of October 31, 2018.This information is current as

Model of DNA-PKcs DeficiencySCID in Jack Russell Terriers: A New Animal

Hirschhorn and Tom BellRochelleMichael J. Dark, Patrick J. Venta, Maryann L. Huie,

Katheryn Meek, Laura Kienker, Clarissa Dallas, Wei Wang,

http://www.jimmunol.org/content/167/4/2142doi: 10.4049/jimmunol.167.4.2142

2001; 167:2142-2150; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/167/4/2142.full#ref-list-1

, 28 of which you can access for free at: cites 51 articlesThis article

        average*  

4 weeks from acceptance to publicationFast Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

Submit online. ?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 2001 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on October 31, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on October 31, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

SCID in Jack Russell Terriers: A New Animal Model ofDNA-PKcs Deficiency1

Katheryn Meek,2* Laura Kienker,† Clarissa Dallas,* Wei Wang,* Michael J. Dark,*Patrick J. Venta,* Maryann L. Huie,‡ Rochelle Hirschhorn,‡ and Tom Bell*

We recently described the incidence of a SCID disease in a litter of Jack Russell terriers. In this study, we show that the moleculardefect in these animals is faulty V(D)J recombination. Furthermore, we document a complete deficit in DNA-dependent proteinkinase activity that can be explained by a marked diminution in the expression of the catalytic subunit DNA-dependent proteinkinase catalytic subunit (DNA-PKcs). We conclude that as is the case in C.B-17 SCID mice and in Arabian SCID foals, thedefective factor in these SCID puppies is DNA-PKcs. In mice, it has been clearly established that DNA-PKcs deficiency producesan incomplete block in V(D)J recombination, resulting in “leaky” coding joint formation and only a modest defect in signal endligation. In contrast, DNA-PKcs deficiency in horses profoundly blocks both coding and signal end joining. Here, we show thatalthough DNA-PKcs deficiency in canine lymphocytes results in a block in both coding and signal end joining, the deficit in bothis intermediate between that seen in SCID mice and SCID foals. These data demonstrate significant species variation in theabsolute necessity for DNA-PKcs during V(D)J recombination. Furthermore, the severity of the V(D)J recombination deficits inthese three examples of genetic DNA-PKcs deficiency inversely correlates with the relative DNA-PK enzymatic activity expressedin normal fibroblasts derived from these three species. The Journal of Immunology, 2001, 167: 2142–2150.

T he DNA-dependent protein kinase (DNA-PK)3 is an ex-tremely large multiprotein complex comprised of a regu-latory subunit, the heterodimeric Ku protein comprised of

70- and 86-kDa polypeptides, and the catalytic subunit DNA-de-pendent protein kinase catalytic subunit (DNA-PKcs; a 465-kDapolypeptide) (reviewed in Refs. 1–3). DNA-PK is specifically ac-tivated to phosphorylate its target proteins in the presence of DNAends, suggesting a role for this enzyme in DNA repair (4). Indeed,genetic experiments have demonstrated unequivocally thatDNA-PK functions in the repair of DNA double-strand breaks(DSBs) (reviewed in Refs. 5 and 6). In eukaryotes, two majorpathways exist to repair DSBs: homologous recombination andnonhomologous DNA end joining (NHEJ). Vertebrates rely mostheavily on the NHEJ pathway whereas yeast preferentially usehomologous recombination to repair DSBs. DNA-PK functions inthe NHEJ pathway (7). Furthermore, recent evidence indicates thatat least part of DNA-PK’s role in this pathway is dependent on itskinase activity (8, 9).

In higher eukaryotes, deficiencies in any of DNA-PK’s threecomponent polypeptides effectively disrupt the organism’s capac-ity to efficiently repair DNA DSBs, resulting in extreme sensitivityto ionizing radiation. This radiosensitivity is not in and of itselflethal. However, in normal environments, DNA-PK deficienciesare generally lethal postnatally because vertebrate immune sys-tems are dependent on the NHEJ pathway to complete V(D)J re-combination, the site-specific recombination process that providesfor assembly of unique Ag receptor genes (10–12). If V(D)J re-combination is disrupted, lymphocyte development is blocked atthe prolymphocyte stage, resulting in a profound deficiency of ma-ture B and T lymphocytes and the disease SCID.

The lymphocyte-specific endonuclease which initiates V(D)J re-combination is comprised of the recombinase- activating gene(RAG) 1 and RAG2 proteins and is targeted to immune receptorgene segments by recombination signal sequences (RSS) whichabut the coding sequences of all functional immune receptor genesegments (13, 14). During V(D)J recombination, the RAG proteinsintroduce two DSBs generating four DNA ends with structurallydistinct termini (15, 16). The coding ends have covalently sealedhairpinned ends, whereas signal ends have blunt and 5�-phosphor-ylated termini. The NHEJ pathway mediates the resolution of bothcoding and signal ends.

Two spontaneous germline mutations of DNA-PKcs have beencharacterized previously. The first was discovered in a colony ofC.B-17 mice which were shown to have markedly reduced num-bers of both B and T lymphocytes (17). Later it was determinedthat the lack of mature lymphocytes was due to a severe defect inV(D)J recombination (18). This defect in V(D)J recombinationproduced a dramatic deficit in coding joint formation, but had anegligible effect on signal end joining, and, in 1995, DNA-PKcswas implicated as the defective factor (19). In 1975, McGuire andPoppie (20) described an autosomal recessive form of combinedimmunodeficiency that occurs at high frequency in Arabian foals(20). Like SCID mice, these animals are profoundly deficient inboth B and T cells. In 1995, we characterized the molecular defect

*College of Veterinary Medicine and Department of Veterinary Pathology, MichiganState University, East Lansing, MI 48824;†Department of Internal Medicine, HaroldC. Simmons Arthritis Research Center, University of Texas Southwestern MedicalCenter, Dallas, TX 75390; and‡Department of Medicine, New York University Med-ical Center, New York, NY 10016

Received for publication December 26, 2000. Accepted for publication June 7, 2001.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby markedadvertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by Public Health Service Grants AI32600 and AI32600from the National Institutes of Health, the Harold C. Simmons Research Center, andthe American Heart Association, Texas Affiliate (to K.M.).2 Address correspondence and reprint requests to Dr. Katheryn Meek, Department ofVeterinary Pathology, Michigan State University, 350 Food Safety and Toxicology,East Lansing, MI 48824. E-mail address: kmeek@msu.edu3 Abbreviations used in this paper: DNA-PK, DNA-dependent protein kinase; DNA-PKcs, DNA-PK catalytic subunit; NHEJ, nonhomologous DNA end joining; RAG,recombinase-activating gene; RSS, recombination signal sequence; PI3, phosphati-dylinositol 3-kinase; ADA, adenosine deaminase; PNP, purine nucleoside phosphor-ylase; LMPCR, ligation-mediated PCR; SNP, single nucleotide polymorphism.

