Genetic Adjuvantation of a Cell-Based Therapeutic Vaccine for Amelioration of Chagasic Cardiomyopathy

ABSTRACT

Chagas disease, caused by infection with the protozoan parasite Trypanosoma cruzi, is a leading cause of heart disease (“chagasic cardiomyopathy”) in Latin America, disproportionately affecting people in resource-poor areas. The efficacy of currently approved pharmaceutical treatments is limited mainly to acute infection, and there are no effective treatments for the chronic phase of the disease. Preclinical models of Chagas disease have demonstrated that antigen-specific CD8⁺ gamma interferon (IFN-γ)-positive T-cell responses are essential for reducing parasite burdens, increasing survival, and decreasing cardiac pathology in both the acute and chronic phases of Chagas disease. In the present study, we developed a genetically adjuvanted, dendritic cell-based immunotherapeutic for acute Chagas disease in an attempt to delay or prevent the cardiac complications that eventually result from chronic T. cruzi infection. Dendritic cells transduced with the adjuvant, an adenoviral vector encoding a dominant negative isoform of Src homology region 2 domain-containing tyrosine phosphatase 1 (SHP-1) along with the T. cruzi Tc24 antigen and trans-sialidase antigen 1 (TSA1), induced significant numbers of antigen-specific CD8⁺ IFN-γ-positive cells following injection into BALB/c mice. A vaccine platform transduced with the adenoviral vector and loaded in tandem with the recombinant protein reduced parasite burdens by 76% to >99% in comparison to a variety of different controls and significantly reduced cardiac pathology in a BALB/c mouse model of live Chagas disease. Although no statistical differences in overall survival rates among cohorts were observed, the data suggest that immunotherapeutic strategies for the treatment of acute Chagas disease are feasible and that this approach may warrant further study.

INTRODUCTION

Chagas disease, also known as American trypanosomiasis, is caused by infection with the protozoan parasite Trypanosoma cruzi (1). It is a leading cause of heart disease in Latin America, with up to 10 million infected people in the Western Hemisphere (2). The disease burden of Chagas, based on disability-adjusted life years (DALYs), is five times greater than that of malaria and approximately one-fifth that of HIV/AIDS in the Latin American/Caribbean region (3). Additionally, the annual economic toll for the treatment of infected patients exceeds $7 billion globally (4). Most of the deaths and disability attributed to Chagas disease result from chronic Chagas cardiomyopathy (CCC) (5) which develops in approximately 30% of infected individuals years to decades after the initial infection due to the cascading effects of parasite-induced pathological changes, including inflammation, cardiomyocyte hypertrophy, and fibrosis (6–8). CCC patients develop conduction disturbances associated with arrhythmias and sudden death or an end stage characterized by gross enlargement with high right or left ventricular apical aneurysm. Histologically, diffuse and patchy chronic myocarditis with mononuclear cell infiltrates and fibrosis is evident (9, 10). Two drugs, benznidazole and nifurtimox, have been used for treatment since the 1970s and have limited efficacy and significant side effects. Both drugs have up to 100% efficacy in congenital infection when administered within the first years of life and 65 to 80% efficacy in children treated during the acute phase. However, less than 35% efficacy is achieved in adults treated during the chronic phase (11, 12). A recent meta-analysis concluded that these drugs are of questionable efficacy in preventing the onset of chagasic cardiomyopathy, and almost 20% of patients fail to complete the months-long drug regimen due to significant associated toxicities (13–15). New chemotherapeutics, such as posaconazole, show promise in preclinical testing but have been of limited efficacy in human studies (12, 16). In a recent trial, trypanocidal therapy with benznidazole in patients with established Chagas cardiomyopathy significantly reduced serum parasite detection but did not significantly reduce cardiac clinical deterioration through 5 years of follow-up (17). Thus, there remains an urgent need to develop new therapies, including vaccines, to achieve sustained parasitological cure and a decreased incidence of sudden cardiac death.

Preclinical studies have revealed the essential role of antigen-specific immune responses, primarily T-cell responses, in the control of T. cruzi parasite burden and cardiac disease. Several candidate antigens, including SA85-L1, Tc52, trans-sialidase antigen 1 (TSA1), and Tc24, have shown efficacy in reducing parasitemia and endpoint cardiac pathology when used therapeutically (18–21). These reductions in parasite burden, cardiac disease, and mortality are due to T_H1-type T-cell immunity and in part to the induction of antigen-specific CD8⁺ T-cell responses (18, 22). To date, vaccines that combine Tc24 and TSA1 have provided the most dramatic reductions in parasite loads and cardiac inflammation in preclinical experimental systems. Mice vaccinated therapeutically during the acute phase with a combination Tc24-TSA1 DNA vaccine exhibited up to a 75% reduction in peak parasitemia, an approximately 60% reduction in cardiac parasite burden, and a significant reduction in endpoint cardiac pathology in comparison to unvaccinated mice (18, 20, 21). Similarly, mice vaccinated with a combination vaccine during the chronic phase demonstrated up to 80% survival at 180 days postinfection (dpi) and reduced endpoint cardiac pathology (20, 21, 23). Additionally, in a dog model of Chagas disease, the same therapeutic DNA vaccine construct reduced the number of dogs that developed cardiac arrhythmias at 50 days postinfection by 40% (24). Similar studies have shown protective efficacy using the recombinant protein counterparts of these DNA vaccines (2, 25–27).

