Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (105K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Ellison, A.
Right arrow Articles by Bishop, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ellison, A.
Right arrow Articles by Bishop, J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© 1996 Oxford University Press 2073-2079

Footnote

Initiation of herpes simplex virus thymidine kinase polypeptides

Initiation of herpes simplex virus thymidine kinase polypeptides Aaron R. Ellison + and John O. Bishop*

Centre for Genome Research, University of Edinburgh, King's Buildings, Edinburgh EH9 3JQ, UK and Department of Biological Sciences, UMBC, Catonsville , MD 21228, USA

Received February 22, 1996; Revised and Accepted April 17, 1996

ABSTRACT

When employed as a transgene reporter, the herpes simplex type 1 virus (HSV1) thymidine kinase gene ( tk ) is ectopically expressed in mouse testis. The principal testicular mRNA lacks the 5 ' -end of the tk reading frame. As a result the principal translation products, P2 and P3, are N-terminally truncated. These co-migrate in SDS-PAGE with polypeptides synthesised during HSV1 infection that were previously thought to be initiated at methionine codons ATG 46 and ATG 60 . Prompted by these observations we generated modified tk genes each carrying only one of the first three ATG codons. Transfected cells expressed both full-length enzyme (P1) and P2 when only ATG 1 was unmodified, P2 and P3 when only ATG 46 was unmodified or P2 and a fourth polypeptide (P4) when only ATG 60 was unmodified. Our observations indicate that P3 is initiated at ATG 46 rather than ATG 60 , while P2 is initiated at a non-ATG codon rather than ATG 46 and P4 is initiated at ATG 60 . When either of two putative non-ATG initiation codons was modified P2 was no longer produced. Cells mainly expressing either P1 or P3 exhibited the same sensitivity to Ganciclovir as cells transfected with the unaltered tk gene. P1 and P3 both have TK activity while P4 probably has none.

INTRODUCTION

BHK cells infected with herpes simplex type 1 virus (HSV1) synthesise three polypeptides from the thymidine kinase gene ( tk ) that can be separated by polyacrylamide gel electrophoresis ( 1 ) and react with antiserum directed against the enzyme (TK) ( 2 ). The 5'-end of the tk coding sequence contains three ATG codons, all in the same reading frame, that could account for the three polypeptides if polypeptide synthesis were initiated at each of them ( 1 ). This interpretation was reinforced by hybrid-arrest translation and by the effects on the pattern of polypeptide synthesis of a 5'-terminal deletion of the tk gene and of a nonsense mutation located between the first and second ATG codons ( 2 , 3 ). However, the N-terminal sequences of the smaller polypeptides have never been determined and the interpretation has remained hypothetical.

Transgenic mice carrying the tk gene as reporter, under the control of a variety of tissue-specific promoters, express the reporter gene ectopically in the testis and this frequently renders the male transgenic carriers infertile ( 4 - 7 ). In the course of a study designed to obviate this undesirable effect of the tk reporter, we generated three variant tk genes each of which carries only one of the three ATG codons, the others being replaced by CTG codons, and a fourth variant in which all three are replaced by CTG codons. When the tk gene that carries only ATG 1 was employed as a reporter in transgenic mice, male-fertile transgenic lines were generated ( 7 ).

Here we describe experiments that explore the properties of the variant tk genes. By transfecting cells with plasmids that carry them we were able to determine which polypeptides each expresses and the relative level of HSV-TK activity due to each. The results indicate that the polypeptide previously thought to be initiated at ATG 46 is initiated at a non-ATG codon, while the polypeptide previously thought to be initiated at ATG 60 is initiated at ATG 46 .

MATERIALS AND METHODS

Plasmid construction

Plasmid pTK1 ( 8 ) carries a 3.5 kb Bam HI fragment from HSV1 (strain CL101) containing the viral thymidine kinase gene cloned in the pAT153 vector. To facilitate sequence changes a Bgl II site was introduced into the polylinker of pBluescript KS(-) and a 502 bp Bgl II- Sac I fragment of pTK1 which contains the region of interest was subcloned between Bgl II and Sac I sites of the vector (plasmid pBS). ATG 1 was altered to CTG by the Kunkel method ( 9 ), employing the antisense oligonucleotide 5'-TACGAAGCCAGACGCGCTTCT-3', generating plasmid pBS[beta][gamma]. ATG 46 and ATG 60 were altered to CTG codons by replacing a 79 bp Dsa I- Fok I fragment with a pair of complementary nucleotides that contained the required sequence (pBS[alpha]). A sequence containing all three altered ATG codons was generated by substituting an 83 bp Bgl II- Mlu I fragment of pBS[beta][gamma], containing the CTG alteration, for the corresponding fragment of pBS[alpha] (pBS0). Sequences containing only ATG 46 (pBS[beta]) and ATG 60 (pBS[gamma]) were generated by exchanging 202 bp Bgl II- Bpm I fragments between pBS[beta][gamma] and pBS0. Bgl II- Sac I fragments from pBS[alpha], pBS[beta], pBS[gamma] and pBS0 were substituted into pTK1 to generate plasmids pTK1[alpha], pTK1[beta], pTK1[gamma] and pTK1.0, respectively.

