxylin and eosin. The size of the adipocytes was determined using the National Institutes of Health Image J image software.
RNA was isolated from adipose tissue using Trizol (Invitrogen). Total RNA was purified using the RNeasy Mini Kit (Qiagen). RNA was reverse-transcribed from 1 μg of total RNA in a final volume of 50 μL using Superscript III transcriptase (Invitrogen). All procedures were performed according to the manufacturer’s directions. The list of 5′-3′ primer sets used for the quantification of respective genes is as follows:
Toll-like receptor (TLR) 2: Forward (F): TTAAGCGAGTCTGCTTTCCT; Reverse (R) CTTCAGGAGCTAAAGC.
TLR4: F: ACTTTATTCAGAGCCGTTGG; R: CCATTCCAGGTAGGTGTTTC.
TLR9: F: AGCCTCCGAGACAACTACCT; R: GC TGAGGTTGACCTCTTTCA.
PPAR-γ: F: GTCTGTGGGGATAAAGCATC; R: CTGATGGCATTGTGAGACAT.
GAPDH: F: AACTTTGGCATTGTGGAAGG; R: ACACATTGGGGGTAGGAACA.
NFκB: F: CTACGGAACTGGGCAAATGT; R: TCGAAATCCCCTCTGTTTTG.
Adipoq: F: AGACAGGAGATGTTGGAATGAC; R: TACACCGTGATGTGGTAAGAGA.
F4/80: F: CCCAGCTTATGCCACCTGCA; R: TCCAGGCCCTGGAACATTGG.
HGPRT: F: TGTTGTTGGATATGCCCTTG; R: TTGCGCTCATCTTAGGCTTT.
Additionally, the primers for the parasite load analysis have been previously described . The quantitative real-time polymerase chain reaction (qPCR) was run using PCR SYBR Green Master Mix (Roche Applied Science) and magnesium chloride in the iQ5 LightCycler (Bio-Rad). To normalize the gene expression, messenger RNA (mRNA) expression of the housekeeping gene HPRT was amplified. The genes were amplified in triplicate using the reaction conditions and analytical parameters described previously . Fold increases in target mRNA expression were calculated as described earlier .
RT2 Profiler PCR array mouse inflammatory cytokines and receptors (PAMM-011A) (n = 3) were used to analyze the expression of a focused panel of genes. Data analysis was performed using the ΔΔCT method according to the manufacturer’s protocol (SABiosciences).
The presence of macrophages 15 days after infection was determined by immunohistochemical (IHC) analysis and qPCR. For IHC, we used the macrophage-specific marker Iba1, and for qPCR we used F4/80 mRNA. Macrophage staining increased in both BAT and WAT, which is evident in both low-and high-magnification images (Figure 1). Consistent with this observation, there was a significant upregulation of the macrophage marker F4/80 mRNA: 10-fold in infected BAT and 6.5-fold in WAT.
Adipose tissue carried significantly more (P < 0.5) parasites compared with heart and spleen of the infected mice during acute infection. The average number of parasites/cell in adipose tissue decreased from 15 days after infection to 30 days after infection as analyzed by qPCR . The number of parasites per cell for heart, spleen, BAT, and WAT 15 days after infection was 0.36, 0.02, 3.99, and 7.6, respectively. The number of parasites 30 days after infection was 0.69, 0.005, 0.145, and 3.4, respectively.
We examined the expression of cytokines, chemokines, and chemokine receptors by PCR arrays and found that the mRNA levels of interferon γ (IFN-γ) and tumor necrosis factor α (TNF-α) were significantly elevated (10.7- and 8.15-fold, respectively) in infected BAT (Table 1). In infected BAT, the CC chemokines (CCL2, CCL4, CCL5, CCL6, CCL7, CCL8 and CCL12) were upregulated, as were the levels of CXCL9 and CXCL10. The levels of interleukin 1α (IL-1α), interleukin 1β (IL-1β), interleukin 17β (IL-17β), and interleukin 18 (IL-18) were also increased. There was increased expression of cytokine and chemokine receptors Tnfrsf1b, CCR1, 2, 3 and 5, CXCR3, interleukin 2 receptor β (IL-2 Rβ), and interleukin 2 receptor γ (IL-2 Rγ). The mRNA level of CD40 ligand (CD40lg) was increased.
In WAT, there were no significant increases in IFN-γ and TNF-α. Additionally, there were no significant changes in the mRNA levels of IL-1α, IL-1β, IL-17β, and IL-18. However, the levels of CC chemokines (CCL2, CCL4, CCL5, CCL6, CCL7, CCL8, and CCL12) were upregulated, and CXCL9 and CXC10 showed a 9.9- and 2.5-fold increase in expression, respectively. Several cytokines and chemokine receptors, such as CCR2, CCR3, CCR5, and CCR7 and IL-2 Rγ, were increased in infected WAT (Table 1).
