Important roles of the human leukocyte antigen class I and II molecules and their associated genes in the autoimmune coagulation factor XIII deficiency via whole-exome sequencing analysis.

Autoimmune coagulation factor XIII deficiency is a bleeding disorder caused by the formation of autoantibodies against the coagulation factor XIII (FXIII); however, the molecular mechanism underlying this process is unknown. Therefore, in the present study, we aimed to elucidate this mechanism by performing whole-exome sequencing analysis of 20 cases of autoimmune FXIII deficiency. We identified approximately 21,788-23,916 variants in each case. In addition to their ability to activate T cells, present antigens, and immune tolerance, the candidate alleles were further narrowed down according to their allelic frequencies and the magnitude of damage caused by the substitution of amino acids. After selecting 44 candidate alleles, we investigated whether they were associated with the FXIII inhibitory titers and/or the anti-FXIII autoantibodies. We found that two polymorphisms whose variant allele frequencies were significantly lower in the patients tended to decrease FXIII inhibitory titers as the number of variant alleles increased. We also found that five polymorphisms whose variant allele frequencies were significantly higher in the patients tended to increase the levels of the anti-FXIII autoantibodies as the number of variant alleles increased. All of these polymorphisms were found in the human leukocyte antigen (HLA) class I and II molecules and their associated genes. In particular, the HLA class II molecule and its associated genes were found to be involved in the presentation of foreign antigens as well as the negative regulation of the proliferation of T-cells and the release of cytokines. Polymorphisms in the HLA class II molecules and the cytotoxic T lymphocyte antigen 4 have been reported to be associated with the development of autoantibodies in acquired hemophilia A. Therefore, we hypothesized that these polymorphisms may be associated with the development of autoantibodies in autoimmune FXIII deficiency.

Introduction

Coagulation factor XIII (FXIII) is a plasma pro-transglutaminase consisting of two catalytic A subunits (FXIII-A) and two carrier B subunits (FXIII-B). It plays an important role in maintaining hemostasis by cross-linking and stabilizing fibrin clots and increasing the resistance to mechanical stress and fibrinolysis [1,2]. FXIII deficiency results in severe bleeding diathesis, with the affected patients often requiring lifelong replacement therapy. The clinical symptoms of congenital and acquired FXIII deficiencies are very similar, ranging from multiple cutaneous mucosal bleeding to fatal intracavitary hemorrhage.

Acquired FXIII deficiency can either be an autoimmune or non-autoimmune disorder. Autoimmune FXIII deficiency is a rare autoimmune hemorrhagic disease [3-8], formerly known as autoimmune hemorrhaphilia due to anti-FXIII autoantibodies (AH13) [3] or autoimmune acquired factor XIII/13 deficiency (AAXIII/13D) [4]. Over the years, the incidence rate of autoimmune FXIII deficiency has been increasing in Japan; growing from 8 cases before 2000 to 51 cases in 2017 [3]. About half of the autoimmune FXIII deficiency cases are idiopathic in nature, while the other half are associated with underlying diseases [4]; however, all of them occur as a result of the spontaneous production of autoantibodies against endogenous FXIII. Autoimmune FXIII deficiency is characterized by a sudden onset of bleeding, which is often life‐threatening, in patients with no history of bleeding, without either prolonged prothrombin time or prolonged activated partial thromboplastin time.

Approximately 50% of the total cases of acquired hemophilia A (AHA), are considered to be idiopathic [9]. AHA is associated with high frequencies of the human leukocyte antigen (HLA) class II alleles and single nucleotide polymorphisms (SNPs) of the cytotoxic T-lymphocyte antigen 4 (CTLA-4) gene [9-13]. These genetic factors are also associated with the development of factor VIII inhibitors in patients with severe hemophilia A [14-16]. HLA class II alleles play essential roles in the presentation of factor VIII peptides to the cluster of differentiation (CD)-4+ T-lymphocytes, while CTLA-4 acts as a negative regulator of the activation of T-cells. The variants of these genes are thought to be associated with the development of AHA in combination with other genetic and/or environmental factors.

Autoimmune FXIII deficiency, like other autoimmune diseases, is thought to be caused by the disruption of the immune system. A combination of genetic and environmental factors can impair immune tolerance, leading to the development of autoantibodies against FXIII along with the ageing of the immune system in elderly individuals. Nearly 47% of the patients with autoimmune FXIII deficiency have underlying diseases, including other autoimmune diseases (17%), diabetes (9%), and cancer (6%), while the rest of the cases, with no underlying diseases are classified as idiopathic [4]. The genetic factors associated with autoimmune FXIII deficiency have not yet been identified and the etiology of autoimmune FXIII deficiency also remains unknown.

To identify the genetic factors at risk of producing the FXIII inhibitors in patients with autoimmune FXIII deficiency, we performed whole-exome sequencing (WES) analysis of 20 autoimmune FXIII deficiency cases and investigated whether polymorphisms were associated with FXIII inhibitory titers and levels of anti-FXIII autoantibodies in these patients.

Materials and methods

Materials

Recombinant FXIII-A was kindly provided by Zymogenetics (Seattle, WA, USA). Anti-FXIII-A monoclonal antibody was obtained from Prof. Reed (Massachusetts General Hospital, Boston, MA, USA). Peroxidase-conjugated anti-human IgG antibodies were purchased from MP Biomedicals (Solon, OH, USA). Tetramethylbenzidine peroxidase substrate kits were purchased from Bio-Rad Laboratories (Hercules, CA, USA).

Clinical samples

We were consulted by physicians from all over Japan, from Hokkaido in the north to Okinawa in the south, in charge of the patients with unexplained hemorrhage. For this study, we recruited patients with severe bleeding who did not present any personal or family history of bleeding, from June 2003 to Aug 2016. A total of 48 cases of autoimmune FXIII deficiency with the FXIII activities measured using the amine incorporation assay [17] below the standard value of 0.7 IU/mL and with anti-FXIII autoantibodies [4] were included in this study. Of these, we collected the peripheral blood cells from 31. This study was approved by the Institutional Review Board of Yamagata University School of Medicine. All procedures were conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all individuals.

The data about the Japanese population was obtained from "Population by Sex and Sex ratio for Prefectures—Total population, Japanese population, October 1, 2016" on the official statistics portal site of Japan (https://www.e-stat.go.jp/en/stat-search/files?page=1&layout=datalist&toukei=00200524&tstat=000000090001&cycle=7&year=20160&tclass1=000001011679&tclass2val=0).

NGS library and template preparation

Genomic DNA was extracted from the peripheral blood cells of each of the 31 patients using standard phenol/chloroform methods [18]. The lengths of the DNA fragments were measured by capillary electrophoresis using a 2200 Tape Station Instrument with a High Sensitivity D1000 Screen Tape and Reagents (Agilent Technologies Japan, Ltd., Tokyo, Japan) and the DNA integrity number was calculated [ranging from 1 (highly degraded genomic DNA) to 10 (intact genomic DNA)]. As DNA with low integrity does not provide sufficient information in WES, we excluded the samples having a DNA integrity number less than 5. Finally, 20 samples were selected and next-generation amplicon-based sequencing (Ion Proton™ System; Life Technology Japan Ltd., Tokyo, Japan) was performed. DNA samples were amplified using premixed AmpliSeq primer pools and an Ion AmpliSeq HiFi mix (Ion AmpliSeq Library Kit v2.0; Life Technology Japan Ltd.). The resulting multiplex amplicons were treated with the FuPa reagent (Life Technology Japan Ltd.) to partially digest the primer sequences and phosphorylate the amplicons. Then, the amplicons were ligated to the Ion Xpress barcode adapters (Life Technology Japan Ltd.) according to the manufacturer’s instructions. Library quantification was performed using a 2200 Tape Station Instrument with a High Sensitivity D1000 Screen Tape and Reagents. The amplified library was subjected to an emulsion polymerase chain reaction using the Ion OneTouchTM 2 Instrument with the Ion PITM Template OT2 200 Kit v3 (Life Technology Japan Ltd.). Ion sharing particles were concentrated using the Ion OneTouch ES (Life Technology Japan Ltd.) and loaded onto the Ion PI Sequencing 200 Kit v3 (Life Technology Japan Ltd.).

Ion Torrent data analysis

Signal Processing, base calls, and barcode deconvolution were performed using the Torrent SuiteTM Software v 5.0.2 (Life Technology Japan Ltd.) [19]. Alignment to the HG19 Genome Reference Consortium Human Build 37 (GRCh37) was performed using the Torrent Mapping Alignment Program (TMAP; Life Technology Japan Ltd.) and the alignment output was in the BAM format. Torrent SuiteTM Software v 5.0.2 was also used to generate relevant execution metrics, such as the total number of sequences per sample. Individual amplicon coverage metrics were calculated using the coverage Analysis plug-in in Torrent Suite™ Software. Variants were identified using the variant Caller plug-in in Torrent Suite™ Software. The output results were in the VCF format, and were filtered and selected if their coverage was 15 x or higher within the exome target region.

All variants were filtered based on the minor allelic frequencies (< 0.01) and their potential harmful effects to accurately identify the rare and harmful variants. Allelic frequencies were obtained from the Single Nucleotide Polymorphism Database (dbSNP) (https://www.ncbi.nlm.nih.gov/snp/), including several public databases, such as the Allele Frequency Aggregator (ALFA), 1000 Genomes, Exome Aggregation Consortium (ExAC), and the Genome Aggregation Database (GenomeAD)-Genomes and GenomeAD-Exomes databases. The pathogenicity of the missense changes was assessed using the following in silico predictions: PROVEAN (http://provean.jcvi.org/genome_submit_2.php) and PolyPhen-2 (http://genetics.bwh.harvard.edu/pph2/).

Allele call thresholds

A simple designation for the alleles was used based on more stringent allelic frequency thresholds, with reference to a previously reported paper [19]. SNP amplicons with allelic frequencies ≥ 90% were considered to be homozygous for that allele, while those with allelic frequencies between 10% and 90% were considered to be heterozygous. All amplicons with allelic frequencies < 10% were ignored.

Quantification of the FXIII inhibitory titer by ammonia release assay

The plasma samples of the patients were diluted 2-fold with saline and incubated with equal volumes of standard human plasma (Sysmex Corporation, Kobe, Japan) at 37°C for 2 h. The FXIII activity of the reaction mixture was measured using the Berichrom® FXIII ammonia release assay (Sysmex Corporation) according to the manufacturer’s instructions. One Bethesda unit (BU) is defined as the amount of inhibitor that results in a residual activity of 50% in the mixture.

Detection of anti-FXIII-A autoantibodies using enzyme-linked immunosorbent assay (ELISA)

ELISA was performed to detect the anti-FXIII-A autoantibodies in the plasma samples of the patients, as previously described [17]. Patient plasma (0.5 μL) was diluted 10-fold with 20 mM Tris-buffered saline (pH 7.5) containing 2% bovine serum albumin and incubated with 100 ng of recombinant FXIII-A at 37°C for 1 h. The reaction mixture was further diluted 100-fold with the same buffer, pipetted into a 96-well plate coated with an anti-FXIII-A monoclonal antibody (100 ng), and incubated at 37°C for 1 h. The plate was incubated with peroxidase-conjugated anti-human IgG. Detection of the anti-FXIII-A autoantibodies bound to recombinant FXIII-A was performed using the tetramethylbenzidine substrate as previously described [17]. The relative absorbance of autoimmune FXIII deficiency-1 at 450 nm was set as 1.0.

