Human complement receptor type 1 (CR1) protein levels and genetic variants in chronic Chagas Disease

Complement is an essential element in both innate and acquired immunity contributing to the immunopathogenesis of many disorders, including Chagas Disease (CD). Human complement receptor 1 (CR1) plays a role in the clearance of complement opsonized molecules and may facilitate the entry of pathogens into host cells. Distinct CR1 exon 29 variants have been found associated with CR1 expression levels, increased susceptibility and pathophysiology of several diseases. In this study, CR1 plasma levels were assessed by ELISA and CR1 variants in exon 29 by sequencing in a Brazilian cohort of 232 chronic CD patients and 104 healthy controls. CR1 levels were significantly decreased in CD patients compared to controls (p < 0.0001). The CR1 rs1704660G, rs17047661G and rs6691117G variants were significantly associated with CD and in high linkage disequilibrium. The CR1*AGAGTG haplotype was associated with T. cruzi infection (p = 0.035, OR 3.99, CI 1.1-14.15) whereas CR1*AGGGTG was related to the risk of chagasic cardiomyopathy (p = 0.028, OR 12.15, CI 1.13-113). This is the first study that provides insights on the role of CR1 in development and clinical presentation of chronic CD.

role in the infection process, while the parasite deactivates the lectin complement pathway 16 , which ultimately could favor T. cruzi cell internalization mediated by receptors for both molecules, including CR1.
The complement system is essential in both innate and acquired immunity 17 , contributing to the immunopathogenesis of a variety of diseases, including CD 10,11,18,19 . CR1, or CD35, is a multi-functional polymorphic glycoprotein, which occurs as a soluble or transmembrane protein expressed on peripheral blood cells including monocytes and erythrocytes, natural killer cells as well as on B and T cells 17,20 . CR1 is known to enhance phagocytosis of particles opsonized with C3b, C4b, C1q, MBL, and ficolin-2 as well as to facilitate the clearance of immune complexes by binding to CR1 on erythrocytes and macrophages for further disposal 21,22 . The CR1 gene is located on chromosome 1q32.2 (OMIM 120620) and belongs to the Regulator of Complement Activation family, which is characterized by small consensus repeats, also known as complement control protein repeats 17,22 . Genetic variability may influence CR1 expression including its molecular weight and the density of CR1 molecules on cell surfaces 22,23 .
It has been demonstrated that CR1 is involved in the pathogenesis of several of infectious diseases either by facilitating pathogens entry into host cells in some cases or by down-modulating complement activation in others 24,25 . CR1 was shown to mediate immune opsonization of Leishmania amastigotes and promastigotes 26,27 , Plasmodium falciparum 28 , Mycobacterium tuberculosis 29 , M. leprae 30 , HIV 31,32 , SARS-CoV 33 , adenovirus 34 hepatitis C virus 31 and West Nile Virus 35 .
Besides the role of CR1 in facilitating the entry of intracellular pathogens into host cells, CR1 protein levels were shown to be associated with the pathogenesis of different diseases including malaria 28 , tuberculosis 36 , lepromatous leprosy 37 , severe acute respiratory syndrome 33 , chronic liver diseases 38 , HIV infection among others 39 . The CR1 genetic variants in exon 29 evaluated in this study (rs17259045, rs41274768, rs17047660, rs17047661, rs4844609 and rs6691117) are of particular interest since all are non-synonymous variants (https://www.ensembl. org) that are situated at the binding site for C1q, ficolins and MBL having thereby potential to influence the complement induced phagocytosis 21,22 . The present study aimed to assess if the genetic variants in exon 29 and CR1 levels are associated with development and clinical presentation of chronic CD.

Results
CR1 plasma levels. CR1 plasma levels were significantly lower in CD patients compared to controls (p < 0.0001), (Fig. 1). When comparing controls to each clinical form separately, statistical differences were also observed for CR1 levels between controls and the indeterminate form (p = 0.0002), cardiac form (p < 0.0001), digestive form (p < 0.0001), and cardiodigestive form (p < 0.0001) (Fig. 1). Comparison of CR1 levels between asymptomatic (indeterminate form) and symptomatic patients showed no statistical difference.
