Restoration of Adenosine Deaminase-Deficient Human Thymocyte Development In Vitro by Inhibition of Deoxynucleoside Kinases This information is current as of September 26, 2021. Michelle L. Joachims, Patrick A. Marble, Aletha B. Laurent, Peter Pastuszko, Marco Paliotta, Michael R. Blackburn and Linda F. Thompson J Immunol 2008; 181:8153-8161; ; doi: 10.4049/jimmunol.181.11.8153 Downloaded from http://www.jimmunol.org/content/181/11/8153
References This article cites 54 articles, 20 of which you can access for free at: http://www.jimmunol.org/ http://www.jimmunol.org/content/181/11/8153.full#ref-list-1
Why The JI? Submit online.
• Rapid Reviews! 30 days* from submission to initial decision
• No Triage! Every submission reviewed by practicing scientists by guest on September 26, 2021 • Fast Publication! 4 weeks from acceptance to publication
*average
Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts
The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2008 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology
Restoration of Adenosine Deaminase-Deficient Human Thymocyte Development In Vitro by Inhibition of Deoxynucleoside Kinases1,2
Michelle L. Joachims,* Patrick A. Marble,* Aletha B. Laurent,* Peter Pastuszko,3† Marco Paliotta,† Michael R. Blackburn,‡ and Linda F. Thompson4*
Mutations in the gene encoding adenosine deaminase (ADA), a purine salvage enzyme, lead to immunodeficiency in humans. Although ADA deficiency has been analyzed in cell culture and murine models, information is lacking concerning its impact on the development of human thymocytes. We have used chimeric human/mouse fetal thymic organ culture to study ADA-deficient human thymocyte development in an “in vivo-like” environment where toxic metabolites accumulate in situ. Inhibition of ADA during human thymocyte development resulted in a severe reduction in cellular expansion as well as impaired differentiation, Downloaded from largely affecting mature thymocyte populations. Thymocyte differentiation was not blocked at a discrete stage; rather, the paucity of mature thymocytes was due to the induction of apoptosis as evidenced by activation of caspases and was accompanied by the accumulation of intracellular dATP. Inhibition of adenosine kinase and deoxycytidine kinase prevented the accumulation of dATP and restored thymocyte differentiation and proliferation. Our work reveals that multiple deoxynucleoside kinases are involved in the phosphorylation of deoxyadenosine when ADA is absent, and suggests an alternate therapeutic strategy for treatment of
ADA-deficient patients. The Journal of Immunology, 2008, 181: 8153–8161. http://www.jimmunol.org/
utations in the gene encoding adenosine deaminase deficient infants suggests a profound defect in thymocyte devel- (ADA),5 a purine metabolic enzyme that catalyzes the opment (4). With the demonstration of elevated dATP in the RBC M deamination of both adenosine (Ado) and deoxyade- of ADA-deficient patients (5, 6), the ADA substrate dAdo became nosine (dAdo), cause severe combined immunodeficiency in hu- the prime candidate for causing lymphoid toxicity. However, little mans (1). ADA-deficient humans have a pronounced deficiency of is known about the specific stage or stages of lymphocyte devel- T cells with variable numbers of B and NK cells, and they die early opment that might be sensitive to a lack of ADA or of specific in life of infections unless treated with enzyme replacement ther- mechanism(s) responsible for the lack of thymocyte differentiation by guest on September 26, 2021 apy, bone marrow transplantation, or gene therapy (reviewed in and/or proliferation in ADA deficiency because of the difficulty in Refs. 2, 3). The paucity of lymphoid cells in thymuses from ADA- obtaining thymus samples from these rare patients and the lack of convenient culture systems for studying human thymocyte *Immunobiology and Cancer Program, Oklahoma Medical Research Foundation, development. Oklahoma City, OK 73104; †Department of Thoracic Surgery; University of Okla- The successive stages of human thymocyte maturation can be ‡ homa Health Sciences Center, Oklahoma City, OK 73104; and Department of Bio- categorized as double negative (DN), double positive (DP), or sin- chemistry and Molecular Biology, University of Texas-Houston Medical School, Houston, TX 77030 gle positive (SP) according to the presence or absence of the CD4 Received for publication November 16, 2007. Accepted for publication September 25, and CD8 coreceptors. Immature human DN thymocytes are clas- 2008. sified based on expression of CD34, CD38, and CD1a (reviewed in The costs of publication of this article were defrayed in part by the payment of page Ref. 7). The transition from the DN to DP stage occurs via a CD4 charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. immature single-positive (CD4 ISP) stage (8), accompanied by TCR gene rearrangements and the first developmental checkpoint, 1 This work was supported by National Institutes of Health grants F32 HD008709 (to M.L.J.), HD36044 (to L.F.T.), and AI43572 (to M.R.B.) and was conducted in a -selection. -selection involves the pairing of the TCR protein facility constructed with support from Research Facilities Improvement Program with the pre-T␣ protein to produce a membrane-localized pre-TCR Grant C06 RR14570-01 from the National Center for Research Resources, National  Institutes of Health. L.F.T. holds the Putnam City Schools Distinguished Chair in that signals survival, expansion, and allelic exclusion. -selection Cancer Research. occurs over a range of phenotypic stages from CD34ϩCD1aϩ cells 2 Parts of this work were presented in the Purine and Pyrimidine Society Meeting, to DP cells (9, 10). Successful -selection leads to the expression July 2007, in Chicago, IL. of the ␣TCR, followed by positive and negative selection, and 3 Current address: Rady Children’s Hospital, San Diego, CA 92123. development into mature CD4 and CD8 SP T cells (7, 11). Thy- 4 Address correspondence and reprint requests to Dr. Linda F. Thompson, Oklahoma muses from ADA-deficient patients have not been analyzed for Medical Research Foundation, 825 NE 13th Street, Oklahoma City, OK 73104. E-mail address: [email protected] their proportions of these normal human thymocyte subsets, nor 5 Abbreviations used in this paper: ADA, adenosine deaminase; Ado, adenosine; AK, has human thymocyte development been studied in vitro under adenosine kinase; dAdo, 2Ј-deoxyadenosine; 5ЈA5ЈdAdo, 5Ј-amino-5Ј-deoxyade- ADA-deficient conditions. nosine; dCF, 2Ј-deoxycoformycin; dCK, deoxycytidine kinase; dCyd, 2Ј-deoxycyti- Mouse models for ADA deficiency have, however, yielded im- dine; DN, double negative; DP, double positive; EDP, early double positive; FTOC, fetal thymic organ culture; hu/moFTOC, human/mouse chimeric fetal thymic organ portant clues to the pathogenesis of the immunodeficiency. In an in culture; ISP, immature single positive; SAH, S-adenosylhomocysteine; TCRic, in- vitro model for murine thymocyte development, fetal thymic organ tracellular TCR. culture (FTOC) (12–14), inhibition of ADA led to a marked re- Copyright © 2008 by The American Association of Immunologists, Inc. 0022-1767/08/$2.00 duction in lymphocyte development secondary to the accumulation www.jimmunol.org 8154 ADENOSINE DEAMINASE-DEFICIENT HUMAN THYMOCYTE DEVELOPMENT