Detection of anti-FXIII-A autoantibodies using the immunochromatographic test (ICT)

ICT was performed to detect the anti-FXIII-A autoantibodies as previously described [20]. The plasma samples were diluted with saline and incubated with equal volumes of standard human plasma at 37°C for 5 min. Then, the anti-FXIII-A autoantibodies in the reaction mixture were detected using the in-house anti-FXIII-A monoclonal antibodies applied to a nitrocellulose strip and anti-human Ig (G+M+A)-colloidal gold conjugate. The line intensity proportional to the amount of anti-FXIII-A autoantibodies visualized using colloidal gold was read using a reader device (Fact Scan; Denken Co., Ltd., Oita, Japan) and expressed as a unit of absorbance relative to the absorbance of the positive control plasma [assigned as 1 arbitrary unit (AU)].

Statistical analysis

The variant allelic frequencies of the patients with autoimmune FXIII deficiency were calculated as follows: for chromosomes 1–22, [2*(number of cases with homozygous variant alleles) + (number of cases with heterozygotes)]/[2*(total number of cases)]; for chromosome X, [2*(number of female cases with homozygous variant alleles) + (number of female cases with heterozygotes) + (number of male cases with variant alleles)]/[2*(total number of female cases) + (total number of male cases)]; and for chromosome Y, [(number of male cases with variant alleles)/(total number of male cases)]. The frequency of the reference allele was calculated as: [1-(variant allelic frequency)]. The odds ratio (OR) was calculated as follows: [(ratio of variant allelic frequency to reference allelic frequency in all autoimmune FXIII deficiency cases)/(ratio of variant allelic frequency to reference allelic frequency registered in the database)]. In this study, we focused on polymorphisms with ORs < 0.67 or > 1.5. Comparisons of the autoimmune FXIII deficiency case distribution and relative allelic frequency risk were performed using a chi-square test or two-tailed Fisher’s exact test in the JMP software v.12.2.0 (SAS Institute, Cray, NC, USA). Statistical significance was set at P < 0.05.

Results

WES analysis of autoimmune FXIII deficiency

We performed next-generation sequencing analysis of the whole-exome of 20 autoimmune FXIII deficiency cases from 19 institutions. The distribution of autoimmune FXIII deficiency was not significantly different in different areas of Japan; however, two cases were identified in the Gunma prefecture (S1 and S2 Tables). This study included 12 males and 8 females, aged 55–88 years with a median age of 75 (Table 1). Chromosome 1 had the highest number of variants, accounting for approximately 11% of the total variants (S1). Homozygous variants accounted for approximately 42%, while heterozygous variants accounted for the remaining 58% of all variants (S1). The SNP number in each patient was observed from 21,026–23,037, with a median value of 22,415, which accounted for approximately 97% of the total variants (S1 and Table 1). The number of multiple nucleotide polymorphisms (MNPs) was 153–195 with a median value of 176. The deletion number was 268–431 with a median value of 358, and the insertion number was 217–291 with a median value of 265. Approximately 87% of these variants were derived from exons, while the remaining 13% were derived from the introns close to exons. When classified on the basis of the amino acid (AA) mutations due to exon variants, approximately 52% of the cases involved 1-AA substitutions, while 46% exhibited synonymous mutations (S1). Approximately 9% of the exon variants were predicted to be damaged, while the remaining 91% were predicted to be tolerated.

Table 1

Summary of variants in 20 autoimmune FXIII deficiency cases.

Case No.	Sex	Age (yr)	FXIII activity (U/mL)	Underling disease	SNP	MNP	Del	Ins
1	M	55	0.06	None	22,598	171	369	283
4	M	78	0.27	Unknown	22,382	184	347	291
6	M	79	<0.02	AAA	22,896	185	364	285
8	F	76	0.26	None	22,708	176	397	280
9	M	75	<0.02	AAA + IP	22,235	184	410	254
10	F	83	0.06	DM	22,830	171	318	266
11	F	78	0.09	HT + DL	21,150	153	268	217
15	M	63	0.15	Unknown	21,981	187	342	252
17	F	73	0.26	None	22,316	162	314	247
19	M	75	0.60	None	21,026	175	431	259
20	M	88	0.37	DVT	21,038	164	354	242
27	M	71	0.02	BT	23,037	195	402	282
28	M	70	<0.02	SLE + AIHA	22,570	164	312	251
29	M	65	0.04	Unknown	21,659	158	311	238
35	F	77	0.04	None	22,360	167	328	264
39	F	68	0.18	DL	21,955	187	299	245
40	F	80	0.27	DM + HT	22,448	193	375	281
42	M	71	<0.02	None	22,613	174	367	274
44	F	68	0.10	RA	22,591	180	389	268
48	M	UNK	0.62	Unknown	22,465	178	362	279

M and F indicate male and female, respectively. FXIII activity was measured using an ammine-incorporation assay. The abbreviations for underlying diseases are as follows: AAA; abdominal aortic aneurysm, IP; interstitial pneumonia; DM; diabetes mellitus, HT; hypertension, DL; dyslipidemia, DVT; deep venous thrombosis, BT; bladder tumor, SLE; systemic lupus erythematosus, AIHA; autoimmune hemolytic anemia, RA; rheumatoid arthritis. The number of SNPs, MNPs, deletions (Del), and insertions (Ins) is entered in the table.

Selection of candidate alleles associated with the development of anti-FXIII autoantibodies in autoimmune FXIII deficiency

We selected candidate alleles in three ways (Fig 1). First, we focused on the variants of the exon regions in F13A1, F13B, CTLA4, HLA-DRB1, and HLA-DQB1. Second, we focused on variants that caused AA changes including frameshift mutations in the genes associated with the Gene Ontology (GO) terms “T cell activation,” “antigen presentation,” and/or “immune tolerance.” Third, we selected the best candidate alleles to efficiently identify rare and damaging variants.

Fig 1

Data analysis and selecting candidate alleles above the criteria from genetic variants identified in whole exome sequencing of 20 autoimmune FXIII deficiency cases.

The analytical dataset was obtained by merging all the annotated variants called in the 20 autoimmune FXIII deficiency cases into a single dataset. The candidate alleles were subsequently narrowed down by three selection methods.

Exon variants in F13A1, F13B, CTLA4, HLA-DRB1, and HLA-DQB1 genes (Selection of Candidate 1)

We hypothesized that F13A1 and F13B variants might also be associated with the development of autoantibodies in patients with autoimmune FXIII deficiency as F8 is known to be associated with the development of autoantibodies in patients with AHA [13]. We identified four F13A1 variants and three F13B variants. All the F13A1 variants and one F13B variant caused a single AA change, while two of the four F13B variants caused a synonymous change. In these variants, one F13A1 variant (rs5982, p.Pro565Leu) had an OR < 0.67 against datasets of all databases from the Asian or East Asian regions. Another of F13A1 variant (rs76451285, p.Ala395Val) exhibited an OR > 1.5 except for ALFA database. The F13B variant (rs6003, p.Arg115His) was found to be homozygous in autoimmune FXIII deficiency; however, its allelic frequencies were found to be > 0.9 in all the databases.

We also investigated CTLA4, HLA-DRB1, and HLA-DQB1, whose variant allelic frequencies in patients with AHA differ from those in the control cohort [9-11]. We identified 2, 6, and 28 CTLA4, HLA-DRB1, and HLA-DQB1 variants, respectively (Tables 2 and S3). In these variants, one CTLA4 and two HLA-DQB1 variants exhibited an OR > 1.5 against the datasets from all databases of Asia or East Asia, while two HLA-DRB1 and six HLA-DQB1 variants exhibited an OR < 0.67.

Table 2

F13A1, F13B, CTLA4, HLA-DRB1, and HLA-DQB1 variants in autoimmune FXIII deficiency cases.

Chr	Pos	Ref	Var	Gene ID	Pos (AA)	Ref (AA)	Var (AA)	Type	Prediction (cutoff = 0.05)	dbSNP_ID	Cases	ExAc
												Asian
												Frequency	OR
1	197009798	A	G	F13B	602	N	N	Synonymous	Tolerated	rs5998	8.50E-01	6.71E-01	2.78
1	197030201	T	C	F13B	152	T	T	Synonymous	Tolerated	rs5997	1.00E+00	9.35E-01	ND
1	197031021	C	T	F13B	115	R	H	Single AA Change	Tolerated	rs6003	1.00E+00	9.29E-01	ND
2	204732714	A	G	CTLA4	17	T	A	Single AA Change	Tolerated	rs231775	5.75E-01	NA	ND
2	204737478	C	G	CTLA4	169	P	A	Single AA Change	Damaging	rs74808460	2.50E-02	2.00E-04	128.18
6	6152137	C	G	F13A1	652	E	Q	Single AA Change	Tolerated	rs5988	1.00E-01	1.72E-01	0.54
6	6152140	C	T	F13A1	651	V	I	Single AA Change	Tolerated	rs5987	1.00E-01	1.13E-01	0.87
6	6174866	G	A	F13A1	565	P	L	Single AA Change	Tolerated	rs5982	2.00E-01	3.30E-01	0.51
6	6197488	G	A	F13A1	395	A	V	Single AA Change	Tolerated	rs76451285	5.00E-02	3.70E-03	14.17
6	32548581	A	G	HLA-DRB1	235	F	F	Synonymous	Tolerated	rs113175445	1.50E-01	2.95E-01	0.42
6	32549525	C	G	HLA-DRB1	154	G	A	Single AA Change	Damaging	rs111965977	5.00E-02	1.90E-01	0.22
6	32549531	T	C	HLA-DRB1	152	Y	C	Single AA Change	Damaging	rs112796209	5.00E-02	1.90E-01	0.22
6	32549596	T	C	HLA-DRB1	130	V	V	Synonymous	Tolerated		1.25E-01	NA	ND
6	32549611	T	C	HLA-DRB1	125	Q	Q	Synonymous	Tolerated	rs1071752	1.25E-01	NA	ND
6	32549613	GG	CA	HLA-DRB1	125	Q	E	Single AA Change	Tolerated		1.25E-01	NA	ND
6	32629755	G	A	HLA-DQB1	217	T	I	Single AA Change	Tolerated	rs1130399	3.25E-01	2.31E-01	1.60
6	32629764	C	T	HLA-DQB1	214	S	N	Single AA Change	Tolerated	rs1130398	4.75E-01	3.89E-01	1.42
6	32629802	A	G	HLA-DQB1	201	D	D	Synonymous	Tolerated	rs1049092	7.00E-01	6.97E-01	1.02
6	32629809	C	T	HLA-DQB1	199	R	H	Single AA Change	Tolerated	rs701564	3.75E-01	NA	ND
6	32629847	A	G	HLA-DQB1	186	T	T	Synonymous	Tolerated	rs1049133	8.00E-01	8.92E-01	0.49
6	32629859	A	G	HLA-DQB1	182	N	N	Synonymous	Tolerated	rs1049130	8.00E-01	7.00E-01	1.71
6	32629868	A	G	HLA-DQB1	179	L	L	Synonymous	Tolerated	rs1049088	1.00E-01	1.94E-01	0.46
6	32629889	G	A	HLA-DQB1	172	A	A	Synonymous	Tolerated	rs1049087	5.50E-01	5.39E-01	1.05
6	32629891	C	T	HLA-DQB1	172	A	T	Single AA Change	Tolerated	rs1063323	4.75E-01	3.61E-01	1.60
6	32629904	A	G	HLA-DQB1	167	D	D	Synonymous	Tolerated	rs1049086	7.00E-01	6.94E-01	1.03
6	32629920	C	T	HLA-DQB1	162	R	Q	Single AA Change	Tolerated	rs41544112	1.25E-01	3.01E-02	4.61
6	32629935	C	G	HLA-DQB1	157	G	A	Single AA Change	Tolerated	rs1063322	4.75E-01	5.44E-01	0.76
6	32629936	C	T	HLA-DQB1	157	G	S	Single AA Change	Tolerated	rs1049107	1.00E-01	2.02E-01	0.44
6	32629955	C	T	HLA-DQB1	150	S	S	Synonymous	Tolerated	rs1063321	4.75E-01	3.41E-01	1.75
6	32629963	C	T	HLA-DQB1	148	V	I	Single AA Change	Tolerated	rs1049100	1.00E-01	2.12E-01	0.41
6	32632744	C	T	HLA-DQB1	70	A	A	Synonymous	Tolerated	rs1049082	3.25E-01	3.67E-01	0.83
6	32632745	G	A	HLA-DQB1	70	A	V	Single AA Change	Tolerated	rs1063318	2.75E-01	4.44E-01	0.47
6	32632749	A	C	HLA-DQB1	69	Y	D	Single AA Change	Tolerated		2.25E-01	NA	ND
6	32632770	A	G	HLA-DQB1	62	Y	H	Single AA Change	Tolerated		1.75E-01	NA	ND
6	32632777	C	T	HLA-DQB1	59	V	V	Synonymous	Tolerated	rs1049068	1.50E-01	1.02E-01	1.55
6	32632781	AGA	TAA/CCC	HLA-DQB1	58	L	Y/G	Single AA Change	Tolerated/Tolerated		3.50E-01/1.00E-01	NA	ND
6	32632790	C	A	HLA-DQB1	55	R	L	Single AA Change	Damaging	rs41540813	5.00E-02	1.94E-02	2.65
6	32632795	C	G	HLA-DQB1	53	T	T	Synonymous	Tolerated	rs1049079	7.50E-02	NA	ND
6	32632801	G	A	HLA-DQB1	51	N	N	Synonymous	Tolerated	rs3204373	2.25E-01	1.43E-01	1.74
6	32632818	T	G	HLA-DQB1	46	M	L	Single AA Change	Tolerated	rs1130368	2.50E-02	1.78E-01	0.12
6	32632820	C	G	HLA-DQB1	45	G	A	Single AA Change	Tolerated	rs1130375	3.75E-01	NA	ND
6	32632832	A	T	HLA-DQB1	41	F	Y	Single AA Change	Tolerated	rs9274407	5.25E-01	7.79E-01	0.31
6	32632833	A	G	HLA-DQB1	41	F	L	Single AA Change	Tolerated	rs12722107	2.25E-01	NA	ND

