Nucleic Acids Research, Vol. 18, No. 19 5723
Expression of a human 06-alkylguanine-DNAalkyltransferase cDNA in human cells and transgenic mice C.Y.Fan, P.M.Potter, J.Rafferty, A.J.Watson, L.Cawkwell, P.F.Searle1, P.J.O'Connor and G.P.Margison* Cancer Research Campaign, Department of Chemical Carcinogenesis, Paterson Institute for Cancer Research, Manchester M20 9BX and 1Department of Cancer Studies, University of Birmingham Medical School, Birmingham B15 2TJ, UK Received June 29, 1990; Revised and Accepted September 10, 1990
ABSTRACT A truncated human 06-alkylguanine-DNAalkyltransferase (ATase) cDNA was ligated into an expression vector under the control of the mouse metallothionein-1 gene promotor and upstream of part of the human growth hormone gene to provide splice and polyadenylation signals. Transfection of this construct into human cells resulted in very high levels of ATase expression (more than 300 fmoles/mg protein versus less than 2 fm/mg protein in parent vector transfected control cells). Microinjection of a 4.2 kb fragment of this vector into B6D2F2 mouse embryos and implantation of survivors into pseudopregnant females has so far generated 35 offspring. Southern analysis of tail tip DNA has shown that 11 of the offspring are transgenic for the human ATase gene, between 1 and at least 30 copies of the gene being detected. Human ATase transcripts were detected in total RNA extracted from liver obtained from two male transgenic mice by partial hepatectomy. Cell free extracts of liver samples from five transgenic mice showed up to 4 times higher ATase levels than control livers. INTRODUCTION Alkylating agents exert a wide range of biological effects in both pro and eukaryotes and there is ever increasing evidence that these are mediated via interactions with DNA. The majority of evidence, much of it correlative, indicates that O6-alkylguanine is a mutagenic, toxic and probably carcinogenic lesion (1-4). Repair of 06-alkylguanine in DNA can be mediated by ATase and cells or tissues that express low levels of ATase are generally more susceptible to the toxic and mutagenic effects of alkylating agents than high level expressors (3,4). More compelling evidence for the importance of this lesion in DNA has come from experiments in which mammalian cells sensitive to the toxic effects of alkylating agents have attained increased resistance following transfection and expression of the cloned E. coli ATase genes ada (4), and more recently ogt (5). *
To whom
correspondence, should be addressed
To further examine the possible role of 06-alkylguanine in carcinogenesis by alkylating agents, we have produced ada (6) and ogt (5) transgenic mice in attempts to generate a strain with greatly increased repair capacity. However, we have so far found little orrno evidence of expression of the bacterial ATases (5,6) although more recently, ada-expressing transgenic mice have been described (7,8). The availability of the human ATase cDNA nucleotide sequence (9) has allowed us to produce a human ATase cDNA inducible expression vector which we have shown to achieve high levels of expression in human cells. We also report the generation of human ATase transgenic mice, the presence of human ATase transcripts and up to 4 times higher levels of ATase activity in liver tissue.
MATERIALS AND METHODS Isolation of human alkyltransferase cDNA Oligonucleotides 1075, 207 and 238 (Figure la) were synthesised on a Dupont Coder 3000 using standard phosphoramidite chemistry and purified by preparative polyacrylamide gel electrophoresis (10). Oligo-dT selected RNA (10%g) prepared from human CEM cells was converted to cDNA by specific priming of the first strand using an oligonucleotide (1075, Figure la) complementary to the 3' region of the ATase sequence (9). Second strand synthesis was performed according to Gubler and Hoffmann (11) and the cDNA then amplified by the polymerase chain reaction using human ATase specific oligonucleotides 207 and 238 (Figure la) (9). These create Bam HI restriction endonuclease recognition sites 4bp upstream of the initiation and 12bp downstream of the termination codons respectively. Approximately lOOng of dscDNA was amplified in a total volume of 501d containing lOmM Tris-HCl pH8.3, 1.5mM MgCl2, 1 unit of Taq polymerase (Cetus), lOOng of each oligonucleotide and 400 1tM of each dNTP. After overlaying with 75 yd of paraffin oil, 35 cycles of amplification were performed (93°C-20 seconds, 60°C-1 minute, 72°C-1 minute). Aliquots (5 Id) were then subjected to agarose gel electrophoresis to determine the extent of amplification.
5724 Nucleic Acids Research, Vol. 18, No. 19
a Oligonucleotide 1075 Human ATase cDNA
Oligonucleotide 207 Human ATase cDNA
GTTACACGTGTGTGTCGCTCM CAATGTGCACACACAGCGAGTTTGT
..... AGT
CM
CTTGGTACTTGGATCC ATG GAC MG GAT TGT G CGCTGMCCATGMCCTT TAC CTG TTC CTA ACA C
b rTracription
FcoRl
Oligonucleotide 238 Human ATase cDNA
hAT
I MNT-1 wrww-I
CGCTCAMCATGGATCCTACTGCAC
BamHl
GCGAGTTTGTAGGTAGGATGACGTGTATA AGT CM
Hindll!
