β2-Microglobulin deficiency causes a complex immunodeficiency of the innate and adaptive immune system

Immune deficiencies, infection, and systemic immune disorders

b2-Microglobulin deficiency causes a complex immunodeficiency of the innate and adaptive immune system € u € seyin Onay, MD, PhD,c Sandra Ammann, MSc,b Caroline Keck, MSc,d € r Ardeniz, MD,a Susanne Unger, MSc,b Hu Om e € Gerc¸eker, MD,f Bianca Martin, PhD,g Ilka Fuchs, MSc,b Ulrich Salzer, MD,b Corina Cianga, MD, PhD, Bengu h _  ulları, MD, Deniz Gu € log  lu, PhD,h Tug  rul Dereli, MD,f Robert Thimme, MD,g Stephan Ehl, MD,b Aydan Ikinciog Klaus Schwarz, MD,i Annette Schmitt-Graeff, MD,d Petru Cianga, MD, PhD,e Paul Fisch, MD,d and Klaus Warnatz, MDb _ Izmir and Ankara, Turkey, Freiburg and Ulm, Germany, and Iasi, Romania Background: Most patients with MHC class I (MHC-I) deficiency carry genetic defects in transporter associated with antigen processing 1 (TAP1) or TAP2. The clinical presentation can vary, and about half of the patients have severe skin disease. Previously, one report described b2-microglobulin (b2m) deficiency as another monogenetic cause of MHC-I deficiency, but no further immunologic evaluation was performed. Objective: We sought to describe the molecular and immunologic features of b2m deficiency in 2 Turkish siblings with new diagnoses. Methods: Based on clinical and serologic findings, the genetic defect was detected by means of candidate gene analysis. The immunologic characterization comprises flow cytometry, ELISA, functional assays, and immunohistochemistry. From athe Internal Medicine Division of Allergy and Clinical Immunology, c the Department of Medical Genetics, and fthe Department of Dermatology, b _ Ege University Medical Faculty, Izmir; the Center for Chronic Immunodeficiency, University Medical Center Freiburg and University of Freiburg; dthe Institute of Pathology and gthe Department of Internal Medicine II, University Medical Center Freiburg; eGrigore T. Popa University of Medicine and Pharmacy, Department of Immunology, Iasi; hthe Department of Pediatric Immunology and Allergy, Ankara University School of Medicine; and ithe Institute for Transfusion Medicine, University of Ulm, and the Institute for Clinical Transfusion Medicine and Immunogenetics Ulm, German Red Cross Blood Service, Baden-W€urttemberg-Hessen, Ulm. Supported by the German Federal Ministry of Education and Research (BMBF 01EO1303 to K.W., S.U., S.E., and U.S.), grant DJCLS R10/34f from the Deutsche Jose Carreras Leuk€amie Stiftung (to P.F.), and Romanian Academy of Medical Sciences, VIASAN grant #328 (to P.C.). C.C. and P.C. have received research support from the ‘‘Grigore T. Popa’’ University of Medicine and Pharmacy, Iasi, Romania (grant 7365/2010). Disclosure of potential conflict of interest: P. Fisch has received support for travel to the 2014 Gamma Delta T Cell Conference (Chicago), the 2013 KIR Workshop (Minneapolis), the 2014 LMB Cambridge Alumni Reunion (2014), and the 2012 and 2013 IMBS Symposia (Buenos Aires). K. Warnatz has received lecture fees from Baxter, GlaxoSmithKline, CSL Behring, Pfizer, Biotest, Novartis Pharma, Stallergenes AG, Roche, Meridian Health Comms, Octapharma, and the American Academy of Allergy, Asthma & Immunology; has received payment for manuscript preparation from UCB Pharma; and has received payment for development of educational presentations from ESID. The rest of the authors declare that they have no relevant conflicts of interest. Received for publication May 8, 2014; revised October 21, 2014; accepted for publication December 18, 2014. Available online February 19, 2015. € ur Ardeniz, MD, Ege University, Medical Faculty, Corresponding author: Om€ Department of Internal Medicine, Division of Allergy and Clinical Immunology, _ Izmir, Turkey. E-mail: [email protected]. 0091-6749/$36.00 Ó 2015 American Academy of Allergy, Asthma & Immunology http://dx.doi.org/10.1016/j.jaci.2014.12.1937

392

Results: Here we provide the first extensive clinical and immunologic description of b2m deficiency in 2 siblings. The sister had recurrent respiratory tract infections and severe skin disease, whereas the brother was fairly asymptomatic but had bronchiectasis. Not only polymorphic MHC-I but also the related CD1a, CD1b, CD1c, and neonatal Fc receptor molecules were absent from the surfaces of b2m-deficient cells. Absent neonatal Fc receptor surface expression led to low serum IgG and albumin levels in both siblings, whereas the heterozygous parents had normal results for all tested parameters except b2m mRNA (B2M) expression. Similar to TAP deficiency in the absence of a regular CD8 T-cell compartment, CD81 gd T cells were strongly expanded. Natural killer cells were normal in number but not ‘‘licensed to kill.’’ Conclusion: The clinical presentation of patients with b2m deficiency resembles that of patients with other forms of MHC-I deficiency, but because of the missing stabilizing effect of b2m on other members of the MHC-I family, the immunologic defect is more extensive than in patients with TAP deficiency. (J Allergy Clin Immunol 2015;136:392-401.) Key words: b2-Microglobulin deficiency, human, MHC class I, hypogammaglobulinemia, neonatal Fc receptor, CD1, gd T cells, CD8 T cells

MHC class I (MHC-I) deficiency syndromes are rare forms of primary immunodeficiency most commonly caused by a deficiency of transporter associated with antigen processing (TAP) 1 or 2.1,2 Patients typically present with recurrent bacterial infections of the upper and lower respiratory tract, and about half have an ulcerating granulomatous skin disease.2 Increased numbers of infections and cutaneous manifestations might start in childhood or adulthood. In some patients necrotizing granulomatous lesions lead to complete destruction of the nasal cartilage. Loss of TAP expression is usually due to homozygous mutations in either TAP1 or TAP2, and most reported patients are of consanguineous descent.3 In patients with TAP deficiency, defective loading of the MHC-I a-chain/b2-microglobulin (b2m) heterodimers with peptides leads to retention of the complex in the endoplasmic reticulum (ER) and severely reduced expression of the MHC-I complex on the cell surface.4 Reduced MHC-I expression is associated with alterations of T-cell receptor (TCR) ab1 CD8 T-cell and natural killer (NK) cell compartments because MHC-I molecules select cytotoxic T cells in the thymus