When the OR of autoimmune FXIII deficiency against Asia database in ExAc was > 1.5 or < 0.67, the OR was represented in bold letters. NA; not available because the variant frequency was not registered in the database. ND; not determined because the variant frequency was not registered in the Asian database. In addition, OR was not determined by following two; 1) the variant frequency registered in the database was 0.0, or 2) the variant frequency of autoimmune FXIII deficiency was 1.0.

Damaged exon variants associated with GO terms “T cell activation,” “antigen presentation,” and “immune tolerance” (Selection of Candidate 2)

We investigated 818 genes, without duplication, including 677 genes associated with the GO term, “T cell activation,” 233 with the GO term, “antigen presentation,” and 26 with the GO term, “immune tolerance.” A total of 7,055 variants, including 422 redundant genes, were identified (S4 Table). The top 11 genes, including HLA-DQB1, possessed > 100 variants (five variants/case). We summarized these 11 genes, including their ORs compared with the allelic frequencies registered in the databases (S5 Table). Among these variants, three BTNL2 variants, two SPINK5, HLA-C, and HLA-DPB1 variants, and one HLA-A, HLA-B, MICA, HLA-DQB1, and SIRPA variant exhibited ORs > 1.5 (compared with the allelic frequencies registered in the databases), while four HLA-DPB1 variants, three SPINK5 variants, two HLA-B, MICA, and HLA-DQB1 variants exhibited ORs < 0.67.

We also narrowed down the candidate variants based on the damage caused by the change in nucleotides. We identified a total of 1,029 variants, including those in 191 redundant genes (S6 Table). The top 9 genes, including MICA and HLA-C, possessed 20 or more variants (one variant/case). We also summarized these 9 genes, including their ORs and compared them with the allelic frequencies registered in the databases (S7 Table). In these variants, one variant each of ITPKB, MICA, and PSMA7 exhibited ORs > 1.5 (compared with the allelic frequencies registered in the databases), while one variant of P2RX7 had an OR < 0.67.

Damaged exon variants with allelic frequencies less than 0.01 (Selection of Candidate 3)

We narrowed down candidate alleles based on the damage described above, exhibiting variant allelic frequencies < 0.01 in the databases, but ≥ 0.05 in autoimmune FXIII deficiency cases. As a result, 64 variants met the criteria, while 440 variants were not registered in the allelic frequency database (S8 Table). A total of 504 of these were selected as candidate alleles. With respect to the number of variants per chromosome, chromosome 3 had the largest number of variants, accounting for approximately 16% of the total variants (S2). Homozygous variants accounted for about 43%, while heterozygous variants accounted for the remaining 57% of the total variants (S2). The deletion number in each patient was observed from 30 to 68 with a median of 43, which occupied approximately 50% of the total (S2). The SNP number was 13–37 with a median of 27, the MNP number was 4–10 with a median of 6, and the insertion number was 4–15, with a median of 11. When classified by codon mutations due to exon variants, the frameshift mutations were found to be the highest, about 55% (S2). We further selected five candidate alleles based on 10 or more (half or more) cases with the terms, “heterozygous” or “homozygous.”

Association of selected candidate alleles with FXIII inhibitory titers and/or levels of anti-FXIII-A autoantibodies

We then investigated the association of these variants with the FXIII inhibitory titers and/or levels of anti-FXIII-A autoantibodies. Twenty autoimmune FXIII deficiency cases were divided into three groups using each polymorphism as an index, i.e., cases with variant allele homozygotes, cases with heterozygotes, and cases with reference allele homozygotes. We compared the FXIII inhibitory titers and/or levels of anti-FXIII-A autoantibodies measured by ELISA or ICT in each genotype of 44 candidate alleles (Fig 2, Tables 3–5 and S9). Of these, 21 had significantly lower variant frequencies in autoimmune FXIII deficiency cases and 23 had significantly higher variant frequencies than those registered in the database. Among polymorphisms with significantly lower variant allele frequencies in autoimmune FXIII deficiency cases, we found that two HLA-DPB1 polymorphisms (rs1126504 and rs1126509) tended to decrease FXIII inhibitory titers as the number of variant alleles increased (Fig 2A, Tables 3 and S9). Similar results were obtained from the levels of anti-FXIII autoantibodies measured by ELISA in each genotype of 44 candidate alleles (Tables 4 and S9). On the other hand, among polymorphisms with significantly higher frequencies in autoimmune FXIII deficiency cases, we also found that five polymorphisms, HLA-B (rs1050723), MICA (rs1131897), BTNL2 (rs2076530), HLA-DQB1 (rs41544112), and HLA-DPB1 (rs1042131), tended to increase levels of anti-FXIII-A autoantibodies measured using ICT as the number of variant alleles increased (Fig 2, Tables 5 and S9).

Fig 2

FXIII inhibitory titers and anti-FXIII autoantibody levels in various variants.

A, FXIII inhibitory titers (Bethesda unit; BU) in reference allele (C/C) and heterozygous allele (C/G) at chromosome 6; 33048457 in HLA-DPB1 (rs1126504). B, Anti-FXIII autoantibody levels measured using ICT (arbitrary unit; AU) as described in “Materials and Methods” in reference allele (T/T), heterozygous allele (T/C) and homozygous allele (C/C) at chromosome 6; 32363816 in BTNL2 (rs2076530).

Table 3

FXIII inhibitors in each genotypes of MHC class I and II molecules and their associated genes.

Frequency	Chr	Pos	Ref	Var	Gene ID	dbSNP_ID	N			FXIII inhibitor (BU)
										Median
							Ref	Hetero	Homo	Ref	Hetero	Homo
High	2	204737478	C	G	CTLA4	rs74808460	19	1	0	6.90	27.00
High	6	29912856	A	T	HLA-A	rs2231119	1	3	16	6.90	0.80	9.10
High	6	31236853	G	A	HLA-C	rs1065711	0	4	16		3.60	12.30
High	6	31238155	G	A	HLA-C	rs1050328	3	6	11	21.00	2.40	9.70
High	6	31323321	G	A	HLA-B	rs1050723	15	5	0	6.40	9.70
Low	6	31324506	C	T	HLA-B	rs1050388	19	1	0	8.50	0.70
Low	6	31324549	T	C	HLA-B	rs1050570	17	2	1	0.75	5.80	51.50
High	6	31379134	C	G	MICA	rs1131897	15	5	0	6.40	9.70
Low	6	31379807	C	T	MICA	rs1051798	19	1	0	8.50	0.90
Low	6	31379823	C	G	MICA	rs1051799	19	1	0	8.50	0.90
High	6	32362741	C	T	BTNL2	rs28362677	10	8	2	8.30	7.45	25.75
High	6	32362745	G	A	BTNL2	rs28362678	10	8	2	8.30	7.45	25.75
High	6	32363816	T	C	BTNL2	rs2076530	4	11	5	3.25	8.50	9.70
Low	6	32549525	C	G	HLA-DRB1	rs111965977	18	2	0	6.65	27.35
Low	6	32549531	T	C	HLA-DRB1	rs112796209	18	2	0	6.65	27.35
Low	6	32629868	A	G	HLA-DQB1	rs1049088	16	4	0	9.10	6.10
High	6	32629920	C	T	HLA-DQB1	rs41544112	15	5	0	6.40	9.70
Low	6	32629936	C	T	HLA-DQB1	rs1049107	16	4	0	9.10	6.10
Low	6	32629963	C	T	HLA-DQB1	rs1049100	16	4	0	9.10	6.10
Low	6	32632745	G	A	HLA-DQB1	rs1063318	10	9	1	8.30	4.00	26.40
High	6	32632801	G	A	HLA-DQB1	rs3204373	12	7	1	8.30	4.00	26.40
Low	6	32632818	T	G	HLA-DQB1	rs1130368	19	1	0	6.90	51.50
Low	6	32632832	A	T	HLA-DQB1	rs9274407	7	5	8	8.50	0.90	12.30
Low	6	33048457	C	G	HLA-DPB1	rs1126504	17	3	0	9.70	0.10
Low	6	33048461	T	A	HLA-DPB1	rs1126509	17	3	0	9.70	0.10
High	6	33048542	C	T	HLA-DPB1	rs1042121	1	1	18	6.90	9.70	7.45
High	6	33048602	C	A	HLA-DPB1	rs1042131	7	5	8	6.90	27.00	5.20
Low	6	33048661	A	G	HLA-DPB1	rs1042151	19	1	0	8.50	0.00
Low	6	33048663	G	A	HLA-DPB1	rs1042153	19	1	0	8.50	0.00

If variant allelic frequency (compared with that in the database) was significantly high, the term “High” was described in column 1. If the frequency was significantly low, the term “Low” was described in column 1. The missing values were displayed in a gray box.

Table 5

Anti-FXIII-A autoantibodies measured by ICT in each genotypes of MHC class I and II molecules and their associated genes.

