| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Laboratorio de Biologia Molecular de la Enfermedad de Chagas, Instituto de Investigaciones en Ingeniería Genética y Biología Molecular (INGEBI), 1428, Buenos Aires, Argentina
1Correspondence: Laboratorio de Biologia Molecular de la Enfermedad de Chagas, INGEBI, Vuelta de Obligado 2490, 1428 Buenos Aires, Argentina. E-mail: mlevin{at}dna.uba.ar
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: Chagas disease • ß1-adrenoreceptor • G-protein-coupled receptors • arrhythmia • supraventricular tachycardia
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Sudden cardiac death due to sustained ventricular tachyarrhythmia is exceedingly common (2)
. Therefore, the conventional electrocardiogram (ECG) is the simplest and most widely available technique for diagnosis of chronic chagasic heart compromise. Histological examination of heart necropsies reveals mononuclear cell infiltration, myocyte damage, and fibrosis (1
, 2)
. Detection of parasite rests at the site of myocarditis indicate that heart parasitation is a major stimulus for perpetuation of myocardial inflammation and the generation of a strong anti-parasite response, directed mainly against intracellular parasite components, that may cross-react with heart tissue (3
, 4)
. In patients with cChHD, antibodies against ribosomal P proteins are prevalent (5
6
7
8
9)
. They are known as anti-P antibodies and are directed against the carboxyl-terminal regions of the T. cruzi ribosomal P0 (TcP0) and P2ß (TcP2ß) proteins (5
6
7
8
9
10
11)
. Fine epitope mapping demonstrated that these antibodies are directed to the acidic portions of the corresponding carboxyl-terminal epitopes, which enables them to react with the acidic epitope of the second extracellular loop of the ß1-adrenergic receptor (11
12
13)
. Due to this cross-reactive property, anti-P antibodies from cChHD patients exert a positive chronotropic effect in vitro on cardiomyocytes, suggesting they may be involved in the induction of arrhythmias and/or other electrical disorders characteristic of this chronic infection (14)
. However, no direct proof of their potential arrhythmogenic properties is available.
The mouse model of T. cruzi infection reproduces certain features of the human disease including chronic, focal, and diffuse myocarditis as well as changes in heart beating rate, complex arrhythmias, atrioventricular dissociation, and a high incidence of supraventricular tachycardia (15
16
17)
.
T. cruzi ribosomes or ribosomal proteins have also been associated with the induction of heart dysfunction in experimental models (18
19
20
21)
. Indeed, immunization of rabbits or mice with parasite ribosome-enriched fractions induced inflammatory lesions in heart as well as electrocardiographic disturbances similar to those found in chronically infected mice (18
, 21)
. In agreement with these results, immunization of mice with a single ribosomal component, the recombinant ribosomal protein TcP2ß, generated a strong reaction against the carboxyl-terminal portion of the T. cruzi ribosomal P protein that correlated with the recording of anomalous electrocardiograms (22)
.
To explore the connection between antibodies to the ribosomal P proteins of T. cruzi and arrhythmias and/or other heart disturbances that are usual pathological findings of cChHD, we immunized mice with either TcP0 or TcP2ß recombinant protein. The antibody response was monitored, mapped, and evaluated as to its functionality on cultured neonatal rat cardiocytes. Each ribosomal P protein elicited a strong and specific antibody response against their 13 residue-long carboxyl-terminal epitopes, P013 and R13 for TcP0 and TcP2ß protein, respectively, which cross-reacted with and stimulated the ß1-adrenergic receptor. High antibody titers of anti-P013 and anti-R13 adrenergic-stimulating activities correlated significantly with the recording of supraventricular arrhythmia, repolarization abnormalities, and conduction disturbances, revealing unequivocally their pathogenic nature.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
gt11 T. cruzi cDNA library derived from mRNA of bloodstream trypomastigotes allowed isolation of recombinant phages gt11-TcP2ß and gt11-CH3. The cDNA from gt11-TcP2ß encoded 102 of the 107 residues of the T. cruzi ribosomal P2ß protein (TcP2ß) (22)
. Recombinant clone gt11-CH3 expresses an almost complete copy of the T. cruzi ribosomal P0 protein (TcP0) (23)
. CH3 codifies 291 of 321 amino acids corresponding to the carboxyl-terminal end of TcP0 lacking only 30 residues of the nonantigenic NH2 terminus. To produce enough recombinant protein for immunization experiments, the cDNA inserts were subcloned in convenient expression plasmids. Consequently, CH3 cDNA was amplified by PCR using primers S1 (5' GAGCACGTCAGGATCCGCGGAAT 3') and S2 (5' GCGACCGAAGCTTAGCTGGAATTC 3'), then cloned into pMal-c2 plasmid (New England Biolabs, Cambridge, MA) in the BamHI-HindIII sites. The underlined sequences correspond to the BamHI-HindIII sites. The pMal-CH3 plasmid was digested with EcoRI and the CH3 cDNA was cloned into the EcoRI site of the pGex-1T plasmid (Amersham-Pharmacia Biotech, Piscataway, NJ). The TcP2ß cDNA that had been cloned into pMal-c2 (22)
was also subcloned into the EcoRI site of pGex-1T. Production and purification of the maltose binding protein (MBP) and glutathione-S-transferase (GST) fusion proteins MBP-TcP0, MBP-TcP2ß, GST-TcP0, and GST-TcP2ß were performed as indicated by the suppliers.
