Variability of clinical and laboratory features among patients with ribonuclease mitochondrial RNA processing endoribonuclease gene mutations

Variability of clinical and laboratory features among patients with ribonuclease mitochondrial RNA processing endoribonuclease gene mutations Fotini D. Kavadas, MD,a* Silvia Giliani, PhD,b* Yiping Gu, PhD,a Evelina Mazzolari, MD,b Andrea Bates, BS,a Eleonora Pegoiani, BS,b Chaim M. Roifman, MD, FRCPC,a  and Luigi D. Notarangelo, MDb,c  Toronto, Ontario, Canada, Brescia, Italy, and Boston, Mass Background: Cartilage hair hypoplasia is an autosomal recessive type of metaphyseal chondrodysplasia, caused by mutations in the ribonuclease mitochondrial RNA processing (RMRP) gene. Typical features of cartilage hair hypoplasia include short stature, a predisposition to malignancy, and a variable degree of impairment of cellular immunity. Objective: We sought to describe the heterogeneity of clinical and immunologic phenotype in 12 consecutive patients with RMRP mutations who were referred to 2 different institutions for immunologic evaluation. Methods: We have retrospectively analyzed the clinical and laboratory features in 12 patients with molecular defects in the RMRP gene. T-cell repertoire was investigated by quantitating Vb families’ expression and analyzing their diversity. T-cell receptor excision circle analysis was used to study thymic output. Results: All 12 patients had significant immune abnormalities, leading to severe immune deficiency in 9. CD8 lymphocytopenia was identified as a novel phenotype associated with RMRP mutations. Significant, even intrafamilial, phenotypic heterogeneity was observed. In 3 cases, severe immunodeficiency was the only phenotypic manifestation associated with RMRP mutations, a novel finding. Mutations leading to significant immune defects were most often located in the promoter, and the first case of a compound heterozygote for 2 such mutations is reported. Conclusion: This report broadens the spectrum of phenotypes associated with RMRP mutations and suggests that mutations in this gene should be considered when evaluating patients with From athe Division of Immunology and Allergy, Hospital for Sick Children, University of Toronto; bthe Division of Pediatric Hematology-Oncology, Spedali Civili, and ‘‘Angelo Nocivelli’’ Institute for Molecular Medicine, University of Brescia; and c the Division of Immunology, Children’s Hospital Boston, Harvard Medical School. *These authors contributed equally to this work.  Co-senior authors. Supported by the Canadian Immunodeficiency Society, the Canadian Centre for Primary Immunodeficiency, the Donald and Audrey Campbell Chair of Immunology, the ‘‘Angelo Nocivelli’’ Foundation, and the European Union, grant 006411 (EURO-POLICYPID) to L.D.N. Disclosure of potential conflict of interest: The authors have declared that they have no conflict of interest. Received for publication June 23, 2008; revised July 28, 2008; accepted for publication July 29, 2008. Available online September 22, 2008. Reprint requests: Chaim M. Roifman, MD, FRCPC, Division of Immunology/Allergy and Infection, Immunity, Injury and Repair Program, Hospital for Sick Children, 555 University Avenue, Toronto, Ontario, M5G 1X8, Canada. E-mail: chaim.roifman@ sickkids.ca; or Luigi D. Notarangelo, MD, Division of Immunology, Children’s Hospital Boston, Karp Family Building, 9th Floor, Room 9210, 1 Blackfan Circle, Boston, MA 02115. E-mail: [email protected]. 0091-6749/$34.00 Ó 2008 American Academy of Allergy, Asthma & Immunology doi:10.1016/j.jaci.2008.07.036

1178

combined immune deficiency, regardless of the presence of other manifestations. (J Allergy Clin Immunol 2008;122:1178-84.) Key words: Cartilage hair hypoplasia, RNase mitochondrial RNA processing endoribonuclease, severe combined immunodeficiency, Omenn syndrome, CD8 lymphopenia