Frequency	Chr	Pos	Ref	Var	Gene ID	dbSNP_ID	N			Anti-FXIII-A autoantibodies (ICT, AU)
										Median
							Ref	Hetero	Homo	Ref	Hetero	Homo
High	2	204737478	C	G	CTLA4	rs74808460	19	1	0	0.46	0.61
High	6	29912856	A	T	HLA-A	rs2231119	1	3	16	0.35	0.45	0.50
High	6	31236853	G	A	HLA-C	rs1065711	0	4	16		0.40	0.50
High	6	31238155	G	A	HLA-C	rs1050328	3	6	11	0.61	0.40	0.46
High	6	31323321	G	A	HLA-B	rs1050723	15	5	0	0.35	0.93
Low	6	31324506	C	T	HLA-B	rs1050388	19	1	0	0.49	0.05
Low	6	31324549	T	C	HLA-B	rs1050570	17	2	1	0.45	0.48	0.65
High	6	31379134	C	G	MICA	rs1131897	15	5	0	0.35	0.93
Low	6	31379807	C	T	MICA	rs1051798	19	1	0	0.49	0.23
Low	6	31379823	C	G	MICA	rs1051799	19	1	0	0.49	0.23
High	6	32362741	C	T	BTNL2	rs28362677	10	8	2	0.41	0.53	0.35
High	6	32362745	G	A	BTNL2	rs28362678	10	8	2	0.41	0.53	0.35
High	6	32363816	T	C	BTNL2	rs2076530	4	11	5	0.14	0.49	0.64
Low	6	32549525	C	G	HLA-DRB1	rs111965977	18	2	0	0.48	0.50
Low	6	32549531	T	C	HLA-DRB1	rs112796209	18	2	0	0.48	0.50
Low	6	32629868	A	G	HLA-DQB1	rs1049088	16	4	0	0.47	0.51
High	6	32629920	C	T	HLA-DQB1	rs41544112	15	5	0	0.45	0.93
Low	6	32629936	C	T	HLA-DQB1	rs1049107	16	4	0	0.47	0.51
Low	6	32629963	C	T	HLA-DQB1	rs1049100	16	4	0	0.47	0.51
Low	6	32632745	G	A	HLA-DQB1	rs1063318	10	9	1	0.48	0.49	0.22
High	6	32632801	G	A	HLA-DQB1	rs3204373	12	7	1	0.48	0.49	0.22
Low	6	32632818	T	G	HLA-DQB1	rs1130368	19	1	0	0.46	0.65
Low	6	32632832	A	T	HLA-DQB1	rs9274407	7	5	8	0.35	0.45	0.53
Low	6	33048457	C	G	HLA-DPB1	rs1126504	17	3	0	0.49	0.04
Low	6	33048461	T	A	HLA-DPB1	rs1126509	17	3	0	0.49	0.04
High	6	33048542	C	T	HLA-DPB1	rs1042121	1	1	18	0.35	0.64	0.48
High	6	33048602	C	A	HLA-DPB1	rs1042131	7	5	8	0.35	0.61	0.53
Low	6	33048661	A	G	HLA-DPB1	rs1042151	19	1	0	0.49	0.04
Low	6	33048663	G	A	HLA-DPB1	rs1042153	19	1	0	0.49	0.04

Table 4

Anti-FXIII-A autoantibodies measured by ELISA in each genotypes of MHC class I and II molecules and their associated genes.

Frequency	Chr	Pos	Ref	Var	Gene ID	dbSNP_ID	N			Relative anti-FXIII-A autoantibodies against autoimmune FXIII deficiency-1 set to 1.0 (ELISA)
										Median
							Ref	Hetero	Homo	Ref	Hetero	Homo
High	2	204737478	C	G	CTLA4	rs74808460	19	1	0	1.10	1.67
High	6	29912856	A	T	HLA-A	rs2231119	1	3	16	0.90	1.29	1.17
High	6	31236853	G	A	HLA-C	rs1065711	0	4	16		0.77	1.35
High	6	31238155	G	A	HLA-C	rs1050328	3	6	11	1.56	0.70	1.24
High	6	31323321	G	A	HLA-B	rs1050723	15	5	0	1.24	1.10
Low	6	31324506	C	T	HLA-B	rs1050388	19	1	0	1.24	0.30
Low	6	31324549	T	C	HLA-B	rs1050570	17	2	1	1.05	1.32	2.04
High	6	31379134	C	G	MICA	rs1131897	15	5	0	1.24	1.10
Low	6	31379807	C	T	MICA	rs1051798	19	1	0	1.24	1.00
Low	6	31379823	C	G	MICA	rs1051799	19	1	0	1.24	1.00
High	6	32362741	C	T	BTNL2	rs28362677	10	8	2	1.25	1.17	1.04
High	6	32362745	G	A	BTNL2	rs28362678	10	8	2	1.25	1.17	1.04
High	6	32363816	T	C	BTNL2	rs2076530	4	11	5	0.92	1.24	1.05
Low	6	32549525	C	G	HLA-DRB1	rs111965977	18	2	0	1.08	1.53
Low	6	32549531	T	C	HLA-DRB1	rs112796209	18	2	0	1.08	1.53
Low	6	32629868	A	G	HLA-DQB1	rs1049088	16	4	0	1.17	1.09
High	6	32629920	C	T	HLA-DQB1	rs41544112	15	5	0	1.24	1.05
Low	6	32629936	C	T	HLA-DQB1	rs1049107	16	4	0	1.17	1.09
Low	6	32629963	C	T	HLA-DQB1	rs1049100	16	4	0	1.17	1.09
Low	6	32632745	G	A	HLA-DQB1	rs1063318	10	9	1	1.15	1.10	1.87
High	6	32632801	G	A	HLA-DQB1	rs3204373	12	7	1	1.15	1.10	1.87
Low	6	32632818	T	G	HLA-DQB1	rs1130368	19	1	0	1.10	2.04
Low	6	32632832	A	T	HLA-DQB1	rs9274407	7	5	8	1.10	1.00	1.35
Low	6	33048457	C	G	HLA-DPB1	rs1126504	17	3	0	1.29	0.04
Low	6	33048461	T	A	HLA-DPB1	rs1126509	17	3	0	1.29	0.04
High	6	33048542	C	T	HLA-DPB1	rs1042121	1	1	18	0.90	1.05	1.27
High	6	33048602	C	A	HLA-DPB1	rs1042131	7	5	8	1.05	1.67	0.70
Low	6	33048661	A	G	HLA-DPB1	rs1042151	19	1	0	1.24	0.04
Low	6	33048663	G	A	HLA-DPB1	rs1042153	19	1	0	1.24	0.04

Discussion

To the best of our knowledge, this is the first study to identify the genetic factors associated with the development of anti-FXIII autoantibodies in autoimmune FXIII deficiency. In this study, we performed WES analysis of autoimmune FXIII deficiency and narrowed down the candidate alleles based on their allelic frequencies and the magnitude of damage caused by AA substitutions. We also investigated the relationship between the 44 selected candidate alleles and the FXIII inhibitory titers and/or the levels of anti-FXIII autoantibodies via ELISA and ICT. We found that two polymorphisms tended to decrease FXIII inhibitory titers as the number of variant alleles increased and these polymorphisms were significantly lower variant allele frequencies in autoimmune FXIII deficiency cases. On the other hand, we found that five polymorphisms tended to increase levels of anti-FXIII-A autoantibodies via ICT as the number of variant alleles increased and these polymorphisms were significantly higher frequencies in autoimmune FXIII deficiency cases. All these polymorphisms were exclusively found in the HLA class I and II molecules and their associated genes.

The production of autoantibodies is considered to be caused by the disruption of the mechanism of immune tolerance [21,22]. The specific disruption mechanism has not yet been completely elucidated; however, several mechanisms have been proposed, such as abnormal apoptosis, abnormal regulatory T-cells, and molecular homology with foreign antigens. We found four F13A1 variants and one F13B variant exhibiting single AA changes. Among these variants, the allelic frequency of one variant (rs5982, p.Pro565Leu) was lower in autoimmune FXIII deficiency cases than that registered in the five databases, while that of one variant (rs76451285, p.Ala395Val) was higher. However, Ala395 was hidden inside the molecule, while Pro565 was exposed outside [23]. Of the 20 cases of autoimmune FXIII deficiency, 12 had reference alleles at the F13A1 variant (rs5982), while 8 cases exhibited heterozygosity. If the variant was associated with the development of anti-FXIII autoantibodies, the levels of the anti-FXIII autoantibodies of heterozygotes would be lower than those of the reference allele because the allelic frequency of the variant in autoimmune FXIII deficiency was lower than that registered in the databases. However, in heterozygous cases, the FXIII inhibitory titers and anti-FXIII autoantibody levels were higher than those in the reference allele cases, and there were no significant differences between these two groups. Therefore, it is unlikely that the F13A1 variants are involved in the production of autoantibodies; however, this need to be investigated further in future studies.

CTLA-4 is an important negative regulator of the immune system, exhibiting several polymorphisms associated with susceptibility to autoimmune diseases [22,24]. One of the polymorphisms (rs231775, p.Thr17Ala) was detected at a significantly higher frequency in all AHA cases [9,11]. In the present study, the allelic frequencies of this polymorphism in autoimmune FXIII deficiency cases was 0.58, which was comparable to that of the AHA cases. However, the frequency in Asia or East Asia was much higher than the global average and almost the same as that in all the autoimmune FXIII deficiency cases. On the other hand, we found other polymorphism (rs74808460) that may be associated with the development of autoantibodies in the patients. Whether these polymorphisms are actually associated with the development of autoantibodies should be confirmed by increasing the number of patients.

Various HLA alleles are known to contribute to susceptibility/protection to autoimmunity and play a definite role in the regulation of T-cell signaling [21,22]. Here, we identified two HLA-DPB1 polymorphisms (rs1126504 and rs1126509) associated with both FXIII inhibitory titers and levels of anti-FXIII autoantibodies measured by ELISA. However, these two polymorphisms might be on the same haplotype because the genotype pattern was exactly the same for each autoimmune FXIII deficiency patient.

We also identified five polymorphisms associated with the levels of anti-FXIII autoantibodies measured by ICT. These were, one HLA-B polymorphism (rs1050723), one MICA polymorphism (rs1131897), one BTNL2 polymorphism (rs2076530), one HLA-DQB1 polymorphism (rs41544112), and one HLA-DPB1 polymorphism (rs1042131). Of these, the HLA-B and MICA polymorphisms might be on the same haplotype as well as two HLA-DPB1 polymorphisms.

There are two possible causes for the differences between the autoantibody levels measured using ELISA and ICT. First, in the case of measuring autoantibody levels using ICT, especially when they are extremely high, FXIII in the sample that binds to the autoantibody becomes saturated and insufficient, resulting in insufficient quantification. Second, as the sample concentration used in ICT is much higher than that used in ELISA, it is possible that low-affinity autoantibodies are also detected. The amount of autoantibodies with higher affinity is considered to be similar to the results of ELISA. The results of the FXIII inhibitory titer and levels of autoantibodies measured using ELISA were in good agreement. Therefore, we hypothesized that the HLA-DPB1 polymorphism (rs1126504) is important for the development of autoantibodies in autoimmune FXIII deficiency. However, ICT is effective at diagnosing autoimmune FXIII deficiency with a specificity of 0.87 and sensitivity of 0.94 [20], and polymorphisms associated with the levels of autoantibodies measured using ICT cannot be excluded from the candidate alleles; therefore, the results of ICT were also considered.

HLA-B is an HLA class I molecule, and MICA is its associated gene whose name is an abbreviation for “major histocompatibility complex (MHC: synonymous with HLA) class I polypeptide-related sequence A.” HLA-DQB1 and HLA-DPB1 are HLA class II molecules and BTNL2, the associated gene of HLA class II molecules is also known as “butyrophilin-like protein 2” [25]. BTNL2 shares sequence homology with the B7 proteins that regulate T-cell activation and tolerance [26,27]. The BTNL2 mutation has been recently associated with inflammatory autoimmune diseases such as sarcoidosis and myositis [28-30]. In fact, the BTNL2 polymorphism (rs2076530) has been registered in the dbSNP database as a risk factor for sarcoidosis.

