Elevated levels of autoantibodies against EXD2 and PHAX in the sera of patients with chronic thromboembolic pulmonary hypertension

While circulating autoantibodies have been detected in patients with several cardiovascular diseases, such studies have not been performed for chronic thromboembolic pulmonary hypertension (CTEPH) and pulmonary arterial hypertension (PAH). Here we investigated the production of certain auto-antibodies in CTEPH patients. Initial screening was performed in 5 CTEPH patients and 5 healthy donors (HDs) using a ProtoArray Human Protein Microarray v5.1 containing 9,375 human proteins, and we selected 34 antigens recognized by IgG antibodies more strongly in the sera of CTEPH patients than in the sera of HDs. In subsequent second/third analyses, we validated the auto-antibody level using amplified luminescent proximity homogeneous assay-linked immunosorbent assay (AlphaLISA) in 96 CTEPH patients and 96 HDs as follows: At the second screening, we used 63 crude peptides derived from those selected 34 antigens and found that the serum levels of autoantibodies for 4 peptides seemed higher in CTEPH patients than in HDs. In third analysis, we used the purified peptides of those selected in second screening and found that serum antibodies against peptides derived from exonuclease 3'-5' domain-containing 2 (EXD2) and phosphorylated adaptor for RNA export (PHAX) were significantly higher in CTEPH patients than in HDs. The serum antibody levels to these antigens were also elevated in PAH patients. The titers against EXD2 peptide decreased after surgical treatment in CTEPH patients. These autoantibodies may be useful as biomarkers of CTEPH and PAH, and further investigations may provide novel insight into the etiology.

Introduction Chronic thromboembolic pulmonary hypertension (CTEPH) is a form of pulmonary hypertension (PH) caused by persistent thromboemboli of the pulmonary arteries. Various etiological factors, including infection, inflammation, genetic susceptibilities, and insufficient angiogenesis [1], have been discussed as important pathogenetic factors [2]. However, the etiology of CTEPH is not completely understood, and disease-specific, non-invasive biomarkers have not been identified.
Circulating autoantibodies have been detected in patients with several cardiovascular diseases, such as atherosclerosis [3,4] and other cardiovascular diseases, including coronary artery diseases [5]. As a typical example, anti-phospholipid antibodies reportedly enhance the uptake of oxidized LDL by macrophages, which leads to foam cell formation [5][6][7]. Recently, we established the auto-antibody screening method using an amplified luminescent proximity homogeneous assay-linked immunosorbent assay (AlphaLISA) and found that anti-adiponectin antibody levels were significantly higher in patients with coronary artery disease, cerebral infarction and diabetes mellitus than in HDs [8]. However, autoantibodies in the context of CTEPH and pulmonary arterial hypertension (PAH) have not yet been thoroughly explored.
In the present study, we comprehensively screened autoantigens recognized by IgG antibodies in the sera of patients with CTEPH using a protein array. We then selected and identified the autoantibodies elevated in the sera of CTEPH patients and also investigated whether or not PAH patients had the same autoantibodies.

Ethical statement
The protocol for the analysis of the sera from CTEPH and PAH patients was approved by the Local Ethical Review Board of the Chiba University Graduate School of Medicine (approval number 1248). The protocol for the serum analysis in healthy donors (HDs) and the patients with sleep apnea syndrome (SAS) was also approved by the Local Ethical Review Board of the Chiba University Graduate School of Medicine (approval number 973). Written informed consent was obtained from all participating patients before sera were collected.

Patients and healthy donor sera
We collected serum samples from patients diagnosed with CTEPH and PAH at Chiba University Hospital between 2001 and 2015. Serum samples were collected from HDs who underwent annual medical checkups at Port Square Kashiwado Clinic. We also collected serum samples from patients with SAS, as previously reported [9] [10]. Each serum sample was centrifuged at 3,000 × g for 10 min at room temperature, and then the supernatant was stored at -80˚C until use (no other freeze-thaw cycles).