Human and mouse sera
Serum samples from individuals with cChHD were obtained as described previously (5)
. Sera from mice chronically infected with T. cruzi were kindly provided by Dr. González Cappa, Departamento de Microbiología, Facultad de Medicina, Universidad de Buenos Aires, Argentina (24)
.
Immunization schedule
Groups of Balb/c mice, aged 6–8 wk, were immunized intraperitoneally (i.p.) with 6 doses (days 1, 14, 28, 42, 56, and 70) of purified MBP-TcP0 or MBP-TcP2ß (50 µg/mouse) plus incomplete Freund adjuvant (IFA) (Sigma, St. Louis, MO). A similar protocol was used for sex- and age-matched control groups receiving MBP plus IFA or IFA alone. Mice were bled on days 0 (bleed 1), 26 (bleed 2), 52 (bleed 3), 66 (bleed 4), and 82 (bleed 5). Approximately 50–100 µl of serum was obtained from each mouse in bleeds 1–3. In bleeds 4 and 5, larger volumes were extracted for purification of immunoglobulin G (IgG) fractions.
ECG records
Electrocardiograms were performed on days 0, 40, 68, and 82. Mice were anesthetized with pentobarbital according to the protocol described by Pilgrim and DeOme (25)
. This method does not modify the ECG. Records were taken 10–20 min after the injection of pentobarbital. ECGs were obtained with the six standard leads (I, II, III, AVR, AVL, AVF) at 50 mm/s of paper speed and at 20 mm/mV amplitude using a Fukuda-Denshi Fx-2111 electrocardiograph (Tokyo, Japan). Electrocardiographic analysis included measurements of heart rate, P wave duration and amplitude, QRS complex duration and amplitude, P-R interval duration and a search for disturbances of rhythm, conduction, and repolarization. Effect of isoproterenol (Sigma) on cardiac frequency was assessed by recording ECGs 2 min after i.p. injection of 0.2 ng/g body weight of isoproterenol.
SDS-PAGE and Western blots
Standard procedures for SDS-PAGE and Western blots were used (6)
. Recombinant proteins (2 µg/lane) were loaded onto 10% polyacrylamide gels and stained with Coomassie blue or blotted to nitrocellulose. Sera from cChHD patients and T. cruzi chronically infected mice were tested on immunoblots at 1:200 dilution. Bound IgGs were detected with the Vectastain ABC kit (Vector Laboratories, Burlingame, CA) using the manufacturer’s instructions. In inhibition experiments, 2 ml of diluted sera was incubated for 2 h at 37°C with 10 µg of GST recombinant protein or 100 µM of synthetic peptides.
Enzyme-linked immunoassay (ELISA) determinations
ELISA experiments were performed as described previously (7)
. Polystyrene immunoplates were coated overnight at 4°C with 50 µl of 2 µM bovine serum albumin (BSA) -conjugated peptides, 50 µM of uncoupled peptides, or 2 µg/ml of GST recombinant proteins in 0.05 M bicarbonate-carbonate buffer (pH 9.6). BSA and GST were used as controls. Dilution of the serum samples was 1:200 except on titration experiments. To titrate the sera, 1:2 serial dilutions (from 1:100 to 1:102,400) were assayed against the corresponding antigen. The end point of ELISA titer was defined as the last serum dilution that gave an absorbance value greater than the mean plus 2.5 SDs of normal serum samples.