Dendritic cells (DCs) are the most important of the professional antigen-presenting cells (APCs) that initiate and direct adaptive immune responses. Upon the detection of danger signals, DCs migrate to local lymph nodes where they induce a variety of immunogenic responses (28–32). Signals induced by the ligation of pattern recognition receptors, major histocompatibility complex (MHC) molecules, costimulatory molecules, and inflammatory cytokine receptors activate DCs, enabling them to drive immunogenic T-cell responses (33). Src homology region 2 (SH2) domain-containing tyrosine phosphatase 1 (SHP-1) is expressed in a wide variety of immune cells, where it plays a largely inhibitory role in cell signaling initiated through a range of stimuli (34). SHP-1 inhibition in DCs leads to increased proinflammatory cytokine production and the activation of Akt, enhancing DC survival. Importantly, mice vaccinated with SHP-1-inhibited DCs mount effective immune responses against both melanoma and prostate tumors, demonstrating that SHP-1 is an intrinsic negative regulator of DC activation signaling. Previous work demonstrated that a dominant negative kinase-inactivated SHP-1 adjuvant (dnSHP) delivered to DCs by means of an adenoviral vector could substantially enhance downstream T-cell responses in the context of vaccination (35).

A therapeutic vaccine is a promising alternative intervention to delay or halt the progression of chagasic cardiomyopathy due to the high toxicities and poor efficacies of current pharmaceutical interventions (13, 14). Several vaccine platforms, including plasmid DNA, viral vectors, and recombinant proteins, have been pursued in preclinical studies and have demonstrated the ability to induce antigen-specific CD8⁺ responses against viral and protozoal pathogens (36–38). In the present proof-of-principle study, we demonstrated that a candidate cell-based vaccine combining the T. cruzi-specific antigens Tc24 and TSA1 with the genetic adjuvant dnSHP exhibits therapeutic efficacy against T. cruzi infection and the ability to ameliorate cardiac pathology in a rodent model of Chagas disease.

RESULTS

Bicistronic adenoviral constructs encoding functional adjuvants and parasite-specific antigens.

Shuttle plasmids were constructed with the T. cruzi antigens Tc24 and TSA1 downstream of the genetic adjuvants caAKT (constitutively activated Akt) (39), iMC (inducible MyD88/CD40) (40), and dnSHP (35) (Fig. 1A). Ampicillin-resistant positive clones were sequenced, followed by restriction digestions with XbaI and PmeI to confirm successful cloning (Fig. 1B). The constructs were excised from the shuttle plasmid backbone and ligated into the adenoviral backbone according to the manufacturer's instructions. Ampicillin-resistant clones were selected, and confirmation of positive clone selection was performed by means of Psce-I and Iceu-I restriction digestion followed by direct sequencing (Fig. 1C). We next tested the efficacy of each adenoviral construct in generating antigen-specific immune responses. Mice were vaccinated with each of the adenoviral constructs (caAKT [constitutively activated Akt]-TC24/TSA1, iMC-Tc24/TSA1, and dnSHP-Tc24/TSA1), and antigen-specific responses were measured by gamma interferon (IFN-γ) secretion by restimulated splenocytes. Compared to caAKT and iMC, genetic adjuvantation with dnSHP gave a significantly better enhancement of the production of Tc24-specific IFN-γ responses (Fig. 2). Low-titer adenovirus was produced for the construct with dnSHP as the genetic adjuvant. Western blot analysis of HEK293T cells and DCs transduced with viral particles (vp) and probed with antihemagglutinin (anti-HA), anti-SHP-1, anti-Tc24, and anti-TSA1 proteins confirmed the expression of the genetic adjuvant and/or antigenic fusion protein (Fig. 3A to D). DCs were transduced with different titrations of the viral particles, and cell lysates were probed with anti-SHP-1 antibody to determine the optimum titer at which the antigen, adjuvant, and fusion proteins could be detected. We observed that good expression could be obtained by transducing DCs with doses as low as 100 vp (10² vp) per cell (Fig. 3E) as described previously (39).

FIG 1.