Cartilage hair hypoplasia (CHH) was originally described by McKusick et al1 in the Amish population as an autosomal recessive form of metaphyseal chondrodysplasia with short stature and light-colored hypoplastic hair. Patients may also present with bone marrow dysplasia, abnormal intestinal plexus neurons, impaired spermatogenesis, and predisposition to malignancy.2-5 The disease is more common among selected populations, such as the Amish and the Finns.2 The CHH gene encodes for the untranslated RNA component of the ribonuclease mitochondrial RNA processing (RMRP) complex, which also includes multiple proteins.6 A common 70A>G mutation has been identified in 92% of Finnish and 48% of non-Finnish CHH patients7; however, more than 70 different mutations have been identified worlwide.6-15 Disease-causing mutations in the RMRP gene result in disruption of ribosomal processing and cell cycle progression in rapidly dividing cells such as lymphocytes and chondrocytes, thus explaining the pleiotropy of clinical manifestations.6,13,14,16,17 A variable degree of immunodeficiency has been reported in CHH that predominantly affects cell-mediated immunity.2,18,19 Some affected individuals have been shown to present with severe combined immunodeficiency (SCID), prompting the need for hematopoietic cell transplantation (HCT).6,20 We have recently shown that mutations in the RMRP gene may also cause Omenn syndrome (OS).21 HCT has resulted in successful immune reconstitution in patients with CHH who present with SCID or with OS.20-23 On the other hand, immunologic abnormalities may be subtle or even absent in other affected individuals. Although no strict genotype-phenotype correlation has been established in CHH, the common 70A>G mutation has been often associated with a relatively mild immunologic phenotype.5 On the other hand, no extensive studies of immune function have been performed on a significant number of patients with CHH with other mutations. We report on the clinical and immunologic presentation and long-term outcome in 12 patients with RMRP mutations from 10 unrelated families. We show that a severe immunologic phenotype is most often associated with RMRP mutations other than 70A>G, and demonstrate for the first time that RMRP mutations may lead to significant immune deficiency without short stature or skeletal involvement, thus expanding the spectrum of RMRP-associated clinical phenotypes.

J ALLERGY CLIN IMMUNOL VOLUME 122, NUMBER 6

Abbreviations used CHH: Cartilage hair hypoplasia HCT: Hematopoietic cell transplantation OS: Omenn syndrome RMRP: Ribonuclease mitochondrial RNA processing endoribonuclease SCID: Severe combined immunodeficiency sj: Signal joint TREC: T-cell receptor excision circle

KAVADAS ET AL 1179

(Applied Biosystems, Rotkreuz, Switzerland). The number of TRECs in a given sample was compared with a value obtained with 10-fold serial dilutions of an internal standard provided by Dr Daniel Douek (Vaccine Research Center, National Institute of Allergy and Infectious Diseases, Bethesda, Md).

Quantification of T-cell receptor Vb genes Analysis of T-cell repertoire was performed by quantifying expression of T-cell receptor Vb families using flow cytometry (Coulter ELITE, Beckman Coulter, Miami, Fla) and by heteroduplex analysis of single amplified T-cell receptor beta variable (TCRBV) products, as previously described.27,28

METHODS Two centers specializing in the management and diagnosis of primary immunodeficiency participated in this study: the Hospital for Sick Children in Toronto, Ontario, Canada, and the Department of Pediatrics at the University of Brescia, Italy. This is a retrospective analysis of data derived from patient charts and databases over a period of 16 years. All 12 patients who were diagnosed with RMRP mutations and were referred to the 2 centers with clinical or laboratory evidence of immunologic abnormalities were included. Three of these patients have been previously reported.21,22 Written consent for participation in this study was obtained from the patients (if of appropriate age) or their parents or legal guardians.

Mutation analysis of the RMRP gene Genomic DNA was extracted by standard procedures from blood samples and was analyzed by PCR amplification and direct sequencing of RMRP. PCR primers were designed to amplify a genomic DNA fragment between 60 bp upstream and 10 bp downstream of the RMRP RNA gene transcript.24 The PCR primers used were 59-TCACGCCACCAACTTTCTCACC-39 and 59CTGCAGTGAGCCGTGGTCTCG-39. The PCR products were then analyzed by direct cycle sequencing using the Big Dye Terminator kit (Applied Biosystems, Foster City, Calif) on an ABI PRISM 3130 automatic sequencer (Applera, Norwalk, Conn). For detection of insertions, the PCR fragments were cloned into a TOPO-TA cloning vector (Invitrogen, Paisley, United Kingdom). At least 10 single colonies were picked and sequenced.

Lymphocyte markers and T-cell proliferative responses The surface phenotype of blood mononuclear cells obtained by FicollHypaque density gradient centrifugation was determined by flow cytometry as described.25 Lymphocyte proliferative responses to mitogens including PHA were determined by tritiated thymidine incorporation using the microtiter plate technique.26 All assays were performed in triplicate and were compared with those simultaneously performed on normal controls.

Serum concentration of immunoglobulins Serum concentrations of immunoglobulins were measured by nephelometry. Levels of serum antibodies to tetanus were measured by ELISA, and polio antibody titers were determined by complement fixation.26,27

Quantification of T-cell receptor excision circles by real-time PCR The number of signal joint (sj) T-cell receptor excision circles (TRECs), adjusted to CD41 and CD81 T subsets, was determined by real-time quantitative PCR, as previously described.27 In brief, genomic DNA was isolated from PBMCs. For TREC PCR, the following primers and probe were used: sj-50 forward: CACATCCCTTTCA ACCATGCT (900 nmol/L); sj-30 reverse: GCCAGCTGCAG GGTTTAGG (900 nmol/L); and the oligo 50 FAM-ACCTCTGGTTTTT GTAAAGGTGCCCACT- TAMRA p-30 (250 nmol/L) as a deletion probe. PCR (2 minutes at 508C followed by 958C for 10 minutes, then 40 cycles at 958C for 15 seconds and 608C for 1 minute) was performed in the ABI PRISM 7900 Sequence Detector TaqMan system