I Baml
-j
hGH
Figure 1. (a) Sequences of oligonucleotides used for cDNA synthesis (1075) and amplification (207 and 238). (b)Structure of the 4.2 kb microinjected fragment metallothionein-1 gene (mMT-1) promotor region nucleotides -1700 to -1. of the human ATase (hAT) cDNA expression vector pMThAThGH. mMT-1 5' untranslated region nucleotides + 1 to +68. hGH, transcription unit of the human growth hornone gene. The 648 bp hAT cDNA was ligated into the unique BamHI site. For further details, see text. ,
The amplified 671 bp fragment was digested with BamHI and ligated into BamHI restricted pUC 8.1 to generate phATI. Eukaryotic expression vector construction The 648 bp Bam HI ATase fragment was isolated from phATI and ligated into the vector p(RG48)MT-hGH which consists of the mouse metallothionein-1 gene promotor and most of the transcribed region of the human growth hormone gene containing a number of introns and a polyadenylation signal. This vector is basically similar to that described in (12) but it contains a unique BamHI site. A 4.2 kb EcoRI-HindIll fragment of this construct (Figure lb) was isolated from 2 sequential low melting point agarose gels and made up at 2 ng/4l in buffer for microinjection. Expression of ATase construct in mammalian cells In order to confirm that pMThAThGH was capable of expressing the human ATase the intact plasmid (or p(RG48)MT-hGH in control experiments) was transferred by lipofection (Gibco BRL) into SV40 transformed Xeroderna pigmentosum fibroblasts (XP12ROSV ref 13): 18 ug of plasmid DNA was cotransfected with 2,tg of pZipneoSV(X)l (14) and cells were selected in MEM containing 10% foetal calf serum and 0.5 mg/ml G418. Resistant clones were isolated after 23 days, expanded and extracts prepared for ATase assay. Production of human ATase transgenic mice B6D2F2 mouse embryos were microinjected using standard procedures (15). Drop cultures were incubated at 37°C in an atmosphere of 95% air 5% CO2 overnight and surviving embryos were implanted into pseudopregnant females. Offspring were weaned at 28 days and tail tips were removed for DNA isolation and analysis. Partial hepatectomy (1/3) provided liver tissue for DNA and total RNA extraction, and for ATase assay. DNA isolation and analysis DNA was isolated (15) and 5 /tg samples were BamHI digested, electrophoresed in 1% agarose gels and transferred to nylon filters (Hybond N, Amersham International) essentially as previously described (10). Prehybridisation and hybridisation were as described using a 32P-labelled nick-translated human ATase 648 bp probe (16); Filters were washed sequentially in 6 x, 1 x and 0.1 x SSC at 65°C and exposure was with an intensifying screen for up to 4 days between washes.
E 20 E~~~ 0r
50
100 150 ,ug protein assayed
200
Figure 2. Alkyltransferase activity in extracts of G418 resistant clones isolated after transfection of Xerodema pigmentosum cells with the human ATase cDNA conltaining vector pMThAThGH (0-0, clone Sb; 0-0, clone 6B) or control vector (LI-I, clone lb) see text for further details.
RNA isolation and analysis Fresh liver samples were homogenised using the guanidinium method as recommended (10). The samples were then repeatedly (3 x) phenol:chloroform extracted before the RNA was recovered by sequential isopropanol precipitations. Denatured RNA samples were separated on 1 % agarose gels containing 2.2M formaldehyde and 1 x MOPS running buffer as described (10), except 2yu of 0.5 jig/ml EtBr was added to each sample before loading. This allows visualisation of the major RNA species without the usual background fluorescence associated with this technique. Following electrophoresis (40V, 6 hrs) the gels were washed in water prior to capillary transfer of RNA to Hybond-N in the usual manner (10). An RNA ladder (BRL) was used as a size marker. All hybridisations were performed in 40% formamide, 6 x SSC, 1 % SDS, I x Denhardt's solution and 250 /Ag/ml denatured salmon sperm DNA at 420C.
Alkyltransferase assay Extracts of cells or tissues were prepared and assayed for ATase activity as described (17) except that in some experiments incubation volumes were 500,lI and the amount of 4M perchloric acid added following incubation (2h at 37°C) was reduced to 200,1. Specific activities were calculated from the linear portion of the graph of fmoles [3H]methyl transferred to protein versus amount of protein assayed.
}.1'?8:M.:s,'}.: ,~ ~.
Nucleic Acids Research, Vol. 18, No. 19 5725
Table 1. Status of mice surviving to weaning age Mouse Number
Sex
transgenic status (by BamHl) digest)
Mouse 51 0 42 41 40 3.4 3.3 3.2 3.13-0 23 22 2-1 2.0 1 4 1 3 1.2 1.1 1.0 approx. copy number
}. .... ..... Y
.
.:X::!::}i.:}.::.i:::.:} .:: :.:: :
.. ''''' . . ..
..
..
..
:....
...
..
..
:l: :' ..:....::.: :, .
1.0 1.1 1.2 1.3 1.4 2.0 2.1 2.2 2.3 3.0 3.1 3.2 3.3 3.4 4.0 4.1 4.2 5.0 5.1 5.2 5.3 6.0
6.1 6.2 6.3 6.4 7.0 7.1 7.2 7.3 7.4 8.0 8.1 8.2 8.3
female female female female female female female female female female female female female female female female female female female female female male male male male male male male male male male male male male male
.e..
......::
.::::.:..::..
. a:e: : :.
64Bbp-
;
..' ...
~~~~~~ :..: ~~. .: :...~~~~ :X . . ... . ... .. . . .
.'
....
..
_
_
5
+
30
+
20
+
5 5
+
I
Mouse
52 53 60 61 6-2 63 64 70 7-1 72 73 7480 81 82 83
648bp-
mm
46 4a
+ +
5
+
10
+
20
Figure 3. Southern analysis of tail-tip DNA. 5 jug of DNA was digested with Bam HI, electrophoresed in 1 % agarose and transferred following denaturation to Hybond-N. The filter was prehybridised and hybridised with a 32P-labelled human ATase cDNA probe as described in the text. The position of the 648 bp fragment is shown.
