Abstract
Purine analogs such as 6-thioguanine (6TG) cause myelotoxicity upon conversion into nucleotides by hypoxanthine-guanine phosphoribosyltransferase (HPRT). Here we have developed a novel and highly efficient strategy employing 6TG as a single agent for both conditioning and in vivo chemoselection of HPRT-deficient HSC. The dose-response and time course of 6TG myelotoxicity were first compared in HPRT-wild type mice and HPRT-deficient transgenic mice. Dosage and schedule parameters were optimized to employ 6TG for myelo-suppressive conditioning, immediately followed by in vivo chemoselection of HPRT-deficient transgenic donor bone marrow (BM) transplanted into syngeneic HPRT-wild type recipients. At appropriate doses, 6TG induced selective myelotoxicity without any adverse effects on extra-hematopoietic tissues in HPRT-wild type mice, while HSC deficient in HPRT activity were highly resistant to its cytotoxic effects. Combined 6TG conditioning and post transplant chemoselection consistently achieved ~95% engraftment of HPRT-deficient donor BM, with low overall toxicity. Long-term reconstitution of immunophenotypically normal BM was achieved in both primary and secondary recipients. Our results provide proof-of-concept that single-agent 6TG can be used both for myelo-suppressive conditioning without requiring irradiation, and for in vivo chemoselection of HPRT-deficient donor cells. Our results show that by applying the myelosuppressive effects of 6TG both before (as conditioning) and after transplantation (as chemoselection), highly efficient engraftment of HPRT-deficient hematopoietic stem cells can be achieved.
Keywords: Hematopoietic stem cell transplantation, bone marrow, hypoxanthine-guanine phosphoribosyltransferase, 6-thioguanine, conditioning, in vivo chemoselection
Introduction
Clinical efficacy of ex vivo gene therapy using hematopoietic stem cells remains dependent on imparting a selective advantage to the transplanted cells [1, 2]. In order to enhance engraftment and decrease the time needed for lymphohematopoietic reconstitution, in vivo selection strategies employing drug resistance genes such as dihydrofolate reductase (DHFR) [3] or multiple drug-resistance gene 1 (MDR1) [4, 5] have been tested, but have generally failed due to unacceptable toxicity [6] or insufficient selection efficiency [7]. Currently, mutant forms of O6-methylguanine-DNA-methyltransferase (MGMT) are being tested for their ability to confer chemoprotection against BCNU or temozolomide in combination with O6-benzylguanine [8, 9], but these agents also pose a considerable risk of toxicity, and recent observations suggest that mutant MGMT may confer a selective disadvantage when expressed at high levels [10].
Notably, these approaches have generally relied upon transplantation of hematopoietic progenitors overexpressing an exogenous drug resistance gene into recipients preconditioned with myeloablative irradiation; however, chemo-resistance may also be conferred by reduced levels of endogenous enzymes that are normally essential for activation of cytotoxic drugs. In this context, we have previously observed that high levels of the purine nucleotide salvage pathway enzyme hypoxanthine-guanine phosphoribosyltransferase (HPRT) cause increased sensitivity to the purine analog 6-thioguanine (6TG) [11]. The first step of the metabolic conversion of 6TG is catalyzed by HPRT [12], which mediates the addition of ribose 5-phosphate to generate thioguanosine monophosphate (TGMP). Thus, 6TG cytotoxicity is essentially predicated on its HPRT-mediated conversion to thio-dGTPs, which are then incorporated into DNA, inducing futile mismatch repair and subsequent apoptosis.
To confer myeloprotection through reduced activity, the endogenous drug-activating enzyme should normally be highly expressed in hematopoietic progenitors, yet non-essential for normal hematopoiesis. In fact, hematopoietic progenitors normally express high levels of HPRT [13–16], which makes them extremely sensitive to 6TG. Indeed, purine analogs such as 6-mercaptopurine (6MP), azathioprine (Aza), and 6TG have been used clinically for the treatment of leukemia, particularly in pediatric patients, for half a century [17], as well as for immunosuppression in organ transplant patients, and more recently for autoimmune diseases. At higher doses, myelotoxicity is the most frequent and consistent adverse effect of 6TG when used clinically, and when administered at appropriate concentrations for short periods of time, 6TG is strongly myelosuppressive with little toxicity to other tissues in normal HPRT-wild type animals [11].
In contrast, bone marrow (BM) from HPRT-deficient animals is highly resistant to 6TG [11]. Notably, however, we [11] and others [18] have found that hematopoiesis is phenotypically and functionally normal in Hprt-knockout animals, and although cases of megaloblastic anemia have been associated with hereditary HPRT deficiency (Lesch-Nyhan syndrome) in humans [19], this has been reported to respond well to oral administration of adenine [20]. Furthermore, HPRT deficiency does not appear to be associated with any gross impairment of the immune system in humans or in animals [21].
These observations suggest that HPRT-deficient hematopoietic progenitor cells, which are 6TG-resistant but otherwise normal, should have a selective advantage when transplanted into HPRT-wild type recipients undergoing 6TG treatment, and that this strategy can be used to improve the outcome of ex vivo gene therapy. In fact, Porter and DeGregori [22] previously demonstrated the feasibility of transducing HSC with a lentiviral vector expressing Hprt-targeted shRNA and enriching these engineered hematopoietic cells in vivo by 6TG chemoselection in mice. However, in this previous report, 6TG was employed either at dosages chosen to be only moderately myelosuppressive or only over short periods of selection that were initiated 4 weeks or more following transplantation, and despite preconditioning by total body irradiation, engraftment results were relatively modest and highly variable, ranging from 5–50% [22].
We have now systematically examined the effects of modifying the dose, timing, and duration of 6TG administration on engraftment and hematopoietic reconstitution after transplantation of HPRT-deficient bone marrow. In order to exclude vector transduction efficiency as a variable, we employed BM from Hprt-knockout animals as ‘ideal’ donor cells, thereby enabling us to focus on the effects of modifying 6TG dosage and scheduling parameters both (i) pre- transplantation for myelosuppressive conditioning of HPRT-wild type recipients, and (ii) post-transplantation for chemoselective amplification of HPRT-deficient donor cell populations. Consequently, here we report the development of a novel regimen that sequentially employs 6TG as a single agent for both preconditioning and in vivo chemoselection, and show that this combination regimen rapidly and consistently achieves highly efficient engraftment and long-term reconstitution.
Materials and methods
Mice
Hprt-deficient B6.129P2-Hprt1b-m3/J (CD45.2) mice, Hprt-wild type (wt) C57BL/6J, and B6.SJL-PtprcaPepcb/BoyJ (CD45.1) mice were originally obtained from the Jackson Laboratory (Bar Harbor, ME). B6.129P2-Hprt1b-m3/J mice carry a 55 kb deletion spanning the promoter and the first 2 exons of the Hprt-gene [23]. Mice were bred and maintained in the institutional specific-pathogen-free animal facility under standard conditions according to institutional guidelines.
