ADA-deficient SCID is associated with a specific microenvironment and bone phenotype characterized by RANKL/OPG imbalance and osteoblast insufficiency

IMMUNOBIOLOGY

ADA-deficient SCID is associated with a specific microenvironment and bonephenotype characterized by RANKL/OPG imbalance and osteoblast insufficiencyAisha V. Sauer,1,2 Emanuela Mrak,3 Raisa Jofra Hernandez,1 Elena Zacchi,3 Francesco Cavani,4 Miriam Casiraghi,5

Eyal Grunebaum,6 Chaim M. Roifman,6 Maria C. Cervi,7 Alessandro Ambrosi,8 Filippo Carlucci,9 Maria Grazia Roncarolo,1,2

Anna Villa,1,10 Alessandro Rubinacci,3 and Alessandro Aiuti1,11

1San Raffaele Telethon Institute for Gene Therapy (HSR-TIGET), Milan, Italy; 2Universita Vita-Salute San Raffaele, Milan, Italy; 3Bone Metabolic Unit, ScientificInstitute San Raffaele, Milan, Italy; 4Department for Anatomy and Histology, University of Modena, Modena, Italy; 5Pediatric Clinical Research Unit, HSR-TIGET,Milan, Italy; 6Division of Allergy and Clinical Immunology, Hospital for Sick Children, Toronto, ON; 7Department of Puericultura and Pediatria, University of SaoPaulo, Ribeirao Preto, Brasil; 8University Statistics Centre for Biomedical Sciences (CUSSB), San Raffaele University, Milan, Italy; 9Institute for Biochemistry andEnzymology, University of Siena, Siena, Italy; 10Consiglio Nazionale delle Ricerche Istituto Tecnologie Biomediche (ITB-CNR), Segrate, Milan, Italy; and11Department of Public Health and Cell Biology, Tor Vergata University, Rome, Italy

Adenosine deaminase (ADA) deficiencyis a disorder of the purine metabolismleading to combined immunodeficiencyand systemic alterations, including skel-etal abnormalities. We report that ADAdeficiency in mice causes a specific bonephenotype characterized by alterations ofstructural properties and impaired me-chanical competence. These alterationsare the combined result of an imbalancedreceptor activator of nuclear factor-�B li-gand (RANKL)/osteoprotegerin axis, caus-ing decreased osteoclastogenesis and an

intrinsic defect of osteoblast function withsubsequent low bone formation. In vitro,osteoblasts lacking ADA displayed an al-tered transcriptional profile and growth re-duction. Furthermore, the bone marrow mi-croenvironment of ADA-deficient miceshowed a reduced capacity to support invitro and in vivo hematopoiesis. TreatmentofADA-deficient neonatal mice with enzymereplacement therapy, bone marrow trans-plantation, or gene therapy resulted in fullrecovery of the altered bone parameters.Remarkably, untreated ADA–severe com-

bined immunodeficiency patients showed asimilar imbalance in RANKL/osteoprote-gerin levelsalongsideseveregrowthretarda-tion. Gene therapy with ADA-transducedhematopoietic stem cells increased serumRANKL levels and children’s growth. Ourresults indicate that the ADA metabolismrepresents a crucial modulatory factor ofbone cell activities and remodeling. Thetrials were registered at www.clinicaltrials.gov as #NCT00598481 and #NCT00599781.(Blood. 2009;114:3216-3226)

Introduction

Genetic defects in the adenosine deaminase (ADA) gene are amongthe most common causes for severe combined immunodeficiency(SCID).1 Lack of ADA causes accumulation of purine metabolitesin plasma, lymphoid tissues, and red blood cells. ADA-SCID patientshave lymphopenia, absent cellular and humoral immunity, failure tothrive, and recurrent infections.2 The additional presence of skeletal,hepatic, renal, lung, and neurologic abnormalities underlines that ADAdeficiency is a multiorgan pathology.1,3

Approximately 50% of early-onset ADA-deficient patients exhibitradiologically detectable bone defects.4 Lack of organized cartilagecolumnar formation, large lacuni containing hypertrophied cells, lack oftrabecular formation with uninterrupted areas of calcified cartilage, aswell as few osteoblasts (OBs) and osteoclasts (OCs) with normalmineralizing osteoid have been reported.4-6 Nonetheless, the underlyingcellular and molecular mechanisms have remained unclear because ofthe complexity of the skeletal phenotype and the fact that boneabnormalities are observed also in other immunodeficiencies.7,8 Increas-ing evidence underlines the importance of an intense crosstalk betweenimmune and bone cells regulating not only bone remodeling but alsohematopoiesis.9 Therefore, ADA deficiency represents an importantmodel to study both the impact of altered purine metabolism andimmunodeficiency on bone and bone marrow (BM) stroma.

Bone is a highly complex organ that participates in mineralmetabolism,10,11 provides structural integrity for the body, and supportshematopoiesis. Through a well-organized balance of bone resorptionand formation in a time- and space-dependent manner, bone remodelingenables bone mechanical competence and adaptation to various mechani-cal demands.12 This complex process requires interaction betweendifferent cell types and is regulated by a variety of mechanical andmolecular factors. OCs derived from monocyte/macrophage precursorcells differentiate into multinucleated giant cells specialized in boneresorption.13 Monocyte/macrophage function has been described to bemodulated by adenosine receptor activation.14

OBs are bone-forming cells derived from mesenchymal origin.They secrete an extracellular matrix consisting mainly of type Icollagen, which they later mineralize. Human OBs possess all4 adenosine receptor subtypes and produce extracellular adenosine,which modulates their secretion of interleukin-6 (IL-6) and osteo-protegerin (OPG).15 Adenosine exhibits a potent mitogenic effecton murine calvarial OBs,16 whereas adenosine triphosphate is aknown inhibitor of bone formation.17

Besides their established role in bone remodeling, OBs are acrucial component of the hematopoietic stem cell (HSC) niche.18-20

The interaction of HSCs with OBs is critical for maintaining stem

Submitted March 14, 2009; accepted July 20, 2009. Prepublished online asBlood First Edition paper, July 24, 2009; DOI 10.1182/blood-2009-03-209221.

The online version of this article contains a data supplement.

The publication costs of this article were defrayed in part by page chargepayment. Therefore, and solely to indicate this fact, this article is herebymarked ‘‘advertisement’’ in accordance with 18 USC section 1734.

© 2009 by The American Society of Hematology

3216 BLOOD, 8 OCTOBER 2009 � VOLUME 114, NUMBER 15

For personal use only.on May 5, 2016. by guest www.bloodjournal.orgFrom

cell properties, including self-renewal capacity and the ability todifferentiate into multiple lineages.21

OBs and OCs interact closely to maintain bone homeostasis.Their crosstalk is mediated by RANKL and its decoy receptorOPG. RANKL is produced by OBs but also activated T andB lymphocytes. The source of OPG had historically been attributedto OBs; until recently, BM B cells have been described as a majorsource of BM OPG.22 Because the RANKL-to-OPG ratio regulatesOC activity and formation, it is conceivable that T�B� immunode-ficiencies, such as ADA-SCID, are accompanied by a generalmisbalance of bone homeostasis.

In ADA-deficient patients, as in other forms of SCID, bonemarrow transplantation (BMT) is an effective treatment.23 Enzymereplacement therapy (ERT) with polyethylene glycol-conjugatedbovine ADA (PEG-ADA) provides metabolic detoxification butoften insufficient immune reconstitution.24 Recently, gene therapy(GT) with BM CD34� cells transduced with a �-retroviral vectorhas been shown to correct both the immune and metabolic defectsof ADA-SCID children pretreated with low-intensity condition-ing.25,26 However, little information is available on the correctiveeffects of these treatments on the associated bone defects.7,27,28

Because of the complexity of the interactive scenario describedherein and the limited availability of biologic materials obtainedfrom ADA-deficient patients, we first focused our study on thebone phenotype of the ADA�/� mouse model. ADA�/� mice retainmany features associated with ADA deficiency in humans, includ-ing T- and B-cell lymphopenia and a profound metabolic defect.29

Elevated adenosine levels cause abnormal alveolar development,leading ADA�/� mice to die postnatally within 3 weeks.