Fluorography SDS-PAGE and fluorography was carried out as previously described (23) except that 16% gels were used.
RESULTS Expression of transgenic vector in human cells The XP cell line used has previously been reported to express very low levels of ATase (less than 2 fm/mg protein, 12). Following lipofection of the ATase vector (pMThAThGH) together with the selectable vector pZipneoSV(x)l, 10 G418 resistant clones were isolated and 3 of the 8 assayed showed marked increases in ATase activity. Fig 2 shows the results of the ATase assay of clones 5b and 6b which had specific activities of 350 and 158 fmoles/mg protein respectively. In contrast, three G418 resistance clones isolated after lipofection of the parent vector together with pZipneoSV(X)1 expressed very low levels of ATase (also 2fm/mg protein). Production of transgenic mice Following microinjection of 700 eggs, 240 survivors were implanted into 12 pseudopregnant D6B1 females (10 per oviduct). DNA was isolated from tail tips (removed after weaning) of the
N 71 73
9.57.544-
2414-
03Figure 4. Northern analyses of total liver RNA from a normal mouse (lane N) and from transgenic mice 7.1 (lane 2) and 7.3 (lane 3). Exactly 10 jig of total liver RNA (as determined by E26 measurements) were run in each lane. Sizes of markers in kb are shown.
5726 Nucleic Acids Research, Vol. 18, No. 19
3 to 0
E
500
400l
-200 1-00-
Controls
60&2 6411 73
Fme 5. Alyltsferase specific acfivities (fmoles/mg total protein) in extracts of normal and transgenic mouse liver.
35 mice born and analysis (see below) indicated that 11 were transgenic, one with in excess of 30 copies of the ATase/cell (Table 1). Nucleic Acid analysis The results of Southern analyses of tail tip DNA are shown in Fig 3. A hybridisation signal at 648 bp was taken as indicative of the presence of the human ATase cDNA. Under the conditions used, hybridisation to the endogenouse mouse ATase gene was not seen. Approximate human ATase cDNA copy numbers are presented in Table 1. Northern analyses showed little or no hybridisation to normal liver RNA whereas in the two transgenic mice that were analysed signals were easily seen at approximately 2.4 kb. These were exactly the size expected from this eukaryotic expression vector (pMThAThGH) and were considerably more intense for mouse 7.3 than for mouse 7.1 (Fig 4). Assay of hepatic aikyltnderase activity As shown in Fig 5, extracts of liver obtained by partial hepatectomy of six norml male control animals contained 96 to 148 fmoles ATase/mg total protein. Extracts of the liver of two male transgenic mice contained ATase activity close to control levels (mice 6.0 and 6.2) two contained twice as much (mice 6.4 and 7. 1) and one more than 4 times as much (mouse 7.3). Fluorography of liver extacts of control and transgenic (7. 1 and 7.3) mice (Fig. 6) showed a band of around 22 kDa, the intensity of which corresponded to the ATase activities shown in Fig. 5.
DISCUSSION The carcinogenic and other biological effects of alkylating agents in mice are well established (18). Of the various lesions produced in DNA, 06-alkylation of guanine in has been implicated in the carcinogenic effects. However, the evidence is correlative and not entirely consistent (19). One approach to providing more conclusive evidence would be to generate transgenic mice that express very high levels of an o6- alkylguanine repair function and to assess whether this protects against these biological effects.
FE1 6. Fluorograph of extracts of the liver of a control mouse (N; ATase sp.act 148 fm/mg protein) and transgenic mice 7.1 and 7.3 following incubation with [3H]MNU-methylated substrate DNA, SDS-PAGE and transfer to nitrocellulose. A 3mm thick 16% acrylamide gel was used and each sample contained 200 gig total protein.
This approach has been used very successfully in a variety of mammalian cells, and more recently in yeast (20) by a number of different groups (reviewed in 4). Although our attempts to produce E. coli ada gene transgenic mice that express high levels of the bacterial ATase have been unsuccessful (6), other groups using different expression vectors have been able to show low levels of ada expression (7,8). The inherent disadvantage of using the ada gene is that the 39 kDa protein encoded contains two ATase active sites that act on alkylphosphotriesters and on 06-alkylguanine and 04-alkylthymine (21). Thus it would be difficult to exclude a possible contribution from any of these products if changes in the responses of such mice to the biological effects of alkylating agents are observed. The ATase encoded by the E.coli ogt gene (22,23) more closely resembles the mammalian equivalent in as much as it lacks activity on alkylphosphotriesters (22). However, though ogt transgenic mice have been produced (5) no indication of expression has been found. The cloning and sequencing of the human ATase cDNA (9) has enabled us to generate a transgenic vector containing this gene. In the vector used, ATase expression is controlled by the heavy metal inducible mouse metallothionein-1 gene promoter: the presence of the human growth hormone sequences downstram provide introns which have been reported to increase expression in transgenic mice (24). The ability of the construct to express at high levels in human cells (Fig. 1) and to increase expression following addition of Zn2+ to the growth medium of these cells (data not shown) confirmed its integrity before proceeding with transgenic mice generation. In the present experiment approx 30 percent of the offspring surviving to weaning age were shown to be transgenic by the presence of a 648 bp BamHI fragment seen on Southern analysis of tail-tip DNA: ATase gene copy numbers ranged from 1 to, in one mouse, greater than 30 per genome. A more comprehensive molecular analysis of these mice will be presented elsewhere. In two of the male transgenic mice subjected to partial hepatectomy, human ATase message-containing ranscripts were
