Association of L-Ficolin Levels and FCN2 Genotypes with Chronic Chagas Disease

Background L-ficolin (encoded by FCN2) binds to acetylated sugar moieties of many pathogens, including Trypanosoma cruzi, promoting their phagocytosis and lysis by the complement system. Methods We investigated L-ficolin levels in 160 T. cruzi infected patients with chronic Chagas disease and 71 healthy individuals, and FCN2 polymorphisms (−986 G>A, −602 G>A, and −4 A>G in the promoter and A258S in exon 8) in 243 patients, being 88 indeterminate (asymptomatic), 96 with cardiac, 23 with digestive and 33 with cardiodigestive manifestations (two were unspecified) and 305 controls (135 for A258S). Results Patients presented lower L-ficolin plasma levels than controls (p<0.0001). Among the different groups of cardiac commitment, individuals with moderate forms had higher L-ficolin levels than the severe forms (P = 0.039). Lower L-ficolin levels were found associated with the 258S variant in the patients (P = 0.034). We found less −4A/G heterozygotes in the cardiac patients, than in the controls (OR = 0.56 [95% CI = 0.33–0.94], P = 0.034). Heterozygote −4A/G genotypes with the 258S variant and 258SS homozygotes were nevertheless more frequent among cardiodigestive patients than in controls (OR = 14.1 [95% CI = 3.5–56.8], P = 0.0001) and in indeterminate patients (OR = 3.2 [95% CI = 1.1–9.4], P = 0.037). We also found an association of the allelic frequency of the 258S variant with cardiodigestive Chagas disease compared to controls (OR = 2.24 [95% CI = 1.1–4.5], P = 0.037). Thus, decreased patient levels of L-ficolin reflect not only protein consumption due to the disease process, but also the higher frequency of the 258S variant in patients with cardiodigestive symptoms. Conclusion The very first study on Brazilian cohort associates both L-ficolin plasma levels and FCN2 variants to Chagas disease and subsequent disease progression. The prognostic value of L-ficolin levels and the FCN2*A258S polymorphism should be further evaluated in other settings.


Introduction
Chagas disease (CD) occurs in 18 different countries, mostly throughout South and Central America, and affects approximately 15 million people worldwide [1]. CD pathogenesis, caused by the flagellated protozoan Trypanosoma cruzi, is still poorly understood and there is no available marker that indicates the progression to the different clinical forms or prognosis of chronic disease. Despite the fact that approximately 50% of the individuals infected by T. cruzi stay in the indeterminate or asymptomatic form throughout their lives, which in general present good prognosis, each year about 2-5% of them progress to symptomatic forms of the disease, presenting irreversible cardiac and/or digestive and/or disorders [2]. A plausible presumption is that individuals who remain asymptomatic are able to reduce parasite numbers in the early phase of infection, and down modulate the immune response, limiting the development of pathology. On the other hand, individuals who will develop cardiac disease, although capable of parasite control, may not be capable of mounting efficient immunoregulatory mechanisms, thus leading to establishment of persistent inflammation [3].
The complement system has been shown to play an important role in the control of experimental T. cruzi infection, as well in clinical evolution of CD [4][5][6][7][8]. Ficolins are pattern-recognition proteins which bind to specific pathogen-associated molecular patterns (PAMP) on microorganism surfaces, promoting activation of the complement cascade through the lectin pathway thereby triggering the innate immune response [9]. Other putative functions of ficolins include binding to late apoptotic cells, apoptotic bodies and necrotic cells, enhancing their uptake by macrophages [10]. The ficolins are synthesized as a single polypeptide containing N-collagen-tails and C-terminal fibrinogen-like binding domains, which are oligomerized into higher oligomeric forms [9]. In humans, three ficolin genes have been identified: FCN1, FCN2 and FCN3, which encode M-ficolin (ficolin-1), L-ficolin (ficolin-2) and H-ficolin (Hakata antigen or ficolin-3), respectively. Single nucleotide polymorphisms (SNPs) in the promoter region of the FCN2 gene have been associated with the variability in the individual serum concentrations of the protein. Studies have shown that the presence of the nucleotide adenine (A) at positions 2986 A.G and 2602 A.G as well as the nucleotide guanine (G) at the position 24G.A are related to higher L-ficolin serum levels [11]. In addition, two polymorphisms located in exon 8, encoding the fibrinogen-like domain (containing T236M and A258S) are associated with decreased and increased ability of binding to acetylated residues, respectively [11][12][13]. Lficolin levels and FCN2 polymorphisms have been reported to be associated with different diseases [14][15][16][17][18][19][20].
Both L-ficolin and H-ficolin are able to interact with the mannan-binding lectin (MBL)-associated serine proteases, promoting activation of the complement cascade [21]. L-ficolin recognizes N-acetylated molecules, including N-acetylglucosamine (GlcNAc) or N-acetylgalactosamine (GalNAc) [22,23]. The surface of T. cruzi contains a large diversity of N-linked and O-linked carbohydrate-rich molecules and it was recently demonstrated that T. cruzi activates the lectin pathway [7]. The authors have shown that a fast binding of L-ficolin occurs on the surface of T. cruzi and that the depletion of these molecules in the serum leads to failure of parasite clearance by the complement, indicating that the lectin pathway obviously plays an important role in host defense against this pathogen. Although it has been demonstrated that L-ficolin is involved in the host defense against T. cruzi infection, there are no studies on ficolins and chronic CD so far. In this study we aim to evaluate whether L-ficolin levels and FCN2 polymorphisms could be possible prognostic markers for susceptibility to the different clinical forms of CD.