Synthetic peptides
Peptides were prepared by the solid-phase method of Merrifield as described by Müller et al. (26)
with a semiautomatic multisynthesizer NPS 4000 (Neosystem, Strasbourg, France). P013 (EDDDDDFGMGALF) and R13 (EEEDDDMGFGLFD) were derived from the 13 carboxyl-terminal amino acids of TcP0 and TcP2ß protein, respectively. Peptides were coupled at a molar ratio of 1:30 to BSA (Sigma) with 0.05% glutaraldehyde as described (26)
. The peptide H26R (HWWRAESDEARRCYNDPKCCDFVTNR) corresponds to amino acids 197–222 of the human ß1-adrenoreceptor (12)
.
Alanine mutation scanning
Fine epitope mapping of the carboxyl-terminal region of TcP2ß (R13) and TcP0 (P013) was performed with the SPOTs® kit (Genosys Biotechnologies Inc., Cambridgeshire, UK). Two sets of 14 peptides representing the R13 and P013 sequences and its 13 alanine replacement analogs were synthesized on a membrane by the supplier. The interactive residues were defined after incubation of the membranes with 1:200 dilutions of serum samples from TcP2ß or TcP0 immunized mice for 2 h at room temperature, followed by 1 h incubation with peroxidase-conjugated anti-mouse IgG (Sigma) and revealed by the ECL kit (Amersham).
IgG fractionation and affinity purification
IgG fractions of the sera were prepared by diluting the sera 1:5 in phosphate-buffered saline (PBS), pH 7.4, and further precipitation with 40% (NH4)2SO4. The precipitate was redissolved in PBS at 1:1 dilution. Monospecific polyclonal antibodies against epitopes R13 and P013 were affinity purified using Act-Ultragel AcA 22 resin (Sepracor, Villeneuve La Garenne, France) coupled to the corresponding peptide as described by the manufacturer. Purified antibodies to be used in functional assays were previously dialyzed against PBS.
Functional assay
Spontaneously beating cultured neonatal rat cardiomyocytes were used to assess the functional effects of IgG fractions from immunized mice (11)
. Single cells were dissociated from the minced heart of Wistar rats with a trypsin-collagenase solution. Myocytes were cultured for 4 days at 37°C in a 5% CO2 atmosphere as monolayers in DMEM/F-12 (Life Technologies, Gaithersburg, MD) containing 5% calf serum. The baseline beating rate was measured in 10 different fields at 37°C on the heated stage of an inverted microscope. The baseline beating rate was 120 ± 19 beats/min (bpm). Measurements were repeated 1 h after exposure to IgG fractions at a 1:50 dilution and after subsequent addition of the specific antagonists atropine (10-6 M), bisoprolol (10-6 M), and propranolol (10-6 M). The blocking properties of the peptides were studied by preincubating the IgGs with the peptides (60 µM) and, in the case of an absence of response, adding the specific agonist to confirm that the peptides had no blocking effect per se. All reported data were the mean of at least three measurements. The functional test was considered positive for the presence of anti-ß-adrenergic receptor antibodies if addition of IgG induced a statistically significant increase in the beating rate (compared with the control IgG) that was neutralized by propranolol. It was considered positive for the presence of anti-M2 cholinergic receptor antibodies whenever addition of IgG led to a statistically significant reduction in the beating rate (compared with the control IgG) that was antagonized by atropine. Data are shown as increase in bpm with respect to the control beating rate.
cAMP measurements in COS-7 transfected cells
This assay was performed as described (14)
. COS-7 cells were transfected with 1 µg of pBC expression vector containing cDNA encoding for the human ß1-adrenergic receptor. The principle of competitive protein binding was used to assess cAMP production in transfected cells after incubating with a 1:50 dilution of the IgGs for 60 min (14)
.
Statistical analysis
Statistical analyses were done by Fisher exact test or paired Student’s t test where appropriate. Medians in Fig. 2
were compared by using the Wilcoxon Rank Sum test. Correlation between the anti-R13 antibody titer and heart rate was calculated by the Simple Linear Regression module of the Statistix for Windows Software (Advance Investment Technologies, La Jolla, CA). The level of statistical significance for all tests was set at P < 0.05.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
7
8
9)
. This finding led to characterization of the peculiar ribosomal P protein system of T. cruzi composed by a least four different immunodominant polypeptides. This set of parasite proteins can be classified in two main types, depending on the sequence of their carboxyl-terminal epitopes: the high molecular weight type composed of the 34 KDa TcP0 protein with the carboxyl-terminal antigenic motif P013, EDDDDDFGMGALF (peptide P013); and the low molecular weight type composed of the ribosomal TcP1, TcP2
, and TcP2ß proteins, ranging between 10 and 12 KDa, sharing the 13 residue-long carboxyl-terminal motif EEEDDDMGFGLFD, peptide R13 (27)
. To assess the humoral, functional, and electrocardiographic responses induced by immunization with both P protein types, mice were immunized with either the ribosomal TcP0 or the TcP2ß recombinant MBP protein. The specific anti-P response was followed by using either GST-TcP0 or -TcP2ß fusion proteins and synthetic peptides representing their corresponding carboxyl-terminal ends.
