Article Text

Pathogenic UBA1 variants associated with VEXAS syndrome in Japanese patients with relapsing polychondritis
  1. Naomi Tsuchida1,2,
  2. Yosuke Kunishita1,
  3. Yuri Uchiyama2,3,
  4. Yohei Kirino1,
  5. Makiko Enaka4,
  6. Yukie Yamaguchi5,
  7. Masataka Taguri6,
  8. Shoji Yamanaka7,
  9. Kaoru Takase-Minegishi1,
  10. Ryusuke Yoshimi1,
  11. Satoshi Fujii4,7,
  12. Hideaki Nakajima1,
  13. Naomichi Matsumoto2
  1. 1 Department of Stem Cell and Immune Regulation, Yokohama City University Graduate School of Medicine, Yokohama, Japan
  2. 2 Department of Human Genetics, Yokohama City University Graduate School of Medicine, Yokohama, Japan
  3. 3 Department of Rare Disease Genomics, Yokohama City University Hospital, Yokohama, Japan
  4. 4 Department of Molecular Pathology, Yokohama City University Graduate School of Medicine, Yokohama, Japan
  5. 5 Department of Environmental Immuno-Dermatology, Yokohama City University Graduate School of Medicine, Yokohama, Japan
  6. 6 Department of Data Science, School of Data Science, Yokohama City University, Yokohama, Japan
  7. 7 Department of Pathology, Yokohama City University Hospital, Yokohama, Japan
  1. Correspondence to Dr Yohei Kirino, Department of Stem Cell and Immune Regulation, Yokohama City University Graduate School of Medicine, Yokohama 236-0004, Japan; kirino{at}


Objectives To determine clinical and genetic features of individuals with relapsing polychondritis (RP) likely caused by pathogenic somatic variants in ubiquitin-like modifier activating enzyme 1 (UBA1).

Methods Fourteen patients with RP who met the Damiani and Levine criteria were recruited (12 men, 2 women; median onset age (IQR) 72.1 years (67.1–78.0)). Sanger sequencing of UBA1 was performed using genomic DNA from peripheral blood leukocytes or bone marrow tissue. Droplet digital PCR (ddPCR) and peptide nucleic acid (PNA)-clamping PCR were used to detect low-prevalence somatic variants. Clinical features of the patients were investigated retrospectively.

Results UBA1 was examined in 13 of the 14 patients; 73% (8/11) of the male patients had somatic UBA1 variants (c.121A>C, c.121A>G or c.122T>C resulting in p.Met41Leu, p.Met41Val or p.Met41Thr, respectively). All the variant-positive patients had systemic symptoms, including a significantly high prevalence of skin lesions. ddPCR detected low prevalence (0.14%) of somatic variant (c.121A>C) in one female patient, which was subsequently confirmed by PNA-clamping PCR.

Conclusions Genetic screening for pathogenic UBA1 variants should be considered in patients with RP, especially male patients with skin lesions. The somatic variant in UBA1 in the female patient is the first to be reported.

  • immune system diseases
  • systemic vasculitis
  • polymorphism
  • genetic

Data availability statement

Data are available upon reasonable request. The datasets used and analysed during the current study are available from the corresponding author upon reasonable request.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Key messages

What is already known about this subject?

  • Pathogenic somatic variants in the ubiquitin-like modifier activating enzyme 1 gene (UBA1) were discovered in individuals with systemic inflammation of cartilage, skin and blood vessels accompanied by haematological abnormalities, named VEXAS (vacuoles, E1 enzyme, X-linked, autoinflammatory, somatic) syndrome, affecting only male patients.

What does this study add?

  • The clinically oriented phenotype-first approach used in this study of 14 Japanese individuals with relapsing polychondritis (RP) and subsequent genetic screening revealed that 73% (8/11) of the male patients with RP had p.Met41 variants in UBA1 and the key feature of these patients was skin involvement.

  • Two highly sensitive PCR methods identified a low-prevalence somatic variant in UBA1 in a female patient, the first time that such a pathogenic variant has been found in a female patient with RP.

How might this impact on clinical practice or future developments?

  • Genetic screening for UBA1 should be considered in patients with RP, especially in males with skin lesions.


Relapsing polychondritis (RP) is a rare, systemic, idiopathic inflammatory disease that affects the cartilage of ear, nose, and trachea, as well as skin, eyes and joints.1–3 Patients with RP typically show middle-age onset and the sex ratio is close to 1.2–4 Subgroups of patients with RP have different clinical phenotypes, and men, cardiac abnormalities and myelodysplastic syndrome are clinically associated with poor prognosis.4–6 The chronic inflammation of RP is often intractable, difficult to keep in remission and its treatment is empirical and non-standardised.1–3 RP is considered as an immunological disease and antibodies to type II collagen have been detected in some patients.7 Genetic factors, such as HLA-DR4, HLA-DRB1*16:02, HLA-DQB1*05:02 and HLA-B*67:01, have been associated with RP.2 3 8 The rarity of RP has hampered exhaustive investigation to reveal its pathogenesis.