We hypothesized that the altered purine metabolism in ADAdeficiency impairs OBs and OCs genesis and activity throughimmunodependent and -independent processes resulting in a spe-cific bone phenotype. We characterized the in vivo bone phenotypeof ADA�/� mice and evaluated possible OB and/or OC defectsin vitro. In parallel, we analyzed the ADA�/� BM stromal cellcompartment and assessed its capacity to support hematopoiesis.We extended our study to 15 ADA-SCID patients, either naive fortreatment, under ERT or after GT to assess bone parameters.Results were discussed in light of potential correction of the ADAbone phenotype by current treatment regimens.

Methods

Mice

ADA-deficient mice have been described previously.29 Breeding pairs forFVB;129-Adatm1MW-TgN(PLADA)4118Rkmb were purchased from TheJackson Laboratory. Rag2�/��c�/� mice30 on BALB/c background wereobtained from the Central Institute for Experimental Animals. Matchedwild-type controls (BALB/c) were purchased from The Jackson Laboratory.All animals were bred and maintained in a specific pathogen-free animalfacility. Procedures were performed according to protocols approved by theCommittee for Animal Care and Use of San Raffaele Scientific Institute(Institutional Animal Care and Use Committee 318).

Patients and clinical trials

Patients were enrolled in 2 subsequent phase 1/2 clinical protocolsapproved by San Raffaele Scientific Institute’s Ethical Committee andItalian National Regulatory Authorities (www.clinicaltrials.gov,#NCT00598481 and #NCT00599781). Orphan Drug Status to ADAvector-transduced CD34� cells was granted by the European MedicinesAgency (EMEA/OD/053/05) to the Italian Telethon Foundation. Pa-tients’ parents signed informed consent to experimental treatment. GT

treatment was performed as described.31 Only patients not displayingany other congenic or endocrine diseases were included. In case ofhealthy donors or patients on ERT not enrolled in either trial, parents oradult subjects signed informed consent in accordance with the Declara-tion of Helsinki for research studies on peripheral blood.

Peripheral quantitative computed tomography

Measurements were performed using a Stratec Research SA� peripheralquantitative computed tomography (pQCT) scanner (Stratec) at voxel sizeof 0.070 mm3 and scan speed of 3 mm/second. All images were obtainedwith 360 projections and section thickness of 500 or 100 �m.

To separate muscle from bone, scans were analyzed twice with pQCTsoftware 6.00B, using contour mode 1 and peel mode 2, threshold of�50 mg/cm3, inner threshold of 40 mg/cm3, filter 2 F03F04 for calculationof total area, and threshold of 280 mg/cm3, inner threshold of 400 mg/cm3

for the calculation of bone area. Bones alone were analyzed using contourmode 2 and peel mode 2 with a threshold of 350 mg/cm3 for the calculationof trabecular and total bone parameters in metaphysis and with a thresholdof 600 mg/cm3 for cortical bone parameters in diaphysis.

Histomorphometry

Femora from 19-day-old mice were fixed, dehydrated, and embedded inmethyl methacrylate without decalcification. Histomorphometric analysiswas performed on toluidine blue– and Alizarin Red S–stained sectionsusing a light microscope (Nikon Axiophot) equipped with an image analysissystem (Nikon DS-5Mc Videocamera; NIS Elements AR 2.20 NikonSoftware).

Calvarial OB cultures and lentiviral transduction

Calvaria from newborn mice were explanted, digested with collagenase,and grown until confluence. In the absence of stimuli, ADA�/� OBs weretransduced with pCCLsin.cPPT.hPGK.hADA.Wpre (PGK-ADA) at a mul-tiplicity of infection of 100 as described previously.32 Median ADA activitywas 13 460.5 plus or minus 1904 nmol/h per milligram of protein. At days3, 5, 8, 10, 12, and 15, replated (15 000/well) cells were harvested andcounted using an automated cell counter (Coulter Counter ZM, Electronics).

Human OB-like cultures

Human bone cell cultures were established using trabecular bone samplesobtained from waste materials during orthopedic surgery.33 Cells weretested for alkaline phosphatase and osteocalcin production after1,25(OH)2D3 10�8 M to ensure that they were endowed with OBcharacteristics.

In vitro osteoclastogenesis

Flushed total BM was plated in presence of 100 ng/mL macrophage colony-stimulating factor (M-CSF; PeproTech) for 3 days. Cells were replated at5000/well in 96-well-plates and cultured in presence of 25 ng/mL M-CSF and100 ng/mL RANKL (PeproTech). Cultures were fixed, thrombin receptoractivating peptide (TRAP) staining was performed according to the manufactur-er’s instructions (Sigma-Aldrich), and multinucleated (3 or more nuclei) TRAP�

cells were scored.

Alamar Blue viability assay

OBs, M-CSF–dependent BM macrophages, and stromal layers forlong-term culture-initiating cell (LTC-IC) assays were isolated asdescribed in “Calvarial OB cultures,” “In vitro osteoclastogenesis,” and“Stromal cultures (CFU-F, LTC-IC)” and plated in 96-well-plates;10 �L Alamar Blue (Biosource) reagent was added per well. After3 hours, viability was measured (excitation 530 nm, emission 590 nm)using a Victor3 Microplate Reader (PerkinElmer Life and AnalyticalSciences).

ADA-SCID BONE PHENOTYPE 3217BLOOD, 8 OCTOBER 2009 � VOLUME 114, NUMBER 15

For personal use only.on May 5, 2016. by guest www.bloodjournal.orgFrom

ADA enzymatic activity

Intracellular ADA enzyme activity was analyzed by adenosine to inosineconversion followed by high-performance capillary electrophoresis.34,35

Red blood cell lysis was performed on BM and spleen samples.

Gene expression analyses

RNA was extracted using EUROzol (Euroclone) and transcribed into cDNAusing a High-Capacity cDNA Archive Kit (Applied Biosystems). Real-timepolymerase chain reactions were carried out using Assay-on-Demand geneexpression arrays (Applied Biosystems). The relative expression of each genewas normalized to hypoxanthine phosphoribosyl transferase (HPRT) as endoge-nous control. mRNA levels were quantified using the comparative threshold-cycle method. Arrays used included the following: Mm01187117_m1,Mm00801666_g1, Mm01337566_m1, Mm00485009_m1, Mm00435452_m1,Mm00441908_m1, Mm03003491_m1.

FACS analyses

Staining for annexin V and 7-amino-actinomycin D was performed oncultured OBs according to manufacturer’s instructions (both BD Bio-sciences PharMingen). Stainings for OC precursors from flushed total BMwere performed as described previously.32 Rat anti–mouse monoclonalantibodies used included the following: fluorescein isothiocyanate–conjugated anti-CD117 (2B8; BD Biosciences PharMingen), phycoerythrin-conjugated anti-F4/80 (Serotec), allophycocyanin-conjugated anti-CD11b(M1/70; BD Biosciences PharMingen), and Pacific Blue–conjugated anti-CD48 (HM48-1; BioLegend). Samples were analyzed using a BD fluores-cence-activated cell sorter (FACS) Canto and DiVa software (BD Bio-sciences PharMingen).

ELISA

Murine RANKL, OPG (R&D Systems), and murine N-terminal propeptideof type I procollagen (PINP; iDS) were assayed on sera from 19-day-oldmale ADA�/� and ADA�/�, Rag2�c�/� and Rag2�c�/� or 12-week-oldADA�/� and rescued ADA�/� mice. Enzyme-linked immunosorbent assay(ELISA) for fragments of the type I collagen (CTX; RatLaps ELISA;Nordic Biosciences Diagnostics) was performed according to the manufac-turer’s instructions on serum samples from 12-week-old ADA�/� orrescued ADA�/� mice starved for 6 hours. Assessment of CTX in untreatedADA�/� mice of 19 days of age was not feasible because of the starvationprotocol. Increased sensitivity sRANKL and human OPG ELISA (Bio-medica) were performed on plasma from patients and pediatric normaldonors according to the manufacturer’s instructions.