ARDENIZ ET AL 393

J ALLERGY CLIN IMMUNOL VOLUME 136, NUMBER 2

Abbreviations used b2m: b2-Microglobulin CMV: Cytomegalovirus ER: Endoplasmic reticulum FcRn: Neonatal Fc receptor iNKT: Invariant natural killer T MHC-I: MHC class I NK: Natural killer PMA: Phorbol 12-myristate 13-acetate RTI: Respiratory tract infections TAP: Transporter associated with antigen processing TCR: T-cell receptor

and are involved in the maturation and ‘‘licensing’’ of NK cells.1,5,6 Total TCRab1 CD8 T-cell counts are often reduced in TAP-deficient patients, depending on the residual MHC-I expression, whereas total CD8 T-cell counts might be normal, likely because of expansion of gd T cells expressing CD8.3 NK cells of TAP-deficient patients were normal in number but, when examined, lacked cytotoxic activity toward HLA class I–deficient targets.7 In 1990, Waldmann and Terry8 described 2 siblings born of a first-cousin marriage who presented with low IgG and albumin levels caused by increased catabolism. The female sibling reportedly had purple-red spots and ulcerations on her legs after the age of 21 years and died of sepsis and bleeding at the age of 40 years. Her brother demonstrated the same laboratory changes but was asymptomatic and lost to follow-up. The analysis of archived DNA 35 years later revealed a mutation in the b2m gene (B2M),9 which is required for cell-surface expression of nearly all members of the MHC-I family, including the neonatal Fc receptor (FcRn).10 FcRn is not only responsible for the maternofetal transfer of IgG but is also important for normal IgG and albumin serum half-life.11 Because these b2m-deficient siblings were no longer available for further evaluation, the effect of this mutation on the cellular expression of b2m-dependent proteins and on the immune system was never evaluated. Here we describe, for the first time, the multifaceted effect of genetic b2m deficiency on the human innate and adaptive immune system.

METHODS Patients and control subjects Written informed consent was received from all patients and control subjects before inclusion in the study. The study was approved by the institutional ethical board (251/13), according to the Declaration of Helsinki.

Degranulation and cytotoxicity assays Degranulation and NK cell cytotoxicity assays were performed, as described previously.12 PMBCs were incubated with K562 target cells at a 1:1 ratio or with 50 ng/mL PMA and 1 mg/mL ionomycin (both from Sigma Aldrich, St Louis, Mo) for 2 hours at 378C. For Fc receptor stimulation, PBMCs were rested overnight at 378C and then incubated at a 1:1 ratio with L1210 cells alone or L1210 plus anti-CD16 (BD Biosciences, San Jose, Calif) for 2.5 hours at 378C. For cross-linking, L1210 cells were preincubated with 5 mg/mL anti-CD16 for 15 minutes at room temperature. Unbound anti-CD16 antibodies were removed. CD107a expression on CD32CD561 NK cells was measured by using flow cytometry. To assess NK-cell cytotoxicity, 51Cr-labeled K562 target cells were incubated with PMBCs at different effector/target cell ratios for 4 hours at 378C. [51Cr]

release was measured in a Topcount luminometer (PerkinElmer, Waltham, Mass), and specific lysis was calculated. For degranulation of freshly isolated CD8 T cells, PBMCs were rested overnight and then incubated at a 1:1 ratio with P815 cells alone or P815 plus anti-CD3 (Okt3; eBioscience, San Diego, Calif) or medium alone for 2.5 hours at 378C. For cross-linking, P815 cells were preincubated with 5 mg/mL anti-CD3 for 15 minutes at room temperature. Unbound anti-CD3 antibodies were removed. For degranulation of CD8 T-cell blasts, PHA/IL-2 T-cell blasts were simulated with anti-CD3/CD28 beads (Dynal; Invitrogen, Carlsbad, Calif) at a 1:3 ratio for 3 hours at 378C. CD107a expression on CD31CD81 T cells was measured by means of flow cytometry. CD8 T-cell cytotoxicity assays were performed, as described previously.13 PBMCs were primed against irradiated cells from the EBV-transformed 721 B-cell line. On days 10, 17, 24, and 31, priming of the cultures was repeated by adding irradiated 721 cells. Five days after the last restimulation, cells were harvested and tested in a standard [51Cr] release assay with 721 and K562 target cells. Cytotoxicity was calculated as described.

Antibodies for flow cytometry Antibodies are listed in the Methods section in this article’s Online Repository at www.jacionline.org. B-cell subsets were stained and analyzed, as previously described.14 For details on molecular genetics and immunohistochemistry, see the Methods section in this article’s Online Repository at www.jacionline.org.

RESULTS Clinical presentation A 31-year-old woman (IV-7, P1) was referred to the Dermatology Clinic at Ege University with nasal perforation and ulcerated brown-violet–colored skin lesions affecting all 4 extremities (Fig 1, A and B). These lesions had appeared as subcutaneous nodules at 9 years of age during a flu-like infection and progressed slowly over time. At the age of 12 years, she had an abscess of the left lung, and a wedge resection revealed a histologic picture compatible with organizing pneumonia. Because of the granulomatous dermatitis, she was treated for several years unsuccessfully for suspected cutaneous tuberculosis despite negative histology, PCR, and culture results and a negative QuantiFERON-TB Gold In-Tube test (Cellestis, Carnegie, Victoria, Australia) result. Multiple testing for other infectious agents and anti-neutrophil cytoplasmic antibodies yielded negative results. Thoracic computed tomography revealed bronchiectasis (Fig 1, C). The most remarkable finding in the routine blood chemistry was severe hypoproteinemia without signs of renal or gastrointestinal protein loss (Table I). Albumin (2.3 g/dL) and IgG (2.34 g/L) levels were reduced, although normal IgM and IgA levels were detected (Table I). Blood counts and routine lymphocyte phenotyping revealed decreased numbers of B and NK cells (Table I). Specific IgG responses against Epstein-Barr nuclear antigen, rubella (>500 IU/mL), tetanus toxoid, and cytomegalovirus (CMV) were readily detectable, whereas the response to anti-pneumococcal polysaccharide vaccination was slightly reduced (Table I). The index patient was born as the third child of 6 to consanguineous parents (Fig 1, D). Three of the siblings died early in life, and for the 3 siblings who survived infancy, recurrent attacks of diarrhea were reported. The younger sister (IV-9) died at the age of 22 years of unknown cause after she had been given a diagnosis of ‘‘cutaneous tuberculosis.’’ Similar to her sister, the diagnosis was based on granulomatous skin lesions without microbiologic confirmation.