Nucleic Acids Research, Vol. 18, No. 19 5727 readily detectable in total RNA isolated from excised liver: transcripts were very much more abundant in mouse 7.3 than in 7.1. Alkyltransferase activity in normal mouse liver extract was in the range of 96 to 148 fmoles/mg protein, close to that reported for other strains used in such studies (6-8). Two of the five transgenic mice assayed for hepatic ATase activity showed similar levels of activity whereas in the other 3 activity was considerbly higher, particularly in mouse 7.3 which contained more than 4 times the mean control level. Fluorography further confirmed the considerably higher hepatic ATase activity in mouse 7.3 in comparison with 7.1 and a control mouse. However, in the gel system used, no resolution of the transgene and the endogenouse mouse ATases was obtained, probably because of the amounts of protein loaded (200 jig/lane) to detect the ATases. There was a very approximate correlation between the hAT cDNA copy number and the hepatic ATase activity and this correlation was also evident with mRNA abundance, mouse 7.3 appearing to produce considerably more transcripts than mouse 7.1. Large numbers of analyses will be required before any consistent relationships can be established: in the present work we have addressed the question only of whether it is possible to express the human gene in transgenic mice. In E. coli ada gene transgenic mice (7,8) the ATase expressed acts on 06-alkylguanine and alkylphosphotriesters (21,23) and half of the total additional ATase activity is due to the latter function. After correction for this, 06-alkylguanine ATase activity in hAT transgenic mice is approx 3 times higher than in the ada mice. Thus, if 06-alkylguanine is important in the adverse biological effects of alkylating agents, protection against these effects may be expected to be more extensive in the human ATase expressing mice. The tissue-dependence of ATase expression, the production of a homozygous transgenic mouse line and the assessment of the response of these animals to specific biological effects of alkylating agents are in hand.
ACKNOWLEDGEMENTS This work was supported by the Cancer Research Campaign. We thank Miss Michelle D. Boulter and Mrs Eunice M. Guscott for manuscript preparation.
REFERENCES 1. Singer, B. (1979) J. Natl. Cancer Inst., 62, 1329-1339. 2. Saffhill, R., Margison, G.P. and O'Connor, P.J. (1985) Biochim. Biophys. Acta 823, 111-145. 3. Yarosh, D.B. (1985) Mut. Res. 145, 1-16. 4. Margison, G.P. and O'Connor, P.J. (1979) In: Handbook of Experiemtnal Pharmacology, vol. 94/1 (Ed. C.S. Cooper and P.L. Grover), SpringerVerlag Berlin, pp 547-571.5. Harris, L. and Margison, G.P. Unpublished results. 6. Searle, P., Tinsley, J.M., O'Connor, P.J. and Margison, G.P. (1989) Biochem. Soc. Trans. 17, 120-121. 7. Matsukuma, S., Nakatsuru, Y., Nakagawa, K., Utakoji, T., Sugano, H., Kataoka, H., Sekiguchi, M. and Ishikawa, T. (1989) Mut. Res. 218, 197-206. 8. Lim, I.K., Dumenco, L.L., Yun, J., Donovan, C., Warman, B., Gorodetzkaya, N., Wagner, T.E., Clapp, D.W., Hanson, R.W. and Gerson, S.L. (1990) Can. Res. 50, 1701-1708. 9. Tano, K., Shiota, S., Collier, J., Foote, R.S. and Mitra, S. (1990) Proc. Natl. Acad. Sci. USA, 87, 686-690. 10. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbour Press, New York. 11. Gubler, U. and Hoffman, B.J. (1983) Gene 25, 263. 12. Palmiter, R.D., Norstedt, R.E., Gelinas, R.E., Hammer, R.E. and Brinster, R.L. (1983) Science 222, 809-814.
13. Boyle, J.M., Margison, G.P. and Saffhill, R. (1986) Carcinogenesis 9, 1987-1990. 14. Cepko, C.L., Roberts, B.E. and Mulligan, R.C. (1984) Cell 37, 1053- 1062. 15. Hogan, B., Costantini, F. and Lacy, E. (1984) Manipulating the Mouse Embryo: A Laboratory Manual. Cold Spring Harbour Press, New York. 16. Rigby, P.W.J., Dieckmann, M., Rhodes, C. and Berg, P. (1977) J. Mol. Biol. 113, 237. 17. Morten, J.E.N. and Margison, G.P. (1988) Carcinogenesis 9, 45-49. 18. LARC Monographs on the evaluation of the carcinogenic risk of chemcials to humans. Vols. 1 (1972), 4 (1974), 17 (1978) and Supplement No. 2 (1980). 19. Margison, G.P. and O'Connor, P.J. (1979) In Chemical Carcinogenesis and DNA, Vol. 1 (Ed. P.L. Grover). CRC Press Florida, USA pp 111-159. 20. Brozmanova, J., Kleibl, K., Vlckova, V., Skorvaga, M., Cernakova, L. and Margison, G.P. (1990) Nucleic Acids Res., 18, 331-335. 21. Lindahl, T., Sedgwick, B., Sekiguchi, M. and Nakabeppu, Y. (1988) Annu. Rev. Biochem. 57, 133-157. 22. Margison, G.P., Cooper, D.P. and Brennand, J. (1985) Nucleic Acids Res. 13, 1939-1952. 23. Potter, P.M., Wilkinson, M.C., Fitton, J., Carr, F.J., Brennand, J., Copper, D.P. and Margison, G.P. (1987) Nucleic Acids Res. 22, 9177-9193. 24. Brinster. R.L., Allen, J.M., Behringer, R.R., Gelinas, R.E. and Palmiter, R.D. Proc. Natl. Acad. Sci.(1988) 85, 836-840.