Stromal cultures (CFU-F, LTC-IC)

Fibroblast colony-forming unit (CFU-F) assays were performed usingMesenCult medium (StemCell Technologies) according to the manufactur-er’s protocol. Stromal feeder layers for LTC-IC assays were establishedfrom flushed BM using MyeloCult medium (StemCell Technologies)according to the manufacturer’s instructions. Within 1 week after irradia-tion (15 Gy), lineage-negative cells isolated using StemSep Murine Progeni-tor Enrichment separation (StemCell Technologies) were added. ADA�/�

cells showed comparable numbers and viability after irradiation at 15 Gyand 30 Gy, excluding a potential bias because of differential irradiationsensitivity36 (data not shown). After 4 weeks of coculture, adherent andnonadherent cells from each well were transferred into semisolid MethoCult(StemCell Technologies). CFU were scored after 10 days.

Proteome profile

Supernatants from LTC-IC cultures established as described in “Stromalcultures (CFU-F, LTC-IC)” were analyzed using Proteome Profiler Arrays(R&D Systems) according to the manufacturer’s instructions. Densitomet-ric analysis was carried out using ImageQuant software (GE Healthcare).

5-Fluorouracil treatment

Neonatal ADA�/� and ADA�/� mice were transplanted into the temporalvein with 3000 ADA�/� lineage-negative cells, as described.32 Transplantedcells were purified by StemSep Murine Progenitor Enrichment separation(StemCell Technologies). Starting from 6 weeks of age, surviving micereceived weekly intraperitoneal injections with 150 mg/kg 5-fluorouracil(Sigma-Aldrich).

ERT, BMT, and GT treatment

ADA�/� were rescued by weekly intraperitoneal injections (1000 U/kg)with PEG-ADA (Adagen, Enzon), by BMT or GT as described previously.32

A total of 5 � 106 ADA�/� or transduced ADA�/� BM cells were infusedby injection through the temporal vein of irradiated neonatal ADA�/� mice.

Statistical analyses

For comparisons between groups, an unpaired 2-tailed Student t test wasused. A P value less than .05 was considered significant. For analyses ofLTC-IC assays, the difference in scored colonies was evaluated fitting ageneralized linear model based on Poisson distribution, taking into accountpossible differences between experiments and first-order interactions. ForProteome Profile analyses, a linear model was used to assess differences inprotein expression, taking into account possible differences betweenexperiments. P values were adjusted for multiplicity by the Benjamini-Hochberg false discovery rate method. All analyses were performed with Rsoftware (Version 2.8.0, Fedora Project).

Results

ADA-deficient mice display a specific bone phenotype

Because bone defects resulting from alterations of crosstalk betweenhematopoietic and bone cells have been described in other immunodefi-cient mice,37 we studied the ADA�/� bone phenotype compared withdouble-mutant Rag2�c�/� mice.30 The latter lack T, B, and NK cells butdo not have the profound metabolic defect typical for ADA�/� mice.Comparing both immunodeficient models allowed us to distinguish thecontribution of immunodeficiency from that of the metabolic disease tothe ADA bone phenotype. Because of strain differences, ADA�/� andRag2�c�/� mice were compared with their respective sex- and age-matched wild-type controls.

ADA�/� mice were analyzed at 19 days of age, when they aregenerally smaller, whereas Rag2�c�/� mice (x-ray not shown) arecomparable with wild-type controls (supplemental Figure 1A, availableon the Blood website; see the Supplemental Materials link at the top ofthe online article). To characterize bone size and structural parameterspQCT was performed on tibiae and femora retrieved from ADA�/�,ADA�/�, Rag2�c�/�, and Rag2�c�/� mice. As shown in Figure 1A-B,a significantly lower total area and medullary canal area was found inboth immunodeficient mouse models. This difference was more pro-nounced in the ADA�/� compared with Rag2�c�/� mice, resulting insignificant reduction in the canal area to total area ratio (Figure 1C).Consequently, cortical thickness (Figure 1D) was found to be signifi-cantly larger in Rag2�c�/� mice, whereas no difference was measuredin ADA�/� mice. The strength-strain index, direct estimate of bonestrength, was significantly reduced only in ADA�/� mice (Figure 1E).The observed structural differences were not accompanied by changesin cortical bone density (Figure 1G). Interestingly, the trabecular densitywas significantly reduced inADA�/� mice but significantly increased inRag2�c�/� mice (Figure 1H).

ADA�/� mice have significant weight loss during the last daysof life (supplemental Figure 1B). Because developmental changesin bone strength are secondary to increasing loads imposed by

3218 SAUER et al BLOOD, 8 OCTOBER 2009 � VOLUME 114, NUMBER 15

For personal use only.on May 5, 2016. by guest www.bloodjournal.orgFrom

larger muscle forces,38 the functional muscle-bone unit was evalu-ated to distinguish whether any observed bone defect is specificallyrelated to insufficient muscle mass. We found a large reduction inthe muscle cross-sectional area, whereas bone mineral contentremained unchanged, resulting in a significantly higher proportionbetween bone mineral content and muscle cross-sectional area inADA�/� mice (supplemental Figure 1C).

ADA�/� mice display low trabecular bone volume

The low trabecular density, detected by pQCT, was confirmed atstatic histomorphometric evaluation. As shown in Figure 2A,longitudinal growth of ADA�/� femora is impaired compared withcontrols. To assess defects in the trabecular bone network, metaphy-seal sections were analyzed by image analysis software (Figure2B-C). As summarized in supplemental Table 1, the trabecularbone volume and trabecular number in ADA�/� mice weresignificantly lower compared with wild-type. Consequently, trabec-ular separation was significantly increased, whereas trabecularthickness was not different between both groups. Despite early

reports on a selective toxicity of ADA substrates for chondrocytes,6

we did not detect any gross differences between the growth platesof ADA�/� and ADA�/� mice (Figure 2D-E).

Reduced RANKL/OPG ratio and reduced bone formation inADA�/� mice

RANKL and OPG, well-characterized key players of crosstalkbetween OBs or stromal cells with hematopoietic OC precur-sors, synchronize osteoclastogenesis, and bone resorption. Asshown in Figure 3A, both immunodeficient models showedsignificantly reduced RANKL levels, whereas serum OPG levelswere in the normal range for both groups (Figure 3B). Conse-quently, the RANKL/OPG ratio, indicator of in vivo boneturnover, was reduced in ADA�/� and Rag2�c�/� mice (Figure3C). To assess new bone formation, we measured the levels ofthe PINP. As shown in Figure 3D, PINP levels were significantlyreduced in sera of ADA�/� mice, although comparable inRag2�c�/� mice and controls.

Figure 1. Ex vivo femural pQCT analyses of ADA�/� andADA�/� compared with Rag2�c�/� and Rag2�c�/� mice.(A) Total area (mg/cm3). (B) Medullary canal area (mm2).(C) Cortical thickness (mm). (D) Canal area/total area ratio(AU). (E) Strength-strain index (mm3). (F) Schematic viewrepresenting cross-sectional bone sections in diaphysis of all4 groups of mice analyzed. (G) Cortical density (mm/cm3).(H) Trabecular density (mm/cm3). Box and whiskers graphs.***P � .001. **P � .001-.005. *P � .005-.05.ADA�/� (n � 22),ADA�/� (n � 18), Rag2�c�/� (n � 10), and Rag2�c�/�

(n � 14).

ADA-SCID BONE PHENOTYPE 3219BLOOD, 8 OCTOBER 2009 � VOLUME 114, NUMBER 15

For personal use only.on May 5, 2016. by guest www.bloodjournal.orgFrom

ADA activity in bone cells

ADA is ubiquitously expressed, but little information is availableon the specific levels of enzymatic activity in bone cells. Remark-ably, ADA activity in wild-type OBs was found 3-fold highercompared with mesenchymal progenitor cells, 2-fold higher com-pared with OC precursors and therefore lies in the range of otherlymphoid organs, such as the spleen or BM (supplemental Table 2).