Copyright © 2001 by The American Association of Immunologists 0022-1767/01/$02.00

by guest on October 31, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

in these horses as faulty V(D)J recombination that could be ex-plained by a severe deficiency in DNA-PKcs expression (21). Incontrast to SCID mice, lymphocytes from SCID foals have dra-matically reduced levels of signal joints. We first hypothesized thatthis difference might be explained by the specific DNA-PKcs mu-tations in these two animal models of SCID. The DNA-PKcs mu-tation in SCID foals deletes 967 aa from the C terminus, includingthe entire phosphatidylinositol 3-kinase domain (22). In contrast,the mutation in SCID mice deletes only 83 aa from the C terminus,leaving the phosphatidylinositol 3-kinase domain intact and gen-erating a potentially partially active mutant enzyme (23, 24). Thus,we attributed the less severe phenotype in SCID mice to an in-complete mutation. This hypothesis was negated by targeted de-letion of DNA-PKcs (25–28). The phenotype of DNA-PKcs nullanimals is indistinguishable from C.B-17 SCID mice, althoughthese recent studies establish that signal joint formation in SCIDmice is also modestly defective—being both slightly reduced (0-to 10- fold) and less precise than in normal animals.

Another potential explanation for the phenotypic differences be-tween SCID mice and foals is that the equine mutation might gen-erate a functional protein that blocks other components of theNHEJ pathway in a dominant negative manner. We negated thishypothesis using site-directed mutagenesis of DNA-PKcs (29).Furthermore, we established that signal end joining in SCID foalsis reduced by at least 4 logs and that coding end joining is reducedby 5 logs. Thus, both signal and coding joint formation is moredramatically reduced in SCID foals than in SCID mice (wherecoding joints are reduced by 2–3 logs maximally). We hypothesizethat in normal lymphocytes, DNA-PKcs functions in both the ef-ficient joining of coding ends and the efficient and precise joiningof signal ends. In the absence of DNA-PKcs, alternative DNAend-joining pathways may facilitate inefficient and imprecise join-ing of recombination intermediates. Furthermore, this alternativepathway must not be utilized equivalently in all species.

We recently discovered another genetic form of non-B/non-TSCID in a litter of Jack Russell terriers.4 Briefly, 12 of 32 siblingsfrom a single breeding pair of Jack Russell terriers succumbed toopportunistic infections within 8–14 wk of age. To investigate thispotential genetic immunodeficiency, an additional litter waswhelped. From this litter, we determined that four of seven puppiesdisplayed a phenotype of SCID which included extreme lym-phopenia, aggamaglobulinemia, thymic dysplasia, peripheral lym-phoid aplasia, and apparent autosomal recessive inheritance.4 Inthis report, we assess cells or tissues from these animals for thefollowing activities: adenosine deaminase (ADA), purine nucleo-side phosphorylase (PNP), Ig and TCR gene rearrangement, radi-ation resistance, and DNA-PK. We have determined that immu-nodeficiency in these animals is the result of defective V(D)Jrecombination that can be explained by a deficiency in DNA-PKcs.Our data indicate that as in SCID foals, both coding and signaljoint resolution are impaired although the degree of both is inter-mediate between the murine and equine phenotypes.

Materials and MethodsDetection of ADA and PNP activity

The presence of ADA and PNP was detected by electrophoresis in starchgels and in situ staining essentially as described previously (30).

DNA isolation

DNA was prepared from thymus and bone marrow collected from 6-wk-old SCID and normal Jack Russell terrier littermates and from bone mar-row of C.B-17 SCID and BALB/c mice using commercially available DNAextraction buffer (Applied Biosystems, Foster City, CA).

Oligonucleotides

Sequences of oligonucleotides used in this study were as follows: can5�VH-2, GCAGACGCTGTGAAGGGCC; can 5�VH-1, CAGATGAACAGCCTGAGAGC; can 3�VH, GCACAGTAATACATGGCCG; can3�JH, TCTGAGGAGACGGTGACCAGGG; can 5�VB1, CCAGACATCTGTGTACTTCTG; can 3�JB1, AGGACCGTGAGCCTGGTGCC;can VB screen, ATGTTCTGGTATCGACAAGACCC; can 3�VB, CAGAAGTACACAGATGTCTG; can 5�JBds, GGGCGCCTGGTCGGGGAA; can 5�JB inner, CCCTCAGCTTGTGCCAAG; can 5�JB outer,CAGCTAGCTCTGGAGGTG; can 3�DB outer, ACCAGCTAGAGAGACACC; can 3�DB inner: TCCCAGCCACTCCCCATC; can 3�DB probe,CAGGTGGAGGGGGTCCTT; 150, GCTATGTACTACCCGGGAATTCGTG; 149, CACGAATTCCC; 150 � cac, GCTATGTACTACCCGGGAATTCGTGCAC; mur 5�JH4, ATCGATGCATAATGTCTGAGTTGCCC; mur 3�JH4, GACCTGCAGAGGCCATTCTTACCTGAGGAG;5�J558, TCCAGCACAGCCTACATGCAGCTC; mur JH screen,TCACXGTYTCYTCYKCAGGT; 5�DNA-PKcs exon 5, CTCATGGATGAATTTAAAATTGGAGA; 3�DNA-PKcs exon 6, TCACTAGGATGAACTTCACC; 5�DNA-PKcs intron 5, GACCAGATCTTTCTTTCCTG; 3�DNA-PKcs intron 5, GAAATAATTTCATGGAGTGTTTGG; andDNA-PKcs diagnostic primer, AGCAAAAAGAATTCCTCTGTAGTA;altered base is underlined.

PCR

PCR were conducted in a volume of 100 �l using amplification primersdescribed in each figure legend and the indicated amounts of DNA. For theexperiment depicted in Fig. 1, 40 cycles of amplification were performedusing the following conditions: 94°C for 1.5 min, 55°C for 2 min, and 72°Cfor 3 min. For nested PCR experiments, initial amplification conditionswere: 94°C for 1.5 min, 2°C for 2 min, and 72°C for 3 min for 40 cycles.Ten microliters of each reaction was subsequently amplified as follows:94°C for 1 min, 56°C for 1 min, and 72°C for 1 min for 40 cycles. Twentymicroliters of each PCR was analyzed by Southern blot filter hybridizationanalysis. For sequence analysis, amplified rearrangements were gel purifiedand ligated into pCR2.1 (Invitrogen, Carlsbad, CA) and transformed intocompetent Escherichia coli. Recombinant colonies were identified by hy-bridization and sequenced on an Applied Biosystems sequencer (AppliedBiosystems).

Ligation-mediated PCR (LMPCR)

LMPCR was performed similar to the method described by Roth et al. (15).Briefly, 300 pmol of annealed linker (oligonucleotides 150 and 149) wereligated to 20 �g of normal canine lung, normal canine thymus, or SCIDcanine thymus DNA. Ligated DNA was extracted with phenol-chloroformand ethanol precipitated. Subsequently, linker-ligated DNA was used inPCR using primers 150 �cac and 5�JBds for 40 cycles as follows: 94oC for1.5 min, 64oC for 45 min, and 72oC for 45 min.

Cell lines

The JRSC and JN005 cell lines were established from dermal biopsies froma SCID puppy and from an unrelated normal dog, respectively. TheMSU1.1 is a normal immortalized human fibroblast line, the generous giftof Justin McCormick (Michigan State University, East Lansing, MI). The0176 and 1821 cell lines were derived from normal and SCID foals, re-spectively, and have been described previously (21). NS47, a wild-typemouse fibroblast cell line, was generously provided by Dr. K. Ariizumi(University of Texas Southwestern Medical Center, Dallas, TX). Sf19, aSCID mouse fibroblast cell line, was generously provided by Dr. M. Bosma(Fox Chase Cancer Center, Philadelphia, PA).