of intracellular dATP and mitochondria-dependent apoptosis. Few Human neonatal thymus was obtained from children (ages 1 day to 4 thymocytes survived past the point of -selection, most likely be- years) undergoing cardiac surgery at Children’s Hospital in Oklahoma cause most thymocytes fail to pass this developmental checkpoint City, OK, under protocols approved by the Institutional Review Boards of both the University of Oklahoma Health Sciences Center and the Okla- and die by apoptosis providing a source of ADA substrates. Thy- homa Medical Research Foundation. Thymocyte cell suspensions were mocyte differentiation and proliferation could be rescued by the made by forcing the thymic tissue pieces through a nylon filter. Human adenosine kinase (AK) inhibitor 5Ј-amino-5Ј-deoxyadenosine CD34ϩ thymocytes were enriched with anti-CD34 magnetic beads (Dy- (5ЈA5ЈdAdo) (15), since the main route of dAdo phosphorylation nal Biotech) and sorted if needed to obtain Ն95% purity for use in hu/moFTOC. Staining for intracellular TCR (TCRic) expression was in murine cells is via AK (16, 17). performed by fixing cells with 1% formaldehyde and permeabilizing with It is well known that the pathway of dAdo phosphorylation in 0.5% saponin (10). Data were collected using an LSR II or FACSCalibur human thymocytes differs from that in the mouse (17, 18), making flow cytometer and analyzed with CellQuest software (BD Biosciences). it essential that models of human thymocyte development be uti- FITC annexin V (Caltag Laboratories) was used following the manufac- lized to understand the pathogenesis of ADA deficiency. Despite turer’s instructions after surface staining with anti-CD4 and anti-CD8 Abs. Propidium iodide staining for cell cycle assessment was performed as de- many early studies, there is still contention regarding which de- scribed (10) and analyzed using the Dean-Jett-Fox cell cycle analysis al- oxynucleoside kinase is primarily responsible for dAdo phosphor- gorithm in FlowJo analysis software (Tree Star). ylation in human thymocytes. Deoxycytidine kinase (dCK) is usu- Chimeric human/mouse fetal thymic organ culture ally thought to be the major cytoplasmic enzyme responsible for phosphorylation of deoxycytidine, deoxyguanosine, and deoxya- Hu/moFTOCs were performed as previously described (27). Briefly, fetal denosine (19). Human AK primarily utilizes Ado as a substrate, thymic lobes from timed pregnant C57BL/6 mice at day 15 of gestation (plug day ϭ day 0) were placed in FTOC for 4–6 days with 1.35 mM but it can also phosphorylate dAdo when it is present in high 2Ј-deoxyguanosine to deplete endogenous murine thymocytes. The lobes Downloaded from enough concentration (20, 21). Differential roles for these enzymes were then washed in complete medium to remove deoxyguanosine and depend on experimental context. In cell extracts, dCK was the placed in Terasaki wells with 25 l of Yssel’s medium (28) supplemented primary dAdo-phosphorylating enzyme (22), while in intact hu- with 5% FCS/2% human serum (Yssel’s complete medium) containing ϫ 4 man T and B lymphoblastoid cells, AK activity was more impor- 1–2 10 human thymocyte precursors. After 2 days of hanging drop culture, the lobes were transferred to a standard FTOC format in Yssel’s tant (21, 23). The relative roles of each of these nucleoside kinases complete medium and cultured with and without 5 M dCF. 5ЈA5ЈdAdo
in human ADA-deficient thymocyte development have yet to be was used at 5 M, and dCyd was used at 50 M with culture medium http://www.jimmunol.org/ investigated. replenished daily. Cultures were incubated for 1–3 wk and then harvested Herein we report the use of human/mouse chimeric fetal thymic by forcing the thymic lobes through a 70- m nylon mesh. Thymocytes were counted in trypan blue to assess viability, stained with Abs, and organ culture (hu/moFTOC) to study the consequences of ADA analyzed by flow cytometry or processed for dATP analysis. Suspension deficiency upon human thymocyte development. In this model sys- cultures for dATP accumulation were performed in Yssel’s complete me- tem, human CD34ϩ thymic precursor cells reconstitute murine dium with 5 M dCF and the indicated concentration of dAdo. Cultures thymocyte-depleted fetal thymic lobes and develop normally were incubated for 20 h at 37°C, and then cells were counted and processed through positive selection (24, 25). We show that cell expansion for dATP analysis by HPLC. and differentiation are severely impaired when ADA is inhibited. TCR gene rearrangement analysis by guest on September 26, 2021 We also demonstrate the importance of both dCK and AK in the Thymocyte genomic DNA from either control or dCF-treated hu/moF- accumulation of dATP and induction of apoptosis in developing TOCs was prepared using the Puregene kit (Gentra Systems). TCR gene thymocytes, evidenced by the normalization of dATP levels and nomenclature is that of the IMGT (International ImMunoGeneTics data- rescue of ADA-deficient thymocyte development in the presence base; imgt.cines.fr). V20.1 rearrangements to either J1.1–1.6 or J2.1– of multiple deoxynucleoside kinase inhibitors. Our studies lend 2.3 were amplified from 200 ng of genomic DNA using JumpStart TaqDNA polymerase (Sigma-Aldrich) (10). A portion of the RAG2 gene further support to the idea that dATP accumulation is the primary was amplified to normalize for input DNA in the PCR reactions. Primers cause of toxicity to developing human thymocytes under condi- used were: RAG2 forward, 5Ј-TGTGAATTGCACAGTCTTGCCAGG; tions of ADA deficiency, and they underscore differences between RAG2 reverse, 5Ј-GGGTTTGTTGAGCTCAGTTGAATAG; V20.1 for- Ј  Ј murine and human development. Our findings also suggest that ward, 5 -GATCGAGTGCCGTTCCCTGGACTT; J 1.6 reverse, 5 -AC CCTGACCTCCGTTCTTACACTC; J2.3 reverse, 5Ј-GGTGCCTGGG treatment of ADA-deficient patients with a combination of nucle- CCAAAATACTGCGTA. oside kinase inhibitors may be an alternative therapeutic strategy to prevent metabolic toxicity to developing thymocytes. HPLC analysis of dATP from thymocytes and S-adenosylhomocysteine (SAH) hydrolase activity assays Materials and Methods Single-cell suspensions were prepared at 4°C from 30 to 60 lobes of hu/ Mice moFTOC treated with dCF with and without deoxynucleoside kinase in- hibitors for 8–14 days. Aliquots were removed for cell counts and immu- C57BL/6 mice (The Jackson Laboratory) were bred in our animal facility nophenotyping, and then the remainder of the cells were immediately under specific pathogen-free conditions and in compliance with the Okla- pelleted and resuspended in 0.5 ml of ice-cold 60% MeOH. The cells were homa Medical Research Foundation Institutional Animal Care and Use extracted at Ϫ80°C for Ն18 h, then analyzed for dATP content by HPLC Committee specifications. as previously described (29). For SAH hydrolase activity measurements, Drugs and reagents thymocytes were washed with cold PBS and resuspended in 10 mM Tris- HCl (pH 7.5) plus protease inhibitors. Cell extracts were prepared by three The specific ADA inhibitor 2Ј-deoxycoformycin (dCF) (26) was obtained rounds of freeze-thaw lysis, followed by clarification by centrifugation. from SuperGen. 5ЈA5ЈdAdo, dAdo, 2Ј-deoxycytidine (dCyd), and all other SAH hydrolase enzyme activity was determined by measuring the forma- chemical reagents were obtained from Sigma-Aldrich. tion of SAH from homocysteine and radiolabeled adenosine as described previously (29). Abs, thymocyte cell preparation, and immunofluorescent staining Results Abs used were: FITC anti-CD1a, allophycocyanin anti-CD34, PE anti- ␥␦TCR, FITC anti-CD8, and PE anti-␣TCR (BD Pharmingen); PE and Inhibition of human thymocyte development under PE-Texas Red anti-CD8␣, PE-Cy5, PE-Cy5.5 and allophycocyanin anti- ADA-deficient conditions in hu/moFTOC CD4, PE anti-CD34, FITC and PE anti-␣TCR, FITC and allophycocyanin ␥␦  To assess the impact of ADA deficiency on developing human anti-CD3, and FITC anti- TCR (Caltag Laboratories); PE anti-CD8 (Se- ϩ rotec); and PE anti-TCRC (Ancell). Matched isotype control Abs were thymocytes, hu/moFTOCs were seeded with CD34 DN thymo- purchased from all of the above sources. cytes and allowed to develop for various time periods up to 3 wk The Journal of Immunology 8155 Downloaded from
