Microbes and Infection 8 (2006) 2745e2755 www.elsevier.com/locate/micinf
Original article
A DNA vaccine encoding CCL4/MIP-1b enhances myocarditis in experimental Trypanosoma cruzi infection in rats Ester Roffeˆ a, Adriano L.S. Souza a, Bra´ulia C. Caetano a, Patrı´cia P. Machado a, Lucı´ola S. Barcelos a, Remo C. Russo a, Helton C. Santiago a, Danielle G. Souza a, Vanessa Pinho a, Herbert B. Tanowitz b, Elisabeth R.S. Camargos c, Oscar Brun˜a-Romero d, Mauro M. Teixeira a,* a
Departamento de Bioquı´mica e Imunologia, Instituto de Cieˆncias Biolo´gicas, Universidade Federal de Minas Gerais, Avenida Antonio Carlos 6627, Pampulha, Belo Horizonte, Minas Gerais 31270-901, Brazil b Department of Pathology, Albert Einstein College of Medicine, Yeshiva University, Bronx, New York, USA c Departamento de Morfologia, Instituto de Cieˆncias Biolo´gicas, Universidade Federal de Minas Gerais, Avenida Antonio Carlos 6627, Pampulha, Belo Horizonte, Minas Gerais 31270-901, Brazil d Departamento de Microbiologia, Instituto de Cieˆncias Biolo´gicas, Universidade Federal de Minas Gerais, Avenida Antonio Carlos 6627, Pampulha, Belo Horizonte, Minas Gerais 31270-901, Brazil Received 26 June 2006; accepted 7 August 2006 Available online 30 August 2006
Abstract Chagas’ disease, caused by Trypanosoma cruzi, is a major cause of cardiovascular disease in Latin America. Exacerbated inflammation disproportional to parasite load characterizes chronic myocardial lesions in chagasic patients. Chemokines and their receptors are expected to account for the renewed inflammatory processes after the inoculation of the parasite, but their potential unique functions are far from being clear. Herein, we evaluated the effect of a DNA vaccine encoding CCL4/MIP-1b, a CC-chemokine, in T. cruzi-elicited myocarditis in rats. Holtzman rats were given intramuscularly cardiotoxin and the CCL4/MIP-1b DNA-containing plasmid (100 mg) was delivered in this muscular site four times. Fourteen days after last immunization, animals were inoculated with a myotropical CL-Brener T. cruzi clone. Peak of parasitism was observed at day 15 after infection, preceding the peak of myocardial inflammation at day 20. Myocarditis was still intense at day 30, but the inflammatory infiltrates showed a more focal distribution. The expression of CCL2/MCP-1 and CCL4/MIP-1b correlated closely with the kinetics of myocardial inflammation. The CCL4/MIP-1b DNA vaccine induced an increase of the levels of the anti-CCL4/MIP-1b observed in T. cruzi-infected animals. This was associated with an exacerbation of myocardial inflammation and fibrosis, although alterations in parasitemia and myocardial parasitism were not observed. Our data suggest that CCL4/MIP-1b plays a role in preventing excessive inflammation and pathology rather than in controlling parasite replication. Ó 2006 Elsevier Masson SAS. All rights reserved. Keywords: Trypanosoma cruzi; Myocarditis; Chemokine; DNA vaccine; CCL4/MIP-1b
1. Introduction Chagas’ disease is caused by the protozoan Trypanosoma cruzi and affects around 15 million people in Latin * Corresponding author. Tel./fax: þ55 31 3499 2651. E-mail address:
[email protected] (M.M. Teixeira). 1286-4579/$ - see front matter Ó 2006 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.micinf.2006.08.004
America [1]. Chronic inflammation and fibrosis characterize the chronic cardiac form of the disease [2]. The main pathological manifestation of chronic cardiomyopathy is a fibrosing and progressive disease, with structural disarrangement that leads to the loss of function of the heart and results in cardiac failure, ventricular arrhythmias and hypertrophy [3].
2746
E. Roffeˆ et al. / Microbes and Infection 8 (2006) 2745e2755
Chemokines are mediators of inflammation thought to play a role in the pathogenesis of several inflammatory and infectious diseases, including Chagas disease [4]. T. cruzi-infected cardiomyocytes produce CC and CXC-chemokines in vitro and treatment with CXCL1-2/MIP-2, CCL2/MCP-1 and CXCL10/IP-10 alone or in combination induces synthesis of nitric oxide and inhibit parasite multiplication [5]. However, at the chronic phase of infection, when parasite load is scarce, T. cruzi-infected mice are still expressing considerable amounts of CCL5/RANTES, CCL3/MIP-1a and IFNg-induced CXC-chemokines [6,7]. More recently, treatment with Met-RANTES, a selective CCR1 and CCR5 antagonist was described to improve survival, decrease heart inflammation and reduce cardiac deposition of an extracellular matrix component involved in fibrosis, without enhancement of parasitemia and myocardial parasitism in infected mice [8]. This is in contrast to in vitro data showing the importance of CC-chemokines to control T. cruzi multiplication [5,9]. Also, T. cruzi infection of CCR5 knock-out mice was strongly deleterious. Indeed, myocardial inflammation was dramatically inhibited whereas parasitism was exacerbated and mortality of CCR5 deficient infected mice was strongly increased, suggesting that CCR5 receptor could play an important protective role in control of T. cruzi infection [10]. Thus, whereas late pharmacological blockade of CCR1 and CCR5 prevented pathology and lethality, absence of CCR5 was associated with loss of the ability to mount an inflammatory response and loss of control of infection. The role of individual chemokines has not been evaluated in detail in experimental T. cruzi infection. Moreover, all studies to date evaluating the expression and role of chemokines have been carried out in mice. Infection of rats with T. cruzi represents an alternative model to study disease pathogenesis as this animal appears to clear the infection more effectively and develop a more silent disease, resembling the chronic indeterminate form observed in humans [11,12]. Chemokine-based DNA vaccines have been evaluated in experimental models of tissue-destructive autoimmune processes. The rationale underlying such studies is that immunization with a chemokine gene in an immunogenic plasmid facilitates the induction of anti-chemokine antibodies which have the ability to block the action of the chemokine of interest. For example, DNA vaccines encoding chemokine genes were shown to partially or totally inhibit pathology, or reverse established disease in models of experimental allergic encephalomyelitis and chronic poly-adjuvant-induced arthritis [13,14]. In the present study, we have evaluated the expression of cytokines, chemokines, tissue pathology and parasitism in rats infected with the CL-Brener clone of T. cruzi. Our studies showed a marked expression of several chemokines, including CCL4/MIP-1b. CCL4/MIP-1b is a CCR5-acting chemokine which exerts chemotactic and migratory effect in monocytes, T lymphocytes, dendritic cells and NK cells [15]. There are no reported studies evaluating the role of CCL4/MIP-1b in experimental T. cruzi infection. Here, we investigated whether administration of a DNA vaccine encoding CCL4/MIP-1b had any effect in T. cruzi-elicited myocarditis in rats.
