1 Qi \ci i 7‘23 6 W llllllllHlllllllllllllllllllllllllllIIHHHI ~ 1 31293 0056122079 A LIBRARY Michigan State University This is to certify that the dissertation entitled A Study of the Regulation of Mammalian Thymidine Kinase Genes presented by Christine Joyce Stewart has been accepted towards fulfillment of the requirements for Ldegreein Microbiology Major professor Date May 2!), I988 MSU i; an Affirmative/icrian/Equul Opportunity Institution 0-12771 )V1ESI_} RETURNING MATERIALS: Place in book drop to LJBRARJES remove this checkout from a your record. FINES W‘iil v be charged if book is returned after the date stamped below. A STUDY OF THE REGULATION OF MAMMALIAN THYMIDINE KINASE GENES BY Christine Joyce Stewart A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF.PHILOSOPHY Department of Microbiology and Public Health 1988 (TKJ manx to e fact have bot! cull resi drar inc1 incl Peri infe time a2. Wit} Shov 5/ 79457 ABSTRACT A STUDY OF THE REGULATION OF MAMMALIAN THYMIDINE KINASE GENES BY Christine Joyce Stewart The regulation of the cytoplasmic thymidine kinase (TR) gene occurs in a cell cycle growth-phase dependent manner. It is known that cells are mitogenically induced to enter the cell cycle by either addition of serum growth factors or by infection with the papovavirus, SV40. I have studied the regulated expression of the TR gene using both types of induction of quiescent mammalian tissue culture cells. TK mRNA and enzyme levels are low in resting Go/G1 phase simian CV-l cells, but increase dramatically by late S/G phase. To determine whether an 2 increase in the rate of transcription was mediating this increase, nuclear run-on transcription assays were performed at various times after serum stimulation or SV40 infection of quiescent CV-l cells. Assays performed at time points spanning the 61/8 phase interface showed a 6-7 - fold increase in serum induced cells, but only a 2.5-3.5 - fold increase in SV40 infected cells. Studies with Actinomycin D treated synchronized cell cultures showed that the TK messenger RNA is extremely stable when the SV4C witt sugg nece the of t stuc cyc] duri but synl can: that hon] hum: Site homc regj mult gene the drug is added in Go/Gl' S and G2 phase, as well as SV40 infected cells in S phase. Results from experiments with 3' end deletion mutants of a TK mini-gene contruct suggest that a poly-adenylation recognition site is necessary for the mediation of TX mRNA stability, while the relatively long 662 base pair 3’ untranslated region of the TR gene is not. TK mRNA and enzyme levels were studied into and beyond the G2 phase of the cell cycle following serum stimulation. Withdrawal of serum during G2 phase caused a decrease in TX enzyme activity, but the mRNA level remained unaffected. Cessation of DNA synthesis early in S phase by the addition of hydroxyurea caused TK enzyme and mRNA levels to increase more slowly .than control levels, and both kept increasing up to 50 hours post serum stimulation. The promoter region of the human TK gene was sequenced and two transcription start sites were mapped by $1 nuclease analysis. A 40 base pair homology common to the hamster and human TK promoter regions was found. Taken together, these results suggest multiple levels of regulation for the thymidine kinase gene. To Mom and Dad for the support, understanding and encouragement that made this possible. iv dis map fri nig me Acknowledgements I would like to thank the members of my committee; Dr. Ron Patterson, Dr. Barb Sears, Dr. Larry Snyder, and Dr. Jerry Dodgson for all of their help throughout the past four and a half years. Thanks also to my advisor, Dr. Sue Conrad for her support. Thanks to Moriko Ito for all the hard work, great discussions, and especially for being such a good friend and helping me through all the rough times. Thanks to Paul Boyer for help with sequencing and $1 mapping experiments, for being such an incredibly loyal friend, and especially for feeding me during the late night shifts. Thanks to the cousins--Don P. and Tom for teaching me how to handicap and helping to improve my golf game. Thanks to the members of the Brubaker lab for adopting me and showing me that science can actually be a heck of a lot of fun. ‘ Last, but not least, thanks to Rick Mehigh for believing in me and for all the love and encouragement throughout the final phases of my stay here. LIS LIS INT CHA CHAI TABLE OF CONTENTS LIST OF TABLES OOOOOOOOOOOOOOOOOO ....... OOOOOOOOOOOOViii LIST OF FIGURES oto.o00.000000000000000000.0000...ooooix INTRODUCTION.OOOOOOOOOOOOOOOOOOO ..... OOOOOOOOOOOOOOOOOOI CHAPTER I Literature Review Biochemical Aspects of Thymidine Kinase... ..... 3 TR Isozymes....................................9 Expression of TX in the Cell Cycle............10 TK Molecular Structure........................18 Other Cell Cycle Regulated Genes..............24 Viral Induction of TX Activity................29 References....................................41 CHAPTER II ARTICLE: Evidence for Transcriptional and Post- Transcriptional Control of the Cellular Thymidine Kinase Gene.......................52 Abstract....................................53 Introduction................................55 Materials and Methods.......................58 Results.....................................64 Discussion..................................90 References... ...... ......... ................ 97 CHAPTER III ARTICLE: Rhythmic Expression of the Cytoplasmic Thymidine Kinase Gene is Regulated by Multiple Levels of Cell Cycle Dependent Controls....................................100 Abstract............... ........ .............101 Introduction.......... ..... .................103 Materials and Methods.......................106 Results...................... ..... ..........112 Discussion.......... ..... . ....... ..... ...... 133 References........... .................... ...140 vi SUM APP APP CHAPTER IV ARTICLE: Sequence Analysis of the Human Thymidine Kinase Gene Promoter Region.......145 Abstract....................................146 Introduction........................ ..... ...147 Materials and Methods.......................150 Results.....................................154 Discussion..................................164 References............r.....................168 SWARYAND CbNCLUSIONS...00.000.000.000000000.000.001.73 APPENDIX I ARTICLE: Induction of Cellular Thymidine Kinase Occurs at the mRNA Level..............177 APPENDIX II NOTE IN PROOF: TK Transcriptional Regulation in SV40 Infected Cells...................185 vii —I’—* < - - ———.