6TG treatment
C57BL/6J and B6.129P2-Hprt1b-m3/J mice were injected intraperitoneally (i.p.) with 200 μl of varying doses of 6TG (Sigma-Aldrich, Saint Louis, MO) at different time points as described in the figure legends. Control animals were i.p. injected with 200 μl of sterile H2O.
Bone marrow transplantation and 6TG in vivo chemoselection
Female recipient C57BL/6J (HPRT-wt) or B6.SJL-PtprcaPepcb/BoyJ (HPRT-wt, CD45.1) mice were treated with 10 mg/kg 6TG by i.p. injection 48 hrs before HSCT. For HSCT, 0.8–1 × 107 nucleated BM cells isolated from B6.129P2-Hprt1b-m3/J male mice (CD45.2) were intravenously injected into HPRT-wt recipients. 6TG (10 mg/kg) was again administered by i.p. injection 2 hrs later, and every 3 days thereafter, at 5 mg/kg for 4 weeks. Serial transplantation into secondary recipient mice was performed using the same cell dose and 6TG preconditioning / chemoselection regimen as above, but using BM from primary recipient mice that had undergone transplantation with 6TG in vivo chemoselection 6 months previously.
Mouse chromosome X- and Y-specific fluorescence in situ hybridization
FISH was performed on BM and PBL cells using the mouse-specific Whole Chromosome-Y paint probe/ RAB9 (XqF1) DNA probes mix (Kreatech, Amsterdam, Netherlands) according to the manufacturer’s protocol. To determine the percentage of male donor HPRT-wt cells in the female recipient C57BL/6J mice, 200 nuclei per slide were counted using a fluorescence microscope (Zeiss) equipped with appropriate dual and triple-colour filters. The following criteria were applied for analyzing FISH: (1) quality of the nuclei was evaluated via DAPI staining, (2) green fluorescent signal for the Y-chromosome was scored, (3) red fluorescent signal for X-chromosome was scored, (4) absence of the green fluorescent signal for the Y-chromosome nucleus was recorded as female, even when only one X-chromosome was detectable (Supplementary Figure S1).
Immunophenotypic analysis of hematopoietic tissues
After blocking with Mouse BD Fc Block (BD Biosciences, San Jose, CA), BM, PBL, thymus, or spleen cells were stained with FITC-, PE-, PerCP- or APC-conjugated rat anti mouse antibodies against CD45, CD45.2, CD4, CD8, Mac1/Gr1, B220, Sca-1, c-kit, and lineage antibody cocktail. Antibodies were received from Biolegend (San Diego, CA) or BD Biosciences. Flow cytometric data were acquired on a BD LSRII running BD FACSDiva (BD Biosciences) and analyzed using FlowJo software (TreeStar, Ashland, OR) (Supplementary Figure S2).
Histopathological analysis
Necropsy of all mice used in this study and histological examination of all thoracic and abdominal organs and bone marrow was performed by the UCLA Division of Laboratory Animal Medicine Diagnostic Service Laboratory. Tissues were routinely processed or decalcified if necessary, and paraffin sections were cut and stained with hematoxylin and eosin (H&E).
Statistical analysis
Data were analyzed using the QuickCalcs statistical software program (http://www.graphpad.com/quickcalcs/index.cfm). Unpaired t-tests were used to calculate p values, and p<0.05 was considered statistically significant.
Results
Acute myelotoxicity of 6TG in HPRT-wt mice
The high levels of HPRT expression in hematopoetic progenitors are predicted to mediate a selective myelotoxic effect of 6TG and enable its use as a conditioning regimen. Accordingly, we examined the short-term effects on BM after bolus injections of 6TG at various dosages in HPRT-wt mice. Intraperitoneal injection of a single dose of 2.5 mg/kg, 5.0 mg/kg, or 10 mg/kg 6TG was performed on Day 1, or injection of two doses at 10 mg/kg on Days 1 and 3, and BM histology was examined on Day 4. Increasing myelotoxicity was observed as the total 6TG dosage was increased (Fig. 1A). Vascular structures became progressively prominent and cells of both erythroid and myeloid lineages were depleted. Only vascular endothelium, mesenchymal cells, some mature granulocytes and macrophages, and occasional hematopoietic progenitor cells remained at the highest 6TG dose tested (Fig. 1B).
Figure 1. Optimization of 6TG conditioning regimen.
(A) As a dose-finding study, HPRT-wt mice were injected i.p. with vehicle control, or varying 6TG doses ranging from 2.5 to 10 mg/kg, as indicated, on Day 1 (n=3 per group), or with two doses of 10 mg/kg on Day 1 and Day 3 (n=3), respectively. On Day 4 after the first 6TG dose, paraformaldehyde-fixed bone sections were stained with H&E, and BM histology was examined. Representative photomicrographs (40 × magnification) are shown for each 6TG conditioning regimen. (B) Representative photomicrographs showing overall and detailed BM histology at low (10x) and high (100x) magnification from HPRT-wt mice treated with control vehicle, or with the optimized conditioning regimen consisting of two doses of 10 mg/kg 6TG on Day 1 and Day 3. Histological analyses were performed on Day 4, as above.
No readily apparent clinical signs were observed on Day 4 with any of the above regimens, but when the observation period was extended, weight loss and pallor of the extremities was seen by Day 7. Histological examination of BM showed severity increasing between Day 4 and Day 7 even without administration of additional doses of 6TG (Fig. 1A, 2A). In concordance with the clinical and histological findings, BM of HPRT-wt mice treated with two doses of 10 mg/kg 6TG and analyzed on Day 7 showed a significant quantitative decrease in the BM nucleated cell count recovered from one femur and tibia (4.7×106 ± 1.0×106, n=3) compared to that of vehicle control-treated HPRT-wt mice (1.1×107 ± 0.2×107, n=3) (p<0.01).
Figure 2. Lack of progressive myelotoxicity after injection with conditioning doses of 6TG in HPRT-deficient mice, and low engraftment rate with 6TG conditioning alone.
(A) HPRT-wt and HPRT-deficient mice were injected i.p. with 10 mg/kg 6TG on Days 1 and 3. On Day 7 after the first 6TG dose, paraformaldehyde-fixed bone sections were stained with H&E, and BM histology was examined. Representative photomicrographs (40 × magnification) are shown. (B) Treatment schedule: HPRT-wt female recipient mice (n=4) received the first conditioning dose of 6TG (10 mg/kg, i.p.) on Day −2, then were transplanted with HPRT-deficient male BM, followed by a second conditioning dose of 6TG (10 mg/kg, i.p.) on Day 0. BM was analyzed on Day 7 after the first 6TG dose. Paraformaldehyde-fixed bone sections stained with H&E (40 × magnification) are shown.
