Targeted Disruption of the Artemis Murine Counterpart Results in SCID and Defective V(D)J Recombination That Is Partially Corrected with Bon
http://www.100md.com
免疫学杂志 2005年第4期
Abstract
Artemis is a mammalian protein, the absence of which results in SCID in Athabascan-speaking Native Americans (SCIDA). This novel protein has been implicated in DNA double-strand break repair and V(D)J recombination. We have cloned the Artemis murine counterpart, mArt, and generated a mouse with a targeted disruption of mArt. Artemis-deficient mice show a similar T–B– NK+ immunodeficiency phenotype, and carry a profound impairment in coding joint rearrangement, while retaining intact signal ends and close to normal signal joint formation. mArt–/– embryonic fibroblasts show increased sensitivity to ionizing radiation. Hemopoietic stem cell (HSC) transplantation using 500-5000 enriched congenic, but not allogeneic mismatched HSC corrected the T cell and partially corrected the B cell defect. Large numbers (40,000) of allogeneic mismatched HSC or pretreatment with 300 cGy of radiation overcame graft resistance, resulting in limited B cell engraftment. Our results suggest that the V(D)J and DNA repair defects seen in this mArt–/– mouse model are comparable to those in humans with Artemis deficiency, and that the recovery of immunity following HSC transplantation favors T rather than B cell reconstitution, consistent with what is seen in children with this form of SCID.
Introduction
Maturation of T and B lymphocytes requires V(D)J recombination in which the exons that encode for V regions of Ig and TCR are assembled from their germline components, V, D, and J gene segments (1). The significance of this process has been well documented by analyses of animal models carrying germline mutations/deletions in genes encoding essential proteins, and also by studies of some rare types of human T–B–NK+ SCID caused by defects in genes essential for this process (2). It is well established that the RAG1 and RAG2 (RAG1/2) complex initiates V(D)J recombination in lymphoid cells by recognizing and cleaving at the conserved recombination signal sequences that flank the germline V, (D), and J coding elements. Each cleavage results in two blunt signal ends (SE),4 and two covalently sealed (hairpinned) coding ends (CE). The resolution of SEs and CEs leads to the direct fusion of the two blunt SEs mostly in a head-to-head fashion, and the joining of two CEs between the V, (D), and J elements to be assembled. The CEs require further processing before the joining, by which hairpins are opened, and nucleotides added and/or deleted at the junctions of rearranged products. The presence of palindromic (P) nucleotides is sometimes observed at the junctions, which corresponds to the addition of several nucleotides that are complementary to the nucleotides at the end, and is thought to reflect the position in which hairpins are opened (3).
The rejoining is conducted by the general nonhomologous end joining machinery, which is the major repair pathway for DNA double-strand breaks in mammalian cells. The well-characterized factors in this pathway include DNA-dependent protein kinase (DNA-PKcs), Ku heterodimer complex (Ku70/86), XRCC4, DNA ligase IV, and the more recently identified Artemis. DNA-PKcs is a protein kinase, and its kinase activity is essential for the pathway (4). Although it is able to phosphorylate a number of proteins in vitro, the only known substrate that has been demonstrated to be significant for this pathway to date is Artemis (5). Among the other factors, Ku70/86 is believed to function in binding the broken ends and recruiting/activating other components (6, 7, 8), and DNA ligase IV couples with XRCC4 to accomplish the final ligation of the broken DNA ends (9).
Artemis is a mammalian protein, in which mutations were identified first in children with radiation-sensitive SCID and then in Athabascan-speaking children with SCID (SCIDA) (10, 11). Radiation-sensitive SCID represents a small group of patients of various ethnic origins, who show increased radiosensitivity in their fibroblast cells, and impaired coding joint (CJ) formation in an extrachromosomal V(D)J recombination assay (10). SCIDA has a high incidence in the Navajo and Apache populations. SCIDA patients have a T–B–NK+ immunophenotype and also show a profound impairment in V(D)J CJ formation (11, 12, 13). Interestingly, children with SCIDA (13) and other forms of T–B–NK+ SCID (14) do not engraft as efficiently as children with T–B+NK– SCID and often require a chemotherapy preparative regimen. This is thought to be due at least in part to the normal NK cell function in these patients.
In an in vitro study, Artemis was found to be an effective substrate of DNA-PKcs (5). It interacted with DNA-PKcs to form a stable complex and acquired endonuclease activity, enabling it to open RAG-derived hairpin structures, and process 3' ends of sequences. In a recently reported 129/Sv mouse model of Artemis deficiency, the T–B–NK+ phenotype was confirmed, as was the increased sensitivity to ionizing radiation in embryonic fibroblasts (15, 16). In addition, accumulation of hairpins supports the role of Artemis at hairpin processing in CJ formation.
To evaluate the response to hemopoietic stem cell transplantation (HSCT) in a mouse model of SCIDA, we constructed a mouse carrying a germline deletion of the Artemis murine counterpart, mArt. We confirm that Artemis-deficient mice closely resemble DNA-PKcs-deficient mice in that both present with severely impaired CJ formation and near normal signal joint (SJ) formation. We also detected unusual long P nucleotide additions in the rare CJ from Artemis-deficient mice, further supporting the potential role of Artemis in hairpin processing. HSCT resulted in discordant reconstitution of the immune system with a preference toward the T cell compartment, consistent with what has been seen in children with SCIDA.
Materials and Methods
Cloning of mArt in mice and targeted disruption of the mArt gene
The Artemis murine counterpart, mArt, was identified by a database search, RT-PCR, RACE experiments, and sequence assembly. It was then further consolidated by radiation hybrid (RH) mapping using the mouse T31 RH panel and the T31 mouse RH database (The Jackson Laboratory), and genomic organization determination by a series of intron PCR and restriction mapping.
The targeting construct was generated by cloning a 3.2-kb mArt DNA fragment (exons 4–6) and a 9.1-kb fragment (exons 11–14) into pPgk.tk.neo, resulting in the two mArt genomic fragments flanking the neomycin resistance gene, which is driven by the phosphoglycerate kinase promoter, for positive selection. The construct also contains an HSV-thymidine kinase gene, also driven by the phosphoglycerate kinase promoter, to select against nonhomologous integrations of the construct in the genome (negative selection). The construct (25 μg) was linearized at the unique NotI site and used for electroporation to transfect 129/Sv embryonic stem (ES) cells, and the targeted ES cells were enriched by positive selection with G418 (300 μg/ml) and negative selection with 0.2 μM 1-(2'-deoxy-2'-fluoro--D-arabinofuranosyl)-5-ioduracil. After 10 days of selection, visible ES colonies were picked and expanded, genomic DNA was prepared, and standard Southern blot analysis was performed to identify the targeted ES cell clones. Two external probes were used (Fig. 1A): the 5' one detects a 15-kb BamHI fragment for the wild-type (wt) and a 5-kb fragment for the targeted allele, while the 3' one detects a 12-kb EcoRI fragment from the wt allele and a 9-kb fragment from the targeted allele. We also used a neomycin internal probe recognizing an EcoRI fragment (6 kb) from the targeted ES allele.
FIGURE 1. Inactivation of mArt by homologous recombination. A, Diagrammatic representation of the mArt locus (top), the targeting construct (middle), and the targeted allele (bottom) with hybridization probes indicated by small blocks. EcoRI (E) and BamHI (B) restriction sites are indicated, together with the sizes of products. B, Southern blot analysis of EcoRI-digested ES cell DNA showing positive ES cell clones (lanes 2 and 5) with 3' probe and Neo probe. C, DNA from progenies of heterozygote crossing was analyzed by PCR, amplifying a 190-bp product from wt, and a 390-bp product from the targeted allele. D, Western blot analysis showing the absence of Artemis expression in Artemis-deficient thymus.
Generation of Artemis-deficient mice
The positive ES clones with normal karyotype were used for injection into C57BL/6 (B6) blastocysts, which were transferred to CD1 pseudopregnant females. The progenies were monitored by coat color to obtain chimeric males, who were bred with B6 females to generate heterozygotes that carry the targeted allele transmitted through the germline. A PCR assay was used to determine the knockout mice, in which three primers were included in the PCR: E6F, lying 5' to the targeted deletion, i6R from the region that is deleted in the targeted allele, and NeoR; thus, E6F and i6R amplify a 190-bp product from the wt allele, and E6F and NeoR amplify a region (390 bp) from the targeted allele. Finally, the selected male heterozygotes were crossed with wt C57BL/6 females for two generations (N2) and heterozygotes mated to generate F1 homozygous (mArt–/–) progenies.
All of the mice involved in this study were handled under a protocol approved by the University of California Committee on Animal Research following National Institutes of Health approved guidelines. The animals were bred and monitored in the University of California Barrier Facility in autoclaved cages. DNA-PKcs mutant (scid) breeding pairs (CB17-Prkdcscid/SzJ; The Jackson Laboratory) were bred under the same conditions.
Western blot analyses
Whole cell lysates were prepared from lymphoid tissues, resolved in a 7% SDS-PAGE gel, transferred to nitrocellulose membranes, and hybridized with a chicken IgY Ab (1:1000; Genway Biotech) raised against the C-terminal of the human Artemis gene. Anti-actin Ab (1:5000; Sigma-Aldrich) was used as an internal control.
Flow cytometry and lymphocyte function assays
Single cell suspensions were prepared from the thymus, spleen, and bone marrow (BM), and cells were stained with mAbs for flow cytometric analysis on a FACScan (BD Biosciences), as described (17). The conjugated mAbs (BD Pharmingen) were used in combination, as indicated in the figure legends. In addition, we evaluated the lymphocyte proliferative response to Con A (5 μg/ml), solid-phase anti-CD3 (2 μg/ml), and LPS (5 μg/ml), as well as NK cytotoxicity, as described (17).
Histology evaluation
Tissue sections of various organs from 4- to 5-wk-old mice were examined. Briefly, lymphoid and other organs were fixed in Bouin’s solution and embedded in paraffin. Tissue sections were stained with H&E, examined, and photographed in a Zeiss photomicroscope.
PCR analysis of V(D)J recombination products and intermediates
DNA was prepared from lymphoid tissues of 2- to 4-wk-old animals for a PCR-based evaluation (7, 8). For Ig loci, DHL and JH2 were used to evaluate D-JH rearrangements (7), and VH7183y and JH2y to evaluate VH-(D)JH rearrangements. VH7183y (5'-CGATTCAYCATYTCYAGAGABAAT-3') recognizes all 19 VH7183 family members. A JH2-specific probe from the 5' JH2 coding sequence was used to probe and evaluate the amplified rearrangement products. Evaluation for IgL chain was performed with J, V, and V deg (a degenerate primer that amplifies most VK segments) as described (18).
