Radiation response and cell cycle regulation of p53 rescued malignant keratinocytes

Mutations in the tumor suppressor gene p53 were found in more than 90% of all human squamous cell carcinomas (SCC). To study the function of p53 in a keratinocyte background, a tetracycline-controlled p53 transgene was introduced into a human SCC cell line (SCC15), lacking endogenous p53. Conditional expression of wild-type p53 protein upon withdrawal of tetracycline was accompanied with increased expression of p21(WAF1/Cip1) resulting in reduced cell proliferation. Flow-cytometric analysis revealed that these cells were transiently arrested in the G1/S phase of the cell cycle. However, when SCC15 cells expressing p53 were exposed to ionizing radiation (IR), a clear shift from a G1/S to a G2/M cell cycle arrest was observed. This effect was greatly depending on the presence of wild-type p53, as it was not observed to the same extent in SCC15 cells lacking p53. Unexpectedly, the p53- and IR-dependent G2/M cell cycle arrest in the keratinocyte background was not depending on increased expression or stabilization of 14-3-3sigma, a p53-regulated effector of G2/M progression in colorectal cancer cells. In keratinocytes, 14-3-3sigma (stratifin) is involved in terminal differentiation and its cell cycle function in this cell type might diverge from the one it fulfills in other cellular backgrounds.

Th1-Polarizing Capacity of Clinical-Grade Dendritic Cells Is Triggered by Ribomunyl but Is Compromised by PGE2

The maturation state of (monocyte-derived) dendritic cells (DCs) determines the type of T-cell response. Currently, several maturation cocktails are used in clinical trials, most commonly a cocktail of TNF-alpha, PGE2, IL-1beta, and IL-6. The authors studied DC phenotype and functional ability to stimulate TH1 responses after maturation with different cocktails employing clinically approved agents. DCs were stimulated with the microbial agent Ribomunyl combined with IFN-gamma and various inflammatory cytokine cocktails: TNF-alpha/IL-1beta/IFN-gamma and TNF-alpha/PGE2 combined with monocyte-conditioned medium (MCM) or IL-1beta/IL-6. Regardless of the maturation cocktail used, all DCs possessed the characteristic phenotype of mature, migratory DCs (high expression of CD40, CD80, CD83, CD86, CCR7, MHC class I and MHC class II). Ribomunyl/IFN-gamma matured DCs produced high IL-12p70 levels, whereas other maturation stimuli did not. Even more striking, restimulation of Ribomunyl IFN-gamma mDCs with CD40-activating antibody reactivated IL-12p70 production. No IL-12p70 could be detected when DCs were stimulated with TNF-alpha/PGE2 combined with MCM or IL-1beta/IL-6, presumably by suppression by PGE2. Restimulation of these DCs with CD40-activating antibody failed to activate IL-12p70 production. Moreover, low levels of IL-10 were observed, possibly indicating inhibition of TH1-cell responses. Indeed, T cells stimulated with these DCs produced high levels of type 2 cytokine IL-5 and outgrowth of CD4CD25 T cells. This study shows that DC maturation with cytokine cocktails different from those most commonly used could be beneficial for immunotherapy trials in cancer patients.

ATM haplotypes and cellular response to DNA damage: association with breast cancer risk and clinical radiosensitivity

The ATM gene, mutated in the cancer-prone and radiation-sensitive syndrome ataxia-telangiectasia (AT), could predispose to breast cancer (BC) development and adverse radiotherapy responses. Sixteen ATM variants were genotyped in 254 BC cases, 70 of whom were adverse radiotherapy responders (RS-BC), and 312 control subjects and the ATM haplotypes were constructed. Constitutive ATM protein, cell survival, and the p53 response after exposure to ionizing radiation were compared in lymphoblastoid cell lines (LCLs) established from the BC cases, AT, and normal individuals. The tightly linked intronic ATM polymorphisms IVS22-77 T>C and IVS48 + 238 C>G, in the homozygote state were associated with increased BC risk [IVS22-77 CC versus TT odds ratio (OR), 1.67; 95% confidence interval (CI), 1.00-2.81], and in the heterozygote state with clinical radioprotection (IVS22-77 CT versus TT OR, 0.45; 95% CI, 0.24-0.85). Homozygote carriers of the G5557A variant were over-represented in RS-BC cases compared with non-RS-BC cases (OR, 6.76; 95% CI, 1.19-38.43). These three single nucleotide polymorphisms were associated with the three major ATM haplotypes present in >80% of the study population. BC LCLs treated with ionizing radiation exhibited an intermediate cell survival and p53 response between that of normal and AT LCLs, with the response in the RS-BC LCLs being more compromised than in the non-RS-BC LCLs. Our study suggests a general pattern of increased BC risk associated with carrying any one of the ATM variants studied, with a significant association being observed in individuals carrying variants on both ATM alleles (OR, 1.75; 95% CI, 1.09-2.81) and that ATM variants may impact on radiation sensitivity.