Assessment of radiation sensitivity

Cells (103) were exposed to various amounts of ionizing radiation using a60Co source and immediately seeded in complete medium containing 20%FBS. After 14 days, cell colonies were fixed with 2% formaldehyde fol-lowed by 100% methanol. Subsequently, the colonies were stained withtrypan blue and colony numbers were assessed.

4 T. G. Bell, K. L. Butler, H. B. Sill, J. E. Stickle, J. A. Ramos-Vera, and M. J. Dark.Autosomal recessive severe combined immunodeficiency of Jack Russell terriers.Submitted for publication.

2143The Journal of Immunology

by guest on October 31, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

DNA-PK microfractionation and measurement of kinase activity

Whole-cell extracts were prepared by a modification of the method ofFinnie et al. (31). Briefly, 20 � 106 cells were harvested, washed threetimes in PBS, and cell pellets were frozen at �80°C. Frozen cell pelletswere resuspended in 20 �l of extraction buffer (50 mM NaF, 20 mMHEPES (pH 7.8), 450 mM NaCl, 25% (v/v) glycerol, 0.2 mM EDTA, 0.5mM DTT, 0.5 mM PMSF, 0.5 �g/ml leupeptin, 0.5 �g/ml protease inhib-itor, and 1.0 �g/ml trypsin inhibitor), subjected to three freeze/thaw cycles(liquid nitrogen/37°C), and centrifuged at 8160 � g for 7 min at 4°C.Supernatants were stored at �80°C before use and concentrations weredetermined by Bradford analysis using BSA as a standard.

The SignaTECT DNA-PK assay system (Promega, Madison, WI) wasused to assay DNA-PK activity with the following modifications. Twohundred micrograms of extract was incubated with 20 �l of preswollendsDNA-cellulose beads (Amersham Pharmacia Biotech, Piscataway, NJ)for 30 min at 4°C. The dsDNA-cellulose was then washed three times with1 ml of buffer A (25 mM HEPES (pH 7.9), 50 mM KCl, 10 mM MgCl2,10% (v/v) glycerol, 1 mM EDTA, 1 mM EGTA, and 1 mM DTT) beforeit was resuspended in 20 �l of DNA-PK reaction buffer containing 100�g/ml BSA. Kinase reactions were conducted with 10-�l aliquots of theresuspended DNA-PK-absorbed cellulose beads and were performed inboth the presence and absence of a biotinylated DNA-PK p53-derived sub-strate peptide. Terminated reactions were analyzed by spotting onto SAM2

membrane, washing, and counting in a scintillation counter as per the man-ufacturer’s instructions. All assays were performed in duplicate with atleast two different extract preparations.

Immunoblot analysis

The indicated amounts of whole-cell extracts were electrophoresed in a 5%SDS-polyacrylamide gel and transferred to polyvinylidene difluoride. AmAb (32) (18-1, generous gift from Dr. T. Carter, St. Johns University,New York, NY) was used as the primary Ab (1:300). This monoclonal candetect DNA-PKcs from many diverse mammalian species and detects anepitope within the N-terminal 250 kDa of the protein. A goat anti-mouseIgG conjugated to HRP was used as the secondary Ab. The membrane wasthen incubated with a chemiluminescent substrate (ECL; DuPont Pharmaceu-ticals, Wilmington, DE) according to the manufacturer’s recommendations.

Genetic marker in the canine DNA-PKcs gene

Primers were designed to conserved regions in exons 5 and 6 of the humanand mouse DNA-PKcs gene sequences (33, 34). A 650-bp dog DNA-PKcsfragment was amplified in a 25-�l reaction using 50 mM KCl, 10 mM Tris(pH 8.3 at 21°C), 1.5 mM MgCl2, 100 �M dNTPs, 20 �M of each primer,and 50 � 10�9 g of canine genomic DNA. Cycling conditions were 95°Cfor 1 min, 55°C for 2 min, and 72°C for 3 min for 35 cycles. The identityof the sequence was confirmed by performing a BLAST search (35). Twoadditional primers were made to determine the complete sequence of intron5 (GenBank accession number in process). A single nucleotide polymor-phism (SNP) was found in intron 5 using a pool-and-sequence approach(36). Briefly, genomic DNA samples from 10 different dog breeds werepooled, amplified, and sequenced using a Thermosequenase cycle sequenc-ing kit according to the manufacturer’s instructions (United States Bio-chemical, Cincinnati, OH). The SNP was identified as two bands (A and G)at the identical position in the sequencing ladder. An RsaI restriction sitewas created for the G allele by altering one base in a diagnostic primer.Amplification conditions were the same as above, except that cycling con-ditions were 94°C for 1 min, 54°C for 2 min, and 72°C for 3 min for 35cycles. Three microliters of 50 mM MgCl2 and 1 �l of RsaI (10 U) wereadded directly to the PCR product, followed by incubation at 37°C for 1 h.Products were analyzed by electrophoresis on a 2% agarose gel. The Aallele was recognized by the presence of a 202-bp band and the G allelewas recognized by the presence of 178- and 24-bp bands.

ResultsPurine metabolism is normal in SCID puppies

Abnormalities in purine metabolism due to defective ADA or PNPaccount for a significant percentage of the cases of SCID in hu-mans. To date, no such naturally occurring mutations have beendocumented in animals (37). We assessed both ADA and PNPactivity in normal and SCID dogs by assaying enzymatic activityin hemolysates after electrophoresis in starch gels, as describedpreviously (30). In sum, levels of PNP and ADA activity weresimilar in hemolysates from normal and SCID dogs (data not

shown), and we conclude that ADA and PNP defects cannot ex-plain the immunodeficiency observed in Jack Russell terriers.

Ig and TCR coding joints are severely diminished in SCIDpuppies

To determine whether V(D)J recombination is intact in SCID pup-pies, DNA was isolated from thymus and bone marrow samplesfrom affected and unaffected Jack Russell Terrier littermates. Us-ing PCR amplification primers derived from published caninecDNAs, both TCR and Ig coding joints were assessed. As can beseen, unrearranged VH (Fig. 1A, top panel) and VB (Fig. 1C, toppanel) gene segments are equivalently amplified from normal andSCID bone marrow and thymus DNA, respectively. Complete VH-

DHJH (Fig. 1A, bottom panel) or VBDBJB (Fig. 1C, bottom panel)coding joints were only consistently detected from DNA derivedfrom normal animals. Furthermore, coding joints in normal dogswere easily detected using as little as 5 ng of thymus or bonemarrow DNA. In contrast, only extremely low levels of VBDBJB

coding joints could be detected from SCID thymus and detectionof coding joints required 5 �g of SCID thymus DNA. VHDHJH

coding joints were not detected from SCID bone marrow DNA.This level of coding joint depression is representative of experi-ments when comparing TCR (representative of six different exper-iments) and Ig (representative of three different experiments) re-arrangements from two normal and two SCID animals (data notshown). Thus, we conclude that coding end resolution is dimin-ished by at least 4 logs in SCID dog lymphocytes as comparedwith normal canine lymphocytes implicating V(D)J recombinationas the mechanistic defect in this animal model of SCID.