For TCR rearrangement, V8 (8.1, 8.2, and 8.3) to J2.6 CJ formation was examined, as described (8), as well as D2 to J1 (7) and V8 to J50 rearrangements (19). To analyze SJ formation, we used DR21 and DR161 for circular PCR to amplify the SJ formation for the D2 to J1 rearrangement, and ApaLI digestion was used to assess the fidelity of the SJ formation (4, 7). Furthermore, ligation-mediated PCR (LM-PCR) was used to evaluate V(D)J recombination intermediate SE at the 5' end of the D2 locus (7). In addition, the recovered CJs were cloned into a pGEM-T vector (Promega), selected by colony PCR, sequenced, and analyzed.
Extrachromosomal V(D)J recombination and ionizing radiation (IR) sensitivity assays
We used a modification of the extrachromosomal assay for human cells that we have previously described in detail (11). Briefly, mouse embryonic fibroblasts were established from 14-day F1 fetuses of N2 generation Art+/– pregnancies. The cell lines were phenotyped, and the third passage was used for the assays. The murine cells were transiently transfected with 4 μg of RAG1/2, and 2 μg of pJH200 (ampr, SJ) or 2 μg of pJH290 (ampr, CJ) by electroporation in Opti-MEM at 975 μF and 300 V. A full-length human Artemis primary transcript expression construct, pCMV-ART (2 μg), was added to the transfection to assess complementation. pJH200 and pJH290 carry a cam gene that is interrupted from its promoter by a transcriptional terminator flanked by recombination signal sequences. Upon transfection, RAG proteins induce V(D)J recombination in the extrachromosomal plasmid substrates, resulting in the excision of the transcriptional terminator and the activation of chloramphenicol resistance. After 48 h of incubation in DMEM at 37°C, 5% CO2, the pJH200 or pJH290 plasmid was recovered using the alkaline lysis method and digested with DpnI to select for those that had been replicated in the Sv40-transformed murine embryonic fibroblasts (MEFs). The selected plasmids were transformed into DH10B by electroporation to assay for ampicillin and chloramphenicol resistance. The percentage of successful recombination is represented by the ratio of colonies grown on ampicillin/chloramphenicol (from recombined substrate only) vs ampicillin plates (11).
For IR sensitivity assays, varying numbers (2.5 x 102 to 2 x 103) of MEFs were plated in triplicate, and incubated for 5–10 h at 37°C, 5% CO2. Cells were then exposed to the indicated dose of x-rays using a Pantak x-ray generator operating at 320 kV/10 mA with 0.5 mm copper filtration, and returned to the incubator for 7 days. Colonies arising from surviving cells were stained with crystal violet and scored. Percentage of survival was calculated as the number of colonies on treated dishes over those on untreated dishes. The data presented represent the average of three independent experiments.
HSC transplants
Young adult B6 wt congenic or BALB/c allogeneic mismatched mice (The Jackson Laboratory) were used as donors. BM cells were freshly harvested, and several donor cell preparations were used, including whole BM (WBM), T cell-depleted BM (TCDBM), lineage-depleted (lin–) BM, and BM that was lin– and positively selected for Sca-1 (17). TCDBM was prepared using an anti-CD3 mAb and immunomagnetic beads, as previously described (17). The purification of Sca-1-positive cells was done in two steps: first, negative selection of lineage-committed cells using the Hemopoietic Stem Cell Enrichment Kit by StemCell Technologies; second, positive selection using anti-Sca-1 mAb directly labeled to paramagnetic particles (Miltenyi Biotec). The process resulted in a highly pure lin– Sca-1+ HSC population (>95%).
The recipient mice were 7–10 wk of age at the time of transplant, and either 1 x 106 WBM, TCDBM or lin– BM, or 500–40,000 lin– Sca-1+ HSC were injected i.v. Most of the animals received no prior conditioning therapy, but one group was pretreated with 300 cGy of total body irradiation (TBI) using a cesium137 animal irradiator (20, 21). Controls included age-matched mArt–/–, wt, or heterozygotes that received a sham injection, or allogeneic HSCT, or TBI and allogeneic HSCT. T and B cell reconstitution was evaluated by flow cytometry using a mAb specific for BALB/c allogeneic (anti-H2Kb) or congenic (Ly-5.1) donor cells, and V(D)J recombination in the thymus and BM.
Results
Targeted disruption of the mArt gene
The mArt gene that we obtained matches Artemis with an overall homology of 81% (AF387731), with an open reading frame of 2118 bp. RH mapping placed this gene at a region on chromosome 2 that is syntenic to the human Artemis region on chromosome 10p, and furthermore, it has a structure similar to the human Artemis gene, and both contain 14 similarly sized exons and introns (data not shown).
The targeting vector was designed as shown in Fig. 1A. Upon transfection, homologous recombination in the genome resulted in the replacement of the 8-kb mArt fragment (exons 6–11), with the 1.8-kb neomycin gene, leading to the removal of 80% of the mArt gene. We obtained 15 ES clones that were successfully targeted (Fig. 1B). Three of them were used for blastocyst injections, from which we obtained two healthy chimeric males. Both showed successful germline transmission of the targeted ES cells and produced heterozygous mice, which appeared similar to their wt littermates and were crossed to generate 125 animals for this study, 26 of which (21%) were homozygous for the deleted allele (Fig. 1C and data not shown).
The disruption of Artemis function was demonstrated in the Western blot analysis of Artemis expression in lymphoid tissues (thymus, spleen, and lymph nodes). A unique protein band migrating at 105 kDa, a size similar to that observed for human Artemis, was detected in wt and heterozygous littermates, but not in mArt knockout mice (Fig. 1D and data not shown). The data suggest that the targeted deletion is a null allele (mArt–/–).
Early arrest of B and T cell differentiation in mArt–/– mice
We examined five mArt–/– mice at 4–5 wk of age, and four at 2 wk of age, together with 10 wt and heterozygote littermates and 3 scid mice. Gross and histologic examination of various organs revealed abnormalities only in the lymphoid tissue of mArt–/– mice. mArt–/– thymus, spleen, and lymph node were smaller and poorly developed, and the most profound effect on development was seen in the mArt–/– thymus, which weighed 20% of the thymi from wt and heterozygote controls. Thymocyte counts were 0.7–4% of that observed in wt littermates (Table I).
Table I. Lymphoid cellularity of mArt–/– micea
The developmental defects of lymphoid tissues were further demonstrated in microscopic examination. In contrast with wt littermates (Fig. 2A), the thymus of mArt–/– mice showed absence of the lymphocytic cortex, and predominance of connective tissue with scattered lymphoid cells (Fig. 2B). These scattered lymphoid cells showed remarkably abundant mitotic figures (Fig. 2B, high power). Compared with littermate controls (Fig. 2C), the spleens of mArt–/– mice showed notable reduction in fully developed lymphoid follicles (Fig. 2D), in which the majority of lymphoid cells showed larger nuclei with less dense chromatin and prominent nucleoli (Fig. 2D, high power). In contrast to the lymphoid depletion, the interfollicular hemopoietic red pulp was very conspicuous in mArt–/– spleens (Fig. 2D). The mArt–/– lymph nodes were much smaller than those of their littermates (Fig. 2E), and were severely depleted of mature lymphocytes, with complete absence of lymphoid follicles (Fig. 2F).
FIGURE 2. Histologic (10x) evaluation of lymphoid tissues. A, Thymic tissue from wt, and B, from mArt–/– mice; inset shows x40 view. C, Splenic tissue from wt, and D, from mArt–/– animal; inset shows x40 view. E, Lymph node of wt, and F, of mArt–/– animal.
Analyses of B cell maturation in the 2- to 5-wk-old mArt–/– mice revealed a developmental arrest at an early progenitor stage. As presented in Fig. 3A from a representative flow cytometry analysis, the majority of the developing B cells (B220+) were arrested at the progenitor (B220+CD43+) stage, in sharp contrast to wt littermates in which most of the developing B cells have progressed to more mature stages (B220+CD43–). The arrest was also evident in the mArt–/– spleen, where only a very small number of developing B cells (1.5% of splenocytes analyzed) had progressed to immature and mature B cells (B220+IgM+), in contrast to the 27% observed in wt controls (Fig. 3A). Peripheral blood cell counts and B lymphocyte numbers were also similarly reduced (data not shown). Heterozygote littermates showed similar results in all cases to that seen in wt mice (data not shown).
FIGURE 3. Flow cytometric analysis showing early arrest of lymphocyte development in Artemis-deficient mice. A, Analysis of BM and splenocytes from 4-wk-old animals. Numbers represent the percentage of total cells within indicated region. B, Analysis of thymocytes and splenocytes from 2-wk-old animals.
Analysis of mArt–/– thymocytes revealed a similar early arrest of T cell development, which was reflected by >90% of the thymocytes being arrested at the CD4–CD8– stage, compared with 5–10% observed in wt and heterozygous littermates (Fig. 3B). CD25+ analysis of gated CD4–CD8– thymocytes showed that there were at least 20 times more CD4–CD8–CD25+ cells in mArt–/– mice than in wt and heterozygous controls (data not shown). The early arrest was also reflected by the near absence (<0.1%) of CD8+CD3+ and CD4+CD3+ mature single-positive cells in 2-wk-old mArt–/– spleens (data not shown). Analysis of 4- to 5-wk-old animals revealed a small increase in single-positive splenocytes in mArt–/– mice (Fig. 3B). Approximately 2% of the mArt–/– splenocytes analyzed were CD4+CD3+ and 1% CD8+CD3+, but these numbers were still much lower compared with the 10 and 6% observed in wt controls (Fig. 3B and data not shown).
T cell proliferation to Con A and anti-CD3 in mArt–/– mice was reduced to 16.5 and 18.7% of that seen in wt controls, and B cell proliferation to LPS was 13.5% of that in wt. NK cell number and cytotoxicity were normal in mArt–/– mice (data not shown).
Impaired Ig and TCR rearrangement in mArt–/– mice
In PCR-based analyses, DH-JH2 and VH7183-JH2 rearrangements were abundant in wt and mArt+/– controls, but rare in mArt–/– mice at a level comparable to that seen in age-matched scid mice. Both mArt–/– and scid mice consistently showed 10-fold decrease in DH-JH2, and 100-fold decrease in VH7183-JH2 rearrangements (Fig. 4A). TCR CJ formation was reduced 10100-fold in mArt–/– mice (Fig. 4B), and a similar result was seen for D2-J1 CJ rearrangement. The IgL() and TCR rearrangements were undetectable in mArt–/– mice (data not shown).