The non-ATG codons ACG and CTG (shown in Fig. 3 ) were altered by a PCR-based method, employing upstream sense primers (see Fig. 3 ) that incorporated the alterations required (TCG and TTG) and a downstream antisense primer (5'-TCGATGTGTCTGTCCTCCGGAAGG-3') to generate double-stranded oligonucleotides. These were digested with Spl I and Sac I and each was incorporated into both pBS[beta] and pBS[gamma]. The appropriate Bgl II- Sac I fragments of these four plasmids were introduced into pTK1, giving the plasmids pTK1[beta]TTG, pTK1[beta]TCG, pTK1[gamma]TTG and pTK1[gamma]TCG. All alterations were verified by nucleotide sequence determination.

Stably transfected cells

BHKTK - cells were maintained in Dulbecco's modified minimal medium (DMEM) supplemented with 10% fetal calf serum, 5 U/ml penicillin and 5 [mu]g/ml streptomycin, transfected with linearized DNA by CaPO 4 co-precipitation ( 10 ) and selected by growth in HAT medium (DMEM supplemented as above and with 15 [mu]g/ml hypoxanthine, 0.2 [mu]g/ml aminopterin and 5 [mu]g/ml thymidine). Five millimetre colonies were isolated and maintained in HAT medium.

Transient expression

Twenty-four hours prior to treatment 2.5 * 10 5 BHKTK - cells/dish were seeded onto 60 mm diameter dishes. Transfection was by CaPO 4 co-precipitation ( 10 ) employing 2.5 [mu]g circular control plasmid [pA10CAT2 ( 11 )] and 5 [mu]g circular experimental plasmid DNA per dish. Forty-eight hours after transfection cells were washed with ice-cold phosphate-buffered saline (PBS), scraped from the dishes, washed twice in PBS by centrifugation, suspended in 0.5 ml TKB per dish, disrupted by sonication and clarified by centrifugation (8000 g , 15 min), all at 0-4oC. (TKB: 10 mM KCl, 10 mM NaF, 2 mM MgCl 2 , 1 mM ATP, 50 mM [epsilon]-amino caproic acid, 10 mM Tris-HCl, pH 7.5). Protein concentrations were determined by the Bradford method ( 12 ).

Thymidine kinase assays were carried out (in duplicate) as described ( 13 ) employing 1.7 [mu]M [ 3 H]thymidine (29 Ci/mmol) and 170 [mu]M TTP to inhibit cellular thymidine kinases ( 14 ). Aliquots (25 [mu]l) were spotted on 2.5 cm DE-81 disks (Whatman), washed in 10 mM Tris-HCl, pH 7.5 and three times in 95% ethanol and air dried. Radioactivity was estimated by liquid scintillation. One Unit of kinase activity converts 1 pmol of thymidine to a phosphorylated form in 1 min.

Chloramphenicol acetyltransferase (CAT) assays were carried out in duplicate as described ( 15 ), employing 0.5-5 [mu]g protein in 20 [mu]l cell extract in a total volume of 50 [mu]l. After heating for 15 min at 65oC, this was combined with 250 [mu]l CAT reaction mixture containing 0.25 [mu]Ci [ 3 H]acetyl CoA in a 5 ml liquid scintillation vial, overlaid with 3 ml Econofluor-2 (New England Nuclear), and vials were counted for 1 min at regular intervals for 4-6 h. CAT activity was calculated by comparison with parallel reactions containing known amounts of purified CAT (Pharmacia LKB). One Unit of CAT activity acetylates 1 nmol chloramphenicol in 1 min.

Thymidine kinase specific activities were normalised to accommodate different transfection efficiencies by reference to the results of the CAT assays.

Western blotting

Cells were removed from culture dishes as described above and sonicated in a buffer containing 10 mM Tris-HCl, 100 mM NaCl, 1 mM EDTA, 2 [mu]M thymidine, 50 mM [epsilon]-amino caproic acid ( 13 ), 50 mM [rho]-hydroxymercuribenzoate ( 16 ) and 250 [mu]M phenylmethylsulfonyl fluoride. Residual proteolytic activity was assayed by testing for the hydrolysis of fluorescent oligopeptides that contain cleavage sites for a range of serine and cysteine proteases (PepTag, Promega). Only samples free of proteolytic activity were analysed further. Aliquots containing 1-20 [mu]g protein were run on 12% polyacrylamide sodium dodecyl sulphate gels layered with 5% stacking gels ( 17 ) at 2 V/cm for 16 h. Proteins were transferred to Immobilon-P membranes (Millipore) by semi-dry blotting, stained with 0.1% Ponceau S (Sigma) to visualise marker proteins (BioRad, M r 14.5-97.4), treated with 5% non-fat milk (Marvel) in phosphate-buffered saline (PBS) for 2 h and washed with 0.1% Tween-20 in PBS (Tween-PBS). To detect HSV-TK polypeptides membranes were incubated for 3 h with 1/3000 rabbit anti-HSV-TK serum, washed 3 times for 15 min with Tween-PBS, incubated for 2 h with 1/1000 biotinylated donkey anti-rabbit IgG (RPN-1004, Amersham), washed twice in Tween-PBS and once in 0.1% Tween-20 in Tris-buffered saline (Tween-TBS), incubated for 1 h in 1/1000 streptavidin-alkaline phosphatase conjugate (RPN-1234, Amersham), washed three times in Tween-TBS and finally developed with 5-bromo-4- chloro-3-indoyl phosphate and nitro-blue tetrazolium. The apparent sizes of immunoreactive polypeptides measure in this way were (mean M r +- SEM): P1, 43 600 +- 600; P2, 40 400 +- 500; P3, 39 800 +- 500; P4, 36 800.