ADA�/� OB activity is impaired

The reduced periosteal bone apposition in vivo and the high ADAactivity detected in wild-type OBs in vitro suggested defects of theosteoblastic compartment in ADA�/� mice. ADA�/� OBs showed asignificantly lower proliferation rate compared with wild-type cellsin vitro (Figure 4A). This was associated with a significant decreasein viability and a significant increase in the percentage of apoptoticcells (Figure 4B-C). Importantly, transduction with a lentiviralvector encoding for ADA rescued both the growth defects andincreased sensitivity to apoptosis observed in ADA�/� cells (Figure4C-E). We performed quantitative gene expression arrays designedto assess whether OB differentiation is altered in ADA�/� mice(Figure 4D). Runx2, known to contribute to early osteogenic

differentiation, was expressed at comparable levels.39 Althoughexpression levels of collagen type 1- and osteocalcin-recognizedmarkers of OB activity were significantly decreased in ADA�/�

OBs,39 alkaline phosphatase, essential for matrix mineralization,revealed 2-fold increased expression levels. Interestingly, RANKLexpression levels were significantly reduced, whereas OPG levelswere 3-fold increased. ADA inhibition with erythro-9-(2-hydroxy-3-nonyl)-adenine (EHNA; 100 mM) is a commonly used method tomimic ADA deficiency in vitro. For unavailability of sufficientpatient material, we used EHNA to block ADA activity in primaryhuman OBs. As demonstrated in murine OBs (Figure 4E), EHNAefficiently blocked ADA activity, reducing OBs proliferation tolevels comparable with ADA�/� cells. Proliferation of human OBswas significantly inhibited when cultured with EHNA (Figure 4F),culturing with EHNA and adenosine completely ablatedproliferation.

No intrinsic defect of ADA�/� OCs

Consistent with the small medullary canal area observed inADA�/� mice, serum RANKL levels were found to be significantlydecreased, suggesting a block in osteoclastogenesis. To exclude a

Figure 2. Reduced longitudinal growth and trabecular bone in ADA�/�

mice. (A) ADA�/� femora and tibiae (not shown) are significantly shortercompared with wild-type (cm). ***P � .001. Metaphyseal bone sections ofADA�/� (B) and ADA�/� (C) femora (original magnification �2.5). Scalebar represents 600 �m and original magnification �20 of ADA�/� (D) andADA�/� (E). Scale bar represents 50 �m; indicates defects in ossificationbut not in cartilage formation. Histomorphologic analyses were performedon ADA�/� (n � 5) and ADA�/� (n � 5) femora. BV TV indicates trabecularbone volume (%); Tb Th, trabecular thickness (�m); Tb N, trabecularnumber (1/mm); Tb S, trabecular separation (mm).

Figure 3. Reduced RANKL/OPG ratio and PINP levels in ADA-deficient mice. (A) Murine serum RANKL (ng/mL) in ADA�/� (n � 22) andADA�/� (n � 18), Rag2�c�/� (n � 10), and Rag2�c�/� (n � 14) mice.(B) Murine OPG (ng/mL). (C) Murine RANKL/OPG ratio (AU). ***P � .001.**P � .001-.005. *P � .005-.05. N.S. indicates not significant. (D) In vivobone formation rate as assessed by PINP levels are significantly lower inADA�/� mice compared with wild-type controls. Serum PINP levels(ng/mL), ADA�/� (n � 9), ADA�/� (n � 9), Rag2�c�/� (n � 15), andRag2�c�/� (n � 13).

3220 SAUER et al BLOOD, 8 OCTOBER 2009 � VOLUME 114, NUMBER 15

For personal use only.on May 5, 2016. by guest www.bloodjournal.orgFrom

lack of OC precursors, we assessed their percentage in the BM byFACS staining for CD11b, cKit, CD48, and F4/80. The CD11b�/cKit� population was selected as myeloid precursors; within thissubpopulation, we distinguished committed myeloid precursorsfrom macrophages by CD48 and F4/80 staining. The relativepercentages of OC precursors of both the monocyte and macro-phage lineages were increased in the BM of ADA�/� mice(supplemental Figure 2A). This significant difference was main-tained when the percentage of OC precursors was normalized forthe absolute number of total BM cells and the median size of theBM cavity (supplemental Figure 2B). No differences in viability ofOC precursors and in vitro osteoclastogenesis from M-CSF andRANKL-stimulated BM were detected (supplemental Figure 2C-D).

Reduced hematopoietic support of ADA�/� stroma

Given the importance of OB for the maintenance of the HSCniche, profound defects of the osteoblastic compartment inADA�/� mice described herein imply a potential impact on theHSC niche and/or hematopoiesis itself. Using CFU-F assays, weshowed that ADA�/� and ADA�/� mesenchymal progenitorcells grow equally in vitro (Figure 5A). Nonetheless, whencocultured with ADA�/� or ADA�/� lineage-negative cells,ADA�/� stromal cells supported colony formation less effi-ciently than wild-type cells (Figure 5B). This difference wassignificantly different considering wild-type versus ADA�/�

feeder layers (P � .012), but independent from the coculturedlineage-negative cells. Proteome profiling of total cytokine andchemokine production from the supernatant of LTC-IC culturesrevealed reduced levels of M-CSF, IL-6, CXCL1/10, andsICAM-1 (Figure 5C). Interestingly, only the IL-1 receptor

antagonist (IL-1ra) was expressed at higher levels comparedwith wild-type. The secreted form of the IL-1ra is an acute-phase protein intervening in the counterregulation of inflamma-tory processes and is highly expressed in liver and spleen.40

Gene expression arrays for IL1-ra revealed 5.6- to 6.2-foldincreased expression levels in spleen, liver, and BM fromADA�/� mice. Normal IL-1ra expression was detected incultured OBs, stromal, and mesenchymal cells (Figure 5D),indicating that hematopoietic cells are probably the source ofincreased IL-1ra levels detected in proteome profiles fromADA�/� LTC-IC assays. To assess potential defects of the HSCniche in situ, we assessed the percentage of CD150�CD48�

within the lineage-negative, Sca1� and cKit� compartment,which are generally thought to be quiescent HSC with afrequency of 50% repopulating capacity.41 As shown in Figure5E, the percentage of LSK CD150�CD48� HSC is significantlylower in ADA�/� mice compared with wild-type. In addition, wetransplanted neonatal ADA�/� and ADA�/� mice with wild-typelineage-negative cells. Hematopoietic reconstitution was as-sessed by monitoring survival on weekly injections of thecell-cycle-dependent myelotoxic agent 5-fluorouracil.42 As shownin Figure 5F, because of hematopoietic failure, ADA�/� micetransplanted with wild-type cells died earlier compared withADA�/� mice (P � .011).

Recovery of altered bone parameters in rescued ADA�/� mice

To assess the effect of different treatments on the ADA bonephenotype, we rescued newborn ADA�/� mice with ERT, BMT, orex vivo lentiviral-mediated GT. Serum concentrations of RANKL,OPG, PINP, and CTX were assessed in surviving mice at 12 weeks

Figure 4. Defects in ADA�/� OB activity. (A) In vitro ADA�/� OBs growsignificantly less than wild-type cells. *P � .005-.05. (B) Viability as assessedby the Alamar Blue staining is significantly lower in cultured ADA�/� OBs.(C) The percentage of apoptotic cells as assessed by annexin V�/7-amino-actinomycin D� staining is significantly higher in ADA�/� OBs but normalizes incells transduced with a lentiviral vector encoding for ADA (ADA�/� TR).**P � .001-.005. (D) TaqMan gene expression analyses for osteoblasticdifferentiation markers Runx2, type 1 collagen, alkaline phosphatase, andosteocalcin as well as RANKL and OPG; fold expression versus wild-type(represented by dashed line), normalized for HPRT endogenous control;average of 3 independent experiments � SD. (E) Growth of murine ADA�/�

OB after 12 days in vitro is reduced when cultured with EHNA (100 �M) andabolished when cultured with EHNA and adenosine (1 mM). ADA�/� OBsregain growth after lentiviral vector transduction. Growth as percentage ofinitially plated OBs � SD. (F) Growth of human OB after 13 days of in vitro isreduced when cultured with EHNA (100 �M). **P � .01.