Figure 1
shows the expression and affinity purification of MBP-TcP0 (panel A, lanes 1–3) and MBP-TcP2ß (panel B, lanes 1–3). The serum of a cChHD patient and that of a chronically infected mouse were used to test their immunoreactivity. Both human and mouse sera reacted with MBP-TcP0 (Fig. 1A
, lanes 4 and 7) and MBP-TcP2ß (Fig. 1B
, lanes 4 and 7). These reactions were abolished by preincubation with the corresponding GST recombinants (Fig. 1A, B
, lanes 5 and 8) and peptides P013 and R13 (Fig. 1A, B
, lanes 6 and 9), indicating that GST fusions and synthetic peptides were efficient tools to monitor the humoral anti-P immune response.
|
Humoral response induced by immunization with recombinant P proteins
Groups of 12 mice were immunized with either MBP-TcP0 or MBP-TcP2ß as described above. Control groups were immunized with MBP (n=12), IFA alone (n=12), or were not immunized (n=12). During the immunization experiment, a mouse inoculated with MBP and a naive one died establishing the spontaneous mortality rate of our 12 wk protocol. Unexpectedly, within the group of mice immunized with MBP-TcP2ß, a mouse died before the third bleed whereas five mice that developed a strong anti-R13 response died between the fourth and fifth bleeds, resulting in a mortality rate significantly higher than that of controls (P<0.001).
To study the antibody response, serum samples from immunized and control mice were extracted after the second, fourth, fifth, and sixth boost. Mice immunized with TcP0 or TcP2ß recombinants raised a humoral response against the whole parasite protein that reached titers of 51,200 (bleed 5, n=12, P<0.001) and 12,800 (bleed 4, n=11, P<0.001), respectively (Fig. 2
A, B). Ten of 12 (83%) mice immunized with MBP-TcP0 generated anti-P013 antibodies, with the final median titer at bleed 5 of 6400 (n=12, P<0.001) (Fig. 2C
). On the contrary, only 6 of 11 (55%) mice immunized with MBP-TcP2ß reacted with the R13 peptide as measured in samples from the fourth bleed, the median titer of anti-R13 being 1600 (P<0.001) (Fig. 2D
). Antibodies against the recombinant P proteins were first detectable at the second bleed, whereas the response against their carboxyl-terminal ends was measurable only from the third blood extraction onward. Clearly, immunization with TcP0 generated a more intense specific response than immunization with TcP2ß. In accordance with this, the response against P013 reached higher titers than the response against R13. It is noteworthy that the antibody responses against the parasite proteins and peptides were mainly of the IgG1 isotype and, to a lesser extent, IgG2a and IgG2b (data not shown).
Fine epitope mapping of the antibody response against the carboxyl-terminal ends of the ribosomal P proteins
Considering the functional implications of antibodies against the carboxyl-terminal ends of TcP0 and TcP2ß obtained from cChHD patients (11
12
13)
, we performed alanine mutation scanning of peptides P013 and R13 with sera from mice immunized with TcP0 or TcP2ß to assess the relevance of each amino acid in their reactive properties. As shown in Fig. 3
, the essential residues for recognition of P013 were the Asp residues at position 2, 4, and 6 together with the contiguous Phe in position 7 (capital letters in Fig. 3
). The relevant amino acids for peptide R13 were the Glu at position 3, the Asp residues at positions 5 and 6, and amino acids Gly and Phe at positions 8 and 9, respectively (Fig. 3)
.
|
This analysis reveals that although acidic, both carboxyl-terminal epitopes are discontinuous and present different antigenic motifs: DxDxDF for P013 and ExDDxGF for R13. The latter has homology with the ExDE acidic cChHD epitope of the second extracellular loop of the ß1-adrenergic receptor (12)
, which may also be targeted by anti-P013 antibodies (10)
. In fact, the anti-P0 specificity may not only recognize the ExD acidic motif of the second extracellular loop, but also the DF dipeptide at its carboxyl-terminal end if its conformation brings both ends of the loop together (Fig. 3)
(28)
.
