Research ArticleAutoimmunity

Transglutaminase 4 as a prostate autoantigen in male subfertility

See allHide authors and affiliations

Science Translational Medicine  17 Jun 2015:
Vol. 7, Issue 292, pp. 292ra101
DOI: 10.1126/scitranslmed.aaa9186
  • Fig. 1. Identification of TGM4 as a male-specific autoantigen.

    (A) Human protein arrays containing more than 9000 targets were screened with sera from 27 male APS1 patients, 24 female APS1 patients, and 21 healthy control subjects, and gender-imbalanced autoantigens were selected. After an initial filtering removing targets where no patient had an intensity > 5000 or where the negative sample had an intensity > 2000, 429 targets remained. A t test was used to identify targets that differed between APS1 patients and healthy controls, and the top 50 targets were then assessed for gender specificity by comparing the frequency of positive individuals between male and female APS1 patients using Fisher’s exact test (cutoff = average of the healthy + 3 SD). At the 0.05 significance level and after the Bonferroni correction for multiple testing, only one significant target remained—TGM4 [n = 27 + 24; P = 0.00017 (not adjusted), Fisher’s exact test)]. (B) TGM4 autoantibodies were detected in 14 male patients and 1 female patient (marked in red) and were absent in the healthy subjects (cutoff = average of the healthy + 3 SD). The y axis indicates the average fluorescence signal intensity for duplicate TGM4 protein spots on the array, after subtraction of the background signal.

  • Fig. 2. TGM4 autoantibodies are highly specific for male APS1 patients.

    TGM4 autoantibodies were validated in an extended clinical material using an RLBA. (A and B) TGM4 autoantibodies were confirmed in a discovery cohort of 27 male and 24 female APS1 patients (A) and were also demonstrated in a replication cohort of 19 male and 23 female APS1 patients (B). In total, we detected TGM4 autoantibodies in 26 male APS1 patients (57%) and 1 female APS1 patient. (C) To determine the clinical specificity of TGM4 autoantibodies, we also investigated 98 patients with prostatitis, 14 patients with prostate cancer, 160 males with idiopathic infertility, a selection of autoimmune diseases (including 20 patients with each of Addison’s disease, autoimmune thyroiditis, type 1 diabetes mellitus, and Sjögren’s syndrome), and 135 healthy control subjects. TGM4 autoantibodies were detected just above cutoff in one patient with prostatitis and were absent in remaining controls, and thereby showed high specificity for male APS1 patients. The upper limit of the normal range was defined as an index value of 10.

  • Fig. 3. Transglutaminase-related autoimmune disorders and target specificity of TGM4 autoantibodies.

    (A) Phylogenetic tree showing protein sequence similarity between members of the human transglutaminase family. The dominating tissue distribution of each transglutaminase and autoimmune diseases characterized by autoantibodies against the respective transglutaminases are also displayed. (B) Crosswise assessment of TGM4 and TGM2 autoantibodies in APS1 (n = 93) and celiac disease patient sera (n = 50) shows specific reactivity against TGM4 in APS1.

  • Fig. 4. TGM4 is specifically expressed in the prostate epithelium.

    The expression of TGM4 was investigated on mRNA level by digital droplet PCR and on protein level by immunohistochemistry and immuno-rolling circle amplification, and multiple human tissues were assessed. (A) TGM4 mRNA was specifically detected in prostate tissue. (B) Strong TGM4 staining was seen in the prostate epithelium, whereas remaining tissues showed very faint or no staining (from left to right: prostate, kidney, duodenum, epididymis, liver, testis, gallbladder, lung, tonsil, and placenta). (C) TGM4 expression was further restricted to certain regions within the prostate gland, here shown with immuno-rolling circle amplification and exemplified by a tissue section with marked TGM4 expression (left), a transition zone in which the prostate epithelium contains cells that are positive and cells that are negative for TGM4 (middle), and a region without TGM4 expression (right). Scale bars, 200 μm.

  • Fig. 5. TGM4 autoantibodies present at puberty.

    (A) TGM4 autoantibodies were detected in 78% of males more than 30 years of age (n = 18) and in 52% of males 13 to 30 years of age (n = 23), but were absent in the five patients younger than 13 years. (B) TGM4 autoantibodies were assessed in consecutive samples from six male APS1 patients and were first detected by the age of 12 to 16 years. Time points for blood sampling are indicated as dots along the patient age timeline, and TGM4 autoantibody positivity is marked in red. TGM4 autoantibodies were measured in serum using an RLBA, and the upper limit of the normal range was defined as an index value of 10.

  • Fig. 6. (A) Aire-deficient mice develop TGM4 autoantibodies and prostatitis.