FIGURE 4. PCR-based analysis showing profound reduction of CJ formation, and near normal SJ formation in 2- to 5-wk-old Artemis-deficient mice. A, D-JH2 and VH7183-DJH2 CJ formation at IgH loci in BM. B, V8-DJ2.6 and D2-J1 CJ formation at TCR and loci. C, Detection of circular excised D2-J1 SJ formation; two-thirds of the PCR products were digested with ApaLI. D, LM-PCR detection of SE at the 5' end of the D2, and the integrity examination by ApaLI digestion.
The amount of D2 to J1 SJ amplification from mArt–/– mice was similar to that detected in wt and heterozygous controls from independent experiments, and ApaLI digestion of SJ joints in mArt–/– mice also appeared similar to scid mice and wt controls (Fig. 4C). These results suggest near normal SJ formation in mArt–/– mice. LM-PCR evaluation showed that the SEs at 5' of D2 locus are abundant in mArt–/– mice at a level similar to their littermate controls, and furthermore, LM-PCR products from mArt–/– and wt mice were both almost completely digested by ApaLI, suggesting that the SEs are intact in mArt–/– mice (Fig. 4D).
Aberrant rare V(D)J CJ rearrangement events in mArt–/– mice
We analyzed a total of 7 DH-JH2, 12 VH7183-D-JH2, 22 D1-J2.6, and 34 V8-D-J2.6 CJs from mArt–/– mice, together with those from wt controls at these loci (10, 14, 16, and 26, respectively). We observed standard coding junctions for most of the rearrangements at these loci; however, we detected 8 unusual long P elements in 29 mArt–/– D-J CJs analyzed, but none in the 26 D-J CJs from wt mice. We also found relatively larger deletions in mArt–/– mice, i.e., 6 deletions (11 nt) in 75 mArt–/– CJs, but only 1 in the 66 wt CJs analyzed, and 1 of the VH7183-D-JH2 CJs showed a 28-nt deletion in JH2 (Fig. 5 and data not shown).
FIGURE 5. Sequence analysis of D1-J2.6 CJ recovered from wt, mArt–/– mice, and also from post-BMT mArt–/– mice. N and P are for nucleotides not present in the germline sequences.
In vitro V(D)J recombination corrected by transduction with human rART
We used an extrachromosomal V(D)J recombination assay to evaluate V(D)J recombination in MEFs from N2 mice (Table II). In four separate mArt–/– cell lines in two experiments, CJ formation was undetectable in 2 and <1/50 of that seen in mArt+/+ cells. SJ formation was comparable to that seen in wt cells. Transduction of the human Artemis primary transcript expression construct restored CJ formation in mArt–/– MEFs.
Table II. Extrachromosomal V(D)J recombination in mArt–/– and mArt+/+ MEFs
Radiation sensitivity in mArt–/– MEF cells
We used colony formation assays to evaluate the relative radiation sensitivity of mArt–/– and mArt+/+ MEFs (Fig. 6). Like human fibroblasts defective in Artemis, we found that MEFs from mArt–/– mice exhibited elevated sensitivity to IR relative to mArt+/+ MEFs (22, 23).
FIGURE 6. Radiation sensitivity of mArt–/– MEF cells. Percentage of survival of SV40-transformed mArt–/– and mArt+/+ MEF cells is plotted as a function of x-ray dose (rad). These data represent the average of three independent experiments.
T and B cell reconstitution following HSCT
Young adult B6 wt, congenic, or BALB/c allogeneic mismatched mice were used as donors. BM was freshly harvested, and several donor cell preparations were used, including WBM, TCDBM, lineage-depleted (lin–) BM, and BM that was lin– and positively selected for Sca-1. The recipient mice were 4–12 wk of age at the time of transplant. A subgroup in each experiment was pretreated with 300 cGy of TBI using a cesium137 animal irradiator. Control animals were mArt–/– littermates that received a sham injection or mArt+/+ age-matched mice that were injected with similar marrow preparations. Animals were followed until they were 3–4 mo post-HSCT when they were euthanized and analyzed for engraftment and immune reconstitution. All control mice died before the first bleed at 4 wk, while engrafted mArt–/– recipients of congenic or allogeneic BM survived to at least 14 wk post-BM transplantation (BMT); survival correlated with T cell engraftment (data not shown).
To evaluate the effects of cell dose, we transplanted animals with 500 or 5000 congenic or allogeneic lin– Sca-1+ HSC. With either 500 or 5000 HSC, the congenic recipients appeared to have some engraftment with limited immune reconstitution (Table III). With 5000 HSC, the T cell engraftment in the congenic recipients was significantly greater, but with limited evidence for B cell recovery. In the allorecipients, neither 500 nor 5000 HSC resulted in significant engraftment. With low dose TBI, there was significant engraftment (congenic > allogeneic) with some B cell reconstitution in the congenic recipients, but not in the allogeneic recipients.
Table III. Engraftment in blood at 8 wk post-HSCT in mArt–/– mice: 500 vs 5000 cells
We then evaluated the effects of using a larger cell dose of HSC as well as different preparations of BM. Because the results were comparable in recipients of 1 x 106 TCDBM and lin– BM, and 40,000 lin– Sca-1+ BM, the groups were combined and labeled Sca-1 (Table IV). In the two recipients of 1 x 106 congenic WBM cells, some T and B cell reconstitution could be seen at 6 wk. With 40,000 Sca-1+ cells, T cell reconstitution was comparable (14.7% at 6 wk). With 300 cGy of TBI before transplant, there was a significant increase in T cell reconstitution (22.9% at 4 wk and 69.2% at 6 wk). In the allogeneic mismatched recipients of 40,000 Sca-1 HSC (Table IV), there was some B cell engraftment at 6 wk (4.7 ± 1.1%), possibly due to the large cell dose. In the Sca-1/TBI group, donor T cells (66.9 ± 12.6%) and B cells (12.1 ± 4.2%) could also be detected in the blood. The percentage of host T and B cells in this group of irradiated animals was 12.1 ± 4.9% and 1.2 ± 0.8%, respectively, possibly a result of the TBI exposure (24).
Table IV. Immune reconstitution in mArt–/– mice following congenic and allogeneic HSCT with 40,000 bone marrow cells
Fig. 7A shows an analysis at 12 wk post-BMT from an animal that received 40,000 lin– Sca-1+ allogeneic HSC (without TBI), in which the reconstitution of T cell immunity in the thymus and spleen was virtually normal, and the splenocyte proliferative responses to Con A and PHA were restored to normal levels (data not shown). Although there is some evidence for B cell maturation in the BM, i.e., 5.4% B220+ CD43– cells vs 0% in a mArt–/– littermate control and 19.3% in a wt littermate (data not shown), there was very little evidence for B cell maturation in the spleen (0.7% B220+IgM+ in mArt–/– transplant recipient vs 0.1% in mArt–/– untransplanted vs 26.1% in wt). Also, the proliferative response to LPS remained lower in this animal (2199 ± 246 cpm), compared with the wt control (8436 ± 316 cpm), and was similar to the untreated mArt–/– control (1580 ± 40 cpm).
FIGURE 7. Effects of HSCT in mArt–/– mice. A, FACScan analysis of B and T cells in spleen, BM, and thymus in a recipient of lin– Sca-1+ allogeneic BM at 12 wk posttransplant without TBI; B, Histology (x10) of thymus, spleen, and lymph node at 12 wk posttransplant (without TBI); and C, PCR analysis of V(D)J rearrangement for VH7183-DJH2 CJ formation in BM, and V8-DJ2.6 in thymus.
Reconstitution of T cell immunity in mArt–/– mice was also evident in the histology post-BMT (Fig. 7B). Transplanted mice showed normal sized thymi with a remarkable cortex rich in lymphocytes, and scant thymic medulla, very similar to mArt+/+ controls. The spleens of transplanted mArt–/– mice also showed more white pulp than untransplanted mArt–/– mice, but notably less than mArt+/+ controls. Occasional lymphoid follicles were observed in transplanted mArt–/– spleens, but none of them showed pale-staining germinal centers. Similarly, although the lymph nodes of post-BMT mArt–/– mice had abundant lymphoid cells, well-developed lymphoid follicles with germinal centers were absent.
T and B cell reconstitution when pretransplant irradiation was given was reflected in V(D)J recombination activity (Fig. 7C). However, while TCR -chain rearrangement (V8 to DJ2.6) was apparent at a dramatically increased level in the post-BMT thymus, IgH chain rearrangement in BM (VH7183 to DJH2) was only slightly increased and much lower compared with the wt control. Finally, 21 rare CJ of TCR D to J rearrangements that were analyzed all showed normal junction sequences without unusual P element addition (Fig. 5).
Discussion
Artemis is essential for lymphocyte maturation
Our findings in N2 mice on the C57BL/6 background carrying a targeted disruption of Artemis are for the most part comparable to those recently reported in 129/Sv mice (15, 16). The loss-of-Artemis function results in the arrest of T and B progenitor cells in lymphoid tissues and the dramatic decrease of mature T and B cells in spleen and peripheral blood. The immunodeficient phenotype of Artemis-deficient mice is similar to that seen in SCIDA patients, although the arrest may be more complete in humans in that we found some residual in vitro T cell proliferation in response to mitogens (<20% of controls), which is usually <10% of normal in SCIDA patients (13). In an evaluation of SCIDA BM samples, we also found the early arrest of B cell development and a near absence of more mature B cells (B220+IgM+) (data not shown). Our results provide direct evidence that a defect in Artemis is the pathologic mechanism for SCIDA.
Artemis is essential for CJ, but not SJ formation in vivo
mArt–/– MEFs were deficient in CJ, but not SJ formation in the in vitro assay. This could be complemented with the transduction of a human Artemis primary transcript expression construct. Associated with the early arrest of T and B cell differentiation, Artemis-deficient mice showed severely impaired CJ formation in both T and B cell lineages, which provides direct in vivo evidence for the involvement of Artemis in V(D)J recombination. Rare CJs were detected at IgH and TCR loci in Artemis-deficient mice at a significantly decreased level that is comparable to that seen in scid control mice. A leaky V(D)J recombination activity associated with an incomplete SCID phenotype has been observed previously in DNA-PKcs-mutant/deficient and Ku86-deficient mice, but not in animals carrying deletions for RAG1/2, XRCC4, or ligase IV. It seems likely that when RAGs function normally in cutting V, (D), and J segments, these segments can then be assembled at low efficiency into V(D)J joints by either residual nonhomologous end joining activity or some other repair pathway. However, RAG1/2–/– animals are unable to initiate V(D)J recombination, while XRCC4–/– or ligase IV–/– mice are unable to form the joints, in all cases resulting in a total blockage of V(D)J rearrangement and a complete SCID phenotype. In contrast, mArt–/– SE and SJ formation does not appear to be affected to a level that is evident in the LM-PCR- and PCR-based analysis. These results are consistent with our previous results in SCIDA patients, and support the preferential impact of Artemis on the formation of CJ.