The frequencies of CTLA-4, HLA-DRB1, and HLA-DQB1 alleles in AHA cases have been previously compared to those of the healthy controls [9,11]. Here, we compared the frequencies of these genes in the autoimmune FXIII deficiency cases with the frequencies registered in the five databases. However, the frequencies of CTLA-4 and HLA-DRB1 were not significantly different between the autoimmune FXIII deficiency cases and those reported by the database. The frequencies of HLA-DQB1 polymorphism (rs41544112), which is characteristic of DQB1*06, were higher in the autoimmune FXIII deficiency cases than those reported in the database. Therefore, the genetic risk factors of autoimmune FXIII deficiency may differ from those of AHA, as the frequencies of DQB1*0502 have been reported to be higher than those of the healthy controls.

Limitations

As autoimmune FXIII deficiency is a rare disease, it was difficult to recruit a large number of patients and only 20 cases were analyzed in this study. Furthermore, due to the involvement of a rare polymorphism, it was difficult to obtain an appropriate number of patients to perform the statistical analysis. Therefore, it is necessary to verify the results of this study by conducting further investigations with larger sample sizes. In addition, HLA consists of several similar polymorphisms and pseudogenes, which cannot be easily distinguished from each other. In this study, we identified at least two cases in which two polymorphisms were thought to be on the same haplotype. Therefore, future studies can employ specific HLA-typing techniques to obtain novel results. Lastly, as the information available on this rare disease is very limited, the results obtained in this study may require further verification. However, these results may be used as a reference to elucidate the mechanism of pathogenesis of this disease in future studies.

Conclusions

In this study, we found that HLA-DPB1 polymorphisms were important for the development of autoantibodies in autoimmune FXIII deficiency, while the potential involvement of the HLA-DQB1 and BTNL2 polymorphisms was also indicated by the results. We believe that these genetic factors, along with other genetic factors and environmental factors, such as aging, together result in the development of autoimmune FXIII deficiency.

Breakdown of identified variants.

A, Number of variants per chromosomes in each case. B, Number of heterozygous and homozygous variants in each case. C, Number of each genetic variant type in each case. D, Number of each codon mutation type in each case.

(TIF)

Click here for additional data file.

Breakdown of candidate 3.

A, Number of variants per chromosomes in each case. B, Number of heterozygous and homozygous variants in each case. C, Number of each genetic variant type in each case. D, Number of each codon mutation type in each case.

(TIF)

Click here for additional data file.

Summary of FXIII tests in 20 autoimmune FXIII deficiency cases.

(XLSX)

Click here for additional data file.

Comparison of the distribution of the number of autoimmune FXIII deficiency cases used in this study and the population of each prefecture in Japan.

Japanese population data was obtained from "Population by Sex and Sex ratio for Prefectures—Total population, Japanese population, October 1, 2016" in portal site of official statistics of Japan (https://www.e-stat.go.jp/en/stat-search/files?page=1&layout=datalist&toukei=00200524&tstat=000000090001&cycle=7&year=20160&tclass1=000001011679&tclass2val=0) and modified the layout a little. Two-tailed Fisher’s exact test was used to compare differences in distribution between the Japanese population and the autoimmune FXIII deficiency cases used in this study.

(XLSX)

Click here for additional data file.

Variants in F13A1, F13B, CTLA4, HLA-DRB1, and HLA-DQB1 in autoimmune FXIII deficiency cases and its allelic frequency compared with that registered in five databases concerning total and (East) Asia.

When the OR of autoimmune FXIII deficiency against each (East) Asia database was > 1.5 or < 0.67, the OR was represented in bold letters. When the P-value was < 1.00E-8, the value was represented as "<1.00E-8" with a bold letter. Following three cases, chromosome number (Chr), position (Pos), reference nucleotide sequence (Ref), variant nucleotide sequence (Var), and gene ID (Gene ID) were represented in bold letters when the codon mutation type was single AA change. 1) In the case of the OR of autoimmune FXIII deficiency in each database was > 1.5 or < 0.67 no defect was present in all databases. 2) In the case of the OR of autoimmune FXIII deficiency to the non-defective database was all > 1.5 or < 0.67 when there were some defects. 3) In the case of all databases were missing.

(XLSX)

Click here for additional data file.

Number of codon mutations of genes associated with GO terms “T cell activation,” “antigen presentation,” or “immune tolerance” in each case.

Total number > 100 was represented by a bold letter.

(XLSX)

Click here for additional data file.

Variants of genes associated with GO terms “T cell activation,” “antigen presentation,” or “immune tolerance” in autoimmune FXIII deficiency cases.

When the OR of autoimmune FXIII deficiency against each (East) Asia database was > 1.5 or < 0.67, the OR was represented in bold letters. When the P-value was < 1.00E-8, the value was represented as "<1.00E-8" with a bold letter. Following three cases, chromosome number (Chr), position (Pos), reference nucleotide sequence (Ref), variant nucleotide sequence (Var), and gene ID (Gene ID) were represented in bold letters when the codon mutation type was single AA change. 1) In the case of the OR of autoimmune FXIII deficiency to each database was > 1.5 or < 0.67 when there was no defect in all databases. 2) In the case of the OR of autoimmune FXIII deficiency to the non-defective database was > 1.5 or < 0.67 when there were some defects. 3) In the case of all databases were missing. Polymorphisms that do not have a second (Var2) or third variant (Var3) in autoimmune FXIII deficiency are displayed in a gray box.

(XLSX)

Click here for additional data file.

Number of variant genes that thought to cause a damage in variants of S4 Table.

Total number of 20 or greater was represented by a bold letter.

(XLSX)

Click here for additional data file.

Damaging mutations of genes associated with GO terms “T cell activation,” “antigen presentation,” or “immune tolerance” in autoimmune FXIII deficiency cases.

When the OR of autoimmune FXIII deficiency against each (East) Asia database was > 1.5 or < 0.67, the OR was represented in bold letters. When the P-value was < 1.00E-8, the value was represented as "<1.00E-8" with a bold letter. Following 3 cases, chromosome number (Chr), position (Pos), reference nucleotide sequence (Ref), variant nucleotide sequence (Var), and gene ID (Gene ID) were represented in bold letters when the codon mutation type was single AA change. 1) In the case of the OR of autoimmune FXIII deficiency to each database was > 1.5 or < 0.67 when there was no defect in all databases. 2) In the case of the OR of autoimmune FXIII deficiency to the non-defective database was > 1.5 or < 0.67 when there were some defects. 3) In the case of all databases were missing.

(XLSX)

Click here for additional data file.

The list of the codon mutations that probably cause damage with variant allelic frequency < 0.01 in 20 autoimmune FXIII deficiency cases.

When the allelic frequency of the database was < 1.00E-2, the frequency was represented in bold letters. When the OR of autoimmune FXIII deficiency against each (East) Asia database was > 1.5 or < 0.67, the OR was represented in bold letters. When the P-value was < 1.00E-8, the value was represented as "<1.00E-8" with a bold letter. When the case numbers whose genotypes were “Homozygous” or “Heterozygous” were ≥ 10, chromosome number (Chr), position (Pos), reference nucleotide sequence (Ref), and variant nucleotide sequence (Var) were represented in bold letters.

(XLSX)

Click here for additional data file.

FXIII inhibitors and anti-FXIII autoantibody levels measured by ELISA and ICT in each genotypes except for MHC class I and II molecules and their associated genes.

If variant allelic frequency compared with database was significantly high, the term “High” was described in column 1. If the frequency was significantly low, the term “Low” was described in column 1. The missing value are displayed in a gray box.

(XLSX)

Click here for additional data file.

10 May 2021

PONE-D-21-03903

Whole-exome sequencing analysis of autoimmune coagulation factor XIII/13 deficiencies reveals the importance of human leucocyte antigen class I and II genes and their associated genes

PLOS ONE

Dear Dr. Osaki,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

As you can see from the comments of the reviewers, a significant revision of the manuscript is expected for this article to proceed to the next round of evaluation/ selection.

Please submit your revised manuscript by Jun 24 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Arijit Biswas

Academic Editor

PLOS ONE

Additional Editor Comments:

The paper by Osaki et al addressing autoimmune FXIII deficiencies has now been reviewed by two reviewers. Both reviewers have been highly critical of the article. While one reviewer has outright rejected the article, the other reviewer has left open the scope of a major revision. I agree with almost all the criticism raised by the reviewer. The article is inconsistent and lacking in several directions. The use of incorrect nomenclature can still be rectified and the grammar improved. However, the article suffers from poor execution with respect to statistical evaluation, explaining patient selection, clinical profile, genotypic selection etc. However, in spite of the deficiencies, owing to the data that is of interest to the FXIII specific audience I am willing to give the authors a chance to address the issues raised by the reviewers. Only and only if the authors satisfactorily address these issues will this article being continued to the next round of evaluation/ selection.

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. In your Methods section, please provide additional information about the participant recruitment method and the demographic details of your participants. Please ensure you have provided sufficient details to replicate the analyses such as:

a) the institution(s) participants were recruited from,

b) the recruitment date range (month and year),

c) a description of any inclusion/exclusion criteria that were applied to participant recruitment,

d) a table of relevant demographic details,

e) a statement as to whether your sample can be considered representative of a larger population

3. Please provide catalog numbers, sources, and dilutions of anti-F13-A monoclonal antibody and the recombinant F13-A protein used in this study.

4. Please provide citations for PROVEAN and PolyPhen-2.

5. We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match.

When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: No

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: No

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The article presents a study to deep in the genetic basis of the Acquired FXIII deficiency. The authors present the results of a cohort of 20 patients. They sequence the whole exome, and analyse different candidate variants to see if there is a correlation with the FXIII inhibitory titers, and anti-FXIII autoantibodies.

The first thing that surprise is the nomenclature used to refer to the disease. The authors, in a previous article “Recommendation for ISTH/SSC Criterion 2015 for autoimmune acquired factor XIII/13 deficiency” (Thromb Haemost 2016; 116:772-774) refer to this deficiency as AAXIII/13D instead of AiF13D. The same article explains that Coagulation factor XIII is abbreviated as FXIII, instead of F13. Having said that, and as a general evaluation, the Introduction is weak and little argued. In page 4, authors state that “AiF13D occurs as a result of the spontaneous production of autoantibodies against endogenous F13”, while in their previous article (Thromb Haemost 2016; 116:772-774) stated that “About half of AAXIII/13D cases are idiopathic, while the remaining half have an underlying disease(s)” and that there are different pathological mechanisms. In page 5, there are presented part of the results, that not have to be in the introduction.

The Material and Methods section is presented in an inaccurate manner. There is no information about the origin of the samples nor the inclusion criteria. It is not stated if they have to present low FXIII levels and no other alteration in the coagulation factors, or if they have to present anti-FXIII autoantibodies. Only “Patients with severe bleeding who did not have a personal or family history of bleeding were recruited for this study” is stated in page 6. The information about the NGS library and template preparation is insufficient, as AmpliSeq Library Kit is not a specific protocol for WES. In page 8, in the Allele call thresholds section, it is specified that “SNPs with allele frequencies of 95% or higher were considered homozygous for that allele, and those with allele frequencies of 30% to 70% were heterozygous. Allele frequencies of less than 30% and between 70% and 95% were ignored.” Those SNPs with allele frequencies between 70% and 95% are probably heterozygous or homozygous, and it is necessary to justify why they were ignored.