—-—:;...- LIST OF TABLES AB E PAGE CHAPTER II 1 Quantitation of Nuclear Run Ons.......83 2 TR Enzyme Assays-Transfected Cells....89 CHAPTER III 1 TR RNA values post Actinomycin D......116 viii CHAP CHAPf CHAPl APPEE APPEN LIST OF FIGURES FIGURE PAGE CHAPTER I NNH CHAPTER II CDQONUlohNNI-J CHAPTER III mflOXUIJE-le-i CHAPTER IV hLAk)H APPENDIX I R>H APPENDIX II 1 2 Deoxythymidine Triphosphate Synthesis......5 Cell Cycle.. ...... ........................14 SV40 Genome...............................32 TK cDNA Structures........................61 Northern Blot Analysis of TR mRNA.........67 Nuclear Run On-lz Hour Time Points........70 Measurement of DNA Synthesis..............74 Nuclear Run On-l Hour Time Points-Serum...77 Nuclear Run On-l Hour Time Points-SV40....79 Nuclear Run On-Early SV40 Infection.......82 Northern Blot of Transfected Contructs....88 Functional TK cDNAs......................lll TK mRNA Stability in CV-l Cells..........115 Quantitation of Densitometer Scans.......118 DNA Synthesis-Serum Induced CV-l Cells...120 TK 3’ Deletion Mutation Analysis.. ....... 123 Late S/G2 Phase TK Enzyme Activity.......127 Late S/G2 Phase TK mRNA Levels...........129 Densitometer Analysis of Fig. 7..........131 Structure of p5’TKcDNA....... ..... .......153 Human TK Promoter Sequence...............157 Sl Nuclease Experiment ................... 160 Dot Matrix Analysis. ..................... 163 Restriction Map of Human TK Locus........179 Northern Blot Analysis of SV40 Infected CV-l Cells...............................180 SV40 Induction of TK in CV-l Cells.......181 Tewnty-four Hour Time Course of SV40 Infected CV-l Cells.. ..... . ..... .........l8l Densitometer Analysis.... ....... ... ...... 182 Comparison of Serum and SV40 Induction...182 Nuclear Run On-Early SV40 Infection ...... 188 Graphic Analysis of Figure 1 ............. 190 ix IN’l thy 0nl prc am the kno cel 009 cul the Syn stej Ste. We. act. Pap. sym App‘ mRNl Stir — ,N .“ _ __........' WWW-n . INTRODUCTION The S phase dependency of the cytosolic form of the thymidine kinase (TK) enzyme has been known for many years. Only recently, with the use of the cloned TK gene as a probe, have we been able to begin to elucidate how this enzyme is regulated at the molecular level. At the time the studies described here were begun, very little was known about the synthesis or metabolism of TR mRNA in the cell, or how these processes affected the regulation of the cognate enzyme. Previous studies using synchronized cell cultures had shown that increases in TK enzyme activity in the cell cycle were dependent upon both protein and RNA synthesis, but not DNA synthesis. Therefore, our first step in the approach to this problem was to investigate the steady state levels of TR mRNA produced throughout the cell cycle, and to attempt to correlate this to the enzyme activity. Since it was also well known that infection with papovaviruses such as SV40 produced an induction of DNA synthesis enzymes similar to serum growth factor addition, Appendix 1 describes the investigation of the nature of TK mRNA induction using both these mitogenic factors to stimulate growth of quiescent cells. tha man tha tra ana ind the lev to the hel mor. and incl SUb< Pro: regj meek IEgt Pre$ resu Coul The results of our preliminary experiments showed that both TK mRNA and enzyme were regulated in a similar manner under both of these conditions, and this suggested that TK regulation might be occurring at the level of transcription. Chapter 2 and appendix 2 describe an analysis of the rate of transcription in both serum induced and SV40 infected cells. Since the results from these experiments indicated that there might be multiple levels of regulation, chapter three describes an attempt to determine the half-life of the messenger RNA throughout the various phases of the cell cycle. In an attempt to help further understand TK regulation in the cell cycle, a more detailed characterization of the fate of the message and enzyme at late times in the cell cycle is also included in this chapter. The last chapter includes a description of the subcloning and DNA sequence analysis of the genomic promoter region of the human TK gene. An analysis of this region was necessary to begin to understand the molecular mechanisms that might mediate transcriptional cell cycle regulation of this gene. Finally, a summary and conclusions section has been presented at the end. This section is a summation of the results, and attempts to describe areas of research that could be undertaken in the future. tics phos pyri the but phos pyri deox iodc nucl Subs meth (72) inhi dU a reac and' COns kina: Chapter 1. LITERATURE REVIEW Biochemical Aspects of Thymidine Kinase Deoxythymidine kinase (ATP:deoxythymidine 5’ 'phosphotransferase, EC 2.7.1.75) is an enzyme in the pyrimidine salvage pathway for DNA synthesis which catalyzes the reaction: Deoxythymidine + NTP -f> dTMP + NDP (24). Typically, the nucleoside phosphorylated is deoxythymidine, but thymidine kinase (TK) will also catalyze the phosphorylation of deoxyuridine and of several halogenated pyrimidine deoxyribonucleoside analogs of dT: flouro- deoxyuridine (dFu), triflourothymidine (F3dT), and iododeoxyuridine (dIU). Therefore, the specificity for the nucleoside acceptor is that it be a uracil in which the substituent on carbon 5 of the pyrimidine is hydrogen, a methyl group, i.e. deoxythymidine or any of several halogens (72). All of these analogs are known to be competitive inhibitors of the phosphorylation reaction when either dT or dU are employed as nucleoside receptors (24,68). The main reactions in the formation of deoxythymidine triphosphate, and ultimately DNA synthesis, are shown in figure 1 (79). Considering enzymes 1 and 2, thymidine kinase and TMP kinase, respectively, evidence shows that TMP kinase is the Figure 1 Some of the reactions involved in the synthesis of precursors of deoxythymidine triphosphate required for DNA synthesis. The boxed area indicates the reaction catalyzed by thymidine kinase. Numbers in circles: 1 is thymidine kinase, 2 is TMP kinase. dC 4;; dCMP Jl‘ . ii DINA dU Edwin /dUDP \, ® dUTP/ TMP :TDP :TTP T 1? c reggae UTP CONTROL ‘\\UDP : UMP FIGURE 1 cri is of fun tha wid and for bac and of . enz: sou: pur; inw enz} 0f 1 Sum bec< The 100‘ Sep) plac mole critical rate controlling enzyme for TTP synthesis, while TK is more of a "salvage" enzyme (24,50,68,79). The presence of the TR gene is actually of benefit to the cell, which functions by using less energy (ATP) for synthesis of dTMP than biosynthesis of TTP by the de novo pathway. Thymidine kinase has been reported to be present in a wide range of animal, plant and microbiological forms (yeast ,and other fungi being notable exceptions) and is also coded for by many viral genomes (64,68,79,101). Even the bacteriophage T4 carries a functional thymidine kinase gene, and has been shown to induce activity of TX upon infection of a mutant of E; 92;; lacking this activity (15,30). The enzyme has been partially purified from a variety of sources (24,27,68,72,79). The enzyme in E; 99;; was purified approximately 1200 fold and its properties investigated by Okazaki and Kornberg in 1964 (72). The TK enzyme from g; 92;; is an allosteric protein which consists of two monomeric subunits of 42,000 daltons each. These subunits are extremely temperature sensitive, but become quite insensitive in the dimeric form (68,72). The molecular weights of animal cell enzymes are about 80- 100,000 daltons as estimated by gel filtration with Sepharose 6100 or G200. The cytosolic TK from human placenta has been purified to homogeneity. The molecular weight obtained by gel filtration and sucrose density ultracentrifugation is 92,000. The Sprl is '. stu< shm thin the act: pan none inhi mm the rep: but IS“ [<2 7 subunit molecular weight is 44,000, suggesting that the enzyme is a homodimer in its native state (27). The enzyme has also been purified and studied in detail in a variety of other microorganisms such as Tetrahymena pyriformis and Chlorella renoidosa, as well as in a host of eukaryotic viruses: notably the herpesviruses such as HSV types I and II, EBV and CMV (38). In plants, it has been foundin gee gays, in roots of sprouting seedlings and in Trillium (68,79). However, it is TK associated with mammalian tissue that has been studied the most extensively. TK from both mammalian tumors and E; coli has been shown to be quite stable, particularly in the presence of thymidine (79). There is also fairly general agreement that the metal ion Mg++ is absolutely necessary for TK activity (24,33,79). The requirement for Mg++ can be partially fulfilled by Mn++, Co++, Ca++ and Fe++, but none are as effective as Mg++, and higher concentrations are inhibitory in the presence of Mg++ (24). With the highly purified E; Coli enzyme, maximum activity was obtained in the presence of bothoMg++ and Mn++(72). The pH optimum is reported to range from 7.5 to 8.5 in a variety of sources but as pH drops to 7.0, approx. 25% of enzyme activity is lost (24,33). One important mechanism of control of the TR enzyme in vivo, as well as in vitro, involves a regulatory feedback mec its dem act tri abs a v emb hum. inf= sho‘ liv. of to Strt nucj dAT] kin: are Proc can cell aeic‘ cell mechanism in which TK activity drops upon accumulation of its end product, TTP. Okazaki and Kornberg (72) demonstrated this in E; coli, and also discovered an activation of the enzyme by a number of nucleoside di- and triphosphates, such as dCDP, that would accumulate in the absence of TK activity. This same effect has been noted in a variety of mammalian systems as well, including rat embryo, normal adult and regenerating rat liver, matched human normal and cancer tissues, and uninfected or virus infected LM or CV-1 cells (68,79). TDP has also been shown to be inhibitory to TX activity in regenerating rat liver, although TTP was described as the more "potent" of the two inhibitors (11). The inhibition appears to be competitive with thymidine, and is also strongly modulated by the concentrations of both substrates, ATP and thymidine (11,33). A variety of other nucleoside triphosphates, including UTP, dCTP, dGTP and dATP, were found to have no inhibitory effect on thymidine kinase (11). Other nucleoside kinases, such as dCMP kinase, are also inhibited by high concentrations of their end products (68). Therefore, enzymes such as thymidine kinase can be seen as a component of a larger homeostatic cellular mechanism for growth control if deoxyribonucleic acid synthesis takes place at critical concentrations of cellular deoxyribonucleoside triphosphate levels. euk (ct for in lin sti CYt gro DNA Als ele- pH dCT: exp; cul' 31391 300: Pig] elm: reg1 decl mt'm , Stlk TK Isozymes It is now well known that two distinct forms of eukaryotic TK exist--the cytoplasmic form of the enzyme (ctTK) and the mitochondrial form (mtTK). The mitochondrial form of the enzyme was first discovered by Berk and Clayton in 1973 when they found that a mutant mouse cell line resistant to 5-bromodeoxyuridine, designated LMTK-, still expressed about 2% thymidine kinase activity in the mitochondrial fraction while lacking all activity in the cytosol. Further evidence showed that when TK- cells are grown in media containing [3H]-thymidine, mitochondrial DNA! incorporates tritium, while nuclear DNA does not (8). Also, ctTK and mtTK were shown to differ with respect to electrophoretic mobility, Km, .sedimentation coefficient, pH optimum, phosphate donor specificity and inhibition by dCTP (44,91). The two enzymes have characteristic expression during cell growth. In rapidly growing cultures of established cell lines, ctTK accounts for approximately 96-99% of total activity, with mtTK accounting for the rest. In more slowly growing xeroderma pigmentosa cells, however, one finds the mtTK accounting for almost 45% of activity (44,45). CtTK activity is also regulated with the growth state of the cell--its activity declines rapidly as cells enter stationary phase, while mtTK expression is constituitively expressed (38). Some studies have shown a correlation of expression between the TKS tis adu of use is eit eff Alt com Wil chr hyb aha res is ; Sim undc typ; be 1 mit: eacl 10 TKs in fetal and adult tissue. In fetal and tumor tissue, the ctTK is the major molecular form, yet in some adult human and rat tissues, the mtTK accounts for most of the cellular TK activity (12,22,91). One of the most useful ways to distinguish the two