Targeting of adenovirus to human renal cell carcinoma cells

OBJECTIVES

The use of recombinant adenoviruses in cancer gene therapy is limited by the widespread expression of the coxsackievirus and adenovirus receptor on normal human cells. Targeting adenoviral vectors to renal cell carcinoma (RCC) cells may improve their potential in cancer gene therapy of patients with metastatic RCC. The G250 protein, also known as the carbonic anhydrase IX protein, is membranously expressed in all cases of clear cell RCC, and clinical studies have demonstrated exceptional tumor targeting with a G250 monoclonal antibody.

METHODS

A recombinant bispecific single-chain antibody directed against the RCC-associated G250 protein and the adenovirus fiber knob domain was constructed and used to retarget recombinant adenovirus expressing the green fluorescent protein under control of the cytomegalovirus promoter. G250-specific adenoviral transduction of cells was examined by flow cytometric analysis of green fluorescent protein expression.

RESULTS

G250-positive RCC cells displayed enhanced susceptibility for transduction by the green fluorescent protein recombinant adenovirus complexed with the G250-directed bispecific single-chain antibody when compared with native adenovirus. This enhanced transduction was restricted to G250-positive RCC cells and could be abolished completely in the presence of excess G250 protein.

CONCLUSIONS

The results of this study demonstrate the feasibility of immunologic retargeting of adenovirus to RCC cells with the highly tumor-specific G250 protein as the target. This strategy may provide the possibility of improving cancer gene therapy for patients with RCC.

Vaccination of patients with metastatic renal cell carcinoma with autologous dendritic cells pulsed with autologous tumor antigens in combination with interleukin-2: a phase 1 study

Dendritic cells (DC) have been recognized as the most potent antigen presenting cells (APC) of the immune system. We performed a phase 1 study in twelve patients with metastatic renal cell carcinoma (RCC) using autologous immature DC loaded with autologous tumorlysate (TuLy) as a vaccine based on our earlier in vitro observations that such DC can activate tumor-specific cytotoxic T-lymphocytes. The treatment was combined with low-dose interleukin (IL)-2, as this has shown benefit in DC-based therapies. Patients received three intradermal vaccinations at two weekly intervals, and, after each vaccination, IL-2 was administered for 5 consecutive days. In six patients, keyhole-limpet hemocyanin (KLH) was added to the DC culture for immunologic monitoring purposes. In general, DC phenotype was CD14(low), CD86(high), CD40(high), CD80(low), and CD83(low). We noticed that the number of CD14+ cultured DC increased during treatment. Nevertheless, ovalbumin uptake remained high, underlining that these cells were still functional immature DC. The vaccine was able to elicit cellular anti-KLH responses, emphasizing the ability of the injected DC to mount an immunologic response. However, proliferative responses against TuLy were not detected, and humoral responses against TuLy or KLH were absent. Objective clinical responses were not observed, but extended stable disease was noted. The absence of cellular, humoral, or clinical antitumor responses suggests that the vaccination strategy with immature DC has little benefit for patients with advanced RCC. Nevertheless, this study shows the feasibility of a completely autologous DC and tissue culture methodology for the generation of TuLy pulsed DC.

Cellular responses to radiation and risk of breast cancer

Mutations in the ataxia telangiectasia gene (ATM) result in an abnormal p53-mediated cellular response to DNA damage produced by ionising radiation. This deficiency is believed to contribute to the radiosensitivity and high cancer risk seen in ataxia telangiectasia (AT) patients and AT heterozygotes. Epidemiological studies have demonstrated that relatives of AT patients are particularly predisposed to breast cancer. This observation, together with the finding that a relatively high proportion of breast cancer patients display an abnormal severe reaction of normal tissues following radiotherapy, has led to the suggestion that AT heterozygosity plays a role in radiosensitivity and breast cancer development. The cloning of the ATM gene has allowed this possibility to be examined at the molecular level. The studies reported to date remain inconclusive, with the number of AT heterozygotes being found in radiosensitive breast cancer patients being less than would be expected based on the family studies. The potential role of several other recently identified genes which are involved in the cellular DNA damage response to ionising radiation and which could also play a role in radiosensitivity and breast cancer development are reviewed.