2. Materials and methods 2.1. Animals and infection Holtzman rats, 90 days old and 300e400 g of body weight, were obtained from CEBIO/UFMG (Minas Gerais, Brazil) and maintained in the animal facilities of the Laborato´rio de Imunofarmacologia, with filtered water and food ad libitum. Animals were inoculated intraperitoneally with 104 blood forms of the CL-Brener clone of T. cruzi/50 g body weight. All animal procedures had prior approval from the local animal ethics committee (CETEA/UFMG). The myocardium was obtained in different days corresponding to the acute (15, 20 and 30) and chronic (65 and 130) phases of infection. Parasitemia was estimated according to Brener’s method every other day in 5 mL of peripheral blood sampled from the tail vein of T. cruzi-infected mice [16]. 2.2. Cytokines and chemokine detection by ELISA The upper part of the heart was processed in a solution containing protease inhibitors and centrifuged at 10,000 rpm for 10 min in 4 C, as previously described [17]. The supernatant was stocked at 20 C until used to detect IFN-g, IL-4 (Duoset, R&D Systems), TNF-a, IL-10 (gently donated by Dr. Stephen Poole, England) and the chemokines CCL2/ MCP-1 and CCL5/RANTES (Peprotech, USA) by ELISA. 2.3. Histopathological analysis The middle part of the heart was washed in sterile PBS, smoothly dried and immediately fixed in 4% buffered paraformaldehyde. The tissue was embedded in paraffin, cut in 7 mm sections and stained with hematoxylin and eosin, to quantify inflammation and infection, and with Picro-syrius, to evaluate fibrosis. Cardiac parasitism and inflammation of hearts from 5e7 animals per group were analyzed with a Zeiss integrating ¨ berkohen, Germany) at a final mageyepiece with 100 hits (O nification of 320. A total of 4000 hits were evaluated in each section of cardiac tissue. The infection and inflammation indices represent the number of hits covered by amastigote nests and inflammatory cells, respectively. 2.4. RT-PCR The RNA was extracted by using RNAzol (Sigma, USA). The RNA obtained was resuspended in diethylpyrocarbonatetreated water and stocked at 70 C until use. The RNA was quantified and 1 mg was utilized in the reverse transcription reaction. The resulting cDNA was used for amplification of the desired gene with the following primer sequences and specific conditions (annealing temperature and cycles): CCL2/MCP-1 sense 50 -ATG-CAG-GTC-TCT-GTC-ACGCTT-CTG-GGC-30 , anti-sense 50 -CTA-GTT-CTC-TGT-CATACT-GGT-CAC-30 (55 C, 35 cycles); CCL4/MIP-1b sense 50 -ATG-AAG-CTC-TGC-GTG-TCT-GCC-TTC-30 , anti-sense 50 -TCA-GTT-CAA-CTC-CAA-GTC-ATT-CAC-30 (50 C, 33
E. Roffeˆ et al. / Microbes and Infection 8 (2006) 2745e2755
cycles); CCL5/RANTES sense 50 -ATG-AAG-ATC-TCTGCA-GCT-GCA-TCC-30 , anti-sense 50 -CTA-GCT-CAT-CTCCAA-ATA-GTT-G-30 , (50 C, 26 cycles); GAPDH sense 50 -CTC-AAG-ATT-GTC-AGC-AAT-GC-30 ; anti-sense 50 -CAGGAT-GCC-CTT-TAG-TGG-GC-30 (60 C, 26 cycles). 2.5. Quantification of parasite tissue loads by real-time PCR Real-time PCR for parasite quantification was performed as described previously [18], with minor modifications. Briefly, on different days after infection, the hearts were digested with proteinase K, followed by a phenol-chloroform-isoamyl alcohol affinity extraction. Real-time PCR using 50 ng of total DNA was performed on an ABI PRISM 7900 sequence detection system (Applied Biosystems) using SYBR Green PCR Master Mix according to the manufacturer’s recommendations. Purified T. cruzi DNA (American Type Culture Collection) was sequentially diluted for curve generation in aqueous solution containing equivalent amounts of DNA from uninfected mouse tissues. The following primers were used: T. cruzi-specific-primers (TCZ gene e a tandemly repeated genomic sequence, known as ‘‘satellite’’ DNA) GCTCTTGCC CACAMGGGTGC, where M ¼ A or C (S35-forward), and CCAAGCAGCGGATAGTTCAGG (S36-reverse). 2.6. Hydroxyproline quantification Fragments of 100 mg of myocardium were removed for hydroxyproline determination as an indirect measure of collagen content and as previously described [19]. This method uses Chloramine T as an oxidizing agent and compares values of hydroxyproline with a standard curve. 2.7. Rat CCL4/MIP-1b cloning and construction of the vaccination plasmid The cloning of the CCL4/MIP-1b gene (279 bp) was accomplished from the product resulting from the CCL4/MIP1b specific amplification by RT-PCR reaction in myocardial samples from T. cruzi-infected rats 20 days after infection. The DNA was extracted from agarosis gel by a DNA purification kit from agarosis (Promega, USA). The cloning was performed with p-TopoTA cloning kit (Invitrogen, San Diego, CA), as recommended by the supplier. The p-Topo recombinant plasmid was used to transform chemically competent E. coli XL1-blue bacteria. To confirm the CCL4/MIP-1b cloning the plasmid was purified using a miniprep kit (Qiagen, USA), digested with EcoR1 (Promega, USA) and sequenced (MEGABACEÔ). The confirmed CCL4/MIP-1b gene was inserted into the vaccination pcDNA 3.1 plasmid (Invitrogen, USA) with T4 DNA ligase enzyme and transfected into E. coli XL1-blue bacteria. The vaccination construct was in purified (midiprep kit) and sequenced. The correct recombinant colonies were grown in 2,5 L of LB-ampicilin medium and plasmid purification was performed by Endofree gigaprep kit.
