Protection of committed murine haemopoietic progenitors ... - Nature

Leukemia (1999) 13, 1776–1783  1999 Stockton Press All rights reserved 0887-6924/99 $15.00 http://www.stockton-press.co.uk/leu

Protection of committed murine haemopoietic progenitors against BCNU toxicity does not predict protection of primitive, multipotent spleen colony-forming cells – implications for chemoprotective gene therapy N Chinnasamy1,2,3 JA Rafferty1,2 LS Lashford4, D Chinnasamy1,3, GP Margison1, N Thatcher5, TM Dexter2,6 and LJ Fairbairn2 CRC Sections of 1Genome Damage and Repair and 2Haemopoietic Cell and Gene Therapeutics, Paterson Institute for Cancer Research, 4 Academic Unit of Paediatric Oncology, 5Department of Medical Oncology, Christie Hospital NHS Trust, Manchester, UK

The effect of expression of an O6-benzylguanine (O6-beG)resistant mutant (hATPA/GA) of human O6-alkylguanine-DNA alkyltransferase (ATase) on the in vivo toxicity and clastogenicity of the anti-tumour agent N,N⬘-bis(2-chloroethyl)-Nnitrosourea (BCNU) to murine bone marrow has been investigated. When compared with control animals, the bipotent granulocyte–macrophage colony-forming (GM-CFC) progenitor population of the hATPA/GA transduced mice were somewhat more resistant to BCNU (1.4-fold, P = 0.047) and this effect was more significant in the presence of the ATase inactivator O6beG (3.5-fold, P = 0.001). The polychromatic erythrocytes were also significantly protected against BCNU-induced clastogenicity both in the presence (P ⬍ 0.001) and absence of O6-beG (P ⬍ 0.05). The primitive, multipotent spleen colony-forming cells (CFU-S) in these animals also showed moderate (1.6-fold, P = 0.034) protection in the absence of O6-beG but in the presence of the inactivator they remained as sensitive to BCNU toxicity as those in the control animals (P = 0.133). This result contrasts with previous findings demonstrating significant hATPA/GAmediated, O6-beG-resistant protection against the toxicity and clastogenicity of a number of O6-alkylating agents, including temozolomide, fotemustine and chlorozotocin. The possibility that our strategy for protective gene therapy may be highly agent and cell-type specific is unexpected and has possible implications for clinical trials of this approach using BCNU or related agents. Keywords: chemoprotective gene therapy; BCNU; GM-CFC; CFUS; alkyltransferase; bone marrow stroma

Introduction Cytostatic anti-tumour therapies can result in both early doselimiting toxicity and late tissue injury. In the bone marrow acute toxicity frequently manifests as myelosuppression whilst chronic injury is more insidious and may result in the development of a secondary malignancy such as leukaemia.1 Such myelotoxicity constrains the dose and frequency of drug exposure. A possible solution to this problem is to enhance the resistance of sensitive tissues to these effects.2,3 Current clinical strategies include supporting haemopoietic function by administration of growth factors or by supplying progenitor cells harvested from the peripheral blood (PBPC) of patients prior to therapy.4,5 To date, reported growth factor or PBPC support does provide clinical benefit in terms of reduced toxicity but not to a level at which substantial increases in dose intensity could be considered. As an alternative strategy, we have been investigating the effects of the ex vivo transfer and in vivo expression of drug Correspondence: LJ Fairbairn, Paterson Institute for Cancer Research, Christie Hospital (NHS) Trust, Wilmslow Road, Manchester M20 4BX, UK; Fax: 44-161-446-3033 Current addresses: 3CGTB/NHGRI, National Institute of Health, Bethesda, MD 20892-1851, USA; and 6The Wellcome Trust, 183 Euston Road, London, NW1 2BE, UK. Received 1 May 1999; accepted 30 July 1999

resistance factors in normal, early haemopoietic progenitor cells. This genetic chemoprotection approach has a number of theoretical advantages over PBPC or growth factor support. First, with every round of treatment the number of drug-resistant progenitors is anticipated to increase as untransduced cells would remain sensitive to toxicity, thereby potentially facilitating drug dose escalation above currently marrow suppressive levels (or at least facilitating maintenance of dose intensity). Second, while present studies focus on protecting the haemopoietic system, it may ultimately be possible to reduce collateral toxicity in the next dose limiting tissues such as lung or gut in order to extend the general utility of this approach. Third, it may be possible to integrate the gene-based protection strategy with novel approaches to tumour sensitisation. Finally, the appropriate choice of resistance factor may allow protection against the mutagenic effects associated with certain classes of anti-tumour agent. There are a number of drug resistance functions that have been evaluated for their potential to enhance haemopoietic cell survival during chemotherapy. These include the efflux pump proteins (multiple drug resistance-1 and multi-drug resistance protein),6,7 aldehyde dehydrogenase,8,9 glutathionecytidine deaminase,11 dihydrofolate S-transferases,10 reductase12,13 and O6-alkylguanine-DNA-alkyltransferase (ATase).14,15 The latter protein, which confers resistance to the biological effects of O6-alkylating agents comprising the chloroethylnitrosoureas (eg N,N⬘-bis(2-chloroethyl)-N-nitrosourea (BCNU); 1-(2-chloroethyl)-3-cyclohexyl-1-nitrosourea (CCNU); fotemustine) and related methylating agents (eg temozolomide; dimethyltriazenoimidazolecarboxamide (DTIC); streptozocin; procarbazine), has been widely studied in cells and animal systems. These investigations have shown that the expression of ATase in cells which are normally deficient for this function produced a marked increase in cellular resistchromosome damage,17,20,21 ance to toxicity,16–20 16,20,22–25 25–31 mutagenesis and transformation as a consequence of enhanced repair of O6-alkylguanine (O6-alkG). In contrast, mice carrying homozygous knockout mutations for ATase have been demonstrated to be particularly sensitive to the biological effects of O6-alkylating agents.32,33 Recent efforts to improve the chemotherapeutic efficacy of O6-alkylating agents have centred on attempts to exploit the autoinactivating nature of the ATase reaction such that tumour sensitisation is achieved by pretreatment with substrate analogues such as O6-benzylguanine (O6-beG, reviewed in Ref. 34). In in vivo studies, O6-beG can dramatically potentiate the cytotoxicity of BCNU in various tumour xenograft bearing animals (eg Refs 35–37). However, we have recently demonstrated that exposure of primary human or murine bone marrow to O6-beG, in vitro and in vivo at concentrations that would be expected to deplete ATase activity in many tumours, rendered the haemopoietic progenitors more sensitive to killing by temozolomide.38,39