The Statistical analysis section is not enough, as most of the results presented are based on this type of analysis. The statistical approach described is incomplete and, based on the Results section and the information displayed on the tables, possibly incorrect.

-No clear information is given about the strategy followed to compare groups, such as which groups are being compared, or which variables are being considered. This information has to be deduced by the reader from other sections.

-The use of the Kruskal-Wallis non-parametric test is not argued. Moreover, this test is used to compare two or more independent samples and, again, based on the results presented in the tables, only comparisons of two groups are being shown. Therefore, the use of the Kruskal-Wallis test would not be suitable in these cases. Alternatively, and with the aim to conduct a basic statistical analysis to compare two groups, the Mann-Whitney non-parametric test or the parametric T-test could be applied.

-The ODDS ratio calculation is not described and the selection of the relevant thresholds (1.5 and 0.67) is not explicitly justified. In addition, empty cells in tables is not recommended: ODDS ratios out of the considered range (1.5 and 0.67) should be shown if they have been calculated, either in a main table or a supplementary.

-It is necessary to clarify the frequency of what is exactly being compared between the sample of acquired FXIII deficiency and the general population cohort. Generally, the available information from population panels is given as allelic or genotypic frequencies. For this analysis, in the Materials and Methods section, it is not described whether allelic or genotypic frequencies are calculated for the variants identified in the acquired FXIII deficiency sample and it is not specified which one of the latter is being compared. In the case that the raw number of variants detected in the acquired FXIII deficiency sample is being used, then the comparison is not adequate, as variants in this group may be either homozygous (two alternative alleles) or heterozygous (one alternative allele), which is not the frequency of alleles nor genotypes.

- In page 20, the meaning of the sentence “We selected variants whose numbers of the top two genotypes were three or more so that they could be compared by statistical processing” should be clarified and argued. What does “numbers” mean in this sentence? Number of variants? Of alleles? Of carriers (homozygous and heterozygous)? The “statistical processing” refers to the statistical power of this analysis? Then, it should be justified the suitability of “three numbers” to achieve certain statistical power and which the latter is.

About HLA, class I and II are not a gene (as stated in the abstract), it is a group of genes which polymorphisms determine the haplotype. As it is a very polymorphic locus, and there are different highly homologous pseudogenes, the alignment of the sequences of this area is very complex. For this reason, WES is not a good approximation, and it is necessary to use specific sequencing techniques and analysis software to establish the HLA typing.

Finally, the English have to be revised, as there are sentences that result difficult to understand. The quality of the figures is low and they are difficult to read.

For all the stated in this evaluation, I consider that the article is not suitable for publication in PlosOne.

Reviewer #2: The paper by Osaki et al reports the whole-exome sequencing analysis of autoimmune coagulation factor XIII (FXIII) deficiencies and reveals the association of certain leukocyte antigen class I and II genes and their associated genes with the development of autoantibodies.

The report is of potential interest, however, major corrections must be made before potential publication of this manuscript.

1/ The nomenclature of FXIII/FXIII deficiency used in the paper is unacceptable. FXIII, a coagulation factor (protein), cannot be referred to as „F13” -that is the nomenclature of the FXIII gene. Moreover, autoimmune FXIII deficiencies cannot be referred to as „AiF13D”. Also, deficiency subtypes used by the authors, e.g. Aa, Ab and B are not widely accepted. The authors must use the nomenclature that has been provided by the Scientific and Standardization Committee of the ISTH (Kohler et al, JTH 2011;9:1404-6). Autoimmune FXIII deficiency should be described according to a recent review paper published in JTH (Muszbek et al, JTH 2018; 16:822-32).

2/ Patient population is not well described. The paper must clearly describe how the patients were selected, list their symptoms, selection criteria, as well as main laboratory findings including FXIII activity and FXIII antigen levels (preferably FXIII-A2B2 and FXIII-B). Moreover, although a fraction of patients might have idiopathic autoimmune FXIII deficiency, one would expect to see association with cancer or autoimmunity, pregnancy, etc in others. Relevant information related to this must be provided.

3/ I believe that the associations described in the paper between certain tested genotypes and autoimmune FXIII deficiency were not proven to be causal in this paper,thus, the importance of the findings must remain limited in the Discussion and in the title of the manuscript.

4/ Associations described on page 20 (see „Association of selected candidates with FXIII inhibitory titers and/or anti-FXIII-A autoantibody levels” as well as Figure 2) are grossly underpowered statistically. In order to perform such analysis, statistically, at least 20-30 individuals per group must be present, otherwise conclusions may be misleading. In fact, as it is clear from Figure 2, results in all groups show an overlap, but due to the very low number of samples in some groups, results of the calculations are misleading. This section of the manuscript must be omitted or the number of patients/ group must be increased in order to provide correct statistical calculations.

5/ The manuscript should be more concise and I believe it could be significantly shortened to enhance clarity.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

24 Jun 2021

June 24, 2021

Dear Dr. Arijit Biswas:

We appreciate reviewer’s comments very much, for they help us to improve the quality of our paper considerably. According to the reviewer’s comments, the manuscript has been revised with yellow backgrounds and red letters.

Before responding point-by-point, we would like to draw your attention to two major additional changes that were revised according to the reviewers’ point out. First, we changed the title to enhance its readability from "Whole-exome sequencing analysis of autoimmune coagulation factor XIII/13 deficiencies reveals the importance of human leucocyte antigen class I and II genes and their associated genes" to " Important roles of the human leukocyte antigen class I and II molecules and their associated genes in the autoimmune coagulation factor XIII/13 deficiency via whole-exome sequencing analysis". Second, we added “Limitation” in the “Discussion” section to prevent readers from misunderstanding.

Response to Journal Requirements

Comment 1: Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Reply to Comment 1: We followed the rules and worked carefully to ensure a style suitable for publication.

Comment 2: In your Methods section, please provide additional information about the participant recruitment method and the demographic details of your participants. Please ensure you have provided sufficient details to replicate the analyses such as:

Comment 2-1: a) the institution(s) participants were recruited from

Reply to Comment 2-1: According to Editor’s comment, we mentioned the institution in the “Results” section under the heading “WES analysis of AiF13D” (p.14, lane 9), but could not describe details without the consent of the patient. However, of the 20 patients, only 2 patients from the same institution were relatively unbiased.

Comment 2-2: b) the recruitment date range (month and year)

Reply to Comment 2-2: According to Editor’s comment, we described the recruitment date range (month and year) in the “Materials and methods” section under the heading “Clinical samples” (p.7, lane 4).

Comment 2-3: c) a description of any inclusion/exclusion criteria that were applied to participant recruitment

Reply to Comment 2-3: According to Editor’s comment, we described the inclusion/exclusion criteria in “Materials and methods” section under headings “Clinical samples” (p.7, lanes 2–7) and “NGS library and template preparation” (p.8, lanes 1–10).

Comment 2-4: d) a table of relevant demographic details

Reply to Comment 2-4: According to Editor’s comment, we obtained the table of relevant demographic details from "Population by Sex and Sex ratio for Prefectures - Total population, Japanese population, October 1, 2016" in portal site of official statistics of Japan (https://www.e-stat.go.jp/en/stat-search/files?page=1&layout=datalist&toukei=00200524&tstat=000000090001&cycle=7&year=20160&tclass1=000001011679&tclass2val=0) (p.7, lanes 11–15) and added S2 Table.

Comment 2-5: e) a statement as to whether your sample can be considered representative of a larger population

Reply to Comment 2-5: Following the Editor’s comment, we mentioned whether our sample can be considered representative of a larger population in the “Results” section under the heading “WES analysis of AiF13D” (p.14, lanes 9–11). As stated in the manuscript, the distribution of AiF13D cases by region is not significantly different from the distribution of Japanese, so we considered that the sample was representative of a larger population.

Comment 3-1: Please provide catalog numbers, sources, and dilutions of anti-F13-A monoclonal antibody used in this study.

Reply to Comment 3-1: We could not provide catalog number of the anti-FXIII/13-A monoclonal antibody because it was not a commercial available. However, we described the source of the antibody in the “Materials and methods” section under the heading “Materials” (p.6, lanes 11–12). We also described the dilutions of the antibody in the “Results” section under the heading “Detection of anti-FXIII/13-A autoantibodies using enzyme-linked immunosorbent assay (ELISA)” (p.12, lane 3).

Comment 3-2: Please provide catalog numbers, sources, and dilutions of the recombinant F13-A protein used in this study.

Reply to Comment 3-2: We could not provide catalog number of the recombinant FXIII/13-A because it was not a commercial available. However, we described the source of the recombinant protein in the “Materials and methods” section under the heading “Materials” (p.6, lane 10). We had described the dilutions of the recombinant protein in the “Results” section under the heading “Detection of anti-FXIII/13-A autoantibodies using enzyme-linked immunosorbent assay (ELISA)” in the original manuscript (p.12, lanes 1–2).

Comment 4: Please provide citations for PROVEAN and PolyPhen-2.

Reply to Comment 4: According to Editor’s comment, we described both citations for PROVEAN and PolyPhen-2 in the “Materials and methods” section under the heading “Ion Torrent data analysis” (p.10, lanes 11–12).

Comment 5: We note that the grant information you provided in the ‘Funding Information’ and ‘Financial Disclosure’ sections do not match. When you resubmit, please ensure that you provide the correct grant numbers for the awards you received for your study in the ‘Funding Information’ section.

Reply to Comment 5: We have confirmed that we are providing the correct grant number for the award we received for our study in the ‘Funding Information’ section.

Response to Reviewer #1

Comment 1-1: The first thing that surprise is the nomenclature used to refer to the disease. The authors, in a previous article “Recommendation for ISTH/SSC Criterion 2015 for autoimmune acquired factor XIII/13 deficiency” (Thromb Haemost 2016; 116:772–774) refer to this deficiency as AAXIII/13D instead of AiF13D.

Reply to Comment 1-1: As pointed out by Reviewer # 1, abbreviations different from the previous article might confuse the reader, but since the disease name designated as an intractable disease by the Japanese Ministry of Health, Labour and Welfare was AiF13D, we used AiF13D as the abbreviation. We explained in the "Introduction" section of this manuscript (p.4, lanes 5–10).

Comment 1-2: The same article explains that Coagulation factor XIII is abbreviated as FXIII, instead of F13.

Reply to Comment 1-2: As Reviewer #1 pointed out, we abbreviated coagulation factor XIII as FXIII/13 instead of F13. The reason why we did not use FXIII is to avoid confusion with FVIII and FXII for medical safety measures, although we wrote in the manuscript (p.3, lanes 12–13).

Comment 1-3: Having said that, and as a general evaluation, the Introduction is weak and little argued.

Reply to Comment 1-3: As Reviewer # 1 pointed out, “Introduction” was weak, so we clarified the purpose of this study (p.6, lanes 3–6).

Comment 2: In page 4, authors state that “AiF13D occurs as a result of the spontaneous production of autoantibodies against endogenous F13”, while in their previous article (Thromb Haemost 2016; 116:772–774) stated that “About half of AAXIII/13D cases are idiopathic, while the remaining half have an underlying disease(s)” and that there are different pathological mechanisms.

Reply to Comment 2: In fact, as Reviewer # 1 says, about half of AiF13D cases were idiopathic and the other half had underlying disease (Table 1), but we did not know how underlying disease affects antibody production. Therefore, we assumed that AiF13D occurs as a result of the production of autoantibodies against endogenous FXIII/13, and a supplementary explanation was added to "Introduction" (p.4, lanes 12–15).