In our attempt to detect coding joints from SCID dogs and pre-viously from SCID foals (29), a more aggressive PCR strategythan that used in studies of SCID mouse Ig and TCR rearrange-ments was used. Both the quantity of template DNA and the cyclenumber were increased to detect coding joints in SCID dogs.Therefore, to more directly compare the relative coding joint dim-inution in SCID dogs to that observed in SCID mice, which waspreviously reported to be depressed 100- to 1000-fold comparedwith normal animals (18, 25–27), we assessed Ig VHDHJH rear-rangements in murine bone marrow DNA using our PCR strategyand a VH region primer specific for the J558 VH gene family (Fig.1B, bottom panel). As can be seen, the sensitivity of detecting Igrearrangements from normal mouse and dog bone marrow is sim-ilar. Furthermore, Ig VHDHJH rearrangements are depressed by�100-fold in bone marrow from SCID mice, consistent with thereports from other investigators. This level of coding joint depres-sion was also observed using another VH segment primer (the X24gene segment family, data not shown). Thus, we conclude that theblock in coding joint resolution is more dramatic in SCID dogsthan in SCID mice.

V(D)J recombination is initiated normally in SCID puppies

To date, several different spontaneous germline mutations havebeen documented which result in defective V(D)J recombinationand the disease SCID. In children, mutations in either RAG1 orRAG2 have been shown to account for SCID (38). Similarly inmice, targeted mutation of either RAG1 or RAG2 results in a se-vere form of SCID (39, 40) which does not include the “leaky”phenotype seen in DNA-PKcs-deficient mice. RAG-deficient micecompletely lack either signal or coding rearrangements because, inthe absence of the RAG proteins, rearrangement is not initiated andIg and TCR genes remain in germline configuration. This has beendocumented using the technique of LMPCR to assess free signalends that are present in developing lymphocytes as recombinationintermediates (16). We reasoned that the RAG proteins could be

2144 CANINE DNA-PKcs DEFICIENCY

by guest on October 31, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

implicated (or eliminated) as the defective factor in these SCIDpuppies by assessing free signal ends by LMPCR.

Using LMPCR, free JB signal ends were assessed from normalcanine thymus DNA, canine SCID thymus DNA, and normal ca-nine lung DNA. As can be seen (Fig. 2A), free signal ends areequivalently detected in normal and SCID thymus DNA (comparelanes 3 and 4 to lanes 5 and 6). As expected, free signal ends arenot detected in nonlymphoid DNA (lanes 1 and 2). These datasuggest that TCR gene rearrangement is initiated normally inSCID dogs and that the RAG proteins are fully functional.

Signal joints are markedly reduced in SCID puppies

Using amplification primers from conserved regions of DB2 andJB2.1, the intergenic region between the two gene segments wascloned and sequenced so that DBJB signal joints could be exam-ined. Although signal joints were easily detected from normal ca-nine thymus DNA with a standard PCR assay, signal joints werenot detected from canine SCID thymus DNA (data not shown).Thus, a more sensitive nested PCR strategy was used which iscompletely analogous to the strategy used to detect signal jointsfrom SCID foals in our recent study (29). As can be seen (Fig. 2B),signal joints can be detected in as little as 0.5 ng of normal caninethymus DNA; in contrast, signal joints could only be detected in 50ng of canine SCID thymus DNA. This degree of signal joint dim-inution was consistent when comparing two SCID and two normalanimals and is representative of four independent experiments.Thus, we conclude that the frequency of DBJB signal joints in canineSCID thymus is �2 logs lower than in normal canine thymus.

As discussed above, signal joints are generally precise, lackingeither nucleotide addition or deletion. Structurally, signal joints are

head to head ligations of the two heptamer sequences which gen-erates a de novo restriction endonuclease site, ApaLI. It has beendemonstrated that the signal joints formed in SCID mice are lessprecise than in normal animals (27). The fidelity of signal ligationwas examined by subjecting the amplified signal joints to ApaLIrestriction. As can be seen (Fig. 2C), �90% of the signal jointsamplified from normal dogs are sensitive to ApaLI, whereas only10–20% of the joints amplified from SCID dogs are sensitive toApaLI. Thus, we conclude that both the rate and fidelity of signalend resolution in SCID dogs is diminished. Furthermore, the rel-ative diminution in signal end resolution in SCID dogs is interme-diate as compared with SCID mice or SCID horses such that thedegree of signal joint reduction in these three models of SCID is asfollows: equine SCID � canine SCID � murine SCID.

Hybrid joints are modestly reduced in SCID puppies

Hybrid joints are examples of aberrant end resolution and resultfrom ligation of the coding end of one gene segment to the signalend of a second gene segment (41). The frequency of hybrid re-arrangements of endogenous immune receptor gene segments hasbeen shown to be �100- to 1000-fold less than standard joiningdepending on the specific rearrangements analyzed (42, 43). It hasrecently been shown that hybrid rearrangements can be mediatedsolely by the RAG proteins in a reversal of the normal cleavagereaction (44, 45). Thus, in the absence of intact nonhomologousDNA end joining, the relative number of hybrid joints should berelatively normal, as compared with either signal or coding joints.We recently examined hybrid joining of TCR gene segments inequine SCID lymphocytes (29). We found that although certainhybrid rearrangements occurred at relatively normal levels, other

FIGURE 1. Analysis of Ig and TCR coding joints. A, PCR amplification of the germline (top panel) or rearranged (bottom panel) Ig genes from normaldog bone marrow (lanes 1–5), SCID dog bone marrow (lanes 6–10), or no template (lane 11). Amplification of the germline VH gene segment wasperformed as described in Materials and Methods using the oligonucleotides can 5�VH-2 and can 3�VH. Amplification of complete VHDHJH rearrangementswas done using the oligonucleotides can 5�VH-2 and can 3�JH. B, PCR amplification of germline (top panel) or rearranged (bottom panel) murine Ig genesfrom BALB/c bone marrow (lanes 1–5), from C.B-17 SCID bone marrow (lanes 6–10), or no template (lane 11). Amplification of the germline JH4 genesegment was done as described in Materials and Methods using the oligonucleotides mur 5�JH4 and mur 3�JH4. Amplification of complete VHDHJH

rearrangements was done using the oligonucleotides 5�J558 and mur 3�JH4. C, PCR amplification of germline (top panel) or rearranged (bottom panel)canine TCR genes from normal dog thymus (lanes 1–5), from SCID dog thymus (lanes 6–10), or no template (lane 11). Amplification of the germline VB

gene segment (as described in Materials and Methods) was performed using the oligonucleotides can 5�VB1 and can 3�VB. Amplification of completeVBDBJB rearrangements was done using the oligonucleotides can 5�VB1 and can 3�JB. A–C, The quantities of DNA(ng) used in amplifications are indicatedabove the figures.

2145The Journal of Immunology

by guest on October 31, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

hybrid joints were severely diminished, suggesting that some hy-brid rearrangements are mediated by the NHEJ pathway.

Hybrid rearrangements were also assessed in canine thymo-cytes. As can be seen (Fig. 3A), JB coding 3�DB RSS hybrid jointsare diminished by 2 logs in canine SCID thymocytes, similar to thereduction observed in signal joints. These data support our previ-ous conclusion that certain hybrid joints are mediated predomi-nately by the NHEJ pathway.

