Background

Chagas disease (CD), caused by the protozoan parasite Trypanosoma cruzi, represents the third largest tropical disease burden, after malaria and schistosomiasis (Organization WH 2010). Despite the significant reduction in the number of infected people that has occurred worldwide, CD still represents a major public health problem in endemic countries, mainly Latin America, and is becoming an emerging problem in non-endemic countries, such as USA, Spain, Japan, and Australia, due to population migration (Bern 2015; Schmunis 2007). Chronic Chagas cardiomyopathy (CCC) is the most severe form of clinical presentation in CD and may occur up to 20 years after infection, in approximately 20–30% of infected subjects (Rassi and Rassi 2007; WHO 2015). The hallmark of CCC is the presence of a multifocal inflammatory reaction, which leads to myocardial fibrosis, often followed by ventricular dysfunction and arrhythmias (Marin-Neto et al. 2007; Rochitte et al. 2005; Mello et al. 2012). In this stage of the disease, the conventional therapy for CD has not shown significant benefits (Morillo et al. 2015), and there is no effective treatment other than orthotopic heart transplantation.

Although the pathogenesis of CD has not been fully understood, it is well known that the persistent cardiac damage that occurs during the chronic phase is, at least partially, a result of immune responses directed to the parasites, as well as to autoreactive cells, which recognize heart antigens (Soares et al. 2001; Bonney and Engman 2015). Previous studies indicate a role for Th1 lymphocytes, with a high production of IFN-γ, which has been associated with progression to severe forms of the disease in subjects with CCC (Soares et al. 2001; Gomes et al. 2003). IFN-γ and TNF-α are overexpressed in the hearts of mice chronically infected with T. cruzi (Soares et al. 2010) and can activate macrophages, an important cell population in inflammatory sites.

Our group demonstrated, through DNA microarray analysis, that several genes related to inflammation and fibrosis are upregulated in the hearts of mice with CCC (Soares et al. 2011). Among those genes, we described the upregulation of syndecan-4 gene transcription in the heart. Syndecan-4 expression was modulated after treatment with bone marrow mononuclear cells or granulocyte-colony stimulating factor (G-CSF), two protocols that significantly reduced cardiac inflammation and fibrosis in the mouse model of T. cruzi infection (Soares et al. 2011; Vasconcelos et al. 2013), thus suggesting a role of this protein in the pathogenesis of Chagas disease.

Syndecan-4 is a transmembrane protein capable of carrying heparan sulfate and chondroitin sulfate chains, and it is expressed by different cell types, including endothelial cells, smooth muscle cells, and cardiac myocytes, playing a role in processes such as fibroblast growth factor signaling, regulation of cell migration and control of cell adhesion (Tkachenko et al. 2005; Nikkari et al. 1994; Li et al. 1997). In clinical settings, syndecan-4 concentration has been shown to be increased in subjects with heart failure, inversely correlated with left ventricular ejection fraction (Ma et al. 2013), and also after acute myocardial infarction (Kojima et al. 2001). Takahashi and colleagues demonstrated that syndecan-4 might be useful as a possible biomarker to predict cardiovascular events, such as death and hospitalization caused by worsening of heart failure (Takahashi et al. 2011).

In the present study, we aimed to evaluate the pattern of expression of syndecan-4 in heart tissue of mice and subjects with Chagas disease (CD), correlating this expression with inflammation and myocardial fibrosis.

Methods

Animal procedures

Six-to-eight weeks old female C57BL/6 mice were used for T. cruzi infection and as normal controls. All animals were raised and maintained at the animal facility of the Center for Biotechnology and Cell Therapy, Hospital São Rafael, in rooms with controlled temperature (22 ± 2 °C) and humidity (55 ± 10%), and continuous air flow. Animals were housed in a 12 h light/12 h dark cycle (6 am - 6 pm) and provided with rodent diet and water ad libitum. Animals were handled according to the NIH guidelines for animal experimentation. All procedures described had prior approval from the local institutional animal ethics committee at Hospital São Rafael (01/13).