The p53-mediated DNA damage response to ionizing radiation in fibroblasts from ataxia-without-telangiectasia patients

PURPOSE

To assess the functionality of the p53-mediated pathway, activated by the ataxia-telangiectasia gene product (ATM) in response to ionizing radiation, in cells derived from four ataxia-without-telangiectasia patients. These patients exhibit cerebellar ataxia and cellular abnormalities that are compatible with the diagnosis of ataxia-telangiectasia (AT), but the telangiectasias normally seen in AT patients are absent.

MATERIALS AND METHOD

Protein and RNA extracts were prepared from primary fibroblast cultures non- or exposed to 5 Gy of ionizing radiation in order to monitor the modulation in p53 and ATM protein levels by immunologic techniques and WAF1/Cip1(p21) mRNA by Northern blotting.

RESULTS

A sub-optimal response in terms of increased levels of p53 and the transcriptional activation of WAF1/Cip1(p21) was see in the ataxia-without-telangiectasia fibroblast cultures examined over a 4 h period post-irradiation when compared with normal fibroblast cultures. The ATM protein was expressed at much reduced levels in the ataxia-without-telangiectasia and the classical AT fibroblast cultures examined when compared with normal fibroblast cultures.

CONCLUSIONS

Despite the milder clinical phenotypes observed in these ataxia-without-telangiectasia patients and the presence of low levels of ATM protein in the fibroblast cultures, their response to ionizing radiation quantitatively resembles that reported in fibroblast cultures established from classical AT patients.

Nijmegen breakage syndrome cells fail to induce the p53-mediated DNA damage response following exposure to ionizing radiation

The functionality of the p53-mediated pathway, activated in response to DNA damage, has been assessed in primary fibroblast cell cultures and Epstein-Barr virus-transformed lymphoblastoid cell lines derived from Nijmegen breakage syndrome (NBS) patients. This autosomal recessive disease is characterized by microcephaly, growth and mental retardation, chromosomal instability, radiosensitivity, and high cancer incidence. The recent mapping of the NBS gene to chromosome 8q21 demonstrates that NBS is genetically distinct from ataxia telangiectasia (AT). Changes in p53 protein levels were significantly reduced and delayed in all the NBS fibroblast cell cultures and lymphoblastoid cell lines examined compared to normal cultures over a 4-h period postirradiation (5 Gy). The transcriptional activation of p21(WAF1/CIP1) mRNA was also lower in 12 NBS fibroblast cultures examined. In agreement with an abrogated p53 function, NBS cells exposed to ionizing radiation show an abnormal cell cycle arrest at G1-S and a prolonged accumulation of cells in the G2 phase. In contrast, exposure to the alkylating agent methyl methanesulfonate results in similar increases of p53 and p21(WAF1/CIP1) mRNA in both cell types. The ATM gene transcript was found to be expressed at similar levels in NBS and normal cells, whereas it was strongly reduced in the AT homozygote cells examined. These results suggest that the ATM gene product cannot substitute for that of the NBS gene in the signaling of cellular damage produced by ionizing radiation and that both are involved in the activation of p53. The suboptimal p53-mediated response could contribute to the high cancer risk and radiosensitivity seen in NBS patients.

The defect in the AT-like hamster cell mutants is complemented by mouse chromosome 9 but not by any of the human chromosomes