Structural characteristics of TCR hybrid joints in caninelymphocytes

It has previously been reported that hybrid and coding joints iso-lated from animals deficient in Ku or DNA-PKcs exhibit less junc-tional diversity than rearrangements isolated from normal animals(46). This is primarily due to diminished numbers and length of Nsegments (the random nucleotides added at the coding junctions bythe enzyme terminal deoxynucleotidyl transferase). We chose toassess junctional diversification of rare rearrangements present incanine lymphocytes by analyzing hybrid joints (as opposed to cod-ing joints) for two reasons. First, hybrid joints are significantlymore abundant than coding joints; furthermore, it has been shownthat hybrid joints display similar levels of N and P segment addi-tions as compared with coding joints (46), and thus should be

representative with regard to junctional modification. Therefore,hybrid joints were cloned and sequenced from thymus DNA fromboth normal and SCID dogs (Fig. 3B). As can be seen, excessiveP segments (the palindromic nucleotides at the junctions generatedby asymmetric opening of the hairpinned coding termini) derivedfrom the J� coding segment are observed in the SCID rearrange-ments. Fifty percent of the rearrangements have P elements whichrange from 1 to 9 nt. None of the hybrid joints isolated from thenormal animals have P elements. There is a slight decrease in thenumber of N segment-positive rearrangements when comparingnormal animals to SCID as well as a modest decrease in N segmentlength (2.4 vs 8.9 nt). Hybrid and coding joints from Ku-deficientmice have virtually no N segments, whereas SCID mice have onlya modest deficiency in N segment addition (46). Thus, the struc-tural characteristics of these hybrid joints are more analogous tothose observed from SCID mice than from those observed fromKu-deficient mice.

Radiosensitivity of canine SCID fibroblasts

The data presented thus far are consistent with the NHEJ pathwaybeing defective in these animals. Thus, we next assessed radio-sensitivity in fibroblasts derived from a SCID puppy and from anormal dog. As can be seen in Fig. 4A, the normal canine fibroblastcell line is more radioresistant than fibroblasts derived from theSCID puppy. These data suggest that the genetic defect in theseanimals results in impaired NHEJ.

Canine SCID fibroblasts lack DNA-PK activity because ofseverely diminished DNA-PKcs expression

Of the five known factors known to be requisite for both the V(D)Jrecombination and NHEJ, three are components of DNA-PK.Thus, to further investigate the NHEJ pathway in these animals,DNA-PK activity was assessed using whole-cell extracts preparedfrom canine SCID and normal fibroblasts. It is well appreciatedthat DNA-PK activity is abundantly expressed in human cells butmuch more minimally expressed in rodent cells. We previouslyestablished that equine cells express intermediate levels ofDNA-PK activity as compared with human and rodent cells. Tocompare the relative level of DNA-PK in normal canine cells tothe level in human, equine, and murine cells, we assessedDNA-PK activity in normal fibroblast cell lines from these species.As can be seen (Fig. 4B), in normal canine fibroblasts, DNA-PKactivity is intermediate as compared with that observed in horseand mouse fibroblasts. Finally, DNA-PK activity was undetectablein the canine, equine, or murine SCID fibroblasts, implicating ei-ther DNA-PKcs or Ku as the defective factor in SCID dogs.

Next, we performed immunoblot assays to assess levels ofDNA-PKcs, Ku70, and Ku86 in whole-cell extracts from canineSCID fibroblasts compared with normal canine cells. Whereas Kulevels did not differ in SCID cells compared with the normal cells(data not shown), canine SCID fibroblasts were found to be de-fective in DNA-PKcs expression (Fig. 4C). The 465-kDa DNA-PKcs polypeptide is easily detected in the normal canine cells, butnot in the canine SCID fibroblasts, and we conclude that DNA-PKcs expression is markedly diminished in canine SCID cells.

Identification of a SNP in the DNA-PKcs gene and segregationof one allele with the mutant phenotype

To obtain genetic evidence that DNA-PKcs is in fact the defectivefactor in these animals, we attempted to identify a polymorphismwithin the canine DNA-PKcs gene to determine whether within alineage a single allele segregated with the mutant phenotype. A

FIGURE 2. Analysis of JB signal ends and TCR DB2JB2.1 signal joints.A, Assessment of free signal ends as described in Materials and Methodswas done via LMPCR. Amplification products using the indicated amountsof linker-ligated genomic DNA (�g) derived from normal canine lungDNA (lanes 1 and 2), normal canine thymus DNA (lanes 3 and 4), orcanine SCID thymus DNA (lanes 5 and 6). B, Nested PCR amplification ofDBJB signal joints from normal canine thymus (lanes 1–6) or SCID dogthymus (lanes 7–12). Quantities of DNA (ng) used in amplifications areindicated above the figure. Amplification of D BJB signal joints was done asdescribed in Materials and Methods using the oligonucleotides can 5�JBouter and can 3�DB outer, followed by can 5�JB inner and can 3�DB inner.C, Products from amplifications using 5000 ng of thymus DNA from twonormal dogs (lanes 1–4) and two SCID dogs (lanes 5–8) were subjectedto ApaLI digestion (lanes 2, 4, 6, and 8) and analyzed by Southern blotfilter hybridization using the can 3�DB probe oligonucleotide as a hybrid-ization probe.

2146 CANINE DNA-PKcs DEFICIENCY

by guest on October 31, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

polymorphic transition (A/G) was identified at position 95 of in-tron 5 by using a pool-and-sequence method. A diagnostic PCR-based test was then developed to distinguish the two alleles. WhenDNA samples were analyzed from members of the pedigree inwhich the SCID gene was segregating, it was found that segrega-tion of the G allele was consistent with the SCID gene being DNA-PKcs, under the assumption that the G allele is linked with themutation in the gene (Fig. 4D). The three affected animals arehomozygous for the G allele. The three unaffected siblings areeither homozygous for the A allele or heterozygotes (Fig. 4D). Wehave recently tested an additional five animals from a subsequentlitter, and homozygosity for the G allele segregates exclusivelywith the SCID phenotype. Thus, in this limited pedigree analysis,the mutant phenotype segregates with homozygosity of one DNA-PKcs allele. These data are consistent with (although certainly notindicative of) DNA-PKcs being the defective factor in these animals.