Trypanosoma cruzi infection

Trypomastigotes of the myotropic Colombian T. cruzi strain were obtained from culture supernatants of infected LLC-MK2 cells, as previously described (Federici et al. 1964). Infection of C57BL/6 mice was performed by intraperitoneal injection of 1000 T. cruzi trypomastigotes in saline, and was confirmed through evaluation of parasitemia at different time points after infection.

Real time reverse transcription polymerase chain reaction (RT-qPCR)

Total RNA was isolated from heart samples with TRIzol reagent (Invitrogen, Molecular Probes, Oregon, USA) and concentration was determined by photometric measurement. High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Foster City, CA, USA) was used to synthesize cDNA of 1 μg RNA following manufacturer’s recommendations. RT-qPCR assay was performed to detect the expression levels of SDC4. The RT-qPCR amplification mixtures contained template cDNA, Taqman Master Mix and probe (all from Applied Biosystems). The samples were normalized with HPRT1 (endogenous control). All reactions were run in triplicate on an ABI7500 Sequence Detection System (Applied Biosystems) under standard thermal cycling conditions. The threshold cycle (2-∆∆ct) method of comparative PCR was used to analyze the results.

Human samples

The study complied with the Declaration of Helsinki, and was approved by the Ethics Committee of the São Rafael Hospital under the number 51025115.3.0000.0048. Sixty samples were obtained at Messejana Hospital in Fortaleza, Ceará, a specialized medical center for orthotopic heart transplantation in Brazil. Samples consisted of fragments of explanted hearts from 15 subjects with CD, confirmed by serological assay, 21 subjects with ICM, and 24 subjects with IdDCM. Heart samples from left ventricle and septum were included in paraffin, stained with H&E and Sirius Red, and used for immunostaining for detection of syndecan-4, as described below.

Morphometry

Heart sections were analyzed by light microscopy after paraffin embedding, followed by standard hematoxylin and eosin (H&E) staining. Inflammatory cells infiltrating heart tissue were counted using a digital morphometric evaluation system. Images were digitized using the slide scanner Scan Scope (Leica). Morphometric analyses were performed using the software Image Pro Plus v.7.0 (Media Cybernetics). The inflammatory cells were counted in 10 fields (400× magnification) per heart. The percentage of fibrosis was determined using Sirius red-stained heart sections and the Image Pro Plus v.7.0. Two blinded investigators performed the analyses.

Immunofluorescence analysis

Histological sections of paraffin embedded tissues were dewaxed and submitted to a heat-induced antigen retrieval step by incubation in citrate buffer (pH = 6.0). Then, sections were incubated overnight at 4 °C with the following primary antibodies: goat anti-syndecan-4, diluted 1:400 (Santa Cruz Biotechnology), mouse anti-α-smooth muscle actin (1:200, Dako) and rabbit anti-von Willebrand factor (1:100, Diagnostic Biosystem, Pleasanton, CA). Next, the sections were incubated for 1 h with secondary antibodies anti-goat IgG Alexa fluor 488-conjugated, anti-mouse IgG Alexa Fluor 568-conjugated (1:600, Molecular Probes) or anti-rabbit IgG Alexa Fluor 568-conjugated (1:1000, Molecular Probes). Nuclei were counterstained with DAPI (Vector Labs). The presence of fluorescent cells was determined by observation on a FluoView 1000 confocal microscope (Olympus) and A1 confocal microscope (Nikon). Quantifications were performed in 10 random fields captured under 400× magnification, using the Image Pro Plus v.7.0 software (Media Cybernetics).

Statistical analysis

Categorical data were presented as percentages. Continuous variables were tested for normal distribution using Shapiro-Wilk test and were expressed as mean ± SEM or median (interquartile interval). Comparisons of continuous variables among groups were performed with analysis of variance (ANOVA) test, followed by Bonferroni post hoc test for multiple-comparison test, or Kruskal-Wallis, depending on normality assessed by Shapiro-Wilk test. Correlation between continuous variables was evaluated by Pearson or Spearman coefficients, depending on normality. Cases with missing data were not included in the analysis. Analyses were performed using SPSS version 20.0 (IBM) or Prism 6.0 (GraphPad Software), and p < 0.05 was considered statistically significant.