Immunophenotypic analysis of the remaining BM hematopoietic cells by flow cytometry (Table 1) revealed that the relative proportion of KLS (lin−/c-kit+/sca-1+) progenitors, which includes HSC with long-term multilineage reconstitution activity [24], significantly decreased 3-fold by Day 4 (p<0.01), and 10-fold by Day 7 (p<0.001) (Table 1-iii, -iv). The relative percentage of mature CD8+ and CD4+ T cells progressively increased over time after 6TG administration, reaching up to 7-fold compared to the controls by Day 7 (p<0.001), likely due to the massively dilated blood vessels and influx of peripheral blood/hemorrhage, as can be also seen in the BM histology. The relative percentage of B220+ cells showed no significant change by Day 4, but had doubled compared to controls by Day 7 after 6TG administration.
Table 1. Immunophenotypic analysis of BM hematopoietic cells after 6TG conditioning regimen.
BM cells were stained with the following rat anti–mouse antibodies: CD45-FITC, CD4-PE, CD8-APC, B220-PerCP, Mac1/Gr1-PE, Sca-1-PE, and c-kit-FITC, and examined by flow cytometry. Treatment and analysis schedules are indicated by small roman numerals, and are the same as in Figure 1. (i): Vehicle control on Day 1, analysis on Day 4. (ii) Single dose of 10 mg/kg 6TG on Day 1, analysis on Day 4. (iii): Two doses of 10 mg/kg 6TG on Days 1 and 3, analysis on Day 4. (iv) & (v): Two doses of 6TG 10 mg/kg on Days 1 and 3, analysis on Day 7. Please note that the last two columns show the results of the same 6TG dosage and schedule (10 mg/kg 6TG × 2 doses) in HPRT-wt (iv) and HPRT-deficient (v) mice, respectively. Percentages of the indicated hematopoietic cell subpopulations are expressed as mean % ± SD of total CD45+ cells (n=3 per group).
| Cell population | HPRT-wild type
|
HPRT-deficient
|
|||
|---|---|---|---|---|---|
| i | ii | iii | iv | v | |
|
|
|
||||
| CD45+ | 91.5±2.4 | 95.7±0.8 | 95.7±0.1 | 82.8±5.1 | 90.0±2.0 |
| CD4+ | 3.5±0.4 | 11.9±2.5 | 9.6±0.6 | 20.5±3.1 | 2.9±1.7 |
| CD8+ | 3.6±0.1 | 5.3±1.7 | 5.2±0.4 | 25.0±2.7 | 2.7±0.4 |
| B220+ | 28.4±3.3 | 23.1±2.2 | 26.2±1.8 | 56.3±1.5 | 10.7±3.5 |
| Mac1+/Gr1+ | 75.8±1.1 | 85.4±1.8 | 81.0±1.3 | 61.8±3.0 | 79.3±2.9 |
| KLS (HSC) | 4.2±0.6 | 1.3±0.2 | 1.4±0.5 | 0.4±0.2 | 2.0±0.6 |
Notably, liver enzymes were not elevated following this conditioning regimen in HPRT-wt mice (data not shown), indicating the selectivity of 6TG cytotoxicity for hematopoietic progenitors at this dosage, and suggesting its potential for use as a myelosuppressive conditioning regimen.
Lack of myelotoxicity at 6TG conditioning dosage in HPRT-deficient mice
In contrast to the above findings, when HPRT-deficient mice were treated with the maximum dosing (two doses of 10 mg/kg 6TG on Days 1 and 3), bone marrow histology remained completely unaffected at day 7 (Fig. 2A), and was comparable to that of the vehicle control group (Fig. 1A). Importantly, the overall count of nucleated BM cells obtained from one femur and tibia of 6TG-treated HPRT-deficient mice (1.5×107 ±0.3×107, n=3) was also comparable to that of vehicle control-treated HPRT-wt mice (1.1×107 ± 0.2×107, n=3), and was significantly higher than in HPRT-wt mice treated with the same 6TG regimen (4.7×106 ±1.0×106, n=3) (p<0.005).
Furthermore, the ratios of hematopoietic cell subpopulations in the BM of the 6TG-treated HPRT-deficient mice (Table 1-v) were comparable to those of vehicle control-treated HPRT-wt mice (Table 1-i) and untreated HPRT-wt mice (Table 2, BM% column), as well as untreated HPRT-deficient mice (Table 3, BM% column).
Table 2. Immunophenotypic analysis of hematopoietic cells in treatment-naïve HPRT-wt mice.
BM, PBL, spleen (S) and thymus (T) were stained with the following rat anti–mouse antibodies: CD45-FITC (BM, PBL, T, S), CD4-PE (BM, PBL, T, S), CD8-APC (BM, PBL, T, S), Mac1/Gr1-PE (BM, PBL, S), B220-PerCP (BM, PBL, S), Sca-1-PE (BM), and c-kit-FITC (BM), and examined by flow cytometry. Percentages of the indicated hematopoietic cell subpopulations are expressed as mean % ± SD of total CD45+ cells (n=3 per group).
| Cell population |
|
|||
|---|---|---|---|---|
| BM (%) | PBL (%) | Thymus (%) | Spleen (%) | |
| CD45+ | 92.6±3.2 | 98.6±0.5 | 98.7±0.6 | 97.1±1.0 |
| CD4+ | 3.9±1.6 | 12.2±2.6 | 6.5±2.1 | 19.7±1.7 |
| CD8+ | 1.6±0.8 | 11.4±1.7 | 3.6±1.2 | 16.5±0.3 |
| CD4+/CD8+ | 85.6±4.3 | |||
| B220+ | 20.4±8.2 | 47.8±13.2 | 54.4±4.9 | |
| Mac1+/Gr1+ | 87.3±8.1 | 36.1±5.0 | 18.7±3.9 | |
| KLS (HSC) | 3.6±2.4 | |||
Table 3. Immunophenotypic analysis of hematopoietic cells in treatment-naïve HPRT-deficient mice.