Subjects and Samples
A total of 243 chronic CD patients attended at the Chagas Disease Ambulatory of the Clinical Hospital of the Federal University of Paraná (HC-UFPR) were investigated (Mean age 57.3  years; 58% female, 42% male; 76.5% Euro-Brazilian, 18.9% Afro-Brazilian, 4.1% Amerindian, 0.5% Asian). CD diagnosis was performed by serological and clinical examinations. The clinical history of the patients was obtained from medical records and interviews, using a standard questionnaire. Patients younger than 18 years-old, with a history of blood transfusion, recent infections or suspected non-chagasic cardiomyopathy (such as hypertensive cardiopathy) were excluded. The cardiac patients were graded according to the cardiac insufficiency classification of the American Heart Association (AHA), adapted for Chagas disease [24]. Detailed demographic and clinical characteristics of the specific CD forms are given in Table 1. A group of 305 unrelated South Brazilians with negative Chagas (anti-Trypanosoma cruzi) serology and without non-chagasic cardiomyopathy were used as controls (Mean age 41  years; 49.2% female, 50.8% male; 85.2% Euro-Brazilian, 12.8% Afro-Brazilian, 1.3% Amerindian, 0.7% Asian). Ethnic background of patients and controls was determined as previously described [25]. After signing the informed formal written consent, three ml of venous blood from each individual was collected in tubes containing ethylenediamine tetra-acetic acid (EDTA). Samples were maintained on ice after collection and during transportation to the laboratory. After centrifugation at 24uC plasma separation from whole blood was performed as quickly as possible, samples were aliquoted on ice and immediately stored at 280uC until used. Genomic DNA was extracted from peripheral whole blood using commercial kits (GFX TM Genomic Blood DNA Purification Kit, GE Healthcare, São Paulo, Brazil), according to the manufacturer's instructions.

L-fciolin Measurement
The quantification of plasma L-ficolin levels was performed in 71 controls (of which 62 were genotyped) and 160 patients (of which 158 were genotyped, being 54 indeterminate, 54 with some kind of cardiomyopathy, 19 with digestive symptoms, 29 with the cardiodigestive form and two unspecified). L-ficolin levels were determined using a commercially available enzyme linked immunosorbent assay (ELISA) kit (HK336, HycultH Biotech, Uden, Netherlands), according to the manufacturer's instructions. The detecting range of this kit is 16-1000 ng/ml of the protein.

Genotyping of FCN2 Variants
Three promoter SNPs including 2986 G.A (rs3124952), 2602 G.A (rs3124953), 24 A.G (rs17514136) and A258S in exon 8 (rs7851696) were genotyped in the patients using Fluorescence Resonance Energy Transfer (FRET) based real-time PCR assay as previously described [26], and in 135 controls through sequencing [14]. Additionally, 170 patients were genotyped for the SNPs 2986 G.A (rs3124952), 2602 G.A (rs3124953) and 24 A.G (rs17514136) using PCR amplification with sequence-specific primers (SSP) ( Table 2). In order to reconstruct the haplotypes, we performed three reactions called FCN2 Prom12, FCN2 Prom23 and FCN2Prom13, using the primers listed in table 2. In order to validate the reactions, we added two generic primers, either HGHf and HGHr or After a 5 min initial DNA denaturation step at 94uC, the PCR protocol was followed by 30 cycles of 30 sec at 94uC, 45 sec at the specific annealing temperature and 40 sec at 72uC, ending with 5 min at 72uC in the final DNA extension step. A touch-down strategy decreasing the annealing temperatures every 10 cycles was performed, using 68uC-64uC-62uC for FCN2 Prom12, 65uC-62uC-59uC for FCN2 Prom23 and 69uC-67uC-65uC for FCN2Prom13. According to a previous study, this strategy assures higher specificity to the amplification and provides a larger amount of the desired PCR product [27]. The promoter haplotypes amplified by a pair of SSPs were identified by the presence or absence of specific bands when run on a 1.2% agarose gel electrophoresis. Positive controls reassured the quality of the reactions.