394 ARDENIZ ET AL

J ALLERGY CLIN IMMUNOL AUGUST 2015

FIG 1. A, Clinical presentation, pedigree, and molecular characterization of b2m deficiency. A-C, Midface destruction (Fig 1, A), skin lesions (Fig 1, B), and bronchiectasis (Fig 1, C) in patient IV-7 (P1). D, Pedigree of the index patient. IV-10 (P2) is the younger brother. E, Both had a homozygous mutation in intron 1 (c.6711G>T). F, Abrogating full-length mRNA expression, as shown by RT-PCR. GAPDH, Glyceraldehyde-3-phosphate dehydrogenase. G, b2m protein was absent on the surfaces of patients’ CD451 lymphocytes. FMO, Fluorescence Minus One; T control, transport control.

The physical examination of the 18-year-old brother (IV-10, P2) was unremarkable. Despite the absence of recurrent respiratory tract infections (RTIs), thoracic computed tomography showed bronchiectasis. Serum albumin and IgG levels were diminished, whereas IgM and IgA levels were normal or even increased (Table I). Anti–Epstein-Barr nuclear antigen and anti-CMV IgG were detectable. B-cell counts were found to be diminished on flow cytometry. Given the clinical resemblance to MHC-I deficiency combined with severe hypoalbuminemia and low IgG but normal IgA and IgM levels, b2m deficiency was suspected and corroborated based on undetectable serum levels of b2m in both siblings (<0.22 mg/L). The parents’ b2m levels were within the normal range (Table I). Both siblings began to receive 800 mg/kg intravenous immunoglobulin every 2 weeks in conjunction with antibiotic prophylaxis. Although IgG replacement was able to decrease the frequency of airway infections during the 4 years of follow-up, IgG serum levels at 2 weeks after substitution had decreased to trough levels, and no improvement of skin lesions was recorded.

Molecular characterization of b2m deficiency and HLA genotypes B2M consists of 3 exons. Sequencing the genomic DNA for B2M of both patients revealed a novel homozygous splice site mutation (c.6711G>T) in intron 1 that was not detected in

200 chromosomes of healthy control subjects (Fig 1, E). Both parents were heterozygous carriers of this mutation. Splice site prediction software (Splice Site Prediction by Neural Network, Berkeley Drosophila Genome Project) predicted that this mutation results in the use of a cryptic splice site 4 nucleotides downstream of exon 1, leading to a frame shift and a premature stop codon in exon 2 after 104 bp. This prediction was supported by the absence of transcripts spanning across exons 1 and 2 (Fig 1, F, and see Fig E1 in this article’s Online Repository at www. jacionline.org). The severe reduction of transcript levels of exon 1 suggests a nonsense-mediated mRNA decay of the truncated B2M product (Fig 1, F). As suggested by the absent serum levels, there was no detectable b2m protein expression on the surfaces of lymphocytes from either patient (Fig 1, G). Although B2M mRNA levels were reduced in the heterozygous parents, b2m protein surface levels were comparable with control levels (Fig 1, G, and see Fig E1). Two-digit resolution of HLA genotyping revealed that the female patient was heterozygous and the male sibling was homozygous for the HLA locus (see Table E1 in this article’s Online Repository at www.jacionline.org).

Expression of b2m-associated proteins Surface expression of several molecules depends on coexpression of b2m. Consistently, MHC-I surface expression was absent on lymphocytes of both siblings (Fig 2, A). Because

ARDENIZ ET AL 395

J ALLERGY CLIN IMMUNOL VOLUME 136, NUMBER 2

TABLE I. Blood and serum parameters in patients with b2m deficiency Parameter

Leukocytes (/mL) Lymphocytes (/mL) T cells (/mL) CD4 T cells (/mL) CD81 T cells (/mL) TCRab1 CD8 T cells (/mL) CD42CD82 gd T cells (/mL) CD81 gd T cells (/mL) B cells (/mL) NK cells (/mL) IgG (g/L) IgA (g/L) IgM (g/L) Albumin (g/dL) b2m (mg/L) Iron (mg/dL) Total iron-binding capacity (mg/dL) Ferritin (ng/mL) Transferrin (mg/dL) Hematocrit (%) Postvaccination Anti–tetanus toxoid (IU/mL) Anti-PnPS (mU/mL) Anti-rubella (IU/mL) Serostatus Anti-EBV Anti-CMV

P1 (female)

P2 (male)

Mother

Father

Normal range

6,680 1,600 1,542 1,096 317 78 130 223 16 38 2.34* 1.53 1.16 2.3 <0.22 32 373 38.1 294 36

5,800 1,500 1,175 604 467 13 94 444 47 273 2.37* 4.29 0.44 3.0 <0.22 147 330 43.93 321 56.6

8,690 1,400 1,119 753 370 354 24 4 148 1,132 9.57 3.4 0.37 4.1 1.20 NA NA NA NA NA

6,960 1,500 1,358 830 563 513 80 46 50 92 10.22 1.37 0.92 4.8 1.03 NA NA NA NA NA

4,000-10,000 1,200-3,600 700-2,100 300-1,400 200-900 172-665 11-154 1-35 100-500 90-600 7.00-16.00 0.70-4.00 0.40-2.30 3.5-5.2 0.8-2.2 f:37-145; m: 59-158 228-428 Female: 13-150; male: 30-400 200-360 Female: 35-45; male: 37.7-53.7 >0.1 (>40) (10-500)