Analysis of the rare mArt–/– CJ sequences reveals an overall normal N nucleotide addition, but unusual P nucleotide additions and large deletions that we did not see in wt and heterozygous controls. Longer P additions and larger deletions have been frequently observed in scid mice (25, 26). Because the P elements are thought to reflect the position in which hairpinned CEs are opened, and the deletion modification at the junction is believed to result from nuclease activity, the unusual longer P additions and larger deletions observed in DNA-PKcs-deficient and mArt–/– CJs support the functional involvement of DNA-PKcs and Artemis in hairpin opening and CE processing. These in vivo results are consistent with the in vitro data that suggest the association of these two proteins and their involvement in hairpin opening/CE processing (5).
HSCT at least partially corrects the mArt–/– defect
The results of HSCT in the mArt–/– mice are for the most part comparable to what is seen in children with this form of T–B–NK+ SCID. Specifically, even with congenic wt BM, as in children with SCIDA receiving HLA-matched sibling transplants (13), there is a preferential reconstitution of T cell immunity, which can be partially overcome with sublethal TBI, similar to what we have seen in children with SCIDA (13). However, murine congenic donors are more like syngeneic and less like HLA-matched siblings, which probably explains the degree of multilineage engraftment seen in the murine model (Table III). There was also a similar resistance to allogeneic mismatched HSCT at the lower cell doses (500–5000 HSC) in the Artemis-deficient recipients that we and others have seen in children with T–B–NK+ SCID who receive haplomismatched donor grafts (13). At higher cell doses, we found that this resistance could be overcome, something that has not been tested in humans. A haplomismatched donor-recipient pair might have resulted in a different outcome; however, the mutation currently only exists on either the B6 or 129 backgrounds. We plan to cross the mArt–/– mutation onto the BALB/c background in the future. Despite the T cell reconstitution post-HSCT, we found only limited B cell recovery, a problem that is also seen in human SCIDA (13). To demonstrate that immune reconstitution could be enhanced in mArt–/– recipients by providing a growth advantage in the marrow for donor cells, we treated them with low dose TBI using a dose of radiation that has been used previously in scid and DNA-PKcs-deficient mouse models that are also radiation sensitive (20, 21). We saw no acute adverse effects of this therapy, although we did find that mArt–/– MEFs had increased sensitivity to ionizing radiation, comparable to what was reported in the other Artemis-deficient murine model (15). We did not follow these animals out sufficiently long to detect long-term effects of the radiation that are likely to lead to increased tumor formation and possibly growth delays due to their DNA repair defect.
In summary, our results demonstrate that V(D)J CJ formation is impaired after the RAG-mediated cleavage in mArt–/– mice, resulting in defects in hairpin opening and coding end processing, and severely impaired CJ formation. These defects result in a severe maturation arrest of T and B lymphocyte and a T–B–NK+ phenotype, which can be partially corrected by HSCT.
Acknowledgments
We thank Peter Goebel for many helpful discussions, and Brian Wardwell, Peter Skewes-Cox, and Elisa Ng for their valuable technical support.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by grants from the National Institutes of Health (AI 28339), the March of Dimes (6-FY00-301), the Children’s Health Initiative, Lucille Packard Foundation, and the Nigel Gough Memorial Fund.
2 L.L. and E.S. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Morton J. Cowan, Department of Pediatrics, University of California, 505 Parnassus Avenue, San Francisco, CA 94143-1278. E-mail address: mcowan@peds.ucsf.edu
4 Abbreviations used in this paper: SE, signal end; BM, bone marrow; BMT, BM transplantation; CE, coding end; CJ, coding joint; DNA-PKcs, DNA-dependent protein kinase; ES, embryonic stem; HSC, hemopoietic stem cell; HSCT, HSC transplantation; IR, ionizing radiation; LM-PCR, ligation-mediated PCR; MEF, Sv40-transformed murine embryonic fibroblast; P, palindromic; RH, radiation hybrid; SCIDA, Athabascan-speaking children with SCID; SJ, signal joint; TBI, total body irradiation; TCDBM, T cell-depleted BM; WBM, whole BM; wt, wild type.
Received for publication March 3, 2004. Accepted for publication December 2, 2004.
References
Grawunder, U., R. B. West, M. R. Lieber. 1998. Antigen receptor gene rearrangement. Curr. Opin. Immunol. 10:172.
Bassing, C. H., W. Swat, F. W. . 2002. Alt: the mechanism and regulation of chromosomal V(D)J recombination. Cell 109:S45.
Lewis, S. M.. 1994. The mechanism of V(D)J joining: lessons from molecular, immunological and comparative analyses. Adv. Immunol. 56:27.
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, V. L. M. Herrera. 1998. Targeted disruption of the catalytic subunit of the DNA-PK gene in mice confers severe combined immunodeficiency and radiosensitivity. Immunity 9:355.
Ma, Y., U. Pannicke, K. Schwarz, M. R. Lieber. 2002. Hairpin opening and overhang processing by an Artemis/DNA-dependent protein kinase complex in nonhomologous end joining and V(D)J recombination. Cell 108:781.
Featherstone, C., S. P. Jackson. 1999. Ku, A DNA repair protein with multiple cellular functions. Mutat. Res. 434:3.
Zhu, C., M. A. Bogue, D.-S. Lim, P. Hasty, D. B. Roth. 1996. Ku86-deficient mice exhibit severe combined immunodeficiency and defective processing of V(D)J recombination intermediates. Cell 86:379.
Ouyang, H., A. Nussenzweig, A. Kurimasa, V. da Costa Soares, X. Li, C. Cordon-Cardo, W. H. Li, N. Cheong, M. Nussenzweig, G. Iliakis, et al 1997. Ku70 is required for DNA repair but not for T cell antigen receptor recombination in vivo. J. Exp. Med. 186:921.
Grawunder, U., D. Zimmer, P. Kulesza, M. R. Lieber. 1998. Requirement for an interaction of XRCC4 with DNA ligase IV for wild-type V(D)J recombination and DNA double-strand break repair in vitro. J. Biol. Chem. 273:24708.
Moushous, D., I. Callebaut, R. de Chasseval, B. Corneo, M. Cavazzana-Calvo, F. le Deist, I. Tezczn, O. Sanal, Y. Bertrand, N. Philippe, et al 2001. Artemis, a novel DNA double-strand break repair/V(D)J recombination protein, is mutated in human severe combined immune deficiency. Cell 105:177.
Li, L., D. Moshous, Y. Zhou, J. Wang, G. Xie, E. Salido, D. Hu, J. P. de Villartay, M. J. Cowan. 2002. A founder mutation in Artemis causes severe combined immunodeficiency disease in Athabascan-speaking Native Americans. J. Immunol. 168:6323.
Moshous, D., L. Li, R. de Chasseval, N. Philippe, N. Jabado, M. J. Cowan, A. Fisher, J. P. de Villartay. 2000. A new gene involved in DNA double-strand break repair and V(D)J recombination is located on human chromosome 10p. Hum. Mol. Genet. 9:583.
O’Marcaigh, A., K. DeSantes, D. Hu, H. Pabst, B. Horn, L. Li, M. J. Cowan. 2001. Bone marrow transplantation for T–B– severe combined immunodeficiency disease in Athabascan-speaking Native Americans. Bone Marrow Transplant. 27:703.
Bertrand, Y., P. Landais, W. Friedrich, B. Gerritsen, G. Morgan, A. Fasth, M. Cavazzana-Calvo, F. Porta, A. Cant, T. Espanol, et al 1999. Influence of severe combined immunodeficiency phenotype on the outcome of HLA non-identical, T-cell-depleted bone marrow transplantation. J. Pediatr. 134:740.
Rooney, S., J. Sekeguchi, C. Zhu, H. L. Cheng, J. P. Manis, S. Whitlow, J. DeVido, D. Foy, J. Chaudhuri, D. Lombard, et al 2002. Leaky scid phenotype associated with defective V(D)J coding end processing in Artemis-deficient mice. Mol. Cell 10:1379.
Rooney, S., F. W. Alt, D. Lombard, S. Whitlow, M. Eckersdorff, J. Fleming, S. Fugmann, D. O. Ferguson, D. G. Schatz, J. Sekeguchi. 2003. Defective DNA repair and increased genomic instability in Artemis-deficient murine cells. J. Exp. Med. 197:553.
Bhattacharyya, S., A. Chawla, K. Smith, Y. Zhou, S. Talib, M. J. Cowan. 2002. Increased engraftment with minimal graft-vs-host-disease following in utero transplantation of S-59 psoralen/UVA light-treated sensitized T cells and adult T cell depleted bone marrow in fetal mice. J. Immunol. 169:6133.
Schlissel, M. S., D. Baltimore. 1991. Activation of immunoglobulin gene rearrangement correlates with induction of germline gene transcription. Cell 58:1001.
Zhu, C., M. A. Bogure, D. B. Roth. 1996. Thymocyte differentiation in -irradiated severe-combined immunodeficient mice: characterization of intermediates and products of V(D)J recombination at the T cell receptor A locus. Eur. J. Immunol. 26:2859.
Lapidot, T., O. Kollet. 2002. The essential roles of the chemokine SDF-1 and its receptor CXCR4 in human stem cell homing and repopulation of transplanted immune-deficient NOD/SCID and NOD/SCID/B2m(null) mice. Leukemia 16:1992
Greiner, D. L., R. A. Hesselton, L. D. Shultz. 1998. SCID mouse models of human stem cell engraftment. Stem Cells 16:166.
Nicolas, N., D. Moshous, M. Cavazzana-Calvo, D. Papadopoulo, R. de Chasseval, F. Le Deist, A. Fischer, J. P. de Villartay. 1998. A human severe combined immunodeficiency (SCID) condition with increased sensitivity to ionizing radiations and impaired V(D)J rearrangements defines a new DNA recombination/repair deficiency. J. Exp. Med. 188:627.
Dudasova, Z., M. Chovanec. 2003. Artemis, a novel guardian of the genome. Neoplasma 50:311.
Williams, C. J., I. Grandal, D. J. Vesprini, U. Wojtyra, J. Dansda, C. J. Guidos. 2001. Irradiation promotes V(D)J joining and RAG-dependent neoplastic transformation in SCID T-cell precursors. Mol. Cell. Biol. 21:400.
Fish, S. M., M. J. Bosma. 1994. Abnormal deletions in the T-cell receptor locus of mouse thymocytes. Mol. Cell. Biol. 14:4455.