FIGURE 2. Mature thymocytes are most affected in ADA-deficient hu/ moFTOC. A, Cells from hu/moFTOC with and without 5 M dCF were harvested at 3 wk, stained with Abs to ␣TCR and ␥␦TCR, and analyzed by flow cytometry (shaded histograms). Isotype control Ab staining is in- dicated by the black line histograms. B, Absolute numbers of human thy- http://www.jimmunol.org/ mocyte subpopulations in dCF-treated hu/moFTOC at 3 wk expressed as a FIGURE 1. Human thymocyte yield and differentiation are severely im- percentage of control cultures (mean Ϯ SD, n Ն 3). paired in ADA-deficient hu/moFTOC. A, CD34ϩ DN cells from human thymus were cultured in hu/moFTOC in Yssel’s complete medium with and without 5 M dCF for 3 wk, then harvested and stained with Abs to the absolute numbers of thymocyte subpopulations revealed that CD4 and CD8. Three individual experiments are shown. B, Cell yield from both ␣ϩ and ␥␦ϩ thymocytes were decreased (Fig. 2B). The in- dCF-treated cultures expressed as a percentage of control cultures har- creased percentages of ␥␦ϩ thymocytes likely reflects their devel- Ϯ Ն vested at wk 1, 2, and 3 of culture (mean SD,n 6). Live cells per lobe opment in an environment with very little overall cell expansion. were determined by trypan blue staining and counting. Fig. 2B also shows that the earliest stage of thymocyte develop- ment (DN) is least impacted by ADA deficiency. by guest on September 26, 2021 in the presence and absence of the specific tight-binding ADA Human thymocyte differentiation is not blocked at a discrete inhibitor dCF. DP thymocyte development was efficient in control stage in the absence of ADA cultures (Fig. 1A, left panels) but was dramatically impaired by the Because the yield of more mature populations of thymocytes was addition of dCF (Fig. 1A, right panels). A broad range of devel- so severely compromised under ADA-deficient conditions, we in- opmental inhibition was observed (from Ͻ5% to Ͼ50% DP thy- vestigated whether this might be the result of a block at a particular mocytes) as illustrated by data from three independent experi- differentiation step. Key early steps in the differentiation of ␣ ments. The decreased percentage of DP thymocytes was always lineage T cells involve the processes of TCR recombination and accompanied by increased percentages of CD4 ISP and DN thy- -selection. We first examined whether TCR gene rearrange- mocytes, indicating an inhibition of differentiation or a lack of ments might be inhibited in thymocytes developing in ADA-defi- survival and/or expansion of more mature thymocytes (or a com- cient conditions. V20.1, known as V2 in previous nomencla- bination of these). While cell yields progressively increased in ture, was chosen for this analysis because it is a commonly used control conditions, lack of ADA activity caused a cumulative de- V gene segment in the human T cell repertoire (30). Fig. 3A crease in cell yields, which were typically inhibited by Ն95% rel- shows that complete V 3 DJ TCR gene rearrangements to both ative to controls at 3 wk (Fig. 1B). Note that while the percentages D-J gene clusters, as analyzed by PCR, were readily detectable in of DP thymocytes varied from culture to culture, the absolute num- both control and dCF-treated hu/moFTOCs and were comparable bers of DP thymocytes were consistently decreased by Ն90% in in amounts to the rearrangements detected in total human thymo- dCF-treated hu/moFTOCs. Similar results were obtained when cyte DNA. Therefore, the absence of ADA activity did not lead to starting the cultures with the earliest thymocyte progenitors, an inability to rearrange TCR genes. ϩ Ϫ CD34 CD1a cells, or with more mature populations such as Next, we assessed the expression of both intracellular TCR and CD4 ISPs or early double positives (EDPs), although the kinetics CD8 in ADA-inhibited hu/moFTOC as a way to gauge the effi- of DP formation differed with the maturity of the starting popula- ciency of the -selection process and later differentiation events. tion (data not shown). Similar results were also obtained when dCF Staining thymocytes for TCRic expression is one way to evaluate was included in the hanging drop cultures. whether an in-frame TCR gene rearrangement has occurred. DP cells from control and dCF-treated cultures both showed similar Later stages of thymocyte development are most severely high levels of TCRic expression (Fig. 3B), indicating that this impacted by ADA deficiency requirement for -selection was met in ADA-deficient cultures. Percentages of ␣ϩ thymocytes were always severely reduced in Secondary to -selection, EDP CD8␣-expressing thymocytes up- ADA-inhibited cultures, while percentages of ␥␦ϩ thymocytes regulate CD8 expression (31). To determine whether this transi- were increased after 3 wk of culture (Fig. 2A). An examination of tion occurred normally in ADA-inhibited hu/moFTOC, DP cells 8156 ADENOSINE DEAMINASE-DEFICIENT HUMAN THYMOCYTE DEVELOPMENT Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021
FIGURE 3. Human thymocyte development in the absence of ADA is FIGURE 4. Induction of apoptosis, inhibition of cell proliferation, and not blocked at a particular stage. A, Cells from hu/moFTOC with and accumulation of dATP in cells from ADA-inhibited hu/moFTOC. A–D, without 5 M dCF were harvested at 3 wk and genomic DNA was pre- Hu/moFTOCs were incubated for 11 days with and without 5 M dCF; pared. TCR gene rearrangements from human V20.1 to either the J1or then cells were harvested and stained. One representative experiment out of the J2 region were analyzed by PCR and agarose gel electrophoresis. three is shown. A, Surface staining with anti-CD4 and anti-CD8 of total Genomic DNA from total human thymocytes (Thymus) or human fibro- thymocytes. B, Annexin V staining on total thymocytes from A. C, CD4/ blast cells (A549) were used as positive and negative controls, respectively. CD8 phenotype of annexin Vϩ cells in A. D, Cell cycle analysis of thy- B and C, Cells from cultures were harvested as in A, stained with Abs to mocytes from 11-day culture stained with propidium iodide as described in CD4, CD8␣, and CD8 (C), and analyzed using flow cytometry. Following Materials and Methods. E, Cells from 11-day hu/moFTOCs were pro- ic the surface stains, the cells were stained for TCR (B) as described in cessed for dATP analysis as described in Materials and Methods. dATP is Materials and Methods. The dot plots shown are gated on all live cells, and expressed in pmol/million cells (mean Ϯ SD, n ϭ 4). F, Total and CD34ϩ ic the histograms for TCR expression are shown gated on the DP cells thymocytes were incubated with 5 M dCF with and without dAdo for from each condition. Isotype control staining is indicated by the black line 20 h in suspension culture; then cells were harvested and analyzed for histograms. dATP as in E (mean Ϯ SD, n ϭ 4). were stained for the expression of cell surface CD8␣ and CD8. Even though percentages of DP cells were greatly inhibited rela- ADA inhibition results in induction of apoptosis, inhibition of tive to controls, approximately the same percentages of cells that cellular proliferation, and accumulation of intracellular dATP were CD4ϩCD8␣ϩ were also CD8␣ϩ, indicating that develop- ADA-deficient patients have elevated levels of the ADA substrates ment of EDP to true DP cells was not blocked in dCF-treated dAdo and Ado in the plasma, and dATP in RBCs (32). dATP can cultures (Fig. 3C). These results demonstrate that the differentia- induce mitochondrial cytochrome c release (33) and is a part of the tion of cells in ADA-inhibited dCF was not blocked at a discrete apoptosome (34, 35). Therefore, we examined the induction of stage of development, as assessed starting from the CD34ϩ thy- apoptosis in thymocytes from ADA-deficient hu/moFTOC by an- mocyte stage. Therefore, the selective loss of more mature thymo- nexin V staining. The CD4/CD8 phenotype at 11 days showed the cyte populations during ADA-inhibited development is likely due typical reduction in DP percentages (Fig. 4A), and annexin V stain- to increased cell death