1.42 37 >500

1.88 14 NA

NA NA NA

NA NA NA

Positive Positive

Positive Positive

NA NA

NA NA

Abnormal values are indicated in boldface. NA, Not available; PnPS, pneumococcal polysaccharide. *IgG levels at diagnosis.

b2m is essential for assembly and transport of MHC-I molecules from the ER to the cell surface, intracellular b2m-free MHC-I a chains (reactive with the mAbs HC10 and Q1/28) but not b2m-bound MHC-I molecules (reactive with mAbs B.1.23.2 and Q1/28) were readily detected by the respective mAbs in the EBV line of P2 (Fig 2, B). The b2m-deficient EBV line Daudi was used as a control. Cell-surface b2m-free MHC-I a chains were clearly detectable in control 721 cells but not in Daudi cells or EBV-transformed cells from P2. When analyzing skin sections of both siblings, we could detect CD1a expression in a reduced number of Langerhans cells in the epithelial layer but not the dermis (Fig 2, C). Because immunohistochemistry does not allow differentiation of intracellular and surface expression, we tested the induction of CD1 molecules on monocyte-derived dendritic cells after 6 days of in vitro differentiation. In contrast to healthy control subjects, no surface expression of CD1a, CD1b, or CD1c could be detected in either patient (Fig 2, D, left). In contrast, b2m-deficient monocytes express CD1d on their surfaces but partially, with reduced levels when compared with those seen in control subjects (Fig 2, D, right). Similar to CD1a, expression of the MHC-I–like FcRn was absent from the surfaces of monocytes in both patients (Fig 2, E) but could be detected by using immunohistochemistry in skin sections (Fig 2, F) in the epidermis of P2.

T-cell compartment in patients with b2m deficiency Absolute numbers of T cells, including CD81 T-cell numbers from both siblings, were within the normal range (Fig 3, A, and

Table I). However, in both b2m-deficient patients, the CD81 compartment consisted mainly of gd T cells expressing CD8 (Fig 3, A). Similar to control subjects, the majority of patients’ CD81 gd T cells were CD8aa1 (see Fig E2 in this article’s Online Repository at www.jacionline.org). Only in P1 were a relevant amount of TCRab1 CD8 T cells with a highly biased Vb repertoire (see Fig E3 in this article’s Online Repository at www.jacionline.org) and a terminal effector or effector memory 3 phenotype (EM3; Fig 3, B) present. Tetramer analysis could not detect HLA-B*27–restricted EBV- or CMV-specific CD8 T cells in P1, despite her being seropositive for both viruses (see Fig E4 in this article’s Online Repository at www.jacionline. org). The differentiation of TCRab1 CD4 T-cell subpopulations was fairly normal (Table II). Only increased HLA-DR expression (Table II) and skewing of the Vb repertoire of CD4 T cells (see Fig E3) suggested an ongoing T-cell activation in P1, although not in P2. The percentage of IL-41 and IL-171 CD4 T cells resembled the frequency in control subjects, but IFN-g1 CD4 T cells were relatively expanded when compared with those in control cells (see Fig E5 in this article’s Online Repository at www.jacionline.org). Importantly, no CD1d-restricted Va241Vb111 NKT cells were detectable in either sibling (Fig 3, C). All CD81 gd T cells of patients and control subjects were CCR72 and part of the effector/memory pool. In P1 all CD81 gd T cells were CD45RA1CD272CD282, whereas in her brother CD45RA2CD271CD281 T cells were present also (Fig 3, D). CD42 and especially CD45R02 T cells expressed high IFN-g levels (see Fig E5). CD42CD82 gd T cells resembled the normal gd T-cell population in healthy control subjects in regard to numbers (Table I). Both CD82 and CD81 gd T cells

396 ARDENIZ ET AL

J ALLERGY CLIN IMMUNOL AUGUST 2015

FIG 2. Expression of b2m-associated proteins in patients with b2m deficiency. A, Surface HLA-A, HLA-B, and HLA-C expression on CD451 lymphocytes. B, Expression of MHC-I a-chains bound to b2m (B.1.23.2 and Q1/28) and unbound (HC10). C, Intraepithelial CD1a1 Langerhans cells in the skin of P2 and a control subject. D, Surface expression of CD1a, CD1b, and CD1c on monocyte-derived dendritic cells and of CD1d on monocytes. E and F, FcRn surface expression on monocytes (Fig 1, E) and in the skin (Fig 1, F). FMO, Fluorescence Minus One; T control, transport control.

FIG 3. T-cell compartment in patients with b2m deficiency. A, Distribution of CD4, CD8, and gd T cells gated on CD31 T cells. B, Phenotyping of TCRab1 CD8 T cells. NA, Not available. C, Va241Vb111 NKT cells gated on lymphocytes. D, Phenotyping of CD81 gd T cells.

had reduced expression of CD3 and TCRgd (see Fig E6 in this article’s Online Repository at www.jacionline.org).

Cytotoxicity in patients with b2m deficiency A substantial proportion of circulating CD4 T cells in P1, but only a few in P2 and none in the healthy control subject, expressed perforin (Fig 4, A). Compatible with their effector memory phenotype, TCRab1 CD8 T cells (see Fig E7, A, in this article’s Online Repository at www.jacionline.org) and CD81 gd T cells

(see Fig E7, B) expressed perforin and CD57, revealing the presence of antigen-experienced effector cells with an indistinguishable phenotype in both patients and control subjects. Degranulation of freshly isolated, rested CD81CD571 T cells of both patients was slightly reduced compared with that seen in control subjects after anti-CD3 stimulation (see Fig E8, A, in this article’s Online Repository at www.jacionline.org), whereas it was comparable for P2 after 2 days of in vitro expansion with PHA plus IL-2 (Fig 4, B). The assay repeatedly failed for P1 because of poor survival of CD81 cells during expansion. Despite