Shuler, W., N. R. Ruetsch, M. Amsler, M. J. Bosma. 1991. Coding joint formation of endogenous T cell receptor genes in lymphoid cells from scid mice: unusual P-nucleotide additions in VJ-coding joints. Eur. J. Immunol. 21:589.(Lanying Li, Eduardo Salid)
Artemis is a mammalian protein, the absence of which results in SCID in Athabascan-speaking Native Americans (SCIDA). This novel protein has been implicated in DNA double-strand break repair and V(D)J recombination. We have cloned the Artemis murine counterpart, mArt, and generated a mouse with a targeted disruption of mArt. Artemis-deficient mice show a similar T–B– NK+ immunodeficiency phenotype, and carry a profound impairment in coding joint rearrangement, while retaining intact signal ends and close to normal signal joint formation. mArt–/– embryonic fibroblasts show increased sensitivity to ionizing radiation. Hemopoietic stem cell (HSC) transplantation using 500-5000 enriched congenic, but not allogeneic mismatched HSC corrected the T cell and partially corrected the B cell defect. Large numbers (40,000) of allogeneic mismatched HSC or pretreatment with 300 cGy of radiation overcame graft resistance, resulting in limited B cell engraftment. Our results suggest that the V(D)J and DNA repair defects seen in this mArt–/– mouse model are comparable to those in humans with Artemis deficiency, and that the recovery of immunity following HSC transplantation favors T rather than B cell reconstitution, consistent with what is seen in children with this form of SCID.
Introduction
Maturation of T and B lymphocytes requires V(D)J recombination in which the exons that encode for V regions of Ig and TCR are assembled from their germline components, V, D, and J gene segments (1). The significance of this process has been well documented by analyses of animal models carrying germline mutations/deletions in genes encoding essential proteins, and also by studies of some rare types of human T–B–NK+ SCID caused by defects in genes essential for this process (2). It is well established that the RAG1 and RAG2 (RAG1/2) complex initiates V(D)J recombination in lymphoid cells by recognizing and cleaving at the conserved recombination signal sequences that flank the germline V, (D), and J coding elements. Each cleavage results in two blunt signal ends (SE),4 and two covalently sealed (hairpinned) coding ends (CE). The resolution of SEs and CEs leads to the direct fusion of the two blunt SEs mostly in a head-to-head fashion, and the joining of two CEs between the V, (D), and J elements to be assembled. The CEs require further processing before the joining, by which hairpins are opened, and nucleotides added and/or deleted at the junctions of rearranged products. The presence of palindromic (P) nucleotides is sometimes observed at the junctions, which corresponds to the addition of several nucleotides that are complementary to the nucleotides at the end, and is thought to reflect the position in which hairpins are opened (3).
The rejoining is conducted by the general nonhomologous end joining machinery, which is the major repair pathway for DNA double-strand breaks in mammalian cells. The well-characterized factors in this pathway include DNA-dependent protein kinase (DNA-PKcs), Ku heterodimer complex (Ku70/86), XRCC4, DNA ligase IV, and the more recently identified Artemis. DNA-PKcs is a protein kinase, and its kinase activity is essential for the pathway (4). Although it is able to phosphorylate a number of proteins in vitro, the only known substrate that has been demonstrated to be significant for this pathway to date is Artemis (5). Among the other factors, Ku70/86 is believed to function in binding the broken ends and recruiting/activating other components (6, 7, 8), and DNA ligase IV couples with XRCC4 to accomplish the final ligation of the broken DNA ends (9).
Artemis is a mammalian protein, in which mutations were identified first in children with radiation-sensitive SCID and then in Athabascan-speaking children with SCID (SCIDA) (10, 11). Radiation-sensitive SCID represents a small group of patients of various ethnic origins, who show increased radiosensitivity in their fibroblast cells, and impaired coding joint (CJ) formation in an extrachromosomal V(D)J recombination assay (10). SCIDA has a high incidence in the Navajo and Apache populations. SCIDA patients have a T–B–NK+ immunophenotype and also show a profound impairment in V(D)J CJ formation (11, 12, 13). Interestingly, children with SCIDA (13) and other forms of T–B–NK+ SCID (14) do not engraft as efficiently as children with T–B+NK– SCID and often require a chemotherapy preparative regimen. This is thought to be due at least in part to the normal NK cell function in these patients.
In an in vitro study, Artemis was found to be an effective substrate of DNA-PKcs (5). It interacted with DNA-PKcs to form a stable complex and acquired endonuclease activity, enabling it to open RAG-derived hairpin structures, and process 3' ends of sequences. In a recently reported 129/Sv mouse model of Artemis deficiency, the T–B–NK+ phenotype was confirmed, as was the increased sensitivity to ionizing radiation in embryonic fibroblasts (15, 16). In addition, accumulation of hairpins supports the role of Artemis at hairpin processing in CJ formation.
To evaluate the response to hemopoietic stem cell transplantation (HSCT) in a mouse model of SCIDA, we constructed a mouse carrying a germline deletion of the Artemis murine counterpart, mArt. We confirm that Artemis-deficient mice closely resemble DNA-PKcs-deficient mice in that both present with severely impaired CJ formation and near normal signal joint (SJ) formation. We also detected unusual long P nucleotide additions in the rare CJ from Artemis-deficient mice, further supporting the potential role of Artemis in hairpin processing. HSCT resulted in discordant reconstitution of the immune system with a preference toward the T cell compartment, consistent with what has been seen in children with SCIDA.
Materials and Methods
Cloning of mArt in mice and targeted disruption of the mArt gene
The Artemis murine counterpart, mArt, was identified by a database search, RT-PCR, RACE experiments, and sequence assembly. It was then further consolidated by radiation hybrid (RH) mapping using the mouse T31 RH panel and the T31 mouse RH database (The Jackson Laboratory), and genomic organization determination by a series of intron PCR and restriction mapping.
The targeting construct was generated by cloning a 3.2-kb mArt DNA fragment (exons 4–6) and a 9.1-kb fragment (exons 11–14) into pPgk.tk.neo, resulting in the two mArt genomic fragments flanking the neomycin resistance gene, which is driven by the phosphoglycerate kinase promoter, for positive selection. The construct also contains an HSV-thymidine kinase gene, also driven by the phosphoglycerate kinase promoter, to select against nonhomologous integrations of the construct in the genome (negative selection). The construct (25 μg) was linearized at the unique NotI site and used for electroporation to transfect 129/Sv embryonic stem (ES) cells, and the targeted ES cells were enriched by positive selection with G418 (300 μg/ml) and negative selection with 0.2 μM 1-(2'-deoxy-2'-fluoro--D-arabinofuranosyl)-5-ioduracil. After 10 days of selection, visible ES colonies were picked and expanded, genomic DNA was prepared, and standard Southern blot analysis was performed to identify the targeted ES cell clones. Two external probes were used (Fig. 1A): the 5' one detects a 15-kb BamHI fragment for the wild-type (wt) and a 5-kb fragment for the targeted allele, while the 3' one detects a 12-kb EcoRI fragment from the wt allele and a 9-kb fragment from the targeted allele. We also used a neomycin internal probe recognizing an EcoRI fragment (6 kb) from the targeted ES allele.
FIGURE 1. Inactivation of mArt by homologous recombination. A, Diagrammatic representation of the mArt locus (top), the targeting construct (middle), and the targeted allele (bottom) with hybridization probes indicated by small blocks. EcoRI (E) and BamHI (B) restriction sites are indicated, together with the sizes of products. B, Southern blot analysis of EcoRI-digested ES cell DNA showing positive ES cell clones (lanes 2 and 5) with 3' probe and Neo probe. C, DNA from progenies of heterozygote crossing was analyzed by PCR, amplifying a 190-bp product from wt, and a 390-bp product from the targeted allele. D, Western blot analysis showing the absence of Artemis expression in Artemis-deficient thymus.
Generation of Artemis-deficient mice
The positive ES clones with normal karyotype were used for injection into C57BL/6 (B6) blastocysts, which were transferred to CD1 pseudopregnant females. The progenies were monitored by coat color to obtain chimeric males, who were bred with B6 females to generate heterozygotes that carry the targeted allele transmitted through the germline. A PCR assay was used to determine the knockout mice, in which three primers were included in the PCR: E6F, lying 5' to the targeted deletion, i6R from the region that is deleted in the targeted allele, and NeoR; thus, E6F and i6R amplify a 190-bp product from the wt allele, and E6F and NeoR amplify a region (390 bp) from the targeted allele. Finally, the selected male heterozygotes were crossed with wt C57BL/6 females for two generations (N2) and heterozygotes mated to generate F1 homozygous (mArt–/–) progenies.
All of the mice involved in this study were handled under a protocol approved by the University of California Committee on Animal Research following National Institutes of Health approved guidelines. The animals were bred and monitored in the University of California Barrier Facility in autoclaved cages. DNA-PKcs mutant (scid) breeding pairs (CB17-Prkdcscid/SzJ; The Jackson Laboratory) were bred under the same conditions.
Western blot analyses
Whole cell lysates were prepared from lymphoid tissues, resolved in a 7% SDS-PAGE gel, transferred to nitrocellulose membranes, and hybridized with a chicken IgY Ab (1:1000; Genway Biotech) raised against the C-terminal of the human Artemis gene. Anti-actin Ab (1:5000; Sigma-Aldrich) was used as an internal control.
Flow cytometry and lymphocyte function assays
Single cell suspensions were prepared from the thymus, spleen, and bone marrow (BM), and cells were stained with mAbs for flow cytometric analysis on a FACScan (BD Biosciences), as described (17). The conjugated mAbs (BD Pharmingen) were used in combination, as indicated in the figure legends. In addition, we evaluated the lymphocyte proliferative response to Con A (5 μg/ml), solid-phase anti-CD3 (2 μg/ml), and LPS (5 μg/ml), as well as NK cytotoxicity, as described (17).
Histology evaluation
Tissue sections of various organs from 4- to 5-wk-old mice were examined. Briefly, lymphoid and other organs were fixed in Bouin’s solution and embedded in paraffin. Tissue sections were stained with H&E, examined, and photographed in a Zeiss photomicroscope.
PCR analysis of V(D)J recombination products and intermediates
DNA was prepared from lymphoid tissues of 2- to 4-wk-old animals for a PCR-based evaluation (7, 8). For Ig loci, DHL and JH2 were used to evaluate D-JH rearrangements (7), and VH7183y and JH2y to evaluate VH-(D)JH rearrangements. VH7183y (5'-CGATTCAYCATYTCYAGAGABAAT-3') recognizes all 19 VH7183 family members. A JH2-specific probe from the 5' JH2 coding sequence was used to probe and evaluate the amplified rearrangement products. Evaluation for IgL chain was performed with J, V, and V deg (a degenerate primer that amplifies most VK segments) as described (18).