Thin layer chromatography

Thymidine kinase assays were carried out as described above. An AL SilG/UV plate (Whatman) was prepared by spotting 40 nmol each of thymidine, TMP, TDP and TTP on each station. Eight [mu]L of each reaction mix was spotted in the same positions and the plate was developed with 60% n-propanol, 20% ammonia v/v in water. After migration plates were dried and viewed under UV light to locate the markers. The spots were scraped from the plates into scintillation vials and their radioactivity determined.

Sensitivity of stably-transfected cells to antiherpetic drugs

Sensitivity to three nucleoside analogues, namely 9-[(1,3- dihydroxy-2-propoxy)methyl]guanine (Ganciclovir), (E)-5-(2- bromovinyl)-2'-deoxyuridine (BVdU) and 1-(2-deoxy-2-fluoro- [beta]-D-arabinofuranosyl)-5-iodouracil (FIAU), was tested. 10 5 cells were seeded into each well of two 6-well culture dishes and grown overnight in 5 ml HAT medium. The nucleoside analogue was added to 11 of the wells to final concentrations of 0.1, 0.17, 0.3, 1.0, 3.0, 5.5, 10, 30, 100, 200 and 300 [mu]M. Seventy-two hours later the medium was removed, the wells were washed twice with PBS and cells were stripped with 500 [mu]l 0.05% w/v trypsin, 0.02% w/v EDTA, 0.085% NaCl. Each sample was adjusted to 20 ml with 100 mM NaCl, 10 mM Na-citrate, 0.01% formaldehyde and cell density was estimated as the average of four separate aliquots counted using a Coulter Model ZM counter. The best fit sigmoid competition curve relating cell number to log 10 (drug concentration) was derived by iteration to evaluate the IC 50 value (InPlot 3.5, GraphPad Software Inc.).

Minimum free energy RNA structures

These were computed for the 1308 bp sequence of HSV1 tk mRNA (accession number J02224) using the MFOLD program of the Wisconsin Package (Version 8, September 1994, Genetics Computer Group, 575 Science Drive, Madison, WI 53711, USA) which employs the algorithm of Zuker ( 18 , 19 ) and displayed with the Squiggles program employing the EmuTek PC interface.

RESULTS

ATG 1 was changed to a CTG codon by site-specific mutagenesis. The ATG 46 and ATG 60 codons of a different copy of the tk gene were simultaneously altered to CTG codons by replacing a 79 bp Dsa I- Fok I fragment with a double-stranded oligonucleotide carrying the desired changes (plasmid pTK1[alpha], Fig. 1 ). Restriction enzyme fragments were rearranged to generate plasmids pTK1[beta] (containing ATG 46 but not ATG 1 or ATG 60 ), pTK1[gamma] (containing ATG 60 but not ATG 1 or ATG 46 ) and pTK1.0 (containing none of these ATG codons) and the alterations were confirmed by DNA sequencing. (See the Materials and Methods section for details of plasmid construction.)


Figure 1 . Diagram illustrating modified tk genes. The top line shows the configuration of transcription and translation signals in the wild-type tk gene carried in plasmid pTK1. The lines below show the arrangement of ATG and CTG codons in the modified genes carried in the plasmids listed.

Expression of HSV-TK activity in transfected BHK cells

BHK tk - cells were transfected with plasmids carrying the wild-type and mutant tk genes and TK + colonies were selected in HAT medium. Abundant colonies developed in cultures transfected with pTK1, pTK1[alpha] and pTK1[beta], but none developed in cultures transfected with pTK1[gamma] or pTK1.0, presumably because these plasmids conferred on the cells either no HSV-TK activity or a level insufficient for survival in HAT medium. HSV-TK activities due to pTK1 (70.9 +- 16.3 sem U/mg protein), pTK1[alpha] (76.3 +- 12.2 sem U/mg) and pTK1[beta] (70.3 +- 11.9 sem U/mg) were essentially the same. A similar result previously obtained with tk when under the control of different cellular promoters was attributed to narrow selective limits imposed by HAT medium on cells expressing HSV-TK ( 20 ).

To avoid effects of this sort due to selection pressure and obtain estimates of the activities of the proteins encoded by the different plasmids, HSV-TK activity was measured during transient expression. So as to eliminate differences due to transfection efficiency, each plasmid was co-transfected with pA10CAT2, which expresses CAT. Results (Table 1 ) are expressed as the ratio of HSV-TK to CAT activity, and so are not comparable in absolute terms to the results obtained with stably transfected cells. In this series of experiments cells transfected with pTK1.0 exhibited only the basal TK activity found in cells transfected with a control plasmid. Cells transfected with pTK1[gamma] had a low HSV-TK activity, which was significantly different from the basal activity in cells transfected with pTK1.0 or the control plasmid (Student's t test: P = 0.0490 and 0.0697, respectively). Also, cells transfected with pTK1[alpha] had a lower HSV-TK activity than cells transfected with either pTK1 or pTK1[beta] (Student's t test: P = 0.0748 and 0.0212 respectively).