ARDENIZ ET AL 397

J ALLERGY CLIN IMMUNOL VOLUME 136, NUMBER 2

TABLE II. Lymphocyte subpopulations in patients with b2m deficiency

CD4/CD8 TCRab1CD42CD82 (DNT) [% of lymphocytes] Naive CD41 T cells (% of CD4 T cells) iNKT cells HLA-DR1 (% of CD4 T cells) TCRab (% of CD8 T cells) CD81 (% of gd T cells) Percentage of TCRab1 CD8 T cells CD45RA1CCR71 (naive) CD45RA2CCR71 CM CD45RA1CCR72CD271CD281 pE1 CD45RA1CCR72CD271CD282 pE2 CD45RA1CCR72CD272CD282 E CD45RA2CCR72CD271CD281 EM1 CD45RA2CCR72CD271CD282 EM2 CD45RA2CCR72CD272CD282 EM3 CD45RA2CCR72CD272CD281 EM4 Percentage of CD81 gd T cells CD45RA1CCR71 CD45RA2CCR71 CD45RA1CCR72CD271CD281 CD45RA1CCR72CD271CD282 CD45RA1CCR72CD272CD282 CD45RA2CCR72CD271CD281 CD45RA2CCR72CD271CD282 CD45RA2CCR72CD272CD282 CD45RA2CCR72CD272CD281 CD191 total B cells Percentage of B cells IgD1CD272 naive IgD1CD271 MZ-like B cells IgD2CD271 switched memory B cells IgA1 B cells IgG1 B cells CD21low B cells IgM11IgD11CD2411CD3811 transitional B cells CD202CD3811CD271 plasmablasts

Day Normal control range (%)

P1

P2

3.67 0.2

2.15 0.3

1-3.6 0.4-2.5

23.1 37.4 0.00 0.00 0.09 20.7 6.1 24.6 2.7 89.5 53.8 73.6 29.1

21-58 0.01-0.44 3.0-12.0 86.7-98.3 2.5-39.4

0.9 0.7 0 0.5 37.0 1.7 1.3 51.3 0.4

NA NA NA NA NA NA NA NA NA

18.6 3.3 2.5 5.2 14.6 31.1 6.15 2.7 3.3

16.9-64.3 0.8-7.8 0.3-19.6 0.7-16.3 0.2-16.7 5.3-39.3 0.3-13.0 0.2-24.7 0.6-4.9

1.0 0.1 0.1 1.3 73.5 0.4 1.6 13.3 0.1 0.9

1.4 1.8 0.1 2.9 35.7 39.4 1.9 6.2 1.4 3.5

0.7 2.1 1.3 13.3 17.2 12.3 16.6 16.9 3.3

0-14.7 0-11.5 0-43.8 1.8-31.6 0-78.7 0.1-54.2 0-30.9 0-27.0 0-4.5 6-19

44.0 18.0 24.0 10.0 NA 23.0 3.0

59.8 13.3 18.8 10.6 6.3 9.6 1.4

43.2-82.4 7.2-30.8 6.5-29.2 2.7-14.8 3.8-13.6 0.8-7.7 0.6-3.5

1.0

13.9

0.4-3.6

Abnormal values are indicated in boldface. DNT, Double-negative T cells; E, effector; EM, effector memory; NA, not available; pE, peripheral effector.

intact degranulation, it was not possible to generate antigen-specific cytotoxic T cells from either sibling’s PBMCs after 4 rounds of in vitro stimulation by irradiated allogeneic EBV-transformed 721 cells, whereas this was easily possible in 2 healthy control subjects (see Fig E8, B). No relevant nonspecific cytotoxicity against K562 cells was present in these in vitro–primed cultures, implying that most of the killing of 721 target cells by the PBMCs from the healthy control subjects was mediated by antigen-specific cytotoxic T cells rather than nonspecific lymphokine-activated killer activity. The numbers of circulating NK cells were decreased in P1 but normal in P2 (Table I). CD56bright and CD56dim NK cells were detectable in both patients (Fig 4, C). However, NK cells of both siblings did not degranulate on stimulation with K562 target cells on 2 different occasions, whereas degranulation on Fc receptor stimulation was detectable but reduced and even comparable

with control values on phorbol 12-myristate 13-acetate (PMA)/ ionomycin (Fig 4, C, and see Fig E8, C). A 51chromium release assay revealed no NK cell–mediated cytotoxic activity in P2 (Fig 4, D).

B-cell compartment in patients with b2m deficiency In both siblings B cells were reduced in number, but the relative distribution of the subpopulations was unremarkable, except for a relative expansion of activated CD21low B cells (Table II). Inflammatory skin lesions in patients with b2m deficiency A common feature seen in both siblings was dense collagen fibrosis involving the dermis and the superficial subcutaneous tissue accompanied by a loss of hair follicles and reduced sweat glands (Fig 5, A, E, and I). The severe skin disease of P1 was associated with granulomatous inflammation with giant cells of the deep dermis (Fig 5, E-H). Immunohistochemistry showed scattered intradermal, mainly perivascular CD4 (Fig 5, B) and intraepithelial CD8 (Fig 5, C) T-cell infiltration. The latter comprised mainly TCRgd1 cells in small intraepithelial clusters (Fig 5, D). A subpopulation of cytotoxic T cells in the granuloma expressed perforin (Fig 5, H). In P2 numerous T cell– and macrophage-rich inflammatory lesions developed in the fibrotic skin (Fig 5, I-L) that lacked giant cells. DISCUSSION Here we describe the first 2 living siblings of consanguineous descent given diagnoses of b2m deficiency. The recurrent bacterial upper and lower RTIs and bronchiectasis were first interpreted in the context of hypogammaglobulinemia, but with the onset of mutilating granulomatous skin lesions and nasal septum perforation in the female sibling, the clinical manifestation resembled that of MHC-I deficiency,2 leading to the diagnosis of b2m deficiency. This was corroborated by absent serum b2m and absent MHC-I expression on PBMCs. Genetic analysis in both siblings revealed a homozygous splice site mutation associated with a severely reduced, alternatively spliced short mRNA transcript abrogating any detectable protein expression. This differs from the original b2m-deficient couple in whom a single missense mutation (c.913G>C) in exon 1 of B2M allowed for 20% of the normal surface expression of MHC-I.9 All b2m-deficient patients had low serum albumin and IgG levels, normal to high IgA levels, and variable IgM levels. Interestingly, although both male patients had mild disease, both female patients had an ulcerative skin disease during adolescence reminiscent of that seen in patients with TAP deficiency.15 The clinical resemblance to TAP deficiency extends to a possible asymptomatic course16 in the historic male patient8 and RTIs and bronchiectasis in both siblings reported here.15,17 As previously reported in b2m-deficient cells, the heavy chain of MHC-I molecules accumulates in the ER and is not transported to the surface.18 Thus MHC-I heavy chains were detectable intracellularly but not on the surface of PBMCs of both b2mdeficient siblings. In this regard the block is complete compared with most TAP1/2-deficient patients, who typically show residual surface expression of MHC-I molecules.3 As in murine models of TAP deficiency19 and b2m deficiency20 and reported in a few