and/or lack of cell expansion. ing showed that ADA-inhibited cultures had nearly 3-fold higher The Journal of Immunology 8157 percentages of apoptotic cells compared with medium controls (Fig. 4B). Caspase activation assessed with a fluorogenic caspase substrate that can bind to all activated caspases (FAM-VAD-fmk) showed 4-fold elevated percentages of activated caspases in ADA- inhibited cells (data not shown). In both dCF-treated and control cultures, DP cells were the primary population undergoing apo- ptosis (Fig. 4C), although higher fractions of less mature cells (DNs and CD4 ISPs) were also seen in the dCF-treated cultures. Because cell yields were so dramatically inhibited in the ab- sence of ADA activity, we next analyzed the proliferative capacity of the developing thymocytes. Propidium iodide staining at 11 days of culture showed that nearly two-thirds fewer cells were in cycle in dCF-treated cultures compared with medium controls (Fig. 4D). Ki-67 staining supported this result, as fewer cells were in cycle in all phenotypic compartments in ADA-inhibited cultures (data not shown). We next asked whether the induction of apoptosis in ADA- deficient hu/moFTOC might be caused by the accumulation of dATP as was found in ADA-deficient murine FTOCs (12). HPLC Downloaded from analyses revealed that dATP accumulated to high levels (Ͼ15-fold over control) in 11-day cultures of ADA-inhibited hu/moFTOC ( p ϭϽ0.00001; Fig. 4E). dATP could be detected as early as 8 days of culture, peaked at days 10–12, and declined thereafter to ϳ2-fold over control at 3 wk of culture (data not shown). ϩ
Hu/moFTOCs were initiated with immature CD34 DN thymo- http://www.jimmunol.org/ cytes whose deoxynucleoside metabolism is known to change with differentiation (36). Because DN cells were least affected by ADA inhibition (Fig. 2B), we compared the ability of these cells to ac- cumulate dATP to that of total thymocytes (ϳ80% DP cells). CD34ϩ thymocytes and total thymocytes were incubated in sus- pension cultures with dCF and varying concentrations of dAdo for FIGURE 5. Rescue of ADA-deficient phenotype and dATP accumula- 20 h, followed by harvest of the cells for dATP analysis. On av- ϩ tion in hu/moFTOC by inhibition of deoxynucleoside kinases. A, Human erage, CD34 thymocytes accumulated 10-fold as much dATP per CD34ϩ cells were cultured in hu/moFTOC in medium alone, 5 M by guest on September 26, 2021 cell as total thymocytes, even at relatively low dAdo concentra- 5ЈA5ЈdAdo, 50 M dCyd, or 50 M dCyd ϩ 5 M5ЈA5ЈdAdo Ϯ 5 M tions (Ͻ10 M) (Fig. 4F). Therefore, the relative survival advan- dCF for 10 days, then harvested, counted, and stained with Abs to CD4 and tage of immature thymocytes cannot be explained by an inability CD8. B, Cell yield from experiments as in A, expressed as a percentage of to accumulate dATP. In fact, these high levels of dATP did not each control condition (mean Ϯ SD, n ϭ 4). C, Analysis of dATP accu- induce any more caspase activation or apoptosis than was observed mulation for experiments as in A. dATP values are expressed as pmol/ in cultures of total thymocytes at comparable concentrations of million cells (mean Ϯ SD, n ϭ 4). D, Hu/moFTOCs incubated with the Ј Ј dAdo (data not shown). indicated combinations of 5 M dCF, 5 M5A5 dAdo, and 50 M dCyd were harvested after 3 wk, stained with Abs to either ␣TCR or ␥␦TCR, Rescue of human thymocyte development in ADA-deficient and analyzed by flow cytometry. hu/moFTOC by inhibition of deoxynucleoside kinases The intracellular dATP in ADA-deficient thymocytes is derived dCyd and 5ЈA5ЈdAdo were used, a complete rescue of DP thymo- from the ADA substrate dAdo by the action of cellular de- cyte development and cell expansion was observed (Fig. 5, A and oxynucleoside kinases. To attempt to prevent the accumulation of B). Cells from dCF- and dCFϩ5ЈA5ЈdAdo-treated cultures had dATP and rescue thymocyte development in ADA-deficient hu/ similar amounts of accumulated dATP, at levels ϳ15-fold over moFTOC, we inhibited the activity of dCK and AK, either sepa- control cultures (Fig. 5C). Interestingly, cells from dCF ϩ dCyd- rately or in combination (Fig. 5). AK activity can be efficiently treated cultures produced significantly higher levels of accumu- inhibited by a structural analog of dAdo, 5ЈA5ЈdAdo, which has lated dATP, consistently ϳ2.5-fold higher than cells from cultures nanomolar affinity for AK (15), while micromolar concentrations treated with dCF alone ( p ϭ 0.003). However, there was no further of dCyd competitively inhibit the phosphorylation of dAdo by decrease in cell yield (Fig. 5B) or increase in apoptosis (data not dCK and promote feedback inhibition of dCK by the phosphory- shown). This is consistent with the dATP concentration having lated product, dCTP (37). We first assessed the impact on devel- peaked over a “threshold” value needed to induce the effects ob- opment at an early stage (10–12 days) when robust development served with inhibition of ADA, above which there is little further and expansion of DP thymocytes occur in control cultures. The effect. The use of both kinase inhibitors (dCyd ϩ 5ЈA5ЈdAdo) in addition of the inhibitors alone, either singly or in combination, did dCF-treated cultures corrected both the phenotype and cell yields not significantly perturb the expansion and development of thy- of the cultures, as well as normalized dATP levels to nearly those mocytes in hu/moFTOC (Fig. 5A and data not shown). The addi- of the control. tion of either dCyd or 5ЈA5ЈdAdo to the dCF-treated cultures did We next asked whether dCyd ϩ 5ЈA5ЈdAdo could also rescue not significantly restore the development of DP thymocytes (Fig. longer (3 wk) ADA-inhibited cultures. Fig. 5D shows an assess- 5A). Although the cell yields increased slightly with the addition of ment of ␣TCR and ␥␦TCR thymocyte development by flow cy- either dCyd or 5ЈA5ЈdAdo during dCF treatment (Fig. 5B), these tometry. Low percentages of ␣ϩ thymocytes and higher percent- increases were not statistically significant. However, when both ages of ␥␦ϩ thymocytes were observed in dCF-treated and dCF ϩ 8158 ADENOSINE DEAMINASE-DEFICIENT HUMAN THYMOCYTE DEVELOPMENT
vivo-like model for ADA deficiency, dCF-treated hu/moFTOC. Hu/moFTOC is an ideal model system for investigating human thymocyte development because the early events in T cell differ- entiation take place spontaneously without the need for exogenous cytokines or growth factors. Furthermore, the early murine fetal thymic lobe lacks a thick capsule and is small enough to permit optimal exchange of gases and nutrients when cultured at the air/ medium interface (40). Relatively little has been known about how human thymocytes develop in an ADA-deficient environment. Studies on patient RBCs or measurements of metabolites from patient fluids constituted the bulk of information known about pu- rine metabolism in ADA-deficient patients. ADA-inhibited hu/ FIGURE 6. SAH hydrolase activity is inhibited in rescued ADA-defi- moFTOC approximates the metabolic and developmental environ- cient hu/moFTOC. SAH hydrolase activity was measured as described in ment of an ADA-deficient thymus, since toxic metabolites are Materials and Methods in cell extracts of thymocytes harvested from hu/ generated in situ, rather than being provided exogenously, as in moFTOC cultured for 12 days. Results are expressed as pmol/h/106 cells. previous models (41). Although we cannot be certain that the con- The data shown are representative of two separate experiments. centrations of toxic metabolites are exactly the same as in the thymuses of ADA-deficient patients, we show that the addition of dCyd-treated cultures, indicating little correction in long-term de- the ADA inhibitor dCF to hu/moFTOC causes a cumulative loss of Downloaded from ϩ velopment. While there was no increase in the percentage of ␣ mature thymocytes, accompanied by induction of apoptosis, accu- thymocytes in dCF ϩ 5ЈA5ЈdAdo-treated cultures, the percentage mulation of intracellular dATP, and inhibition of proliferation, but ϩ of ␥␦ thymocytes was reduced by ϳ50% relative to cultures not a block in differentiation. Importantly, we show that normal treated with dCF alone. The addition of both kinase inhibitors to thymocyte development, including the generation of TCRϩ cells, ϩ the dCF-treated cultures restored the development of TCR thy- is restored when the activities of AK and dCK are inhibited during