Bicistronic adenoviral constructs encoding functional adjuvants and parasite-specific antigens. Replication-deficient human recombinant adenovirus type 5 vectors were constructed with the genetic adjuvants caAKT (constitutively activated Akt), iMC (inducible MyD88/CD40), or dnSHP (dominant negative SHP-1) upstream of the T. cruzi antigens Tc24 and TSA1. (A) HA-tagged genetic adjuvants (Akt, iMC, and dnSHP) were cloned into the XbaI site of the pShuttle plasmid driven by the cytomegalovirus (CMV) promoter. The T. cruzi antigens Tc24 and TSA1 were cloned as a fusion fragment downstream of the adjuvants between the NheI and NotI sites. A single glycine hexamer linker separated the antigens, while a P2A sequence inserted between the antigens and adjuvant permitted the cleavage of the two. The expression cassette was cloned into the PmeI site of the pShuttle plasmid. (B) Restriction digestion of ampicillin-resistant positive shuttle plasmid clones with PmeI and XbaI generated the shuttle backbone fragment (4 kb), a fragment corresponding to the adjuvant (1.8 kb), and a fusion fragment corresponding to the antigens (2.5 kb). The clones were later confirmed by direct sequencing. (C) Psce-I and Iceu-I digestion of the shuttle plasmid released the construct, which was then ligated into the adenoviral backbone according to the manufacturer's instructions. Ampicillin-resistant clones were selected, and further confirmation of the positive clones was performed by means of Psce-I and Iceu-I restriction digestion followed by direct sequencing. The standard 1-kb ladder indicates the DNA fragment size with a doublet at 2.0 and 1.65 kb. The dotted line between the 1-kb ladder and shuttle plasmid digestions indicates the absence of irrelevant, unrelated digestion lanes spliced out of the image.

FIG 2.

Genetic adjuvantation with dnSHP induces significantly enhanced antigen-specific IFN-γ responses. Mice were vaccinated with each of the adenoviral constructs (caAKT-TC24/TSA1, iMC-TC24/TSA1, and dnSHP-TC24/TSA1), and antigen-specific responses were measured by the IFN-γ secretion of splenocytes stimulated with and without 10 μg/ml Tc24 protein by an ELISpot assay. Compared to caAKT and iMC, genetic adjuvantation with dnSHP induced a significant enhancement of the production of Tc24-specific IFN-γ responses. NS, not significant; *, P ≤ 0.05.

FIG 3.

Expression of the genetic adjuvant and parasite-specific antigens in DCs transduced with the adenoviral vector. Low-titer adenovirus was made with dnSHP as the genetic adjuvant. (A to D) HEK293T cells and dendritic cells were transduced with adenovirus expressing dnSHP (genetic adjuvant) and Tc24 and TSA-1 (antigens). Western blots of cell lysates were probed with anti-HA, anti-SHP, anti-Tc24, and anti-TSA1 antibodies to confirm the expression of the adjuvant, antigen, and fusion protein products. (E) Dendritic cells were transduced with different titrations of the viral particles, and cell lysates were probed with anti-SHP-1 antibody. Doses as low as 100 vp (10² vp) per cell gave good expression of the adjuvant, as confirmed by Western blotting. Protein molecular masses (in kilodaltons) are indicated on the left-hand side of each blot.

DC vaccines transduced with the dnSHP adenoviral construct and loaded with the Tc24 recombinant protein reduce T. cruzi parasite burdens and significantly improve cardiac pathology.

To test the therapeutic efficacy of the dnSHP-based DC vaccine, mice were challenged intraperitoneally (i.p.) with 500 blood-form T. cruzi trypomastigotes and treated therapeutically 7 days later by using a variety of different DC-based platforms. These platforms included unloaded DCs, DCs transduced with the dnSHP vector alone, DCs loaded with only the Tc24 recombinant protein, DCs transduced with the dnSHP vector and simultaneously loaded with the Tc24 recombinant protein to provide antigen-specific T-cell help (41–43), as well as untreated controls (Fig. 4A). Following infection and treatment, mice were monitored for 50 days and then sacrificed for analysis. Most mice lived until day 50; however, a few died prior to day 50 and were not available for all analyses. Although a Kaplan-Meier survival plot is provided in Fig. S1A in the supplemental material, there were no statistically significant differences in overall survival rates observed between groups. Analysis of serum antibody titers (Fig. 4B) demonstrated that mice treated with vector- and protein-loaded DCs exhibited significantly elevated Tc24-specific IgG1 antibody titers (P < 0.004). Mice treated by any other means exhibited IgG1 titers indistinguishable from those stimulated by infection alone. There were no statistically significant differences observed among Tc24-specific IgG2a or IgG2b antibody isotype titers, nor were there significant differences observed in total Tc24-specific serum IgG levels (not shown). Furthermore, there were no statistically significant differences in IFN-γ-secreting T cells in peripheral circulation, as evidenced by an IFN-γ enzyme-linked immunosorbent spot (ELISpot) assay (not shown).