BM, PBL, spleen (S) and thymus (T) were stained with the following rat anti–mouse antibodies: CD45-FITC (BM, PBL, T, S), CD4-PE (BM, PBL, T, S), CD8-APC (BM, PBL, T, S), Mac1/Gr1-PE (BM, PBL, S), B220-PerCP (BM, PBL, S), Sca-1-PE (BM), and c-kit-FITC (BM), and examined by flow cytometry. Percentages of the indicated hematopoietic cell subpopulations are expressed as mean % ± SD of total CD45+ cells (n=3 per group).
| Cell population |
|
|||
|---|---|---|---|---|
| BM (%) | PBL (%) | Thymus (%) | Spleen (%) | |
| CD45+ | 92.3±1.3 | 96.1±2.7 | 98.7±0.4 | 94.9±1.8 |
| CD4+ | 4.9±1.4 | 15.9±0.8 | 6.5±1.0 | 18.4±2.7 |
| CD8+ | 2.4±0.2 | 11.7±0.7 | 2.9±0.1 | 11.8±1.9 |
| CD4+/CD8+ | 86.9±0.9 | |||
| B220+ | 28.2±2.7 | 55.7±2.9 | 51.3±5.5 | |
| Mac1+/Gr1+ | 79.7±2.2 | 30.1±4.0 | 9.9±1.8 | |
| KLS (HSC) | 3.3±0.5 | |||
Taken together, these results suggested that up to two doses of 10 mg/kg 6TG could be employed as an effective conditioning regimen that would be well tolerated for up to 3 days prior to HSCT, with progressively increasing myelotoxicity occurring over a period of 7 days.
Transplantation of HPRT-deficient BM after 6TG conditioning in HPRT-wt recipients
Based on the schedule established above, we employed 6TG (10 mg/kg i.p.) as a conditioning regimen in HPRT-wt recipients (CD45.1), with one dose administered 48 hours prior (now designated Day −2, rather than Day 1) and one dose administered on the day of transplantation (designated Day 0, rather than Day 3) per the schedule established above. After conditioning, HPRT-wt recipients were then transplanted with BM from HPRT-deficient congenic donors (CD45.2),
At day 4, the marrow showed reduced cellularity with fewer early progenitor cells and increased vascularity (Fig. 2B), and the overall count of nucleated BM cells recovered from these 6TG-conditioned HPRT-wt mice transplanted with HPRT-deficient BM (3.8×106 ± 0.5×106, n=4) was still significantly reduced (p<0.001) compared to that of the vehicle control-treated HPRT-wt group (1.1×107 ± 0.2×107, n=3). Flow cytometric analyses of BM showed that at day 4 only 17.8% ± 4.4% of the total cell population was derived from donor HPRT-deficient CD45.2+ hematopoietic cells.
Threshold for myelotoxicity after chronic low-dose 6TG treatment
As our results above indicated that 6TG preconditioning alone may not be sufficient to achieve high levels of engraftment, we next performed a dose-finding study in non-transplanted mice given chronic treatment with lower doses of 6TG. HPRT-wt and HPRT-deficient mice (n=3 per group) were injected i.p. with vehicle alone, 0.25 mg/kg, 0.5 mg/kg, 1.0 mg/kg, 2.5 mg/kg, or 5.0 mg/kg 6TG every 3 days for up to 60 days. In HPRT-wt mice, the vehicle control group as well as the 0.25 mg/kg and 0.5 mg/kg 6TG groups showed 100% survival over a period of 60 days, and histological examination of BM in the 0.25 mg/kg and 0.5 mg/kg 6TG groups showed normal cellularity on Day 60 (Fig. 3).
Figure 3. Dose-response and time course of chronic low-dose 6TG myelotoxicity in HPRT-wt vs. HPRT-deficient mice.
HPRT-wild type mice and HPRT-deficient mice were treated every 3 days with different dosages of 6TG or vehicle control (n=3 per group), as shown above each panel. For HPRT-wild type mice, histology was examined at the following time points up to 60 days after initiation of treatment: Vehicle control (Day 60), 6TG 0.25 mg/kg (Day 60), 6TG 0.5 mg/kg (Day 60), 6TG 1.0 mg/kg (Day 38), 6TG 2.5 mg/kg (Day 28), 6TG 5.0 mg/kg (Day 22). For HPRT-deficient mice, histology was examined on Day 60 for all animals. Representative photomicrographs of paraformaldehyde-fixed bone sections stained with H&E (40× original magnification) are shown.
In contrast, in the HPRT-wt 1.0 mg/kg 6TG group, deaths occurred on Day 38 (13 mg/kg total dosage), Day 42 (14 mg/kg total dosage), and Day 51 (17 mg/kg total dosage) (Table S1). At higher dosages, HPRT-wt mice receiving repeated injections of 2.5 mg/kg or 5 mg/kg 6TG consistently showed progressive clinical signs of distress (inactivity, hunched posture, lack of grooming, anorexia), anemia (pallor of extremities) and >10% weight loss, necessitating sacrifice per institutional guidelines on Day 28 (2.5 mg/kg 6TG group; 22.5 mg/kg total dosage) and Day 22 (5.0 mg/kg 6TG group; 35 mg/kg total dosage), respectively. Histological examination of BM from the 1.0 mg/kg 6TG group showed several apoptotic figures and more blast cells than in the lower dosage groups were observed, likely reflecting an initial activation response to the injury (Fig. 3; Day 38). The higher dosage groups also showed significantly reduced cellularity and expansion of vascular structures, with the severity of lesions proportional to the cumulative dosage of 6TG in each group. Mice treated with repeated doses of 2.5 mg/kg 6TG showed reduced cellularity and widespread apoptosis (Day 28). At the highest chronic dosage of 5 mg/kg, BM was markedly depleted, with most of the surviving cells being of myeloid lineage (Day 22).
In contrast to the HPRT-wt mice, all HPRT-deficient mice survived for the duration of the experiment (60 days), independent of the injected 6TG dosage (up to 105 mg/kg maximum total dose, administered according to the same dosing schedules as above) (Table S1). No significant BM pathology was observed in any HPRT-deficient mice, regardless of 6TG dosage, at the terminal time point of the experiment on Day 60 (Fig. 3). No significant treatment-related abnormalities were observed in any other tissues examined, including heart, lung, liver, pancreas, kidney, and spleen. Thus, HPRT-deficient mice showed no toxic effects of chronic 6TG treatment at the 1.0 mg/kg, 2.5 mg/kg, and 5.0 mg/kg doses that caused lethal myelotoxicity in HPRT-wt mice.
Combined 6TG conditioning and in vivo chemoselection achieves consistent and highly efficient engraftment of HPRT-deficient BM
Based on the above dose-finding study, we then asked whether 6TG preconditioning combined with continued administration of lower doses of 6TG, beginning immediately after transplantation of HPRT-deficient BM in HPRT-wt recipients, could achieve further chemoselective amplification of engrafted donor cells. Accordingly, female HPRT-wt mice were preconditioned with 6TG (10 mg/kg i.p.), with one dose administered 48 hours prior (Day −2) and one dose administered on the day of transplantation (Day 0) per the conditioning schedule established previously. The HPRT-wt female recipients were then transplanted with HPRT-deficient male BM, and based on the chronic myelotoxicity results above, the recipients were further treated with repeated doses of 2.5 mg/kg 6TG every 3 days for 2 weeks (30 mg/kg total dosage) or 4 weeks (42.5 mg/kg), or repeated doses of 5.0 mg/kg 6TG every 3 days for 2 weeks (40 mg/kg total dosage) or 4 weeks (65 mg/kg total dosage). Analysis was performed immediately after the in vivo chemoselection period at 2 weeks or 4 weeks, respectively (Fig. 4).