For TCR rearrangement, V8 (8.1, 8.2, and 8.3) to J2.6 CJ formation was examined, as described (8), as well as D2 to J1 (7) and V8 to J50 rearrangements (19). To analyze SJ formation, we used DR21 and DR161 for circular PCR to amplify the SJ formation for the D2 to J1 rearrangement, and ApaLI digestion was used to assess the fidelity of the SJ formation (4, 7). Furthermore, ligation-mediated PCR (LM-PCR) was used to evaluate V(D)J recombination intermediate SE at the 5' end of the D2 locus (7). In addition, the recovered CJs were cloned into a pGEM-T vector (Promega), selected by colony PCR, sequenced, and analyzed.
Extrachromosomal V(D)J recombination and ionizing radiation (IR) sensitivity assays
We used a modification of the extrachromosomal assay for human cells that we have previously described in detail (11). Briefly, mouse embryonic fibroblasts were established from 14-day F1 fetuses of N2 generation Art+/– pregnancies. The cell lines were phenotyped, and the third passage was used for the assays. The murine cells were transiently transfected with 4 μg of RAG1/2, and 2 μg of pJH200 (ampr, SJ) or 2 μg of pJH290 (ampr, CJ) by electroporation in Opti-MEM at 975 μF and 300 V. A full-length human Artemis primary transcript expression construct, pCMV-ART (2 μg), was added to the transfection to assess complementation. pJH200 and pJH290 carry a cam gene that is interrupted from its promoter by a transcriptional terminator flanked by recombination signal sequences. Upon transfection, RAG proteins induce V(D)J recombination in the extrachromosomal plasmid substrates, resulting in the excision of the transcriptional terminator and the activation of chloramphenicol resistance. After 48 h of incubation in DMEM at 37°C, 5% CO2, the pJH200 or pJH290 plasmid was recovered using the alkaline lysis method and digested with DpnI to select for those that had been replicated in the Sv40-transformed murine embryonic fibroblasts (MEFs). The selected plasmids were transformed into DH10B by electroporation to assay for ampicillin and chloramphenicol resistance. The percentage of successful recombination is represented by the ratio of colonies grown on ampicillin/chloramphenicol (from recombined substrate only) vs ampicillin plates (11).
For IR sensitivity assays, varying numbers (2.5 x 102 to 2 x 103) of MEFs were plated in triplicate, and incubated for 5–10 h at 37°C, 5% CO2. Cells were then exposed to the indicated dose of x-rays using a Pantak x-ray generator operating at 320 kV/10 mA with 0.5 mm copper filtration, and returned to the incubator for 7 days. Colonies arising from surviving cells were stained with crystal violet and scored. Percentage of survival was calculated as the number of colonies on treated dishes over those on untreated dishes. The data presented represent the average of three independent experiments.
HSC transplants
Young adult B6 wt congenic or BALB/c allogeneic mismatched mice (The Jackson Laboratory) were used as donors. BM cells were freshly harvested, and several donor cell preparations were used, including whole BM (WBM), T cell-depleted BM (TCDBM), lineage-depleted (lin–) BM, and BM that was lin– and positively selected for Sca-1 (17). TCDBM was prepared using an anti-CD3 mAb and immunomagnetic beads, as previously described (17). The purification of Sca-1-positive cells was done in two steps: first, negative selection of lineage-committed cells using the Hemopoietic Stem Cell Enrichment Kit by StemCell Technologies; second, positive selection using anti-Sca-1 mAb directly labeled to paramagnetic particles (Miltenyi Biotec). The process resulted in a highly pure lin– Sca-1+ HSC population (>95%).
The recipient mice were 7–10 wk of age at the time of transplant, and either 1 x 106 WBM, TCDBM or lin– BM, or 500–40,000 lin– Sca-1+ HSC were injected i.v. Most of the animals received no prior conditioning therapy, but one group was pretreated with 300 cGy of total body irradiation (TBI) using a cesium137 animal irradiator (20, 21). Controls included age-matched mArt–/–, wt, or heterozygotes that received a sham injection, or allogeneic HSCT, or TBI and allogeneic HSCT. T and B cell reconstitution was evaluated by flow cytometry using a mAb specific for BALB/c allogeneic (anti-H2Kb) or congenic (Ly-5.1) donor cells, and V(D)J recombination in the thymus and BM.
Results
Targeted disruption of the mArt gene
The mArt gene that we obtained matches Artemis with an overall homology of 81% (AF387731), with an open reading frame of 2118 bp. RH mapping placed this gene at a region on chromosome 2 that is syntenic to the human Artemis region on chromosome 10p, and furthermore, it has a structure similar to the human Artemis gene, and both contain 14 similarly sized exons and introns (data not shown).
The targeting vector was designed as shown in Fig. 1A. Upon transfection, homologous recombination in the genome resulted in the replacement of the 8-kb mArt fragment (exons 6–11), with the 1.8-kb neomycin gene, leading to the removal of 80% of the mArt gene. We obtained 15 ES clones that were successfully targeted (Fig. 1B). Three of them were used for blastocyst injections, from which we obtained two healthy chimeric males. Both showed successful germline transmission of the targeted ES cells and produced heterozygous mice, which appeared similar to their wt littermates and were crossed to generate 125 animals for this study, 26 of which (21%) were homozygous for the deleted allele (Fig. 1C and data not shown).
The disruption of Artemis function was demonstrated in the Western blot analysis of Artemis expression in lymphoid tissues (thymus, spleen, and lymph nodes). A unique protein band migrating at 105 kDa, a size similar to that observed for human Artemis, was detected in wt and heterozygous littermates, but not in mArt knockout mice (Fig. 1D and data not shown). The data suggest that the targeted deletion is a null allele (mArt–/–).
Early arrest of B and T cell differentiation in mArt–/– mice
We examined five mArt–/– mice at 4–5 wk of age, and four at 2 wk of age, together with 10 wt and heterozygote littermates and 3 scid mice. Gross and histologic examination of various organs revealed abnormalities only in the lymphoid tissue of mArt–/– mice. mArt–/– thymus, spleen, and lymph node were smaller and poorly developed, and the most profound effect on development was seen in the mArt–/– thymus, which weighed 20% of the thymi from wt and heterozygote controls. Thymocyte counts were 0.7–4% of that observed in wt littermates (Table I).
Table I. Lymphoid cellularity of mArt–/– micea
The developmental defects of lymphoid tissues were further demonstrated in microscopic examination. In contrast with wt littermates (Fig. 2A), the thymus of mArt–/– mice showed absence of the lymphocytic cortex, and predominance of connective tissue with scattered lymphoid cells (Fig. 2B). These scattered lymphoid cells showed remarkably abundant mitotic figures (Fig. 2B, high power). Compared with littermate controls (Fig. 2C), the spleens of mArt–/– mice showed notable reduction in fully developed lymphoid follicles (Fig. 2D), in which the majority of lymphoid cells showed larger nuclei with less dense chromatin and prominent nucleoli (Fig. 2D, high power). In contrast to the lymphoid depletion, the interfollicular hemopoietic red pulp was very conspicuous in mArt–/– spleens (Fig. 2D). The mArt–/– lymph nodes were much smaller than those of their littermates (Fig. 2E), and were severely depleted of mature lymphocytes, with complete absence of lymphoid follicles (Fig. 2F).
FIGURE 2. Histologic (10x) evaluation of lymphoid tissues. A, Thymic tissue from wt, and B, from mArt–/– mice; inset shows x40 view. C, Splenic tissue from wt, and D, from mArt–/– animal; inset shows x40 view. E, Lymph node of wt, and F, of mArt–/– animal.
Analyses of B cell maturation in the 2- to 5-wk-old mArt–/– mice revealed a developmental arrest at an early progenitor stage. As presented in Fig. 3A from a representative flow cytometry analysis, the majority of the developing B cells (B220+) were arrested at the progenitor (B220+CD43+) stage, in sharp contrast to wt littermates in which most of the developing B cells have progressed to more mature stages (B220+CD43–). The arrest was also evident in the mArt–/– spleen, where only a very small number of developing B cells (1.5% of splenocytes analyzed) had progressed to immature and mature B cells (B220+IgM+), in contrast to the 27% observed in wt controls (Fig. 3A). Peripheral blood cell counts and B lymphocyte numbers were also similarly reduced (data not shown). Heterozygote littermates showed similar results in all cases to that seen in wt mice (data not shown).
FIGURE 3. Flow cytometric analysis showing early arrest of lymphocyte development in Artemis-deficient mice. A, Analysis of BM and splenocytes from 4-wk-old animals. Numbers represent the percentage of total cells within indicated region. B, Analysis of thymocytes and splenocytes from 2-wk-old animals.
Analysis of mArt–/– thymocytes revealed a similar early arrest of T cell development, which was reflected by >90% of the thymocytes being arrested at the CD4–CD8– stage, compared with 5–10% observed in wt and heterozygous littermates (Fig. 3B). CD25+ analysis of gated CD4–CD8– thymocytes showed that there were at least 20 times more CD4–CD8–CD25+ cells in mArt–/– mice than in wt and heterozygous controls (data not shown). The early arrest was also reflected by the near absence (<0.1%) of CD8+CD3+ and CD4+CD3+ mature single-positive cells in 2-wk-old mArt–/– spleens (data not shown). Analysis of 4- to 5-wk-old animals revealed a small increase in single-positive splenocytes in mArt–/– mice (Fig. 3B). Approximately 2% of the mArt–/– splenocytes analyzed were CD4+CD3+ and 1% CD8+CD3+, but these numbers were still much lower compared with the 10 and 6% observed in wt controls (Fig. 3B and data not shown).
T cell proliferation to Con A and anti-CD3 in mArt–/– mice was reduced to 16.5 and 18.7% of that seen in wt controls, and B cell proliferation to LPS was 13.5% of that in wt. NK cell number and cytotoxicity were normal in mArt–/– mice (data not shown).
Impaired Ig and TCR rearrangement in mArt–/– mice
In PCR-based analyses, DH-JH2 and VH7183-JH2 rearrangements were abundant in wt and mArt+/– controls, but rare in mArt–/– mice at a level comparable to that seen in age-matched scid mice. Both mArt–/– and scid mice consistently showed 10-fold decrease in DH-JH2, and 100-fold decrease in VH7183-JH2 rearrangements (Fig. 4A). TCR CJ formation was reduced 10100-fold in mArt–/– mice (Fig. 4B), and a similar result was seen for D2-J1 CJ rearrangement. The IgL() and TCR rearrangements were undetectable in mArt–/– mice (data not shown).
FIGURE 4. PCR-based analysis showing profound reduction of CJ formation, and near normal SJ formation in 2- to 5-wk-old Artemis-deficient mice. A, D-JH2 and VH7183-DJH2 CJ formation at IgH loci in BM. B, V8-DJ2.6 and D2-J1 CJ formation at TCR and loci. C, Detection of circular excised D2-J1 SJ formation; two-thirds of the PCR products were digested with ApaLI. D, LM-PCR detection of SE at the 5' end of the D2, and the integrity examination by ApaLI digestion.