FIG 4.

DC vaccines transduced with the dnSHP adenoviral construct and loaded with the recombinant Tc24 protein reduce the T. cruzi parasite burden. (A) To test the therapeutic efficacy of the vaccine, naive BALB/c mice were infected intraperitoneally with 500 blood-form trypomastigotes. At 7 dpi, the mice were therapeutically immunized intraperitoneally with 250,000 DCs transduced with the dnSHP vector alone, DCs loaded with the recombinant Tc24 protein only, and DCs transduced with the dnSHP vector and simultaneously loaded with the recombinant Tc24 protein. Unloaded DCs served as controls. Boost vaccination was given at 14 dpi. At 50 dpi, all remaining mice were sacrificed and analyzed. (B) Serum antibody titers analyzed for the mice at the end of the study demonstrate that mice treated with vector-plus-protein-loaded DCs exhibited significantly elevated Tc24-specific IgG1 antibody titers (P < 0.004). (C) Endpoint cardiac parasite DNA was quantified by qPCR analysis using T. cruzi-specific primers. A trend toward low/no detectable parasite DNA (P = 0.08) was observed among mice that received vector-plus-protein-loaded DCs compared to the other groups. (D and E) H&E staining was performed on cardiac tissue at the end of the study, and pathological scores on a scale of 0 to 5 were given for lymphocytic infiltration, with 5 being maximum infiltration. Randomized and blind pathological scoring demonstrated that mice which received vector-plus-protein-loaded DCs exhibited substantially lower pathological index scores, some on par with those of uninfected mice (P < 0.0006). The endpoint cardiac size (enlargement) as assessed by cross-sectional distance was nearly 20% smaller (5.04 mm versus 6.16 mm; P < 0.0001) among mice that received vector-plus-protein-loaded DCs and similar to that of uninfected mice. For all panels, error bars indicate SEM.

Despite the only modest immunological differences observed between groups on study day 50, there were significant differences observed with regard to objective cardiac pathology. Quantitative PCR (qPCR) analysis of cardiac samples using T. cruzi-specific primers indicated a clear trend toward low levels of or no detectable parasite DNA among mice that received vector- and protein-loaded DCs (76% less parasite DNA than in protein-only DCs, 80% less than in vector-only DCs, and >99% less than in either unloaded DCs or unvaccinated controls) (Fig. 4C), despite the lack of clear differences in blood-form parasite burdens between groups (Fig. S1B; the area under the parasitemia curve is shown for each animal in Fig. S1C). Hematoxylin-and-eosin (H&E)-stained cardiac tissues were scored for lymphocytic infiltration in a randomized and blind fashion. A score of 0 indicated minimum or no infiltration, and a score of 5 indicated maximum infiltration. The pathological scoring demonstrated that mice that received vector- and protein-loaded DCs exhibited substantially lower pathological index scores (Fig. 4D), some on par with those of uninfected mice (P < 0.0006). Furthermore, cardiac size (enlargement) as assessed by multiple measurements of the gross cross-sectional distance was nearly 20% smaller (5.04 mm versus 6.16 mm; P < 0.0001) among mice that received vector- and protein-loaded DCs and similar to that of uninfected mice (Fig. 4E). Representative images indicating the cardiac histopathology of each group at a magnification of ×100 are shown in Fig. 5A to E, and the presence of amastigote nests is shown in images at a ×400 magnification in Fig. 5F to J. Significant numbers of amastigote nests were observed in at least one animal of each of the five groups, with the exception of the vector- and protein-loaded DC group, in which no amastigote nests were observed in any sample from any animal. The presence of inflammatory fibrosis was also characterized by Masson's trichrome staining with blind image capture and ImageJ quantitation. As indicated in the representative images (Fig. 5K to O), extensive inflammatory damage was observed in all groups except for the vector-plus-protein group and, to a lesser extent, the unvaccinated group. Nonetheless, among infected animals, the level of trichrome staining was significantly lower only in the vector-plus-protein group (P < 0.01 by one-way analysis of variance [ANOVA]) (Fig. 6).

FIG 5.

DC vaccines transduced with the dnSHP adenoviral construct and loaded with the recombinant Tc24 protein improve cardiac pathology. (A to E) Representative endpoint cardiac H&E histopathology reveals amastigote nest formation in mice from all groups except the ones treated with DCs loaded with the vector and protein. Images are shown at a ×100 magnification. Bar = 20 μm. (F to J) Representative H&E-stained images of the amastigote nests in cardiac tissues of each group of mice. Images are shown at a ×400 magnification. Bar = 5 μm. (K to O) Representative Masson's trichrome-stained images showing the extent of cardiac inflammatory fibrosis among mice of each treated group. Images are shown at a ×100 magnification. Bar = 20 μm.