Patients with cardiomyopathy, graded according to the classification of cardiomyopathy as outlined in the Methods section, were compared to asymptomatic patients. The rs17047661G allele occurred more frequently in patients with cardiomyopathy without ECHO alteration (p = 0.028, OR 2.8, 95%CI 1.14-7.16) and in patients with cardiomyopathy without heart failure (p = 0.0065, OR 2.8, 95%CI 1.33-6.02) than in asymptomatic patients. Both rs17047661AG and rs17047661GG genotypes were observed more frequently among patients with cardiomyopathy without ECHO alteration (p = 0.031, OR 3.3, 95%CI 1.11-9.75) and in patients with cardiomyopathy without heart failure (p = 0.02, OR 1.7, 95%CI 1.08-2.79) than in patients without overt symptoms ( Table 2).
Comparing patients with cardiomyopathy and considering the different stages of cardiac pathology, the rs17047661G alleles (p = 0.0017, OR 0.19, 95%CI 0.06-0.59) and rs17047661AG and rs17047661GG genotypes (p = 0.007, OR 0.15, 95%CI 0.03-0.60) were more frequent in patients without heart failure than in those with heart failure ( Table 2).

Discussion
In order to maintain its life cycle after transmission by triatomine vectors to a human host, T. cruzi needs to evade host immune attack and develops further intracellularly. The successful entrance of T. cruzi into host cells depends on the down-regulation of complement activation by parasite regulatory molecules and by its binding to complement receptors such as CR1 10,11,24 . Thus, complement system and CR1 have an important role both in the establishment of T. cruzi infection and sustenance of the chronic phase. In this study, the CR1 genetic variants in exon 29 were investigated in patients with chronic CD in order to assess their role in the modulation of CR1 levels as well as in the development and in the clinical progression of the disease. Patients with chronic CD had significantly decreased levels of CR1 compared to healthy controls. Plasma levels observed in the control group were in accordance with those reported in other studies 38,40 . The reduced CR1 expression on erythrocytes combined with increased levels of immune complexes has been demonstrated in the pathogenesis of HIV, SARS-CoV, M. tuberculosis and M. leprae infections 33,36,37,39 . In leprosy, AIDS, and tuberculosis, the reduction of CR1 levels is disease regulated, demonstrating that this condition is acquired rather than inherited 36,39 . Moreover, it is known that, similarly to mechanisms used by other pathogens, T. cruzi uses C1q to promote C1-dependent phagocytosis as well as MBL and ficolin-2 to promote opsonization via CR1 as a strategy to evade the host immune system and infect host cells 15 .
Interestingly, patients with cardiomyopathy had lower CR1 plasma levels than asymptomatic patients, which might indicate either consumption due to increased complement activation or lower production associated with this clinical manifestation. In fact, chagasic cardiomyopathy is known to be associated with inflammatory process and tissue damage, as observed in various inflammatory and infectious conditions [41][42][43][44] . Moreover, one of the consequences of the persistent myocardial damage in CD is left ventricular dilation with systolic dysfunction 45 . For this reason, left ventricular systolic function was evaluated in CD patients using the Left Ventricular Ejection Fraction (LVEF). Despite the important role of the complement system in cardiovascular diseases 46   such as atherosclerosis 48,49 , myocardial infarction 50,51 , and acute ischaemic stroke 52 , no correlation between CR1 levels and LVEF was found. This finding corroborates with data from a study on patients with acute myocardial infarction 53 . It is known that CR1 levels may be influenced by infections and that their expression is associated with genetic as well as acquired factors 39 . It was observed in this study that lower levels of CR1 were associated with rs6691117GG genotype in the controls, but not in patients. Two other studies found this genotype associated with lower erythrocyte sedimentation rate 54 and with preterm birth 55 . These findings indicate that rs6691117GG genotype may modulate CR1 expression. Since there was no association between the rs6691117GG and CR1 levels in the patients, the reduction of CR1 levels in chronic CD is probably due to the disease process. An anti-inflammatory role for CR1 was already observed in experimental studies where CR1 was able to prevent tissue injury induced by complement activation 56 . Considering that chronic CD is associated with inflammation, it is possible that the low levels of CR1 in CD patients may be related to its anti-inflammatory effect and consumption due to complement activation. However, the exact mechanism, which controls the expression of CR1 in CD patients is still unclear.