Table 1 . HSV-TK activity in extracts of cells transfected with wild-type and modified tk genes during transient expression
Plasmid a

Mean HSV-TK activity b +- SE

Range

pTK1

92.2 +- 24.0

33.5-134.4

pTK1[alpha]

40.2 +- 8.9

24.9-65.4

pTK1[beta]

65.1 +- 13.3

38.6-97.3

pTK1[gamma]

11.3 +- 1.6

7.8-11.8

pTK1.0

4.50 +- 1.87

0.8-9.5

None

4.58 +- 1.25

1.1-6.8

a Cells were transfected with 2.5 [mu]g pA10CAT2 per 6 cm dish and 5[mu]g of the tk plasmid stipulated in column 1 or no tk plasmid, and protein extracts prepared 48 h later. b Extracts were assayed for HSV-TK activity and chloramphenicol acetyltransferase activity. The values given record the ratio of HSV-TK to CAT activity, both expressed in Units. The values and standard errors are based on four independent transfection experiments in each case.

HSV-TK polypeptides synthesised in transfected cells

Cells were disrupted in the presence of protease inhibitors, and only extracts that tested negative for residual protease activity using fluorescent indicator oligopeptides (Promega Pep-Tag, see Methods section) were analysed further. HSV-TK polypeptides were detected using a polyclonal rabbit antiserum following SDS-PAGE. Qualitatively very similar results were observed with stably and transiently transfected cells in those instances in which both could be generated (Fig. 2 ). In cells transfected with pTK1 and pTK1[alpha] the most prominent polypeptide (P1) had an apparent M r of ~43 000, the apparent size previously recorded for HSV-TK ( 1 , 2 , 16 ). An additional prominent M r 39 800 polypeptide (P3) was present in cells transfected with pTK1, but not pTK1[alpha]. This band was very prominent in cells transfected with pTK1[beta]. A polypeptide (P4) with approximate M r 36 800 was present only in cells transfected with pTK1[gamma], which contains ATG 60 but neither ATG 1 nor ATG 46 . These results are consistent with the interpretation that P1, the full-length enzyme, is initiated at ATG 1 , while P3 and P4 are initiated at ATG 46 and ATG 60 respectively. Note that the prominent truncated polypeptide present in HSV1-infected cells co-migrates with P3. P3 cannot be initiated at ATG 60 , which is not present in the pTK1[beta] plasmid, and most probably is initiated at ATG 46 , the only one of the three in-frame initiation codons present in that plasmid.


Figure 2 . HSV-TK polypeptides present in extracts of cultured transfected BHK cells and in transgenic tissues. ( a ) Stably transfected cells and transgenic tissues. Lane 1, cells infected with HSV1; lanes 2-7, BHKTK - cells stably transfected with the plasmids indicated; each track corresponds to a different clone; lane 8, non-transfected BHKTK - cells; lanes 9 and 10, thyroid and testis extracts from transgenic line TG66.19, which carries the bTG-tk1 transgene (41). ( b ) Extracts of transfected BHKTK - cells prepared during transient expression. Lane 1, mock-transfected cells; lanes 2-9, cells transfected with the plasmids indicated; each lane represents a separate cell culture. The approximate M r values of P1-P4 are 43 000, 40 400, 39 800 and 37 000 respectively. Extracts were run on SDS-PAGE gels and transferred to Immobilon membranes, and HSV-TK polypeptides were identified by specific immunoreactivity, all as described in Materials and Methods.


Figure 3 . Expression of tk genes with altered non-ATG codons. ( Top ) The sequence of the tk reading frame between nucleotides 1 and 138. Suspected non-ATG initiation codons are shown italicised and underlined, and the single nucleotide substitutions introduced are shown below. The substitutions were made by PCR using as 5' primers oligodeoxynucleotides complementary to the regions indicated by lines above the sequence. The same downstream reverse primer was employed in both cases (see Materials and Methods). The designations of the plasmids into which the modified sequences were introduced are also shown. ( Bottom ) Western blot of extracts of BHKTK - cells transfected with various plasmids during transient expression. The transfecting plasmids were: lane 1, pTK1; 2, pTK1[alpha]; 3, pTK1[beta]; 4, pTK1[beta]TCG; 5, pTK1[beta]TTG; 6, pTK1[gamma]; 7, pTK1[gamma]TCG; 8, pTK1[gamma]TTG; 9, pTK1.0. Lanes 10 and 11 contain extracts of non-transfected cells and HSV1-infected cells respectively. The approximate M r values of P1-P3 are 43 000, 40 400 and 39 800, respectively.