mocytes to proportions similar to those in control cultures, indi- culture. http://www.jimmunol.org/ cating that inhibition of both deoxynucleoside kinases in long-term The effects of ADA inhibition in hu/moFTOC were cumulative, hu/moFTOCs could effectively restore the development of ␣ and with the loss of mature thymocytes more pronounced with longer ␥␦ thymocytes. Although the cell yields were not completely res- cultures (Figs. 1B,2B, and data not shown). Differentiation was not cued (ϳ40% of controls at 3 wk), the absolute numbers of both DP blocked at a discrete developmental stage, evidenced by detection ϩ and ␣ thymocytes showed substantial (50-fold) improvements. of complete TCR rearrangements and expression of DP differen- In summary, our data suggest that both dCK and AK play a role in tiation markers (TCRic and CD8; Fig. 3). It is unknown whether the phosphorylation of the ADA substrate dAdo, leading to dATP the earliest multipotent thymic progenitors or stem cells are af- accumulation in hu/moFTOC. Further, our data show that inhibi- fected by ADA deficiency. Cultures initiated with pure sorted tion of both dCK and AK normalized dATP levels in hu/moFTOC CD34ϩCD1aϪ thymocytes (the earliest thymocyte progenitors) by guest on September 26, 2021 and overcame the effect of ADA deficiency, allowing mature T cell readily generated CD4 ISPs in the presence of dCF and gave re- development. sults that were indistinguishable from those initiated with CD34ϩ DN cells (primarily CD1aϩ, data not shown). Although these data Rescue of ADA-deficient hu/moFTOC by deoxynucleoside kinase do not address this issue directly, we think it is unlikely that stem inhibition does not prevent inhibition of SAH hydrolase cell differentiation is impaired by ADA deficiency since enzyme SAH hydrolase is a high-affinity adenosine-binding protein that replacement therapy provides an effective treatment for ADA-de- catalyzes the reversible reaction, Ado ϩ L-homocysteine N SAH. ficient patients. It may be that reduced cell division (Fig. 4D)isthe In the absence of ADA, two factors contribute to dysregulation of more significant factor in the lack of thymocyte development in SAH metabolism. First, the equilibrium constant of this reaction ADA-deficient thymus. Histological studies of ADA-deficient thy- favors SAH formation, and increased Ado levels potentiate the mic tissue from fetuses revealed a small thymic tissue rudiment accumulation of SAH, a product and potent inhibitor of numerous devoid of lymphoid cells or normal thymic architecture (4). Lack S-adenosyl methionine-dependent methyl transferase reactions. of thymocyte expansion would lead to a thymus devoid of lym- This inhibition of transmethylation reactions likely accounts for phocytes, without which a normal thymic architecture fails to nucleotide-independent adenosine toxicity to lymphoblasts (38) develop. and has been considered as a potential contributing mechanism to Alhough the absolute numbers of DP thymocytes were always the immunodeficiency seen in ADA-deficient patients. Second, inhibited by Ͼ90% relative to controls, the proportions of DP thy- SAH hydrolase activity is inhibited by dAdo in an active-site di- mocytes in ADA-deficient hu/moFTOC varied from culture to cul- rected, “suicide-like” process (39). To assess the role of SAH hy- ture (Fig. 1A). While some of this is likely due to human sample drolase inhibition in the development of human thymocytes in heterogeneity, most could be explained by the nonsynchronous ADA-deficient hu/moFTOC, we measured SAH hydrolase activity occurrence of -selection in human thymocyte development. Thy- in cells from ADA-inhibited and rescued cultures (Fig. 6). SAH mocytes that fail -selection provide the first major source of ADA hydrolase activity was inhibited under ADA-deficient conditions substrate to other thymocytes through degradation of apoptotic cell as expected and remained inhibited in cultures that had been res- DNA. Cells that fail -selection at an early stage (i.e., at the cued with kinase inhibitors. It is therefore unlikely that inhibition CD34ϩCD1aϩ or CD4 ISP stage) generate substrate (dAdo) that of SAH hydrolase plays a significant role in the toxicity to devel- can be taken up by adjacent cells and converted to dATP, inducing oping thymocytes in ADA-deficient hu/moFTOC. additional apoptosis before significant numbers of thymocytes reach the DP stage. Cultures where -selection occurs earlier Discussion would yield low percentages of DP thymocytes (Fig. 1A, top In this study, we provide the first description of the impact of panel). However, thymocytes with a delayed -selection process, ADA deficiency on human thymocyte development using an in occurring mainly in the DP stage, would not generate toxic levels The Journal of Immunology 8159 of ADA substrates until many DP thymocytes had already been oxynucleoside diphosphate reduction by ribonucleotide reductase produced, leading to higher percentages of surviving DP thymo- (42, 46) and allow more dADP and dATP accumulation. Clearly, cytes (Fig. 1A, bottom panel). Our studies of TCRic expression in some facets of these enzymes’ regulation are not well understood isolated ex vivo thymocytes suggest that the point of -selection and will need to be further studied to provide an explanation for varies substantially from thymus to thymus (10) and provide an the increased dATP accumulation when ADA and dCK are both explanation for the variable proportions of surviving DP thymo- inhibited. cytes in ADA-deficient hu/moFTOC. Because both Ado and dAdo are elevated in ADA-deficient pa- Cells from ADA-inhibited cultures had high levels of induced tients, it has been difficult to discern the relative contribution of apoptosis accompanied by accumulation of intracellular dATP Ado and dAdo in the toxicity to developing lymphocytes. Al- (Fig. 4). A consistent finding was that immature thymocytes were though the primary focus in more recent years has been on the relatively resistant to ADA inhibition (Fig. 2B). Our initial hypoth- effects of aberrant dAdo metabolism, some studies have suggested esis was that they had less ability to accumulate dATP. However, a possible role for Ado in ADA deficiency through the action of ϩ ϳ data in Fig. 4F show that CD34 thymocytes accumulate 10- Ado receptors and blocking of NF-B activity (47, 48). Studies fold higher levels of dATP per cell than do total thymocytes with murine FTOCs convincingly ruled out a major role for Ado in (which are ϳ80% DP thymocytes), likely a consequence of their inhibition of thymocyte development under ADA-deficient condi- relatively high deoxynucleoside kinase activities, coupled with low tions (13). Our own studies also do not support a role for Ado in activity of enzymes (nucleotidases) that can carry out the reverse ϩ the inhibition of human thymocyte development in dCF-treated reaction (36, 42). Furthermore, CD34 thymocytes do not undergo hu/moFTOC, as adenosine receptor antagonists failed to protect dATP-induced apoptosis in suspension cultures to a greater degree Downloaded from than more mature cells (43), perhaps because their greater size from the consequences of ADA deficiency and adenosine receptor decreases the actual intracellular concentration of dATP. On av- agonists had no appreciable effect on thymocyte development (data erage, CD34ϩ thymocytes are 8.7 m in diameter, while