Results

Analysis of inflammation, fibrosis and syndecan-4 expression in the hearts of mice during the course of infection with T. cruzi

Groups of mice infected with T. cruzi were euthanized at different time points after infection for histological and immunofluorescence analyses. H&E-stained heart sections obtained from mice during the acute phase (30 days) of infection showed the presence of intense and diffuse inflammatory infiltrates mainly composed by mononuclear cells. Parasite nests and perivascular inflammation were also frequently present (Fig. 1a and Additional file 1: Figure S1A). A marked decrease on inflammation was seen 90 days after infection, followed by a slight increase of a multifocal inflammatory infiltrate, and areas of myocytolysis thereafter (Fig. 1a and c). The presence of arterioles and veins with thickened wall due to VSMC proliferation was frequently seen during the chronic phase (Additional file 1: Figure S1B). The presence of diffuse interstitial fibrosis was observed in Sirius red-stained sections, and morphometric analysis showed a progressive increase in cardiac fibrosis with time of infection (Fig. 1b and d).

Fig. 1
figure 1

Histopathological analysis of mouse hearts at different time points after infection. a, Heart sections stained with H&E, showing the presence of inflammatory infiltrate at 30, 90 and 180 days after infection with T. cruzi. b, Heart sections stained with Sirius red, showing areas of fibrosis at 30, 90 and 180 days after infection with T. cruzi. Arrows indicate the presence of parasite nests in the acute phase (30 days of infection). c and d, Quantification of inflammation and fibrosis by morphometric analysis in the hearts of infected mice. Data represent the mean ± SEM of 5–6 mice/group. * P < 0.05; ** P < 0.01; *** P < 0.001. Calibration bars = 50 μM

Full size image

Sections of mouse hearts collected at different time points after infection and of uninfected controls were stained with anti-syndecan-4 antibodies. Syndecan-4 staining was seen in cardiomyocytes and blood vessels, both in mice euthanized during the acute and chronic phases of infection (Fig. 2a-d). A statistically significant increase in syndecan-4+ blood vessels was seen at the three time points evaluated, when compared to uninfected controls (Fig. 2e and 2f).

Fig. 2
figure 2

Expression of syndecan-4 in the heart is increased after T. cruzi infection. Heart sections of uninfected (a) and T. cruzi-infected mice, 30 (b), 90 (c) and 180 (d) days after infection were stained with anti-syndecan-4 antibody (green) and DAPI (blue) for nuclei, and observed by confocal microscopy. e, Quantification of syndecan-4+ blood vessels in heart sections. Data represent the mean ± SEM of 5–6 mice/group. f, Evaluation of cardiac Sdc4 gene expression by qRT-PCR. ** P < 0.01; *** P < 0.001 compared to other groups

Full size image

To better characterize the expression of syndecan-4, we performed confocal microscopy analysis in heart sections co-stained for α-smooth muscle actin (α-SMA) and von Willebrand factor (vWF). Co-localization between vWF, a marker for endothelial cells, and syndecan-4 was observed (Fig. 3a and b). However, co-localization of syndecan-4 and α-SMA was much more intense and frequently found, indicating that vascular smooth muscle cells (VSMCs) are the main cell population expressing syndecan-4 in this model (Fig. 3c and d). We also performed double staining for the pan-leukocyte marker CD45 and syndecan-4, but did not observe co-staining (Fig. 3e and f).