Ability of hATPA/GA expression to provide protection against BCNU N Chinnasamy et al

An approach to circumventing the dose limiting toxicity of O6-alkylating agents, whilst simultaneously sensitising tumours via O6-beG, would be to enhance the repair capacity of bone marrow for O6-alkG via expression of a mutant version of human ATase (hAT) that is both refractory to O6-beG inactivation and can efficiently repair O6-alkG in DNA. Increasing the expression of hAT40–45 or O6-beG-resistant mutants of hAT,46–50 either in murine bone marrow or in human primary haemopoetic cells, has been shown to afford some degree of protection against O6-alkylating agentinduced toxicity in the bipotent granulocyte–macrophage progenitor population. These observations have stimulated interest in a clinical trial to protect bone marrow against BCNUinduced myelosuppression, even though evidence of enhanced resistance to BCNU in multipotent haemopoietic progenitors has not yet been reported. Recently, we have investigated whether expressing a mutant of hAT (hATPA/GA)51,52 that is essentially resistant to O6-beG in bone marrow of mice can protect day 12 spleen colony-forming cells (CFU-S)53 from the toxic effects of temozolomide. The day 12 CFU-S, under certain conditions of transplantation, can exhibit long-term repopulating ability despite their relatively low self-maintenance capacity.53,54 Thus, while most day 12 CFU-S have only a transient reconstitution potential,53,54 they are nonetheless somewhat characteristic of the multipotent cells that represent the ideal target for genetic chemoprotection. These studies showed a reduction in toxicity even in the presence of O6-beG.49,50 To determine if this protection extended to chloroethylating agents we have, in the present study, exposed mice reconstituted with bone marrow virally transduced with a cDNA encoding hATPA/GA, to BCNU in the presence or absence of O6-beG. We observed protection of bipotent GM-CFC and polychromatic erythrocytes against BCNU toxicity and clastogenicity, respectively, but demonstrated that the CFU-S of hATPA/GA-transduced animals remained as sensitive to the toxicity of BCNU as those of control animals in the presence of O6-beG. The significance of these data for clinical trials of this approach are discussed. Materials and methods

Materials BCNU was purchased from Bristol-Myers Pharmaceuticals, Hounslow, UK. It was freshly dissolved in ethanol (25 mg/ml) and appropriately diluted in phosphate-buffered saline (PBS) for intraperitoneal (i.p.) injection into mice. O6-beG was a generous gift from Dr RS McElhinney and Prof B McMurry, Trinity College, Dublin, Ireland and was homogeneously suspended in corn oil using a TEFLON glass homogeniser (Scientific Laboratory Supplies, Nottingham, UK) for i.p. injection into mice. 5-Fluorouracil (5-FU; 250 mg/ml solution) was purchased from David Bull Laboratories, Warwick, UK. Recombinant human IL-6 was from Sandoz, Basle, Switzerland; recombinant murine IL-3 was from R&D Systems, Abingdon, UK; and recombinant rat stem cell factor was from Amgen, Thousand Oaks, CA, USA.

Producer cells The LhATPA/GA vector which expresses the human hATPA/GA protein52 (driven by Moloney murine leukemia virus long terminal repeat) was constructed by cloning the

appropriate cDNA into a BamHI site in the retroviral vector pLX derived from the LN series of vectors.55 The above mentioned vector construct and pcDNA3neo (10:1) were cotransfected into the ecotropic producer cells GP + E86 using Lipofectin (Life Technologies, Paisley, UK). G418-resistant LhATPA/GA-positive clones were cocultured with amphotropic GP + envAM12 cells for 7 days in a ‘ping-pong’ method to increase the viral titre. Amphotropic producer cells were then selected by culturing in the presence of 200 ␮g/ml hygromycin for 7 days to eliminate the ecotropic producers. The viral titre of supernatant from the amphotropic cells was tested using NIH3T3 cells and this indicated a titre in excess of 5 × 105 infectious particles/ml.

Animal studies B6D2F1 male donor mice (9–12 weeks old) were treated with 150 mg/kg 5-FU by intravenous injection. Two days later bone marrow cells were harvested from the femurs and cultured for 2 days at 37°C in DMEM supplemented with 20% foetal calf serum, 0.1% BSA, 2 mM glutamine, 10 ng/ml recombinant murine IL-3, 100 ng/ml recombinant rat stem cell factor and 200 U/ml recombinant human IL-6. The bone marrow cells were scraped from the culture flasks and overlayed on either GP+envAM 12 (mock), or LhATPA/GA retroviral producer cells for 2 days in the same medium with the addition of 4 ␮g/ml polybrene. Supernatant cells were then collected, washed and used for cytospin preparation, GM-CFC colony plating in methylcellulose for PCR analysis and transplantation into 8- to 10-week-old bone marrow ablated (15.2 Gy, Cobalt 60 source for 16 h female B6D2F1 mice (5 × 106 cells per mouse). To assess the gene transduction frequency among CFU-S, 10 recipient mice were transplanted with 1 × 106 cells each. The mice were sacrificed 12 days later and either spleens were fixed in 70% ethanol for immunocytochemical analysis or individual spleen colonies were carefully dissected out for DNA isolation and subsequent PCR analysis. One month after transplantation groups of five mice were treated with BCNU (5 mg/kg, i.p. in 3% ethanol in PBS) either alone or 2 h after receiving a single dose of O6-beG (30 mg/kg, i.p. in corn oil): 24 h later the treated mice were sacrificed by cervical dislocation and femoral bone marrow cells were harvested and used for CFU-S and GM-CFC analysis as described previously.39 The mean numbers of GM-CFC in untreated mice were 17 850 ± 2930 and 14 075 ± 3250 per femur in mock control and hATPA/GA mice, respectively. The mean numbers of CFU-S in untreated mice were 867 ± 99 and 897 ± 18 per femur in mock control and hATPA/GA mice, respectively.