FIG 6.

Quantitation of Masson's trichrome staining. To quantify cardiac fibrosis, 5-μm sections of heart tissue were stained with Masson's trichrome stain. Images of three to five representative sections from each mouse were captured at a ×100 magnification by using a Fisher Micromaster microscope and Micron software. Images were evaluated by a reviewer who was blind to the treatment groups and analyzed by using ImageJ FIJI software to quantify the area of fibrosis and the total tissue area. The level of fibrosis in the vector-plus-protein group was significantly lower than those in all other infected groups as determined by one-way ANOVA (**, P < 0.01). Data for uninfected age-matched control mice are shown at the far right for comparison. Error bars indicate SEM.

DISCUSSION

The objective of the present proof-of-principle study was to determine the therapeutic efficacy of DC-based vaccines to prevent the progression of acute Chagas disease to chagasic cardiomyopathy. Vaccines using replication-deficient human recombinant type 5 adenoviruses have been viewed as an attractive strategy to deliver antigens for the generation of specific immune responses. In the present study, antigen-specific cellular immune responses against Tc24 were observed following vaccination with an adenoviral vector-transduced dendritic cell-based vaccine, further substantiating the use of adenoviral vectors for DNA delivery. This study made use of the in vitro transduction of dendritic cells, thereby bypassing legitimate concerns about the ability of preexisting adenovirus type 5 (Ad5)-neutralizing antibody titers to hinder or abolish in vivo transduction in previously exposed individuals. Genetic adjuvantation is an attractive strategy by which to augment the generation of immune responses (44). SHP-1 is expressed in a wide variety of immune cells, where it plays a largely inhibitory role in cell signaling (34). Previous work showed that DCs transduced with a dominant negative kinase-inactivated SHP-1 adjuvant (dnSHP) can substantially enhance antitumor T-cell responses in the context of cancer vaccination (35), suggesting that such a strategy might be beneficial in enhancing T-cell-mediated immunity against other diseases. However, natural mutation of SHP-1 is also known to result in several characteristic autoimmune phenotypes that vary in severity from enhanced dermatitis to early postnatal death (35, 45, 46). Despite this theoretical potential for the development of autoimmune inflammation, no such inflammation was observed in previous studies (35) or in the present studies, regardless of whether the genetic adjuvant was delivered to mice by in vivo plasmid transfection, in vivo adenoviral transduction, or administration of pretransduced dendritic cells.

The administration of DNA vaccines encoding a trans-sialidase antigen (TSA1) or the secreted antigen Tc24 (47) following T. cruzi infection was able to reduce parasitemia and cardiac inflammatory reactions while increasing the survival of treated mice (48). Therapeutic DNA vaccination with plasmids encoding Tc24 and TSA1 has demonstrated efficacy across different strains of mice, with no antigenic interference and/or genetic restriction of vaccine efficacy (18). As in mice, the therapeutic administration of two doses of DNA vaccines encoding TSA1 and Tc24 during the acute phase was also tested in dogs, and electrocardiogram (EKG) recordings indicated a decrease in the severity of disease-associated cardiac arrhythmia (49). In the present study, we demonstrate that the expression of the Tc24 and TSA1 antigens through dendritic cell-based vaccination generated effective antigen-specific immune responses that not only aided in parasite clearance but also mitigated cardiac manifestations of the parasite. In this model system, vaccination with MHC class I and II presentation of Tc24 and TSA1 significantly decreased the total fibrotic cardiac tissue area and overall heart width. Detailed studies of the chagasic infection process in vitro have shown that invasion of T. cruzi into cardiac cells induces hypertrophy and increased production of collagen IV and fibronectin (50). Furthermore, T. cruzi invasion activates transforming growth factor β (TGF-β), inducing intracellular signaling cascades that can drive the expression of profibrotic genes (51–55). Thus, we propose that vaccine efficacy can be attributed in part to the induction of an immune response that enhances parasite killing, decreasing total cardiomyocyte invasion by T. cruzi with a concomitant reduction of TGF-β-activated profibrotic gene expression.

Other animal models are available for the study chagasic cardiomyopathy, including canine and nonhuman primate models (49, 56–58). These models possess much if not all of the cardiac pathology seen in classic human chagasic cardiomyopathy. However, rodents are also a widespread natural reservoir for the parasite, and laboratory mice have been used extensively to study vaccine efficacy as well as the effects of T. cruzi on cardiac pathology induced by both acute and chronic Chagas disease (20, 22). Inbred rodent models also represent the most accurate, cost-effective, and reproducible systems with which to effectively model translational research with statistically reliable results.