The amount of D2 to J1 SJ amplification from mArt–/– mice was similar to that detected in wt and heterozygous controls from independent experiments, and ApaLI digestion of SJ joints in mArt–/– mice also appeared similar to scid mice and wt controls (Fig. 4C). These results suggest near normal SJ formation in mArt–/– mice. LM-PCR evaluation showed that the SEs at 5' of D2 locus are abundant in mArt–/– mice at a level similar to their littermate controls, and furthermore, LM-PCR products from mArt–/– and wt mice were both almost completely digested by ApaLI, suggesting that the SEs are intact in mArt–/– mice (Fig. 4D).
Aberrant rare V(D)J CJ rearrangement events in mArt–/– mice
We analyzed a total of 7 DH-JH2, 12 VH7183-D-JH2, 22 D1-J2.6, and 34 V8-D-J2.6 CJs from mArt–/– mice, together with those from wt controls at these loci (10, 14, 16, and 26, respectively). We observed standard coding junctions for most of the rearrangements at these loci; however, we detected 8 unusual long P elements in 29 mArt–/– D-J CJs analyzed, but none in the 26 D-J CJs from wt mice. We also found relatively larger deletions in mArt–/– mice, i.e., 6 deletions (11 nt) in 75 mArt–/– CJs, but only 1 in the 66 wt CJs analyzed, and 1 of the VH7183-D-JH2 CJs showed a 28-nt deletion in JH2 (Fig. 5 and data not shown).
FIGURE 5. Sequence analysis of D1-J2.6 CJ recovered from wt, mArt–/– mice, and also from post-BMT mArt–/– mice. N and P are for nucleotides not present in the germline sequences.
In vitro V(D)J recombination corrected by transduction with human rART
We used an extrachromosomal V(D)J recombination assay to evaluate V(D)J recombination in MEFs from N2 mice (Table II). In four separate mArt–/– cell lines in two experiments, CJ formation was undetectable in 2 and <1/50 of that seen in mArt+/+ cells. SJ formation was comparable to that seen in wt cells. Transduction of the human Artemis primary transcript expression construct restored CJ formation in mArt–/– MEFs.
Table II. Extrachromosomal V(D)J recombination in mArt–/– and mArt+/+ MEFs
Radiation sensitivity in mArt–/– MEF cells
We used colony formation assays to evaluate the relative radiation sensitivity of mArt–/– and mArt+/+ MEFs (Fig. 6). Like human fibroblasts defective in Artemis, we found that MEFs from mArt–/– mice exhibited elevated sensitivity to IR relative to mArt+/+ MEFs (22, 23).
FIGURE 6. Radiation sensitivity of mArt–/– MEF cells. Percentage of survival of SV40-transformed mArt–/– and mArt+/+ MEF cells is plotted as a function of x-ray dose (rad). These data represent the average of three independent experiments.
T and B cell reconstitution following HSCT
Young adult B6 wt, congenic, or BALB/c allogeneic mismatched mice were used as donors. BM was freshly harvested, and several donor cell preparations were used, including WBM, TCDBM, lineage-depleted (lin–) BM, and BM that was lin– and positively selected for Sca-1. The recipient mice were 4–12 wk of age at the time of transplant. A subgroup in each experiment was pretreated with 300 cGy of TBI using a cesium137 animal irradiator. Control animals were mArt–/– littermates that received a sham injection or mArt+/+ age-matched mice that were injected with similar marrow preparations. Animals were followed until they were 3–4 mo post-HSCT when they were euthanized and analyzed for engraftment and immune reconstitution. All control mice died before the first bleed at 4 wk, while engrafted mArt–/– recipients of congenic or allogeneic BM survived to at least 14 wk post-BM transplantation (BMT); survival correlated with T cell engraftment (data not shown).
To evaluate the effects of cell dose, we transplanted animals with 500 or 5000 congenic or allogeneic lin– Sca-1+ HSC. With either 500 or 5000 HSC, the congenic recipients appeared to have some engraftment with limited immune reconstitution (Table III). With 5000 HSC, the T cell engraftment in the congenic recipients was significantly greater, but with limited evidence for B cell recovery. In the allorecipients, neither 500 nor 5000 HSC resulted in significant engraftment. With low dose TBI, there was significant engraftment (congenic > allogeneic) with some B cell reconstitution in the congenic recipients, but not in the allogeneic recipients.
Table III. Engraftment in blood at 8 wk post-HSCT in mArt–/– mice: 500 vs 5000 cells
We then evaluated the effects of using a larger cell dose of HSC as well as different preparations of BM. Because the results were comparable in recipients of 1 x 106 TCDBM and lin– BM, and 40,000 lin– Sca-1+ BM, the groups were combined and labeled Sca-1 (Table IV). In the two recipients of 1 x 106 congenic WBM cells, some T and B cell reconstitution could be seen at 6 wk. With 40,000 Sca-1+ cells, T cell reconstitution was comparable (14.7% at 6 wk). With 300 cGy of TBI before transplant, there was a significant increase in T cell reconstitution (22.9% at 4 wk and 69.2% at 6 wk). In the allogeneic mismatched recipients of 40,000 Sca-1 HSC (Table IV), there was some B cell engraftment at 6 wk (4.7 ± 1.1%), possibly due to the large cell dose. In the Sca-1/TBI group, donor T cells (66.9 ± 12.6%) and B cells (12.1 ± 4.2%) could also be detected in the blood. The percentage of host T and B cells in this group of irradiated animals was 12.1 ± 4.9% and 1.2 ± 0.8%, respectively, possibly a result of the TBI exposure (24).
Table IV. Immune reconstitution in mArt–/– mice following congenic and allogeneic HSCT with 40,000 bone marrow cells
Fig. 7A shows an analysis at 12 wk post-BMT from an animal that received 40,000 lin– Sca-1+ allogeneic HSC (without TBI), in which the reconstitution of T cell immunity in the thymus and spleen was virtually normal, and the splenocyte proliferative responses to Con A and PHA were restored to normal levels (data not shown). Although there is some evidence for B cell maturation in the BM, i.e., 5.4% B220+ CD43– cells vs 0% in a mArt–/– littermate control and 19.3% in a wt littermate (data not shown), there was very little evidence for B cell maturation in the spleen (0.7% B220+IgM+ in mArt–/– transplant recipient vs 0.1% in mArt–/– untransplanted vs 26.1% in wt). Also, the proliferative response to LPS remained lower in this animal (2199 ± 246 cpm), compared with the wt control (8436 ± 316 cpm), and was similar to the untreated mArt–/– control (1580 ± 40 cpm).
FIGURE 7. Effects of HSCT in mArt–/– mice. A, FACScan analysis of B and T cells in spleen, BM, and thymus in a recipient of lin– Sca-1+ allogeneic BM at 12 wk posttransplant without TBI; B, Histology (x10) of thymus, spleen, and lymph node at 12 wk posttransplant (without TBI); and C, PCR analysis of V(D)J rearrangement for VH7183-DJH2 CJ formation in BM, and V8-DJ2.6 in thymus.
Reconstitution of T cell immunity in mArt–/– mice was also evident in the histology post-BMT (Fig. 7B). Transplanted mice showed normal sized thymi with a remarkable cortex rich in lymphocytes, and scant thymic medulla, very similar to mArt+/+ controls. The spleens of transplanted mArt–/– mice also showed more white pulp than untransplanted mArt–/– mice, but notably less than mArt+/+ controls. Occasional lymphoid follicles were observed in transplanted mArt–/– spleens, but none of them showed pale-staining germinal centers. Similarly, although the lymph nodes of post-BMT mArt–/– mice had abundant lymphoid cells, well-developed lymphoid follicles with germinal centers were absent.
T and B cell reconstitution when pretransplant irradiation was given was reflected in V(D)J recombination activity (Fig. 7C). However, while TCR -chain rearrangement (V8 to DJ2.6) was apparent at a dramatically increased level in the post-BMT thymus, IgH chain rearrangement in BM (VH7183 to DJH2) was only slightly increased and much lower compared with the wt control. Finally, 21 rare CJ of TCR D to J rearrangements that were analyzed all showed normal junction sequences without unusual P element addition (Fig. 5).
Discussion
Artemis is essential for lymphocyte maturation
Our findings in N2 mice on the C57BL/6 background carrying a targeted disruption of Artemis are for the most part comparable to those recently reported in 129/Sv mice (15, 16). The loss-of-Artemis function results in the arrest of T and B progenitor cells in lymphoid tissues and the dramatic decrease of mature T and B cells in spleen and peripheral blood. The immunodeficient phenotype of Artemis-deficient mice is similar to that seen in SCIDA patients, although the arrest may be more complete in humans in that we found some residual in vitro T cell proliferation in response to mitogens (<20% of controls), which is usually <10% of normal in SCIDA patients (13). In an evaluation of SCIDA BM samples, we also found the early arrest of B cell development and a near absence of more mature B cells (B220+IgM+) (data not shown). Our results provide direct evidence that a defect in Artemis is the pathologic mechanism for SCIDA.
Artemis is essential for CJ, but not SJ formation in vivo
mArt–/– MEFs were deficient in CJ, but not SJ formation in the in vitro assay. This could be complemented with the transduction of a human Artemis primary transcript expression construct. Associated with the early arrest of T and B cell differentiation, Artemis-deficient mice showed severely impaired CJ formation in both T and B cell lineages, which provides direct in vivo evidence for the involvement of Artemis in V(D)J recombination. Rare CJs were detected at IgH and TCR loci in Artemis-deficient mice at a significantly decreased level that is comparable to that seen in scid control mice. A leaky V(D)J recombination activity associated with an incomplete SCID phenotype has been observed previously in DNA-PKcs-mutant/deficient and Ku86-deficient mice, but not in animals carrying deletions for RAG1/2, XRCC4, or ligase IV. It seems likely that when RAGs function normally in cutting V, (D), and J segments, these segments can then be assembled at low efficiency into V(D)J joints by either residual nonhomologous end joining activity or some other repair pathway. However, RAG1/2–/– animals are unable to initiate V(D)J recombination, while XRCC4–/– or ligase IV–/– mice are unable to form the joints, in all cases resulting in a total blockage of V(D)J rearrangement and a complete SCID phenotype. In contrast, mArt–/– SE and SJ formation does not appear to be affected to a level that is evident in the LM-PCR- and PCR-based analysis. These results are consistent with our previous results in SCIDA patients, and support the preferential impact of Artemis on the formation of CJ.