PCR analysis PCR analyses were carried out on individual GM-CFC colonies (n = 20) picked from methylcellulose and DNA isolated from 18 day 12 spleen colonies. Using the primers 5⬘ATGGACAAGGATTGTGAAATGAAACG3⬘ and 5⬘TCAGTTTCGGCCAGCAGGCGG3⬘, a PCR product of 624 bp was amplified corresponding to the hAT cDNA. A DNA fragment of the murine, Y-chromosome-specific Sry gene was amplified using the primers 5⬘AAGCGCCCCATGAATGCATT3⬘ and 5⬘TCCCAGCTGCTTGCTGATCT3⬘ to detect male (donor) cells in the female recipients based on the presence of a 105-bp PCR product.56

1777


1778

Immunocytochemistry Cytospin preparations of transduced bone marrow cells and 3-␮m sections of day 12 spleen colonies were stained with a rabbit polyclonal antibody for human ATase which recognises hATPA/GA using a streptavidin biotin peroxidase complex detection system (Vector Laboratories, Peterborough, UK) in a modified procedure as previously described.57

Micronucleus test Groups of five control and LhATPA/GA transduced mice were treated with a single dose of BCNU (0.75 mg/kg i.p. in 3% ethanol in PBS) either alone or 2 h after receiving a single dose of O6-beG (30 mg/kg i.p. in corn oil). The treated mice were sacrificed 24 h later by cervical dislocation and femoral marrow cells were flushed into foetal calf serum. Bone marrow smears were prepared from the cell pellets, fixed in methanol for 10 min and subsequently stained with acridine orange as described.58 The slides were scored independently by two separate investigators for micronucleated polychromatic erythrocytes (MPCE) among 2000 polychromatic erythrocytes (PCE) and results expressed as MPCE/1000 PCE.

Statistical analysis The unpaired Student’s t-test was used to compare the difference between the control and experimental animals. Results

Efficiency of transduction of primary murine bone marrow progenitors The DNA from day 12 spleen colonies was used to determine if the haemopoiesis in the study animals was derived from the male donor marrow used for reconstitution. A PCR product derived from a murine Y-chromosome-specific sequence was demonstrated in all spleen colonies analysed confirming donor haemopoiesis. The same DNA samples were examined for the presence of the hATPA/GA cDNA and of 18 colonies analysed from the five recipient mice, nine proved positive for this sequence, indicating that about half of the reconstituted marrow arose from transduced cells. A similar transduction efficiency was calculated via PCR analysis of GM-CFC DNA isolated from the ex vivo-transduced marrow.

Immunocytochemical analysis of ATase expression in bone marrow and spleen colony cells Cytospin preparations of in vitro-transduced total murine bone marrow were analysed for evidence of ATase expression using a highly specific polyclonal antibody raised against hAT which does not cross-react with murine ATase. Bone marrow that had been co-cultured with hATPA/GA producers clearly showed the presence of nuclear staining antigen in about 40– 45% of the cells (Figure 1A). In comparison, bone marrow cultured with naive packaging cells showed no evidence of discrete antibody localisation, either in the nucleus or otherwise (Figure 1B). The expression of ATase in the day 12 spleen colonies was

also investigated and about 50% of colonies derived from mice reconstituted with hATPA/GA-transduced bone marrow (Figure 1C), were found to contain high levels of ATase protein. These data are consistent with the transduction frequency as determined by PCR. As expected, when detected the protein was localised mainly in the nucleus of the spleen colony cells.

In vivo toxicity of BCNU in hATPA/GA-transduced and mock-transduced bone marrow progenitors Mice reconstituted with hATPA/GA-transduced or mock-transduced bone marrow were treated with BCNU (5 mg/kg) either alone or in combination with O6-beG (30 mg/kg) approximately 5 weeks post-transplant. These treatments had been previously demonstrated to result in significant toxicity to murine bone marrow.39 The effect of in vivo drug treatment on the colony-forming ability of the treated marrows was assessed by comparison of the levels of CFU-S and GM-CFC in the hATPA/GA and mock transduced populations of mice. The GM-CFC levels in the mock-transduced animals treated with BCNU were reduced to 68% and 8% in the absence and presence of O6-beG, respectively (Figure 2). In the hATPA/GAtransduced mice, a modest protective effect against BCNU alone was seen in the GM-CFC, which was significant at the 5% level (P = 0.047) and equated to a 1.4-fold increased resistance over controls. In the presence of O6-beG, however, a more pronounced protective effect (3.5-fold, P = 0.001) was observed. In the CFU-S assayed from the mock-transduced animals, toxicity was also observed and pretreatment with O6-beG significantly reduced CFU-S numbers from 38% (BCNU alone) to 5% of those in untreated controls (Figure 3). As with the GM-CFC population, expression of hATPA/GA conferred modest (1.6-fold, P = 0.034) protection to the CFU-S against BCNU alone. However, this survival advantage was lost when the animals were exposed to O6-beG prior to BCNU treatment (P = 0.133).

In vivo clastogenicity of BCNU in transduced and mock-transduced polychromatic erythrocytes Five weeks post reconstitution, both hATPA/GA and mocktransduced mice were treated with a dose of BCNU or O6beG plus BCNU that was known from previous studies to be highly clastogenic to mouse bone marrow polychromatic erythrocytes.39 In the mock-transduced controls, there was a significant elevation in micronucleus frequency above that of the no treatment controls and this was further increased by pretreatment with O6-beG (Figure 4). In those animals reconstituted with hATPA/GA-transduced marrow there was a small but significant (P ⬍ 0.05) reduction in the number of micronuclei induced relative to the normal controls. The difference between the two groups, however, was more pronounced when mice were exposed to a combination of O6-beG and BCNU. In this case there was a 50% reduction (P ⬍ 0.001) in the micronucleus frequency in the animals reconstituted with the transduced compared with the mock-transduced marrow (again consistent with the observed transduction frequency).


1779

Figure 1 Immunohistochemical analysis of hATPA/GA-transduced (A, C) or mock-transduced (B, D) bone marrow (A, B) or spleen colonies (C, D) using the anti-human ATase antibody.

Figure 2 Survival of GM-CFC from mock-transduced (solid bars) or hATPA/GA-transduced (open bars) mice following exposure in vivo to either BCNU (5 mg/kg) or to the combination of BCNU and O6beG (30 mg/kg). Data shown are the mean of five observations ± the standard error of the mean. Significance by Student’s t-test comparing mock vs transduced means for the same treatment protocol; *P ⬍ 0.05 and ***P ⬍ 0.001.