DiscussionIn this study, we have characterized the third example of a geneticDNA-PKcs deficiency. Like mice and horses, DNA-PKcs defi-

ciency in dogs results in SCID because of defective V(D)J recom-bination. Similarly, cells from SCID dogs are hypersensitive toionizing radiation. The data presented here demonstrate significantdifferences between species for the absolute requirement for DNA-PKcs during V(D)J recombination. It is well established that lackof DNA-PKcs in mice results in a “leaky” block in V(D)J recom-bination such that coding joint resolution is depressed by as muchas 100- to 1000-fold, but with considerably less effect on certain“privileged” immune receptor loci, which are apparently unaf-fected and generate normal levels of coding joints (25–28). Signaljoint ligation is even less affected, being virtually normal in somecells and depressed by only as much as 10-fold in some experi-mental systems. In contrast, DNA-PKcs deficiency in horses re-sults in a much more substantial block of V(D)J recombination,resulting in a 5-log depression in coding joints and a 4-log depres-sion in signal joints (21, 29, 47). The data presented in this reportdemonstrate that DNA-PKcs deficiency in dogs results in V(D)Jrecombination defects which are intermediate between those ob-served in horses and mice. Thus, we conclude that the absoluterequirement for DNA-PKcs in V(D)J recombination varies in these

FIGURE 3. Analysis of TCR� hybrid rearrangements. A, Nested PCR amplification of JB coding to 3�DB RSS hybrid joints from normal canine thymus(lanes 1–4) or SCID dog thymus (lanes 5–8). Quantities of DNA (ng) used in amplification are indicated above the figure. Amplification of hybrid jointswas done as described in Materials and Methods using the oligonucleotides can 3�JB outer and can 3�DB outer and then can 3�JB inner and can 3�DB inner.B, Nucleotide sequences of DBJB hybrid joints isolated from normal and SCID canine thymus are aligned to germline DB and JB gene segments. Nucleotidespotentially derived from P and N addition are depicted.

2147The Journal of Immunology

by guest on October 31, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

three species. This conclusion presents several unanswered ques-tions regarding the biology of DNA-PK.

First, are all cell types affected by species-dependent differencesin the absolute requirement for DNA-PKcs? Alternatively, are justlymphocytes from dogs and horses more dependent on DNA-PKcsthan murine lymphocytes? In some situations, DNA-PKcs hasbeen implicated as being crucial in preventing apoptosis (48) al-though in others it has been suggested to promote apoptosis (49).Also, it has been demonstrated previously that unrepaired recom-bination intermediates in the thymus of SCID mice results in p53activation (23) which may facilitate apoptosis. Our data could beexplained by an increased susceptibility to apoptosis of equine andcanine prolymphocytes as compared with mouse prolymphocytes.If canine and equine SCID prolymphocytes undergo rapid apopto-sis in response to unrepaired recombination intermediates, any po-tential leaky rearrangements might not be detected. Thus, the morestringent DNA-PKcs requirement we observe might not actuallyreflect a more stringent requirement for end joining per se, butinstead reflect a lack of rescue from apoptosis or perhaps morerapid apoptosis in prolymphocytes. However, in our previous

study, a severe deficit in signal end resolution was demonstrated inSCID horse fibroblasts using a transient recombination assay (21);this assay should be insensitive to apoptotic effects. Using thesame assay in our laboratory, most SCID mouse fibroblasts supportwild-type levels of signal end joining (29). This suggests that themore stringent requirement for DNA-PKcs in some species is im-portant for DNA end joining, as opposed to rescue from apoptosis,and that this more stringent requirement may be consistent in allcell types. Finally, it has recently been demonstrated that the NHEJpathway is important for the maintenance of genomic stability (50,51). Whereas loss of either XRCC4 or DNA ligase IV in mouseembryonic fibroblasts results in a significant increase in variouschromosomal aberrations, loss of DNA-PKcs has a clear althoughless dramatic effect (50). This difference in chromosomal stabilityin mice with mutations in distinct components of the NHEJ path-way is reminiscent of differences in the rate and fidelity of signalend joining in the same animals (52). It follows that DNA-PKcsdeficiency in dogs and horses may result in more significant chro-mosomal instability than in mice. This question is currently underexamination.

FIGURE 4. Assessment of NHEJ and DNA-PK. A, Radiation resistance canine SCID JRSC fibroblasts and a normal canine fibroblast cell line, JN005.Data are presented as percent survival of unirradiated controls and represents an average of triplicate samples. B, Whole-cell extracts (100 �g) preparedfrom normal (nontransformed) human fibroblast cell line (MSU1–1), normal equine fibroblast cell line (0176), equine SCID fibroblasts (1821), normalcanine fibroblasts (JN005), canine SCID fibroblasts (JRSC), normal mouse fibroblasts (NS47), and murine SCID fibroblasts (SF19) were assayed forDNA-PK activity as described in Materials and Methods. Each cell extract was tested in duplicate, and at least two independent extracts were tested foreach cell line. Bars, SD. C, Immunoblot analysis of whole-cell extracts from the canine SCID fibroblast cell line JRSC (lane 1; 200 �g) or the canine tumorcell line MDCK (lanes 2–5; 100, 50, 25, and 12.5 �g, respectively). Position of the 200-kDa molecular mass marker is indicated. D, An intronicpolymorphism was defined within the canine DNA-PKcs gene as described in Materials and Methods. Amplification across this polymorphism results inproducts which can be differentiated by RsaI digestion, resulting in DNA fragments of 202 bp (A allele, uncut) or 178 (G allele, cut). The pedigree of testedanimals is shown above. As can be seen, all three SCID animals tested are homozygous for the G allele; the three normal littermates tested were eitherheterozygous or homozygous for the A allele.

2148 CANINE DNA-PKcs DEFICIENCY

by guest on October 31, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

A second question is why are there species differences in abso-lute DNA-PKcs requirements? The severity of the V(D)J recom-bination defects in these three examples of genetic DNA-PKcsdeficiency (this report; Refs, 18, 25–27, and 29) appears to in-versely correlate with two factors: life span and normal DNA-PKenzymatic activity. The correlation between severity of V(D)J re-combination defects in each of the DNA-PKcs deficiencies and therelative DNA-PK activity expressed in normal cells from the samespecies is illustrated in Fig. 5. As can be seen, there is a highlysignificant correlation between relative DNA-PK activity and thelog depression in either signal or coding joint formation. Fig. 5also illustrates that DNA-PKcs deficiency in each of these speciesaffects signal end joining more severely than coding end joining.

It is well appreciated that rodent cells express 50- to 100-foldless DNA-PK activity than human cells (31). This difference inDNA-PK activity can largely be explained by a similar differencein DNA-PKcs and Ku protein levels in human vs rodent cells.Similarly, using mAbs which recognize DNA-PKcs in many spe-cies, we have documented that the differences in DNA-PK activitybetween human cells and either canine or equine fibroblasts arelargely the result of protein expression levels (data not shown).Thus, to date, there is no evidence that the specific activity ofhuman DNA-PK differs from that of DNA-PK from other species.

Here, we show that horses (life span �28 years) and dogs (lifespan �14 years) express intermediate levels of DNA-PK activityas compared with rodents and humans. Thus, these data extend thisgeneral correlation between relative DNA-PK activity and naturallife span. This correlation might predict that loss of DNA-PKcsexpression in humans would result in the most severe phenotype.It is interesting that three spontaneous DNA-PKcs germline mu-tations have now been documented in animals whereas no germ-line mutations of DNA-PKcs have been implicated in the numer-ous incidences of genetic combined immunodeficiency disease in

humans. This raises the question of whether a germline DNA-PKcsdeficiency in humans might be lethal.

It seems reasonable that animals that express very high levels ofan enzymatic activity might have a more absolute requirement forthat activity in vivo as compared with animals which express min-imal levels of the activity. Thus, horse cells that express relativelyhigh levels of DNA-PK have a more stringent requirement forDNA-PKcs than mouse cells. Perhaps expression of high DNA-PKlevels reflects an evolutionary preference for NHEJ in some spe-cies over other end joining pathways. The question remains: arealternative end joining pathways less active in species which ex-press abundant levels of DNA-PK? If so, perhaps the low expres-sion of this alternative pathway explains the relative lack of “leak-iness” in the V(D)J recombination deficits in SCID dogs and SCIDhorses.