Figure 4. Optimization of combined 6TG conditioning and in vivo chemoselection strategy.
Treatment schedule: HPRT-wt female recipient mice received the first conditioning dose of 6TG (10 mg/kg, i.p.) on Day −2, then were transplanted with HPRT-deficient male BM along with a second conditioning dose of 6TG (10 mg/kg, i.p.) on Day 0. In vivo chemoselection was then performed with repeated i.p. injections of 2.5 mg/kg or 5.0 mg/kg 6TG every 3 days for a period of 2 weeks (A) or 4 weeks (B), as indicated. Representative photomicrographs of bone marrow from paraformaldehyde-fixed sections stained with H&E (40× original magnification) are shown.
As expected, the combined 6TG conditioning, HSCT, and 6TG in vivo chemoselection procedures were well-tolerated, and no signs of distress were observed. In all 6TG-treated groups, 100% of the transplanted animals survived (2 week treatment: n=3 and 4 week treatment: n=8). Body weights decreased initially during the first week after transplantation in the 6TG-treated animals, but stabilized and all animals regained normal weight thereafter. Histopathological analysis showed that overall cellularity and hematopoiesis in the transplanted animals were indistinguishable from the untreated HPRT-wt control, regardless of 6TG chemoselection dosage or duration (Fig. 4).
Chromosome XY-FISH [25, 26] showed that in the groups receiving 2 weeks of 6TG chemoselection, the BM at that time point was already highly reconstituted with donor-derived marrow at levels of 89.3%±1.7 (2.5 mg/kg group) and 95.5%±1.2 (5 mg/kg group). The percentages of donor-derived peripheral blood leukocytes (PBL) at 2 weeks were 13.0%±4.6 (2.5 mg/kg group) and 12.7%±2.9 (5.0 mg/kg group), respectively (Table 4). When in vivo chemoselection was continued for up to 4 weeks after HSCT, the percentage of donor-derived BM cells was again found to be extremely high in both the 2.5 mg/kg group (95.3%±0.9) and the 5.0 mg/kg group (96.7%±1.2). Notably, the percentage of donor-derived PBL was significantly higher in the 5.0 mg/kg group (39.7%±3) in comparison to the 2.5 mg/kg group (29.9%±1.9) at 4 weeks (p<0.005), as well as significantly higher (p<0.0002) than in the groups receiving 2 weeks of chemoselection at either dose (Table 4). Thus, preconditioning with 10 mg/kg 6TG on Day −2 and Day 0, combined with on-going in vivo chemoselection with 5 mg/kg 6TG every 3 days for 4 weeks, yielded maximal levels of BM engraftment by HPRT-deficient donor cells as well as the highest levels of donor-derived PBL; this regimen was employed in further studies.
Table 4. Survival and engraftment after combined 6TG conditioning and in vivo chemoselection.
Treatment schedules for in vivo chemoselection are as indicated: In vivo chemoselection with 2.5 mg/kg 6TG for 2 weeks, 5.0 mg/kg 6TG for 2 weeks, 2.5 mg/kg 6TG for 4 weeks or 5.0 mg/kg for 4 weeks. All treatment groups showed 100% survival. Engraftment of HPRT-deficient male hematopoietic cells in HPRT-wt female recipient BM and PBL was determined by chromosome XY-FISH (mean% ± SD).
| Treatment | Survival (%) | Engraftment
|
|
|---|---|---|---|
| BM (%) | PBL (%) | ||
| 6TG 2.5 mg/kg × 2 weeks | 100 (n=3) | 89.3±1.7 | 13.0±4.6 |
| 6TG 5.0 mg/kg × 2 weeks | 100 (n=3) | 95.5±1.2 | 12.7±2.9 |
| 6TG 2.5 mg/kg × 4 weeks | 100 (n=3) | 95.3±0.9 | 29.9±1.9 |
| 6TG 5.0 mg/kg × 4 weeks | 100 (n=8) | 96.7±1.2 | 39.7±3.0 |
Combined 6TG conditioning and in vivo chemoselection results in long-term reconstitution of HPRT-deficient BM
The durability of engraftment by HPRT-deficient donor BM using the combined 6TG conditioning and in vivo chemoselection regimen established above was examined 4 months, 7 months, or 12 months after transplantation (i.e., 3 months, 6 months, and 11 months after the end of the 4-week in vivo chemoselection period, respectively). All transplanted animals (4 months: n=8; 7 months: n=6; 12 months: n=5) remained alive and well, showing no signs of any morbidity or discomfort, at all time points examined. Gross pathological and histological examination of these animals revealed no significant abnormalities.
Multi-lineage reconstitution of lymphohematopoiesis by donor-derived progenitor cells was evaluated 4 months after HSCT with combined 6TG preconditioning and in vivo chemoselection, i.e., 3 months after the end of the 4-week chemoselection period. Immunophenotyping of BM, PBL, thymus, and spleen was performed in a congeneic CD45.1/CD45.2 transplant setting using HPRT-deficient CD45.2 mice as donors and HPRT-wt CD45.1 mice as recipients (n=5). All hematopoietic tissues showed high engraftment of CD45.2+ donor cells at levels exceeding 75% of total bone marrow (Table 5). Immunophenotyping of the donor-derived CD45.2+ population showed that the relative percentages of T cells (CD4/CD8), B220+ cells, and macrophages/granulocytes (Mac-1/Gr1) were comparable to those of the treatment-naïve controls (Tables 2 and 3).
Table 5. Immunophenotypic analysis of hematopoietic tissues 4 months post-transplantation using 6TG conditioning and in vivo chemoselection regimen in a congeneic CD45.1/CD45.2 transplant model.