RESULTS Patients At the time of diagnosis, patients ranged in age from 2.5 months to 44 years (Table I). The male-to-female ratio was 1:2. Six patients were of French or Northern European descent, 1 was Finnish, and 5 were Italians. Only 1 patient was born to consanguineous parents, and there was only 1 family in our series with multiple affected siblings (patients 10-12). No patients were lost to follow-up, and the median length of follow-up from presentation was 7.09 years (ranging from 0.5 year to 22 years). Eleven of 12 patients (92%) are still alive, with only 1 death (patient 10). Clinical presentation The main clinical features observed in the 12 patients are reported in Table I. Short stature, sparse hair, and skeletal abnormalities are typical features of CHH and were seen in 9 patients (67%). Normal hair was documented in 3 patients (9, 10, and 12). With the exception of patient 12, all had low stature (<3rd percentile) at the time of diagnosis. This was associated with low weight (<3rd percentile) in 8 patients. Interestingly, after HCT, stature remained low in 4 but normalized (50th percentile) in 2 patients (8 and 9). Search for skeletal abnormalities by radiograph was performed in 7 patients. Four of these had evidence of metaphyseal dysplasia, with irregularity, serration, and sclerosis of the metaphyseal edges in the femur (patients 2-5), and shortening, invagination, and sclerosis of the metacarpal metaphyses and cone-shaped or cup-shaped deformities of the phalangeal metaphyses (patients 2-4). The remaining 3 patients (1, 8, and 9) had no evidence of metaphyseal dysplasia. Altogether, typical features of CHH (persistently low stature, metaphyseal dysplasia, and/or hair abnormalities) were present in 9 of 11 surviving patients. A history of infections suggestive of immune deficiency was documented in 8 patients. Of these, 4 patients (1-3 and 8) had typical features of OS (extensive erythroderma, protracted diarrhea, lymphadenopathy, enlarged liver and spleen, eosinophilia), and 2 of them (2 and 3) also had Pneumocystis jiroveci interstitial pneumonia. One patient (10) presented at 3 months of age with failure to thrive, interstitial pneumonia caused by cytomegalovirus infection, Aspergillus pneumonia, and protracted diarrhea. She developed severe neurologic deterioration, with seizures and hypotonia, and died 6 days after admission. A diagnosis of SCID was established in this infant on the basis of clinical and laboratory findings. Her youngest sister (patient 12) developed a transient rash at 4 months of age and bronchiolitis caused by respiratory syncytial virus at 6 months of age, before progressive deterioration of immune function was discovered.

1180 KAVADAS ET AL

J ALLERGY CLIN IMMUNOL DECEMBER 2008

TABLE I. Clinical features in 12 patients with RMRP mutations

General features Age at diagnosis Sex Outcome Clinical features Failure to thrive Absent or sparse hair Erythroderma Lymphadenopathy Hepatomegaly Splenomegaly Respiratory infections Protract/recurrent diarrhea Other features Growth Height (<1 y)* Weight (<1 y)* Height (current)* Weight (current)*

P1

P2

P3

P4

3.5 mo M A&W, 1.5 y after MUD-HCT

5 mo F A&W 14 y after MUD- HCT

3 mo F A&W 22 y after MSD- HCT

3 mo F A&W 4 y after MUD-HCT

6y F Alive, 8 y

1 1 Severe 1 1 1 1

1 1 Severe 1 1 1 P jiroveci

1 1 Mild 1 1 1 P jiroveci

1 1 -

1 1 Bronchiectasis

1 Oral thrush

1

-

Chronic OM

<3rd <3rd <3rd 25th

<3rd <3rd <3rd 50th

<3rd <3rd <3rd 25th

<3rd 25% birth <3rd <3rd

<3rd <3rd <3rd <3rd

P5

A&W, Alive and well; CMV, cytomegalovirus; F, female; HCT, hematopoietic cell transplantation; IP, interstitial pneumonia; M, male; MSD, matched sibling donor; MUD, matched unrelated donor; OM, otitis media; P, patient; RSV, respiratory syncytial virus. *Percentile.

TABLE II. Immunologic parameters in 12 patients with RMRP mutations

Age at evaluation ALC (cells/mL) CD31 (cells/mL) CD41 (cells/mL) CD41CD45RA1/CD41 (%) CD81 (cells/mL) CD191 (cells/mL) CD161/CD561 (cells/mL) TRECs (per 106 PBMCs)* T-cell repertoire Proliferative response to PHA (SI)  Proliferative response to anti-CD3 (SI)  IgG (mg/dL) IgA (mg/dL) IgM (mg/dL) IgE (kU/L)