X-ray sensitive Chinese hamster V79 cells mutants, V-C4, V-E5 and V-G8, show an abnormal response to X-ray-induced DNA damage. Like ataxia telangiectasia (AT) cells, they display increased cell killing, chromosomal instability and a diminished inhibition of DNA synthesis following ionizing radiation. To localize the defective hamster gene (XRCC8) on the human genome, human chromosomes were introduced into the AT-like hamster mutants, by microcell mediated chromosome transfer. Although, none of the human chromosomes corrected the defect in these mutants, the defect was corrected by a single mouse chromosome, derived from the A9 microcell donor cell line. In four independent X-ray-resistant microcell hybrid clones of V-E5, the presence of the mouse chromosome was determined by fluorescent in situ hybridization, using a mouse cot-1 probe. By PCR analysis with primers specific for different mouse chromosomes and Southern blot analysis with the mouse Ldlr probe, the mouse chromosome 9, was identified in all four X-ray-resistant hybrid clones. Segregation of the mouse chromosome 9 from these hamster-mouse microcell hybrids led to the loss of the regained X-ray-resistance, confirming that mouse chromosome 9 is responsible for complementation of the defect in V-E5 cells. The assignment of the mouse homolog of the ATM gene to mouse chromosome 9, and the presence of this mouse chromosome only in the radioresistant hamster cell hybrids suggest that the hamster AT-like mutant are homologous to AT, although they are not complemented by hamster chromosome 11.

The role of Ataxia telangiectasia and the DNA-dependent protein kinase in the p53-mediated cellular response to ionising radiation

The DNA-dependent protein kinase (DNA-PK), whose catalytic subunit shows structural similarities to the Ataxia telangiectasia (AT) gene product (ATM), has also been implicated in the p53-mediated signal transduction pathway that activates the cellular response to DNA damage produced by ionizing radiation. DNA-PK activity however was not found to be related to the transcriptional induction of WAFl/CIP1(p2l) in AT lymphoblastoid cell lines, following treatment with ionizing radiation. Normal protein and transcription levels of Ku70 and Ku80, as well as DNA-PK activity, were found in six different AT cell lines, 1-4 h following exposure to ionizing radiation, timepoints where reduced and delayed transcriptional induction of WAF1/CIP1 (p21) was observed. WAF1/CIP1 (p21) was found to be transcriptionally induced by p53 in normal cell lines over this same time period following exposure to ionizing radiation. These results suggest that despite the findings that in vitro DNA-PK may phosphorylate p53, in vivo it would not appear to play a central role in the activation of p53 as a transcription factor nor can it substitute for the ATM gene product in the cellular response following exposure to ionizing radiation.

A gene that regulates DNA replication in response to DNA damage is located on human chromosome 4q

Inhibition of replicative DNA synthesis following gamma-irradiation is observed in eukaryotic cells but is defective in cells derived from patients with the cancer-prone inherited disorder ataxia-telangiectasia (A-T) and in A-T-like Chinese hamster cell mutants. Chinese hamster cells show a less pronounced inhibition of DNA synthesis after gamma-irradiation when compared to irradiated human HeLa or mouse A9 cells. Therefore, to identify new human genes involved in the regulation of DNA replication in response to ionizing radiation in mammalian cells, single human chromosomes were introduced into Chinese hamster cells by microcell-mediated chromosome transfer. It is found that a new gene on human chromosome 4q inhibits DNA synthesis following gamma- and UV irradiation in hamster cells. However, this delay of DNA replication did not improve cell survival or the level of chromosomal aberrations induced by X-rays, indicating that the lack of the inhibition of DNA synthesis after X-irradiation is not a prerequisite for the X-ray sensitivity and chromosomal instability, which is observed in A-T and A-T-like hamster cells.

Complementation analysis of the murine scid cell line

It has been shown that several X-ray-sensitive Chinese hamster cell mutants defective in repair of DNA double-strand breaks (DSBs) are also impaired in the process of V(D)J recombination. The hamster mutants with this phenotype represent three distinct complementation groups, represented by the xrs series, XR-1 and V-3. The murine scid cell line also shows the same phenotype, and therefore we examined whether the scid mutant represents a new complementation group or belongs to one of the existing groups. Scid cells were fused with hamster cell mutants representing the three complementation groups. Hybrids between V-3 and scid cells were only partially complemented for X-ray sensitivity, whereas hybrids derived from fusions with the other mutants were resistant to X rays. These results suggest that V-3 and scid cells are defective in the same gene. To confirm this finding, a single human chromosome 8, which is known to carry the scid gene, was introduced into V-3 cells by microcell-mediated chromosome transfer. Nine hybrid clones derived from V-3 and carrying human chromosome 8 were obtained, and seven were found to be partially complemented for X-ray sensitivity. When human chromosome 8 was introduced into scid cells, seven of eight hybrid clones became resistant to X rays. The results indicate that the defective genes in V-3 and scid are both localized on human chromosome 8. This supports the results from the fusion analysis that V-3 and scid cells are defective in the same gene.