6TG conditioning and in vivo chemoselection was performed as described in text. Recipient BM and PBL were stained with the following rat anti–mouse antibodies: CD45.2-FITC (BM, PBL, T, S), CD4-PE (BM, PBL, T, S), CD8-APC (BM, PBL, T, S), Mac1/Gr1-PE (BM, PBL), B220-PerCP (BM, PBL, S), and examined by flow cytometry. Percentages of the indicated hematopoietic cell subpopulations are expressed as mean % ± SD of total CD45.2+ cells (n=5 per group).
| Cell population |
|
|||
|---|---|---|---|---|
| BM (%) | PBL (%) | Thymus (%) | Spleen (%) | |
| CD45.2+ | 76.1±9.3 | 73.8±5.2 | 89.7±8.3 | 65.9±5.2 |
| CD4+ | 3.2±1.3 | 17.0±3.0 | 10.4±2.8 | 22.6±3.6 |
| CD8+ | 3.0±1.3 | 12.7±0.9 | 4.3±0.8 | 14.9±2.1 |
| CD4+/CD8+ | 79.3±4.2 | |||
| B220+ | 29.3±6.3 | 45.4±3.8 | 58.2±8.6 | |
| Mac1+/Gr1+ | 80.68±6.1 | 28.2±3.3 | 14.1±1.6 | |
Engraftment levels were also evaluated by chromosome XY-FISH at the post-transplant 4-month, 7-month, and 12-month time points. Stable long-term reconstitution by donor-derived BM at high levels was observed at all time points, with engraftment levels of 97.7% ± 0.5% (4 months), 94.7% ± 1.9% (7 months), and 93.0% ± 0.8% (12 months), respectively, after transplantation (Fig. 5). Furthermore, the percentage of donor-derived PBL was significantly increased (67.4% ± 10.6% at 4 months, p=0.01; 73.3% ± 7.9% at 7 months, p<0.01; 73.0% ± 10.7% at 12 months, p<0.01) compared to that immediately following the 4-week 6TG selection period (39.7% ± 3.0%).
Figure 5. Long-term hematopoietic reconstitution after transplantation of HPRT-wt recipients with HPRT-deficient donor-derived BM using combined 6TG conditioning and chemoselection.
Bar graphs show the percentage of donor-derived cells in bone marrow (BM) and peripheral blood leukocytes (PBL) as determined by XY chromosome FISH analysis at 4 weeks (i.e., immediately after the end of the chemoselection period), and at 4 months, 7 months, and 12 months after transplantation, as indicated.
The relative percentages of CD4+ and CD8+ cells, B220+ cells, and Mac-1+/Gr1+ cells, as well as KLS (lin−/sca-1+/c-kit+) HSC, were determined in BM by immunophenotyping at 4 months, 7 months, and 12 months (Table 6), and compared to treatment-naïve female HPRT-wt (Table 2) and treatment-naïve male HPRT-deficient (Table 3) control mice. At all three time points, the relative percentage of each cell population was comparable to that in the controls, although KLS cells were elevated at 4 months and 7 months, and within the normal range at 12 months after HSCT. Thus, HPRT-deficient donor-derived marrow was able to achieve long-term reconstitution of normal hematopoiesis for at least 12 months post-transplantation.
Table 6. Immunophenotypic analysis of BM at 4 months, 7 months, and 12 months post-transplantation with HPRT-deficient BM using 6TG conditioning and in vivo chemoselection regimen.
At the indicated time points after HSCT with 6TG conditioning and in vivo chemoselection, recipient BM cells were stained with the following rat anti–mouse antibodies: CD45-FITC, CD4-PE, CD8-APC, Mac1/Gr1-PE, B220-PerCP, Sca-1-PE, and c-kit-FITC, and examined by flow cytometry. Percentages of the indicated hematopoietic cell subpopulations are expressed as mean % ± SD of total CD45+ cells.
| Cell population |
|
||
|---|---|---|---|
| 4 months BM (%) |
7 months BM (%) |
12 months BM (%) |
|
| CD45+ | 91.6±1.9 | 94.3±2.1 | 82.5±4.9 |
| CD4+ | 2.5±0.6 | 3.8±0.2 | 2.8±2.7 |
| CD8+ | 1.9±0.1 | 2.2±0.4 | 1.8±0.6 |
| B220+ | 51.8±2.4 | 32.9±3.8 | 11.5±2.8 |
| Mac1+/Gr1+ | 77.0±13.0 | 80.8±2.3 | 72.3±5.4 |
| KLS (HSC) | 7.8±1.0 | 7.0±1.1 | 1.8±0.6 |
Hematopoietic reconstitution of secondary recipients after serial transplantation of HPRT-deficient donor BM with combined 6TG conditioning and chemoselection
To further evaluate whether the optimized regimen combining 6TG conditioning with chemoselection selects for long-term repopulating HSCs, we next transplanted BM from primary recipients at 7 months post-transplantation into secondary recipients [27] using the same regimen. The secondary transplant recipients were then maintained for 3 months after the end of their 4-week course of 6TG in vivo chemoselection (Supplementary Figure S3). After 6TG conditioning and chemoselection, HPRT-deficient male donor cells that had engrafted primary recipients were able to serially repopulate secondary female recipients at high levels, (95.5% ± 1.1%, as determined by XY-FISH). Immunophenotypic analysis revealed that the percentages of all cell populations examined in BM and PBL of secondary recipients (Table 7) were again comparable to those of treatment-naïve controls (Tables 2 and 3).
Table 7. Immunophenotypic analysis of hematopoietic cells in secondary recipients after serial transplantation of HPRT-deficient BM with 6TG conditioning and in vivo chemoselection.
Serial transplantation with 6TG conditioning and in vivo chemoselection regimen was performed as described in Figure S3. Secondary recipient BM and PBL were stained with the following rat anti–mouse antibodies: CD45-FITC (BM, PBL), CD4-PE (BM, PBL), CD8-APC (BM, PBL), Mac1/Gr1-PE (BM, PBL), and B220-PerCP (BM, PBL), and examined by flow cytometry. Percentages of the indicated hematopoietic cell subpopulations are expressed as mean % ± SD of total CD45+ cells (n=6 per group).
| Cell population |
|
|
|---|---|---|
| BM (%) | PBL (%) | |
| CD45+ | 84.6±4.7 | 95.8±1.4 |
| CD4+ | 4.1±1.0 | 7.0±3.3 |
| CD8+ | 2.4±0.8 | 7.0±3.2 |
| B220+ | 19.1±2.9 | 33.3±12.3 |
| Mac1+/Gr1+ | 78.3±4.7 | 31.6±11.7 |
Discussion
We have developed an optimized regimen that employs 6TG as a single agent for pre-transplant conditioning as well as continued post transplant in vivo chemoselection of HPRT-deficient donor HSC. This combined 6TG conditioning and chemoselection strategy achieved efficient HSC engraftment with low overall toxicity, through a progressive and simultaneous replacement process, in which recipient hosts showed little if any distress and 100% survival, while their BM was rapidly and almost completely replaced by HPRT-deficient donor cells, consistently achieving ~95% engraftment by XY-FISH, and >75% by CD45.2 immunophenotyping. The percentage of donor-derived PBL increased significantly over time (4 and 7 months post-transplantation) compared to that immediately following the 4-week 6TG selection period. Residual recipient cells in PBL likely reflect a lower turnover of non-dividing mature cells in PBL. Stable long-term reconstitution of the BM was achieved in both primary and secondary recipients. Immunophenotyping analysis of BM, PBL, spleen, and thymus showed that, after long-term reconstitution, hematopoietic differentiation was unaffected by 6TG in vivo chemoselection.