Comment 3: In page 5, there are presented part of the results that not have to be in the introduction.

Reply to Comment 3: Following the Reviewer #1’s suggestions, we have removed some of the results from the last paragraph of “Introduction” section and replaced with the purpose of this study (p.6, lanes 3–6).

Comment 4-1: The Material and Methods section is presented in an inaccurate manner. There is no information about the origin of the samples nor the inclusion criteria.

Reply to Comment 4-1: As Reviewer #1 pointed out, information about the origin of the samples and the inclusion criteria were inadequate, so we added them. We already listed the sex and age of the patient in Table 1 of original manuscript, but newly added the patient’s medical institution (prefecture) to S1 Table. We also described the inclusion criteria in “Materials and methods” section under headings “Clinical samples” (p.7, lanes 2–7) and “NGS library and template preparation” (p.8, lanes 1–10).

Comment 4-2: It is not stated if they have to present low FXIII levels and no other alteration in the coagulation factors, or if they have to present anti-FXIII autoantibodies. Only “Patients with severe bleeding who did not have a personal or family history of bleeding were recruited for this study” is stated in page 6.

Reply to Comment 4-2: Following the Reviewer #1’s recommendation, we added the patient's underlying disease and its FXIII/13 activity to Table 1, and also added the FXIII/13-A, F-XIII/13-B, and FXIII/13-A2B2 antigen level, FXIII/13 inhibitory titer, and anti-FXIII/13 autoantibody level to S1 Table.

Comment 5: The information about the NGS library and template preparation is insufficient, as AmpliSeq Library Kit is not a specific protocol for WES.

Reply to Comment 5: As Reviewer #1 pointed out, we explained the NGS library and template preparation in detail (p.8, lanes 10–16).

Comment 6: In page 8, in the Allele call thresholds section, it is specified that “SNPs with allele frequencies of 95% or higher were considered homozygous for that allele, and those with allele frequencies of 30% to 70% were heterozygous. Allele frequencies of less than 30% and between 70% and 95% were ignored.” Those SNPs with allele frequencies between 70% and 95% are probably heterozygous or homozygous, and it is necessary to justify why they were ignored.

Reply to Comment 6: As Reviewer #1 points out, between 70% and 95% are probably heterozygous or homozygous. Following a previous report (Daniel R et al. Forensic Sci Int Genet. 2015;14:50–60), we changed 10% to 90% to be heterozygous, 90% or greater to be homozygous, and less than 10% to be ignored (p.10, lane 16–p.11, lane 2). According to this criterion, the number of variants changed a little, so I changed it in Table 1, S4, S6, and S8 Tables. The relevant parts of the manuscript have been altered accordingly.

Comment 7: The Statistical analysis section is not enough, as most of the results presented are based on this type of analysis. The statistical approach described is incomplete and, based on the Results section and the information displayed on the tables, possibly incorrect. No clear information is given about the strategy followed to compare groups, such as which groups are being compared, or which variables are being considered. This information has to be deduced by the reader from other sections.

Reply to Comment 7: As Reviewer #1 pointed out, there was not enough information. Therefore, we added a detailed explanation to the subsection entitled “Statistical analysis” in the “Materials and methods” section (p.13, lane 16–p.14, lane 4). AiF13D case distribution comparison and relative risk allele frequency comparisons were performed using a chi-square test or two-tailed Fisher's exact test. Comparisons of FXIII/13 inhibitory titer or levels of anti-FXIII/13 autoantibodies between two allele groups were performed using the Mann-Whitney U-test.

Comment 8: The use of the Kruskal-Wallis non-parametric test is not argued. Moreover, this test is used to compare two or more independent samples and, again, based on the results presented in the tables, only comparisons of two groups are being shown. Therefore, the use of the Kruskal-Wallis test would not be suitable in these cases. Alternatively, and with the aim to conduct a basic statistical analysis to compare two groups, the Mann-Whitney non-parametric test or the parametric T-test could be applied.

Reply to Comment 8: As Reviewer #1 pointed out, we changed "Kruskal-Wallis test" to "Mann-Whitney U-test" (p.14, lanes 1–3).

Comment 9-1: The ODDS ratio calculation is not described and the selection of the relevant thresholds (1.5 and 0.67) is not explicitly justified.

Reply to Comment 9: According to the opinion of Reviewer #1, we described the odds ratio calculation (p.13, lanes 12–15) and thresholds (p.13, lanes 15–16) in “Statistical analysis” in “Materials and methods” section.

Comment 9-2: In addition, empty cells in tables is not recommended: ODDS ratios out of the considered range (1.5 and 0.67) should be shown if they have been calculated, either in a main table or a supplementary.

Reply to Comment 9-2: The empty cells in Table 2 and S3, S5, S7, and S8 Tables does not mean out of the considered range (1.5 and 0.67), and we could not calculate the odds ratio because variant frequency of the SNP was not registered in the database or the variant allele frequency registered in the database was 0 or the frequency of AiF13D was 1. In the tables, we added NA instead of the empty cells.

Comment 10-1: It is necessary to clarify the frequency of what is exactly being compared between the sample of acquired FXIII deficiency and the general population cohort.

Reply to Comment 10-1: Follow the advice of Reviewer #1, we clarified that it is a comparison with the allelic frequencies in “Statistical analysis” in “Materials and methods” section (p.13, lane 16–p.14, lane 1).

Comment 10-2: Generally, the available information from population panels is given as allelic or genotypic frequencies. For this analysis, in the Materials and Methods section, it is not described whether allelic or genotypic frequencies are calculated for the variants identified in the acquired FXIII deficiency sample and it is not specified which one of the latter is being compared.

Reply to Comment 10-2: As Reviewer #1 stated, we did not clarify either the allelic frequencies or the genotypic frequencies, so we clarified it as the allelic frequencies (p.13, lane 16–p.14, lane 1).

Comment 10-3: In the case that the raw number of variants detected in the acquired FXIII deficiency sample is being used, then the comparison is not adequate, as variants in this group may be either homozygous (two alternative alleles) or heterozygous (one alternative allele), which is not the frequency of alleles nor genotypes.

Reply to Comment 10-3: We calculated the allele frequency of the variant as (2n+m)/40 (20; total patient number x 2; number of alleles per person, except chromosomes X and Y), where n is the number of patients with variant homozygotes and m is the number of patients with heterozygotes. We added the calculation method for variant allelic frequencies to the subsection entitled “Statistical analysis” in the “Materials and methods” section (p.13, lanes 5–12).

Comment 11-1: In page 20, the meaning of the sentence “We selected variants whose numbers of the top two genotypes were three or more so that they could be compared by statistical processing” should be clarified and argued. What does “numbers” mean in this sentence? Number of variants? Of alleles? Of carriers (homozygous and heterozygous)?

Reply to Comment 11-1: We meant the number of carriers in each carrier group (heterozygote, and variant and reference allele homozygotes). However, it is difficult to understand, so we changed the expression (p.23, lanes 9–12).

Comment 11-2: The “statistical processing” refers to the statistical power of this analysis?

Reply to Comment 11-2: Yes, we meant the “statistical processing” refers to the statistical power of this analysis.

Comment 11-3: Then, it should be justified the suitability of “three numbers” to achieve certain statistical power and which the latter is.

Reply to Comment 11-3: As Reviewer # 1 mentions, there is no justification for the “three numbers”, so we analyzed all applicable candidates (p.23, lanes 12–14).

Comment 12-1: About HLA, class I and II are not a gene (as stated in the abstract), it is a group of genes which polymorphisms determine the haplotype.

Reply to Comment 12-1: As Reviewer #1 pointed out, HLA class I and II are not genes, so we corrected the notation (Title, p.3, lane 1, Title of Tables 3–5, p.29, lane 11, p.33, lanes 1–5, and Title of S9 Table).

Comment 12-2: As it is a very polymorphic locus, and there are different highly homologous pseudogenes, the alignment of the sequences of this area is very complex. For this reason, WES is not a good approximation, and it is necessary to use specific sequencing techniques and analysis software to establish the HLA typing.

Reply to Comment 12-2: We are grateful to Reviewer #1 for pointing out. We described the difficulty of aligning the HLA region with WES in “Limitation” with reference to the pointed out content (p.34, lanes 9–13).

Comment 13-1: Finally, the English have to be revised, as there are sentences that result difficult to understand.

Reply to Comment 13-1: As shown in the attachment (Certificate_of_editing-AKNOS_29.pdf), we have already asked Editage (www.editage.com) to edit the original manuscript to make it easier to understand in English. I requested that the revised points be edited separately.

Comment 13-2: The quality of the figures is low and they are difficult to read.

Reply to Comment 13-2: As Reviewer #1 pointed out, we increased the resolution of the figures.

Response to Reviewer #2

Comment 1: The nomenclature of FXIII/FXIII deficiency used in the paper is unacceptable.

Comment 1-1: FXIII, a coagulation factor (protein), cannot be referred to as “F13” that is the nomenclature of the FXIII gene.

Reply to Comment 1-1: As Reviewer #2 pointed out, we abbreviated coagulation factor XIII as FXIII/13 instead of F13. The reason why we did not use FXIII is to avoid confusion with FVIII and FXII for medical safety measures, although we wrote in the manuscript (p.3, lanes 12–13).

Comment 1-2: Moreover, autoimmune FXIII deficiencies cannot be referred to as “AiF13D”.

Reply to Comment 1-2: As Reviewer # 2 pointed out, autoimmune FXIII deficiencies may not be described as "AiF13D", but since the disease name designated as an intractable disease by the Japanese Ministry of Health, Labor and Welfare was AiF13D, we used AiF13D as the abbreviation. We explained in the "Introduction" section of this manuscript (p.4, lanes 5–10).

Comment 1-3: Also, deficiency subtypes used by the authors, e.g. Aa, Ab and B are not widely accepted. The authors must use the nomenclature that has been provided by the Scientific and Standardization Committee of the ISTH (Kohler et al, JTH 2011;9:1404–6). Autoimmune FXIII deficiency should be described according to a recent review paper published in JTH (Muszbek et al, JTH 2018; 16:822–32).

Reply to Comment 1-3: Following the instructions of Reviewer #2, we removed the description of Aa, Ab and B.

Comment 2-1: Patient population is not well described. The paper must clearly describe how the patients were selected, list their symptoms, selection criteria, as well as main laboratory findings including FXIII activity and FXIII antigen levels (preferably FXIII-A2B2 and FXIII-B).

Reply to Comment 2-1: As Reviewer #2 stated, patient information is important in interpreting the results. We described the selection criteria in “Materials and methods” section under headings “Clinical samples” (p.7, lanes 2–7) and “NGS library and template preparation” (p.8, lanes 1–10). We added the FXIII/13 activity to Table 1, and listed FXIII/13 antigen levels, including FXIII/13-A, FXIII/13-B, and FXIII/13-A2B2, in S1 Table.

Comment 2-2: Moreover, although a fraction of patients might have idiopathic autoimmune FXIII deficiency, one would expect to see association with cancer or autoimmunity, pregnancy, etc in others. Relevant information related to this must be provided.

Reply to Comment 2-2: As Reviewer #2 pointed out, information on the underlying disease is important. We added the underlying disease to Table 1.

Comment 3: I believe that the associations described in the paper between certain tested genotypes and autoimmune FXIII deficiency were not proven to be causal in this paper, thus, the importance of the findings must remain limited in the Discussion and in the title of the manuscript.