Expression of a human 06-alkylguanine-DNAalkyltransferase cDNA in human cells and transgenic mice C.Y.Fan, P.M.Potter, J.Rafferty, A.J.Watson, L.Cawkwell, P.F.Searle1, P.J.O'Connor and G.P.Margison* Cancer Research Campaign, Department of Chemical Carcinogenesis, Paterson Institute for Cancer Research, Manchester M20 9BX and 1Department of Cancer Studies, University of Birmingham Medical School, Birmingham B15 2TJ, UK Received June 29, 1990; Revised and Accepted September 10, 1990
ABSTRACT A truncated human 06-alkylguanine-DNAalkyltransferase (ATase) cDNA was ligated into an expression vector under the control of the mouse metallothionein-1 gene promotor and upstream of part of the human growth hormone gene to provide splice and polyadenylation signals. Transfection of this construct into human cells resulted in very high levels of ATase expression (more than 300 fmoles/mg protein versus less than 2 fm/mg protein in parent vector transfected control cells). Microinjection of a 4.2 kb fragment of this vector into B6D2F2 mouse embryos and implantation of survivors into pseudopregnant females has so far generated 35 offspring. Southern analysis of tail tip DNA has shown that 11 of the offspring are transgenic for the human ATase gene, between 1 and at least 30 copies of the gene being detected. Human ATase transcripts were detected in total RNA extracted from liver obtained from two male transgenic mice by partial hepatectomy. Cell free extracts of liver samples from five transgenic mice showed up to 4 times higher ATase levels than control livers. INTRODUCTION Alkylating agents exert a wide range of biological effects in both pro and eukaryotes and there is ever increasing evidence that these are mediated via interactions with DNA. The majority of evidence, much of it correlative, indicates that O6-alkylguanine is a mutagenic, toxic and probably carcinogenic lesion (1-4). Repair of 06-alkylguanine in DNA can be mediated by ATase and cells or tissues that express low levels of ATase are generally more susceptible to the toxic and mutagenic effects of alkylating agents than high level expressors (3,4). More compelling evidence for the importance of this lesion in DNA has come from experiments in which mammalian cells sensitive to the toxic effects of alkylating agents have attained increased resistance following transfection and expression of the cloned E. coli ATase genes ada (4), and more recently ogt (5). *
To whom
correspondence, should be addressed
To further examine the possible role of 06-alkylguanine in carcinogenesis by alkylating agents, we have produced ada (6) and ogt (5) transgenic mice in attempts to generate a strain with greatly increased repair capacity. However, we have so far found little orrno evidence of expression of the bacterial ATases (5,6) although more recently, ada-expressing transgenic mice have been described (7,8). The availability of the human ATase cDNA nucleotide sequence (9) has allowed us to produce a human ATase cDNA inducible expression vector which we have shown to achieve high levels of expression in human cells. We also report the generation of human ATase transgenic mice, the presence of human ATase transcripts and up to 4 times higher levels of ATase activity in liver tissue.
MATERIALS AND METHODS Isolation of human alkyltransferase cDNA Oligonucleotides 1075, 207 and 238 (Figure la) were synthesised on a Dupont Coder 3000 using standard phosphoramidite chemistry and purified by preparative polyacrylamide gel electrophoresis (10). Oligo-dT selected RNA (10%g) prepared from human CEM cells was converted to cDNA by specific priming of the first strand using an oligonucleotide (1075, Figure la) complementary to the 3' region of the ATase sequence (9). Second strand synthesis was performed according to Gubler and Hoffmann (11) and the cDNA then amplified by the polymerase chain reaction using human ATase specific oligonucleotides 207 and 238 (Figure la) (9). These create Bam HI restriction endonuclease recognition sites 4bp upstream of the initiation and 12bp downstream of the termination codons respectively. Approximately lOOng of dscDNA was amplified in a total volume of 501d containing lOmM Tris-HCl pH8.3, 1.5mM MgCl2, 1 unit of Taq polymerase (Cetus), lOOng of each oligonucleotide and 400 1tM of each dNTP. After overlaying with 75 yd of paraffin oil, 35 cycles of amplification were performed (93°C-20 seconds, 60°C-1 minute, 72°C-1 minute). Aliquots (5 Id) were then subjected to agarose gel electrophoresis to determine the extent of amplification.
5724 Nucleic Acids Research, Vol. 18, No. 19
a Oligonucleotide 1075 Human ATase cDNA
Oligonucleotide 207 Human ATase cDNA
GTTACACGTGTGTGTCGCTCM CAATGTGCACACACAGCGAGTTTGT
..... AGT
CM
CTTGGTACTTGGATCC ATG GAC MG GAT TGT G CGCTGMCCATGMCCTT TAC CTG TTC CTA ACA C
b rTracription
FcoRl
Oligonucleotide 238 Human ATase cDNA
hAT
I MNT-1 wrww-I
CGCTCAMCATGGATCCTACTGCAC
BamHl
GCGAGTTTGTAGGTAGGATGACGTGTATA AGT CM
Hindll!