forms in tissue extracts is to calculate the ratio of enzyme activity obtained with either CTP or ATP as phosphate donor, as mtTK can efficiently utilize CTP, while ctTK cannot (22). Although the TR isozymes are found in different subcellular compartments, they are both encoded by nuclear DNA. Willecke et. al. in 1977 were able to assign human mtTK to chromosome 16 by the use of human-mouse somatic cell hybrids. which allowed all human chromosomes to be analyzed. In contrast, the human ctTK, has been shown to reside on chromosome 17 (23,100). WW As mentioned in the last section, the activity of ctTK is a growth stage, or cell cycle, dependent phenomenon. Since TX is a DNA synthesis enzyme, it follows that to understand it’s regulation, one must first understand a typical mammalian cell cycle. By cytological observation of the cell, the cycle can be broken down into two basic parts: interphase and mitosis. Interphase itself can be divided into three parts each defined by the status of the chromosomes. The G1 the chr con gen pro par man beg eve res com is div app cel are Whi abs Spe. trat: r83] mum Che: what 11 (first gap) period is defined as the time during which the daughter cells from mitosis retain a diploid set of chromosomes (52). As chromosomes are released from their condensed mitotic state, specific regions of the genome become accessible to RNA polymerases, and RNA and protein synthesis resume at a rapid rate. In particular, the enzymes for DNA metabolism (such as TR) and many proteins associated with cell differentiation functions begin to be synthesized in G1 (70). The crucial control events for the regulation of growth therefore seem to reside in G1. Evidence suggests a restriction point or commitment point in mid to late G1 phase, where a decision is made whether to initiate DNA synthesis and undergo cell division or to cease proliferation (75,102). This process appears to be serum growth factor dependent (51). In 3T3 cells, processes immediately responsive to . serum are completed at approximately 3.2 hours into G1, after which progression to mitosis can be completed in the absence of serum growth factors (101,102). Growth factors present in serum interact through specific surface receptors. The biochemical steps that transfer the mitogenic signal from the receptor to the responding genes have not been resolved, although numerous reactions which are possibly involved have been characterized (76). Therefore, it remains to be seen what exact effect serum induction provides for the im‘ Ce] Ch] ple re; pos ren At COI‘ for and (94 bei and whe Che tim Cha obt The Pep 12 initiation of DNA synthesis in the mammalian cell cycle. Cells leave G1 when they begin to replicate their chromosomes, and the period in which DNA synthesis takes place is termed S phase. At the end of S, a cell has replicated its entire genetic complement and therefore possesses two diploid sets of chromosomes (52). Cells remain in this state for the G2 period until mitosis begins. At the end of G2' when the cells enter mitosis, chromosome condensation occurs (52,70). Somatic cells which have ceased to divide are usually found in G1. This is true of most differentiated cells and of cells in culture that display contact inhibition (94). Cells that have stopped dividing are described as being in the Go state. It is not yet understood if G1 and G0 represent genuinely different states of the cell or whether G is a special and reversible example of the 0 G state (52). l The time spent in each phase of the cell cycle is characteristic of the particular cell type. A diagram of a typical mammalian cell cycle with a 24 hour doubling time is shown in figure 2 (88). In order to study the cell cycle and characterize its biochemical events, one has to be able to obtain a synchronous population of cells in culture. There are several ways to achieve synchrony in a Population : blocking by drugs, serum or amino acid 13 Figgre 2 Cell cycle of a mammalian tissue culture cell with a 24 hour doubling time. .om . .253 \I w 02.70 C . C O O O O O 02:10 «.0... . o o .. . . _ «a: v: o o o o . o o . v: on «a «F o m .8 3.... m— o < FIGURE 3 71 that the chicken and simian genes are not very homologous. The pattern of transcription from the c-myc gene is similar to that seen for both TR and H3.2, with a 3-fold increase at 12 and 24 hours. No hybridization to the Sp6 control was seen in this or any other experiment. The patterns of transcription in the SV40 infected cells was similar to that seen in serum stimulated cells, with small increases at 12 and 24 hours (Figure 3A). The obvious exception is the PJYl control, which shows that SV40 transcription begins by 12 hours, and reaches very high levels at 24 and 36 hours after infection. Itlappears from these results that changes in the rate of TX gene transcription are not responsible for the large increase in TK mRNA seen during 8 phase in serum stimulated and SV40 infected CV-l cells. In fact, the rate of transcription appears relatively constant throughout the cell cycle in these experiments. Transcription at the Gle phase interface. Although the experiments reported in the previous section did not detect any large changes in TX gene transcription during the cell cycle, it seemed possible that the time points chosen were not those where TK transcription was at a maximum. It was recently shown that transcription of the mouse DHFR gene peaks at the Gl/S phase interface, where a 72 7-fold increase over the level found in resting cells was seen. This increase occurs just prior to or concomitantly with the initiation of DNA synthesis in serum stimulated mouse 3T6 cells containing multiple amplified copies of the DHFR gene (2). We therefore decided to measure TK transcription rates at and around the Gl/S interface in serum stimulated and SV40 infected CV-l cells. We have previously shown that DNA synthesis is induced by 12 hours following serum stimulation and 24 hours following SV40 infection of CV-l cells (24). In order to determine the exact onset of DNA synthesis, cells were pulse labeled for one hour at 1 hour intervals surrounding the approximate initiation of S phase, and incorporation was monitored as described in Materials and Methods. The results of these experiments, shown in figure 4, indicated that DNA synthesis begins between 8-9 hours after serum stimulation and 17-18 hours after SV40 infection. Nuclear run-on transcription assays were performed at 0,6,7,8,9,10,11 and 12 hours post serum stimulation in order to span the Gl/S phase boundary. The assays were performed as described above, except that the c-myc cRNA was omitted and cRNA to a human fiZ-microglobulin cDNA was included on the filters as a control for a gene that is expressed at relatively constant levels throughout the 73 Epgppg_4. Measurements of DNA synthesis in serum stimulated and SV40 infected cells. CV-l cells were arrested and then serum stimulated or infected with SV40 (MOI=15) as described in Materials and Methods. At the indicated times after treatment, they were labeled with 3H-thymidine'for 1 hour, and the cpm incorporated/cell was determined by TCA precipitation of a given number of cells. 