398 ARDENIZ ET AL

J ALLERGY CLIN IMMUNOL AUGUST 2015

FIG 4. Cytotoxicity in patients with b2m deficiency. A, Perforin expression in CD31CD41CD82 T cells. B, Degranulation of CD8 T-cell blasts. C and D, Degranulation (Fig 4, C) and killing (Fig 4, D) of target K562 cells by NK cells. T control, Transport control.

TAP-deficient patients,3 the absent MHC-I expression is associated with an altered differentiation of CD8 T cells and lack of functional NK cells.7 Although TCRab1 CD8 T cells account for more than 80% of the CD8 T-cell pool in healthy persons, they were severely reduced in the absence of b2m in both siblings. However, CD81 gd T cells were expanded, as previously documented for human TAP deficiency3 and b2m-deficient mice.21 The expansion of this population might be favored by improved access to growth factors, such as IL-15, in the absence of TCRab CD8 T cells22 or by unknown environmental or endogenous antigens. Despite undetectable MHC-I expression, a few TCRab1 CD8 T cells were found in both siblings, particularly in the sister. Inflammatory skin disease can be found in about half of patients with TAP deficiency and those with b2m deficiency.3 The factors triggering and maintaining the discordant skin manifestation in patients with MHC-I deficiency remain unknown. In the absence of regular CD8 T cells, the infiltrating cells comprise autoreactive NK and CD81 gd T cells2 and perforin-producing, possibly CD272CD282 CD4 T cells, as previously described in patients with Wegener granulomatosis23 and in our case. Degranulation of CD81 gd T cells on anti-CD3/CD28 stimulation appeared only partly impaired in rested cells and capable of recovering in vitro. Despite the preserved degranulation, both patients failed to kill allogeneic target cells after in vitro priming. These data indicate that in the absence of polymorphic MHC-I molecules, no alloreactive T-cell precursors are generated.24 Similarly, NK cells degranulated after PMA/ ionomycin or Fc receptor stimulation but did not degranulate and lyse MHC-I–negative targets, as has been previously shown for TAP-deficient patients25 and b2m-deficient mice.26 This functional inactivation of the whole NK compartment protects MHC-I–deficient ‘‘missing-self’’ cells from MHC-unrestricted

lysis.6 Despite these abnormalities in NK and CD8 T-cell activation, MHC-I deficiency is not associated with an overt susceptibility toward viral infection in patients or mice.27-33 Additional protective mechanisms include the induction and recruitment of cytotoxic CD4 T cells33,34 and other effector mechanisms that do not depend on b2m.35-38 In addition to immunologic similarities between b2m deficiency and TAP deficiency, there are several clear-cut differences. Through the formation of heterodimers, b2m not only stabilizes surface expression of MHC-I but also of the other members of the MHC-I family: FcRn, CD1a, HLA-G, HLA-E, and hemochromatosis protein.39 In our patients CD1a, CD1b, and CD1c surface expression was absent on monocyte-derived dendritic cells. The positive staining of Langerhans cells that we observed in the skin was probably because of intracellular accumulation of CD1a molecules in the absence of b2m.40 Although CD1d surface expression appears to be less dependent on b2m expression,41 as described in b2m-deficient mice,42 our patients lacked classical CD1d-dependent invariant natural killer T (iNKT) cells, expanding the spectrum of primary immunodeficiency disorders lacking classical iNKT cells.43 In patients with b2m deficiency, the lack of iNKT cells might be due to impaired intracellular trafficking and glycosylation of CD1d.41 Neither sibling showed laboratory findings of iron overload as observed in b2m-deficient mice because of disturbed expression of hemochromatosis protein.44 Unlike TAP deficiency,3 b2m deficiency presents with hypogammaglobulinemia and hypoalbuminemia caused by hypercatabolism in the absence of regular FcRn expression. Waldmann and Terry8 reported a reduced half-life of serum albumin (6 days; normal, 16.2 6 1 days) and IgG (3.3 days; normal, 22.9 6 4.0 days) in the original b2m-deficient patients. Although b2m deficiency might permit intracellular expression

ARDENIZ ET AL 399

J ALLERGY CLIN IMMUNOL VOLUME 136, NUMBER 2

FIG 5. Inflammatory skin changes in patients with b2m deficiency. A-H, Characteristic morphologic features of a skin biopsy specimen of P1 adjacent to a florid ulcerated lesion. Fig 5, A, Dense fibrosis involving the dermis and subcutaneous tissue accompanied by a reduction in hair follicles and sweat glands. Fig 5, B, Discrete perivascular CD41 infiltrates in the superficial dermis. Fig 5, C, Predominantly CD81 intraepithelial lymphocytes. Fig 5, D, Presence of gd T cells occasionally forming small intraepithelial clusters (arrowhead). Fig 5, E-H, Granulomatous lesions in the deep fibrotic dermis (Fig 5, E and F, arrows) comprising numerous CD681 macrophages, scattered giant cells (Fig 5, G, asterisk), and rare perforin-positive cytotoxic T cells (Fig 5, H). I-L, Fibrotic skin (Fig 5, I) from P2 developing focal granulomatous CD41 inflammatory lesions (Fig 5, J) characterized by a high content of mononuclear CD681 macrophages (Fig 5, K) and CD81 T cells (Fig 5, L) but without evidence of giant cells.