Recently, pathogenic somatic variants in the ubiquitin-like modifier activating enzyme 1 gene (UBA1) clustered at p.Met41 in a severe adult-onset autoinflammatory disease named vacuoles, E1 enzyme, X-linked, autoinflammatory, somatic (VEXAS) syndrome were detected using a genotype-first approach.9 Patients with VEXAS have a wide range of phenotypes, including chondritis, rashes and haematological abnormalities. Some patients with VEXAS were diagnosed as having giant cell arteritis or polyarteritis nodosa, suggesting that VEXAS may include heterogeneous clinical conditions.9–11 Because VEXAS is a newly identified disease, patients with VEXAS should be examined to fully reveal the clinical spectrum. In this study, we focused on patients with RP because 60% of patients with VEXAS were reported to have an RP phenotype.9 We report detailed genetic and clinical features of Japanese patients with RP with UBA1 variants in a single-centre cohort recruited from two hospitals.


Study subjects

We retrospectively examined clinical records of patients suspected of having RP between 2011 and 2020 at Yokohama City University Hospital and Yokohama Minami Kyousai Hospital. Sixteen patients with RP (labelled RP01–16) were recruited, but two were excluded (RP06 and RP09) because they did not meet the classification criteria.12–14

Genetic screening

Sanger sequencing, exome sequencing, droplet digital PCR (ddPCR) and peptide nucleic acid (PNA)-clamping PCR were used for genetic screening. ddPCR, exome sequencing and data analysis were performed as described previously.15 16 Methods for DNA extraction, ddPCR, exome sequencing, PNA-clamping PCR and measurement of anti-type II collagen antibody are provided in online supplemental methods and the primers and probes used in this study are listed in online supplemental table S1.

Statistical analysis

Categorical variables were analysed using the χ2 test and exact unconditional z-pooled test ( 18 Continuous variables were examined using the Mann-Whitney U test. A p value<0.0028 (=0.05/18) was considered statistically significant using Bonferroni’s corrected p value for multiple testing (N=18).


Clinical features of patients with RP in this study

Fourteen patients with RP who fulfilled the classification criteria were included in the analysis. The patient screening process is illustrated in online supplemental figure S1. The clinical features of the 14 patients are summarised in table 1. The onset ages were high (median (IQR), 72.1 years (67.1–78.6)), and 86% (12/14) of the patients were male. Among the patients, 86% (12/14), 36% (5/14) and 50% (7/14) presented with auricular, nasal or respiratory tract chondritis, respectively, and the pathological findings showed cartilage inflammation compatible with RP in 12 of the 13 patients who had a cartilage biopsy (figure 1 and online supplemental figure S3, table 1). Five of the seven patients with a chest CT abnormality had subjective symptoms such as hoarseness or dyspnoea; the other two were asymptomatic. Other clinical features included high fever (57%, 8/14), skin erythema (57%, 8/14), myelodysplasia (43%, 6/14) and scleritis (36%, 5/14) (figure 1 and online supplemental figure S2, table 1 and online supplemental table S2). Bone marrow aspiration revealed vacuoles in the myeloid and erythroid precursor cells in seven of the eight patients who received aspiration (figure 1 and online supplemental figure S4, table 1 and online supplemental table S2). Detailed clinical courses for these patients are described in the online supplemental results.

Table 1

Clinical features of the 14 patients with relapsing polychondritis and UBA1 variants detected by Sanger sequencing

Figure 1

Clinical features of patient RP04 who had a somatic variant in UBA1. (A) Auricular swelling. (B) Erythema nodosum-like lesions of the upper extremity. (C,D) Biopsy of auricular cartilage. Infiltration of inflammatory cells (mainly segmented neutrophils) in perichondral regions with H&E staining. Square region in (C) (magnitude 4×) was amplified in (D) (20×). (E) Three-dimensional CT reconstruction image of deformed trachea. (F) Skin biopsy from erythema. Infiltration of inflammatory cells in subcutaneous tissues and perivascular regions (H&E, 1.25×). (G) Bone marrow biopsy showed hypercellular, decreased megakaryocyte and erythroblastic islands (H&E, 40×). Chromosome analysis showed deletion of 5q, consistent with 5q− syndrome in myelodysplastic syndrome. Bone marrow aspiration. Vacuolisation in myeloid (H) and erythroid (I) precursor cells (May-Giemsa staining). (J) Electropherogram indicating a somatic variant in UBA1 in genomic DNA derived from peripheral blood leukocytes. UBA1, ubiquitin-like modifier activating enzyme 1.