Figure 3 Survival of CFU-S from mock-transduced (solid bars) or hATPA/GA-transduced (open bars) mice following exposure in vivo to either BCNU (5 mg/kg) or to the combination of BCNU and O6beG (30 mg/kg). Data shown are the mean of five observations ± the standard error of the mean. Significance by Student’s t-test comparing mock vs transduced means for the same treatment protocol; *P ⬍ 0.05.

Discussion

bone marrow progenitors, both in vitro and in vivo, against the toxicity of a range of O6-alkylating agents has been demonstrated.40–45,59 Some of these studies have made use of mutant ATases, which exhibit various degrees of resistance to the tumour sensitiser O6-beG,46–50 and have shown that haemopoietic cell protection is maintained under conditions

A number of groups are currently investigating the utility of gene transfer of ATase for the protection of bone marrow from collateral toxicity during O6-alkylating agent chemotherapy. ATase-mediated protection of lineage-specific or bipotent


1780

Figure 4 Frequency of micronucleated polychromatic erythrocytes (PCE) in bone marrow of mock-transduced (solid bars) or hATPA/GAtransduced (open bars) mice following exposure in vivo to either BCNU (0.75 mg/kg) or to the combination of BCNU (0.75 mg/kg) and O6-beG (30 mg/kg). Data shown are the mean of five observations ± the standard error of the mean. Significance by Student’s t-test comparing mock vs transduced means for the same treatment protocol; * P ⬍ 0.05 and ***P ⬍ 0.001.

where an otherwise resistant tumour might be expected to be sensitised to treatment. Despite the fact that this strategy is often referred to as stem cell protection, little direct qualitative evaluation of any protective effects that may occur in primitive haemopoietic cells has been undertaken. This is of particular importance if the strategy is to provide long-term protection against toxicity or against mutagenic, clastogenic and carcinogenic effects. To address this issue we have made use of the multipotent day 12 CFU-S population as a surrogate marker of primitive cells in mouse bone marrow.53 We have previously shown significant in vivo protection of day 12 CFU-S against the toxic effects of the methylating agent temozolomide, in combination with O6-beG, following reconstitution of mice with bone marrow which had been induced to express hATPA/GA as a result of retroviral gene transfer.49 In the present study, we have investigated the ability of hATPA/GA expression to provide protection against BCNU, which has been used in a number of preclinical studies by other workers and is a potential agent for use in a clinical trial of ATase-mediated myeloprotection. Following gene transfer, transduction frequencies of around 50% were seen in bone marrow GM-CFC and spleen colonies, and this correlated well with the frequency of ATase-positive cells/colonies detected by immunocytochemical analysis of bone marrow and spleen colonies derived from reconstituted mice. When the in vivo sensitivity of mock-transduced and hATPA/GA-transduced bone marrow to the cytotoxic effects of BCNU was tested, significant protection of the GMCFC compartment was seen both in the absence and presence of O6-beG. The size of the protective effect in committed progenitors (1.4- to 3.5-fold) was similar to that previously reported by others following transduction of bone marrow with wild-type or single mutant forms of hAT.40,42,43,46,59

Given this effect and from our experience with temozolomide and mitozolomide,49,50 we expected to see a similar protective effect in the more primitive day 12 CFU-S compartment. However, whilst a significant effect on survival was seen in the absence of O6-beG there was no protective effect in the presence of O6-beG. The reason for this lack of correlation between O6-beGresistant protection of the GM-CFC and CFU-S compartments is not obvious. Clearly, transduction of CFU-S was achieved in the original bone marrow cultures, and expression of hATPA/GA was detected in day 12 spleen colonies following reconstitution of mice. Furthermore the protection observed in the granulocyte/macrophage and erythroid lineages, 5 weeks after transplantation clearly demonstrates transduction of, and expression in, the progeny of primitive haemopoietic progenitors. One possible explanation might be that, whilst day 12 CFU-S were transduced in the original cultures, even more primitive (stem) cells, which would be expected to give rise to the day 12 CFU-S assayed after 5 weeks either were not transduced or alternatively did not express sufficient hATPA/GA. The former of these two possibilities is clearly untrue, since mice from the same cohort of animals as those treated with BCNU exhibited significant resistance of the d12 CFU-S population to two other O6-alkylating agents, temozolomide and mitozolomide.49,50 However, given the stochiometric, autoinactivating nature of ATase, excess damage in genomic DNA can lead to depletion of the repair capacity of cells and, if sufficient extra repair protein is not synthesised prior to cell division to complete the DNA repair process, cell death will ensue. Chloroethylating agents such as BCNU, form irreversible cross-links in DNA within about 6 h of exposure of cells to drug.60,61 Thus one possible explanation might be that the CFU-S express hATPA/GA, but at insufficient levels to effect repair of all of the initial damage caused by BCNU exposure and that fixation of damage leads to death despite further hATPA/GA synthesis. This clearly cannot be the case since, in the absence of O6beG, a protective effect was seen in the day 12 CFU-S which was at least as good as that seen in the GM-CFC and thus the levels of hATPA/GA in the CFU-S population must be adequate to confer a survival advantage following BCNU exposure. Only in the presence of O6-beG is the protective effect of hATPA/GA expression in CFU-S lost. This is consistent with inactivation of an O6-beG-sensitive (eg wild-type) form of hAT prior to BCNU treatment of the animals. In a previous study, we have examined the effect of pretreatment of normal mice with O6-beG on the toxicity of either temozolomide or BCNU against GM-CFC and CFU-S. For temozolomide, the extent of sensitisation of both the GM-CFC and CFU-S populations by O6-beG was the same (1.63- and 1.68fold, respectively). However, when the effect of O6-beG pretreatment on BCNU-induced cell loss was assessed, it was noted that the effect on the CFU-S population (8.25-fold sensitisation) was much greater than that on the GM-CFC population (4.57-fold sensitisation). This result seemed consistent with the inactivation of wild-type hAT in the CFU-S population, followed by fixation of BCNU-induced DNA damage prior to resynthesis. In the current studies, however, the mutant form of hAT used has been previously demonstrated to be resistant to inactivation by O6-beG and to confer O6beG-insensitive protection against O6-alkylating agents to human and murine haemopoietic cells.48,49,52 Furthermore, in this study the GM-CFC compartment of hATPA/GA-transduced mice showed a highly significant protective effect against the combination of BCNU and O6-beG. Thus a direct