The generation of differential IgG subclasses has been indicated in antibody responses (59, 60) as well as CD8 responses (61, 62) after T. cruzi infection. In the present study, analysis of the IgG subclasses from mice immunized with the adenoviral dendritic cell vaccine indicated that IgG1 was the dominant response in immunized BALB/c mice. In a previous study by Farrow et al. (63), a strong IgG2b response induced by the Ad5-gp83 vaccine facilitated antibody-dependent cellular cytotoxicity (ADCC) and complement-dependent cytotoxicity in addition to eliciting neutralizing antibodies. Coordinated cellular (T_H1) and humoral (T_H2) responses are important for the effective clearance of intracellular pathogens by the adaptive immune system. In conjunction with an adjuvant, protective T_H2 immunity can be achieved by the administration of antigens. In the case of intracellular infection, protective cell-based responses can be generated only by the administration of an attenuated live pathogen that induces subclinical infection. In the present study, we observed that the loading of DCs with adenovirus and recombinant protein together resulted in a significant amelioration of cardiac pathology in comparison to that of mice that received DCs loaded by any other means. While the reasons for this observation are not elucidated here, they may be related to the previously observed and reported phenomenon of homologous T-cell help (41–43).

Interestingly, some of the vaccinated groups in this acute model developed significant cardiac fibrosis. Evident fibrosis takes time to occur, and with only 7 weeks between inoculation and sacrifice, significant cardiac fibrosis was an unexpected finding. Cardiac fibrosis is not caused directly by the pathogen itself but by T_H2-biased (64–66) and T_H17-biased (67–70) immune responses mounted against T. cruzi over time. Vaccination of mice with individually loaded vector and protein vaccines appeared to accelerate the development of the T_H2- and T_H17-biased responses that are well-known mediators of fibrosis. In contrast, vaccination of mice with vector-plus-protein-loaded DCs, a methodology that we previously showed (41–43) promotes antifibrotic T_H1-biased (71–73) responses, did not generate significant fibrotic damage. Given that there was no vaccine-mediated acceleration of pathogen-specific immune responses among unvaccinated mice, the level of fibrotic damage in this group was also relatively low. Presumably, unvaccinated mice would have eventually developed levels of cardiac fibrosis similar to those of the vector-only- and protein-only-loaded groups if they were sacrificed at a significantly later time point. This result provides a cautionary tale for others attempting the amelioration of Chagas disease by immunotherapeutic means. T_H1-polarized adaptive responses may be important, not solely for control of the pathogen but also for the minimization of immune-mediated damage to vital cardiac tissue.

The mouse model of acute T. cruzi H1 infection used for these studies is well established and has been shown by several investigators to induce detectable parasitemia, cardiac parasites, and cardiac inflammation (18, 20, 25, 26). While we found that the dendritic cell vaccine construct did not result in appreciable reductions in blood parasite burdens, we observed a significant reduction in cardiac inflammation, which would translate to improved cardiac health. The parasite burden following immunosuppression would be an important parameter to evaluate if there was evidence that the treatment reduced blood parasite burdens to undetectable levels at key time points, for example, at the expected time of peak parasitemia (74, 75). However, in our model, the blood parasite burden was not reduced to undetectable levels, and thus, there was not strong evidence that sterilizing immunity had been achieved. Furthermore, the vaccine construct used in this model is consistent with DNA- and protein-based vaccines using the Tc24 and TSA1 antigens that achieve immune control of the parasites and significantly reduce parasite burdens and cardiac pathology but without achieving sterilizing immunity (20, 21, 25, 26). As 60 to 70% of chronically infected individuals naturally remain in the indeterminate stage without ever developing CCC, many hypothesize that immune control of the parasite through vaccination or other immune modulation, without the complete elimination of the parasite, remains a viable option for reducing the disease burden and improving cardiac health (2, 27). Despite the fact that sterilizing immunity was likely not achieved with the dendritic cell vaccine constructs described here, the results nonetheless demonstrate that the generation of antigen-specific immune responses against T. cruzi infection through genetic adjuvantation can successfully mitigate cardiac pathology associated with chagasic cardiomyopathy and that follow-up studies are warranted.

MATERIALS AND METHODS

Construction of an adenoviral vaccine vector.