P1

P2

P3

P4

P5

P6

4 mo 1300 1112 1034 ND 17 373 317 Low Skewed 36 (230) 5 (70) Und 14 40 1

5 mo 1560 340 289 ND 21 170 980 ND Skewed 3 (64) 5 (80) 560 <0.1 <0.1 ND

8 mo 950 237 133 ND 142 475 47 ND Skewed 14 (107) 7 (92) 620 <0.1 112 ND

3 mo 840 105 51 ND 20 480 229 Low ND 30 (231) ND 100 <0.1 60 ND

8y 1100 616 149 ND 344 75 299 Low Skewed 15 (451) 6 (82) 1200 240 210 ND

69 y 1400 60.4 (841) 40 (557) ND 21 (294) 11.7 (163) 24.9 (347) ND ND 436 (947) 213 (207) 910 540 70 ND

P, Patient; ND, not done; Und, undetectable. *Normal values for TRECs are >100,000 if <1 y old, >30,000 if 1-3 y.  Values for healthy controls are reported in parentheses. àCD31 T-cell receptor ab1 cells were 3.5%, and T-cell receptor gd1 cells were 39.2% of lymphocytes.

Patient 5 had multiple episodes of otitis media and pneumonia, resulting in bronchiectasis. Two patients (2 and 9) had oral thrush, and patient 9 also had multiple episodes of diarrhea.

Immunologic evaluation As shown in Table II, 10 patients (83.3%) already demonstrated a low absolute lymphocyte number at the time of presentation. One patient (12) started with normal lymphocyte numbers that gradually declined over time. A progressive decline of CD81

lymphocytes was observed in patient 9. Eventually all 12 patients, irrespective of the absolute lymphocyte number, had a low number of CD31 T cells and CD81 T cells for age. Eleven of them also had a reduced count of CD41 T cells. In 4 patients (1, 2, 8, and 12), the count of CD81 lymphocytes was disproportionately low relative to the number CD41 cells, reminiscent of CD81 T-cell lymphocytopenia. Six patients (2, 5, 6, 8-10) had CD191 values that were below normal. Natural killer cells were normal in 9 patients, low in 2, and elevated in 1. The proliferative response to PHA ranged from 1.1% to 59% of controls, whereas in vitro responses to CD3 cross-

KAVADAS ET AL 1181

J ALLERGY CLIN IMMUNOL VOLUME 122, NUMBER 6

TABLE I. (Continued) P6

P7

P8

P9

P 10

P 11

44 y F Alive, 60 y

18 mo M Alive, 24 mo

2.5 mo F A&W 9 y after MSD-HCT

5 mo M A&W 5 y after MUD-HCT

3 mo F deceased (3 mo)

5y M Alive, 7 y

3y F Alive, 5 y

1 -

1 -

1 1 Severe 1 1

-

Transient rash

1

-

1 -

Arthritis

-

1

1 Oral thrush

<3rd <3rd <3rd <3rd

<3rd 50th <3rd <3rd

<3rd <3rd <50th 10th

<3rd 3rd <50th 50th-75th

1 IP (CMV), Aspergillus pneumonia 1 Seizures <3rd <3rd Deceased Deceased

P 12

Bronchiolitis (RSV) -

<3 10th-25th <3 10th-25th

25th-50th 25th 25th-50th 25th

TABLE II. (Continued) P7

P8

2y 1510 57.5 (867) 28.5 (430) ND 13.5 (204) 27.8 (419) 12.1 (182) ND ND 156 (380) 256 (57.3) 830 60 90 ND

2.5 mo 1780 1121 1032 <1 160 214 392 ND ND 10 (323) 26 (245) 257 58 102 11

P9 5 mo 930 248 107 2 133 232 439 ND Skewed ND ND 1280 276 272 11.6

2 y 2 mo 1105 636à 102 1 4 152 146 75 Skewed 4 (154) 8 (161) 1835 215 85 20.1

P 10

P 11

P 12

P 12

P 12

4 mo 260 35 21 ND 15 152 45 ND ND <1 (89) <1 (35) ND ND ND ND

2 y 1 mo 2480 1148 543 27 312 756 305 ND ND 113 (376) 169 (246) 595 29 52 ND

Birth 3500 1533 1011 21 308 1064 395 Normal Polyclonal 223 (375) 179 (246) ND ND ND <2

1 y 4 mo 1500 681 382 37 55 417 333 Low Skewed 99 (139) ND 860 30 85 5.3

2 y 11 mo 1120 575 351 13 62 329 135 Und Skewed 16 (228) 56 (155) 1004 59 61 17.8

linking ranged from 7.1% to >100% of controls. A milder or delayed clinical phenotype was observed in patients 6, 7, 11, and 12, who had better proliferative responses to PHA and anti-CD3 (ranging from 30% to 59% and 69% to 100% of controls, respectively) at the time of initial assessment. However, patient 12 experienced a decline in T-cell function with the PHA response falling from 59% at birth to 7% at almost 3 years of life, which coincided with an increased rate of infection. T-cell receptor repertoire was determined by quantitative PCR and flow cytometry in 4 patients (1-3, 5) and by heteroduplex

analysis in 2 patients (9 and 12). All patients tested had a skewed repertoire (Table II) that in most cases was characterized by overrepresentation of some Vb families and underrepresentation of others. For example, patient 1 (who presented with an OS phenotype) had an overrepresentation of Vb 13.1 but an underrepresentation of 12 other Vb families compared with control specimen, whereas patient 5 had an overrepresentation of 3 families, and 18 other Vb families were underrepresented (Fig 1). Thymopoiesis was assessed by measuring TRECs, a byproduct of the variable (diversity) joining recombination process