74 2:0: 9:0: www.mpmom 8 m. 2 mm 3 32829 0 m. a. r. o.. m m m WinIIao ......”w»....rrs.T +v.... ....p w -.o. it” .... -.om rm in” Iron .r [In lrov OV>W Ftp—mm m < 8-01 x HBO/was FIGURE 4 75 cell cycle. Figure 5 shows that there is a sharp increase in TK transcription at 9 hours following serum stimulation, and that the level falls by 11 and 12 hours. We have quantitated these results by comparing the levels of TR transcription to the 02 microglobulin levels at each time point and determined that there is a 6-7-fold increase in TK transcription at the 9 hour time point (Figure SB and Table I). The H3.2 histone gene again shows a small increase at the 9-12 hour time ponts, and the 62 microglobulin control is relatively constant throughout the time course. This experiment has been repeated, and each time the TK gene showed approximately a 6-fold increase in transcription at the Gl/S boundary (Table I). These results indicate that the TK gene is transcriptionally regulated in serum stimulated cells, with the rate of transcription showing a sharp but transient increase as the cells enter S phase. We next examined TK transcription at the Gl/S boundary in SV40 infected cells. CV-l cells were infected with SV40 at an MOI of 15, and transcription assays were performed at 0,17,18,19,20,21,22 and 23 hours after infection. The results of this experiment, shown in Figure 6 and Table I, were somewhat surprising. While the transcription of the fi2 microglobulin control was 76 Figpre 5. Nuclear transcription assays performed at one hour intervals spanning the 61/8 boundary in serum stimulated CV-l cells. Nuclei were prepared at the times indicated following serum stimulation, and used for transcription assays. 2.0 x 106 cpm of labeled RNA were added to each hybridization. (A) Autoradiogram showing results of hybridization. (B) Graphical analysis of the results in (A). TR (0), H3.2 (a), and.fi-2 microglobulin (A). m -.._., 15:"; “n ____ .-e a, A 77 ° 6 7 8 9 10 11 12 O O o o TK ‘ ' ' ‘ ' 0 0 113.2 ' 0 1 ' 0 . - 32 3500. 3000. 2500— .. 2000— o o \ 2 O. 0 1500... 1000. . - 500— :v‘ ‘2 ‘ ‘ ‘ 1 I l I l r l .I O 6 7 8 9 1O 11 12 HOURS POST SERUM STIMULATION FIGURE 5 ‘ 78 Figpp§_§. Nuclear transcription assays performed at one hour intervals spanning the 61/8 boundary in SV40 infected CV-l cells. Nuclei were prepared at the times indicated following viral infection, and used for transcription assays. 2.5 x 106 cpm of labeled RNA were added to each hybridization. (A) Autoradiograms showing the results of hybridization. (B) Graphical analysis of the results in (A). TK (O), H3.2 (a), fi-2 microglobulin (A) and SV40 (A---A) . 79 A 0 17 18 19 20 1':- t O O .. O 1. . O 0' (I 1. C’ 1. 0 o O O B p—-—-¢—--¢——a—-—-¢ i 7‘ 2000-'1- / I I / f. I I 1500:» I '.. O D \ E 1000-- C) 500‘- 4 I 17 18 19 2O 21 22 23 .. .L d:- - .— .1. 01)- HOURS POST SV40 INFECTION FIGURE 6 80 relatively constant throughout the time course, transcription of both TK and histone H3.2 showed a small increase at 20 hours. The magnitude of the increase in TK transcription was significantly less than in serum stimulated cells, however, peaking at less than 2-fold the resting cell level. This experiment was repeated, and no significant increase in TX transcription was seen (Table I). One explanation for our inability to detect transcriptional induction of TX following SV40 infection is that we missed the correct time point, and that induction was occurring prior to the 17 hour time point. In order to examine this possibility, we measured the levels of TR mRNA and transcription at one hour intervals from 12-18 hours post-infection. The results of these experiments, shown in Figure 7 and Table I, indicate that although TK mRNA starts to accumulate by 14 hours, no transcriptional induction is detected at any of the time points tested. These results are in contrast to those measuring mRNA levels (24 and Figure 2), where TK mRNA accumulates to a higher level in virally infected than in serum stimulated cells, and suggest that SV40 is utilizing an alternate mechanism for inducing high levels of TX mRNA. ...—.13.“- -..—...! 81 Figpre 7. Measurements of TR polyA+ RNA levels and transcription rates in SV40 infected cells. Resting CV-1 cells were infected with SV40 (MOI=15) at t = 0, and at the times indicated plates were harvested for either the preparation of polyA+ RNA or nuclear transcription assays. (a) Northern gel analysis of polyA+ RNA hybridized to TR and fi-2 microglobulin probes. (b) Hybridization of 32P-RNA synthesized in isolated nuclei to dot blots containing 5 ug of TK, 02 and Sp6 cRNA and 5 ug of RJYl (cloned SV40) DNA. 82 A 025mmmnm fi'ggflfl , TK gab... £2 FIGURE 7 TK SV40 Sp6 83 TABLE 1 aTranscription assays were performed as described in Materials Methods. In order to quantitate the level of TX transcription, dots taining 32P-1abeled RNA hybridized to either 1'1: or 3-2 microglobulin were cut out and counted in a scintillation counter. The ratio of hybridized to TK)/(cpm hybridized to 8-2) was then calculated For and con- cRNA (cpm each time point, and the ratio at time 0 was defined as 1.0. Levels of TK transcripticn at the other time points within one experiment expressed relative to this ratio. are 84 TABLE 1. Levels of TK gene transcription in serum stimulated and SVHO infected CV1 cellsa Serum §159 Experiment Experiment 1 2 3 1 2 3 Hours Hours post-serum post-SV40 stimulation infection 0 1.0 1.0 1.0 0 1.0 1.0 1.0 1.0 6 0.7 1 2 2.7 12 0.9 0.8 7 0.5 3.9 3.7 13 0.8 0.” 