ADA-SCID BONE PHENOTYPE 3221BLOOD, 8 OCTOBER 2009 � VOLUME 114, NUMBER 15

For personal use only.on May 5, 2016. by guest www.bloodjournal.orgFrom

Figure 5. Stromal insufficiency to support hematopoiesis. (A) CFU-Fassays show comparable frequency of mesenchymal progenitor cells inADA�/� or ADA�/� total BM. (B) LTC-IC assays of ADA�/� or ADA�/� feederlayers cocultured with ADA�/� or ADA�/� lineage� cells. ADA�/� or ADA�/�

LTC-IC grow significantly less when cocultured with ADA�/� feeder layers(P � .012). (C) Proteome profile from supernatants of ADA�/� and ADA�/�

cocultures, performed in triplicate; average � SD. **P � .001-.005.*P � .005-.05. (D) TaqMan gene expression analyses for IL-1ra. Spleen,liver, BM, and bone in vivo. Stromal cells as used for LTC-IC, mesenchymalprogenitor cells as cultured for CFU-F assays, OBs in vitro; fold expressionversus wild-type (represented by dashed line), normalized for HPRT endog-enous control (average of 3 experiments � SD. (E) The percentage ofquiescent HSCs, as assessed by FACS staining for lineage�, Sca1�, cKit�,CD150�, CD48� cells is significantly lower in ADA�/� (n � 8) comparedwith ADA�/� mice (n � 8). ***P � .001. (F) Survival curves of ADA�/� andADA�/� mice transplanted with ADA�/� lineage� cells after multiple5-fluorouracil injections at days 7, 14, 21, and 28. P � .011.

Figure 6. Rescue of the ADA bone phenotype by ERT, GT, or BMT.Comparable serum levels of (A) RANKL (ng/mL), (B) OPG (ng/mL),*P � .005-.05; **P � .001-.005. (C) RANKL/OPG ratio (AU), (D) PINP(ng/mL), and (E) CTX (ng/mL) at 12 weeks of age. Scatter plots plusaverage. PEG-ADA (n � 7), GT (n 4), BMT (n � 10), and wild-typecontrols (n � 7). (F) Strength-strain index (mm3); average � SD; PEG-ADA (n � 4), GT (n � 5), BMT (n � 5), and wild-type controls (n � 4).

3222 SAUER et al BLOOD, 8 OCTOBER 2009 � VOLUME 114, NUMBER 15

For personal use only.on May 5, 2016. by guest www.bloodjournal.orgFrom

of age. As shown in Figure 6, mice rescued by all 3 differenttreatments displayed full correction of serum RANKL and, conse-quently, RANKL/OPG levels (Figure 6A-C). Serum PINP levelswere completely rescued, whereas CTX levels were comparable withcontrols (Figure 6D-E). pQCT analyses confirmed that all bone parame-ters previously observed to be altered in naive ADA�/� mice recoveredto normal (Figure 6F, supplemental Figure 3).

ADA-SCID patients show delayed growth and altered boneremodeling

From the few case reports in ADA-SCID patients, it has remainedunclear whether their severe growth delay is associated withspecific defects in bone remodeling. To assess whether boneturnover is impaired, we measured RANKL and OPG ratio inplasma of 5 ADA-deficient patients naive for treatment anddisplaying severe growth retardation (Table 1). In all patients,RANKL levels were severely reduced or undetectable (P � .005).Interestingly, serum OPG levels were significantly increased(P � .003), resulting in RANKL versus OPG ratio significantlyreduced compared with age-matched normal donors (P � .011).

GT but not enzyme replacement normalizes RANKL inADA-SCID patients

As reported in Table 2, patients receiving ERT long-term(1-21 years), displayed low or undectable serum RANKL levels(P � .004), whereas OPG was in the normal range (P � .504).This resulted in a significantly reduced RANKL/OPG ratio(P � .033). All patients displayed significantly reduced heightcompared with age-matched standards. We next analyzed RANKL andOPG in patients treated with HSC-GT combined to reduced intensityconditioning.31 As reported in Table 3, in patients treated with GT,RANKL, and consequently the RANKL/OPG ratio significantly in-creased one to 2 years after treatment (P � .005; P � .031). Impor-tantly, both serum RANKL (P � .979) and the RANKL/OPG ratio(P � .515) were not significantly different from the range observed in

pediatric normal donors 1 to 5 years of age (Table 1). Consistently, 8 of9 patients displayed an increase in percentile of height at the lastfollow-up after treatment. However, the percentile of height andbone age (Greulich/Pyle) was not fully normalized in all patients.

Discussion

Skeletal defects have been reported to be a common feature ofADA deficiency in humans3-5,8,27,28; nonetheless, it has remainedunclear whether they are a consequence of SCID or caused byalterations in purine metabolism. The present study has filled thisgap by providing evidence that ADA deficiency in mice isassociated with a specific bone phenotype, characterized byalterations in structural properties and mechanical competence.However, these structural alterations being only partially superim-posable to those observed in T�B� Rag2�c�/� mice, could not beascribed to immunodeficiency per se. While the small medullarycanal area appeared a joint feature in both immunodeficientmodels, low bone mass and strength-strain index were observedonly in ADA�/� mice. The reduced trabecular density, specificallyaffected by ADA deficiency, was related to a low trabecularnumber, rather than reduced trabecular thickness. Because fetalbone development predicts bone mass accrual as a consequence ofincreasing trabecular bone volume in function of growth,43 thedecreased trabecular number might be the expression of impairedossification of the cartilaginous template with subsequent deteriora-tion of bone structural properties. However, no mineralizationdefects were measured in the cortical bone of ADA�/� mice, and noosteoid seams were observed in static histomorphometric evalua-tion, indicating an appropriate acquisition of tissue materialproperties. Although no gross alteration of the growth platemorphology could be observed, the contribution of reduced growthplate activity for the determination of the bone phenotype isplausible because both ADA-deficient mice and patients display

Table 1. ADA-SCID patients naive for therapy display significantly reduced growth

Age, y Percentile height, Z-score RANKL, pg/mL OPG, pg/mL RANKL/OPG, � 102

Patient no. 1 1 �1.66 1.42 181.9 0.78

Patient no. 2 1.9 �3.91 1.03 129.6 0.79

Patient no. 3 1.8 �1.83 0 181.9 0

Patient no. 4 1.6 �1.90 0 145.2 0

Patient no. 5 0.5 �2.21 3.08 257.3 1.19

ND 1-5 years of age (n � 10), mean (SEM) 11.49 (3) 104.1 (7) 13.4 (5)

P .005 .003 .011

ND indicates normal donor.

Table 2. ADA-SCID patients on ERT with PEG-ADA display reduced growth

Age at last follow-up, y PEG-ADA, yPercentile height,

Z-score RANKL, pg/mL OPG, pg/mL RANKL/OPG, � 102

Patient no. 6 5.6 5 �1.69 0 78.1 0

Patient no. 7 1.5 1 �2.06 0 120.6 0

Patient no. 8 1.4 1 �0.29 1.61 153.2 1.05

Patient no. 9 1.3 1 1.29 0 145.2 0

Patient no. 10 6.2 6 �0.38 2.85 18.4 15.48

Patient no. 11 2.5 1 �2.14 2.57 173.8 1.48

Patient no. 12 12.5 10 �3.61 0 93.3 0

Patient no. 13 22.5 21 �0.93 0 167.5 0

Patient no. 14 19 17 �1.98 0 135.3 0

ND 1-15 years of age (n � 14), mean (SEM) 8.06 (1.6) 106.9 (5.6) 8.4 (1.7)

P .004 .504 .033

ADA-SCID BONE PHENOTYPE 3223BLOOD, 8 OCTOBER 2009 � VOLUME 114, NUMBER 15

For personal use only.on May 5, 2016. by guest www.bloodjournal.orgFrom

significantly reduced longitudinal growth. Contrarily, the contribu-tion of muscle dysfunction can be considered minimal because thehigher muscle-bone-unit here observed was related to the fasterloss of muscle than bone mass because of metabolic toxification ata later stage of survival.