I Baml
-j
hGH
Figure 1. (a) Sequences of oligonucleotides used for cDNA synthesis (1075) and amplification (207 and 238). (b)Structure of the 4.2 kb microinjected fragment metallothionein-1 gene (mMT-1) promotor region nucleotides -1700 to -1. of the human ATase (hAT) cDNA expression vector pMThAThGH. mMT-1 5' untranslated region nucleotides + 1 to +68. hGH, transcription unit of the human growth hornone gene. The 648 bp hAT cDNA was ligated into the unique BamHI site. For further details, see text. ,
The amplified 671 bp fragment was digested with BamHI and ligated into BamHI restricted pUC 8.1 to generate phATI. Eukaryotic expression vector construction The 648 bp Bam HI ATase fragment was isolated from phATI and ligated into the vector p(RG48)MT-hGH which consists of the mouse metallothionein-1 gene promotor and most of the transcribed region of the human growth hormone gene containing a number of introns and a polyadenylation signal. This vector is basically similar to that described in (12) but it contains a unique BamHI site. A 4.2 kb EcoRI-HindIll fragment of this construct (Figure lb) was isolated from 2 sequential low melting point agarose gels and made up at 2 ng/4l in buffer for microinjection. Expression of ATase construct in mammalian cells In order to confirm that pMThAThGH was capable of expressing the human ATase the intact plasmid (or p(RG48)MT-hGH in control experiments) was transferred by lipofection (Gibco BRL) into SV40 transformed Xeroderna pigmentosum fibroblasts (XP12ROSV ref 13): 18 ug of plasmid DNA was cotransfected with 2,tg of pZipneoSV(X)l (14) and cells were selected in MEM containing 10% foetal calf serum and 0.5 mg/ml G418. Resistant clones were isolated after 23 days, expanded and extracts prepared for ATase assay. Production of human ATase transgenic mice B6D2F2 mouse embryos were microinjected using standard procedures (15). Drop cultures were incubated at 37°C in an atmosphere of 95% air 5% CO2 overnight and surviving embryos were implanted into pseudopregnant females. Offspring were weaned at 28 days and tail tips were removed for DNA isolation and analysis. Partial hepatectomy (1/3) provided liver tissue for DNA and total RNA extraction, and for ATase assay. DNA isolation and analysis DNA was isolated (15) and 5 /tg samples were BamHI digested, electrophoresed in 1% agarose gels and transferred to nylon filters (Hybond N, Amersham International) essentially as previously described (10). Prehybridisation and hybridisation were as described using a 32P-labelled nick-translated human ATase 648 bp probe (16); Filters were washed sequentially in 6 x, 1 x and 0.1 x SSC at 65°C and exposure was with an intensifying screen for up to 4 days between washes.
E 20 E~~~ 0r
50
100 150 ,ug protein assayed
200
Figure 2. Alkyltransferase activity in extracts of G418 resistant clones isolated after transfection of Xerodema pigmentosum cells with the human ATase cDNA conltaining vector pMThAThGH (0-0, clone Sb; 0-0, clone 6B) or control vector (LI-I, clone lb) see text for further details.
RNA isolation and analysis Fresh liver samples were homogenised using the guanidinium method as recommended (10). The samples were then repeatedly (3 x) phenol:chloroform extracted before the RNA was recovered by sequential isopropanol precipitations. Denatured RNA samples were separated on 1 % agarose gels containing 2.2M formaldehyde and 1 x MOPS running buffer as described (10), except 2yu of 0.5 jig/ml EtBr was added to each sample before loading. This allows visualisation of the major RNA species without the usual background fluorescence associated with this technique. Following electrophoresis (40V, 6 hrs) the gels were washed in water prior to capillary transfer of RNA to Hybond-N in the usual manner (10). An RNA ladder (BRL) was used as a size marker. All hybridisations were performed in 40% formamide, 6 x SSC, 1 % SDS, I x Denhardt's solution and 250 /Ag/ml denatured salmon sperm DNA at 420C.
Alkyltransferase assay Extracts of cells or tissues were prepared and assayed for ATase activity as described (17) except that in some experiments incubation volumes were 500,lI and the amount of 4M perchloric acid added following incubation (2h at 37°C) was reduced to 200,1. Specific activities were calculated from the linear portion of the graph of fmoles [3H]methyl transferred to protein versus amount of protein assayed.
}.1'?8:M.:s,'}.: ,~ ~.
Nucleic Acids Research, Vol. 18, No. 19 5725
Table 1. Status of mice surviving to weaning age Mouse Number
Sex
transgenic status (by BamHl) digest)
Mouse 51 0 42 41 40 3.4 3.3 3.2 3.13-0 23 22 2-1 2.0 1 4 1 3 1.2 1.1 1.0 approx. copy number
}. .... ..... Y
.
.:X::!::}i.:}.::.i:::.:} .:: :.:: :
.. ''''' . . ..
..
..
..
:....
...
..
..
:l: :' ..:....::.: :, .
1.0 1.1 1.2 1.3 1.4 2.0 2.1 2.2 2.3 3.0 3.1 3.2 3.3 3.4 4.0 4.1 4.2 5.0 5.1 5.2 5.3 6.0
6.1 6.2 6.3 6.4 7.0 7.1 7.2 7.3 7.4 8.0 8.1 8.2 8.3
female female female female female female female female female female female female female female female female female female female female female male male male male male male male male male male male male male male
.e..
......::
.::::.:..::..
. a:e: : :.
64Bbp-
;
..' ...
~~~~~~ :..: ~~. .: :...~~~~ :X . . ... . ... .. . . .
.'
....
..
_
_
5
+
30
+
20
+
5 5
+
I
Mouse
52 53 60 61 6-2 63 64 70 7-1 72 73 7480 81 82 83
648bp-
mm
46 4a
+ +
5
+
10
+
20
Figure 3. Southern analysis of tail-tip DNA. 5 jug of DNA was digested with Bam HI, electrophoresed in 1 % agarose and transferred following denaturation to Hybond-N. The filter was prehybridised and hybridised with a 32P-labelled human ATase cDNA probe as described in the text. The position of the 648 bp fragment is shown.