An additional polypeptide (P2) with M r 40 400 is present in most extracts irrespective of the status of the ATG codons, and is prominent in extracts of cells transfected with pTK1[beta] and pTK1[gamma]. In the latter extract it is the more prominent of the two bands. Synthesis of this polypeptide was previously thought to be initiated at ATG 46 ( 1 , 2 , 16 ). The new observations presented here indicate that the synthesis of this polypeptide is initiated upstream of ATG 46 (because it migrates more slowly than P3) and downstream of ATG 1 . If this interpretation is correct, P2 must be initiated at a non-ATG codon. The sequence between ATG 1 and ATG 46 contains at least two likely non-ATG initiation codons ( 21 , 22 ). In order to test the new interpretation, each of these was altered separately (Fig. 3 ) and the altered sequences were introduced into the pTK1[beta] and pTK1[gamma] plasmids. Extracts of transfected cells were analysed as before (Fig. 3 ). Unexpectedly, both alterations abolished the synthesis of P2. We conclude from this result that both alterations changed the context ( 21 - 23 ) or secondary structure ( 23 , 24 ) of the mRNA in such a way as to abolish non-ATG translational initiation, which may originate at either of these codons or at another codon.

The bTG-tk1 transgene consists of the tk reporter under the control of the bovine thyroglobulin promoter. In mice carrying this transgene a full-length mRNA initiated at the thyroglobulin cap site is expressed in thyroid follicle cells, while mRNA ectopically expressed in the testis is initiated principally at several sites downstream of ATG 1 ( 25 ). In the thyroid follicle cells the principal product is P1 and relatively very small amounts of P2 and P3 are also synthesised (Fig. 2 ). In the testis the main products are P2 and, at a much higher level, P3, in proportions similar to those obtaining in cells stably transfected with pTK1[beta] (Fig. 2 ). In common with HSV1-infected cells, P4 is not present in either thyroid or testis of the transgenic mice.

When the altered tk [alpha] gene was substituted for tk in the thyroglobulin promoter-driven construct ( bTG-tk1 [alpha]) no P3 was detected in the testis of transgenic animals. However, a significant amount of P2 was present together with, as expected, a trace amount of P1 (data not shown). These mice had much reduced thymidine kinase activity in the testis relative to bTG-tk1 transgenic mice and were fertile ( 7 ). Their thyroids had a high level of P1 and a trace of P2 but, unlike bTG-tk1 thyroids, they contained no P3 (data not shown, cf. Fig. 2 ).

Thymidylate kinase activity of HSV-TK polypeptides

HSV-TK has in addition a TMP kinase activity ( 3 , 25 , 26 ). To investigate the possibility that different TK polypeptides have more or less TMP kinase activity, [ 3 H]thymidine was incubated with extracts of cells stably transfected with pTK1, pTK1[alpha] and pTK1[beta], and the different phosphorylated forms were separated by TLC and quantified (Fig. 4 ). Our experiments differ from those of previous authors in two ways. First, we employ reading frames altered to contain a different complement of initiation codons rather than a deletion and nonsense codons. Secondly, we employ transfected rather than virus-infected cells with their attendant problem of host-cell degeneration. Extracts of cells transfected with pTK1[alpha] produced a higher ratio of TTP to TMP than extracts of cells transfected with pTK1[beta] (0.28 versus 0.09). The amounts of TDP present were very low in all cases, but TDP can be assumed to be raised rapidly to TTP by cellular kinases, and consequently provides a measure of HSV-TK-dependent TDP formation. In agreement with previous reports ( 3 , 26 ) these results show that P1, the predominant polypeptide in cells transfected with pTK1[alpha], has more thymidylate kinase activity than P3, the predominant polypeptide in cells transfected with pTK1[beta].


Figure 4 . Relative amounts of different phosphorylated forms of thymidine generated during incubation of [ 3 H]thymidine with extracts of BHKTK - cells stably transfected with plasmids carrying wild-type and modified tk genes. Reactions were carried out as described in the Materials and Methods section, allowing the phosphorylation of <7% of the radiolabelled substrate. Phosphorylated and non-phosphorylated forms were resolved by TLC. Each value is the average of at least four independent determinations. Standard errors for TMP were 3.4 (pTK1), 4.1 (pTK1[alpha]) and 1.1% (pTK1[beta]) and for TTP 1.8 (pTK1), 2.9 (pTK1[alpha]) and 2.0% (pTK1[beta]).

Sensitivity of transfected cells to antiherpetic drugs

Previous studies described the antiviral effects of antiherpetic agents on cells infected with normal ( 27 - 29 ) and mutant ( 26 ) HSV1 and cytostatic effects on cells transfected with wild-type HSV1-tk ( 30 - 32 ). The experiments to be described are specifically directed towards induction of cell death, in cells carrying wild-type and altered forms of tk , by agents that are candidates for use in the selective destruction of TK-positive cells.

BHKTK + cells and also BHKTK - cells stably transfected with three of the plasmids were challenged with Ganciclovir, BVdU and FIAU and cell proliferation was measured in replicate experiments (Table 2 ). Proliferation of cells transfected with all three plasmids was equally sensitive to Ganciclovir and BVdU, with inhibition of proliferation (IC 50 ) values ~100-fold (Ganciclovir) and 10-fold (BVdU) less than those of the BHKTK + cells, which contain the normal complement of cellular nucleotide kinases. FIAU was more toxic to normal cells and, as in the case of BVdU, the IC 50 values of the transfected cells were ~10-fold lower; the observed differences between cells transfected with pTK1 and either pTK1[alpha] or pTK1[beta] are statistically significant ( P = 0.034 and 0.069, respectively) but this is not understood. Clearly the greatest discrimination between normal and transfected cells is effected by Ganciclovir in all cases.