2747
The DNA resulting was measured at 260 and 280 nm in spectrophotometer and stocked at 70 until use. 2.8. Immunization schedule Cardiotoxin (10 mM, Sigma) was injected intramuscularly and 7 days later animals received an intramuscular injection of 100 mg of DNA in 300 ml of apyrogenic ultra-pure water, divided in two injections of 150 ml of DNA into each tibial muscle. Animals received 4 boosters every 7 days. To investigate whether DNA immunization causes any effect per se, uninfected animals were immunized with CCL4/MIP-1b. As there were no differences in all myocardial parameters studied in comparison to non-vaccinated uninfected rats (data not shown), data in uninfected rats were pooled for facilitation of presentation. 2.9. Detection of serum anti-chemokine antibodies Commercial recombinant rat CCL4/MIP-1b (Peprotech) was coated onto ELISA plates at 10 ng/well. Serial dilutions of rat sera in PBS-BSA 1% were added. Anti-rat IgG streptoavidin-conjugated antibody (Sigma) was used at 1:1000. Absorbance was measured at 492 nm in spectrophotometer. 2.10. In vivo readout assay for screening the neutralizing activity of the rat anti-CCL4/MIP-1b antibodies To address the in vivo neutralizing action of antibodies elicited by CCL4/MIP-1b-encoding DNA vaccination, serum was obtained at 20 days after infection of CCL4/MIP-1b- or control plasmid-vaccinated animals (n ¼ 6). The sera were pooled, diluted at 1:20 in PBS and given subcutaneously to rats. After 1 h, an optimal dose of CCL4/MIP-1b (300 ng/cavity) or sterile PBS was injected intraperitoneally and animals killed after 18 h. Cells were harvested with 15 ml PBS, total cell counts performed in a modified Neubauer chamber using Turk’s stain and differential cell counts on cytospin preparations (Shandon III) stained with May-Grumwald-Giemsa using standard morphologic criteria to identify cell types. There was no effect of the sera on the baseline recruitment of PBS (data not shown). 2.11. Statistical analysis Results are shown as means S.E.M. Differences between groups were compared using Student’s t test (two sets of data) or one-way ANOVA (three or more sets of data), followed by the Student-NewmaneKeuls post hoc test. Differences were considered significant at p < 0.05. 3. Results 3.1. Parasitemia and myocardium inflammation in rats Infected rats had low parasitemia. Parasites were first detected in blood at day 6 and levels peaked at day 12 after
E. Roffeˆ et al. / Microbes and Infection 8 (2006) 2745e2755
2748
infection (around 2 104 parasites/mL of blood). The parasitemia subsided rapidly and parasites could not be detected in blood at day 19 after infection or later (data not shown). There was not lethality in the present group of experiments. Myocardial alterations after T. cruzi infection are shown in Fig. 1. Intense myocarditis was already observed at day 15 and peaked at day 20 after infection (Fig. 1A). At this stage, inflammatory infiltrates were composed predominantly by mononuclear cells with a disperse distribution in the
tissue. Amastigote pseudocysts, frequently accompanied by inflammatory infiltrates, were also observed in the myocardium at days 15 and 20 after infection (Fig. 1C). The peak of parasitism was observed at day 15, thus preceding the peak of myocardial inflammation (Figs. 1A,B). At 20 days after infection, there were increased numbers of infiltrates that were more intense and showed a more disperse distribution in the cardiac tissue (Fig. 1C). The myocarditis was still intense at day 30 after infection, but the inflammatory
A
fg parasite DNA/ 50ng host DNA
T. cruzi DNA in myocardium
50
% points with amastigote nests/ microscopic field (320x)
% points with inflammatory cells/ microscopic field (320x)
B ** *
40 30
** *
20 10 0
15
20
30
65
Days after infection
0.4 0.3
250 200 150 100 50
1.9x10-7 ND
0
15
20
30
65
Days after infection
0.2 0.1 0.0
ND 15
20
30
65
Days after infection
C
D
E
F
Fig. 1. Histopathological alterations in myocardium of T. cruzi-infected animals. Quantification of (A) inflammation; and (B) infection correspond to the quantification of points containing inflammatory cell nuclei or amastigote pseudocysts, respectively. Quantification was performed by using an ocular containing 100 points/microscopic field in a final magnification of 200. A total number of 40 microscopic fields were analyzed by section, in a total of 4000 points. The insert shows the quantification of T. cruzi DNA by Real time-PCR. Results are the means S.E.M. of 5e7 animals in each group. The myocardium from infected animals was excised at days 20 (C); 30 (D); and 65 (E) after infection. The heart of a non-infected animal is shown in (F). ND, not determined. *, ** and *** for P < 0.05, 0.01 and 0.001, respectively, when comparing infected versus non-infected animals.