effect of O6-beG/BCNU on the hATPA/GA-transduced CFU-S is unlikely. A similar argument also discounts the effects both of other DNA lesions produced by BCNU, or of damage to other cellular components (eg carbomylation of proteins62) either of which may be potentially lethal but not repairable by hAT: any effect on CFU-S in the presence of O6-beG, should also be evident in the absence of O6-beG. All of the above leads us to the conclusion that the lack of protection of hATPA/GA-transduced CFU-S from BCNU toxicity following pretreatment with O6-beG is likely to be a function of inactivation of wild-type hAT in a cell type other than the CFU-S itself. One possible explanation for this effect may lie in the nature of the CFU-S assay, which is not strictly a measure of cell viability, but is a functional measure of the ability of cells to proliferate and differentiate after seeding within the spleens of recipient mice. The bone marrow microenvironment plays a crucial role in regulating haemopoietic cell production.63 BCNU has been demonstrated to cause quantitative changes in bone marrow stromal matrix components64 and to impair the capacity of haemopoietic stroma to support haemopoiesis in vitro.65 Similar effects to those described for BCNU have been seen in vivo following treatment of mice with 239Pu,66 ␥-irradiation67 or busulphan.68 That is, a defect in stromal function leading to a reduced capacity to support CFU-S survival/proliferation. Thus it may be possible that a detrimental effect of BCNU on the bone marrow microenvironment is exacerbated by pretreatment with O6-beG and that this leads to loss of functional CFU-S, either through differentiation or through commitment to apoptosis as a result of insufficient levels of growth/survival factors. The issue as to whether the murine bone marrow microenvironment following transplantation is derived from donor or host is not fully resolved. However, the cumulative data published thus far suggest that the endothelial and fibroblast populations are likely to be of host origin.69–71 Moreover, in the mice used in these experiments, reconstitution of the bone marrow was achieved using the non-adherent fraction of cells collected following co-cultivation with retroviral producers. It would seem unlikely, therefore, that many (adherent) fibroblast or endothelial cells were transferred and that this portion of the microenvironment would be host derived. Thus in hATPA/GA-transduced mice, any ATase activity in the (host-derived) fibroblast/endothelial portion of the bone marrow stroma would be wild type and thus susceptible to inactivation by O6-beG. As a consequence of this, these components of the bone marrow microenvironment may be sensitised to BCNU and thus loss of hATPA/GA-transduced CFU-S following treatment with the combination of O6-beG and BCNU may reflect increased damage to the bone marrow microenvironment and not direct toxicity of BCNU on the CFU-S compartment (although it will clearly be important to investigate this further using appropriate in vitro assays). Similar evidence of host origin of the bone marrow stroma has also been seen in humans following allogeneic transplantation.72–75 Given this, it seems unlikely that ex vivo transfer of drug resistance genes to bone marrow will confer full chemoprotection to the bone marrow stroma and it is possible that exacerbated microenvironment damage may also lead to a reduction in stem cell protection in patients undergoing clinical trial. The above data indicate that the protective effects seen thus far in pre-clinical studies may not be entirely predictive of the ability or otherwise to provide genetic protection of stem cells with ATase, when the active drug is BCNU. If this is the case, and a similar senario applies in humans, then clinical trials of

this approach in combination with O6-beG/BCNU may show only short-term protective effects, with longer-term toxicities unaffected. This could limit the ability to provide sustained marrow recovery in the face of repeated, dose-intensive chemotherapy. It remains to be seen whether the observations reported here with BCNU also apply to other O6-alkylating agents. Certainly our experience with temozolomide is that good protection of the CFU-S compartment from the combination of this agent and O6-beG is seen following transfer of the hATPA/GA gene into bone marrow. We have also seen evidence of protection of CFU-S against the toxicity of the cross-linking imidazotriazine, mitozolomide, both in the presence and absence of O6-beG.50 However, there is as yet no evidence of protection of CFU-S from the effects of any chloroethylnitrosourea in combination with O6-beG. The BCNUrelated agent, CCNU is currently being used in a clinical trial of ATase-mediated chemoprotection.76 This study makes use of the wild-type hAT gene, without O6-beG and thus effects on primitive cells may not prove problematic. However, since the next logical extension of this trial would be to incorporate a mutant ATase with O6-beG sensitisation of tumours, it would seem prudent to investigate whether a protective effect in CFU-S can be demonstrated. These studies are in hand.

Acknowledgements We would like to thank Mrs L Woolford and Ms G Foster for technical assistance. This work was supported by grants from the Medical Research Council and the Cancer Research Campaign.

References 1 Boffetta P, Kaldor JM. Secondary malignancies following cancer chemotherapy. Acta Oncol 1994; 33: 591–598. 2 Rafferty JA, Hickson I, Chinnasamy N, Lashford LS, Margison GP, Dexter TM, Fairbairn LJ. Chemoprotection of normal tissues by transfer of drug resistance genes. Cancer Metastasis Rev 1996; 15: 365–383. 3 Baum C, Margison G, Eckert HG, Fairbairn LJ, Ostertag W, Rafferty JA. Gene-transfer to augment the therapeutic index of anticancer chemotherapy. Gene Therapy 1996; 3: 1–3. 4 Crowther D, Scarffe JH, Howell A, Thatcher N, Bronchud M, Steward WP, Testa N, Dexter M. Growth factor-assisted chemotherapy – the Manchester experience. Ciba Found Symp 1990; 148: 201–210. 5 Boogaerts MA, Brugger W, Carella AM, Cortes Funes H, Fibbe WE, Hows J, Khayat D, Linch DC, Link H, Moore MA, Testa NG. Peripheral blood progenitor cell transplantation: where do we stand? Chairman’s Summary of the European School of Oncology Task Force meeting Peripheral Blood Progenitor Cells held September 29–30, 1995. Ann Oncol 1996; 2 (Suppl. 7): 1–4. 6 D’ Hondt V, Caruso M, Bank A. Retrovirus-mediated gene transfer of the multidrug resistance-associated protein (MRP) cDNA protects cells from chemotherapeutic agents. Hum Gene Ther 1997; 8: 1745–1751. 7 Goldstein LJ, Pastan I, Gottesman MM. Multidrug resistance in human cancer. Crit Rev Oncol Hematol 1992; 12: 243–253. 8 Moreb J, Schweder M, Suresh A, Zucali JR. Overexpression of the human aldehyde dehydrogenase class I results in increased resistance to 4-hydroperoxycyclophosphamide. Cancer Gene Ther 1996; 3: 24–30. 9 Moreb JS, Schweder M, Gray B, Zucali J, Zori R. In vitro selection for K562 cells with higher retrovirally mediated copy number of aldehyde dehydrogenase class-1 and higher resistance to 4-hydroperoxycyclophosphamide. Hum Gene Ther 1998; 9: 611–619. 10 Letourneau S, Greenbaum M, Cournoyer D. Retrovirus-mediated gene transfer of rat glutathione S-transferase Yc confers in vitro