DP thy- not shown). Furthermore, rescue of dCF-treated hu/mo FTOC with mocytes are 6.0 m (data not shown). However, the diameters of a combination of dCyd and an AK inhibitor should increase intra- the nuclei of CD34ϩ and DP thymocytes are similar, so the cyto- cellular Ado concentrations. Thus, while it is possible that Ado ϩ plasmic volume of CD34 cells is actually more than three times signaling through AdoR could play a role in immunomodulation of http://www.jimmunol.org/ that of DP thymocytes. It may also be that immature cells are the few surviving T and B cells in the periphery of ADA-deficient inherently more resistant to apoptotic signals, as suggested by the patients (49), it is less likely that Ado has a major impact on de- fact that immature T cell leukemias are very difficult to treat (44). velopment of T cells in the thymus. We propose that as thymocytes mature, the marked decrease in cell It is also highly unlikely that SAH hydrolase inhibition, as size from the CD34ϩ to the DP stage increases the intracellular occurs in the RBCs of ADA-deficient patients (39), has a large dATP concentration enough to induce cytochrome c release from impact on thymocyte development during ADA deficiency. mitochondria, triggering apoptosis in the vast majority of cells. SAH hydrolase remained inhibited under ADA-inhibited con- This idea is consistent with the finding that dATP was greatly ditions where dATP accumulation was prevented and develop- reduced in cells from 3 wk dCF-treated hu/moFTOC compared ment was rescued by the use of deoxynucleoside kinase in- by guest on September 26, 2021 with earlier time points. We postulate that the only surviving cells hibitors (Fig. 6). Furthermore, when murine thymocyte in longer cultures were rare cells that failed to accumulate dATP or development was assessed in FTOC under conditions of inhib- to undergo apoptosis as a part of normal development. In the res- ited methylation, thymocyte development was perturbed not by cued cultures (Fig. 5), the elimination of both kinase activities apoptosis, but by inhibition of transcription of molecules lowered dATP enough to allow survival when the cells transi- needed for development, such as the TCR and CD4/CD8 (50). tioned to the DP stage. Inhibition of SAH hydrolase during ADA deficiency is there- Our observation of hyperelevated dATP in ADA-deficient hu/ fore likely to be a consequence, but not a major cause, of the moFTOC treated with dCyd (Fig. 5C) provides a plausible expla- pathology induced by aberrant dAdo metabolism during thymo- nation for why dCyd therapy provided no clinical benefit in ADA- cyte development. deficient patients (45). Treatment of ADA-inhibited thymocytes Treatments for ADA deficiency usually entail a bone marrow with dCyd in the presence of exogenous dAdo in suspension cul- transplant or, if a suitable donor is not available, enzyme replace- tures prevented the accumulation of dATP (42, 43) and was no ment therapy using pegylated bovine ADA enzyme. While the best doubt the rationale for the use of dCyd therapy in several ADA- chance for a cure is provided by bone marrow transplant, this is deficient patients. However, our data suggest that AK also plays an often not a feasible option. Although initially efficacious, treatment important role in dAdo phosphorylation in our model. Therefore, of ADA-deficient patients using pegylated bovine ADA enzyme analogous to the differences in enzymes that phosphorylate dAdo results in low long-term lymphocyte restoration, with chronic de- in cell extracts compared with intact cells, differences clearly exist in the phosphorylation of dAdo by thymocytes in suspension vs in terioration of T cell responses (51). Furthermore, a recent report organ culture, a more in vivo-like environment. Because RBCs has revealed that a high proportion of bone marrow transplant- from ADA-deficient patients accumulate high levels of dATP even treated ADA-deficient patients are now experiencing progressive though erythrocytes have virtually no detectable dCK activity (2), neurological abnormalities, indicating that hematopoietic reconsti- there is in vivo precedence for the involvement of AK in the for- tution is insufficient to completely correct the systemic metabolic mation of dATP during ADA deficiency. The fact that AK is re- abnormalities associated with lack of ADA enzyme activity in all sponsible for dATP accumulation in vivo leads us to hypothesize other organ systems (52). Although it is unknown whether the that the hyperelevation in dATP in hu/moFTOC treated with mechanisms of neurotoxicity are similar to those of the dATP- dCF ϩ dCyd could be due to a shift to AK as the major dAdo- induced apoptosis occurring in lymphocytes, autopsy material phosphorylating enzyme when dCK is inhibited, consistent with from an ADA-deficient patient clearly showed accumulation of previous studies in kinase-deficient human T lymphoblastoid cell dATP in nonlymphoid tissues such as brain (53). lines (21). Alternatively, the addition of dCyd may be stimulating Our studies provide rationale for consideration of an additional the formation of excess dTTP, which will in turn increase de- therapeutic alternative for ADA-deficient patients: treatment with 8160 ADENOSINE DEAMINASE-DEFICIENT HUMAN THYMOCYTE DEVELOPMENT deoxynucleoside kinase inhibitors. Deoxycytidine tested in ADA- 16. Ullman, B., L. J. Gudas, A. Cohen, and D. W. Martin, Jr. 1978. Deoxyadenosine deficient patients was systemically nontoxic (45). Adenosine ki- metabolism and cytotoxicity in cultured mouse T lymphoma cells: a model for immunodeficiency disease. Cell 14: 365–375. nase inhibitors are currently being developed for treatments of var- 17. Carson, D. A., J. Kaye, and D. B. Wasson. 1980. Differences in deoxyadenosine ious neurological manifestations such as stroke, as AK is the main metabolism in human and mouse lymphocytes. J. Immunol. 124: 8–12. enzymatic activity responsible for the regulation of extracellular 18. Snyder, F. F., and T. Lukey. 1980. Purine ribonucleoside and deoxyribonucleo- side metabolism in thymocytes. Adv. Exp. Med. Biol. 122B: 259–264. adenosine levels (54) and adenosine is known to be neuroprotec- 19. Arner, E. S., and S. Eriksson. 1995. Mammalian deoxyribonucleoside kinases. tive. Interestingly, the AK inhibitor used in our study, 5ЈA5ЈdAdo, Pharmacol. Ther. 67: 155–186. lacks in vivo efficacy due to its deamination by ADA (15); clearly, 20. Hurley, M. C., B. Lin, and I. H. Fox. 1986. Regulation of deoxyadenosine and nucleoside analog phosphorylation by human placental adenosine kinase. Adv. this would be not be an issue in ADA-deficient patients. Thus, Exp. Med. Biol. 195B: 141–149. treatment with deoxynucleoside kinase inhibitors could be a useful 21. Hershfield, M. S., J. E. Fetter, W. C. Small, A. S. Bagnara, S. R. Williams, B. Ullman, D. W. Martin, Jr., D. B. Wasson, and D. A. Carson. 1982. Effects of adjunct to using pegylated bovine ADA enzyme therapy and/or mutational loss of adenosine kinase and deoxycytidine kinase on deoxyATP ac- bone marrow transplant and might ameliorate the neurotoxicity cumulation and deoxyadenosine toxicity in cultured CEM human T-lymphoblas- associated with lack of ADA activity in the CNS. toid cells. J. Biol. Chem. 257: 6380–6386. 22. Hershfield, M. S., and N. M. Kredich. 1980. Resistance of an adenosine kinase- deficient human lymphoblastoid cell line to effects of deoxyadenosine on growth, Acknowledgments S-adenosyl-homocysteine hydrolase inactivation, and dATP accumulation. Proc. Natl. Acad. Sci. USA 77: 4292–4296. The authors thank Drs. Paul Kincade and Justin Van De Wiele for critical 23. Ullman, B., B. B. Levinson, M. S. Hershfield, and D. W. Martin, Jr. 1981. A review of the manuscript. biochemical genetic study of the role of specific nucleoside kinases in deoxya- denosine phosphorylation by cultured human cells. J. Biol. Chem. 256: 848–852.