Functional activity of antibodies induced by immunization with recombinant P proteins
Functional recognition of cardiac receptors by the IgG fraction of serum samples from TcP0 or TcP2ß immunized mice was assessed on spontaneously beating neonatal rat cardiomyocytes. This system has been used before to study anti-receptor antibodies in Chagas disease (11
, 12
, 14)
. Figure 4
A–C depicts the chronotropic effects of IgG fractions. The baseline beating rate was not modified by either the IgG of naive mice or by IgG from mice immunized with IFA or MBP alone (Fig. 4C
). IgG fractions obtained from sera of mice immunized with TcP0 or TcP2ß that did not react with the corresponding carboxyl-terminal peptides also failed to modify the baseline beating rate (sera 37 and 40 in Fig. 4A
; sera 26 and 27 in Fig. 4B
). Only IgG fractions from immunized mice presenting anti-P013 or -R13 antibodies exerted positive chronotropic effects (Fig. 4A, B
, respectively), albeit with specific features. The latter induced an increase in cardiocyte beating frequency that was completely blocked by the ß1 specific blocker bisoprolol and by peptides R13 and H26R (Fig. 4B
). The IgG fractions with anti-P013 antibodies also increased the beating rate of cardiocytes, but this effect was slightly increased by the addition of atropine and was not completely abolished by the ß1 antagonist (Fig. 4A
) or by propanolol (not shown). Accordingly, preincubation of the IgG fraction with peptide H26R did not completely block the effect of antibodies, which was nevertheless completely abolished by P013. The functional response to antagonists and peptides suggests that anti-P013 antibodies may be acting on membrane structures other than the mentioned cardiac receptors, which also influence cardiocyte automatism.
|
Functional testing of the immunopurified anti-P013 and anti-R13 antibodies was in complete agreement with that performed with IgG fractions, since preincubation with the specific peptide (P013 or R13) or H26R abolished the functional activity of these monospecific antibodies (Fig. 4D, E
). To unequivocally establish their functional nature, we also immunopurified anti-P antibodies from mice that had high anti-P013 and -R13 antibody titers but normal ECGs (see below; mice 30 and 39 [bleed 5] for anti-P013, mouse 24 [bleed 4] for anti-R13, Fig. 4D, E
, Table 2
). In all cases these immunopurified antibodies showed the same ß-adrenergic-stimulating activity as that derived from mice with altered ECG (mice 32 and 38 [bleed 5] for anti-P013; mouse 18 [bleed 3] for anti-R13) (Fig. 4D, E
). As expected (14)
, IgG fractions containing anti-P013 and anti-R13 antibodies and the corresponding immunopurified antibodies, which exerted a positive chronotropic action on neonatal rat cardiomyocytes, also induced an increase in cAMP production of COS-7 cells transfected with the ß1-adrenergic receptor, confirming the catecholamine-like properties of these antibodies (not shown).
|
Immunization with P proteins induced electrocardiographic changes
To analyze the ECG of immunized mice and considering the slight variations found in the ECGs of different mouse strains (16
, 17
, 29
, 30)
, we established the parameters of the ECG of normal Balb/c mice. Figure 5
shows the P wave, the P-R segment, the QRS (QRST) complex, and normal values of the measured parameters. In a normal tracing, the T wave corresponding to ventricular repolarization starts immediately after QRS deflection and is represented as a gentle slope rather than a distinctive peak, resulting in a QRST complex as reported for some other mouse strains (29
, 30)
. This feature impairs characterization of the S-T segment as well as the voltage and amplitude of the T wave (Fig. 5)
. The frequency of the sinus rhythm of naive mice and mice immunized with IFA and MBP did not change throughout the experiment (Table 1
). Electrocardiographic tracings of naive mice and those of mice immunized with IFA or MBP showed a complete absence of disturbances of the cardiac rhythm, as well as the first ECGs recorded from mice immunized with the recombinant parasite proteins. The second ECGs from all immunized mice presented no abnormality even though anti-TcP0 and anti-TcP2ß antibody levels were already detected in bleeds 2 and 3 (Fig. 2)
.