The ProtoArray human protein microarrays analysis
Serum samples from 5 CTEPH patients and 5 HDs were profiled on a ProtoArray Human Protein Microarrays v5.1 containing 9,375 human proteins. The serum samples were profiled at a 1:500 dilution, utilizing one ProtoArray Human Protein Microarray per sample. Alexa Fluor 647-anti-human IgG detection reagent was used to quantify the IgG level of associated autoantibodies. Pairwise comparisons were made between the two sample populations. Assays were performed by Thermo Fisher Scientific (Waltham, MA, USA) according to the manufacturer's instructions. support in the form of salaries for authors [AN, NT, TJS, AS], but did not have any additional role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the "author contributions" section. Research grants from the Respiratory Failure Research Group (H26-Intractable diseases-General-076) from the Ministry of Health, Labour and Welfare, Japan to KT. A grant to the Pulmonary Hypertension Research Group (15ek0109127h0001) from the Japan Agency for Medical Research and Development (AMED) to KT. A research grant from the Ministry of Education, Culture, Sports, Science and Technology of Japan (25461148) to NT.
Competing interests: Akira Naito is a member of the Joint Collaborative Department with Teijin Pharma, Ltd. (Tokyo, Japan) and Ono Pharmaceutical, Ltd. (Osaka, Japan). Nobuhiro Tanabe, Takayuki Jujo Sanada, and Ayako Shigeta are members of the endowed department of Actelion Pharmaceuticals Japan (Tokyo, Japan).

Epitope prediction and peptide synthesis
Possible epitope sites in the selected antigenic proteins were predicted using the software program ProPred (http://www.imtech.res.in/raghava/propred/) as described previously [11].

Statistical analyses
All statistical analyses were performed with a commercially available software program (JMP 10.0.2, Japanese version; SAS Institute Tokyo, Japan). The results are expressed as the mean ± standard deviation for continuous variables and as the number and percentage for categorical variables unless otherwise indicated. Comparisons between two groups were performed by Mann-Whitney U test, correlation analysis between two continuous variables were performed by Spearman's correlation analysis, and multiple comparisons were performed by Steel-Dwass test.

Initial screening of CTEPH-specific antigens by a protein array
Using a ProtoArray loaded with 9375 proteins, we examined the sera from 5 CTEPH patients and 5 HDs to identify CTEPH-associated antigens. We selected the antigens for which the absorbance ratio (CTEPH:HD) exceeded 1.5. As a result, the 34 proteins that were more strongly recognized by IgG antibodies in the sera of CTEPH patients than in HDs were selected (Table 1).

Second screening using crude peptides
The amino acid sequence of the 63 peptides shown in Table 2 were predicted as epitope sites of the 34 candidate antigen proteins selected in the first screening using ProtoArray.
In the second screening, these 63 peptides were synthesized and used as antigens for the analysis of the serum antibody levels by AlphaLISA using sera from 48 CTEPH patients and 48 HDs. The levels of autoantibodies for nine peptides (No. 8, 10, 11, 19, 34, 35, 43, 46, 52) seemed higher in CTEPH patients than in HDs. We then performed an additional AlphaLISA analysis for those 9 autoantibodies using sera from 96 CTEPH patients and 96 HDs. These patients' basic characteristics are summarized in Table 3.
Finally, the serum levels of autoantibodies against four peptides (No. 11, 35, 43, 52) seemed higher in CTEPH patients than in HDs (Fig 1) (A statistically significant difference in the auto-antibody level was noted for peptide No. 46 between the two groups; however, both groups had an extremely low alpha count, so the difference was not deemed to be etiologically significant).
The selected antigenic peptides were as follows:

Correlation between antibodies' titers and clinical parameters in CTEPH patients
We then analyzed correlations between the antibody levels and the clinical data of the patients, including the gender, age, body mass index, symptom duration, and hemodynamic parameters measured by right heart catheter. We found the antibody levels of bEXD2-307 showed a mild negative correlation between the arterial oxygen pressure in CTEPH patients (Table 4) and did decrease after successful pulmonary endarterectomy in CTEPH patients (Fig 3).