E. Roffeˆ et al. / Microbes and Infection 8 (2006) 2745e2755
2749
expression of CCL5/RANTES mRNA was slightly increased throughout the observation period after infection (Figs. 2A,C). The expression of CCL2/MCP-1 and CCL5/RANTES protein in the myocardium of T. cruzi-infected animals is shown in Fig. 3. Overall, there was a good correlation between the expression of mRNA, as detected by RT-PCR, and protein, as detected by ELISA, for the chemokine CCL2/MCP-1. The trend was also similar for CCL5/RANTES in both assays. Thus, CCL2/MCP-1 expression was noticed from days 12 to 30 whereas CCL5/RANTES was not detected above levels found in non-infected rats (Figs. 3A,B). A commercial kit for CCL4/MIP-1b was not available at the time these experiments were conducted. The expression of several cytokines thought to play a role in the pathophysiology of T. cruzi infection was also evaluated. IFN-g, TNF-a, IL-4 and IL-10 are increased during the
infiltrates showed a more focal distribution (Fig. 1D). Parasite nests were uncommon at day 30, but few amastigotes could be observed inside the infiltrates. By day 65, there was a significant reduction in the myocarditis and parasites were not found (Fig. 1E). 3.2. Expression of chemokines and cytokines in the myocardium of T. cruzi-infected rats The RT-PCR results are shown in Fig. 2A and the mean of the densitometric analysis on Figs. 2BeD. CCL2/MCP-1 mRNA was already expressed at day 12 after infection and peaked at day 20 (Fig. 2). Thereafter, expression returned to background levels. The expression of CCL4/MIP-1b mRNA followed a similar pattern but the chemokine was only detected at days 15 and 20 after infection (Fig. 2C). Myocardial
Days after infection
A N
15
12
20
30
65
CCL2
CCL4
ND
CCL5
GAPDH
B
C
CCL2 (MCP-1)
Arbitrary units
Arbitrary units
75 50 25 0
CCL4 (MIP-1 ) 100
100
12
15
20
30
75 50 25 0
65
12
Days after infection
D
15
20
30
65
Days after infection CCL5 (RANTES)
Arbitrary units
100 75 50 25 0
12
15
20
30
65
Days after infection Fig. 2. Kinetics of mRNA expression of CCL4/MIP-b in cardiac tissue of T. cruzi-infected animals. A) RT-PCR bands; BeD) densitometry of RT-PCR bands. The organs were excised and submitted to RNA extraction by Phenol-chloroformium technique. The resulting RNA was quantified by spectrophotometry and used to reverse transcription. The cDNA was amplified by PCR reaction using specific primer to the chemokine CCL4/MIP-b. The dotted lines across the bars represent background values in non-infected animals. The result was expressed in arbitrary units according to the constitutively expressed GAPDH gene. Results are the means of 3 animals in each group. ND; Not done.
E. Roffeˆ et al. / Microbes and Infection 8 (2006) 2745e2755
2750
A
B
CCL2/MCP-1 ***
g cytokine/100mg heart tissue
g cytokine/100mg heart tissue
*** 4000
***
2000
0
CCL5/RANTES 3000
6000
12
15
20
30
12
Days after infection
C *
g cytokine/100mg heart tissue
g cytokine/100mg heart tissue
D
750 500 250 0
12
15
20
30
65
** 50
25
12
15
20
30
65
Days after infection
IL-4
IL-10
F 750
***
g cytokine/100mg heart tissue
g cytokine/100mg heart tissue
30
TNF-
0
65
1800
1200
600
0
20
75
Days after infection
E
15
***
Days after infection
IFN1000
***
1000
0
65
*** 2000
12
15
20
ND 30
ND 65
Days after infection
500
250
0
12
15
20
30
65
Days after infection
Fig. 3. Kinetics of chemokine protein production in cardiac tissue of T. cruzi-infected animals. The organs were excised and processed for the measurement of (A) CCL2/MCP-1; (B) CCL5/RANTES; (C) IFN-g; (D) TNF-a; (E) IL-4; and (F) IL-10 by ELISA. The dotted lines across the bars represent background values in non-infected animals. Results are the means S.E.M. of 5e7 animals in each group. *, ** and *** for P < 0.05, 0.01 and 0.001, respectively, when comparing infected versus non-infected animals.
acute phase (Figs. 3CeF). The peak of IFN-g production was earlier than that of IL-4 (15 vs 20 days after infection) and was not detected 30 days after infection (Figs. 3C,E). Moreover, during the chronic phase, levels of these cytokines were near those found in non-infected rats (Fig. 3). 3.3. Effects of the vaccination with a DNA vaccine encoding for CCL4/MIP-1b in the course of T. cruzi infection in rats The functional role of CCL4/MIP-1b was investigated by using a CCL4/MIP-1b-encoding DNA vaccine. Anti-chemokine antibodies are commonly elevated in autoimmune diseases, such as experimental autoimmune encephalomyelitis and arthritis, and vaccination with DNA coding chemokines by the schedule used here has been known to induce an increase in levels of anti-chemokine antibodies [13,14]. In comparison to non-infected rats, higher levels of anti-CCL4/MIP-1b
antibodies were detected in infected animals at 30 days but not at 20 days after infection. Moreover, levels of antibodies were enhanced in T. cruzi-infected animals that were vaccinated (Fig. 4A). We also addressed whether the antibodies elicited by the CCL4/MIP-1b vaccination could indeed prevent the function of CCL4/MIP-1b. Pre-treatment with sera from CCL4/MIP-1b-immunized animals (diluted to 1:20) was effective at blocking the migration of leukocytes to the peritoneal cavity after injection of CCL4/MIP-1b. In contrast, sera from animals immunized with control plasmid failed to affect the recruitment induced by CCL4/MIP-1b (Fig. 4B). The immunization with empty or CCL4/MIP-1b plasmids did not cause any significant effect on parasitemia neither in infection-associated lethality (data not shown). There was an exacerbation of the inflammatory response in the heart of CCL4/MIP-1b immunized rats at 20 and 30 days after infection, as compared to non-immunized and plasmid control immunized animals (Figs. 5A,C vs B,D and E). There did not
E. Roffeˆ et al. / Microbes and Infection 8 (2006) 2745e2755
A
Anti-CCL4/MIP-1 antibodies in serum 1.25
Absorbance (492nm)
## ##
1.00
**
0.75 0.50 0.25 0.00
N
20
30
Days after infection Total leukocytes in peritoneum
B Number of cells x 104 per cavity
2400
and infected animals (Fig. 6B), there was staining of collagen fibers dispersed in the cardiac tissue, especially around inflammatory infiltrates (Figs. 6A,B) and blood vessels (data not shown). Blood vessels were stained in hearts of non-infected animals also, but no staining of fibers in this tissue was observed (data not shown). In CCL4/MIP-1b-vaccinated animals, the same profile of staining was observed, but the intensity and area of the staining was greater (Fig. 6A) and there was an increased number of inflammatory infiltrates which were stained (data not shown). Quantification of collagen content by hydroxyproline measurement concurred with the staining analysis (Figs. 6C,D). Immunization with CCL4/ MIP-1b had no significant effect on the infection-induced increase in chemokine and cytokine levels in myocardium at day 20 and day 30 after infection (data not shown).