Analysis of the rare mArt–/– CJ sequences reveals an overall normal N nucleotide addition, but unusual P nucleotide additions and large deletions that we did not see in wt and heterozygous controls. Longer P additions and larger deletions have been frequently observed in scid mice (25, 26). Because the P elements are thought to reflect the position in which hairpinned CEs are opened, and the deletion modification at the junction is believed to result from nuclease activity, the unusual longer P additions and larger deletions observed in DNA-PKcs-deficient and mArt–/– CJs support the functional involvement of DNA-PKcs and Artemis in hairpin opening and CE processing. These in vivo results are consistent with the in vitro data that suggest the association of these two proteins and their involvement in hairpin opening/CE processing (5).
HSCT at least partially corrects the mArt–/– defect
The results of HSCT in the mArt–/– mice are for the most part comparable to what is seen in children with this form of T–B–NK+ SCID. Specifically, even with congenic wt BM, as in children with SCIDA receiving HLA-matched sibling transplants (13), there is a preferential reconstitution of T cell immunity, which can be partially overcome with sublethal TBI, similar to what we have seen in children with SCIDA (13). However, murine congenic donors are more like syngeneic and less like HLA-matched siblings, which probably explains the degree of multilineage engraftment seen in the murine model (Table III). There was also a similar resistance to allogeneic mismatched HSCT at the lower cell doses (500–5000 HSC) in the Artemis-deficient recipients that we and others have seen in children with T–B–NK+ SCID who receive haplomismatched donor grafts (13). At higher cell doses, we found that this resistance could be overcome, something that has not been tested in humans. A haplomismatched donor-recipient pair might have resulted in a different outcome; however, the mutation currently only exists on either the B6 or 129 backgrounds. We plan to cross the mArt–/– mutation onto the BALB/c background in the future. Despite the T cell reconstitution post-HSCT, we found only limited B cell recovery, a problem that is also seen in human SCIDA (13). To demonstrate that immune reconstitution could be enhanced in mArt–/– recipients by providing a growth advantage in the marrow for donor cells, we treated them with low dose TBI using a dose of radiation that has been used previously in scid and DNA-PKcs-deficient mouse models that are also radiation sensitive (20, 21). We saw no acute adverse effects of this therapy, although we did find that mArt–/– MEFs had increased sensitivity to ionizing radiation, comparable to what was reported in the other Artemis-deficient murine model (15). We did not follow these animals out sufficiently long to detect long-term effects of the radiation that are likely to lead to increased tumor formation and possibly growth delays due to their DNA repair defect.
In summary, our results demonstrate that V(D)J CJ formation is impaired after the RAG-mediated cleavage in mArt–/– mice, resulting in defects in hairpin opening and coding end processing, and severely impaired CJ formation. These defects result in a severe maturation arrest of T and B lymphocyte and a T–B–NK+ phenotype, which can be partially corrected by HSCT.
Acknowledgments
We thank Peter Goebel for many helpful discussions, and Brian Wardwell, Peter Skewes-Cox, and Elisa Ng for their valuable technical support.
Footnotes
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
1 This work was supported by grants from the National Institutes of Health (AI 28339), the March of Dimes (6-FY00-301), the Children’s Health Initiative, Lucille Packard Foundation, and the Nigel Gough Memorial Fund.
2 L.L. and E.S. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Morton J. Cowan, Department of Pediatrics, University of California, 505 Parnassus Avenue, San Francisco, CA 94143-1278. E-mail address: mcowan@peds.ucsf.edu
4 Abbreviations used in this paper: SE, signal end; BM, bone marrow; BMT, BM transplantation; CE, coding end; CJ, coding joint; DNA-PKcs, DNA-dependent protein kinase; ES, embryonic stem; HSC, hemopoietic stem cell; HSCT, HSC transplantation; IR, ionizing radiation; LM-PCR, ligation-mediated PCR; MEF, Sv40-transformed murine embryonic fibroblast; P, palindromic; RH, radiation hybrid; SCIDA, Athabascan-speaking children with SCID; SJ, signal joint; TBI, total body irradiation; TCDBM, T cell-depleted BM; WBM, whole BM; wt, wild type.
Received for publication March 3, 2004. Accepted for publication December 2, 2004.
References
Grawunder, U., R. B. West, M. R. Lieber. 1998. Antigen receptor gene rearrangement. Curr. Opin. Immunol. 10:172.
Bassing, C. H., W. Swat, F. W. . 2002. Alt: the mechanism and regulation of chromosomal V(D)J recombination. Cell 109:S45.
Lewis, S. M.. 1994. The mechanism of V(D)J joining: lessons from molecular, immunological and comparative analyses. Adv. Immunol. 56:27.
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, V. L. M. Herrera. 1998. Targeted disruption of the catalytic subunit of the DNA-PK gene in mice confers severe combined immunodeficiency and radiosensitivity. Immunity 9:355.
Ma, Y., U. Pannicke, K. Schwarz, M. R. Lieber. 2002. Hairpin opening and overhang processing by an Artemis/DNA-dependent protein kinase complex in nonhomologous end joining and V(D)J recombination. Cell 108:781.
Featherstone, C., S. P. Jackson. 1999. Ku, A DNA repair protein with multiple cellular functions. Mutat. Res. 434:3.
Zhu, C., M. A. Bogue, D.-S. Lim, P. Hasty, D. B. Roth. 1996. Ku86-deficient mice exhibit severe combined immunodeficiency and defective processing of V(D)J recombination intermediates. Cell 86:379.
Ouyang, H., A. Nussenzweig, A. Kurimasa, V. da Costa Soares, X. Li, C. Cordon-Cardo, W. H. Li, N. Cheong, M. Nussenzweig, G. Iliakis, et al 1997. Ku70 is required for DNA repair but not for T cell antigen receptor recombination in vivo. J. Exp. Med. 186:921.
Grawunder, U., D. Zimmer, P. Kulesza, M. R. Lieber. 1998. Requirement for an interaction of XRCC4 with DNA ligase IV for wild-type V(D)J recombination and DNA double-strand break repair in vitro. J. Biol. Chem. 273:24708.
Moushous, D., I. Callebaut, R. de Chasseval, B. Corneo, M. Cavazzana-Calvo, F. le Deist, I. Tezczn, O. Sanal, Y. Bertrand, N. Philippe, et al 2001. Artemis, a novel DNA double-strand break repair/V(D)J recombination protein, is mutated in human severe combined immune deficiency. Cell 105:177.
Li, L., D. Moshous, Y. Zhou, J. Wang, G. Xie, E. Salido, D. Hu, J. P. de Villartay, M. J. Cowan. 2002. A founder mutation in Artemis causes severe combined immunodeficiency disease in Athabascan-speaking Native Americans. J. Immunol. 168:6323.
Moshous, D., L. Li, R. de Chasseval, N. Philippe, N. Jabado, M. J. Cowan, A. Fisher, J. P. de Villartay. 2000. A new gene involved in DNA double-strand break repair and V(D)J recombination is located on human chromosome 10p. Hum. Mol. Genet. 9:583.
O’Marcaigh, A., K. DeSantes, D. Hu, H. Pabst, B. Horn, L. Li, M. J. Cowan. 2001. Bone marrow transplantation for T–B– severe combined immunodeficiency disease in Athabascan-speaking Native Americans. Bone Marrow Transplant. 27:703.
Bertrand, Y., P. Landais, W. Friedrich, B. Gerritsen, G. Morgan, A. Fasth, M. Cavazzana-Calvo, F. Porta, A. Cant, T. Espanol, et al 1999. Influence of severe combined immunodeficiency phenotype on the outcome of HLA non-identical, T-cell-depleted bone marrow transplantation. J. Pediatr. 134:740.
Rooney, S., J. Sekeguchi, C. Zhu, H. L. Cheng, J. P. Manis, S. Whitlow, J. DeVido, D. Foy, J. Chaudhuri, D. Lombard, et al 2002. Leaky scid phenotype associated with defective V(D)J coding end processing in Artemis-deficient mice. Mol. Cell 10:1379.
Rooney, S., F. W. Alt, D. Lombard, S. Whitlow, M. Eckersdorff, J. Fleming, S. Fugmann, D. O. Ferguson, D. G. Schatz, J. Sekeguchi. 2003. Defective DNA repair and increased genomic instability in Artemis-deficient murine cells. J. Exp. Med. 197:553.
Bhattacharyya, S., A. Chawla, K. Smith, Y. Zhou, S. Talib, M. J. Cowan. 2002. Increased engraftment with minimal graft-vs-host-disease following in utero transplantation of S-59 psoralen/UVA light-treated sensitized T cells and adult T cell depleted bone marrow in fetal mice. J. Immunol. 169:6133.
Schlissel, M. S., D. Baltimore. 1991. Activation of immunoglobulin gene rearrangement correlates with induction of germline gene transcription. Cell 58:1001.
Zhu, C., M. A. Bogure, D. B. Roth. 1996. Thymocyte differentiation in -irradiated severe-combined immunodeficient mice: characterization of intermediates and products of V(D)J recombination at the T cell receptor A locus. Eur. J. Immunol. 26:2859.
Lapidot, T., O. Kollet. 2002. The essential roles of the chemokine SDF-1 and its receptor CXCR4 in human stem cell homing and repopulation of transplanted immune-deficient NOD/SCID and NOD/SCID/B2m(null) mice. Leukemia 16:1992
Greiner, D. L., R. A. Hesselton, L. D. Shultz. 1998. SCID mouse models of human stem cell engraftment. Stem Cells 16:166.
Nicolas, N., D. Moshous, M. Cavazzana-Calvo, D. Papadopoulo, R. de Chasseval, F. Le Deist, A. Fischer, J. P. de Villartay. 1998. A human severe combined immunodeficiency (SCID) condition with increased sensitivity to ionizing radiations and impaired V(D)J rearrangements defines a new DNA recombination/repair deficiency. J. Exp. Med. 188:627.
Dudasova, Z., M. Chovanec. 2003. Artemis, a novel guardian of the genome. Neoplasma 50:311.
Williams, C. J., I. Grandal, D. J. Vesprini, U. Wojtyra, J. Dansda, C. J. Guidos. 2001. Irradiation promotes V(D)J joining and RAG-dependent neoplastic transformation in SCID T-cell precursors. Mol. Cell. Biol. 21:400.
Fish, S. M., M. J. Bosma. 1994. Abnormal deletions in the T-cell receptor locus of mouse thymocytes. Mol. Cell. Biol. 14:4455.
Shuler, W., N. R. Ruetsch, M. Amsler, M. J. Bosma. 1991. Coding joint formation of endogenous T cell receptor genes in lymphoid cells from scid mice: unusual P-nucleotide additions in VJ-coding joints. Eur. J. Immunol. 21:589.(Lanying Li, Eduardo Salid)