Fluorography SDS-PAGE and fluorography was carried out as previously described (23) except that 16% gels were used.
RESULTS Expression of transgenic vector in human cells The XP cell line used has previously been reported to express very low levels of ATase (less than 2 fm/mg protein, 12). Following lipofection of the ATase vector (pMThAThGH) together with the selectable vector pZipneoSV(x)l, 10 G418 resistant clones were isolated and 3 of the 8 assayed showed marked increases in ATase activity. Fig 2 shows the results of the ATase assay of clones 5b and 6b which had specific activities of 350 and 158 fmoles/mg protein respectively. In contrast, three G418 resistance clones isolated after lipofection of the parent vector together with pZipneoSV(X)1 expressed very low levels of ATase (also 2fm/mg protein). Production of transgenic mice Following microinjection of 700 eggs, 240 survivors were implanted into 12 pseudopregnant D6B1 females (10 per oviduct). DNA was isolated from tail tips (removed after weaning) of the
N 71 73
9.57.544-
2414-
03Figure 4. Northern analyses of total liver RNA from a normal mouse (lane N) and from transgenic mice 7.1 (lane 2) and 7.3 (lane 3). Exactly 10 jig of total liver RNA (as determined by E26 measurements) were run in each lane. Sizes of markers in kb are shown.
5726 Nucleic Acids Research, Vol. 18, No. 19
3 to 0
E
500
400l
-200 1-00-
Controls
60&2 6411 73
Fme 5. Alyltsferase specific acfivities (fmoles/mg total protein) in extracts of normal and transgenic mouse liver.
35 mice born and analysis (see below) indicated that 11 were transgenic, one with in excess of 30 copies of the ATase/cell (Table 1). Nucleic Acid analysis The results of Southern analyses of tail tip DNA are shown in Fig 3. A hybridisation signal at 648 bp was taken as indicative of the presence of the human ATase cDNA. Under the conditions used, hybridisation to the endogenouse mouse ATase gene was not seen. Approximate human ATase cDNA copy numbers are presented in Table 1. Northern analyses showed little or no hybridisation to normal liver RNA whereas in the two transgenic mice that were analysed signals were easily seen at approximately 2.4 kb. These were exactly the size expected from this eukaryotic expression vector (pMThAThGH) and were considerably more intense for mouse 7.3 than for mouse 7.1 (Fig 4). Assay of hepatic aikyltnderase activity As shown in Fig 5, extracts of liver obtained by partial hepatectomy of six norml male control animals contained 96 to 148 fmoles ATase/mg total protein. Extracts of the liver of two male transgenic mice contained ATase activity close to control levels (mice 6.0 and 6.2) two contained twice as much (mice 6.4 and 7. 1) and one more than 4 times as much (mouse 7.3). Fluorography of liver extacts of control and transgenic (7. 1 and 7.3) mice (Fig. 6) showed a band of around 22 kDa, the intensity of which corresponded to the ATase activities shown in Fig. 5.
DISCUSSION The carcinogenic and other biological effects of alkylating agents in mice are well established (18). Of the various lesions produced in DNA, 06-alkylation of guanine in has been implicated in the carcinogenic effects. However, the evidence is correlative and not entirely consistent (19). One approach to providing more conclusive evidence would be to generate transgenic mice that express very high levels of an o6- alkylguanine repair function and to assess whether this protects against these biological effects.
FE1 6. Fluorograph of extracts of the liver of a control mouse (N; ATase sp.act 148 fm/mg protein) and transgenic mice 7.1 and 7.3 following incubation with [3H]MNU-methylated substrate DNA, SDS-PAGE and transfer to nitrocellulose. A 3mm thick 16% acrylamide gel was used and each sample contained 200 gig total protein.
This approach has been used very successfully in a variety of mammalian cells, and more recently in yeast (20) by a number of different groups (reviewed in 4). Although our attempts to produce E. coli ada gene transgenic mice that express high levels of the bacterial ATase have been unsuccessful (6), other groups using different expression vectors have been able to show low levels of ada expression (7,8). The inherent disadvantage of using the ada gene is that the 39 kDa protein encoded contains two ATase active sites that act on alkylphosphotriesters and on 06-alkylguanine and 04-alkylthymine (21). Thus it would be difficult to exclude a possible contribution from any of these products if changes in the responses of such mice to the biological effects of alkylating agents are observed. The ATase encoded by the E.coli ogt gene (22,23) more closely resembles the mammalian equivalent in as much as it lacks activity on alkylphosphotriesters (22). However, though ogt transgenic mice have been produced (5) no indication of expression has been found. The cloning and sequencing of the human ATase cDNA (9) has enabled us to generate a transgenic vector containing this gene. In the vector used, ATase expression is controlled by the heavy metal inducible mouse metallothionein-1 gene promoter: the presence of the human growth hormone sequences downstram provide introns which have been reported to increase expression in transgenic mice (24). The ability of the construct to express at high levels in human cells (Fig. 1) and to increase expression following addition of Zn2+ to the growth medium of these cells (data not shown) confirmed its integrity before proceeding with transgenic mice generation. In the present experiment approx 30 percent of the offspring surviving to weaning age were shown to be transgenic by the presence of a 648 bp BamHI fragment seen on Southern analysis of tail-tip DNA: ATase gene copy numbers ranged from 1 to, in one mouse, greater than 30 per genome. A more comprehensive molecular analysis of these mice will be presented elsewhere. In two of the male transgenic mice subjected to partial hepatectomy, human ATase message-containing ranscripts were