The positive association of AG and GG genotypes (in variants rs17047661, rs6691117) and the G allele (in variant rs17047660G) observed with chronic CD may be related to the functional properties of the CR1 molecule. These variants lead to the substitutions of amino acids in the CR1 molecule which may affect the folding and the affinity of CR1 to C3b, C4b and C1q/MBL/ficolin-2 17,20-22,57,58 . The allele rs6691117G was also related to a low ratio of CR1 expression in erythrocyte membranes 54 . In addition, the alleles rs17047660G and rs17047661G were previously associated with severe malaria 59 , sickle cell anemia 60 , and showed to have protective effects against M. leprae 30 and M. tuberculosis infection 61 , while allele rs6691117G increased risk of Alzheimer disease 62 , gastric cancer 63 , non-small cell lung cancer 64 and preterm birth 55 .
Moreover, the allele rs17047661G and CR1*AGGGTG and AGAGTG haplotypes were related to early stages of CD cardiac form indicating that these variants might predispose to clinical progression of chronic patients with CD. Since the pathogenesis of CCC involves parasite persistence in different tissues as well as continuous low-grade parasitemia, inflammatory process and immune mediated-myocardial injury, it is possible that protein products of these CR1 variants may augment T. cruzi binding with consequent cellular internalization besides having an immunomodulatory effect.
A limitation of the present study is the lack of baseline CR1 plasma levels in patients with acute CD. Acute CD patients are difficult to diagnose clinically and hence the measurement of CR1 levels was not possible during their early stages of infection that might serve as a baseline measurement. The Ambulatory of Chagas Disease of Hospital das Clínicas (Federal University of Paraná) enrolls only chronic CD patients, thus making the access to acute CD patients impossible.
In conclusion, this study reports that CR1 variants are associated with the risk of T. cruzi infection and to progression to chagasic cardiomyopathy. Besides that, the low of CR1 levels observed in CD patients is possibly due to the disease process. This is the first study that provides insights on the role of CR1 in development and clinical presentation of chronic CD. Nevertheless, further studies are necessary to confirm these findings.       Statistical Analysis. CR1 plasma levels were compared between groups using nonparametric Kruskal-Wallis and Mann-Whitney tests. The distribution of each variable was assessed by the Shapiro-Wilk test. Multiple logistic regression was executed with adjustment for age, sex, and ethnic group. Multiple comparisons were corrected using a Benjamini-Hochberg procedure applying a false discovery rate of 0.10 and raw p-values that remained significant after this correction were considered in the study. Odds ratios (OR) and their 95% confidence intervals (CI) were calculated using the STATA software (v. 12.0, StataCorp, College Station, Texas, USA). Correlation analyses were performed by non-parametric Spearman's rank coefficient tests. Allele frequencies were obtained by direct counting. Genotype and haplotype frequencies were analyzed by gene counting and expectation-maximum (EM) algorithms and the significance of deviation from Hardy-Weinberg equilibrium was tested using the random-permutation procedure as implemented in the Arlequin v. 3.5.2.2 software (http://lgb.unige.ch/arlequin). Linkage disequilibrium (LD) analysis was performed using Haploview v. 3.2 (http://broadinstitute.org/haploview). Possible associations of CR1 alleles, genotypes, and haplotypes with different clinical forms were evaluated with two-tailed Fisher exact tests. P-values < 0.05 were considered significant.