Mononuclear Neutrophils
1800
*
1200
PBS
600
0
2751
empty plasmid serum
CCL4/MIP-1ß serum
CCL4/MIP-1ß (300 ng/cavity) Fig. 4. Quantification and in vivo effects of anti-CCL4/MIP-b antibodies in serum of T. cruzi-infected animals previously immunized with CCL4/MIPb (black bar) or control plasmid (grey bar), and non-infected (open bar) animals. In (A), the serum was diluted to 1:20 in PBS-BSA 1% and used in direct ELISA for determination of levels. Results are the means S.E.M. of 5e7 animals in each group. ** and ## for P < 0.01 when comparing infected versus non-infected animals, and CCL4/MIP-b- versus control plasmid-immunized animals, respectively. (B), Blocking activity of antibodies elicited by CCL4/ MIP-1b-encoding DNA vaccination. Animals were pre-treated with sera from T. cruzi-infected animals that were previously immunized with CCL4/ MIP-1b- or control plasmid. The sera was pooled from 6 animals and diluted to 1:20 in PBS and given subcutaneously 1 h before the injection of CCL4/ MIP-1b (300 ng/cavity). Results are shown as number of cells 104 per cavity. Mononuclear cells (black bar) and neutrophils (open bar) are shown. The dotted lines across the bars represent background values in animals given PBS. Results are the means S.E.M. of 5 animals in each group. * and ** for P < 0.5 and 0.01 when comparing CCL4/MIP-1b- versus control plasmid-immunized animals, respectively.
appear to be any major change in the composition of inflammatory lesions in CCL4/MIP-1b- and vector controlimmunized animals (data not shown). However, the inflammatory infiltrates in the heart of CCL4/MIP-1b-immunized animals were multi-focal (Figs. 5A,C) in comparison with a more focal response that was observed in plasmid control-immunized animals (Figs. 5B,D). Myocardial parasitism was scanty and not different in CCL4/MIP-1b and plasmid control-immunized rats (Fig. 5F). An increase in collagen content, an index of tissue fibrosis, was evidenced in heart sections of CCL4/MIP-1b-immunized animals at days 20 (data not shown) and 30 after infection (Fig. 6A), as shown by picro-syrius staining. In non-immunized (data not shown) and in plasmid control-immunized
4. Discussion Expression of inducible chemokines is thought to be a vertebrate cellular ‘‘SOS response’’ to recruit leukocytes to an area of injury [20]. Although certain chemokines expressed in T. cruzi-infected hearts may be important for parasitic containment, their expression may extend injury through several mechanisms, including the maintenance of chronic inflammation [21]. So, chemokines are suggested to play an ambiguous role in T. cruzi infection: in the acute phase, these mediators may be important to facilitate movement of cells and in the acquisition of a parasite-specific immune response that will control parasite replication; in contrast, in the chronic phase, parasite-induced IFN-g and chemokines may play a role in the maintenance of inflammation. In such way, it is crucial to identify which chemokines are essential for parasite control in the acute phase of infection and which chemokines may be deleteriously produced during the chronic phase. Such knowledge may aid in the development of novel pharmacological interventions to control unwanted chronic inflammation. Inoculation of rats with T. cruzi represents an experimental model of resistance to infection [22]. Indeed, we found very low parasitemia, and other studies suggest that parasitemia intensity correlates directly to resistance in different inbred strains of rats [23]. Despite the low levels of myocardial parasitism, there is significant acute myocardial pathology. The cardiac disease does not progress to heart dilatation and failure, but infected rats manifest electrocardiographic alterations in the chronic phase, probably associated with the focal myocarditis and myocardial fibrosis [24]. Thus, despite continued infection and a degree of heart damage, infected rats manage to regulate the immune and inflammatory response and prevent severe disease after infection with certain strains of T. cruzi. On the other hand, T. cruzi-elicited myocarditis in rats share important features with the more commonly studied murine model, including the predominance of mononuclear cells in inflammatory lesions and the predominance of TNF-a and type 1 cytokines in the heart (IFN-g, CCL5/RANTES). Thus, it appears that T. cruzi infection in rats is a valid model to study the immune response involved in myocarditis caused by this protozoan parasite.