8 0.9 6.7 5.2 1” 0.3 0.8 9 6.7 “.6 2 7 15 0.5 0.6 10 5.9 3.9 4.7 16 0.5 0.7 11 3.” “.3 U 2 .17 1 O 2.0 0 5 0 9 12 3.1 1.4 2 7 18 1 O 1.4 0 7 0 8 19 1 1 0.3 20 1.4 1 6 21 1.1 1 u 22 1.0 1.5 23 1.3 u." 85 TX induction in cell lipes transfected with human TK cDNA contructs. In order to identify the DNA sequences required in cis for the regulation of TK gene expression, we have isolated several different TK cDNA contructs and studied their expression after transfection into Rat—3 TK- cells. These plasmids, which are diagrammed in Figure 1, are as follows: pHuTKcDNA7 is the original TK cDNA clone that we isolated from the Okayama and Berg library (18), and contains the SV40 early promoter, an SV40 splice donor and acceptor, the SV40 late polyadenylation signal and virtually the entire HuTK cDNA sequence; p5’TKcDNA was derived from pHuTKcDNA7 by replacing the SV40 promoter and splice signals with the 5’ region from the genomic human TK locus. The 5' genomic region was linked to the cDNA at an SmaI site within the first exon of the gene. These two plasmids were transfected into Rat-3 TK- cells, and TK+ transformants were selected and propagated both as pools of approximately 50 colonies each and as clonal cell lines derived from single colonies. The resulting cell lines were tested for their ability to regulate TK expression in an S phase specific manner by measuring TK mRNA and enzyme levels in resting and serum stimulated cells. The regulation of TR mRNA levels was investigated by isolating poly A+ RNA at 0, 12 86 and 24 hours after serum stimulation and quantitating this RNA by Northern blot analysis as described previously (24). The results of these experiments are shown in Figure 8. In Figure 8A three cell lines containing pHuTKcDNA7 were examined, with cell lines 1C and 2A being derived from single colonies and cell line m being derived from a pool of colonies. In Figure 83 the results are shown from cell lines containing pS'TKcDNA, where DB and A1 are derived from single colonies and m from pooled colonies. All of these cell lines show increased levels of TR mRNA by 12 hours after serum addition, and similar results were obtained with several other transfected cell lines tested (data not shown). Rat-1 (TK+) and Rat-3 TK- cell lines were also included in this experiment as positive and negative controls. The results of TX enzyme assays on the cell lines described above are given in Table II. These results are in agreement with the mRNA data, with peak levels of TX enzyme activity being seen at either 12 or 24 hours following serum stimulation. Thus, both pHuTKcDNA and p5’TKcDNA are cell cycle regulated after serum treatment of transfected cell lines, indicating that sequences within the body of the cDNA are sufficient to confer S phase specific expression upon these hybrid genes. An alternative explanation would be 87 F;gppg_§. Northern blot analysis of TR polyA+ RNA in cell lines transfected with TK cDNA constructs. Rat-3 TK- cells were transfected with either pHuTKcDNA7 or p5'TKcDNA, and HATr colonies were selected. Transfected cell lines were growth arrested and serum stimulated as described in Materials and Methods, and poly A+ RNA was prepared at 0, 12 and 24 hours following serum addition. RNA was analyzed by Northern blot analysis, and probed with a HuTK cDNA probe. (A) Rat-3 TK- negative control and cell lines containing pHuTKcDNA7. Cell lines 1C and 2A were derived from isolated colonies, and m from a pool of approximately 50 colonies. (B) Rat-1 (TK+) control and cell lines containing p5'TKcDNA. Cell lines D3 and A1 were derived from single colonies and m from a pool of approximately 50 colonies. 88 A pHuTKcDNA7 Rats m 10 2A 01224 01224 01224 01224 B p5'TKcDNA Rat1 m 03 A1 0122401224 01224 01224 FIGURE 8 89 TABLE II: TK enzyme assays in cell lines transfected with HuTK cDNA constructsa Hours post serum - pHuTKcDNA7 pS‘TKcDNA stimulation Rat-3 Rat-1 Mass 1C 2A Mass D3 A1 0 - 3.4 3.7 1.2 0.9 1.1 2.0 0.3 12 - 8.5 7.4 13.3 6.4 7.1 11.1 1.6 211 .003 9.2 3.9 7.2 ‘ 4.6 12.6 11.1 3.7 36 - 11.8 2.7 1.2 0.9 0.7 0.8 0.9 aTable II. TK enzyme assays from cell lines transfected with TK cDNA constructs. The cell lines described in the legend to figure 6 were growth arrested and serum stimulated as described in Materials and Methods. At 0,12 and 24 hours after serum addition, TK extracts were prepared and enzyme assays performed as described previously (24). Data is presented as TK units x 10'", where one unit is defined as the amount of enzyme required to convert one nmole of dT t0 dTMP per microgram of protein per minute of reaction at 37°C. 90 that the SV40 early promoter is itself regulated in an S- phase specific manner. Although we have not tested this hypothesis directly, experiments in the laboratory of Dr. N. Heintz have indicated that this promoter is not activated during the transition of cells from G1 into S phase (N. Heintz, personal communication). DISCUSSION The results presented here suggest that the thymidine kinase gene is regulated at both transcriptional and post- transcriptional levels in serum stimulated cells. Transcription of the gene is low in resting cells, and remains low during G1 after serum stimulation. As the cells enter S phase, which is at approximately 8-9 hours following serum stimulation, there is a 6-7-fold increase in the rate of TK gene transcription. Transcription drops to 3-fold the level in resting cells by 12 hours following serum addition, and continues to decrease at 24 hours. These results are similar to those reported for the DHFR gene in mouse cells, where there is'a 7 fold increase in transcription at the Gl/S boundary (2). In contrast, no sharp increase in TK transcription was detected in SV40 infected CV-l cells in these experiments, although the TR mRNA accumulates to higher levels in virally infected than 91 in serum stimulated cells. Thus the mechanisms of induction by these two mitogenic agents appears to differ in some or all aspects. This was also suggested by earlier experiments, where it was shown that induction of DHFR by polyoma virus was not sensitive to the same inhibitors as was induction by serum (10). The molecular basis of these differences remain to be determined. There-are several reasons why we believe that the increased transcription seen in serum stimulated CV-l cells cannot completely account for the increase in TX . mRNA levels. First, the