|
|
All 12 mice immunized with TcP0 presented antibodies against the whole protein, but only 10 developed a humoral response against P013 (Fig. 2
, Table 2
). Seven of 10 mice with anti-P013 antibodies presented alterations in the ECG at the fourth recording (P<0.001, Fig. 6
, Table 3
). The ECG abnormalities are summarized in Tables 2
and 3
. Thirty-three percent of the mice with anti-P013 antibodies (3 of 10) displayed supraventricular arrhythmias (Fig. 6A, B
). Two mice displayed AV node tachycardia characterized by AV dissociation by interference, as shown in Fig. 6A
. The third mouse of this group presented changes in P wave morphology and voltage, evident in high-voltage derivatives II, III, aVR, and aVF, characteristic of multifocal atrial tachycardia (only derivative II is shown in Fig. 6B
). Another group of three mice immunized with TcP0 and anti-P013 positive response presented ventricular repolarization abnormalities (Fig. 6C
). The ECGs of these mice were the only ones among the 60 mice studied that showed a clearly identifiable T wave, albeit with aberrant shapes. Figure 6C
1 shows the tracing of a mouse presenting the T wave as a distinct high-voltage peak close to the QRS complex. This was the only ECG abnormality displayed as early as the third recording. A second animal showed a high-voltage inverted T wave and a depressed S-T segment (Fig. 6C
2) whereas the third showed a wide T wave separated from the QRS complex (Fig. 6C
3). The position of T waves in Fig. 6C
2, C3 define a clear S-T interval that is not registered as such in the normal recording (Fig. 5)
. One animal presented a fourth ECG recording with a first degree AV block (1/10, 10%), a typical finding in chronic human and experimental infections (Fig. 6D
). This mouse evolved from a third normal ECG register to an abnormal fourth, where it showed (as a distinctive and permanent feature) an increase in the P-R segment and an infra level of the P-R segment (Fig. 6D
, arrow in lead II).
|
|
The group of mice immunized with MBP-TcP2ß presented a significant increase in heart rate (523±30, P<0.001, Table 1
) at the third ECG. This increase was restricted to those mice positive for anti-R13 antibodies, all of which presented ECGs consistent with the recording of supraventricular tachycardia (Table 2
and Fig. 7A
). Remarkably, a statistical correlation could be established between increased heart rates and high levels of anti-R13 reactivity (Fig. 8
C, P<0.01).
|
|
To further confirm that this arrhythmia was adrenergically driven, we compared the fourth ECG tracing of a R13 strongly reactive animal (Fig. 7A
) with that of a naive mouse after i.p. administration of isoproterenol (Fig. 7B
). The similarity between the recordings in Fig. 7
(TcP2ß and isoproterenol) show clearly that anti-R13 specificities with adrenergic-stimulating activity exert a catecholamine-like effect, resulting in supraventricular tachycardia analogous to that induced by isoproterenol.
Correlation between ECG disturbances and prevalence of anti-P013 and -R13 antibodies with a marked adrenergic-stimulating activity
Analysis of the variations in the levels of functional anti-P013 and -R13 antibodies during the immunization protocol and the corresponding changes in the ECGs, accompanied by death in the case of TcP2ß, revealed significant associations. Anti-P013 antibodies were measured as early as the third bleed (52 days after the beginning of the protocol), with antibody titers reaching 1:1600 in one case. However, no changes in the ECG were recorded. This trend was maintained 2 wk later, with the exception of mouse 38, which presented a high anti-P013 titer (1:6400) and an ECG pattern of abnormal repolarization. The remaining TcP0 immunized mice (the majority with titers ranging between 1:1600 and 1:6400) had normal ECGs (Table 2)
. The last ECG recording and bleed were performed 2,5 wk later. By that time, 70% of the 10 anti-P013-positive mice presented electrogenic or conduction disturbances in their ECG (Tables 2
and 3)
. All mice with 1:12800 anti-P013 antibody titers presented either supraventricular arrhythmia or abnormal repolarization, whereas first degree AV block was recorded in mouse 31 (Table 2)
, evolving from a 1:3200 to a 1:6400 titer. Another mouse with supraventricular arrhythmia (mouse 32, Table 2
) evolved similarly whereas mice 30 and 39 maintained a normal ECG tracing. Considering there were two TcP0 immunized mice negative for anti-P013 antibodies with normal ECGs, a significant association between high anti-P013 antibody levels and altered ECG recordings was evident (Fig. 8A
, 8P
<0.001). Since the levels of antibodies against the complete TcP0 protein were similar in both groups, the recording of altered ECGs was linked to the anti-P013 response (Fig. 8A
).