of FcRn, b2m-free expression is associated with decreased function.45 Compatible with the wide expression range of FcRn, its role extends beyond the bidirectional transport of IgG by the human intestinal epithelial cells to intracellular trafficking in MHC class II presentation and cross-presentation of antigens.11 Several findings suggest that altered FcRn expression might contribute to the susceptibility for RTIs in patients with

b2m deficiency.46 Unlike more common primary antibody deficiency disorders, such as common variable immunodeficiency,47 hypogammaglobulinemia in human b2m deficiency is not associated with notably disturbed B-cell differentiation. Given the persistent B-cell memory, one should exploit the capacity of boosting memory by active vaccination of b2mdeficient patients.31

400 ARDENIZ ET AL

Taken together, b2m deficiency has a broad effect on the innate and adaptive immune system through its effects on the expression of several members of the MHC-I family. Clinically, it resembles other MHC-I deficiencies but differs in the associated hypogammaglobulinemia and hypoalbuminemia caused by altered FcRn expression. The clinical presentation can vary, but it can be readily diagnosed by the absence of b2m serum levels. We thank Dr Nihal Mete G€okmen for referring the patients to the € immunology outpatient clinic and G€ul Ozeken, Anna Gsch€opf, Julia Beutelschieß, Daniela Fuest, Christina Schwehr, and Mehmet Arca for excellent technical support. Anti-human MHC-I mAbs (Q1/28, HC10, and B1.23.2) were kindly provided by Soldano Ferrone (Massachusetts General Hospital, Harvard, Boston, Mass).

Clinical implications: b2m deficiency presents like other forms of MHC-I deficiency, mainly through RTIs and possibly granulomatous skin disease. Absent surface expression of FcRn in patients with b2m deficiency leads to hypogammaglobulinemia and hypoalbuminemia.

REFERENCES 1. de la Salle H, Hanau D, Fricker D, Urlacher A, Kelly A, Salamero J, et al. Homozygous human TAP peptide transporter mutation in HLA class I deficiency. Science 1994;265:237-41. 2. Gadola SD, Moins-Teisserenc HT, Trowsdale J, Gross WL, Cerundolo V. TAP deficiency syndrome. Clin Exp Immunol 2000;121:173-8. 3. Zimmer J, Andres E, Donato L, Hanau D, Hentges F, de la Salle H. Clinical and immunological aspects of HLA class I deficiency. QJM 2005;98:719-27. 4. Grandea AG 3rd, Androlewicz MJ, Athwal RS, Geraghty DE, Spies T. Dependence of peptide binding by MHC class I molecules on their interaction with TAP. Science 1995;270:105-8. 5. Ploegh HL, Orr HT, Strominger JL. Major histocompatibility antigens: the human (HLA-A, -B, -C) and murine (H-2K, H-2D) class I molecules. Cell 1981;24: 287-99. 6. Orr MT, Lanier LL. Natural killer cell education and tolerance. Cell 2010;142: 847-56. 7. Zimmer J, Donato L, Hanau D, Cazenave JP, Tongio MM, Moretta A, et al. Activity and phenotype of natural killer cells in peptide transporter (TAP)-deficient patients (type I bare lymphocyte syndrome). J Exp Med 1998;187:117-22. 8. Waldmann TA, Terry WD. Familial hypercatabolic hypoproteinemia. A disorder of endogenous catabolism of albumin and immunoglobulin. J Clin Invest 1990;86: 2093-8. 9. Wani MA, Haynes LD, Kim J, Bronson CL, Chaudhury C, Mohanty S, et al. Familial hypercatabolic hypoproteinemia caused by deficiency of the neonatal Fc receptor, FcRn, due to a mutant beta2-microglobulin gene. Proc Natl Acad Sci U S A 2006;103:5084-9. 10. Simister NE, Mostov KE. An Fc receptor structurally related to MHC class I antigens. Nature 1989;337:184-7. 11. Rath T, Kuo TT, Baker K, Qiao SW, Kobayashi K, Yoshida M, et al. The immunologic functions of the neonatal Fc receptor for IgG. J Clin Immunol 2013;33(suppl 1):S9-17. 12. Bryceson YT, Pende D, Maul-Pavicic A, Gilmour KC, Ufheil H, Vraetz T, et al. A prospective evaluation of degranulation assays in the rapid diagnosis of familial hemophagocytic syndromes. Blood 2012;119:2754-63. 13. Chen BP, Lam V, Kraus EE, DeMars R, Sondel PM. Restriction of Epstein-Barr virus-specific cytotoxic T cells by HLA-A, -B, and -C molecules. Hum Immunol 1989;26:137-47. 14. Warnatz K, Schlesier M. Flowcytometric phenotyping of common variable immunodeficiency. Cytometry B Clin Cytom 2008;74:261-71. 15. de la Salle H, Zimmer J, Fricker D, Angenieux C, Cazenave JP, Okubo M, et al. HLA class I deficiencies due to mutations in subunit 1 of the peptide transporter TAP1. J Clin Invest 1999;103:R9-13. 16. de la Salle H, Saulquin X, Mansour I, Klayme S, Fricker D, Zimmer J, et al. Asymptomatic deficiency in the peptide transporter associated to antigen processing (TAP). Clin Exp Immunol 2002;128:525-31. 17. Zimmer J, Sleiman M, Poli A, Michel T, Hentges F. TAP deficiency is also a cause of bronchiectasis. Thorax 2013;68:490-1.