24. Plum, J., M. De Smedt, B. Verhasselt, T. Kerre, D. Vanhecke, Downloaded from Disclosures B. Vandekerckhove, and G. Leclercq. 2000. Human T lymphopoiesis: in vitro The authors have no financial conflicts of interest. and in vivo study models. Ann. NY Acad. Sci. 917: 724–731. 25. Spits, H., P. Res, and A. C. Jaleco. 2000. Techniques for studying development of human natural killer cells and T cells. Methods Mol. Biol. 121: 25–38. References 26. Agarwal, R. P., T. Spector, and R. E. Parks, Jr. 1977. Tight-binding inhibitors: 1. Giblett, E. R., J. E. Anderson, F. Cohen, B. Pollara, and H. J. Meuwissen. 1972. IV. Inhibition of adenosine deaminases by various inhibitors. Biochem. Pharma- Adenosine-deaminase deficiency in two patients with severely impaired cellular col. 26: 359–367. immunity. Lancet 2: 1067–1069. 27. Res, P., B. Blom, T. Hori, K. Weijer, and H. Spits. 1997. Downregulation of CD1 2. Hershfield, M. S., and B. S. Mitchell. 2001. Immunodeficiency diseases caused marks acquisition of functional maturation of human thymocytes and defines a http://www.jimmunol.org/ by adenosine deaminase deficiency and purine nucleoside phosphorylase defi- control point in late stages of human T cell development. J. Exp. Med. 185: ciency. In The Metabolic and Molecular Bases of Inherited Disease, Vol. 2. 141–151. C. R. Scriver, W. S. Sly, B. Childs, A. L. Beaudet, D. Valle, K. Kinzler, and 28. Yssel, H., J. E. De Vries, M. Koken, W. Van Blitterswijk, and H. Spits. 1984. B. Vogelstein, eds. McGraw-Hill, New York, pp. 2585–2625. Serum-free medium for generation and propagation of functional human cyto- 3. Blackburn, M. R., and R. E. Kellems. 2005. Adenosine deaminase deficiency: toxic and helper T cell clones. J. Immunol. Methods 72: 219–227. metabolic basis of immune deficiency and pulmonary inflammation. Adv. Immu- 29. Wakamiya, M., M. R. Blackburn, R. Jurecic, M. J. McArthur, R. S. Geske, nol. 86: 1–41. J. Cartwright, Jr., K. Mitani, S. Vaishnav, J. W. Belmont, R. E. Kellems, et al. 4. Linch, D. C., R. J. Levinsky, C. H. Rodeck, K. A. Maclennan, and 1995. Disruption of the adenosine deaminase gene causes hepatocellular impair- H. A. Simmonds. 1984. Prenatal diagnosis of three cases of severe combined ment and perinatal lethality in mice. Proc. Natl. Acad. Sci. USA 92: 3673–3677. immunodeficiency: severe T cell deficiency during the first half of gestation in 30. van den Beemd, R., P. P. Boor, E. G. van Lochem, W. C. Hop, A. W. Langerak, fetuses with adenosine deaminase deficiency. Clin. Exp. Immunol. 56: 223–232. I. L. Wolvers-Tettero, H. Hooijkaas, and J. J. van Dongen. 2000. Flow cytometric by guest on September 26, 2021 5. Donofrio, J., M. S. Coleman, and J. J. Hutton. 1978. Overproduction of adenine analysis of the V repertoire in healthy controls. Cytometry 40: 336–345. deoxynucleosides and deoxynucleotides in adenosine deaminase deficiency with 31. Hori, T., J. Cupp, N. Wrighton, F. Lee, and H. Spits. 1991. Identification of a severe combined immunodeficiency disease. J. Clin. Invest. 62: 884–887. novel human thymocyte subset with a phenotype of CD3Ϫ CD4ϩ CD8 alphaϩ 6. Cohen, A., R. Hirschhorn, S. D. Horowitz, A. Rubinstein, S. H. Polmar, R. Hong, betaϪ: possible progeny of the CD3Ϫ CD4Ϫ CD8Ϫ subset. J. Immunol. 146: and D. W. Martin, Jr. 1978. Deoxyadenosine triphosphate as a potentially toxic 4078–4084. metabolite in adenosine deaminase deficiency. Proc. Natl. Acad. Sci. USA 75: 32. Coleman, M. S., J. Donofrio, J. J. Hutton, and L. Hahn. 1978. Identification and 472–476. ␣ quantitation of adenine deoxynucleotides in erythrocytes of a patient with aden- 7. Spits, H. 2002. Development of T cells in the human thymus. Nat. Rev. osine deaminase deficiency and severe combined immunodeficiency. J. Biol. Immunol. 2: 760–772. Chem. 253: 1619–1626. 8. Galy, A., S. Verma, A. Barcena, and H. Spits. 1993. Precursors of 33. Yang, J. C., and G. A. Cortopassi. 1998. dATP causes specific release of cyto- CD3ϩCD4ϩCD8ϩ cells in the human thymus are defined by expression of CD34: chrome C from mitochondria. Biochem. Biophys. Res. Comm. 250: 454–457. delineation of early events in human thymic development. J. Exp. Med. 178: 34. Li, P., D. Nijhawan, I. Budihardjo, S. M. Srinivasula, M. Ahmad, E. S. Alnermi, 391–401. and X. Wang. 1997. Cytochrome c and dATP-dependent formation of Apaf-1/ 9. Dik, W. A., K. Pike-Overzet, F. Weerkamp, D. de Ridder, E. F. E. de Haas, caspase-9 complex initiates an apoptotic protease cascade. Cell 91: 479–489. M. R. M. Baert, P. van der Spek, E. E. L. Koster, M. J. T. Reinders, et al. 2005. ⅐ c New insights on human T cell development by quantitative T cell receptor gene 35. Zou, H., Y. Li, X. Liu, and X. Wang. 1999. An APAF-1 cytochrome multi- J. Biol. rearrangement studies and gene expression profiling. J. Exp. Med. 201: meric complex is a functional apoptosome that activates procaspase-9. Chem. 1715–1723. 274: 11549–11556. 10. Joachims, M. L., J. L. Chain, S. W. Hooker, C. J. Knott-Craig, and 36. Ma, D. D. F., T. A. Sylwestrowicz, S. Granger, M. Massaia, R. Franks, L. F. Thompson. 2006. Human ␣ and ␥␦ thymocyte development: TCR gene G. Janossy, and A. V. Hoffbrand. 1982. Distribution of terminal deoxynucleotidyl rearrangements, intracellular TCR expression, and ␥␦ developmental potential: transferase and purine degradative and synthetic enzymes in subpopulations of differences between men and mice. J. Immunol. 