The described ADA bone phenotype is consistent with thehypothesis that the altered purine metabolism impairs OC andOB genesis, proliferation and activity through immuno-dependent and -independent processes.14-17 Because of reducedRANKL levels, the serum RANKL/OPG ratio was significantlyreduced in both ADA�/� and Rag2�c�/� mice. The small medul-lary canal area fits this observation, suggesting a shift of the boneremodeling sequence toward reduced osteoclastogenesis. Interest-ingly, proteome profiling of cytokine and chemokine productionfrom LTC-IC cultures revealed reduced levels of M-CSF and IL-6,factors involved in OC activity and generation.9 It is doubtful thatthis shift is the result of an intrinsic defect of OC precursorsbecause in the presence of M-CSF and RANKL they equally formTRAP� cells in vitro.

The high ADA activity measured in wild-type OBs indicates astrong dependency of this cell type on the ADA metabolic pathway.Interestingly, enzymatic activity measured in mesenchymal progeni-tor cells was 3-fold lower, indicating that differentiating OBs mustup-regulate ADA expression considerably. It is therefore conceiv-able that, with increasing dependency on the ADA enzyme duringproliferation or differentiation, OB function and viability becomeseverely affected.

The reduced outward displacement of the thin cortex as well asthe lower serum PINP levels in ADA�/� mice are in agreement withthe hypothesis that OBs are the major casualty of ADA substrates’toxicity. The lower proliferation rate of ADA�/� OBs observed invitro, associated with decreased viability and increased apoptosis,confirms this hypothesis and underlines the importance of purinemetabolism in OB function. Importantly, correction of ADA expressionby lentiviral vector transduction leads to full correction of ADA�/� OBgrowth defects and apoptosis, providing direct evidence that the OBsinsufficiency state is cell intrinsic. Proliferation of primary murine andhuman OBs was similarly reduced, when mimicking ADA deficiencyusing theADAinhibitor EHNA. Thus, low OB numbers may contributeto the bone phenotype observed in both ADA-deficient mice andpatients. In accordance with impaired bone formation, the expressionprofile of murine ADA�/� OBs is characterized by low collagen type 1and osteocalcin levels. However, it is likely that OBs compensate fortheir low numbers expressing elevated levels of alkaline phosphatase, sothat no differences of bone material properties were detected. Similar to

the low RANKL/OPG ratio observed in ADA�/� sera, their respectiveRNA expression levels were altered also in ADA�/� OBs. Being thedominant mediators of osteoclastogenesis, their misbalance confirmsthe central role of OBs determining the ADA bone phenotype andsuggests the activation of a compensatory mechanism for bone masspreservation.9

Because RANKL ligand is also produced by B and T cells, itsreduced serum levels can be partially attributed to the lymphopeniain both ADA�/� and Rag2�c�/� mice. Nonetheless, lymphocytesare severely reduced but not completely absent in BM and thymusof both immunodeficient models, which probably accounts for theresidual serum RANKL levels. In addition, RANKL expression, incontrast to OPG, is down-regulated in ADA�/� OBs. Its reducedexpression might be induced by the increased levels of IL-1ra,detected in ADA�/� mice, which may act by preventing IL-1signaling, a potent stimulator of bone resorption. IL-1 modulatesOC activity directly44 or indirectly through its ability to stimulateRANKL production by OBs45; therefore, IL-1ra secreted at el-evated levels in ADA�/� mice probably restricts IL-1–inducedbone resorption and RANKL expression. It can be speculated thatthe reduced endosteal resorption might counterbalance the lowperiosteal bone apposition to fulfill the physical requisite of bonestrength for loading and lightness for mobility.46 Considering thatBM B cells are a major producer of OPG,22 it was unexpected toobserve normal serum OPG levels in ADA�/� mice. Nonetheless,OPG may be produced by other cell sources; and indeed, we founda 3-fold up-regulation of OPG mRNA levels in ADA�/� OBs.

These data suggest an OB insufficiency state as consequence ofADA deficiency that, given the interplay between osteogenesis andhematopoiesis, implies an impact on the BM microenvironmentand the HSC niche.18-20 In vivo depletion of OBs caused loss ofHSCs, followed by a marked decrease in BM cellularity.20 Consis-tently, the OB insufficiency in ADA�/� mice is associated with areduced BM cellularity (not shown). The capacity of ADA�/�

stromal cells to support colony formation from lineage-negativecells was significantly lower compared with wild-type. Thisreduced capacity might be associated with the low production ofIL-6, crucial survival factor for HSCs and is in agreement with thedefective hematopoietic support reported in IL-6–deficient mice.47

The hypothesis that the HSC niche in ADA deficiency is specifi-cally affected was further supported by the reduced percentage ofLSK CD150�CD48� HSCs in ADA�/� BM and by the hematopoi-etic failure and premature death of ADA�/� mice transplanted withwild-type HSC when challenged with 5-fluorouracil. Our data may

Table 3. ADA-SCID patients treated with hematopoietic stem cell gene therapy display a significant increase in RANKL

Patientno.

Age at therapy,y

Age at lastfollow-up, y

Percentile height, Z-score RANKL, pg/mL OPG, pg/mLRANKL/OPG,

� 102 Percentageof bone age/actual age, y

(post GT)PrePost (at lastfollow-up) Pre Post Pre Post Pre Post

1 1 7.6 �1.66 �1.12 1.4 3.65 181.9 113.1 0.78 3.23 (2) 96% (6)

2 1.9 8 �3.91 �1.48 1.0 16.1 129.6 62.3 0.79 25.95 109% (6)

4 1.6 3.3 �1.9 �2.11 0 0.82 195.7 172.8 0 0.47 (1) 96% (1)

5 0.5 1.5 �2.21 �2.07 3.0 18.3 257.3 191.5 1.19 9.59 (1) ND

6 5.6 10.7 �1.69 �1.22 0 2.3 78.1 157.5 0 1.46 (1) 81% (4)

7 1.5 4.6 �2.06 �0.69 0 3.54 120.6 412.9 0 0.85 (2) 65% (3)

8 1.4 3.6 �0.29 1.46 1.6 9.1 153.2 198.6 1.05 4.58 (1) 74% (2)

9 1.3 2.9 1.29 0.20 0 6.84 145.2 148.1 0 4.61 (1) 64% (1)

15 1.6 6.3 �3.11 �0.99 2.2 8.94 121.2 64.2 1.87 13.93 ND

P .005 .698, NS .031

NS indicates not significant; and ND, not done.Numbers in parentheses indicate year after GT.

3224 SAUER et al BLOOD, 8 OCTOBER 2009 � VOLUME 114, NUMBER 15

For personal use only.on May 5, 2016. by guest www.bloodjournal.orgFrom

provide a possible explanation for the higher toxicity of pretrans-plantation conditioning and transplantation failures observed inADA-SCID patients treated with BMT24,48 as well as for frequentfindings of hematopoietic abnormalities1 (and A. Aiuti, unpub-lished results, January 2009). Future studies will be needed toassess whether current therapeutic approaches may be improved bysimultaneous targeting of the niche. In this regard, transplantationof normal or ADA engineered mesenchymal stem cells mayfacilitate HSC engraftment or contribute to the reconstitution of thebone cell compartment.