J ALLERGY CLIN IMMUNOL AUGUST 2015

18. Ploegh HL, Cannon LE, Strominger JL. Cell-free translation of the mRNAs for the heavy and light chains of HLA-A and HLA-B antigens. Proc Natl Acad Sci U S A 1979;76:2273-7. 19. Van Kaer L, Ashton-Rickardt PG, Ploegh HL, Tonegawa S. TAP1 mutant mice are deficient in antigen presentation, surface class I molecules, and CD4-81 T cells. Cell 1992;71:1205-14. 20. Koller BH, Marrack P, Kappler JW, Smithies O. Normal development of mice deficient in beta 2M, MHC class I proteins, and CD81 T cells. Science 1990; 248:1227-30. 21. French JD, Roark CL, Born WK, O’Brien RL. Gammadelta T lymphocyte homeostasis is negatively regulated by beta2-microglobulin. J Immunol 2009; 182:1892-900. 22. French JD, Roark CL, Born WK, O’Brien RL. gd T cell homeostasis is established in competition with ab T cells and NK cells. Proc Natl Acad Sci U S A 2005;102: 14741-6. 23. Lamprecht P, Moosig F, Csernok E, Seitzer U, Schnabel A, Mueller A, et al. CD28 negative T cells are enriched in granulomatous lesions of the respiratory tract in Wegener’s granulomatosis. Thorax 2001;56:751-7. 24. Sullivan LC, Hoare HL, McCluskey J, Rossjohn J, Brooks AG. A structural perspective on MHC class Ib molecules in adaptive immunity. Trends Immunol 2006;27:413-20. 25. Cerundolo V, de la Salle H. Description of HLA class I- and CD8-deficient patients: insights into the function of cytotoxic T lymphocytes and NK cells in host defense. Semin Immunol 2006;18:330-6. 26. Liao NS, Bix M, Zijlstra M, Jaenisch R, Raulet D. MHC class I deficiency: susceptibility to natural killer (NK) cells and impaired NK activity. Science 1991;253:199-202. 27. Manickan E, Rouse BT. Roles of different T-cell subsets in control of herpes simplex virus infection determined by using T-cell-deficient mouse-models. J Virol 1995;69:8178-9. 28. Polic B, Jonjic S, Pavic I, Crnkovic I, Zorica I, Hengel H, et al. Lack of MHC class I complex expression has no effect on spread and control of cytomegalovirus infection in vivo. J Gen Virol 1996;77:217-25. 29. Tay CH, Welsh RM, Brutkiewicz RR. NK cell response to viral infections in beta 2-microglobulin-deficient mice. J Immunol 1995;154:780-9. 30. Wu QH, Trymbulak W, Tatake RJ, Forman SJ, Zeff RA, Shanley JD. Replication of human cytomegalovirus in cells deficient in beta 2-microglobulin gene expression. J Gen Virol 1994;75:2755-9. 31. Epstein SL, Misplon JA, Lawson CM, Subbarao EK, Connors M, Murphy BR. Beta 2-microglobulin-deficient mice can be protected against influenza A infection by vaccination with vaccinia-influenza recombinants expressing hemagglutinin and neuraminidase. J Immunol 1993;150:5484-93. 32. Spriggs MK, Koller BH, Sato T, Morrissey PJ, Fanslow WC, Smithies O, et al. Beta 2-microglobulin-, CD81 T-cell-deficient mice survive inoculation with high doses of vaccinia virus and exhibit altered IgG responses. Proc Natl Acad Sci U S A 1992;89:6070-4. 33. Muller D, Koller BH, Whitton JL, LaPan KE, Brigman KK, Frelinger JA. LCMV-specific, class II-restricted cytotoxic T cells in beta 2-microglobulindeficient mice. Science 1992;255:1576-8. 34. Marusic-Galesic S, Udaka K, Walden P. Increased number of cytotoxic T cells within CD418- T cells in beta 2-microglobulin, major histocompatibility complex class I-deficient mice. Eur J Immunol 1993;23:3115-9. 35. Correa I, Bix M, Liao NS, Zijlstra M, Jaenisch R, Raulet D. Most gamma delta T cells develop normally in beta 2-microglobulin-deficient mice. Proc Natl Acad Sci U S A 1992;89:653-7. 36. Thoma-Uszynski S, Emoto M, Kaufmann SH. CD8alphaalpha T cells in lesions of Listeria monocytogenes-infected beta2m-deficient mice. Microb Pathog 1997;23:101-6. 37. Fujiura Y, Kawaguchi M, Kondo Y, Obana S, Yamamoto H, Nanno M, et al. Development of CD8 alpha alpha1 intestinal intraepithelial T cells in beta 2-microglobulin- and/or TAP1-deficient mice. J Immunol 1996;156:2710-5. 38. Koh YI, Kim HY, Meyer EH, Pichavant M, Akbari O, Yasumi T, et al. Activation of nonclassical CD1d-restricted NK T cells induces airway hyperreactivity in beta 2-microglobulin-deficient mice. J Immunol 2008;181:4560-9. 39. Blumberg RS. Current concepts in mucosal immunity. II. One size fits all: nonclassical MHC molecules fulfill multiple roles in epithelial cell function. Am J Physiol 1998;274:G227-31. 40. Bauer A, Huttinger R, Staffler G, Hansmann C, Schmidt W, Majdic O, et al. Analysis of the requirement for beta 2-microglobulin for expression and formation of human CD1 antigens. Eur J Immunol 1997;27:1366-73. 41. Kim HS, Garcia J, Exley M, Johnson KW, Balk SP, Blumberg RS. Biochemical characterization of CD1d expression in the absence of beta2-microglobulin. J Biol Chem 1999;274:9289-95. 42. Ohteki T, MacDonald HR. Major histocompatibility complex class I related molecules control the development of CD418- and CD4-8- subsets of natural

J ALLERGY CLIN IMMUNOL VOLUME 136, NUMBER 2

killer 1.11 T cell receptor-alpha/beta1 cells in the liver of mice. J Exp Med 1994; 180:699-704. 43. Zeissig S, Blumberg RS. Primary immunodeficiency associated with defects in CD1 and CD1-restricted T cells. Ann N Y Acad Sci 2012;1250:14-24. 44. Miranda CJ, Makui H, Andrews NC, Santos MM. Contributions of beta2-microglobulin-dependent molecules and lymphocytes to iron regulation: insights from HfeRag1(2/2) and beta2mRag1(2/2) double knock-out mice. Blood 2004;103:2847-9.

ARDENIZ ET AL 401

45. Praetor A, Hunziker W. beta(2)-Microglobulin is important for cell surface expression and pH-dependent IgG binding of human FcRn. J Cell Sci 2002;115:2389-97. 46. Spiekermann GM, Finn PW, Ward ES, Dumont J, Dickinson BL, Blumberg RS, et al. Receptor-mediated immunoglobulin G transport across mucosal barriers in adult life: functional expression of FcRn in the mammalian lung. J Exp Med 2002;196:303-10. 47. Wehr C, Kivioja T, Schmitt C, Ferry B, Witte T, Eren E, et al. The EUROclass trial: defining subgroups in common variable immunodeficiency. Blood 2008;111:77-85.