Statistical Analysis
Statistics was done using the statistical package for social sciences (SPSS) version 10.0, STATA ver 9.1 and with the Graphpad Prism 5.04 software package. Data were normalized and have been analyzed by STATA. Genotype and allele frequencies were obtained by direct counting. The hypothesis of Hardy-Weinberg equilibrium and of homogeneity between genotype distributions (exact test of population differentiation of Raymond and Rousset) were executed with the ARLEQUIN software package version 3 [28]. Possible associations between FCN2 genotypes/haplotypes/variants and different clinical forms were analyzed with two tailed Fisher's exact test. Additionally multivariate analysis was executed to validate whether variables such as age, gender and ethnicity may possibly influence the clinical outcome by MANOVA. L-ficolin concentrations in the different groups were presented by the median and range and compared between the genotypes using either nonparametric Mann-Whitney or Kruskal-Wallis tests. Also correlation between L-ficolin levels to age was tested using spearman's rank correlation. Unless otherwise stated, two-tailed P-values less than 5% were considered significant. P-values for correlations are provided after correction for false detection rate (FDR).

L-ficolin Levels
Higher L-ficolin plasma concentrations were observed in the controls when compared to Chagas patients (median: 4252 ng/ml vs. 2538 ng/ml, P,0.0001). Similar trend was observed between indeterminate and symptomatic patients (median: 2888 ng/ml vs. 2519 ng/ml, one-tailed P = 0.051, two-tailed P = 0.10) (Figure 1). No significant correlation between L-ficolin levels and age has been observed using Spearman's rank correlation test (Spearman's rho = 20.1402, p = 0.08). Nevertheless L-ficolin levels were higher in the group with 45 to 59 years of age, compared with those aged above 60 years (median: 2834 ng/ml vs. 2265 ng/ml, P = 0.005). Nevertheless they were still lower than the levels found in the 45-59 years group of controls, which had a median of 4311 ng/ml (P = 0.0007) (Figure 2). There was also a trend to increased Lficolin concentration in the healthy adults (P = 0.072). Among the different groups of cardiac commitment, individuals with the less severe B form had higher L-ficolin levels than the C and D forms (median: 2981 ng/ml vs. 2094 ng/ml, P = 0.039) (Figure 2). Among the clinical parameters (listed in Table 1), only mannosebinding lectin correlated weakly with L-ficolin levels in the indeterminate patients (R2 = 0.38, P = 0.004), but not in the other disease stages (not shown). Tests for multivariate analysis revealed no significant effects of age, gender or ethnicity to L-ficolin levels.

Association of L-ficolin Levels with FCN2 Genotypes
Lower L-ficolin levels were found associated with the 258S variant in the patients (median: 2419 ng/ml vs. 2748 ng/ml in genotypes without 258S, P = 0.034). There was no difference