transcriptional induction is both shorter and of lesser magnitude than the change in mRNA levels. The transcriptional induction peaks at approximately Gefold the level found in resting cells, and persists for only about 3 hours, while the mRNA levels increase 10-206fold, and remain high for at least 12 hours. Second, the relative levels of mRNA induction in serum stimulated and SV40 infected cells do not coincide with the levels of transcriptional induction. Finally, we have studied TK regulation in cell lines transfected with hybrid cDNA clones that utilize either the human TK promoter or the SV40 early promoter, and have shown that sequences within the body of the cDNA are sufficient to confer cell cycle regulation upon these constructs. These 92 results are in agreement with those obtained in-other labs with both the chicken (16) and hamster (13) chromosomal TK genes, where it was shown that hybrid genes expressed from the Herpes TK promoter showed cell cycle or growth dependent expression. We have considered two possible explanations for the ability of the various cDNA clones to be regulated in an S phase specific manner. The first, and in our opinion most likely, is that TK RNA metabolism differs somehow during S phase so that it accumulates to high levels. Since p5’TKcDNA contains no introns, it seems unlikely that the difference is in nuclear RNA processing, but may be in transport from the nucleus, or in nuclear or cytoplasmic stability. A second possible explanation is that sequences within the TR cDNA confer cell cycle regulation upon an adjacent promoter, whether it be the genomic TK promoter, the herpes TK promoter or the SV40 early promoter. This seems unlikely, however, since even in the case of CV-l cells with the homologous promoter the level of increased transcription cannot account for the increase in the level of mRNA in all cases. Given these three lines of evidence, it seems likely that both transcriptional and post-transcriptional regulatory mechanisms are operating to increase expression of the TK gene during S phase. 93 We cannot unambiguously determine from our results whether or not the cDNA construct containing the genomic HuTK promoter is more highly induced during 8 phase than the one containing the SV40 promoter, which we might expect if transcriptional control plays an important role in the regulation of this gene. Although the pooled colonies containing pS’TKcDNA always show higher levels of induction than the pools containing pHuTKcDNA7, this is not the case for individual cloned cell lines. Since we have not characterized the integrated plasmids in these cells lines, the variations we see may be due to both the number and location of integration sites. In order to quantitatively compare the levels of induction in transfected cell lines, we will either have to use transient expression assays or carefully compare the patterns of integration in different cell lines. Our results, taken together with the results from other labs, suggest that in general 8 phase specific genes may exhibit multiple levels of regulation. As mentioned previously, both transcriptional and post transcriptional regulation of histone gene expression has been demonstrated for some time (7,21,22,23). In the case of. other 8 phase regulated genes such as DHFR, TS and TK the existence of transcriptional regulation has been in —7———7 V ‘ ""' “—r' 94 question, although post transcriptional control has been demonstrated. In part the difficulty in demonstrating transcriptional control has been that the genes are transcribed at low levels, so the levels of transcription are difficult to measure. Because of this, several investigators used cell lines containing amplified DHFR or T8 genes to measure transcription rates. In this report we have been able to measure TK transcription rates in cell lines containing single copies of the gene by binding single stranded cRNA to filters to hybridize to RNA transcribed in.isolated nuclei. A similar approach was used by Groudine and Casimir (5) to study regulation of the chicken TK_gene. Even with sufficiently sensitive assays, several investigators have failed to detect transcriptional regulation of TR and DHFR gene expression. Our results, and those of Farnham and Schimke (2), suggest that the precise timing of the assay is critical, since the burst of increased transcription is quite transient. The situation has been further complicated by the fact that hybrid TK genes utilizing heterologous promoters are cell cycle regulated in several systems (13,16), and this has led investigators to conclude that transcriptional control is not occurring. In fact, the presence of dual levels of control is now established for 95 many of the S phase regulated genes that have been examined. A complete understanding of the complex pattern of regulation of these genes, and of the relative importance of the different regulatory mechanisms, awaits a more detailed analysis. By utilizing the approaches outlined here, that is the construction of hybrid or mutant genes and the study of their transcription and/or RNA metabolism in cells, we should be able to progress quickly towards that goal. 96 ACKNOWLEDGEMENTS The authors would like to acknowledge the work of Moriko Ito for the TKcDNA transfection experiments. We would also like to thank Ron Patterson and Karen Friderici for helpful comments on this manuscript. This work was supported by Public Health Service Grant CA37144. SEC is a recipient of a Junior Faculty Research Award from the American Cancer Society. 97 LITERATURE CITED 1) Alterman, R.-B., S. Ganguly, D.H. Schulze, W.F. Marzluff, C.L. Schildkraut, and A.I. Skoultchi. (1984). Cell cycle regulation of mouse H3 histone mRNA metabolism. Mol. Cell. Biol. 4 : 123-132. Farnham, P.J., and R. Schimke. (1985). Transcriptional regulation of mouse dihydrofolate reductase in the cell cycle. J. Biol. Chem. ggg : 7675-7680. 2 V 3) Green, M.R., T. Maniatis, and D.A. Melton. (1983). Human fi-globin prc-mRNA synthesized 1p vitro is accurately spliced in Xenopus oocytes nuclei. Cell Q2 : 681-694. Greenberg, M.E., and E.B. Ziff. (1984). Stimulation of 3T3 cells induces transcription of the c-fos protooncogene. Nature 311 : 433-438. 4 V Groudine, M., and C. Casimir. (1984). Post- transcriptional regulation of the chicken thymidine kinase gene. Nucl. Acids Res. 1; 1427-1446. 