E. Roffeˆ et al. / Microbes and Infection 8 (2006) 2745e2755
2752
A
B
C
D
E
F 0.3 # #
30
*
*
20 10 0
20
30
Days after infection
Infection Index
Inflammation Index
40
0.2
0.1
0.0
20
30
Days after infection
Fig. 5. Histopathological alterations in myocardium of T. cruzi-infected animals previously immunized with CCL4/MIP-b (solid bar and symbol) or control plasmid (open bar and symbol). The dotted lines across the bars represent background values in non-infected animals. The myocardium was excised and evaluated 20 (A, B, E, F) or 30 days (C, D, E, F) after infection. Animals were immunized previously with CCL4/MIP-1b-encoding DNA vaccine (A, C) or control plasmid pcDNA3 (B, D) prior to infection. The organs were included in paraffin and sectioned in microtome to posterior hematoxylin and eosin staining. The inflammation (E) and infection (F) indices were quantified using an ocular containing 100 points/microscopic field in a final magnification of 400. A total number of 40 microscopic fields were analyzed by section, in a total of 4000 points. ND, non-determined. Results are the means S.E.M. of 5e7 animals in each group. * and # for P < 0.05 when comparing infected versus non-infected animals, and CCL4/MIP-1b- versus control plasmid-immunized animals, respectively.
Holtzman rats infected with the CL-Brener clone of T. cruzi had low parasitemia. The peak of myocardial parasitism preceded the peak of myocardial inflammation (day 15 vs 20, respectively). Myocarditis was still intense at day 30 after infection, but the inflammatory infiltrates showed a more focal distribution. Parasite nests were uncommon at day 30, but few amastigotes could be observed inside the infiltrates. By day 65, there was a significant reduction in the myocarditis and parasites were not found. The expression of several chemokines and cytokines correlated closely with the kinetics of
myocardial inflammation and both inflammation and cytokine/chemokine levels subsided during the chronic phase of infection. In T. cruzi-infected C57BL/6 mice, it was observed that cytokine expression was dominated by the presence of IFN-g mRNA in the acute phase of infection, whereas the balance of type 1 and type 2 cytokines was switched in favor of IL4 and IL-10 mRNAs in chronic phase [6]. In C3H/He mice, a susceptible mouse strain that present chronic myocarditis, increased expression of TNF-a and IFN-g mRNA was observed
E. Roffeˆ et al. / Microbes and Infection 8 (2006) 2745e2755
B
A
1.3
# 1.2
1.1
1.0
OH-proline (fold increase over T. cruzi-infected animals)
D
C OH-proline (fold increase over T. cruzi-infected animals)
2753
1.5
#
1.4 1.3 1.2 1.1 1.0
Fig. 6. Quantification of collagen content in myocardium of T. cruzi-infected animals previously immunized with CCL4/MIP-b (solid bar) or control plasmid (open bar), and non-infected animals. The myocardium was excised and evaluated 20 after infection. Animals were previously immunized with CCL4/MIP-b-encoding DNA vaccine (A); or control plasmid pcDNA3 (B) prior to infection. The myocardium was included in paraffin, sectioned and stained with Picrus-syrius. Final magnification of 400. Quantification of hydroxyproline was carried out 20 days (C); and 30 days (D) after infection. Results are the means S.E.M. of 5e7 animals in each group. # for P < 0.05 when CCL4/MIP-b- versus control plasmid-immunized animals.
in the cardiac tissue in acute and chronic infection. Interestingly, Talvani et al. noticed an increased expression of IL-4 mRNA only during the chronic infection, when focal inflammatory infiltrates persisted [7]. In the present work we found an increased production of TNF-a, IFN-g and IL-4 only in the acute phase of infection. The latter findings are consistent with the ability of rats to resolve the chronic infection and infection-associated inflammation. CCL4/MIP-1b is a CCR5-acting chemokine which exerts chemotactic and migratory effect in monocytes, T lymphocytes, dendritic cells and NK cells [15]. There are no reported studies evaluating the role of CCL4/MIP-1b in experimental T. cruzi infection. The presented results show that administration of a CCL4/MIP-1b encoding DNA vaccine to T. cruziinfected animals did not affect the parasitism, but resulted in exacerbation of the inflammatory process and collagen deposition in myocardium, alterations more evident 30 days postinfection, at the end of acute phase. Consistent with our studies, in an EAE model in which a CCL4/MIP-1b encoding DNA vaccine was used, there was increased mononuclear cell infiltrates in the central nervous system and significant aggravation of clinical disease; as opposed to CCL2/MCP-1 and CCL3/MIP-1a DNA vaccines that induced protection [13]. Evidences show that CCL4/MIP-1b can recruit regulatory CD4þ CD25þ regulatory T cells (Tregs) [25]. This cell population is characterized by production of high levels of IL-10