Discussion
In this study, L-ficolin levels were measured in plasma of the Chagas patients and controls. L-ficolin levels in the controls were within the range reported for healthy European adults in the early studies [29,30]. A decreased L-ficolin plasma level were observed in the Chagas patients when compared to controls, and a similar trend was observed between symptomatic patients and indeterminate, however, this difference was not significant. Importantly,  serum L-ficolin concentrations were also lower in Bronchiectasis patients than controls from the UK [29]. In contrast to our previous work with the same patients, where MBL levels increased gradually according to age, L-ficolin levels were higher in the group with 45 to 59 years of age, compared with those aged above 60 years. There was also a trend to increased L-ficolin concentration in the healthy adults (P = 0.072), but others did not find any relationship between L-ficolin concentration and age in healthy Scottish adults [29]. There was no significant difference between L-ficolin levels according to the gender of patients and controls, as expected [29]. In addition, the difference of L-ficolin levels was observed among the different groups of cardiac commitment. Of which individuals with the less severe B form had higher L-ficolin levels than the C and D forms. Interestingly, median L-ficolin levels are higher in acute than in chronic hepatitis B conditions [15]. They are also higher in acute malaria, compared with the levels after treatment [16], but lower in Schistosoma haematobium-infected individuals from Nigeria, compared to controls [17]. Our results contrast with those found by our group in the same patients regarding MBL levels, which were much higher (above 1000 ng/ml) in patients with chagasic cardiomyopathy and did not differ between the B C and D stages [6]. Furthermore, a weak correlation between mannose-binding lectin and L-ficolin levels was observed in the indeterminate patients, but not in the other disease stages. Allele and haplotype frequencies in the controls did not differ from those reported for European-derived populations [11,26]. No significant difference of haplotype distribution was observed between different groups, nevertheless, 24 A/G heterozygotes was observed to associate with the disease when compared cardiac patients to the controls as well as cardiac to the indeterminate patients. In addition, the AGG promoter haplotype, whose presence is obligatory in 24 A/G heterozygotes, was also associated with protection against HBV infection in Vietnamese [15], but with susceptibility to schistosomiasis in Nigeria [17]. Our group found the AGA haplotype to be associated with protection against clinical leprosy and rheumatic fever [14,18], whereas others found it associated with susceptibility to cutaneous leishmaniasis [28]. Ficolin-2 seems to be, as MBL, a double-edged sword in immunity, explaining the contrasting results in association studies with different infectious diseases. We also found an association of the allelic frequency of the 258S variant with cardiodigestive Chagas disease (Tables 3 and 4). The 258S variant was also found associated by others with earlier onset of Pseudomonas aeruginosa colonization in cystic fibrosis patients [31]. In contrast to our findings, homozygosity for this variant was associated with protection against cutaneous leishmaniasis in Syria [32] and the presence of this allele, with protection against cytomegalovirus, but not bacterial infections, after orthotopic liver transplantation [33,34]. It was not associated with respiratory tract infections in Dutch children [35], nor with invasive pneumococcal disease [36], rheumatoid arthritis [37] and Behçet's disease in Japan [38]. Interestingly, FCN2*258S, also known as FCN2-C, has increased N-acetyl-D-glucosamine (GlcNAc) binding capacity [11]  and O-linked GlcNAc moieties constitute a common epitope between cruzipain, a major T. cruzi antigen, and either myosin or other cardiac O-GlcNAc-containing proteins [39]. Higher GlcNAc binding capacity of this variant may thus correlate with complement activation and destructive deposition of membrane attack complexes (MAC) on muscle tissues of patients with cardiodigestive manifestations. At least for chronic chagasic cardiomyopathy, increased MAC deposition has been observed, in contrast to patients with idiopathic dilated cardiomyopathy or controls [40]. Additionally, our study also showed that there are associations between 258S variant and L-ficolin level in the patients. The association between low L-ficolin levels and the 258S variant has also formerly been observed in other settings: healthy Danish individuals [13], Polish children with recurrent infections [12], Gabonese malaria patients (but not controls) [16], Vietnamese HBV-infected patients and controls [15].
In conclusion, this is the first report on L-ficolin levels and FCN2 polymorphisms in patients with chronic CD. Activation of the lectin and alternative complement system is a crucial defense mechanism for the control of T. cruzi parasitemia [7] and L-ficolin has been shown to bind metacyclic trypomastigotes [7]. Decreased levels of L-ficolin in chagasic patients reflect, in part, the higher frequency of the 258S variant in those with cardiodigestive symptoms. However, since most of the FCN2 genotypes did not associate with L-ficolin levels, reduction may also be due to the disease process itself, since L-ficolin can bind to T.cruzi antigens [7], to pentraxin 3 [41] and to the C-reactive protein [42], which might affect circulating levels. In addition, inflammatory activity and the resulting tissue fibrosis could be further exacerbated by the higher capacity of GlcNAc binding to the 258S variant. L-ficolin levels and consumption could thus be used as markers of disease activity, but further experiments should be performed to prove this hypothesis. In bold: genotypes with 24 A/G heterozygosity, whose summed frequencies differ between cardiac patients and controls. Gray-shadowed: genotypes AGGA/GGAS and GGAS/GGAS (homozygote for A258S), whose summed frequencies differ between cardiodigestive patients and indeterminate patients or controls (see text). *A258S was investigated by sequencing in a subset of 135 controls. n number of individuals. doi:10.1371/journal.pone.0060237.t004