1781


1782

11

12

13 14

15 16

17 18

19

20

21

22

23

24

25

26

27

28

resistance to alkylating agents in human leukemia cells and in clonogenic mouse hematopoietic progenitor cells. Hum Gene Ther 1996; 7: 831–840. Momparler RL, Eliopoulos N, Bovenzi V, Letourneau S, Greenbaum M, Cournoyer D. Resistance to cytosine arabinoside by retrovirally mediated gene transfer of human cytidine deaminase into murine fibroblast and hematopoietic cells. Cancer Gene Ther 1996; 3: 331–338. Flasshove M, Banerjee D, Bertino JR, Moore MA. Increased resistance to methotrexate in human hematopoietic cells after gene transfer of the Ser31 DHFR mutant. Leukemia 1995; 9 (Suppl. 1): S34–S37. Spencer HT, Sleep SE, Rehg JE, Blakley RL, Sorrentino BP. A gene transfer strategy for making bone marrow cells resistant to trimetrexate. Blood 1996; 87: 2579–2587. Margison GP, O’Connor PJ. Biological consequences of reactions with DNA: role of specific lesions. In: Cooper CS, Grover PL (eds). Handbook of Experimental Pharmacology. Springer-Verlag: Berlin, 1990, pp 547–557. Pegg AE. Mammalian O6-alkylguanine-DNA alkyltransferase: regulation and importance in response to alkylating carcinogenic and therapeutic agents. Cancer Res 1990; 50: 6119–6129. Brennand J, Margison GP. Reduction of the toxicity and mutagenicity of alkylating agents in mammalian cells harboring the Escherichia coli alkyltransferase gene. Proc Natl Acad Sci USA 1986; 83: 6292–6296. Samson L, Derfler B, Waldstein EA. Suppression of human DNA alkylation-repair defects by Escherichia coli DNA-repair genes. Proc Natl Acad Sci USA 1986; 83: 5607–5610. Barrows LR, Borchers AH, Paxton MB. Transfectant CHO cells display expressing O6-alkylguanine-DNA-alkyltransferase increased resistance to DNA damage other than O6-guanine alkylation. Carcinogenesis 1987; 8: 1853–1859. Jelinek J, Kleibl K, Dexter TM, Margison GP. Transfection of murine multi-potent haemopoietic stem cells with an E. coli DNA alkyltransferase gene confers resistance to the toxic effects of alkylating agents. Carcinogenesis 1988; 9: 81–87. Kaina B, Fritz G, Mitra S, Coquerelle T. Transfection and expression of human O6-methylguanine-DNA methyltransferase (MGMT) cDNA in Chinese hamster cells: the role of MGMT in protection against the genotoxic effects of alkylating agents. Carcinogenesis 1991; 12: 1857–1867. White GR, Ockey CH, Brennand J, Margison GP. Chinese hamster cells harbouring the Escherichia coli O6-alkylguanine alkyltransferase gene are less susceptible to sister chromatid exchange induction and chromosome damage by methylating agents. Carcinogenesis 1986; 7: 2077–2080. Abril N, Hera C, Alejandre E, Rafferty JA, Margison GP, Pueyo C. Effect of ogt expression on mutation induction by methyl-, ethyland propylmethanesulphonate in Escherichia coli K12 strains. Mol Gen Genet 1994; 242: 744–748. Aquilina G, Biondo R, Dogliotti E, Meuth M, Bignami M. Expression of the endogenous O6-methylguanine-DNA-methyltransferase protects Chinese hamster ovary cells from spontaneous G:C to A:T transitions. Cancer Res 1992; 52: 6471–6475. Lukash LL, Boldt J, Pegg AE, Dolan ME, Maher VM, McCormick JJ. Effect of O6-alkylguanine-DNA alkyltransferase on the frequency and spectrum of mutations induced by N-methyl-N⬘-nitroN-nitrosoguanidine in the HPRT gene of diploid human fibroblasts. Mutat Res 1991; 250: 397–409. von Hofe E, Fairbairn L, Margison GP. Relationship between O6alkylguanine-DNA alkyltransferase activity and N-methyl-N⬘nitro-N-nitrosoguanidine-induced mutation, transformation, and cytotoxicity in C3H/10T1/2 cells expressing exogenous alkyltransferase genes. Proc Natl Acad Sci USA 1992; 89: 11199–11203. Zaidi NH, Allay E, Ayi TC, Li BF, Dumenco LL, Sy MS, Gerson SL. The immature thymocyte is protected from N-methylnitrosoureainduced lymphoma by the human MGMT-CD2 transgene. Carcinogenesis 1995; 16: 1047–1053. Liu L, Allay E, Dumenco LL, Gerson SL. Rapid repair of O6-methylguanine-DNA adducts protects transgenic mice from N-methylnitrosourea-induced thymic lymphomas. Cancer Res 1994; 54: 4648–4652. Dumenco LL, Allay E, Norton K, Gerson SL. The prevention of