Table 2 . Sensitivity to antiherpetic drugs of cells stably transfected with wild-type and modified tk genes
Plasmid a

Antiherpetic Agent

Ganciclovir

BVdU

FIAU

Mean +- SE

n b

Mean +- SE

n

Mean

n

None c

210 +- 74

6

209

3

20

2

pTK1 d

1.50 +- 0.36

7

31.3

3

2.77

3

pTK1[alpha] d

1.03 +- 0.28

4

19.7 +- 4.0

4

0.55

3

pTK1[beta] d

0.90 +- 0.26

4

30.9 +- 6.8

4

0.71

3

a Cells stably transfected with the stipulated plasmid. b Number of independent determinations. The same cell line was employed for each BHKTK + replicate; in the case of transfected BHKTK - cells, each replicate experiment employed a different clonal isolate. c BHKTK + cells, not transfected. d Stably-transfected BHKTK - cells.

DISCUSSION

The data presented here show that the earlier hypothesis that P2 and P3 are initiated at ATG 46 and ATG 60 is incorrect. The key observations are that P3 is present in cells transfected with pTK1[beta], which does not contain ATG 60 and that P4 is present only in the cells transfected with pTK1[gamma], which contains ATG 60 but not ATG 1 or ATG 46 . The latter observation is consistent with the scanning model of eukaryote translation initiation ( 33 ): ATG 46 lies within a strong Kozak ( 34 ) consensus ( A AA ATG C ) which, when present, may trap all the scanning ribosomes. The fact that cells transfected with pTK1 synthesise both P1 and P3 can be explained by the weaker consensus ( C GT ATG G ) surrounding ATG 1 . Some ribosomes presumably pass over ATG 1 without initiating a polypeptide chain and are able to initiate at ATG 46 when it is present.

The explanation that translation of P2 is initiated from one or more non-ATG codons lying between ATG 1 and ATG 46 would resolve lingering problems associated with the earlier explanation. First, it was problematical that the amount of P2 was always very much less than that of P3, when the Kozak consensus of its supposed initiation codon, ATG 46 , is strong (see above) and precedes the supposed P3 initiation codon, ATG 60 . According to the new explanation the lesser production of P2 is due to the fact that it is initiated at one or more non-ATG codons. Secondly, virus carrying a small deletion between -120 and +81 (i.e. upstream of the non-ATG initiation site) produced P2 but a virus carrying a nonsense mutation two codons upstream of ATG 46 (i.e. downstream of the non-ATG initiation site) did not ( 2 ). The latter result is unexpected in the context of the earlier explanation, because the non-initiated ribosomes would be indifferent to the nonsense codon. However if, as suggested here, P2 is initiated upstream of the nonsense codon, its translation would be terminated by it. Thirdly, the HSV-TK activity present in cells that mainly express P3, i.e. transgenic testis ( 25 ) and now also cells transfected with pTK1[beta] (Table 1 ), is unexpectedly high in view of the fact that the predominant P3, if it were initiated at ATG 60 , would lack a functional ATP-binding pocket ( 35 ). In contrast, according to the new explanation the polypeptides present in those cases are the minor P2 and the predominant P3, both of which will contain an intact ATP binding site and can be expected be enzymatically active. The modest HSV-TK activity that we detected in cells transiently transfected with pTK1[gamma] is presumably due to P2, which appears to be produced in higher amounts from this plasmid.

However, the new explanation is not without problems of interpretation. (i) Cells transfected with pTK1[gamma] express P4 at an unexpectedly low level, and (ii) P2 is expressed at a higher level in cells transfected with pTK1[beta] than with pTK1[gamma], and in cells transfected with pTK1.0 P2 is not expressed at all (Fig. 2 ). (i) It is possible that secondary structure does not favour the utilisation of ATG 60 ( 36 , 37 ) which in addition follows a long (285 nucleotide) untranslated region ( 38 ). (ii) In the absence of ATG 46 the secondary structure of the mRNA may possibly be inimical to non-ATG chain initiation, which is particularly sensitive to perturbation ( 24 , 39 ). Although unexpected, these observations do not outweigh the evidence in favour of the new explanation. However, an unequivocal confirmation of the new explanation will require N-terminal sequencing of the polypeptides.

The abolition of P2 synthesis by alteration of either ACG 38 or CTG 42 was also unexpected. Minimum free energy folding structures were computed for the mRNA encoded by pTK1[beta] and the two altered plasmids, pTK1[beta]TCG and pTK1[beta]TTG. AUG 46 was located in the same 6-base pair hairpin stem-5-base pair loop structure in all three cases. There was no difference in structure between pTK1[beta] and pTK1[beta]TTG RNA (C -> T). A small local difference, an increase in the length of a stem from 9 to 11 base pairs, characterised the pTK1[beta]TCG RNA (A -> T), and brought the first nucleotide of the altered AUG codon into the longer stem, but left the configuration around the CUG codon unchanged. Thus, perhaps not surprisingly, modelling RNA secondary structure in this way gave no indication as to the identity of the non-ATG initiation codon(s). It would appear, again, that a definitive answer could best be obtained by physically isolating P2 and determining its N-terminal amino acid sequence.