Fig. 3
figure 3

Expression of syndecan-4 in endothelial cells, vascular smooth muscle cells. and leukocytes. Heart sections of mice 180 days after infection were stained with syndecan-4 and α-smooth muscle actin - α-SMA (a and b), von Willebrand factor - vWF (c and d) or CD45 (e and f). Sections were co-stained with DAPI and observed by confocal microscopy

Full size image

Inflammation, fibrosis and syndecan-4 expression in human heart samples

Heart sections were prepared from explanted human hearts at end-stage CCC and stained with H&E for histological analysis. We observed the presence of foci of inflammatory infiltrates composed mainly of mononuclear cells near areas of myocytolysis (Fig. 4a, c and e). Alterations in the microvasculature, including perivascular inflammation, fibrosis, proliferation of microvessels and thickening of vessel walls, leading to occlusion, were also observed (Additional file 1: Figure S1C-E). Additionally, extensive areas of interstitial fibrotic scar were found in Sirius red-stained sections (Fig. 4b, d and f).

Fig. 4
figure 4

Presence of inflammation and fibrosis in explanted hearts from subjects with CCC. Heart sections of explants from three subjects with CCC were prepared. a, c and e, Staining with H&E, showing multifocal inflammatory infiltrate composed by mononuclear cells. b, d and f, Sirius red staining, showing extensive areas of interstitial fibrosis

Full size image

The expression of syndecan-4 in human heart samples was evaluated by analysis using confocal microscopy. Syndecan-4 staining was also found in myocytes and mainly in blood vessels, co-localizing with αSMA staining, indicating expression on VSMCs (Fig. 5a and b). Co-localization between syndecan-4 and vWF was found in lower extent (Fig. 5c and d). Analysis of heart samples with CCC was performed to evaluate expression of syndecan-4 expression, inflammation and fibrosis (Table 1). Syndecan-4 expression varied and did not correlate with the degree of inflammation or fibrosis.

Fig. 5
figure 5

Expression of syndecan-4 in heart sections of subjects with CCC. Sections of explanted human hearts from subjects with CCC were stained with anti-syndecan-4 antibody and α-smooth muscle actin - α-SMA (a and b) or von Willebrand factor - vWF (c and d). Sections were co-stained with DAPI and observed by confocal microscopy

Full size image
Table 1 Assessment of number of inflammatory cells, percentage of myocardial fibrosis and expression of syndecan-4
Full size table

Comparison of syndecan-4 expression, inflammation and fibrosis among subjects with end-stage CCC, ICM and idDCM

Intense inflammation and fibrosis are hallmarks of CCC. To investigate if syndecan-4 expression was associated with features of CCC, we compared the expression of syndecan-4 in human heart samples. Sections from explanted hearts were obtained from 60 individuals with end-stage cardiomyopathy. The median number of inflammatory cells/mm2 was 31.0 cells/mm2 (IQI: 16.8–109.8). There was a statistically significant difference in this variable when we compared the three groups of heart samples. The median numbers of inflammatory cells/ mm2 were 127.8 cells/mm2 (IQI: 31.0–260.3), 20.1 cells/mm2 (IQI: 12.4–41.7) and 24.4 cells/mm2 (IQI: 19.4–64.3) in subjects with CD, ICM and IdDCM, respectively (P = 0.035; Fig. 6a).

Fig. 6
figure 6

Comparison of inflammation, fibrosis and syndecan-4 expression in the hearts of subjects with end-stage cardiomyopathy. Heart samples of subjects with Chagas disease (n = 7), ischemic cardiomyopathy (n = 5) and idiopathic dilated cardiomyopathy (n = 8). a, Number of inflammatory cells/mm2 (Kruskal-Wallis one-way analysis of variance, P = 0.035). b, Percentage of myocardial fibrosis (ANOVA, P = 0.610). c, Number of syndecan-4+ blood vessels/mm2 (Kruskal-Wallis one-way analysis of variance, P = 0.466)

Full size image

The mean percentage of myocardial fibrosis was 17.4 ± 8.2%, with no statistically significant difference across the groups of heart samples. The mean percentage of myocardial fibrosis was 19.1 ± 7.7% in subjects with CD; 16.4 ± 8.6% in subjects with ICM; and 17.1 ± 8.4% in subjects with IdDCM (P = 0.610; Fig. 6b).