1182 KAVADAS ET AL

J ALLERGY CLIN IMMUNOL DECEMBER 2008

FIG 1. Proportion of circulating CD31 lymphocytes expressing various T-cell receptor (TCR) Vb families in healthy controls (white bars), patient 1 (solid bars), and patient 5 (hatched bars). Mean values 6 SDs are shown for healthy controls.

that takes place during thymic T-cell development. Of 5 patients who were analyzed, 4 had a low number of TRECs on initial assessment, and another 1 patient (12) started with normal TREC levels that dropped to undetectable levels during follow-up. Patients 2 and 3 had thymic biopsies that revealed very similar pathology, with a loss of corticomedullary architecture, signs of thymic dysplasia, and marked depletion of Hassall corpuscles. The epithelial islands were arranged in small nests of cohesive spindle-shaped cells separated by fibrovascular septa. Immunohistochemical analysis showed few CD31 thymocytes in the medullary region that expressed either CD4 or CD8. No thymic tissue was detected at postmortem analysis in patient 10. In this patient, severe lymphoid depletion, with complete architectural distortion and clear evidence for cytomegalovirus infection, was documented in latero-cervical lymph nodes, appendix, and spleen.

Mutation analysis All patients but 1 were compound heterozygotes for RMRP mutations. Sixteen different mutations were identified (Table III). Only 2 patients had the common 70A>G mutation. The other mutations consisted of single nucleotide substitutions, duplications, triplications, and insertions. Patient 1 had a 15-bp duplication at position -25 in the promoter region on 1 allele and a 154G>C nucleotide substitution in the coding region of the other allele. Patient 2 had on 1 allele a 7-bp deletion followed by a 28-bp insertion in the transcription regulatory region at position -11. The second allele carried a 4C>T point mutation. Patient 3 had a 7-nucleotide insertion at position -20 (ATCTGTG) located between the TATA box and the transcription initiation site on the maternal allele and a 240A>C substitution on the paternal allele. Similarly, patients 10 to 12 were compound heterozygotes for a duplication in the promoter region on 1 allele, and a single

TABLE III. RMRP mutations and clinical phenotype Patient no.

1 2 3 4 5 6 7 8 9 10 11 12

Allele 1

Allele 2

Phenotype

-25_-11dup Complex  -20_-14dup -9_-2dup 242 A>G 242 A>G 70 A>G -25_-13dup -4_-1 dup -25_-5dup -25_-5dup -25_-5dup

154 G>C 4C>T 240 A>C 70A>G 146 G>A 193 G>A 70A>G *5T>C -21_-14trip 146 G>A 146 G>A 146 G>A

CHH1OS1CD8 lymphopenia CHH1OS1CD8 lymphopenia CHH1OS1CD8 lymphopenia CHH1CID1CD8 lymphopenia CHH1CID1CD8 lymphopenia CHH1CD8 lymphopenia CHH1CD8 lymphopenia OS CID1CD8 lymphopenia SCID CHH CID1CD8 lymphopenia

CID, Combined immune deficiency.  -11_-5delCTGAGGAins28bp.

nucleotide substitution in the coding region on the other allele. Patient 4 inherited from her mother an 8-nucleotide duplication at position -9, and the common 70A>G mutation from her father. Patients 5 to 7 had point mutations on both alleles in the transcribed region of the gene, with patient 7 homozygous for the 70A>G mutation. Patient 8 carried a duplication in the promoter region on one allele, and a single nucleotide substitution (*5T>C) after the termination codon on the second allele. One patient (9) had duplications in the promoter region on both alleles. Interestingly, all 5 patients who had SCID (patient 10) or OS (patients 1-3 and 8) had an insertion or duplication on one allele. Within 1 family, the same mutation, a 21-bp duplication on one allele and a single base pair substitution at position 146 on the other allele, resulted in very different clinical and immunologic phenotypes in 3 affected siblings (patients 10-12), 1 of whom had SCID, the second presenting with classic CHH, and the third with

J ALLERGY CLIN IMMUNOL VOLUME 122, NUMBER 6

progressive combined immune deficiency without typical features of CHH. None of these mutations were identified in 100 alleles from 50 unrelated controls for each of the ethnic groups to which the patients reported here belong.