    Autoantibodies against murine TGM4 were measured in 66 mice of the C57BI/6 strain, including 23 male Aire−/−, 24 female Aire−/−, 11 males of wild type, and 8 females of wild type (WT), and were detected in all male Aire−/− mice but were absent in all female and all wild-type mice. (B) The prostate histology was investigated in two Aire−/− mice with TGM4 autoantibodies and in three wild-type mice (all of NOD background), revealing marked signs of an active prostatitis in the Aire−/− and normal histology in wild-type mice. Scale bars, 500 μm. (C) To characterize the prostate-infiltrating T cells, markers for TH1, TH2, TH17, or Treg subsets were assessed in prostate tissue from three Aire−/− and three wild-type NOD mice. Aire−/− mice showed high expression of the TH1-associated cytokine IFNγ and the Treg-specific transcription factor Foxp3 compared to wild-type mice, whereas TH2-associated cytokines (Il4 and Il13) and TH17-associated molecules (Rorc, Il17a, and Il17f) were not significantly increased in the Aire−/− mice, indicating a predominance of TH1 and Treg cell subsets in prostatitis. qPCR results are standardized to β-actin and normalized to wild type. Bars indicate means.

  • Fig. 7. Aire-dependent expression of TGM4 in mTECs.

    Tgm4 mRNA levels were assessed by qPCR in mTECs from Aire−/− mice and wild-type mice of the C57BI/6 strain and were compared with that of Ins2, an established Aire-dependent gene, and Crp, a gene known to be expressed in mTECs independently of Aire. TGM4 mRNA was detected in mTECs from wild-type mice but was undetectable in mTECs from Aire−/− mice. qPCR results are standardized to β2-microglobulin (β2M) and normalized to wild-type mTECs, with bars depicting means ± SD. ND, not detected.

  • Fig. 8. Aire-deficient mouse prostates lack TGM4.

    Tgm4 mRNA levels were assessed in prostate tissue from three Aire-deficient NOD mice with TGM4 autoantibodies and from three wild-type NOD mice. qPCR results are standardized to β-actin and normalized to wild type. Bars indicate means.

Supplementary Materials

  • www.sciencetranslationalmedicine.org/cgi/content/full/7/292/292ra101/DC1

    Fig. S1. Confirmation of TGM4 autoantibodies using an RLBA.

    Fig. S2. TGM4 mRNA expression in multiple human tissues, detected by digital droplet PCR.

    Fig. S3. TGM4 protein expression in multiple human tissues, detected by immunohistochemistry.

    Fig. S4. TGM4 protein expression in multiple human tissues, detected by immuno-rolling circle amplification.

    Fig. S5. Male Aire-deficient NOD mice display TGM4 autoantibodies.

    Fig. S6. Male wild-type NOD mice display TGM4 autoantibodies.

    Table S1. Characterization of APS1 patients.

    Table S2. Autoantibodies in young male APS1 patients.

    Table S3. TGM4 autoantibodies first appear during pubertal age.

    Table S4. mRNA expression in mouse prostates.

    Table S5. TGM4 mRNA expression in mTECs.

  • Supplementary Material for:

    Transglutaminase 4 as a prostate autoantigen in male subfertility

    Nils Landegren,* Donald Sharon, Anthony K. Shum, Imran S. Khan, Kayla J. Fasano, Åsa Hallgren, Caroline Kampf, Eva Freyhult, Brita Ardesjö-Lundgren, Mohammad Alimohammadi, Sandra Rathsman, Jonas F. Ludvigsson, Dan Lundh, Ruben Motrich, Virginia Rivero, Lawrence Fong, Aleksander Giwercman, Jan Gustafsson, Jaakko Perheentupa, Eystein S. Husebye, Mark S. Anderson, Michael Snyder, Olle Kämpe

    *Corresponding author. E-mail: nils.landegren{at}ki.se

    Published 17 June 2015, Sci. Transl. Med. 7, 292ra101 (2015)
    DOI: 10.1126/scitranslmed.aaa9186

    This PDF file includes:

    • Fig. S1. Confirmation of TGM4 autoantibodies using an RLBA.
    • Fig. S2. TGM4 mRNA expression in multiple human tissues, detected by digital droplet PCR.
    • Fig. S3. TGM4 protein expression in multiple human tissues, detected by immunohistochemistry.
    • Fig. S4. TGM4 protein expression in multiple human tissues, detected by immuno-rolling circle amplification.
    • Fig. S5. Male Aire-deficient NOD mice display TGM4 autoantibodies.
    • Fig. S6. Male wild-type NOD mice display TGM4 autoantibodies.
    • Table S1. Characterization of APS1 patients.
    • Table S2. Autoantibodies in young male APS1 patients.
    • Table S3. TGM4 autoantibodies first appear during pubertal age.
    • Table S4. mRNA expression in mouse prostates.
    • Table S5. TGM4 mRNA expression in mTECs.

    [Download PDF]

Navigate This Article