The construction of the replication-deficient human recombinant adenovirus type 5 vector carrying the genetic adjuvants caAKT (constitutively activated Akt) (39), iMC (inducible MyD88/CD40) (40), or dnSHP (dominant negative SHP-1) (34) upstream of the T. cruzi antigens Tc24 (GenBank accession number U70035) and TSA1 (GenBank accession number M58466) was performed as described previously (76). Briefly, HA-tagged genetic adjuvants (Akt, iMC, and dnSHP), generated as described previously (35, 39, 40), were amplified by PCR using XbaI-flanked primers and cloned into the XbaI site of the pShuttle plasmid. The T. cruzi antigens Tc24 and TSA1 were cloned as a PCR-generated fusion fragment downstream of the adjuvants between the NheI and NotI sites. A single primer-encoded glycine hexamer linker separated the antigens, while a primer-encoded P2A sequence inserted between the antigens and adjuvant permitted the cleavage of the two (47). The expression cassette was digested and cloned into the PmeI site of the pShuttle plasmid. The cassette was excised by using the unique restriction sites Psce-I and Iceu-I and ligated into the Adeno-X backbone. By using the BD Adeno-X system, low-titer adenovirus was made by transfecting HEK293 cells with the pAdenoX construct expressing the T. cruzi-specific antigens according to the manufacturer's instructions (BD Biosciences, San Jose, CA).

Western blotting and analysis.

All gel electrophoreses were performed under denaturing, reducing conditions on a 12% polyacrylamide gel, with subsequent transfer to a 0.45-μm nitrocellulose membrane for antibody probing. All blocking and antibody staining steps were carried out with 5% milk, and primary antibodies were applied overnight at 4°C. The Western blot chemiluminescent signal was detected by using a ChemiDoc XRS digital imaging system supported by Image Lab software version 2.0.1 (Bio-Rad Laboratories, Hercules, CA). All Western blots were quantitated by densitometry of Ponceau S (Sigma-Aldrich, St. Louis, MO)-stained membranes. Contamination of supernatants with residual cell lysate or debris from cell death was controlled for by immunoblotting with anti-β-actin (Santa Cruz Biotechnology, Dallas, TX) and additional densitometry. Densitometry was performed by using ImageJ software (NIH, Bethesda, MD). All Western blots are representative of results from at least three independent experiments.

Vaccine production.

DCs were generated from BALB/c mice. Bone marrow leukocytes were flushed from mouse tibia and femur and cultured in AIM-V containing 10% fetal bovine serum (FBS) and 1% antibiotics and supplemented with 50 ng/ml murine granulocyte-macrophage colony-stimulating factor (mGM-CSF) and 10 ng/ml murine interleukin-4 (mIL-4) for 3 days. Cells were cultured in a humidified chamber at 37°C with 5% atmospheric CO₂. The culture medium was removed and replenished with an equal volume of fresh medium supplemented with cytokines on day 3. On day 5, the cells were replenished with fresh AIM-V containing 10% mouse serum and 1% penicillin-streptomycin-amphotericin (anti-anti) and supplemented with mGM-CSF and mIL-4 (R&D Systems, Minneapolis, MN). Immature DCs were harvested on day 6, counted, and incubated with adenovirus (expressing the Tc24 and TSA1 antigens and dnSHP as a genetic adjuvant) and recombinant Tc24 protein (25) (10 μg/ml) in AIM-V supplemented with 5% mouse serum for 3 h. The virus was added to the cells at a concentration of 1,000 viral particles per DC. Cells were plated in a six-well plate at 10⁶ cells/well in AIM-V supplemented with 5% mouse serum. After 3 h of incubation, DCs were allowed to mature for 24 h in AIM-V supplemented with 10% mouse serum, 50 ng/ml GM-CSF, 10 ng/ml IL-4, 10 ng/ml IL-1β (R&D Systems), 10 ng/ml tumor necrosis factor alpha (TNF-α) (R&D Systems), 15 ng/ml IL-6 (R&D Systems), and 1 μg/ml prostaglandin E₂ (PGE₂) (Sigma-Aldrich). The final DC product was extensively characterized as described previously (41, 77).

Mice, parasites, and infection model.

Six- to eight-week-old female BALB/c mice were obtained from Harlan Laboratories (Houston, TX). All mice were maintained in accordance with the specific IACUC requirements of the Baylor College of Medicine (animal welfare assurance number A-3823-01) and in specific accordance with IACUC-approved protocol AN-5973. T. cruzi H1 strain parasites, previously isolated from a human case in Yucatan, Mexico, were maintained by serial passage in mice. Naive mice were infected intraperitoneally with 500 blood-form trypomastigotes as previously described and validated (20). At 7 dpi, the mice were immunized intraperitoneally with 250,000 DCs loaded with adenoviral particles and the Tc24 protein. Blood was collected twice weekly for the quantification of parasitemia. At 50 dpi, all remaining mice were sacrificed and analyzed.

Vaccination, blood draws, and parasitemia.

Seven days after parasite challenge, cohorts of 4 to 5 mice each received primary vaccination with 250,000 DCs per mouse intraperitoneally. Boost vaccination was given 7 days later (41, 77). Blood was collected from the tail vein twice per week, and parasitemia was quantitated by visual microscopy and qPCR using primers specific for T. cruzi nuclear DNA (78). Serum was collected at every blood draw for downstream analysis. Heart tissue was collected on the day of sacrifice for histopathology and qPCR analyses. The experiment characterized is one of three different iterations.