In summary, here we describe the third example of geneticDNA-PKcs deficiency. As is the case with SCID mice and SCIDfoals, DNA-PKcs deficiency in dogs results in a block in V(D)Jrecombination and extreme radiosensitivity. The severity of theV(D)J recombination defects in these three examples of SCID areas follows: equine SCID � canine SCID � murine SCID. Finally,the severity of the V(D)J recombination deficits in these three spe-cies inversely correlates with differences in DNA-PK levels ex-pressed normally.

References1. Jackson, S. P. 1997. DNA-dependent protein kinase. Int. J. Biochem. Cell Biol.

29:935.2. Anderson, C. W., and S. P. Lees-Miller. 1992. The nuclear serine/threonine pro-

tein kinase DNA-PK. Crit. Rev. Eukaryotic Gene Expression 2:283.3. Anderson, C. W., and T. H. Carter. 1996. The DNA-activated protein kinase—

DNA-PK. Curr. Top. Microbiol. Immunol. 217:91.4. Gottlieb, T. M., and S. P. Jackson. 1993. The DNA-dependent protein kinase:

requirement for DNA ends and association with Ku antigen. Cell 72:131.5. Jeggo, P. A. 1998. DNA breakage and repair. Adv. Genet. 38:185.6. Jeggo, P. A., G. E. Taccioli, and S. P. Jackson. 1995. Menage a trois: double

strand break repair, V(D)J recombination and DNA-PK. BioEssays 17:949.7. Kanaar, R., J. H. Hoeijmakers, and D. C. van Gent. 1998. Molecular mechanisms

of DNA double strand break repair. Trends. Cell Biol. 8:483.8. Kurimasa, A., S. Kumano, N. V. Boubnov, M. D. Story, C. S. Tung,

S. R. Peterson, and D. J. Chen. 1999. Requirement for the kinase activity ofhuman DNA-dependent protein kinase catalytic subunit in DNA strand breakrejoining. Mol. Cell. Biol. 19:3877.

9. Kienker, L. J., E. K. Shin, and K. Meek. 2000. Both V(D)J recombination andradioresistance require DNA-PK kinase activity, though minimal levels sufficefor V(D)J recombination. Nucleic Acids Res. 28:2752.

10. Taccioli, G. E., H. L. Cheng, A. J. Varghese, G. Whitmore, and F. W. Alt. 1994.A DNA repair defect in Chinese hamster ovary cells affects V(D)J recombinationsimilarly to the murine scid mutation. J. Biol. Chem. 269:7439.

11. Pergola, F., M. Z. Zdzienicka, and M. R. Lieber. 1993. V(D)J recombination inmammalian cell mutants defective in DNA double-strand break repair. Mol. Cell.Biol. 13:3464.

12. Lewis, S. M. 1994. The mechanism of V(D)J joining: lessons from molecular,immunological, and comparative analyses. Adv. Immunol. 56:27.

13. Schatz, D. G., M. A. Oettinger, and D. Baltimore. 1989. The V(D)J recombina-tion activating gene, RAG-1. Cell 59:1035.

14. Oettinger, M. A., D. G. Schatz, C. Gorka, and D. Baltimore. 1990. RAG-1 andRAG-2, adjacent genes that synergistically activate V(D)J recombination. Sci-ence 248:1517.

15. Roth, D. B., C. Zhu, and M. Gellert. 1993. Characterization of broken DNAmolecules associated with V(D)J recombination. Proc. Natl. Acad. Sci. USA 90:10788.

16. Schlissel, M., A. Constantinescu, T. Morrow, M. Baxter, and A. Peng. 1993.Double-strand signal sequence breaks in V(D)J recombination are blunt, 5�-phos-phorylated, RAG-dependent, and cell cycle regulated. Genes Dev. 7:2520.

17. Bosma, G. C., R. P. Custer, and M. J. Bosma. 1983. A severe combined immu-nodeficiency mutation in the mouse. Nature 301:527.

18. Lieber, M. R., J. E. Hesse, S. Lewis, G. C. Bosma, N. Rosenberg, K. Mizuuchi,M. J. Bosma, and M. Gellert. 1988. The defect in murine severe combined im-mune deficiency: joining of signal sequences but not coding segments in V(D)Jrecombination. Cell 55:7.

19. Blunt, T., N. J. Finnie, G. E. Taccioli, G. C. Smith, J. Demengeot, T. M. Gottlieb,R. Mizuta, A. J. Varghese, F. W. Alt, P. A. Jeggo, et al. 1995. Defective DNA-dependent protein kinase activity is linked to V(D)J recombination and DNArepair defects associated with the murine scid mutation. Cell 80:813.

20. McGuire, T. C., and M. J. Poppie. 1973. Hypogammaglobulinemia and thymichypoplasia in horses: a primary combined immunodeficiency disorder. Infect.Immun. 8:272.

FIGURE 5. Correlation between DNA-PK activity and severity ofV(D)J recombination defects. Correlation between relative deficit (y-axis)in signal joints (�) or coding joints (f) and the relative DNA-PK activityexpressed in fibroblasts from normal individuals (x-axis). Relative codingand signal joint formation for SCID mice are taken from Refs. 18 and25–28 and for SCID horses from Ref. 29.

2149The Journal of Immunology

by guest on October 31, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

21. Wiler, R., R. Leber, B. B. Moore, L. F. VanDyk, L. E. Perryman, and K. Meek.1995. Equine severe combined immunodeficiency: a defect in V(D)J recombi-nation and DNA-dependent protein kinase activity. Proc. Natl. Acad. Sci. USA92:11485.

22. Shin, E. K., L. E. Perryman, and K. Meek. 1997. A kinase-negative mutation ofDNA-PK(CS) in equine SCID results in defective coding and signal joint for-mation. J. Immunol. 158:3565.

23. Danska, J. S., D. P. Holland, S. Mariathasan, K. M. Williams, and C. J. Guidos.1996. Biochemical and genetic defects in the DNA-dependent protein kinase inmurine scid lymphocytes. Mol. Cell. Biol. 16:5507.

24. Blunt, T., D. Gell, M. Fox, G. E. Taccioli, A. R. Lehmann, S. P. Jackson, andP. A. Jeggo. 1996. Identification of a nonsense mutation in the carboxyl-terminalregion of DNA-dependent protein kinase catalytic subunit in the scid mouse.Proc. Natl. Acad. Sci. USA 93:10285.

25. Gao, Y. J., J. Chaudhuri, C. M. Zhu, L. Davidson, D. T. Weaver, and F. W. Alt.1998. A targeted DNA-PKCS-null mutation reveals DNA-PK-independent func-tions for KU in V(D)J recombination. Immunity 9:367.

26. Taccioli, G. E., A. G. Amatucci, H. J. Beamish, D. Gell, X. H. Xiang,M. I. T. Arzayus, A. Priestley, S. P. Jackson, A. M. Rothstein, P. A. Jeggo, andV. L. M. Herrera. 1998. Targeted disruption of the catalytic subunit of theDNA-PK gene in mice confers severe combined immunodeficiency and radio-sensitivity. Immunity 9:355.