Comparative analyses of adult ADA�/� mice treated withERT, GT, or BMT showed full correction of RANKL, RANKL/OPG, and PINP as well as recovery of bone growth. Weobserved a significant increase of serum OPG levels in micetreated with GT or BMT, a phenomenon that has also beenreported in allotransplanted patients.49,50 This is probably re-lated to pretransplantation conditioning rather than the transplan-tation procedure itself. Consistently nonirradiated PEG-ADA–treated mice displayed normal OPG levels.

The results obtained in 15 ADA-SCID patients, either naive orunder different treatments, have confirmed that bone defects are acommon feature associated with ADA deficiency in humans andextended the relevance of our findings to the human diseasephenotype. RANKL and OPG levels in naive ADA-deficientpatients, displaying serious growth retardation, were significantlylower compared with healthy controls. GT with autologous gene-corrected stem cells, but not ERT, resulted in a significant increasein serum RANKL and the RANKL/OPG ratio. The less efficientcorrection of bone parameters in ADA-SCID patients comparedwith mice might be because all mice received treatment within thefirst 5 days of life, whereas patients were treated at variable ages.Moreover, ERT could be more efficient to correct the murine ADAbone phenotype because to rescue ADA�/� mice up to 50 timeshigher doses of PEG-ADA are required compared with standarddoses in humans. Because OBs are not derived from HSCs, they arenot expected to be corrected by either treatment. It is possible thatendogenous ADA production by resident hematopoietic cells ismore efficient to cross-correct OB function compared with circulat-ing PEG-ADA. Moreover, patients treated with GT show superiorimmune reconstitution and lymphocyte counts with respect topatients treated with ERT,31 representing an important source ofRANKL.

In conclusion, we report an ADA-specific bone phenotype,characterized by low bone mass accrual, size acquisition, and

impaired mechanical competence, which is the result of an OBinsufficiency with subsequent impact on the HSC niche. Our resultsemphasize the role of the ADA metabolism in modulating bone cellactivities and add a stromal component to the series of immuno-logic defects described in ADA deficiency. The present data showthat correction of the ADA bone phenotype is feasible with currenttreatment options, but longer follow-up in these patients will beneeded to assess whether bone defects are resolvable over a longerperiod of time.

Acknowledgments

The authors thank Immacolata Brigida, Nicola Carriglio, AntonellaTabucchi, and Bernhard Gentner for technical assistance andhelpful suggestions, Luciano Callegaro for patient data manage-ment, and the physicians and nurses of the Pediatric ClinicalResearch Unit, Italian Telethon Foundation.

This work was supported by the Italian Telethon Foundation(HSR-TIGET grant), the independent drug research program of theItalian Medicines Agency (grant FARM5JRXRM), and the Euro-pean Commission: Concerted Safety and Efficiency Evaluation ofRetroviral Transgenesis in Gene Therapy of Inherited Diseases(grant LSBH-CT-2004-005242).

Authorship

Contribution: A.V.S. designed and performed most research, ana-lyzed data, and wrote the paper; E.M. designed and performed invitro experiments and analyzed data; R.J.H. performed animalexperiments; E.Z. performed pQCT analyses; F.C. performedhistomorphometric analyses; M.C. collected patient data; E.G.,C.M.R., and M.C.C. followed patients; A. Ambrosi performed thestatistical analysis; F.C. conducted the biochemical studies; M.G.R.and A.V. contributed to the study design; A.R. designed theresearch and wrote the paper; A. Aiuti designed the research,analyzed data, and revised the paper; and all authors checked thefinal version of the manuscript.

Conflict-of-interest disclosure: The authors declare no compet-ing financial interests.

Correspondence: Alessandro Aiuti, San Raffaele Telethon In-stitute for Gene Therapy, Via Olgettina 58, 20132 Milano, Italy;e-mail: [email protected].

References

1. Hirschhorn R, Candotti F. Immunodeficiency dueto defects of purine metabolism In: Ochs HD,Smith CIE, Puck JM, eds. Primary Immunodefi-ciency Diseases. Oxford, United Kingdom: Ox-ford University Press; 2006:169-196.

2. Aiuti A. Advances in gene therapy for ADA-deficient SCID. Curr Opin Mol Ther. 2002;4:515-522.

3. Ratech H, Greco MA, Gallo G, Rimoin DL,Kamino H, Hirschhorn R. Pathologic findings inadenosine deaminase-deficient severe combinedimmunodeficiency: I. Kidney, adrenal, andchondro-osseous tissue alterations. Am J Pathol.1985;120:157-169.

4. Cederbaum SD, Kaitila I, Rimoin DL, Stiehm ER.The chondro-osseous dysplasia of adenosinedeaminase deficiency with severe combined im-munodeficiency. J Pediatr. 1976;89:737-742.

5. Kaitila I, Rimoin DL, Cedarbaum SD, Stiehm ER,Lachman RS. Chondroosseous histopathology in

adenosine deaminase deficient combined immu-nodeficiency disease. Birth Defects Orig ArticSer. 1976;12:115-121.

6. Wortmann RL, Tekkanat KK, Veum JA, Meyer RA Jr,Hood JO, Horton WA. Basis for the chondro-osseousdysplasia associated with adenosine deaminasedeficiency: selective toxicity to immature chondro-cytes. Adv Exp Med Biol. 1991;309:265-268.

7. Hirschhorn R, Vawter GF, Kirkpatrick JA Jr,Rosen FS. Adenosine deaminase deficiency: fre-quency and comparative pathology in autosoma-lly recessive severe combined immunodeficiency.Clin Immunol Immunopathol. 1979;14:107-120.

8. MacDermot KD, Winter RM, Wigglesworth JS,Strobel S. Short stature/short limb skeletal dys-plasia with severe combined immunodeficiencyand bowing of the femora: report of two patientsand review. J Med Genet. 1991;28:10-17.

9. Lorenzo J, Horowitz M, Choi Y. Osteoimmunol-

ogy: interactions of the bone and immune sys-tem. Endocr Rev. 2008;29:403-440.

10. Marenzana M, Shipley AM, Squitiero P,Kunkel JG, Rubinacci A. Bone as an ionexchange organ: evidence for instantaneouscell-dependent calcium efflux from bone not dueto resorption. Bone. 2005;37:545-554.

11. Rubinacci A, Covini M, Bisogni C, et al. Bone asan ion exchange system: evidence for a link be-tween mechanotransduction and metabolicneeds. Am J Physiol Endocrinol Metab. 2002;282:E851-E864.

12. Hill P. Bone remodelling. J Orthod. 1998;25:101-107.

13. Tanaka S, Nakamura K, Takahasi N, Suda T. Roleof RANKL in physiological and pathological boneresorption and therapeutics targeting the RANKL-RANK signaling system. Immunol Rev. 2005;208:30-49.

14. Hasko G, Pacher P, Deitch EA, Vizi ES. Shaping

ADA-SCID BONE PHENOTYPE 3225BLOOD, 8 OCTOBER 2009 � VOLUME 114, NUMBER 15

For personal use only.on May 5, 2016. by guest www.bloodjournal.orgFrom

of monocyte and macrophage function by adeno-sine receptors. Pharmacol Ther. 2007;113:264-275.

15. Evans BA, Elford C, Pexa A, et al. Human osteo-blast precursors produce extracellular adenosine,which modulates their secretion of IL-6 and os-teoprotegerin. J Bone Miner Res. 2006;21:228-236.

16. Shimegi S. Mitogenic action of adenosine on os-teoblast-like cells, MC3T3-E1. Calcif Tissue Int.1998;62:418-425.

17. Orriss IR, Knight GE, Ranasinghe S, BurnstockG, Arnett TR. Osteoblast responses to nucleo-tides increase during differentiation. Bone. 2006;39:300-309.

18. Zhang J, Niu C, Ye L, et al. Identification of thehaematopoietic stem cell niche and control of theniche size. Nature. 2003;425:836-841.

19. Calvi LM, Adams GB, Weibrecht KW, et al. Os-teoblastic cells regulate the haematopoietic stemcell niche. Nature. 2003;425:841-846.

20. Visnjic D, Kalajzic Z, Rowe DW, Katavic V,Lorenzo J, Aguila HL. Hematopoiesis is severelyaltered in mice with an induced osteoblast defi-ciency. Blood. 2004;103:3258-3264.