Regarding the expression of syndecan-4 assessment, only 20 samples were successfully stained for the immunostaining analysis (7 from subjects with Chagas disease, 5 from subjects with ischemic cardiomyopathy, and 8 from subjects with idiopathic cardiomyopathy). The overall median syndecan-4 positive vessels/mm2 was 0.54 vessels/mm2 (IQI: 0.27–1.45). No differences were detected comparing the three groups of heart samples. In subjects with CD, the median syndecan-4 positive vessels/mm2 was 0.76 vessels/mm2 (IQI: 0.52–1.66); while in subjects with ICM it was 0.46 vessels/mm2 (IQI: 0.30–1.17); and in subjects with IdDCM it was 0.43 vessels/mm2 (IQI: 0.13–1.70) (P = 0.466; Fig. 6c).

No correlation was found between the number of syndecan-4+ vessels/mm2 and either the degree of myocardial fibrosis (r = 0.261, P = 0.265; Fig. 7a) or the number of inflammatory cells/mm2 (r = 0.098, P = 0.680; Fig. 7b). When we analyzed the results per group of heart samples, we did not find any statistically significant correlation either.

Fig. 7
figure 7

Analysis of correlation between syndecan-4 expression, inflammation and fibrosis in explanted hearts from subjects with end-stage cardiomyopathy. a, Correlation between syndecan-4+ blood vessels/mm2 and number of inflammatory cells (n = 20, Spearman’s correlation). b, Correlation between syndecan-4+ blood vessels/mm2 and percentage of myocardial fibrosis (n = 20, Spearman’s correlation)

Full size image

In CD heart samples (n = 7), we found no significant correlation between the number of syndecan-4 positive vessels/mm2 and either the degree of myocardial fibrosis (r = 0.219, P = 0.637) or the number of inflammatory cells/mm2 (r = 0.314, P = 0.494). The results of these correlations were, respectively, r = 0.835 (P = 0.079) and r = 0.276 (P = 0.653) for heart samples of subjects with ICM (n = 5), and r = 0.303 (P = 0.466) and r = 0.217 (P = 0.606) for heart samples of subjects with IdDCM (n = 8).

Discussion

Heart failure is the final common pathway of a variety of primary cardiovascular diseases regardless of the nature of the cardiomyopathy. Understanding the clinical and pathological differences among its main etiologies is crucial for achieving breakthroughs in the treatment of such a life-threatening disorder. Finding new biomarkers is of special interest, as they may also serve as molecular targets for the development of therapeutic strategies. The current study provides a characterization of syndecan-4 expression in the heart tissue during acute and chronic T. cruzi infection in mice. In addition to the analysis in the mouse model, it shows, for the first time, the expression of syndecan-4 in human tissue samples of subjects who underwent orthotopic heart transplantation due to chronic Chagas disease, ischemic cardiomyopathy, and idiopathic cardiomyopathy.

Syndecan-4 is expressed in a variety of cell types, including cardiomyocytes, endothelial cells, cardiac fibroblasts and smooth muscle cells (Li et al. 1997; Samarel 2013; Julien et al. 2007; Xie et al. 2012). In our study, we found that, although present in cardiomyocytes and endothelial cells, syndecan-4 expression was highly increased in VSMCs upon T. cruzi infection, both in mouse and human hearts. Previous studies have shown that syndecan-4 expression and shedding is increased by mechanical stress in VSMCs (Li and Chaikof 2002). Syndecan-4 co-localizes with integrin heterodimers in focal adhesion complexes, which are affected by physical forces, thus causing the regulating the expression of syndecan-4 (Li and Chaikof 2002). Therefore, it is possible that the increased expression of syndecan-4 is induced by mechanical stress in the vessel walls. The role of syndecan-4 as a mechanosensor has already been demonstrated (Bellin et al. 2009). The implications of this finding to the disease pathogenesis are currently unknown and speculative, and should be further explored in future studies using adequate tools to block or enhance cardiac syndecan-4 expression.