Identification of somatic variants in UBA1 by Sanger sequencing

Genomic DNA (gDNA) was available from 13 of the 14 patients. Sanger sequencing targeting previously reported UBA1 variants9–11 revealed that 8 of the 11 male patients had somatic variants in UBA1 (NM_ 003334.3): c.121A>C, p.Met41Leu in three patients (RP01, RP11 and RP16), c.121A>G, p.Met41Val in two patients (RP03 and RP04) and c.122T>C, p.Met41Thr in three patients (RP05, RP13 and RP15) (table 1 and online supplemental figure S5). Exome sequencing of the gDNA from RP01 showed 74.2% variant allele frequency of p.Met41Leu (17 wild-type and 49 variant alleles). No other genetic variants associated with autoinflammatory diseases were identified (data not shown), but the exome sequencing was performed only for RP01.

On the basis of these results, we compared UBA1 p.Met41 variant-positive and variant-negative patients with RP. The variant-positive patients had significantly more skin involvement and tended to have higher incidences of myelodysplastic syndrome than the variant-negative patients (online supplemental table S3). Although haematological abnormalities were common in the variant-positive patients, one male patient with macrocytic anaemia (RP07) had no bone marrow vacuoles and was negative for the UBA1 variants (table 1 and online supplemental table S2). There was no difference between the variant-positive and variant-negative groups for the positive anti-type II collagen antibody (online supplemental table S3).

Variant detection by ddPCR

The detection limit of somatic variants is 15%–20% of allele ratios by Sanger sequencing;19 therefore, low-prevalence mosaic variants can be easily overlooked. The ddPCR confirmed the pathogenic variants in seven patients with variant fractional abundances (equivalent to allele frequencies) of 72.1% (c.121A>C, RP01), 24.7% (c.121A>G, RP03), 91.5% (c.121A>G, RP04), 75.7% (c.121A>C, RP11), 22.4% (c.122T>C, RP13), 68.8% (c.122T>C, RP15) and 87.1% (c.121A>C, RP16), and four of the five variant-negative patients by Sanger sequencing were negative for any of the three p.Met41 variants detected by ddPCR (online supplemental table S4). Notably, one female patient with RP (RP10) had a fractional abundance of 0.14% of c.121A>C, p.Met41Leu (figure 2A and online supplemental table S4). The detection limit of ddPCR was confirmed using serial dilutions of the gDNA of a variant-positive patient (RP01, UBA1 c.121A>C, 72.1% variant fractional abundance) with control gDNA to obtain variant allele ratios of 5.0%, 1.0%, 0.5%, 0.1%, 0.05% and 0.01%, which confirmed that ddPCR reliably detect variant allele ratios of >0.10% (predicted 21.8 copies) when 66 ng of gDNA was applied per well (figure 2B and online supplemental table S5).

Figure 2

Droplet digital PCR (ddPCR) and peptide nucleic acid (PNA)-clamping PCR detected UBA1 variants in patients with relapsing chondritis (RP). (A) Low-prevalence UBA1 variant in patient RP10 was detected by ddPCR targeting c.121A>C. RP01: variant-positive RP01 (positive control), Control: wild-type healthy (negative) control. Threshold amplitudes for ddPCR (pink lines) were >3500 for variant probe and >2500 for wild-type probe. Blue, green and black dots represent variant (FAM-labelled), wild-type (HEX-labelled) and non-amplification signals, respectively. (B) Detection limit of ddPCR. The genomic DNA (gDNA) of variant-positive patient (RP01, UBA1 c.121A>C) was serially diluted with wild-type gDNA (control) to obtain variant allele ratios of 5.0%, 1.0%, 0.5%, 0.1%, 0.05% and 0.01%. Samples with ratios of 5.0%–0.1% were positively amplified by ddPCR. (C) Detection by PNA-clamping PCR. The gDNA of patients with three known UBA1 p.Met41 variants (RP01: c.121A>C, RP03: c.121A>G, RP15: c.122T>C) was amplified, whereas no amplification of the gDNA of the healthy control with the wild-type allele was detected. The gDNA of the female patient RP10 showed weak amplification (red arrowhead) suggesting the presence of a variant allele. HC, healthy control; Mt, variant control; NC, negative control; UBA1, ubiquitin-like modifier activating enzyme 1; Wt, wild-type control.