Management and bone marrow transplantation Hematopoietic cell transplantation was performed in 6 patients, all of whom are alive at a median of 7.0 years after transplantation (range, 1-15 years; mean, 7.3 years). The reasons for HCT included OS in 4 (patients 1-3 and 8) and combined immune deficiency in 2 (patients 4 and 9). Of these 6 patients, 4 received HCT from a matched unrelated donor, whereas 2 (patients 3 and 8) had a matched sibling. The other patient with combined immunodeficiency (5) presented rather late, at the age of 6 years, when she already had significant morbidity related to bronchiectasis. Robust and sustained T-cell and B-cell engraftment has been observed in all 5 transplanted patients who are alive with a follow-up of at least 4 years after HCT. One of these patients (9) developed antinuclear and antidouble-stranded DNA antibodies after transplant, and then developed nephrotic syndrome, which was reversed with prednisone. None of the patients have developed malignancies. DISCUSSION We describe a group of patients with RMRP mutations and clinical and/or laboratory evidence of immunodeficiency. The mitochondrial RNA-processing ribonuclease is involved in ribosome assembly and cell-cycle regulation.5,11,16,17 Mutations of the RMRP gene in human beings have been associated with a spectrum of autosomal recessive skeletal dysplasias that range from metaphyseal dysplasia without hypotrichosis (MIM 250460) to CHH to anauxetic dysplasia (MIM 607095).11 Among these, immunodeficiency has been reported only in CHH. Although patients with CHH may show wide variability in their immune function, ranging from normal to SCID phenotype, little is known about the mechanisms that underlie such heterogeneity. The largest study of immune function in patients with CHH was performed in Finland, where the vast majority of patients share the 70A>G mutation.18 In that study, abnormalities of cellular immunity were common, but they were rarely severe enough to cause significant clinical problems. Furthermore, in their analysis of 22 patients with CHH with various RMRP mutations, Hermanns et al15 have shown that only a minority of them had significant cellular immunodeficiency. By testing in vitro the impact of RMRP mutations on messenger RNA (mRNA) and ribosomal RNA cleavage, Thiel et al5 have recently demonstrated that mutations that affect cyclin B2 mRNA cleavage are associated with milder skeletal phenotypes, but more pronounced immunodeficiency and hematologic abnormalities. However, the fact that few data are available on patients with CHH with severe immunodeficiency represents a significant limitation to genotype-phenotype correlation studies in this disease. We have reported 12 patients with clinical and/or laboratory evidence of immunodeficiency associated with RMRP mutations. The cohort of patients studied here has an obvious selection bias but has allowed us to provide novel and unanticipated findings that enlarge the phenotypic and molecular spectrum of disorders caused by RMRP mutations.

KAVADAS ET AL 1183

All of the patients described here had immune aberrations regardless of their clinical manifestations. All had low circulating CD31 T cells and significantly reduced in vitro mitogenic responses. Moreover, in all 8 patients tested, the T-cell receptor repertoire, as determined by quantitation of Vb families expressed and analysis of their diversity, was severely restricted. Demonstration of abnormally low TREC levels in all 5 patients tested provided further evidence for poor thymic function. This finding is consistent with the lack of thymus tissue in the postmortem analysis of patient 10 and the severe thymic abnormalities observed in patients 2 and 3. In keeping with this, 9 of 12 patients had a diagnosis of SCID (n 5 1), OS (n 5 4), or combined immunodeficiency (n 5 4). We had previously reported 2 patients with OS and CHH caused by RMRP mutations.21 Two additional patients are reported here. In contrast with what is typically observed in patients with OS, IgE serum levels were normal in both patients (1 and8) with OS and RMRP mutations in which they were measured. We have shown that the level of CD81 T cells was reduced in all patients and was disproportionately low in 4 of them. Such profound CD81 T-cell lymphocytopenia was reminiscent of patients with zeta-associated protein of 70 kDa (ZAP-70) deficiency, who may also have extensive erythematous rash.29,30 It is not immediately clear why patients with RMRP mutations may have CD81 T-cell lymphocytopenia. In yeast, mutations in RMRP lead to a prolonged and abnormal arrest in cell division as the result of an exit from mitosis defect caused by failure in degradation of cyclin B mRNA.16,17 The maturation of thymocytes as well as immune responses is dependent on a high rate of cell division. It is conceivable that RMRP has a similar role in human beings, and perhaps maturation and selection of CD81 cells might be more sensitive than CD41 cells to this mechanism. Following the recognition that RMRP mutations may lead to a spectrum of immunologic abnormalities, we have included screening for RMRP mutations in all patients with otherwise undefined combined immunodeficiency, regardless of the presence of skeletal abnormalities. In our series, 3 patients (8, 9, and 12) had severe immune deficiency without typical features of CHH (sparse hair, persistently low stature, skeletal abnormalities). To our knowledge, this is the first time that RMRP mutations are reported to be associated with a severe immunologic phenotype in the absence of skeletal involvement. Thiel et al5 reviewed the phenotype of 52 patients with respect to the degree of bone dysplasia, hair hypoplasia, immune deficiency, and hematologic abnormalities, and found that all patients had at least some degree of bone dysplasia and/or hair abnormality. Hermanns et al15 reported on 27 patients with RMRP mutations, 2 of whom had initially normal length, but then developed short stature (below 3rd percentile). In that study, growth insufficiency and short-limbed dwarfism were confirmed to be the most typical signs associated with RMRP mutations. Our novel finding that RMRP defects may occur with immune deficiency only, without significant skeletal involvement, expands the clinical spectrum associated with RMRP mutations. Mutations of the RMRP gene may lead to extreme phenotypic variability, even within the same family, as shown for patients 10 to 12, 1 of whom (11) had typical CHH, whereas his siblings, patients 10 and 12, had SCID and combined immune deficiency with progressive CD81 T-cell lymphocytopenia, respectively, without other features of CHH. We have also provided novel findings on the molecular pathophysiology of RMRP defects in human beings. Affected