and may be involved in the regulation of immune responses, as demonstrated in diverse inflammatory models, as autoimmune [26], transplants [27] and allergies [28]. One study has suggested that Tregs are expressed preferentially in patients who do not develop severe Chagas disease [29], but no studies have formally demonstrated a role for these cells in experimental models of T. cruzi infection. Thus, an interesting possibility raised from our studies is that CCL4/MIP-1b may be relevant for the recruitment of Tregs to the cardiac tissue of infected rats. Blockade of CCL4/MIP-1b with vaccine may prevent migration of these cells and, hence, worsen infection. Further studies evaluating the role of Tregs and the role of CCL4/MIP-1b for Tregs recruitment are necessary to resolve these issues. CCL4/MIP-1b binds only to CCR5. However, CCR5 is also a receptor for the chemokines CCL3/MIP-1a and CCL5/ RANTES [30]. Moreover, CCL3/MIP-1a and CCL5/RANTES bind to additional receptors, including CCR1 [31]. In contrast to our findings, T. cruzi infection was much more severe in CCR5-deficient animals when compared to their wild-type controls [10]. It is unclear whether the apparent discrepant role of CCL4/MIP-1b (in rats) versus CCR5 (in mice) blockade represents a differential usage of chemokines and their receptors in different species. Alternatively, in the absence of CCL4/MIP-1b, other chemokines, such as CCL3/MIP-1a and CCL5/RANTES, would still be available for activating
2754
E. Roffeˆ et al. / Microbes and Infection 8 (2006) 2745e2755
the CCR5 receptor, which appears to be essential for leukocyte influx during T. cruzi infection [10]. Preliminary data from our laboratory using DNA vaccination with plasmids containing CCL5/RANTES and CCL3/MIP-1a demonstrate that rats indeed have greater heart parasitism after infection (our unpublished observations). It is of note that the latter chemokines can also activate other receptors, including CCR1. So, absence of CCR5 and unopposed activation of CCR1 could be relevant to infection outcome. Indeed, treatment with Met-RANTES, a CCR1 and CCR5 blocker, resulted in reduced myocarditis without an increase in parasitism and parasitemia [8]. In CCR5/ mice, T cell-mediated hepatitis was exacerbated and accompanied by an increase in infiltration of CCR1þ cells in the liver [32]. Thus, CCR5 activation appears essential for T. cruzi-associated inflammatory and regulatory cell influx. Whereas the influx of regulatory cells appears to be mediated mainly by CCL4/MIP-1b, co-activation of CCR1 and CCR5 by CCL3/MIP-1a and CCL5/RANTES are relevant for inflammatory cell influx, control of parasite replication, and associated with chronic fibrosis. Altogether, these results suggest that an imbalance between the chemokines and chemokine receptors expressed in the cardiac tissue may be a preponderant factor in the perpetuation of the inflammatory process. In conclusion, we have shown here that CCL4/MIP-1b is expressed during the acute phase of T. cruzi infection of rats. The neutralization of CCL4/MIP-1b using a DNA vaccine encoding for CCL4/MIP-1b was deleterious to the host, evidenced by the increase in myocardial inflammation and higher collagen deposition. In contrast, CCL4/MIP-1b appeared not to be relevant for the ability of the host to deal with the acute infection. Our results suggest that CCL4/MIP-1b plays a role in preventing excessive inflammation and pathology rather than in controlling parasite load. Acknowledgements We thank Mariana Sim~ oes and Dr. Guilherme Correˆa Oliveira (CPqRR/Fiocruz, Belo Horizonte, Brazil) for sequencing the constructs, and Valdine´ria Borges, Luı´za Silva and Carlos Henrique Silva, for technical support. This work received finantial support from Fundac¸~ao de Amparo a` Pesquisa do Estado de Minas Gerais (FAPEMIG, Brazil), Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico (CNPq, Brazil), Coordenac¸~ao de Aperfeic¸oamento de Pessoal de Nı´vel Superior (CAPES, Brazil) and Fogarty International Research Collaboration Award (NIH/FIRCA, 1 R03-TW006857-01A1, USA). References [1] World Health Organization, Control of Chagas’ disease, Tech. Rep. Ser. 905 (2002) 1e109. [2] F. Ko¨berle, Chagas’ disease and Chagas’ syndrome: the pathology of American Trypanosomiasis, Adv. Parasitol. 6 (1968) 63e116. [3] M.O.C. Rocha, A.L.P. Ribeiro, M.M. Teixeira, Clinical management of chronic Chagas cardiomyopathy, Front Biosci. 8 (2003) 44e54. [4] M.M. Teixeira, R.T. Gazzinelli, J.S. Silva, Chemokines, inflammation and Trypanosoma cruzi infection, Trends Parasitol. 18 (2002) 262e265.
[5] F.S. Machado, G.A. Martins, J.C. Aliberti, F.L. Mestriner, F.Q. Cunha, J.S. Silva, Trypanosoma cruzi-infected cardiomyocytes produce chemokines and cytokines that trigger potent nitric oxide-dependent trypanocidal activity, Circulation 102 (2000) 3003e3008. [6] A. Talvani, C.S. Ribeiro, J.C. Aliberti, V. Michailowsky, P.V. Santos, S.M. Murta, A.J. Romanha, I.C. Almeida, J. Farber, J. Lannes-Vieira, J.S. Silva, R.T. Gazzinelli, Kinetics of cytokine gene expression in experimental chagasic cardiomyopathy: tissue parasitism and endogenous IFN-gamma as important determinants of chemokine mRNA expression during infection with Trypanosoma cruzi, Microbes Infect. 2 (2000) 851e866. [7] P.V.A. Santos, E. Roffeˆ, H.C. Santiago, R.A. Torres, A.P.M.P. Marino, C.N. Paiva, A.A. Silva, R.T. Gazzinelli, J. Lannes-Vieira, Prevalence of CD8 þ ab T cells in Trypanosoma cruzi-elicited myocarditis is associated with acquisition of CD62L low LFA-1 high VLA-4 high activation phenotype and expression of IFN-g-inducible adhesion and chemoattractant molecules, Microbes Infect. 