Nucleic Acids Research, Vol. 18, No. 19 5727 readily detectable in total RNA isolated from excised liver: transcripts were very much more abundant in mouse 7.3 than in 7.1. Alkyltransferase activity in normal mouse liver extract was in the range of 96 to 148 fmoles/mg protein, close to that reported for other strains used in such studies (6-8). Two of the five transgenic mice assayed for hepatic ATase activity showed similar levels of activity whereas in the other 3 activity was considerbly higher, particularly in mouse 7.3 which contained more than 4 times the mean control level. Fluorography further confirmed the considerably higher hepatic ATase activity in mouse 7.3 in comparison with 7.1 and a control mouse. However, in the gel system used, no resolution of the transgene and the endogenouse mouse ATases was obtained, probably because of the amounts of protein loaded (200 jig/lane) to detect the ATases. There was a very approximate correlation between the hAT cDNA copy number and the hepatic ATase activity and this correlation was also evident with mRNA abundance, mouse 7.3 appearing to produce considerably more transcripts than mouse 7.1. Large numbers of analyses will be required before any consistent relationships can be established: in the present work we have addressed the question only of whether it is possible to express the human gene in transgenic mice. In E. coli ada gene transgenic mice (7,8) the ATase expressed acts on 06-alkylguanine and alkylphosphotriesters (21,23) and half of the total additional ATase activity is due to the latter function. After correction for this, 06-alkylguanine ATase activity in hAT transgenic mice is approx 3 times higher than in the ada mice. Thus, if 06-alkylguanine is important in the adverse biological effects of alkylating agents, protection against these effects may be expected to be more extensive in the human ATase expressing mice. The tissue-dependence of ATase expression, the production of a homozygous transgenic mouse line and the assessment of the response of these animals to specific biological effects of alkylating agents are in hand.
ACKNOWLEDGEMENTS This work was supported by the Cancer Research Campaign. We thank Miss Michelle D. Boulter and Mrs Eunice M. Guscott for manuscript preparation.
REFERENCES 1. Singer, B. (1979) J. Natl. Cancer Inst., 62, 1329-1339. 2. Saffhill, R., Margison, G.P. and O'Connor, P.J. (1985) Biochim. Biophys. Acta 823, 111-145. 3. Yarosh, D.B. (1985) Mut. Res. 145, 1-16. 4. Margison, G.P. and O'Connor, P.J. (1979) In: Handbook of Experiemtnal Pharmacology, vol. 94/1 (Ed. C.S. Cooper and P.L. Grover), SpringerVerlag Berlin, pp 547-571.5. Harris, L. and Margison, G.P. Unpublished results. 6. Searle, P., Tinsley, J.M., O'Connor, P.J. and Margison, G.P. (1989) Biochem. Soc. Trans. 17, 120-121. 7. Matsukuma, S., Nakatsuru, Y., Nakagawa, K., Utakoji, T., Sugano, H., Kataoka, H., Sekiguchi, M. and Ishikawa, T. (1989) Mut. Res. 218, 197-206. 8. Lim, I.K., Dumenco, L.L., Yun, J., Donovan, C., Warman, B., Gorodetzkaya, N., Wagner, T.E., Clapp, D.W., Hanson, R.W. and Gerson, S.L. (1990) Can. Res. 50, 1701-1708. 9. Tano, K., Shiota, S., Collier, J., Foote, R.S. and Mitra, S. (1990) Proc. Natl. Acad. Sci. USA, 87, 686-690. 10. Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbour Press, New York. 11. Gubler, U. and Hoffman, B.J. (1983) Gene 25, 263. 12. Palmiter, R.D., Norstedt, R.E., Gelinas, R.E., Hammer, R.E. and Brinster, R.L. (1983) Science 222, 809-814.
13. Boyle, J.M., Margison, G.P. and Saffhill, R. (1986) Carcinogenesis 9, 1987-1990. 14. Cepko, C.L., Roberts, B.E. and Mulligan, R.C. (1984) Cell 37, 1053- 1062. 15. Hogan, B., Costantini, F. and Lacy, E. (1984) Manipulating the Mouse Embryo: A Laboratory Manual. Cold Spring Harbour Press, New York. 16. Rigby, P.W.J., Dieckmann, M., Rhodes, C. and Berg, P. (1977) J. Mol. Biol. 113, 237. 17. Morten, J.E.N. and Margison, G.P. (1988) Carcinogenesis 9, 45-49. 18. LARC Monographs on the evaluation of the carcinogenic risk of chemcials to humans. Vols. 1 (1972), 4 (1974), 17 (1978) and Supplement No. 2 (1980). 19. Margison, G.P. and O'Connor, P.J. (1979) In Chemical Carcinogenesis and DNA, Vol. 1 (Ed. P.L. Grover). CRC Press Florida, USA pp 111-159. 20. Brozmanova, J., Kleibl, K., Vlckova, V., Skorvaga, M., Cernakova, L. and Margison, G.P. (1990) Nucleic Acids Res., 18, 331-335. 21. Lindahl, T., Sedgwick, B., Sekiguchi, M. and Nakabeppu, Y. (1988) Annu. Rev. Biochem. 57, 133-157. 22. Margison, G.P., Cooper, D.P. and Brennand, J. (1985) Nucleic Acids Res. 13, 1939-1952. 23. Potter, P.M., Wilkinson, M.C., Fitton, J., Carr, F.J., Brennand, J., Copper, D.P. and Margison, G.P. (1987) Nucleic Acids Res. 22, 9177-9193. 24. Brinster. R.L., Allen, J.M., Behringer, R.R., Gelinas, R.E. and Palmiter, R.D. Proc. Natl. Acad. Sci.(1988) 85, 836-840.