Cells transfected with pTK1, which expressed P1 and P3 in a ratio of about 10:1, exhibited the same level of HSV-TK activity as cells transfected with pTK1[beta], which expressed P2 and P3 in a ratio of about 1:10. This indicates that the TK activities of P1 and P3 are essentially the same. However, cells transfected with pTK1[alpha], which expressed P1 almost exclusively, exhibited lower HSV-TK activity than either of the above. These observations point to an effect of one of the amino acid changes introduced into the pTK1[alpha] plasmid. The Met60 -> Leu alteration seems to be neutral. The pTK1[alpha] and pTK1[beta] plasmids, which share the alteration, have different activity levels, while the pTK1 and pTK1[beta] plasmids, which do not share it, have the same activity level. The Met46 -> Leu alteration, on the other hand, is carried by pTK1[alpha], with its low activity level, but not by the other two plasmids. Thus this amino acid substitution is the likely cause of the reduced HSV-TK activity of cells transfected with pTK1[alpha].

HSV-TK activities recorded during transient expression (Table 1 ) avoid the levelling effect of HAT selection on stably-transfected cells and, given that they were normalised for transfection efficiency, should represent accurately the activities of the polypeptides produced. Conversely, TMP and TTP accumulation (Fig. 4 ) were measured using extracts of stably transfected cells, which have approximately equal TK activities, thus avoiding mass-action effects. By combining these measurements we can estimate the relative accumulation of TMP and TTP, and hence the TK and TMPK activities of the polypeptides present during transient expression in cells transfected with the three plasmids (since TDP is rapidly phosphorylated by cellular kinases, TTP accumulation is a measure of HSV TMP kinase activity). Cells transfected with pTK1[alpha] have a higher ratio of TMPK to TK activity than cells transfected with pTK1 (Table 3 ). In principle, the lower TK and higher relative TMP kinase activity of the mutant P1 encoded by pTK1[alpha] could have a common cause in an increased affinity of the enzyme for TMP, which could both decrease the rate of TMP generation and increase the rate at which it becomes phosphorylated. Extracts of cells transfected with pTK1[beta] have a particularly low ratio of TMPK to TK activity. The principal difference between P3, the predominant polypeptide in cells transfected with pTK1[beta], and the two (normal and altered) P1 polypeptides is its N-terminal truncation. However P3 also differs from each of the P1 polypeptides in a different amino acid. Thus N-terminal truncation is the likely explanation for the lower TMPK activity of P3 as previously suggested ( 3 , 26 ), but this interpretation is not unequivocal.

Table 3 . Relative TK and TMP kinase activities due to different tk genes
Cells a

HSV-TK

TTP/TMP c

TMPK

TMPK activity/

activity b

activity d

TK activity

pTK1

92

0.17

13.4

0.15

pTK1[alpha]

40

0.28

8.8

0.22

pTK1[beta]

65

0.09

5.3

0.08

a Cells transfected with the stipulated plasmid. b Average normalised HSV-TK activity of extracts of transiently transfected cells (Table 1). c Ratio of TTP to TMP formed during incubation of extracts of stably transfected cells with [ 3 H]thymidine (Fig. 4). d (TK activity * TTP)/(TMP + TTP).

The sensitivity of cells transfected with pTK1, pTK1[alpha] and pTK1[beta] to Ganciclovir was essentially the same, implying that there are no significant differences in Ganciclovir phosphorylation between the polypeptides specified by these plasmids. In accordance with this finding, thyroid follicle cells of mice carrying a thyroglobulin promoter- HSV1-tk [alpha] transgene ( bTG-tk1 [alpha]) were successfully ablated with the same Ganciclovir administration regime previously used to ablate thyrocytes of thyroglobulin promoter- HSV1-tk ( bTG-tk1 ) transgenic mice ( 7 , 40 ). Unlike mice carrying the bTG-tk1 transgene, lines that carry the bTG-tk1 [alpha] transgene are male-fertile and can be bred to homozygosity ( 7 ). Consequently, the HSV1-tk [alpha] mutant reporter gene is a superior tool for use in conditional ablation protocols.

ACKNOWLEDGEMENTS

We are grateful to Cindy-Lou Dull for preparing rabbit polyclonal anti-HSV-TK antiserum, to Steven Harrison for constructing the ATG 46 /ATG 60 alteration, to A. Phelan and B. Clements for gifts of HSV1-infected cells, to Louise Anderson and Yvonne Harcus for assistance with mouse husbandry, to Syntex (Palo Alto) for gifts of Ganciclovir and to the Maryland Agricultural Experimental Station which supported the early part of the work.

REFERENCES

1 Marsden,H.S., Haarr,L. and Preston,C.M. (1983) J. Virol. 46, 434-445.

2 Haarr,L., Marsden,H.S., Preston,C.M., Smiley,J.R., Summers,W.C. and Summers,W.P. (1985) J. Virol. 56, 512-519.

3 Haarr,L. and Flatmark,T. (1987) J. Gen. Virol. 68, 2817-2829.

4 Al Shawi,R., Burke,J., Jones,C.T., Simons,J.P. and Bishop,J.O. (1988) Mol. Cell. Biol. 8, 4821-4828.

5 Braun,R.E., Lo,D., Pinkert,C.A., Widera,G., Flavell,R.A., Palmiter,R.D. and Brinster,R.L. (1990) Biol. Reprod. 43, 684-693.