In contrast to the evolution of the TcP0 immunized mice, TcP2ß immunization associated with the induction of an anti-R13 response was lethal for five of six mice. Of a majority of mice with undetectable anti-R13 antibody levels at bleed 3 and normal ECG2 (Table 2)
, five evolved to supraventricular tachycardia, with anti-R13 antibody titers between 1:1600 and 1:6400 and heart rates ranging from 520 to 600 bpm (Table 2
and Fig. 8C
). All five died unexpectedly before the end of the protocol. Mortality significantly associated with high antibody levels to R13 and high heart rate frequencies (P<0.001 and P<0.01, respectively), suggesting that death may have been a consequence of supraventricular tachycardia (Fig. 8B
). The only mouse with anti-R13 antibody levels that survived until the end of the experiment (mouse 24, Table 2
) had anti-R13 antibody titers of 1:1600 at bleed 4 and a normal ECG3 (Table 2
, Fig. 8C
, open circles). After the sixth boost, mouse 24 evolved to an anti-R13 antibody titer of 1:3200 and supraventricular tachycardia, with a heart rate of 600 bpm (Fig. 8C
), confirming the link between the anti-R13 adrenergic-stimulating antibodies and this tachyarrhythmia. Light microscopic examination of heart tissue samples obtained from heart necropsies of mice that completed the immunization protocol did not reveal myocyte damage in any of the studied sections.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 31
, 32)
. Our studies of the nature of these antibodies led us to propose that these autoreactive specificities were in fact antibodies directed against intracellular parasite antigens, such as the T. cruzi ribosomal P proteins, with the ability to cross-react and stimulate cardiac receptors (11
12
13)
. This assumption had to be proved by the induction of symptoms in mice either actively immunized with the corresponding parasite protein or passively immunized by the transfer of anti-T. cruzi antibodies (33)
. That the former approach could be successful was supported by two observations: 1) immunization with at least one of the members of the T. cruzi ribosomal P protein family induced an anti-R13 response similar to that elicited by parasite proteins in cChHD, accompanied by induction of changes in the ECG of the immunized mice (22)
; and 2) at least two different functional assays demonstrated that anti-P013 and anti-R13 antibodies from cChHD patients interacted with G-protein-coupled receptors on the cardiocyte membrane (11
12
13)
.
Our results show a highly significant association between circulating anti-P013 or -R13 adrenergic-stimulating antibodies and the induction of supraventricular arrhythmias, repolarization abnormalities, and first degree AV conduction block by the end of immunization with TcP0 or TcP2ß. Of 24 mice, 16 generated antibodies against the carboxyl-terminal end of the corresponding protein, 13 of which showed an altered ECG (P<0.001, 81%). In accordance with the marked adrenergic activities of anti-P013 and -R13 antibodies, 9 of 13 positive mice showed supraventricular arrhythmias whereas abnormal repolarization may be interpreted as a direct consequence of arrhythmias induced by the catecholamine-like effect of circulating IgGs (Tables 2
and 3)
(34)
.
Differences between the TcP0 and the TcP2ß groups were 1) all six R13-positive mice of the TcP2ß group showed supraventricular tachycardia, which was most probably the cause of the high mortality of this group, 2) none of the TcP0 mice died; on the contrary, 10 of 12 developed an anti-P013 response that correlated with the induction of a broader spectrum of electrical and conduction disorders, including a case of first degree AV block, 3) anti-P013 antibodies exerted a complex functional effect on cardiocytes characterized by a major ß1-adrenergic stimulus, which was modulated by varying levels of an anti-M2 cholinergic receptor component and a relative low to moderate proportion of nonadrenergic-noncholinergic activity; in contrast, the anti-R13 antibodies had uniform and exclusive ß1-adrenergic activity, 4) at any given moment of the immunization, the anti-P013 median antibody titers were higher than the anti-R13 ones, but at the third ECG, five of six R13-positive mice presented ECG alterations (supraventricular tachycardia) against 1 of 10 in the TcP0 group (Fig. 2
and Table 2
). These results suggest a higher ß-adrenergic-specific activity for the anti-R13 antibodies and an increased pathogenic potential as judged by the mortality rate of anti-R13-positive mice. Anti-P013 antibodies contained counteracting chronotropic activities that were able to neutralize each other somewhat (Fig. 4A
). This complex physiological balance may explain why all P013 mice completed the immunization protocol. On the other hand, fluctuation of this balance may be at the origin of the recording of a first degree AV block in this group (Table 2)
. It is noteworthy that antibodies cross-reactive with ß1 and M2 cholinergic receptors have also been described in cChHD (12)
.
The functional reactive pattern of both anti-P antibodies is explained by their different fine specificities, as analyzed by alanine mutation scanning (Fig. 3)
. The resulting anti-R13 antigenic structure ExDDxGF confirms the importance of the third Glu residue in R13 (11)
and explains the induction of antibodies that recognize the cChHD acidic epitope AESDE of the second extracellular loop of the ß1-adrenergic receptor. Alanine scanning also disclosed the particular reactive structure of P013, DxDxDF. Its comparison with the sequences of the second extracellular loop of the receptors, particularly that of ß1, allows us to hypothesize that anti-DxDxDF antibodies may recognize a conformational epitope formed by the acidic ESDE and the carboxyl-terminal DF motif (Fig. 3)
brought together by folding of this portion of the receptor, as it has been described for another member of this family of transmembrane proteins (28)
. Similar primary structures present in the ß1-adrenoreceptor and in the M2 muscarinic receptor may explain a certain level of functional cross-reactivity between them and anti-P013 antibodies.