To quantify the effect of T. cruzi infection on BAT and WAT, samples were stained with fat-specific Oil Red O, which specifically stains neutral triglycerides and lipids red (Figure 2A). We observed a change in the lipid content of both BAT and WAT, and the lipid loss appeared more pronounced in WAT. We quantified the Oil Red O content and found that there was a 15% reduction in lipid content of BAT and a 25% loss of lipids in WAT as a result of infection (Figure 2B). Using MRS, we found overall a 38% reduction in fat mass in infected mice (Figure 2C). Immunoblot analysis revealed a significant increase in the expression of lipases (ATGL, HSL, and LPL) in BAT and WAT during acute infection (days 15 and 30 after infection) (Figure 2D and and2E).2E). Hematoxylin-and-eosin staining of adipose tissue demonstrated a significant reduction of adipocyte size in both WAT and BAT obtained from infected mice (Figure 3). This was quantitated using the J Image program. We observed 2.8- and 4.2-fold decreases in the cell area of adipocytes from BAT and WAT, respectively, at 15 days after infection (Figure 3). The results of the Oil Red O and MRS studies taken together with the adipocyte size analysis clearly demonstrate a decrease in adipose tissue early in infection.
TLRs belong to the family of pattern recognition receptors that recognize conserved pathogen-associated molecular patterns [25, 26]. Pattern recognition receptors play a key role in the adipocyte immune response as well as in energy homeostasis. The qPCR analyses revealed a significant increase in the mRNA levels of TLR2 and TLR9 in both infected BAT and WAT (Figure 4A), whereas elevated mRNA levels of TLR4 were observed only in infected WAT. The fold-increase in TLR4 and TLR9 mRNA was significantly higher in WAT (8.4- and 16.7-fold, respectively) as compared with BAT (1.1- and 4.7-fold, respectively).
MAPKs function as signal transducers from cell surface receptors to transcription factors in the nucleus, which consequently triggers long-term cellular responses. Activation of the components of the MAPK pathway (ERK, p38 MAPK, and JNK) play an important role in cell proliferation, differentiation, and cell death [27, 28]. A significant increase in phosphorylated (activated)-ERK, p38 MAPK, and JNK was observed by immunoblot analysis (Figure 4B–E). We observed similar inductions of pERK and phospho-p38 MAPK levels in T. cruzi–infected cultured adipocytes , endothelial cells, smooth muscle cells, and heart [29–31]. Herein there was a significant increase in pJNK levels in both WAT and BAT (Figure 4B–E).
PPAR-γ, a nuclear receptor that functions as a master transcription factor for adipogenesis, plays an important role in differentiation, proliferation, and metabolism of adipocytes. In BAT, T. cruzi infection did not result in changes in protein levels; however, there was a significant reduction in mRNA levels. Infection resulted in a significant increase in both protein and mRNA levels of PPAR-γ in WAT (Figure 5).
Adiponectin levels were significantly reduced in adipose tissue obtained from infected mice as determined both by immunoblot and by qPCR (Figure 6A–C).
The qPCR analyses of infected BAT demonstrated an increase in mRNA levels for Nuclear Factor Kappa B (NFκB) and a decrease in mRNA levels for IKK-γ; conversely, infected WAT displayed a downregulation of NFκB with elevated mRNA levels of IKK-γ (Figure 6D).
Our laboratory group has established that T. cruzi parasitizes adipose tissue and adipocytes resulting in an upregulation of inflammation [22, 23]. Here, we examined inflammatory pathways in adipose tissue obtained from mice 15 days after infection, a time characterized by a lack of blood parasitemia and overt illness. However, our previous observation demonstrated higher levels of parasite load in adipose tissue compared with heart and spleen at later time points. Thus, 15 days after infection was chosen as a latent time point in this particular model because there is usually no peripheral parasitemia and the mice are clinically well without any signs of disease. Surprisingly, we found that even at this early time point, there was already a significant influx of macrophages into adipose tissue. This influx of macrophages was accompanied by a reduction in the expression of adiponectin and PPAR-γ in BAT; both are negative regulators of inflammation. However, PPAR-γ expression was elevated in WAT. In both infected WAT and BAT, there was also an increase in the expression of several cytokines, chemokines, chemokine receptors, and MAPKs. The TLRs were also upregulated, notably TLR2 and TLR9. In the current study, we did not determine which cells were responsible for the upregulation of TLRs. However, previously we demonstrated that T. cruzi–infected cultured adipocytes result in the upregulation of TLRs, and others have shown similar results with other cell types [26, 32]. Oil Red O staining demonstrated a reduction in lipid content, and MRS verified a significant reduction in fat mass by 15 days after infection. This was associated with an increased expression of lipolytic enzymes. Infection-induced upregulation of the inflammation results in lipolysis. Interestingly, brain natriuretic peptide, which is increased in some individuals with chagasic heart disease , has been linked to lipolysis . Inflammation can also result in a reduction in insulin signaling . Finally, our data strongly suggest that at this early, latent stage of infection, the response of BAT and WAT differs.