6 Huttner,K.M., Pudney,J., Milstone,D.S., Ladd,D. and Seidman,J.G. (1993) Biol. Reprod. 49, 251-261.

7 Ellison,A.R., Wallace,H., Al-Shawi,R. and Bishop,J.O. (1995) Mol. Reprod. Dev. 41, 425-434.

8 Wilkie,N.M., Clements,J.B., Boll,W., Mantei,N. and Weissmann,C. (1979) Nucleic Acids Res. 7, 859-877.

9 Kunkel,T. (1985) Proc. Natl. Acad. Sci. USA, 82, 488-492.

10 Spandidos,D.A. and Wilkie,N.M. (1984) In Hames,B.D. and Higgins,S.J. (eds), Transcription and Translation: A Practical Approach. IRL Press, Oxford, pp. 1-48.

11 Rosenthal,N. (1987) Methods Enzymol. 152, 704-720.

12 Bradford,M. (1976) Anal. Biochem. 72, 248-254.

13 Brinster,R.L., Chen,H.Y., Trumbauer,M., Senear,A.W., Warren,R. and Palmiter,R.D. (1981) Cell, 27, 223-231.

14 Jamieson,A.T., Gentry,G.A. and Subak-Sharpe,J.H. (1974) J. Gen. Virol. 24, 465-480.

15 Sambrook,J., Fritsch,E.F. and Maniatis,T. (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

16 Irmiere,A.F., Manos,M.M., Jacobson,J.G., Gibbs,J.S. and Coen,D.M. (1989) Virology, 168, 210-220.

17 Laemmli,U.K. (1970) Nature, 227, 680-685.

18 Zuker,M. (1989) Science, 244, 48-52.

19 Jaeger,J.A., Turner,D.H. and Zuker,M. (1989) Proc. Natl. Acad. Sci. USA, 86, 7706-7710.

20 Lang,J., Wilkie,N.M. and Spandidos,D.A. (1983) J. Gen. Virol. 64, 2679-2696.

21 Grunert,S. and Jackson,R.J. (1994) EMBO J. 13, 3618-3630.

22 Boeck,R. and Kolakofsky,D. (1994) EMBO J. 13, 3608-3617.

23 Kozak,M. (1989) Mol. Cell. Biol. 9, 5073-5080.

24 Kozak,M. (1990) Proc. Natl. Acad. Sci. USA, 87, 8301-8305.

25 Chen,M.S. and Prusoff,W.H. (1978) J. Biol. Chem. 253, 1325-1327.

26 Coen,D.M., Irmiere,A.F., Jacobson,J.G. and Kerns,K.M. (1989) Virology, 168, 221-231.

27 Elion,G.B., Furman,P.A., Fyfe,J.A., de Miranda,P., Beauchamp,L. and Schaeffer,H.J. (1977) Proc. Natl. Acad. Sci. USA, 74, 5716-5720.

28 Cheng,Y.C., Huang,E.S., Lin,J.C., Mar,E.C., Pagano,J.S., Dutschman,G.E. and Grill,S.P. (1983) Proc. Natl. Acad. Sci. USA, 80, 2767-2770.

29 Smee,D.F., Martin,J.C., Verheyden,J.P. and Matthews,T.R. (1983) Antimicrob. Agents Chemother. 23, 676-682.

30 Furman,P.A., McGuirt,P.V., Keller,P.M., Fyfe,J.A. and Elion,G.B. (1980) Virology, 102, 420-430.

31 Uckun,F.M., Dibirdik,I., Smith,R., Tuel Ahlgren,L., Chandan Langlie,M., Schieven,G.L., Waddick,K.G., Hanson,M. and Ledbetter,J.A. (1991) Proc. Natl. Acad. Sci. USA, 88, 3589-3593.

32 Morla,A.O., Schreurs,J., Miyajima,A. and Wang,J.Y. (1988) Mol. Cell. Biol. 8, 2214-2218.

33 Kozak,M. (1987) Mol. Cell. Biol. 7, 3438-3445.

34 Kozak,M. (1986) Cell, 44, 283-292.

35 Liu,Q.Y. and Summers,W.C. (1988) Virology, 163, 638-642.

36 Prats,A., Vagner,S., Prats,H. and Amalric,F. (1992) Mol. Cell. Biol. 12, 4796-4805.

37 Kozak,M. (1988) Mol. Cell. Biol. 8, 2737-2744.

38 Berkner,K.L., Schaffhausen,B.S., Roberts,T.M. and Sharp,P.A. (1987) J. Virol. 61, 1213-1220.

39 Kozak,M. (1991) J. Cell Biol. 115, 887-903.

40 Wallace,H., McLaren,K., Al-Shawi,R. and Bishop,J.O. (1994) J. Endocrinol. 143, 107-120.

41 Wallace,H., Ledent,C., Vassart,G., Bishop,J.O. and Al-Shawi,R. (1991) Endocrinology, 129, 3217-3226.

Return

* To whom correspondence should be addressed + Present address: Cardiovascular Research Institute, School of Medicine, UCSF, San Francisco, CA 94143-0911, USA
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow Print PDF (105K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Ellison, A.
Right arrow Articles by Bishop, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ellison, A.
Right arrow Articles by Bishop, J.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?