Discussion
We screened for autoantibodies in the sera of CTEPH patients using ProtoArray, which resulted in the selection of 34 candidate antigens (Table 1). We then synthetized 63 crude peptides predicted to be the epitope sites of those 34 candidate antigens (Table 2) and evaluated the titers of the autoantibodies for those peptides by AlphaLISA (Fig 1). Based on these results, we synthetized the four purified peptides and re-evaluated the titer of the autoantibodies against those four peptides by AlphaLISA. We ultimately found that CTEPH patients had high titers of autoantibodies for exonuclease 3'-5' domain-containing 2 (EXD2) and phosphorylated adaptor for RNA export (PHAX) than HDs (Fig 2). Furthermore, the antibody levels for EXD2 showed a mild negative correlation between the arterial oxygen pressure in CTEPH patients and decreased after successful pulmonary endarterectomy in CTEPH patients (Fig 3). As an exploratory analysis, we examined the serum antibody levels for those four purified peptides using another set of sera from patients with PAH and SAS. The serum antibody levels to EXD2 and PHAX were also elevated in patients with PAH, similar to our findings in CTEPH patients (Fig 4). Thus, EXD2 and PHAX may be markers associated with PH in general and not specific to CTEPH.
EXD2 is a 621-amino acid protein with a nuclease domain [15]. Its precise role had been unclear; however, Ronan et al. reported that EXD2 is an exonuclease essential for DNA double-strand break (DSB) resection and efficient homologous recombination. EXD2 functionally interacts with the Mre11-Rad50-Nbs1 (MRN) complex to accelerate resection via its 3'-5' exonuclease activity, which efficiently processes double-strand DNA substrates containing nicks [16]. EXD2 also reportedly has a critical functional link with FANCD2, the gene responsible for Fanconi anemia [15]. Fanconi anemia is a severe, congenital form of anemia associated with disorder of DNA repair. It should be noted that the serum anti-EXD2 antibody levels showed mild inverse correlation with arterial oxygen pressure (p = 0.0415) ( Table 4). DSBs can be induced by ischemia such as anemia and low oxygen pressure [17]. Thus, it is possible that EXD2 required for repair of DSBs is expressed to the elevated levels, which may induced the development of its autoantibodies.
PHAX is a key molecule in transporting small nuclear RNA (snRNA) and small nucleolar RNA (snoRNA) from the nucleus to the cytosol. PHAX configures a complex with RNA, CBP80, and CBP20 known as the cap binding complex (CBC), followed by binding with CRM1/RanGTP, which leads to the export of RNA into the cytoplasm [18,19]. snRNAs and snoRNAs play an essential role in the maturation of mRNAs and rRNAs, respectively [20].
As described above, EXD2 and PHAX are reported to be involved in DNA repair and the transportation of snRNA/snoRNA, respectively. The precise mechanisms underlying the production of autoantibodies for these antigens were not determined in the present study; however, the titers of those antigens were elevated in both CTEPH and PAH patients, suggesting that this autoantibody production may share a common etiology in CTEPH and PAH patients, e.g. distal pulmonary artery remodeling or right heart burden. It is reported that oxidative stress and inflammation contribute to vascular remodeling [21], and these factors are also known to favor DNA damage [22]. Indeed, high levels of DNA damage were reported to occur in PAH lungs and remodeled arteries [23] [24] [25]. These previous findings suggest that PHinduced DNA damage followed by the activation of the DNA repair system might account for the overexpression of these antigens and the resultant development of autoantibodies. Regarding DNA damage, both the polyadenylation and deadenylation of mRNA are markedly affected via CBP80 and CBP20 [26]. Thus, the functional roles of EXD2 and PHAX may be interrelated. Further investigations focused on the production of these autoantibodies might give us information on the etiology of CTEPH and PAH as well as prove the utility of these autoantibodies as a new diagnostic tool. Our study had several limitations. First, as the controls were healthy volunteer donors, potential confounding factors between patients with CTEPH and controls (e.g. the age, BMI, hypertension, diabetes, and hyperlipidemia) were not adjusted in the analysis in the current study. Second, we did not investigate the exact mechanism underlying the autoantibody production, so we were unable to determine the pathological etiology of this phenomenon. Third, this study was performed retrospectively at a single institution in Japan. Additional prospective, multicenter investigations are therefore required.
In conclusion, patients with CTEPH and PAH had higher titers for autoantibodies to EXD2 and PHAX than HDs. These autoantibodies may provide novel useful diagnostic and therapeutic tools for treating those patients.