29

30

31

32 33

34 35

36

37

38

39

40

41

42

43

44

45

46

thymic lymphomas in transgenic mice by human O6-alkylguanineDNA alkyltransferase. Science 1993; 259: 219–222. Nakatsuru Y, Matsukuma S, Nemoto N, Sugano H, Sekiguchi M, Ishikawa T. O6-methylguanine-DNA methyltransferase protects against nitrosamine-induced hepatocarcinogenesis. Proc Natl Acad Sci USA 1993; 90: 6468–6472. Becker K, Gregel CM, Kaina B. The DNA repair protein O6methylguanine-DNA methyltransferase protects against skin tumor formation induced by antineoplastic chloroethylnitrosourea. Cancer Res 1997; 57: 3335–3338. Becker K, Dosch J, Gregel CM, Martin BA, Kaina B. Targeted expression of human O6-methylguanine-DNA methyltransferase (MGMT) in transgenic mice protects against tumor initiation in two-stage skin carcinogenesis. Cancer Res 1996; 56: 3244–3249. Sakumi K, Shiraishi A, Shimizu S, Tsuzuki T, Ishikawa T, Sekiguchi M. Methylnitrosourea-induced tumorigenesis in MGMT gene knockout mice. Cancer Res 1997; 57: 2415–2418. Kawate H, Sakumi K, Tsuzuki T, Nakatsuru Y, Ishikawa T, Takahashi S, Takano H, Noda T, Sekiguchi M. Separation of killing and tumorigenic effects of an alkylating agent in mice defective in two of the DNA repair genes. Proc Natl Acad Sci USA 1998; 95: 5116–5120. Dolan ME, Pegg AE. O6-benzylguanine and its role in chemotherapy. Clin Cancer Res 1997; 3: 837–847. Friedman HS, Dolan ME, Moschel RC, Pegg AE, Felker GM, Rich J, Bigner DD, Schold SC Jr. Enhancement of nitrosourea activity in medulloblastoma and glioblastoma multiforme (published erratum appears in J Natl Cancer Inst 1994 Jul 6; 86 (13): 1027). J Natl Cancer Inst 1992; 84: 1926–1931. Dolan ME, Pegg AE, Moschel RC, Grindey GB. Effect of O6-benzylguanine on the sensitivity of human colon tumor xenografts to 1,3-bis(2-chloroethyl)-1-nitrosourea (BCNU). Biochem Pharmacol 1993; 46: 285–290. Mitchell RB, Moschel RC, Dolan ME. Effect of O6-benzylguanine on the sensitivity of human tumor xenografts to 1,3-bis(2chloroethyl)-1-nitrosourea and on DNA interstrand cross-link formation. Cancer Res 1992; 52: 1171–1175. Fairbairn LJ, Watson AJ, Rafferty JA, Elder RH, Margison GP. O6benzylguanine increases the sensitivity of human primary bone marrow cells to the cytotoxic effects of temozolomide. Exp Hematol 1995; 23: 112–116. Chinnasamy N, Rafferty JA, Hickson I, Ashby J, Tinwell H, Margison GP, Dexter TM, Fairbairn LJ. O6-benzylguanine potentiates the in vivo toxicity and clastogenicity of temozolomide and BCNU in mouse bone marrow. Blood 1997; 89: 1566–1573. Allay JA, Dumenco LL, Koc ON, Liu L, Gerson SL. Retroviral transduction and expression of the human alkyltransferase cDNA provides nitrosourea resistance to hematopoietic cells. Blood 1995; 85: 3342–3351. Allay JA, Koc ON, Davis BM, Gerson SL. Retroviral-mediated gene transduction of human alkyltransferase complementary-DNA confers nitrosourea resistance to human hematopoietic progenitors. Clin Cancer Res 1996; 2: 1353–1359. Maze R, Carney JP, Kelley MR, Glassner BJ, Williams DA, Samson L. Increasing DNA-repair methyltransferase levels via bone-marrow stem-cell transduction rescues mice from the toxic effects of 1,3-bis (2-chloroethyl)-1-nitrosourea, a chemotherapeutic alkylating agent. Proc Natl Acad Sci USA 1996; 93: 206–210. Moritz T, Mackay W, Glassner BJ, Williams DA, Samson L. Retrovirus-mediated expression of a DNA repair protein in bone marrow protects hematopoietic cells from nitrosourea-induced toxicity in vitro and in vivo. Cancer Res 1995; 55: 2608–2614. Wang G, Weiss C, Sheng P, Bresnick E. Retrovirus-mediated transfer of the human O6-methylguanine-DNA methyltransferase gene into a murine hematopoietic stem cell line and resistance to the toxic effects of certain alkylating agents. Biochem Pharmacol 1996; 51: 1221–1228. Jelinek J, Fairbairn LJ, Dexter TM, Rafferty JA, Stocking C, Ostertag W, Margison GP. Long-term protection of hematopoiesis against the cytotoxic effects of multiple doses of nitrosourea by retrovirusmediated expression of human O6-alkylguanine-DNA-alkyltransferase. Blood 1996; 87: 1957–1961. Davis BM, Reese JS, Koc ON, Lee K, Schupp JE, Gerson SL. Selection for G156A O6-methylguanine DNA methyltransferase genetransduced hematopoietic progenitors and protection from lethal-