These results also confirm our previous observation that, at appropriate concentrations, 6TG appears to induce selective myelotoxicity without any adverse effects on extra-hematopoietic tissues. HPRT is expressed at low levels in all somatic cells [30], and inherited loss of HPRT gives rise to Lesch-Nyhan syndrome [31] which manifests as severe mental retardation and behavioral abnormalities, as higher levels of HPRT expression and activity have been found in the central nervous system, particularly during neural development [32]. However, since fully differentiated neurons do not undergo replication, these higher levels do not translate into higher 6TG neurotoxicity in mature adults.
In previous work by Porter and DeGregori [22], total body irradiation of 4.5 Gy was used to achieve myeloablation, and 6TG was employed only for chemoselection at later time periods. In contrast, in our regimen, 6TG as a single agent fulfills the dual role of conditioning (cytoreduction of host BM) and chemoselective drug (amplification of donor BM). It is known that 6TG toxicity requires two rounds of DNA replication to result in apoptosis [33] and therefore shows a delayed effect. Our results also indicated a dose- and time-dependent myelotoxic effect of 6TG. Thus, conditioning with 10 mg/kg 6TG prior to transplantation may improve the outcome of 6TG in vivo chemoselection by compensating for the delay in 6TG myelotoxicity and providing an adequate niche for HSC at the time of transplantation.
In addition, Porter et al. employed 6TG at much lower doses ranging from 0.25 – 2 mg/kg over short periods of time, starting more than a month after transplantation, resulting in variable engraftment levels ranging from 5 – 50%. In our current study, we have confirmed that mice transplanted with Hprt-deficient BM can tolerate 6TG doses 5- to 40-fold higher, with administration of the chemoselective drug started immediately post-transplant, and continued over considerably longer selection periods.
In order to translate the use of the 6TG in vivo chemoselection strategy into a clinically feasible approach, it is necessary to develop methods to genetically engineer normal HSC to render them HPRT-deficient and thus 6TG-resistant. Furthermore, it should be emphasized that our current study was limited to employing this strategy for bone marrow transplantation in syngeneic mice, which models the situation in the autologous transplant setting. Whether this strategy will be equally useful in the allogeneic setting remains to be seen, as reduced-intensity conditioning regimens are being used routinely to circumvent the toxicity of traditional myeloablative regimens. Potential issues that may need to be addressed include the possibility of spontaneous 6TG resistance arising in leukemic cells after allogeneic transplantation for hematopoietic malignancies, and possible exacerbation of graft-vs-host disease when allogeneic donor cells are selectively amplified in vivo.
Therefore, on-going studies are aimed at translational application to ex vivo gene therapy in the autologous setting, employing both third-generation lentiviral vectors expressing different HPRT-targeted shRNA candidate sequences, as well as nucleofection of zinc finger nucleases (ZFN) targeting HPRT. In particular, the latter approach promises to be advantageous as only transient expression of the ZFN construct is needed to achieve permanent knockout of the target gene, thereby mitigating the potential for insertional mutagenesis as a result of the genetic engineering procedure. In this context, this strategy is unique in imparting a selective advantage to transplanted cells through an enzyme deficiency, rather than inserting a new transgene to achieve chemoresistance.
Acknowledgments
The authors thank Jessica Scholes (Flow Cytometry Core Facility, UCLA Broad Stem Cell Center) for expert technical assistance with flow cytometric analyses, and the staff of the UCLA Clinical and Molecular Cytogenetics Laboratory for expert technical assistance with FISH analyses.
Footnotes
Authorship
Contributions: K.H. designed and performed experiments, analyzed data, and wrote the manuscript; A.S. designed experiments and wrote the manuscript; N.R. and G.W.L. performed and analyzed histopathology; A.R.C. and N.A.L. performed experiments and analyzed data; J.A. designed, performed, and analyzed preliminary experiments; N.P.R: helped perform FISH studies and analyzed data; G.M.C. analyzed immunophenotyping data and critically revised the manuscript; R.H.S. organized the study and reviewed the manuscript; N.K. designed experiments, analyzed data, and wrote the manuscript. N.K. and R.H.S. should be considered co-senior authors owing to equal contribution to this study.
Conflict-of-interest disclosure:
J.A. is an employee of Pfizer Global Research and Development. The remaining authors declare no competing financial interests.
Support and financial disclosure declaration
This work was supported in part by the NIAID-funded UCLA Center for Biological Radioprotectors (U19 AI 067769) (N.K., R.H.S.), MNIT Foundation (N.K.), and California Institute for Regenerative Medicine (RS1-00402-01) (N.K.).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Biffi A, Cesani M. Human hematopoietic stem cells in gene therapy: pre-clinical and clinical issues. Curr Gene Ther. 2008;8:135–146. doi: 10.2174/156652308784049381. [DOI] [PubMed] [Google Scholar]
- 2.Persons DA, Nienhuis AW. In vivo selection to improve gene therapy of hematopoietic disorders. Curr Opin Mol Ther. 2002;4:491–498. [PubMed] [Google Scholar]
- 3.Williams DA, Hsieh K, DeSilva A, Mulligan RC. Protection of bone marrow transplant recipients from lethal doses of methotrexate by the generation of methotrexate-resistant bone marrow. J Exp Med. 1987;166:210–218. doi: 10.1084/jem.166.1.210. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Mickisch GH, Aksentijevich I, Schoenlein PV, et al. Transplantation of bone marrow cells from transgenic mice expressing the human MDR1 gene results in long-term protection against the myelosuppressive effect of chemotherapy in mice. Blood. 1992;79:1087–1093. [PubMed] [Google Scholar]
- 5.Sorrentino BP, Brandt SJ, Bodine D, et al. Selection of drug-resistant bone marrow cells in vivo after retroviral transfer of human MDR1. Science. 1992;257:99–103. doi: 10.1126/science.1352414. [DOI] [PubMed] [Google Scholar]