27. Bogue, M. A., C. Jhappan, and D. B. Roth. 1998. Analysis of variable (diversity)joining recombination in DNA-dependent protein kinase (DNA-PK)-deficientmice reveals DNA-PK-independent pathways for both signal and coding jointformation. Proc. Natl. Acad. Sci. USA 95:15559.

28. Kurimasa, A., H. Ouyang, L. J. Dong, S. Wang, X. L. Li, C. Cordon-Cardo,D. J. Chen, and G. C. Li. 1999. Catalytic subunit of DNA-dependent proteinkinase: Impact on lymphocyte development and tumorigenesis. Proc. Natl. Acad.Sci. USA 96:1403.

29. Shin, E. K., T. Rijkers, A. Pastink, and K. Meek. 2000. Analyses of TCR�rearrangements substantiate a profound deficit in recombination signal sequencejoining in SCID foals: implications for the role of DNA-dependent protein kinasein V(D)J recombination. J. Immunol. 164:1416.

30. Spencer, W., D. A. Hopkinson, and H. Harris. 1968. Adenosine deaminase poly-morphism in man. Ann. Hum. Genet. 32:9.

31. Finnie, N. J., T. M. Gottlieb, T. Blunt, P. A. Jeggo, and S. P. Jackson. 1995.DNA-dependent protein kinase activity is absent in xrs-6 cells: implications forsite-specific recombination and DNA double-strand break repair. Proc. Natl.Acad. Sci. USA 92:320.

32. Carter, T., I. Vancurova, I. Sun, W. Lou, S. DeLeon, S. 1990. A DNA-activatedprotein kinase from HeLa cell nuclei. Mol. Cell. Biol. 10:6460.

33. Sipley, J. D., J. C. Menninger, K. O. Hartley, D. C. Ward, S. P. Jackson, andC. W. Anderson. 1995. Gene for the catalytic subunit of the human DNA-acti-vated protein kinase maps to the site of the XRCC7 gene on chromosome 8. Proc.Natl. Acad. Sci. USA 92:7515.

34. Fukumura, R., R. Araki, A. Fujimori, M. Mori, T. Saito, F. Watanabe, M. Sarashi,H. Itsukaichi, K. Eguchi-Kasai, K. Sato, et al. 1998. Murine cell line SX9 bearinga mutation in the dna-pkcs gene exhibits aberrant V(D)J recombination not onlyin the coding joint but also in the signal joint. J. Biol. Chem. 273:13058.

35. Altschul, S. F., T. L. Madden, A. A. Schaffer, J. Zhang, Z. Ahang, W. Miller, andD. J. Lipman. 1997. Gapped BLAST and PSI-BLAST: a new generation of pro-tein database search programs. Nucleic Acids Res. 25:3389.

36. Broullette, J. A., J. R. Andrew, and P. J. Venta. 2000. Estimate of nucleotidediversity in dogs using a pool-and-sequence method. Mamm. Genome. 11:1079.

37. Hirschhorn, R. 1999. Immunodeficiency disease due to deficiency of adenosinedeaminase. In Primary Immunodeficiency Diseases. H. D. Ochs, C. I. E. Smith,and J. M. Puck, eds. Oxford Univ. Press, Oxford pp. 121–139.

38. Schwarz, K., G. H. Gauss, L. Ludwig, U. Pannicke, Z. Li, D. Lindner,W. Friedrich, R. A. Seger, T. E. Hansen-Hagge, S. Desiderio, et al. 1996. RAGmutations in human B cell-negative SCID. Science 274:97.

39. Mombaerts, P., J. Iacomini, R. S. Johnson, K. Herrup, S. Tonegawa, andV. E. Papaioannou. 1992. RAG-1-deficient mice have no mature B and T lym-phocytes. Cell 68:869.

40. Shinkai, Y., G. Rathbun, K. P. Lam, E. M. Oltz, V. Stewart, M. Mendelsohn,J. Charron, M. Datta, F. Young, A. M. Stall, et al. 1992. RAG-2-deficient micelack mature lymphocytes owing to inability to initiate V(D)J rearrangement. Cell68:855.

41. Lewis, S. M., J. E. Hesse, K. Mizuuchi, and M. Gellert. 1988. Novel strandexchanges in V(D)J recombination. Cell 55:1099.

42. VanDyk, L. F., T. W. Wise, B. B. Moore, and K. Meek. 1996. ImmunoglobulinD(H) recombination signal sequence targeting: effect of D(H) coding and flank-ing regions and recombination partner. J. Immunol. 157:4005.

43. Sollbach, A. E., and G. E. Wu. 1995. Inversions produced during V(D)J rear-rangement at IgH, the immunoglobulin heavy-chain locus. Mol. Cell. Biol. 15:671.

44. Han, J. O., S. B. Steen, and D. B. Roth. 1997. Ku86 is not required for protectionof signal ends or for formation of nonstandard V(D)J recombination products.Mol. Cell. Biol. 17:2226.

45. Melek, M., M. Gellert, and D. C. van Gent. 1998. Rejoining of DNA by theRAG1 and RAG2 proteins. Science 280:301.

46. Bogue, M. A., C. Wang, C. Zhu, and D. B. Roth. 1997. V(D)J recombination inKu86-deficient mice: distinct effects on coding, signal, and hybrid joint forma-tion. Immunity 7:37.

47. Leber, R., R. Wiler, L. E. Perryman, and K. Meek. 1998. Equine SCID: mech-anistic analysis and comparison with murine SCID. Vet. Immunol. Immunopathol.65:1.

48. Nueda, A., F. Hudson, N. F. Mivechi, and W. S. Dynan. 1999. DNA-dependentprotein kinase protects against heat-induced apoptosis. J. Biol. Chem. 274:14988.

49. Wang, S., M. Guo, H. Ouyang, X. Li, C. Cordon-Cardo, A. Kurimasa, D. J. Chen,Z. Fuks, C. C. Ling, and G. C. Li. 2000. The catalytic subunit of DNA-dependentprotein kinase selectively regulates p53-dependent apoptosis but not cell-cyclearrest. Proc. Natl. Acad. Sci. USA 97:1584.

50. Ferguson, D., J. Sekiguchi, S. Chang, K. Frank, Y. Gao, R. A. DePinho, andF. W. Alt. 2000. The nonhomologous end-joining pathway of DNA repair isrequired for genomic stability and the suppression of translocations. Proc. Natl.Acad. Sci. USA 97:6630.

51. Bailey, S. M., J. Meyne, D. J. Chen, A. Kurimasa, G. C. Li, B. E. Lehnert, andE. H. Goodwin. 1999. DNA double-strand break repair proteins are required tocap the ends of mammalian chromosomes. Proc. Natl. Acad. Sci. USA 96:14899.

52. Gu, Y., J. Sekiguchi, Y. Gao, P. Dikkes, K. Frank, D. Ferguson, P. Hasty,J. Chun, and F. W. Alt. 2000. Defective embryonic neurogenesis in Ku-deficientbut not DNA-dependent protein kinase catalytic subunit-deficient mice. Proc.Natl. Acad. Sci. USA 97:2668.

2150 CANINE DNA-PKcs DEFICIENCY

by guest on October 31, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from