The reactive and autoreactive properties of the anti-P013 and -R13 antibodies and their physiological consequences were specific as they were measured, functionally evaluated, and recorded only in those mice that developed a response against the carboxyl-terminal end of the immunizing protein, not in TcP0-positive/P013-negative or TcP2ß-positive/R13-negative mice.
These studies may be relevant to our understanding of Chagas heart disease, which has an important arrhythmogenic component (2)
. A recent study reported that among 155 cChHD patients, 
70% presented induction of some kind of sustained supraventricular tachyarrhythmia (2)
whereas AV dysfunction was manifested by different degrees of AV block (1
, 2)
. Supraventricular tachycardia, repolarization abnormalities, and AV blocks are also a characteristic of the experimental disease (17)
. A common feature of human and experimental chronic infections is the strong humoral response against intracellular parasite antigens, among which the response against the carboxyl-terminal epitopes of the ribosomal P proteins is prevalent (3
, 5
6
7
8
9)
. The anti-cardiac receptor properties of these antibodies may play a pathogenic role by interfering with the electrophysiological properties of cardiac tissues, generating arrhythmias, and ultimately causing damage to myocardial cells (35)
. The pathological relevance of this anti-P response is stressed by the fact that the anti-cardiac receptor activity of the IgG fraction from cChHD patients can be depleted by incubation with either the R13 or the P013 synthetic peptide (13)
.
Our experimental model cannot mimic the complexity of a chronic T. cruzi infection that not only includes the humoral and cellular responses to the parasite, but also the constant presence of an active and elusive intracellular reproducing microorganism (4
, 36
, 37)
. However, it serves to establish the pathogenic nature of the TcP0 and the TcP2ß humoral response combining for the first time in the same study immunological (ELISA and alanine scanning) and cellular experiments (cardiocyte and receptor transfected cells in vitro functional assays) and electrocardiographic recordings. This allowed us to determine that although similar in amino acid composition, epitopes P013 and R13 generate humoral responses with specific functional activities that induce different disturbances in the cardiac rhythm, the summation of which reproduces (to a certain extent) some features of arrhythmias and conduction defects in cChHD.
Comparison of the antibody profiles induced by immunization with that of experimental chronic T. cruzi infections may allow us to understand why 40% of chronically infected mice with patent myocarditis and high antibody levels to T. cruzi do not display abnormal electrocardiographic recordings (16
, 17)
. This apparent normality may only be related to the relatively short time course of the corresponding protocols. In our case, mouse 24 had a normal electrophysiological condition and a relative high level of anti-R13 antibodies when other mice of the same group and with the same antibody titer showed supraventricular tachycardia (Table 2)
. By the end of the protocol, mouse 24 showed a net increase in antibody titer and a heart rate of 600 bpm (open circles in Fig. 8C
, Table 2
). Similarly, the majority of TcP0 immunized mice showing important anti-P013 antibody titers at the fourth bleed and normal ECGs increased their antibody levels and evolved to altered ECGs as the protocol ended (Table 2)
. It is tempting to hypothesize that a chronic immunization protocol (35)
might have turned the asymptomatic anti-P013 mice (mice 29, 30, and 39, Table 2
) symptomatic.
In establishing conditions for a successful immunization protocol and ECG recordings, we have prepared the tools necessary to continue this type of studies using extended time protocols in our laboratory. Moreover, recently obtained anti-P013 and anti-R13 monoclonal antibodies will be able to test the pathogenic role of anti-P antibodies in passive transfer experiments (38)
.
To summarize, we report that active immunization with recombinant T. cruzi ribosomal P0 and P2ß proteins elicited a strong response against their carboxyl-terminal ends with a marked ß-adrenergic-stimulating activity that strictly correlated with the recording of supraventricular arrhythmias, abnormal repolarizations, or conduction defects. This strongly suggests that these antibody responses may be of pathogenic importance in the development of cChHD. From a more general view, our immunization model with the TcP2ß protein may represent an excellent one to study sudden death by tachyarrhythmia in mice, given the paucity of information on the ECG abnormalities underlying this condition (39)
.
|
ACKNOWLEDGMENTS |
|---|
Received for publication July 6, 2001.
Revision received September 4, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||