IFN-γ ELISpot assays.

Tc24-specific IFN-γ-producing splenocytes were quantified by an ELISpot assay after overnight bulk restimulation with Tc24-loaded DCs using the IFN-γ ELISpot Plus kit (Mabtech, Inc., Cincinnati, OH) according to the manufacturer's instructions. Briefly, filter plates were coated overnight with 15 μg/ml coating antibody. Plates were blocked with Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% FBS at room temperature. Cells and stimuli were added to the plate in a final volume of 0.2 ml. Cells were used at final concentrations of 2.5 × 10⁵ cells per well with 5 μg/ml concanavalin A, 10 μg/ml Tc24 protein, or medium only (DMEM supplemented with 5% FBS and 1× penicillin-streptomycin). The plate was incubated for approximately 18 h at 37°C in 5% CO₂. The amount of bound IFN-γ was quantified by using 1 μg/ml biotinylated detection antibody, a 1:1,000 dilution of streptavidin-horseradish peroxidase (HRP), and the TMB substrate. After drying for a minimum of 18 h, spots were counted manually by using a dissecting microscope.

Serum antibody titers.

Serum antibodies to Tc24 were measured by an enzyme-linked immunosorbent assay (ELISA). Plates were coated with 1.25 μg/ml Tc24 protein in coating solution. Plates were blocked, and serially diluted serum samples were added. Bound antibody was detected with a 1:4,000 dilution of HRP-conjugated goat anti-mouse total IgG, IgG1, or IgG2a secondary antibody. The reaction was developed with the TMB substrate (Thermo Fisher Scientific, Waltham, MA). Titers were recorded as the last positive dilution above a cutoff optical density (OD) as determined by the OD at 450 nm (OD₄₅₀) plus 3 standard deviations (SD) for serum from naive mice.

Evaluation of parasite burdens.

Total DNA was isolated from blood or tissue by using a DNeasy blood and tissue kit (Qiagen, Valencia, CA). T. cruzi levels from 10 ng blood DNA or 50 ng heart tissue DNA were measured by quantitative real-time PCR using TaqMan Fast Advanced master mix (Life Technologies, Carlsbad, CA) and oligonucleotides specific for the satellite region of T. cruzi nuclear DNA (primers 5′-ASTCGGCTGATCGTTTTCGA-3′ and 5′-AATTCCTCCAAGCAGCGGATA-3′ and probe 5′-FAM [6-carboxyfluorescein]-CACACACTGGACACCAA-MGB-3′ [catalog numbers 4304972 and 4316032; Life Technologies]) (41, 78). Data were normalized to values for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (primers 5′-CAATGTGTCCGTCGTGGATCT-3′ and 5′-GTCCTCAGTGTAGCCCAAGATG-3′ and probe 5′-FAM-CGTGCCGCCTGGAGAAACCTGCC-MGB-3′; Life Technologies) (79), and parasite equivalents were calculated based on a standard curve as described previously (80, 81).

Evaluation of cardiac pathology.

Heart tissue was removed from euthanized animals and fixed in 10% formaldehyde for histopathological analysis. Samples were embedded in paraffin, cut into 5-μm sections, and stained with hematoxylin and eosin. For each mouse, a representative section was identified, and amastigote nests were quantified over 20 fields of view at a ×400 magnification in a blind fashion. Lymphocytic infiltration in the representative sections was scored on a scale of 0 to 5, with 0 being minimum or no infiltration and 5 being maximum infiltration (22). Heart enlargement was determined by measurement of multiple cross-sectional slices (>3 for each animal) taken from the center of each paraffin block.

Quantification of cardiac fibrosis.

Heart samples were fixed in 10% neutral buffered formalin and embedded in paraffin. To measure cardiac fibrosis, 5-μm sections were adhered to glass slides and stained with Masson's trichrome stain. Images of three to five representative sections from each mouse were captured at a ×100 magnification by using a Fisher Micromaster microscope and Micron software. Images were evaluated by a reviewer blind to the treatment groups and analyzed by using ImageJ FIJI software to quantify the area of fibrosis and the total tissue area.

Statistical analysis.

Statistical significance was determined by Student's two-tailed t test or one-way ANOVA using Prism software (GraphPad Software, La Jolla, CA). Survival was analyzed by means of Kaplan-Meier analysis. Bonferroni correction was applied when necessary to control for type I errors during multiple comparisons. Data are presented as the means ± standard errors of the means (SEM) unless stated otherwise. Statistical significance was defined as a P value of ≤0.05.