176: 1543–1552. human thymocytes. J. Immunol. 129: 1430–1435. 11. Singer, A., and R. Bosselut. 2004. CD4/CD8 coreceptors in thymocyte develop- 37. Bohman, C., and S. Eriksson. 1988. Deoxycytidine kinase from human leukemic ment, selection, and lineage commitment: analysis of the CD4/CD8 lineage de- spleen: preparation and characteristics of homogeneous enzyme. Biochemistry cision. Adv. Immunol. 83: 91–131. 27: 4258–4265. 12. Thompson, L. F., C. J. Van De Wiele, A. B. Laurent, S. W. Hooker, J. G. Vaughn, 38. Kredich, N. M., and M. S. Hershfield. 1979. S-adenosylhomocysteine toxicity in H. Jiang, K. Khare, R. E. Kellems, M. R. Blackburn, M. S. Hershfield, and normal and adenosine kinase deficient lymphoblasts of human origin. Proc. Natl. R. Resta. 2000. Metabolites from apoptotic thymocytes inhibit thymopoiesis in Acad. Sci. USA 76: 2450–2454. adenosine deaminase-deficient fetal thymic organ cultures. J. Clin. Invest. 106: 39. Hershfield, M. S., N. M. Kredich, D. R. Ownby, and R. Buckley. 1979. In vivo 1149–1157. inactivation of erythrocyte S-adenosylhomocysteine hydrolase by 2Ј-deoxyade- 13. Van De Wiele, C. J., J. G. Vaughn, M. R. Blackburn, C. Ledent, M. Jacobson, nosine in adenosine deaminase-deficient patients. J. Clin. Invest. 63: 807–811. H. Jiang, and L. F. Thompson. 2002. Adenosine kinase inhibition promotes sur- 40. Jenkinson, E. J., and J. J. T. Owen. 1990. T cell differentiation in thymus organ vival of fetal adenosine deaminase-deficient thymocytes by blocking dATP ac- cultures. Semin. Immunol. 2: 51–58. cumulation. J. Clin. Invest. 110: 395–402. 41. Thompson, L. F., and M. R. Blackburn. 1998. At last: experimental models for 14. Van De Wiele, C. J., M. L. Joachims, A. M. Fesler, J. G. Vaughn, adenosine deaminase deficiency. Immunologist 6: 72–75. M. R. Blackburn, S. T. McGee, and L. F. Thompson. 2006. Further differentiation 42. Cohen, A., J. Barankiewicz, H. M. Lederman, and E. W. Gelfand. 1983. Purine of murine double-positive thymocytes is inhibited in adenosine deaminase-defi- and pyrimidine metabolism in human T lymphocytes: regulation of deoxyribo- cient murine fetal thymic organ culture. J. Immunol. 176: 5925–5933. nucleotide metabolism. J. Biol. Chem. 258: 12334–12340. 15. Ugarkar, B. G., J. M. DaRe, J. J. Kopcho, C. E. Browne, III, J. M. Schanzer, 43. Joachims, M. L., P. Marble, C. Knott-Craig, P. Pastuszko, M. R. Blackburn, and J. B. Wiesner, and M. D. Erion. 2000. Adenosine kinase inhibitors: 1. Synthesis, L. F. Thompson. 2008. Inhibition of deoxynucleoside kinases in human thymo- enzyme inhibition, and antiseizure activity of 5-iodotubercidin analogues. J Med. cytes prevents dATP accumulation and induction of apoptosis. Nucleosides Nu- Chem. 43: 2883–2893. cleotides Nucleic Acids 27: 816–820. The Journal of Immunology 8161
44. Uckun, F. M., P. S. Gaynon, M. G. Sensel, J. Nachman, M. E. Trigg, apoptosis and causes defective T cell receptor signaling. J. Clin. Invest. 108: P. G. Steinherz, R. Hutchinson, B. C. Bostrom, H. N. Sather, and G. H. Reaman. 131–141. 1997. Clinical features and treatment outcome of childhood T-lineage acute lym- 50. Benveniste, P., W. Zhu, and A. Cohen. 1995. Interference with thymocyte dif- phoblastic leukemia according to the apparent maturational stage of T-lineage ferentiation by an inhibitor of S-adenosylhomocysteine hydrolase. J. Immunol. leukemic blasts: a Children’s Cancer Group study. J. Clin. Oncol. 15: 155: 536–544. 2214–2221. 51. Chan, B., D. Wara, J. Bastian, M. S. Hershfield, J. Bohnsack, C. G. Azen, 45. Cowan, M. J., D. W. Wara, and A. J. Ammann. 1985. Deoxycytidine therapy in R. Parkman, K. Weinberg, and D. B. Kohn. 2005. Long-term efficacy of enzyme two patients with adenosine deaminase deficiency and severe immunodeficiency replacement therapy for adenosine deaminase (ADA)-deficient severe combined disease. Clin. Immunol. Immunopathol. 37: 30–36. immunodeficiency (SCID). Clin. Immunol. 117: 133–143. 46. Fox, R. M., E. H. Tripp, and M. H. Tattersall. 1980. Mechanism of deoxycytidine rescue of thymidine toxicity in human T-leukemic lymphocytes. Cancer Res. 40: 52. Honig, M., M. H. Albert, A. Schulz, M. Sparber-Sauer, C. Schutz, 1718–1721. B. Belohradsky, T. Gungor, M. T. Rojewski, H. Bode, U. Pannicke, et al. 2007. 47. Hershfield, M. S. 2005. New insights into adenosine-receptor-mediated immu- Patients with adenosine deaminase deficiency surviving after hematopoietic stem nosuppression and the role of adenosine in causing the immunodeficiency asso- cell transplantation are at high risk of CNS complications. Blood 109: ciated with adenosine deaminase deficiency. Eur. J. Immunol. 35: 25–30. 3595–3602. 48. Minguet, S., M. Huber, L. Rosenkranz, W. W. Schamel, M. Reth, and 53. Coleman, M. S., M. J. Danton, and A. Philips. 1985. Adenosine deaminase and T. Brummer. 2005. Adenosine and cAMP are potent inhibitors of the NF-B immune dysfunction: biochemical correlates defined by molecular analysis pathway downstream of immunoreceptors. Eur. J. Immunol. 35: 31–41. throughout a disease course. Ann. NY Acad. Sci. 451: 54–65. 49. Apasov, S. G., M. R. Blackburn, R. E. Kellems, P. T. Smith, and 54. Boison, D. 2006. Adenosine kinase, epilepsy, and stroke: mechanisms and ther- M. V. Sitkovsky. 2001. Adenosine deaminase deficiency increases thymic apies. Trends Pharmacol. Sci. 27: 652–658. Downloaded from http://www.jimmunol.org/ by guest on September 26, 2021