Syndecan-4 has a variety of roles, including cell-cell adhesion, cell-ECM adhesion, binding to growth factors, thus affecting the recruitment of inflammatory cells, as well as cardiac fibroblast activation and fibrosis deposition (Samarel 2013; Li and Chaikof 2002; Gopal et al. 2017; Okina et al. 2012; Strand et al. 2013). The extracellular domain of syndecan-4 has heparan sulfate chains, which bind to several molecules that modulate processes of tissue injury and repair, including cell migration and proliferation, cell-substrate interactions, extracellular matrix remodeling, binding of cytokines and growth factors (Li et al. 2002; Maillard et al. 2014). Recently, Xi and co-workers (2016) have shown that syndecan-4 influences the migration of endothelial progenitor cells. As shown in our study and in previous reports in the literature, CCC is associated with microvasculature alterations, including the proliferation of microvessels (Prado et al. 2011). Thus, it is possible that elevated expression of syndecan-4 participates in the promotion of microvessels formation found in Chagas heart disease.

Although syndecan-4 expression has been linked to inflammatory responses, we did not find a correlation between the degree of inflammation and syndecan-4 expression. Our results showing a higher inflammation in the hearts of subjects with CCC are in accordance with data in the literature regarding the presence of a well-established inflammatory reaction in chronic Chagas disease (Marin-Neto et al. 2007). Interestingly, a significantly higher number of inflammatory cells was found in the hearts of subjects with ICM and IdDCM, in the presence of similar degrees of cardiac fibrosis.

Myocardial fibrosis has been described as prominent not only for CD, but also for ICM and IdCM (Ottaviani et al. 2015; de Leeuw et al. 2001). Explanted hearts were previously examined by de Leeuw and colleagues to determine whether specific histopathologic features were present in the myocardium of patients with end-stage idiopathic dilated cardiomyopathy. Their group revealed marked fibrosis in the hearts of 21 of 37 patients with idiopathic cardiomyopathy and in 26 of 35 patients with heart diseases of known causes (de Leeuw et al. 2001). In 2015, Ottaviani and colleagues compared the gross and histopathological parameters in hearts explanted-native or previously transplanted-from patients with end-stage heart failure with the clinical hemodynamics parameters at the time of orthotopic heart transplantation. The authors showed that both ischemic and non-ischemic cardiomyopathy patients of end-stage heart failure requiring heart transplantation presented high grades of fibrosis, hypertrophy and myocytolysis (Ottaviani et al. 2015). Our study reinforces this finding, since we did not find significant differences in fibrosis area when hearts from the three etiological groups were evaluated.

Additionally, we did not find differences in the degree of myocardial fibrosis or in the density of syndecan-4 positive vessels among the three groups of heart samples. Likewise, we showed no significant correlation between the number of syndecan-4 density and the number of inflammatory cells or the percentage of myocardial fibrosis. These data are complementary and in accordance with our recently published study, which showed lack of association between serum syndecan-4 and myocardial fibrosis measured by magnetic resonance imaging in subjects with CCC (Larocca et al. 2017).

Regarding the human material, the present study was limited by the reduced sample size for the immunostaining analysis, as a result of a surprisingly high number of samples that were not successfully stained, probably due to non-standardized tissue processing or tissue overfixation. Moreover, the inclusion of samples obtained from subjects with advanced heart failure may decrease the ability to detect differences in syndecan-4 expression among different etiologies that could possibly be present in earlier stages of the diseases.

Conclusion

In conclusion, we demonstrated the expression of syndecan-4 on VSMCs in both mice and human hearts in Chagas cardiomyopathy, without any evidence of correlation between either the percentage of myocardial fibrosis or the number of inflammatory cells and the number of syndecan-4 positive blood vessels. Further work is needed to clarify the role of syndecan-4 in the pathogenesis of CCC and its usefulness as a disease serological biomarker.