47

48

49

50

51 52

53 54 55 56 57

58 59

60

ity in mice treated with O6-benzylguanine and 1,3-bis(2chloroethyl)-1-nitrosourea. Cancer Res 1997; 57: 5093–5099. Reese JS, Koc ON, Lee KM, Liu L, Allay JA, Phillips WP Jr, Gerson SL. Retroviral transduction of a mutant methylguanine DNA methyltransferase gene into human CD34+ cells confers resistance to O6-benzylguanine plus 1,3-bis(2-chloroethyl)-1-nitrosourea. Proc Natl Acad Sci USA 1996; 93: 14088–14093. Hickson I, Fairbairn LJ, Chinnasamy N, Lashford LS, Thatcher N, Margison GP, Dexter TM, Rafferty JA. Chemoprotective gene transfer I: transduction of human haemopoietic progenitors with O6-benzylguanine-resistant O6-alkylguanine-DNA alkyltransferase attenuates the toxic effects of O6-alkylating agents in vitro. Gene Therapy 1998; 5: 835–841. Chinnasamy N, Rafferty JA, Hickson I, Lashford LS, Longhurst SJ, Thatcher N, Margison GP, Dexter TM, Fairbairn LJ. Chemoprotective gene transfer II: multilineage in vivo protection of haemopoiesis against the effects of an antitumour agent by expression of a mutant human O6-alkylguanine-DNA alkyltransferase. Gene Therapy 1998; 5: 842–847. Fairbairn LJ, Chinnasamy N, Lashford LS, Chinnasamy D, Rafferty JA. Enhancing haemopoietic drug resistance: a rational for reconsidering the clinical use of mitozolomide. Cancer Gene Ther (in press). Crone TM, Goodtzova K, Edara S, Pegg AE. Mutations in human O6-alkylguanine-DNA alkyltransferase imparting resistance to O6benzylguanine. Cancer Res 1994; 54: 6221–6227. Hickson I, Fairbairn LJ, Chinnasamy N, Dexter TM, Margison GP, Rafferty JA. Protection of mammalian cells against chloroethylating agent toxicity by an O6-benzylguanine-resistant mutant of human O6-alkylguanine-DNA alkyltransferase. Gene Therapy 1996; 3: 868–877. Schofield R, Lord BI, Kyffin S, Gilbert CW. Self-maintenance capacity of CFU-S. J Cell Physiol 1980; 103: 355–362. Ploemacher RE, van der Sluijs JP, Voerman JSA, Brons NHC. An in vitro limiting-dilution assay of long-term repopulating hematopoietic stem cells in the mouse. Blood 1989; 74: 2755–2763. Miller AD, Rosman GJ. Improved retroviral vectors for gene transfer and expression. Biotechniques 1989; 7: 980–986. Han YM, Yoo OJ, Lee KK. Sex determination in single mouse blastomeres by polymerase chain reaction. J Assist Reprod Genet 1993; 10: 151–156. Lee SM, Rafferty JA, Elder RH, Fan CY, Bromley M, Harris M, Thatcher N, Potter PM, Altermatt HJ, Perinat Frey T, Cerny T, O’Connor PG, Margison GP. Immunohistological examination of the inter- and intracellular distribution of O6-alkylguanine DNAalkyltransferase in human liver and melanoma. Br J Cancer 1992; 66: 355–360. Tinwell H, Ashby J. Comparison of acridine orange and Giemsa stains in several mouse bone marrow micronucleus assays – including a triple dose study. Mutagenesis 1989; 4: 476–481. Harris LC, Marathi UK, Edwards CC, Houghton PJ, Srivastava DK, Vanin EF, Sorrentino BP, Brent TP. Retroviral transfer of a bacterial alkyltransferase gene into murine bone-marrow protects against chloroethylnitrosourea cytotoxicity. Clin Cancer Res 1995; 1: 1359–1368. Tong WP, Kirk MC, Ludlum DB. Formation of the cross-link 1[N3-deoxycytidyl], 2-[N1-deoxyguanosinyl]ethane in DNA

61

62

63 64 65 66 67 68

69 70 71

72 73 74 75

76

treated with N,N⬘-bis(2-chloroethyl)-N-nitrosourea. Cancer Res 1982; 42: 3102–3105. Gibson NW, Erickson LC, Hickman JA. Effects of the antitumor agent 8-carbamoyl-3-(2-chloroethyl)imidazo[5,1-d]-1,2,3,5-tetrazin-4(3 H)-one on the DNA of mouse L1210 cells. Cancer Res 1984; 44: 1767–1771. Horgan CM, Tisdale MJ. Antitumour imidazotetrazines IV. An investigation into the mechanism of antitumour activity of a novel and potent antitumour agent, mitozolomide (CCRG 81010, M & B 39565; NSC 353451). Biochem Pharmacol 1984; 33: 2185–2192. Clark BR, Keating A. Biology of bone marrow stroma. Ann NY Acad Sci 1995; 770: 70–78. Ogle KM, Luikart SD. BCNU-induced increase in sulfated glycosaminoglycan production by human bone marrow stromal cells. Exp Hematol 1988; 16: 636–640. Uhlman DL, Verfaillie C, Jones RB, Luikart SD. BCNU treatment of marrow stromal monolayers reversibly alters haematopoiesis. Br J Haematol 1991; 78: 304–309. Lord BI, Mason TM, Humphries ER. Age-dependent uptake and retention of 239Pu: its relationship to haemopoietic damage. Radiat Prot Dos 1992; 41: 163–167. Yang FT, Lord BI, Hendry JH. Gamma irradiation of the fetus damages the developing hemopoietic microenvironment rather than the hemopoietic progenitor cells. Radiat Res 1995; 141: 309–313. Molineux G, Testa NG, Massa G, Schofield R. An analysis of haemopoietic and microenvironmental populations of mouse bone marrow after treatment with busulphan. Biomed Pharmacother 1986; 40: 215–220. Bentley SA, Knutsen T, Whang Peng J. The origin of the hematopoietic microenvironment in continuous bone marrow culture. Exp Hematol 1982; 10: 367–372. Marshall MJ, Nisbet NW, Evans S. Donor origin of the in vitro hematopoietic microenvironment after marrow transplantation in mice. Experientia 1984; 40: 385–386. Perkins S, Fleischman RA. Hematopoietic microenvironment. Origin, lineage, and transplantability of the stromal cells in long-term bone marrow cultures from chimeric mice. J Clin Invest 1988; 81: 1072–1080. Simmons PJ, Przepiorka D, Thomas ED, Torok Storb B. Host origin of marrow stromal cells following allogeneic bone marrow transplantation. Nature 1987; 328: 429–432. Laver J, Jhanwar SC, O’Reilly RJ, Castro Malaspina H. Host origin of the human hematopoietic microenvironment following allogeneic bone marrow transplantation. Blood 1987; 70: 1966–1968. Agematsu K, Nakahori Y. Recipient origin of bone marrow-derived fibroblastic stromal cells during all periods following bone marrow transplantation in humans. Br J Haematol 1991; 79: 359–365. Santucci MA, Trabetti E, Martinelli G, Buzzi M, Zaccaria A, Pileri S, Farabegoli P, Sabattini E, Tura S, Pignatti PF. Host origin of bone marrow fibroblasts following allogeneic bone marrow transplantation for chronic myeloid leukemia. Bone Marrow Transplant 1992; 10: 255–259. Williams DA. Human Gene Market/Therapy Clinical Protocols, Number 173 A pilot study of dose intensified procarbazine, CCNU, vincristine (PCV) for poor prognosis pediatric and adult brain tumours utilising fibronectin-assisted, retroviral-mediated modification of CD34+ peripheral blood cells with O6-methylguanine DNA methyltransferase. Hum Gene Ther 1997; 8: 23–28.

1783