1184 KAVADAS ET AL

subjects from 6 of 10 of the families described here carried at least 1 allele with mutations in the promoter region of the RMRP gene, and patient 9 carried such mutations on both alleles. To our knowledge, this is the first case in which mutations in the promoter region of the RMRP gene have been identified on both alleles of an affected individual. Promoter mutations had been originally thought to result in complete silencing of the RMRP gene transcription.6 Accordingly, it had been hypothesized that the presence of such mutations on both alleles may not be compatible with life. More recent data have shown that mutations that lengthen the interval between the TATA box and the transcription initiation site are permissive for gene transcription, albeit at markedly reduced levels.31 Overall, our data indicate that promoter mutations may be overrepresented among patients with RMRP mutations that present with severe immunodeficiency. In summary, we have demonstrated that RMRP mutations may associate with an expanding spectrum of clinical and immunologic phenotypes. Accordingly, defects in this gene should be considered in the evaluation of patients with combined immunodeficiency (including SCID, OS, and CD81 T-cell lymphocytopenia) or with declining T-cell function, especially (but not exclusively) when associated with short stature. We thank Dr Luisa Imberti for the analysis of TRECs and of T-cell repertoire and the collaboration of all families with RMRP mutations reported.

Clinical implications: RMRP mutations may cause of spectrum of immunodeficiency phenotypes in human beings, even without skeletal involvement. RMRP defects should be considered in patients with combined immunodeficiency of unknown origin. REFERENCES 1. McKusick VA, Eldridge R, Hostetler JA, Ruangwit U, Egeland JA. Bull Johns Hopkins Hosp 1965;116:231-72. 2. Makitie O, Kaitila I. Cartilage-hair hypoplasia: clinical manifestations in 108 Finnish patients. Eur J Pediatr 1993;152:211-7. 3. Ma¨kitie O, Sulisalo T, de la Chapelle A, Kaitila I. Cartilage-hair hypoplasia. J Med Genet 1995;32:39-43. 4. Kuijpers TW, Ridanpaa M, Peters M, de Boer I, Vossen JM, Kaitila I, et al. Short limbed dwarfism with bowing, combined immunodeficiency, and the late onset aplastic anemia caused by novel mutations in the RMPR gene. J Med Genet 2003;40:761-6. 5. Thiel CT, Mortier G, Kaitila I, Reis A, Rauch A. Type and level of RMRP functional impairment predicts phenotype in the cartilage hair hypoplasia-anauxetic dysplasia spectrum. Am J Hum Genet 2007;81:519-29. 6. Ridanpa¨a¨ M, van Eenennaam H, Pelin K, Chadwick R, Johnson C, Yuan B, et al. Mutations in the RNA component of RNase MRP cause a pleiotropic human disease, cartilage hair hypoplasia. Cell 2001;104:195-203. 7. Ridanpa¨a¨ M, Sistonen P, Rockas S, Rimoin DL, Ma¨kitie O, Kaitila I. Worldwide mutation spectrum in cartilage hair hypoplasia: ancient founder of the major 70A to G mutation of the untranslated RMRP. Eur J Hum Genet 2002;10:439-47. 8. Bonafe´ L, Dermitzakis ET, Unger S, Greenber CR, Campos-Xavier BA, Zankl A, et al. Evolutionary comparison provides evidence for pathogenicity of RMRP mutations. PLoS Genet 2005;1:e47. 9. Bonafe L, Schmitt K, Eich G, Giedion A, Superti-Furga A. RMRP gene sequence analysis confirms a cartilage hair hypoplasia variant with only skeletal manifestations and reveals a high density of single nucleotide polymorphisms. Clin Genet 2002;61:146-51.