The increase in macrophages in infected adipose tissue is reminiscent of the findings observed in morbid obesity [36, 37], which may be due in part to lipolysis . The stromal vascular fraction of adipose tissue is comprised of preadipocytes, macrophages, fibroblasts, and other cell types. In the current study, immunohistochemical identification of the macrophage-specific marker Iba1 and the robust increase in F4/80 mRNA in infected adipose tissue demonstrate an increased macrophage infiltration both in BAT and WAT during acute infection. Oil Red O staining further indicates that lipolysis occurred in both BAT and WAT (but was more evident in WAT). Thus, macrophage infiltration is likely the major contributor to adipose tissue inflammation during acute T. cruzi infection.
Adiponectin, when normally expressed, has potent anti-inflammatory, anti-apoptotic, anti-atherosclerotic activity and insulin-sensitizing properties . In contrast, adiponectin expression is reduced in the obese state , and adiponectin levels in serum are negatively correlated with plasma glucose and insulin and serum triglycerides. The suggested mechanism for this observation is that adiponectin has anti-inflammatory properties and there is an inverse relationship between adiponectin levels and serum markers of inflammation [2–7, 37]. Adiponectin levels were significantly reduced by infection in both BAT and WAT. This observation, taken together with results from our previous studies, suggests that T. cruzi infection of cultured adipocytes results in a reduction of adiponectin production. The local inflammation of infected adipose tissue may be a driving force in suppressing adiponectin production. Adiponectin production by adipocytes in culture is inhibited by inflammatory cytokines, such as TNF-α , and this inhibition may be mediated in part by NFκB signaling . We observed differences in the inflammatory phenotypes and NFκB regulation in the 2 adipose tissue types examined here during acute T. cruzi infection, suggesting that fat depots from different anatomical sites play differential roles in the regulation of inflammation during the course of infection.
Adipocytes contribute to innate immunity by modulating the secretion of cytokines and chemokines. IL-1 is a proinflammatory cytokine, and we have reported the upregulation of IL-1 in cultured adipocytes  infected with T. cruzi. Here, we observed an upregulation of IL-1β, IL-1α, and their receptors by qPCR analysis in infected BAT 15 days after infection. IL-1 α and IL-1β induce cellular responses through IL-1R. Interestingly, the adaptor protein Toll interacting protein, which inhibits IL-1–induced signaling pathway, is downregulated (data not shown) in infected BAT.
In BAT, infection with T. cruzi is also marked by a negative regulation of the nuclear receptor PPAR-γ, which regulates lipid metabolism and controls lipid homeostasis and cellular differentiation through gene transcription. Trypanosoma cruzi infection in BAT also causes an upregulation of TNF-α, which negatively regulates lipid homeostasis through downregulation of PPAR-γ. Although PPAR-γ is negatively regulated by TNF-induced inhibition of adipogenesis and de-differentiation of mature adipocytes , activation of PPAR-γ has been shown to upregulate IKK-γ expression, preventing nuclear translocation of the proinflammatory transcription factor NFκB. The increase in PPAR-γ in WAT may explain the observed downregulation of NFκB in conjunction with an elevation of IKK-γ mRNA levels in infected WAT. Appel et al  showed that PPAR-γ inhibits TLR-mediated activation of dendritic cells via ERK, p38 MAPK, and NFκB pathways. Thus, activation of PPAR-γ may constitute a negative feedback response to the infection-initiated inflammatory response in WAT. Conversely, in infected BAT, lipolysis is evident, and the fatty acids are likely used for thermogenesis and no PPAR-γ activation is observed.
Signaling pathways, including those related to inflammation, intersect in BAT and WAT. The differential responses observed in BAT and WAT may reflect a differential parasitic load between the 2 types of adipocytes at this stage of infection. Trypanosoma cruzi may prefer WAT because of the increased levels of free fatty acids availability by lipolysis or other nutrients essential for the rapid propagation of the parasite. These observations demonstrate that even in the early, latent stage of T. cruzi infection, adipose tissue is both a target and a sensor of parasitic infection. Recently T. cruzi has been detected in the adipose tissue of chronically infected humans .
Chagas disease lends itself well to studies of the interaction of infectious agents and adipose tissue. These results have implications for the role of the adipocytes in Chagas disease and other infectious diseases. The persistent inflammation of adipose tissue creates a systemic environment that results in myocardial disease in an analogous fashion to the processes encountered in the metabolic syndrome. It is noteworthy that adiponectin null mice display a cardiomyopathic phenotype similar to chagasic cardiomyopathy [45, 46].
This work was supported in part by the National Institutes of Health (AI-076248, HL-73732, AI-06538 to H. B. T., DK55758, CA112023, and RC1 DK086629 to P. E. S., P60-DK020541 to S. C. C. and G. J. S., 1 R03-TW006857 FIRCA to M. M. T. and H. B. T.); Conselho Nacional de Desenvolvimento Científico e Tecnológico (to F. S. M. and M. M. T.); and Fundação de Amparo à Pesquisa do Estado de Minas Gerais (to F. S. M.). This work was also supported by a pilot grant from the Diabetes Research and Training Center, Albert Einstein College of Medicine.
All authors: No reported conflicts.
All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.