3 (2001) 971e984. [8] A.P.M.P. Marino, A.A. Silva, P.V.A. Santos, L.M.O. Pinto, R.T. Gazzinelli, M.M. Teixeira, J. Lannes-Vieira, Regulated on activation, normal T cell expressed and secreted (RANTES) antagonist (MetRANTES) controls the early phase of Trypanosoma cruzi-elicited myocarditis, Circulation 110 (2004) 1443e1449. [9] J.C. Aliberti, F.S. Machado, J.T. Souto, A.P. Campanelli, M.M. Teixeira, R.T. Gazzinelli, J.S. Silva, Beta-Chemokines enhance parasite uptake and promote nitric oxide-dependent microbiostatic activity in murine inflammatory macrophages infected with Trypanosoma cruzi, Infect. Immun. 67 (1999) 4819e4826. [10] F.S. Machado, N.S. Koyama, V. Carregaro, B.R. Ferreira, C.M. Milanezi, M.M. Teixeira, M.A. Rossi, J.S. Silva, CCR5 plays a critical role in the development of myocarditis and host protection in mice infected with Trypanosoma cruzi, J. Infect. Dis. 191 (2005) 627e636. [11] C.R. Machado, A.L. Ribeiro, Experimental American trypanomiasis in rats: sympathetic denervation, parasitism and inflammatory process, Mem. Inst. Oswaldo Cruz. 84 (1989) 549e556. [12] I.R. Maldonado, M.L. Ferreira, E.R. Camargos, E. Chiari, C.R. Machado, Skeletal muscle regeneration and Trypanosoma cruzi-induced myositis in rats, Histol. Histopathol. 19 (2004) 85e93. [13] S. Youssef, G. Wildbaum, G. Maor, N. Lanir, A. Gour-Lavie, N. Grabie, N. Karin, Long-lasting protective immunity to experimental autoimmune encephalomyelitis following vaccination with naked DNA encoding C-C chemokines, J. Immunol. 161 (1998) 3870e3879. [14] S. Youssef, G. Maor, G. Wildbaum, N. Grabie, A. Gour-Lavie, N. Karin, C-C chemokine-encoding DNA vaccines enhance breakdown of tolerance to their gene products and treat ongoing adjuvant arthritis, J. Clin. Invest. 106 (2000) 361e371. [15] P. Menten, A. Wuyts, J. Van Damme, Macrophage inflammatory protein1, Cytokine Growth Factor Rev. 13 (2002) 455e481. [16] Z. Brener, Therapeutic activity and criterion of cure on mice experimentally infected with Trypanosoma cruzi, Rev. Inst. Med. Trop. Sao Paulo 4 (1962) 389e396. [17] D.G. Souza, D.C. Cara, G.D. Cassali, S.F. Coutinho, M.R. Silveira, S.P. Andrade, S.P. Poole, M.M. Teixeira, Effects of the PAF receptor antagonist UK74505 on local and remote reperfusion injuries following ischaemia of the superior mesenteric artery in the rat, Br. J. Pharmacol. 131 (2000) 1800e1808. [18] K.L. Cummings, R.L. Tarleton, Rapid quantitation of Trypanosoma cruzi in host tissue by real-time PCR, Mol. Biochem. Parasitol. 129 (2003) 53e59. [19] G.K. Reddy, C.S. Enwemeka, A simplified method for the analysis of hidroxyproline in biological tissues, Clin. Biochem. 29 (1996) 225e229. [20] C. Gerard, B.J. Rollins, Chemokines and disease, Nat. Immunol. 2 (2001) 108e115. [21] N.G. Frangogiannis, M.L. Entman, Targeting the chemokines in myocardial inflammation, Circulation 110 (2004) 1341e1342. [22] L.E. Ramirez, V.D. Silva, E. Lages-Silva, E. Chapadeiro, in: T.C. Arau´joJorge, S.L. Castro (Eds.), Doenc¸a de Chagas e Manual para experimentac¸~ao animal, Fiocruz/Instituto Oswaldo Cruz, Rio de Janeiro, 2000, pp. 140e142.
E. Roffeˆ et al. / Microbes and Infection 8 (2006) 2745e2755 [23] M.T. Rivera-Vanderpas, A.M. Rodriguez, D. Afchain, H. Bazin, A. Capron, Trypanosoma cruzi: variation in susceptibility of inbred strains of rats, Acta Trop. 40 (1983) 5e10. [24] E. Chapadeiro, P.S. Beraldo, P.C. Jesus, W.P. Oliveira Junior, L.F. Junqueira Junior, Cardiac lesions in Wistar rats inoculated with various strains of Trypanosoma cruzi, Rev. Soc. Bras. Med. Trop. 21 (1988) 95e103. [25] R.S. Bystry, V. Aluvihare, K.A. Welch, M. Kallikourdis, A.G. Betz, B cells and professional APCs recruit regulatory T cells via CCL4, Nat. Immunol. 12 (2001) 1126e1132. [26] L.A. Stephens, D. Gray, S.M. Anderton, CD4 þ CD25þ regulatory T cells limit the risk of autoimmune disease arising from T cell receptor crossreactivity, Proc. Natl. Acad. Sci. USA 102 (2005) 17418e17423. [27] L.E. Weston, A.F. Geczy, H. Briscoe, Production of IL-10 by alloreactive sibling donor cells and its influence on the development of acute GVHD, Bone Marrow Transplant 37 (2006) 207e212. [28] D.J. Ahern, D.S. Robinson, Regulatory T cells as a target for induction of immune tolerance in allergy, Curr. Opin. Allergy Clin. Immunol. 6 (2005) 531e538.
2755
[29] D.M. Vitelli-Avelar, R. Sathler-Avelar, J.C. Dias, V.P. Pascoal, A. Teixeira-Carvalho, P.S. Lage, S.M. Eloi-Santos, R. Correa-Oliveira, O.A. Martins-Filho, Chagasic patients with indeterminate clinical form of the disease have high frequencies of circulating CD3 þ CD16 CD56þ natural killer T cells and CD4 þ CD25 high regulatory T lymphocytes, Scand. J. Immunol. 62 (2005) 297e 308. [30] M. Samson, O. Labbe, C. Mollereau, G. Vassart, M. Parmentier, Molecular cloning and functional expression of a new human CC-chemokine receptor gene, Biochemistry 35 (1996) 3362e3367. [31] S.B. Su, N. Mukaida, J. Wang, H. Nomura, K. Matsushima, Preparation of specific polyclonal antibodies to a C-C chemokine receptor, CCR1, and determination of CCR1 expression on various types of leukocytes, J. Leukoc. Biol. 60 (1996) 658e666. [32] C. Moreno, T. Gustot, C. Nicaise, E. Quertinmont, N. Nagy, M. Parmentier, O. Le Moine, J. Deviere, H. Louis, CCR5 deficiency exacerbates T-cell-mediated hepatitis in mice, Hepatology 42 (2005) 854e862.