Variant screening by PNA-clamping PCR

To further validate the low-prevalence somatic variant in UBA1 in the female patient with RP, we conducted PCRs with a PNA probe that hybridised to the wild-type UBA1 p.Met41 allele (c.121A or c.122T based on the Sanger sequencing) and specifically amplified only pathogenic variant alleles. Thus, any of the three known UBA1 p.Met41 variants were amplified, whereas the wild-type alleles were not amplified (figure 2C). The low-prevalence UBA1 variant that was detected by ddPCR in the female patient (RP10) showed weak but positive amplification by PNA-clamping PCR, validating the presence of the variant allele.


In this study, three known pathogenic UBA1 variants were identified in 9 of the 13 patients (69%) with RP. In particular, 73% (8/11) of the male patients with RP had the UBA1 p.Met41 variants. A gDNA sample for male patient RP12 was unavailable, so the presence of pathogenic UBA1 variants could not be determined; however, bone marrow tissue from this patient showed evidence of vacuoles, which indicated possible VEXAS (online supplemental figure S4). Genetic screening by Sanger sequencing of UBA1 in male patients with RP may be possible, given the high detection rate of UBA1 variants. Moreover, we showed that ddPCR and PNA-clamping PCR could detect low-prevalence somatic variants with high sensitivity, although such low-prevalence UBA1 variants need careful evaluation. In patient RP15, we conducted ddPCR of gDNA extracted from both peripheral blood leukocytes and bone marrow fluid and obtained positive results for both (online supplemental table S4). An integrated approach using Sanger sequencing, ddPCR and PNA-clamping PCR is important for detecting variant alleles with widely variable frequencies in different tissues at different disease stages in RP.

In UBA1 variant-negative patients, pathogenic variants may be found in the UBA gene family and other genes involved in ubiquitination, inflammatory signalling and protein folding and degradation.20 Deep sequencing of these genes may reveal their somatic variants.

Clinically diagnosed RP may consist of subgroups with different contributions of innate and acquired immunity. Detection of pathogenic UBA1 variants may be important for subgrouping RP to find the appropriate and effective therapeutic interventions.

There are some limitations in this study. Our patients with RP were elderly and male dominant, so were not a representative RP cohort.4–6 Further, the pathogenic impact of the extremely low-prevalence variant is uncertain, especially in the female patient. Therefore, larger cohorts of many patients with UBA1 variants are essential to confirm our results.

In conclusion, given the high prevalence of somatic variants in UBA1, UBA1 screening should be considered in patients with RP, especially in male patients with skin involvement.

Data availability statement

Data are available upon reasonable request. The datasets used and analysed during the current study are available from the corresponding author upon reasonable request.

Ethics statements

Ethics approval

The study was approved by the institutional review boards of Yokohama City University School of Medicine (A121129002) and Minami Kyousai Hospital (1-20-6-14). The study protocol followed the Declaration of Helsinki, and written informed consent was obtained from all the patients.


The authors thank the patients and their families for their participation in this study. The authors also thank Yumiko Takaishi from the Clinical Laboratory Department, Yokohama City University Hospital, Yokohama, and Kaori Takabe, Nobuko Watanabe and Takafumi Miyama from the Department of Human Genetics, Yokohama City University Graduate School of Medicine, for their excellent technical assistance. The authors thank Margaret Biswas, PhD, from Edanz Group ( for editing a draft of this manuscript.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Handling editor Josef S Smolen

  • NT and YKcontributed equally.

  • Contributors Conception and design: YohK. Analysis and interpretation of the data: NT, YosK, YohK, YU, ME, YY, MT and NM. Critical revision for important intellectual content and final approval of the article: all authors. Obtaining of funding: NT, YosK, YU, YohK, RY, HN and NM. Collection and assembly: NT, YosK, YohK and NM.

  • Funding This study was supported by grants from the Japanese Society for the Promotion of Science Grants-in-Aid for Scientific Research JP19H03700 (to YohK), JP20K17428 (to NT), JP19K23847 and JP20K17446 (to YosK), JP19K08914 (to RY), JP20H03714 (to HN), JP19K17865 (to YU) and JP17H01539 (to NM); AMED under grant numbers JP20ek0109486, JP20dm0107090, JP20ek0109301, JP20ek0109348 and JP20kk0205012 (to NM); intramural grants (30-6 and 30-7) of NCNP from the Ministry of Health, Labour and Welfare (to NM) and the Takeda Science Foundation (to NM).

  • Competing interests YohK reports personal fees from Amgen, grants from Chugai and grants from Ono, outside the submitted work.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.