21. Arai F, Suda T. Maintenance of quiescent hema-topoietic stem cells in the osteoblastic niche. AnnN Y Acad Sci. 2007;1106:41-53.

22. Li Y, Toraldo G, Li A, et al. B cells and T cells arecritical for the preservation of bone homeostasisand attainment of peak bone mass in vivo. Blood.2007;109:3839-3848.

23. Grunebaum E, Mazzolari E, Porta F, et al. Bonemarrow transplantation for severe combined im-mune deficiency. JAMA. 2006;295:508-518.

24. Booth C, Hershfield M, Notarangelo L, et al. Man-agement options for adenosine deaminase defi-ciency; proceedings of the EBMT satellite work-shop (Hamburg, March 2006). Clin Immunol.2007;123:139-147.

25. Aiuti A, Slavin S, Aker M, et al. Correction of ADA-SCID by stem cell gene therapy combined withnonmyeloablative conditioning. Science. 2002;296:2410-2413.

26. Aiuti A, Cassani B, Andolfi G, et al. Multilineagehematopoietic reconstitution without clonal selec-tion in ADA-SCID patients treated with stem cellgene therapy. J Clin Invest. 2007;117:2233-2240.

27. Chakravarti VS, Borns P, Lobell J, Douglas SD.

Chondroosseous dysplasia in severe combinedimmunodeficiency due to adenosine deaminasedeficiency (chondroosseous dysplasia in ADAdeficiency SCID). Pediatr Radiol. 1991;21:447-448.

28. Yulish BS, Stern RC, Polmar SH. Partial resolu-tion of bone lesions: a child with severe combinedimmunodeficiency disease and adenosine deami-nase deficiency after enzyme-replacementtherapy. Am J Dis Child. 1980;134:61-63.

29. Blackburn MR, Datta SK, Kellems RE. Adenosinedeaminase-deficient mice generated using a two-stage genetic engineering strategy exhibit a com-bined immunodeficiency. J Biol Chem. 1998;273:5093-5100.

30. Traggiai E, Chicha L, Mazzucchelli L, et al. Devel-opment of a human adaptive immune system incord blood cell-transplanted mice. Science. 2004;304:104-107.

31. Aiuti A, Cattaneo F, Galimberti F, et al. Long-termsafety and efficacy of gene therapy for adenosinedeaminase (ADA)-deficient severe combined im-munodeficiency. N Engl J Med. 2009;360:447-458.

32. Mortellaro A, Jofra Hernandez R, Guerrini MM, etal. Ex vivo gene therapy with lentiviral vectorsrescues adenosine deaminase (ADA)-deficientmice and corrects their immune and metabolicdefects. Blood. 2006;108:2979-2988.

33. Robey PG, Termine JD. Human bone cells invitro. Calcif Tissue Int. 1985;37:453-460.

34. Aiuti A, Vai S, Mortellaro A, et al. Immune recon-stitution in ADA-SCID after PBL gene therapy anddiscontinuation of enzyme replacement. Nat Med.2002;8:423-425.

35. Carlucci F, Tabucchi A, Aiuti A, et al. Capillaryelectrophoresis in diagnosis and monitoring ofadenosine deaminase deficiency. Clin Chem.2003;49:1830-1838.

36. Tullo A, Mastropasqua G, Bourdon JC, et al.Adenosine deaminase, a key enzyme in DNA pre-cursors control, is a new p73 target. Oncogene.2003;22:8738-8748.

37. Villa A, Vezzoni P, Frattini A. Osteopetroses andimmunodeficiencies in humans. Curr Opin AllergyClin Immunol. 2006;6:421-427.

38. Schoenau E, Frost HM. The “muscle-bone unit” inchildren and adolescents. Calcif Tissue Int. 2002;70:405-407.

39. Aubin JE. Regulation of osteoblast formation and

function. Rev Endocr Metab Disord. 2001;2:81-94.

40. Juge-Aubry CE, Somm E, Giusti V, et al. Adiposetissue is a major source of interleukin-1 receptorantagonist: up-regulation in obesity and inflam-mation. Diabetes. 2003;52:1104-1110.

41. Wilson A, Oser GM, Jaworski M, et al. Dormantand self-renewing hematopoietic stem cells andtheir niches. Ann N Y Acad Sci. 2007;1106:64-75.

42. Cheng T, Rodrigues N, Shen H, et al. Hematopoi-etic stem cell quiescence maintained by p21cip1/waf1. Science. 2000;287:1804-1808.

43. Salle BL, Rauch F, Travers R, Bouvier R, GlorieuxFH. Human fetal bone development: histomor-phometric evaluation of the proximal femoral me-taphysis. Bone. 2002;30:823-828.

44. Jimi E, Nakamura I, Duong LT, et al. Interleukin 1induces multinucleation and bone-resorbing ac-tivity of osteoclasts in the absence of osteoblasts/stromal cells. Exp Cell Res. 1999;247:84-93.

45. Hofbauer LC, Lacey DL, Dunstan CR, SpelsbergTC, Riggs BL, Khosla S. Interleukin-1beta andtumor necrosis factor-alpha, but not interleukin-6,stimulate osteoprotegerin ligand gene expressionin human osteoblastic cells. Bone. 1999;25:255-259.

46. Seeman E. Structural basis of growth-relatedgain and age-related loss of bone strength. Rheu-matology (Oxf). 2008;47[suppl 4]:2-8.

47. Rodriguez Mdel C, Bernad A, Aracil M. Interleu-kin-6 deficiency affects bone marrow stromal pre-cursors, resulting in defective hematopoietic sup-port. Blood. 2004;103:3349-3354.

48. Antoine C, Muller S, Cant A, et al. Long-term sur-vival and transplantation of haemopoietic stemcells for immunodeficiencies: report of the Euro-pean experience 1968-99. Lancet. 2003;361:553-560.

49. Baek KH, Oh KW, Lee WY, et al. Changes in theserum sex steroids, IL-7 and RANKL-OPG sys-tem after bone marrow transplantation: influenceson bone and mineral metabolism. Bone. 2006;39:1352-1360.

50. Ricci P, Tauchmanova L, Risitano AM, et al. Im-balance of the osteoprotegerin/RANKL ratio inbone marrow microenvironment after allogeneichemopoietic stem cell transplantation. Transplan-tation. 2006;82:1449-1456.

3226 SAUER et al BLOOD, 8 OCTOBER 2009 � VOLUME 114, NUMBER 15

For personal use only.on May 5, 2016. by guest www.bloodjournal.orgFrom

online July 24, 2009 originally publisheddoi:10.1182/blood-2009-03-209221

2009 114: 3216-3226  

Carlucci, Maria Grazia Roncarolo, Anna Villa, Alessandro Rubinacci and Alessandro AiutiCasiraghi, Eyal Grunebaum, Chaim M. Roifman, Maria C. Cervi, Alessandro Ambrosi, Filippo Aisha V. Sauer, Emanuela Mrak, Raisa Jofra Hernandez, Elena Zacchi, Francesco Cavani, Miriam osteoblast insufficiencybone phenotype characterized by RANKL/OPG imbalance and ADA-deficient SCID is associated with a specific microenvironment and 

http://www.bloodjournal.org/content/114/15/3216.full.htmlUpdated information and services can be found at:

(5380 articles)Immunobiology    (3365 articles)Hematopoiesis and Stem Cells   

Articles on similar topics can be found in the following Blood collections

http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requestsInformation about reproducing this article in parts or in its entirety may be found online at:

http://www.bloodjournal.org/site/misc/rights.xhtml#reprintsInformation about ordering reprints may be found online at:

http://www.bloodjournal.org/site/subscriptions/index.xhtmlInformation about subscriptions and ASH membership may be found online at:

  Copyright 2011 by The American Society of Hematology; all rights reserved.of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American Society

For personal use only.on May 5, 2016. by guest www.bloodjournal.orgFrom