Reply to Comment 3: Following the instructions of Reviewer #2, we changed the title from "Whole-exome sequencing analysis of autoimmune coagulation factor XIII/13 deficiencies reveals the importance of human leucocyte antigen class I and II genes and their associated genes" to " Important roles of the human leukocyte antigen class I and II molecules and their associated genes in the autoimmune coagulation factor XIII/13 deficiency via whole-exome sequencing analysis". We also added “Limitations” in “Discussion” section (p.34, lanes 5–16).

Comment 4-1: Associations described on page 20 (see “Association of selected candidates with FXIII inhibitory titers and/or anti-FXIII-A autoantibody levels” as well as Figure 2) are grossly underpowered statistically. In order to perform such analysis, statistically, at least 20–30 individuals per group must be present, otherwise conclusions may be misleading.

Reply to Comment 4-1: It is understandable that a total of 20 cases is not enough to perform a comparative analysis. However, as added to "Limitations" (p.34, lanes 5–9), AiF13D cases are rare diseases, so it is difficult to secure specimens. In particular, the polymorphism shown here has a polymorphism frequency of 0.26-folds lower than that of healthy subjects, so the number has decreased to three cases. However, we found it worthwhile to share information that at least three patients with alleles, which were more common in healthy subjects than patients, had lower inhibitory titers.

Comment 4-2: In fact, as it is clear from Figure 2, results in all groups show an overlap, but due to the very low number of samples in some groups, results of the calculations are misleading.

Reply to Comment 4-2: As Reviewer # 2 pointed out, Figure 2B is the same polymorphism as Figure 2A, so we omitted Figure 2B.

Comment 4-3: This section of the manuscript must be omitted or the number of patients/ group must be increased in order to provide correct statistical calculations.

Reply to Comment 4-3: As we wrote in the “Reply to Comment 4-1”, we left this section for informational purposes, although it was limited in number.

Comment 5: The manuscript should be more concise and I believe it could be significantly shortened to enhance clarity.

Reply to Comment 5: We eliminated duplication. For example, we deleted the last paragraph of “Introduction” section due to duplicate results. We have already asked Editage (www.editage.com) to edit the original manuscript to make it more concise and clearer, as shown in the attachment (Certificate_of_editing-AKNOS_29.pdf). We requested that the revised points be edited separately.

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

17 Aug 2021

PONE-D-21-03903R1

Important roles of the human leukocyte antigen class I and II molecules and their associated genes in the autoimmune coagulation factor XIII/13 deficiency via whole-exome sequencing analysis

PLOS ONE

Dear Dr. Osaki,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, and taking into account both the points made by both the current reviewers, I feel that the changes made in the first round of revision are quite unsatisfactory (especially with respect to the nomenclature /statistical analysis aspects raised by Reviewer2 who in fact has suggested rejection of the manuscript). However, I would like to give the authors one last chance to improve upon the manuscript especially in the lines of the points raised by Reviewer2 after which I will make a final decision if to accept the article or not. Therefore, I invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Oct 01 2021 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols. Additionally, PLOS ONE offers an option for publishing peer-reviewed Lab Protocol articles, which describe protocols hosted on protocols.io. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols.

We look forward to receiving your revised manuscript.

Kind regards,

Arijit Biswas

Academic Editor

PLOS ONE

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #2: (No Response)

Reviewer #3: (No Response)

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #2: No

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #2: No

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #2: Yes

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #2: Unfortunately, this Reviewer is not satisfied with the revision provided by the authors.

The nomenclature used throughout the paper is still unacceptable and if the authors are not willing to comply with the nomenclature provided by the Scientific and Standardization Committee of the ISTH (Kohler et al, JTH 2011;9:1404–6), I strongly believe that they should not submit and try to publish in Plos One. I think that this a major point as by not using the correct nomenclature, confusion is generated within the scientific community and the paper will not advance this field. Moreover, my other points (e.g. omission of statistically underpowered analysis) were not taken fully into consideration, and only minor changes (e.g. adding a limitation section) were introduced in the paper. I believe that the paper in its current form is misleading both due to the nomenclature used but also due to contents that are not supported by sound statistical analysis and thus, unfortunately, I cannot agree to the publication of this manuscript in its current form.

Reviewer #3: comments:

(1) Please could the authors provide the information about the antigen epitope of anti-FXIII/13 antibody(from Prof.Reed's gift).

(2) For the quantification of FXIII/13 inhibitory titer,the healthy individuals controls should be included.

(3)Is it only limited in FXIII/13 A chain for the detection of against FXIII/13 autoantibodies.Please could the authors comment or illustrate on why autoantibody againist FXIII/13 is not involved in FXIII/13 B chains.

(4)Please could the authors supplement the diagnostic criteria on acquired autoimmune FXIII/13 deficiency or reference in your study series.

(5)In your present studies,you used two methods(ELISA & IST) for evaluating FXIII/13 autoantibodies levels.Could you please determine which method is more sensitive and specific in identifying autoantibodies against FXIII/13;How about the correlation of the both methods.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

27 Aug 2021

August 27, 2021

Dear Dr. Arijit Biswas:

We appreciate reviewer’s comments very much, for they help us to improve the quality of our paper considerably. According to the reviewer’s comments, the manuscript has been revised with yellow backgrounds and red letters.

Before responding point-by-point, we would like to draw your attention to two major additional changes that were revised according to the reviewers’ point out. First, we changed the abbreviations to comply with the nomenclature provided by the Scientific and Standardization Committee of the ISTH. Second, we accept the Reviewer # 2's criticism concerning statistically underpowered analysis, and significantly rewritten “Association of selected candidate alleles with FXIII inhibitory titers and/or levels of anti-FXIII-A autoantibodies” in the “Results” section.

Response to Reviewer #2

Comment 1–1: The nomenclature used throughout the paper is still unacceptable and if the authors are not willing to comply with the nomenclature provided by the Scientific and Standardization Committee of the ISTH (Kohler et al, JTH 2011;9:1404–6), I strongly believe that they should not submit and try to publish in Plos One.

Reply to Comment 1–1: Following the comments of Reviewer #2, we changed the abbreviations to comply with the nomenclature provided by the Scientific and Standardization Committee of the ISTH (Kohler et al, JTH 2011;9:1404–6). Specifically, FXIII/13 was changed to FXIII (94 places including p. 2, lane 3). In addition, AiF13D and AiF8D were changed to autoimmune FXIII deficiency (74 places including p. 2, lane 2) and AHA (9 places including p.7, lane 15).

Comment 1–2: I think that this a major point as by not using the correct nomenclature, confusion is generated within the scientific community and the paper will not advance this field.

Reply to Comment 1–2: Reviewer # 2's comments are justified. So, to avoid confusion in the scientific community, we changed the abbreviations according to the correct nomenclature as described in “Reply to Comment 1–1”.

Comment 2: Moreover, my other points (e.g. omission of statistically underpowered analysis) were not taken fully into consideration, and only minor changes (e.g. adding a limitation section) were introduced in the paper.

Reply to Comment 2: We accept the Reviewer # 2's criticism concerning statistically underpowered analysis, and have deleted the P-value in Tables 3–5, S9 Table, and Fig. 2. In addition, we made the expression a little less than the assertive tone in “Abstract” (p. 2, lanes 12–17), “Results” (p. 23, lane 15–p. 24, lane 9) and “Discussion” (p. 28, lanes 5–11). The relevant parts of the rest of the manuscript have been altered accordingly (p. 14, lane 1 & legends of Tables 3–5 and S9 Table).

Comment 3: I believe that the paper in its current form is misleading both due to the nomenclature used but also due to contents that are not supported by sound statistical analysis and thus, unfortunately, I cannot agree to the publication of this manuscript in its current form.

Reply to Comment 3: Following the comments of Reviewer #2, we changed the abbreviations to comply with the nomenclature provided by the Scientific and Standardization Committee of the ISTH. In addition, as described in “Reply to Comment 2”, we accept the Reviewer # 2's criticism concerning statistically underpowered analysis, and have deleted the P-value in Tables 3–5, S9 Table, and Fig. 2 to avoid misleading. We believe that the revised manuscript is now suitable for publication.

Response to Reviewer #3

Comment 1: Please could the authors provide the information about the antigen epitope of anti-FXIII/13 antibody (from Prof. Reed's gift).

Reply to Comment 1: Unfortunately, we do not have information about the antigen epitope of the anti-FXIII antibody. However, since this antibody recognizes FXIIIa and FXIII-A2B2, it is considered that it does not recognize the activation peptide or the binding site with FXIII-B.

Comment 2: For the quantification of FXIII/13 inhibitory titer, the healthy individual controls should be included.

Reply to Comment 2: I would like to include data for healthy individual controls in order to increase the number, but the inhibitory titer of healthy individual controls cannot be calculated. One Bethesda unit is defined as the amount of inhibitor in 1 mL of plasma that will neutralize 50% of the FXIII activity (residual activity = 50%). The numbers of Bethesda units are calculated according to the dilution of the patient's plasma. To determine the BU/ mL, it is necessary to obtain the residual activity between 25 and 75%. However, FXIII activity of healthy individual controls is generally >70%, and the FXIII activity of 1:1 mixture of standard plasma and healthy individual control plasma is >75%. Therefore, we cannot calculate FXIII inhibitory titer of healthy individual controls.

Comment 3: Is it only limited in FXIII/13 A chain for the detection of against FXIII/13 autoantibodies. Please could the authors comment or illustrate on why autoantibody against FXIII/13 is not involved in FXIII/13 B chains.

Reply to Comment 3: Ninety-five percent of the cases with the autoimmune FXIII deficiency we have identified were due to autoantibodies against FXIII-A and the remaining 5% were due to autoantibodies against FXIII-B. As described in S1 Table, case No. 48 used in this study was due to autoantibodies against FXIII-B. However, in these cases, the decrease in the level of FXIII antigen due to clearance is the cause of the decrease in activity, and since there is no neutralizing activity, there is almost no inhibitory titer.

Comment 4: Please could the authors supplement the diagnostic criteria on acquired autoimmune FXIII/13 deficiency or reference in your study series.

Reply to Comment 4: All patients with acquired autoimmune FXIII deficiency met the diagnostic criteria from the ISTH/SSC FXIII/Fibrinogen Subcommittee (Ichinose A et al, Thromb Haemost 2016; 116: 772–774). Following the comments of Reviewer #3, we supplemented the reference in the “Clinical samples” section of the “Materials and methods” section (p. 7, lane 7).

Comment 5: In your present studies, you used two methods (ELISA & IST) for evaluating FXIII/13 autoantibodies levels. Could you please determine which method is more sensitive and specific in identifying autoantibodies against FXIII/13; How about the correlation of the both methods.

Reply to Comment 5: As mentioned in the “Discussion” section (p. 31, lane 5–p. 32, lane 2), we believe that the ELISA method is more specific in identifying autoantibodies against FXIII-A. In fact, the logarithm of the FXIII inhibitory titer was better correlated with the ELISA method (R=0.86, P<0.0001) than with the ICT method (R=0.63, P=0.0037). However, for patients with many low-affinity antibodies to FXIII-A, the ICT method may be more sensitive. The correlation between the two methods was moderately good (R=0.61, P=0.0043).

Submitted filename: Response to Reviewers.docx

Click here for additional data file.

31 Aug 2021

Important roles of the human leukocyte antigen class I and II molecules and their associated genes in the autoimmune coagulation factor XIII deficiency via whole-exome sequencing analysis

PONE-D-21-03903R2

Dear Dr. Osaki,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Arijit Biswas

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

2 Sep 2021

PONE-D-21-03903R2

Important roles of the human leukocyte antigen class I and II molecules and their associated genes in the autoimmune coagulation factor XIII deficiency via whole-exome sequencing analysis

Dear Dr. Osaki:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Arijit Biswas

Academic Editor

PLOS ONE