A novel type of X-ray-sensitive Chinese hamster cell mutant with radioresistant DNA synthesis and hampered DNA double-strand break repair

It has been shown that the Chinese hamster cell mutant V-C8 is sensitive to different DNA damaging agents, such as mitomycin C (MMC), alkylating agents, UV light, and X-rays. We found that V-C8 is also sensitive to the following radiomimetic agents: bleomycin (approximately 2-fold, based on D10 values), H2O2 (approximately 2-fold), streptonigrin (approximately 11-fold), and etoposide (approximately 8-fold). Two independent spontaneous MMC-resistant revertants isolated from V-C8 cells show a level of cell killing by X-rays, EMS, and UV light which is similar to that of wild-type cells, suggesting that the observed pattern of cross-sensitivity of V-C8 cells to a wide spectrum of DNA damaging agents results from a single mutation. V-C8 cells also display radioresistant DNA synthesis following gamma-irradiation which, however, remained almost unchanged in the V-C8 revertants. The measurement of the level and rate of repair of DNA single- and double-strand breaks (SSBs and DSBs, respectively) by the DNA elution technique showed that the V-C8 mutant has a slower repair of DSBs induced by gamma-rays. The described unique phenotype of V-C8 cells suggested that V-C8 represents a novel type of mutant amongst X-ray-sensitive hamster cell mutants. To confirm this, complementation analysis with other X-ray-sensitive mutants was performed. V-C8 cells were fused with EM9, XR-1, xrs5, sxi-1, V-3, V-E5, irs3, and BLM2 mutant cells, representing different complementation groups. All the obtained hybrids regained X-ray resistance (or bleomycin resistance in the case of V-C8/BLM2 hybrids) similar to that of wild-type cells, indicating that V-C8 represents a new complementation group. The results presented indicate that V-C8 is defective in a gene involved in a pathway operating in the responses to different DNA damaging agents in mammalian cells.

Studies on phenotypic complementation of ataxia-telangiectasia cells by chromosome transfer

Cells derived from patients with the cancer-prone inherited disorder ataxia-telangiectasia (A-T) show an abnormal response to ionizing radiation-induced DNA damage, such as an increased cell killing and a diminished inhibition of DNA synthesis. The enhanced killing of A-T (group D) cells by X-rays can be corrected by multiple cDNAs, mapping to different chromosomes (6, 11, 17, and 18). In order to examine whether genes located on these chromosomes complement AT-D cells, normal neo-tagged chromosomes 6, 11, 17, and 18 were introduced into AT-D cells by microcell-mediated chromosome transfer. However, correction of the enhanced killing of AT-D cells by X-rays could only be achieved by chromosome 11 and by none of the other chromosomes tested. The enhanced killing of A-T (complementation group C) cells was also corrected by chromosome 11. Usually, but not in all microcell hybrid clones, chromosome 11 also corrected the radioresistant DNA synthesis (RDS) phenotype of AT-D and AT-C cells. These results (i) confirm findings by others suggesting assignment of the ATD and ATC genes to chromosome 11, (ii) demonstrate that several genes can modify the cellular radiation response when they are taken out of their normal genomic context and/or control, and (iii) indicate that the RDS phenotype and the enhanced cell killing in A-T are independent pleiotropic features resulting from the primary mutations in A-T. Also, our findings underscore that, in establishing cDNAs as candidate genes for A-T, microcell-mediated chromosome transfer studies are needed to exclude nonspecific correcting effects of these candidate cDNA genes.

Functional complementation studies with X-ray-sensitive mutants of Chinese hamster cells closely resembling ataxia-telangiectasia cells

In order to isolate a human gene complementing the defect in A-T-like hamster cell mutants, the mutants were used as recipients for genomic DNA transfection, using either HeLa chromosomal DNA or DNA from a human cosmid library. Three primary transformants with an intermediate X-ray sensitivity and almost normal sensitivity to MMS, but retaining radioresistant DNA synthesis (RDS), were obtained. To identify the human chromosome that complements the defect in the A-T-like mutants, and to assess the degree of complementation for survival and RDS, microcell-mediated chromosome transfer was used. At least 20 independent hybrid clones between the mutant and each one of the human chromosomes 1, 2, 4, 5, 15, 17 or 18 were isolated. All hybrid clones remained X-ray sensitive, except one with chromosome 4, and another with chromosome 15, both showing an intermediate X-ray sensitivity. By using in situ hybridization we found that this partial correction was due to the presence of a mouse chromosome. In these two hybrids containing the mouse chromosome together with human chromosome 4 or 15, RDS was fully complemented only in the hybrid with chromosome 4 but not in the one containing chromosome 15, suggesting that RDS and X-ray sensitivity may be complemented independently.