J ALLERGY CLIN IMMUNOL DECEMBER 2008

10. Nakashima E, Mabuchi A, Kashimada K, Onishi T, Zhang J, Ohashi H, et al. RMRP mutations in Japanese patients with cartilage hair hypoplasia. Am J Med Genet A 2003;123:253-6. 11. Thiel CT, Horn D, Zabel B, Ekici AB, Salinas K, Gebhart E, et al. Severely incapacitating mutations in patients with extreme short stature identify RNA processing endoribonucelase RMRP as an essential growth regulator. Am J Hum Genet 2005; 77:795-806. 12. Hermanns P, Bertuch AA, Bertic TK, Dawson B, Schmitt ME, Shaw C, et al. Consequences of mutations in the non coding RMRP RNA in cartilage-hair hypoplasia. Hum Mol Genet 2005;14:3723-40. 13. Hirose Y, Nakashima E, Hirofumi O, Mochizuki H, Bando Y, Ogata T, et al. Identification of novel RMRP mutations and specific founder haplotypes in Japanese patients with cartilage-hair hypoplasia. J Hum Genet 2006;51:706-10. 14. Mun˜oz-Robles J, Allende LM, Clemente J, Calleja S, Varela P, Gonzalez L, et al. A novel RMRP mutation in a Spanish patient with cartilage-hair hypoplasia. Immunobiology 2006;211:753-7. 15. Hermanns P, Tran A, Munivez E, Carter S, Zabel B, Lee B, et al. RMRP mutations in cartilage-hair hypoplasia. Am J Med Genet 2006;140:2121-30. 16. Cai T, Aulds J, Gill T, Cerio M, Schmitt ME. The Saccharomyces cerevisiae RNase mitochondrial RNA processing is critical for cell cycle progression at the end of mitosis. Genetics 2002;161:1029-42. 17. Gill T, Cai T, Aulds J, Wierzbicki S, Schmitt ME. RNase MRP cleaves the CLB2 mRNA to promote cell cycle progression: novel method of mRNA degradation. Mol Cell Biol 2004;24:945-53. 18. Ma¨kitie O, Kaitila I, Savilahti E. Susceptibility to infections and in vitro immune functions in cartilage-hair hypoplasia. Eur J Pediatr 1998;157:816-20. 19. Ma¨kitie O, Kaitila I, Savilahti E. Deficiency of humoral immunity in cartilage-hair hypoplasia. J Pediatr 2000;137:487-92. 20. Berthet F, Siegrist CA, Ozsahin H, Tuchschmid P, Eich G, Superti-Furga A, et al. Bone marrow transplantation in cartilage-hair hypoplasia: correction of the immunodeficiency but not of the chondrodysplasia. Eur J Pediatr 1996;155: 286-90. 21. Roifman CM, Gu Y, Coehn A. Mutations in the RNA component of RNase mitochondrial RNA processing might cause Omenn syndrome. J Allergy Clin Immunol 2006;117:897-903. 22. Guggenheim R, Somech R, Grunebaum E, Atkinson A, Roifman CM. Bone marrow transplantation for cartilage-hair-hypoplasia. Bone Marrow Transplant 2006; 38:751-6. 23. Mazzolari E, DeMartiis D, Forino C, Lanfranchi A, Giliani S, Marzollo R, et al. Single-center analysis of long-term outcome after hematopoietic cell transplantation in children with congenital severe T-cell immunodeficiency. Immunol Res. 2008; July 1 [epub ahead of print]. 24. Sulisalo T, van der Burgt I, Rimoin DL, Bonaventure J, Sillence D, Campbell JB, et al. Genetic homogeneity of cartilage hair hypoplasia. Hum Genet 1995;95: 157-60. 25. Roifman CM, Hummel D, Martinex-Valdez H, Thorner P, Doherty PJ, Pan S, et al. Depletion of CD81 cells in human thymic medulla results in selective immune deficiency. J Exp Med 1989;170:2177-82. 26. Sharfe N, Shahar M, Roifman CM. An interleukin-2 receptor gamma chain mutation with normal thymus morphology. J Clin Invest 1997;100:3036-43. 27. Arpaia E, Shahar M, Dadi H, Cohen A, Roifman CM. Defective T cell receptor signaling and CD81 thymic selection in humans lacking ZAP-70 kinase. Cell 1994;76:947-58. 28. Signorini S, Imberti L, Pirovano S, Villa A, Facchetti F, Ungari M, et al. Intrathymic restriction and peripheral expansion of the T-cell repertoire in Omenn syndrome. Blood 1999;94:3468-78. 29. Katamura K, Tai G, Tachibana T, Yamabe H, Ohmori K, Mayumi M, et al. Existence of activated and memory CD41 T cells in peripheral blood and their skin infiltration in CD8 deficiency. Clin Exp Immunol 1999;115:124-30. 30. Turul T, Tezcan I, Artac H, de Bruin-Versteeg S, Barendregt BH, Reisli I, et al. Clinical heterogeneity can hamper the diagnosis of patients with ZAP70 deficiency. Eur J Pediatr 2008; May 29 [epub ahead of print]. 31. Nakashima E, Tran JR, Welting TJ, Pruijn GJ, Hirose Y, Nishimura G, et al. Cartilage hair hypoplasia mutations that lead to RMRP promoter inefficiency or RNA transcript instability. Am J Med Genet A 2007;143:2675-81.