- 6.Zaboikin M, Srinivasakumar N, Schuening F. Gene therapy with drug resistance genes. Cancer Gene Ther. 2006;13:335–345. doi: 10.1038/sj.cgt.7700912. [DOI] [PubMed] [Google Scholar]
- 7.Southgate T, Fairbairn LJ. Genetic manipulation of drug sensitivity in haematopoietic cells. Expert Rev Mol Med. 2004;6:1–24. doi: 10.1017/S146239940400818X. [DOI] [PubMed] [Google Scholar]
- 8.Milsom MD, Williams DA. Live and let die: in vivo selection of gene-modified hematopoietic stem cells via MGMT-mediated chemoprotection. DNA Repair (Amst) 2007;6:1210–1221. doi: 10.1016/j.dnarep.2007.03.020. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Neff T, Beard BC, Kiem HP. Survival of the fittest: in vivo selection and stem cell gene therapy. Blood. 2006;107:1751–1760. doi: 10.1182/blood-2005-06-2335. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Schambach A, Baum C. Vector design for expression of O6-methylguanine-DNA methyltransferase in hematopoietic cells. DNA Repair (Amst) 2007;6:1187–1196. doi: 10.1016/j.dnarep.2007.03.017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Aubrecht J, Goad ME, Schiestl RH. Tissue specific toxicities of the anticancer drug 6-thioguanine is dependent on the Hprt status in transgenic mice. J Pharmacol Exp Ther. 1997;282:1102–1108. [PubMed] [Google Scholar]
- 12.Stout JT, Caskey CT. HPRT: gene structure, expression, and mutation. Annu Rev Genet. 1985;19:127–148. doi: 10.1146/annurev.ge.19.120185.001015. [DOI] [PubMed] [Google Scholar]
- 13.Fontenelle LJ, Henderson JF. An enzymatic basis for the inability of erythrocytes to synthesize purine ribonucleotides de novo. Biochim Biophys Acta. 1969;177:175–176. doi: 10.1016/0304-4165(69)90085-3. [DOI] [PubMed] [Google Scholar]
- 14.Holmsen H, Rozenberg MC. Adenine nucleotide metabolism of blood platelets. 3. Adenine phosphoribosyl transferase and nucleotide formation from exogenous adenine. Biochim Biophys Acta. 1968;157:266–279. [PubMed] [Google Scholar]
- 15.Lajtha LG, Vane JR. Dependence of bone marrow cells on the liver for purine supply. Nature. 1958;182:191–192. doi: 10.1038/182191a0. [DOI] [PubMed] [Google Scholar]
- 16.Lowy BA, Ramot B, London IM. The biosynthesis of adenosine triphosphate and guanosine triphosphate in the rabbit erythrocyte in vivo and in vitro. J Biol Chem. 1960;235:2920–2923. [PubMed] [Google Scholar]
- 17.Elion GB. The purine path to chemotherapy. Science. 1989;244:41–47. doi: 10.1126/science.2649979. [DOI] [PubMed] [Google Scholar]
- 18.Ansell JD, Samuel K, Whittingham DG, et al. Hypoxanthine phosphoribosyl transferase deficiency, haematopoiesis and fertility in the mouse. Development. 1991;112:489–498. doi: 10.1242/dev.112.2.489. [DOI] [PubMed] [Google Scholar]
- 19.Torres RJ, Puig JG. Hypoxanthine-guanine phosophoribosyltransferase (HPRT) deficiency: Lesch-Nyhan syndrome. Orphanet J Rare Dis. 2007;2:48. doi: 10.1186/1750-1172-2-48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.van der Zee SP, Schretlen ED, Monnens LA. Megaloblastic anaemia in the Lesch-Nyhan syndrome. Lancet. 1968;1:1427. doi: 10.1016/s0140-6736(68)92002-3. [DOI] [PubMed] [Google Scholar]
- 21.Seegmiller JE, Watanabe T, Shreier MH, Waldmann TA. Immunological aspects of purine metabolism. Adv Exp Med Biol. 1977;76A:412–433. doi: 10.1007/978-1-4613-4223-6_53. [DOI] [PubMed] [Google Scholar]
- 22.Porter CC, DeGregori J. Interfering RNA-mediated purine analog resistance for in vitro and in vivo cell selection. Blood. 2008;112:4466–4474. doi: 10.1182/blood-2008-03-146571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hooper M, Hardy K, Handyside A, Hunter S, Monk M. HPRT-deficient (Lesch-Nyhan) mouse embryos derived from germline colonization by cultured cells. Nature. 1987;326:292–295. doi: 10.1038/326292a0. [DOI] [PubMed] [Google Scholar]
- 24.Bryder D, Rossi DJ, Weissman IL. Hematopoietic stem cells: the paradigmatic tissue-specific stem cell. Am J Pathol. 2006;169:338–346. doi: 10.2353/ajpath.2006.060312. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Rondon G, Giralt S, Pereira M, et al. Analysis of chimerism following allogeneic bone marrow transplantation by fluorescent-in-situ hybridization. Leuk Lymphoma. 1997;25:463–467. doi: 10.3109/10428199709039033. [DOI] [PubMed] [Google Scholar]
- 26.Brazelton TR, Blau HM. Optimizing techniques for tracking transplanted stem cells in vivo. Stem Cells. 2005;23:1251–1265. doi: 10.1634/stemcells.2005-0149. [DOI] [PubMed] [Google Scholar]
- 27.Krause DS, Theise ND, Collector MI, et al. Multi-organ, multi-lineage engraftment by a single bone marrow-derived stem cell. Cell. 2001;105:369–377. doi: 10.1016/s0092-8674(01)00328-2. [DOI] [PubMed] [Google Scholar]
- 28.Olsen NJ, Kovacs WJ. Effects of androgens on T and B lymphocyte development. Immunol Res. 2001;23:281–288. doi: 10.1385/IR:23:2-3:281. [DOI] [PubMed] [Google Scholar]
- 29.Olsen NJ, Gu X, Kovacs WJ. Bone marrow stromal cells mediate androgenic suppression of B lymphocyte development. J Clin Invest. 2001;108:1697–1704. doi: 10.1172/JCI13183. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Kelley WN, Wyngaarden JB. Clinical syndromes associated with hypoxanthine guanine phosphoribosyltransferase deficiency. In: Stanbury JBJBW, Frederickson DS, Goldstein JL, Brown MS, editors. The Metabolic Basis of Inherited Disease. 5. New York: McGraw Hill; 1983. pp. 1115–1143. [Google Scholar]
- 31.Lesch M, Nyhan WL. A Familial Disorder of Uric Acid Metabolism and Central Nervous System Function. Am J Med. 1964;36:561–570. doi: 10.1016/0002-9343(64)90104-4. [DOI] [PubMed] [Google Scholar]
- 32.Lo YF, Palmour RM. Developmental expression of murine HPRT. I. Activities, heat stabilities, and electrophoretic mobilities in adult tissues. Biochem Genet. 1979;17:737–746. doi: 10.1007/BF00502132. [DOI] [PubMed] [Google Scholar]
- 33.Mojas N, Lopes M, Jiricny J. Mismatch repair-dependent processing of methylation damage gives rise to persistent single-stranded gaps in newly replicated DNA. Genes Dev. 2007;21:3342–3355. doi: 10.1101/gad.455407. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.An DS, Qin FX, Auyeung VC, et al. Optimization and functional effects of stable short hairpin RNA expression in primary human lymphocytes via lentiviral vectors. Mol Ther. 2006;14:494–504. doi: 10.1016/j.ymthe.2006.05.015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Grimm D, Streetz KL, Jopling CL, et al. Fatality in mice due to oversaturation of cellular microRNA/short hairpin RNA pathways. Nature. 2006;441:537–541. doi: 10.1038/nature04791. [DOI] [PubMed] [Google Scholar]