Human chromosome 11 complements ataxia-telangiectasia cells but does not complement the defect in AT-like Chinese hamster cell mutants

It has been shown that the X-ray-sensitive Chinese hamster V79 mutants (V-E5, V-C4 and V-G8) are similar to ataxia-telangiectasia (A-T) cells. To determine whether the AT-like rodent cell mutants are defective in the gene homologous to A-T (group A, C or D), human chromosome 11 was introduced to the V-E5 and V-G8 mutant cells by microcell-mediated chromosome transfer. Forty independent hybrid clones were obtained in which the presence of chromosome 11 was determined by in situ hybridization. The presence of the region of chromosome 11q22-23 was shown by molecular analysis using polymorphic DNA markers specific for the ATA, ATC and ATD loci. Seventeen of the obtained monochromosomal Chinese hamster hybrids contained a cytogenetically normal human chromosome 11, but only twelve hybrid cell lines were shown to contain an intact 11q22-23 region. Despite the complementation of the X-ray sensitivity by a normal chromosome 11 introduced to A-T cells (complementation group D), these twelve Chinese hamster hybrid clones showed lack of complementation of X-ray and streptonigrin hypersensitivity. The observed lack of complementation does not seem to be attributable to hypermethylation of the human chromosome 11 in the rodent cell background, since 5-azacytidine treatment had no effect on the streptonigrin hypersensitivity of the hybrid cell lines. These results indicate that the gene defective in the AT-like rodent cell mutants is not homologous to the ATA, ATC or ATD genes and that the human gene complementing the defect in the AT-like mutants seems not to be located on human chromosome 11.

Cellular characteristics of Chinese hamster cell mutants resembling ataxia telangiectasia cells

The radiosensitive Chinese hamster V79 cell mutants (V-C4, V-E5 and V-G8), isolated previously in our laboratory, have been shown to resemble human ataxia telangiectasia (A-T) cells. These hamster cell mutants were further characterized with respect to cross-sensitivity to different radiomimetic agents and to mutation induction by X-rays. The data on cell survival (D10 values) show that they are hypersensitive to adriamycin (2-3-fold increase), etoposide (3-fold for V-G8 and 6-fold for V-E5 and V-C4), calicheamicin gamma 1I (4-fold) and streptonigrin (3-fold for V-G8 and V-C4, and 12-fold for V-E5). The frequency of X-ray-induced hprt mutations is slightly enhanced in the hamster mutant cells treated with the same dose. However, the mutants show similar mutability as parental V79 cells when considering the same survival level. The overall conclusion from these studies is that these hamster cell mutants mimic the phenotypic characteristics observed in cultured cells from A-T patients and, therefore, may be defective in the same repair pathway as their human counterparts.

Efficient cDNA cloning by direct phenotypic correction of a mutant human cell line (HPRT-) using an Epstein-Barr virus-derived cDNA expression vector

Human cells are, in general, poor recipients of foreign DNA, which has severely hampered the cloning of genes by direct phenotypic correction of deficient human cell lines after DNA mediated gene transfer. In this communication a methodology is presented which largely circumvents this problems. The method relies on the use of a recently developed episomal Epstein-Barr-virus-derived cDNA expression vector (Belt et al. (1989) Gene 84, 407-417). The cloning of hypoxanthine phosphoribosyltransferase (HPRT) cDNA, corresponding to a low abundant mRNA in wild type cells is used as a model system. Size fractionated poly (A)+ RNA from wild type cells, which resulted in an approximately 10 fold enrichment in HPRT mRNA, was used to construct a cDNA library of 25,000 independent clones in the pECV25 vector. An HPRT deficient human cell line was transfected and subsequently selected with hygromycin B for DNA uptake. In a small scale experiment only 7000 hygromycin BR transfectants were sufficient to isolate 2 independent HATR clones which were shown to replicate episomes harbouring HPRT cDNA. The first insert had a 5' untranslated region (UTR) and a 3' UTR perfectly in agreement with published data. The second cDNA clone harboured an unusually long 5' UTR and a shorter 3' UTR due to alternative polyadenylation of the HPRT transcript which has not been previously recognized.