5 V Groudine, M., M. Peretz, and H. Weintraub. (1981). Transcriptional regulation of hemoglobin switching in chicken embryos. Mol. Cell. Biol. 1 : 281-288. 6 v 7 Heintz, N., H.L. Sive, and R.G. Roeder. (1983). Regulation of human histone gene expression: Kinetics of accumulation and changes in the rate of synthesis and in the half-lives of individual histone mRNAs during the HeLa cell cycle. Mol. Cell. Biol. 3 : 539-550. V 8) Jenh, C.-H., P.K. Geyer, and L.F. Johnson. (1985). Control of thymidylate synthase mRNA content and gene transcription in an over-producing mouse cell line. Mol. Cell. Biol. 5 : 2527-2532. 9 Johnson, L., L.G. Rao, and A. Muench. (1982). Regulation of thymidine kinase enzyme levels in serum stimulated mouse 3T6 fibroblasts. Exp. Cell. Res. 13g : 79-85. V 10) 11) 12) 13) 14) 15) 16) 17) 18) 98 Kellems, R.B., U.B. Morhenn, E.A. Pfendt, F.W. Alt, and R.T. Schimke. (1979). Polyoma virus and cyclic AMP-mediated control of dihydrofolate reductase mRNA abundance in methotrexate-resistant mouse fibroblasts. J. Biol. Chem. ggg : 309-318. Kelly, K., E.H. Cochran, C.D. Stiles, and P. Leder. (1983). Cell specific regulation of the c-myc gene by lymphocyte mitogens and platelet derived growth factor. Cell 35 : 603-610. Kit, S., D.R. Dubbs, P.M. Frearson, and J. Melnick. (1966). Enzyme induction in SV40 infected green monkey kidney cultures. Virology 22 : 69-83. Lewis, J.A., and D.A. Matkovich. (1986). Genetic determinants of growth phase-dependent and adenovirus 5-responsive expression of the Chinese hamster thymidine kinase gene are contained within thymidine kinase mRNA sequences. Mol. Cell. Biol. 6 : 2262-2266. Leys, E.J., G.P. Crouse, and R.B. Kellams. (1984). Dihydrofolate reductase gene expression in cultured mouse cells is regulated by transcript stabilization in the nucleus. J. Cell. Biol. 29 : 180-187. Leys, E.J., and R.F. Kellams. (1981). Control of dihydrofolate reductase messenger ribonucleic acid production. Mol. Cell. Biol. 1 : 961-971. Merrill, G.F., S.D. Hauschka, and S.L. McKnight. (1984). TK enzyme expression in differentiating muscle cells is regulated through an internal segment of the cellular TK gene. Mol. Cell. Biol. 4 : 1777-1784. Morello, D., P. Duprey, A. Israel, and D. Babinet. (1985). Asynchronous regulation of mouse H2-D and beta-2 microglobulin RNA transcripts. Immunogenetics 22 : 441-452. Okayama, H., and P. Berg. (1982). High efficiency cloning of full-length cDNA. Mol. Cell. Biol. 2 : 161-169. 19 v 20 v 21) 22 V 23) 24 V 25) 26) 99 Postal, E.H., and A.J. Levine. (1976). The requirement of simian virus 40 gene A product for the stimulation of cellular thymidine kinase activity after viral infection. Virology 13 : 206-215. Santiago, C., M. Collins, and L.F. Johnson. (1984). 1p vitro and 1p vivo analysis of the control of dihydrofolate reductase gene transcription in serum stimulated mouse fibroblasts. J. Cell. Physiol. 11p : 79-86. Schumperli, D. (1986). Cell cycle regulation of histone gene expression. Cell gfi : 471-472. Sittman, D.B., R.A. Graves, and W.F. Marzluff. (1983). Histone mRNA concentrations are regulated at the level of transcription and mRNA degradation. Proc. Natl. Acad. SCi. USA 89 : 1849-1853. Sive, H.L., N. Heintz, and R.G. Roeder. (1984). Regulation of human histone gene expression during the HeLa cell cycle requires protein synthesis. Mol. Cell. Biol. 5 : 2723-2734. Stuart, P., M. Ito, C.J. Stewart, and S.E. Conrad. (1985). Induction of cellular thymidine kinase occurs at the mRNA level. Mol. Cell. Biol. 5 : 1490-1497. Topp, W.C. (1981). Normal rat cell lines deficient in nuclear thymidine kinase. Virology 113 : 408-411. Wu, J.-S., and L.F. Johnson. (1982). Regulation of dihydrofolate reductase gene transcription in methotrexate—resistant mouse fibroblasts. J. Cell. Physiol. 119 : 183-189. 100 CHAPTER III (ARTICLE) Rhythmic Expression of the Cytoplasmic Thymidine Kinase Gene is Regulated by Multiple Levels of Cell Cycle Dependent Controls. by Christine J. Stewart and Susan E. Conrad To be submitted to Molecular and Cellular Biology 101 ABSTRACT The regulation of the cytoplasmic thymidine kinase (TK) gene occurs in a cell cycle-dependent manner. TK has been shown to be mitogenically induced by serum growth factors and by infection with papovaviruses such as SV40. We have studied the stability of TX messenger RNA by the addition of a transcriptional inhibitor in the GO/Gl' S, SV40-S, and G2 phases of the cell cycle. By addition of Actinomycin D to synchronously growing cell populations, we were able to follow the decay of TX mRNA in these various stages. TK mRNA was relatively stable in all phases: 8-9 hours in Go/Gl' 20 hours in S, 32 hours in SV40-S and 9-10 hours in G2. Removal of poly-adenylation recognition sequences destabilized the message (t < 1 hour). However, removal of other 3’ untranslated sequences had no effect on the rate of decay. TK mRNA and enzyme levels were studied into and beyond the G2 phase of the first cell cycle following serum stimulation. Withdrawal of serum in G decreased TK enzyme activity, 2 but TK mRNA remained relatively unaffected. Addition of 102 hydroxyurea in early S phase did not affect the initial induction of TR enzyme or mRNA. However, termination of both TK mRNA steady state levels and enzyme activity kept rising until extremely high levels were seen at late times post serum stimulation, as compared to controls. These results suggest multiple levels of regulation for the TK gene in various phases of the cell cycle. 103 INTRODUCTION The activity of the cytoplasmic thymidine kinase enzyme (EC 2.7.1.75) has been shown in a variety of systems to be dependent on the specific growth state of the eukaryotic cell. In rapidly growing, continuously cycling cells, thymidine kinase (TK) activity is high compared to the levels found in stationary phase or quiescent cells (9). Characteristically, fetal and neoplastic tissues show much higher TK activities than the cognate adult tissue, and adult tissues which proliferate have the highest activities of adult tissue (18, 29). In tissue culture cells synchronized via serum deprivation or drug blocking with butyrate, colcemid, amethopterin or aphidicolin, TR activity is low in Go/G1 phase cells. Following serum refreshment or drug removal, TK activity increases dramatically from levels seen in G1 phase by mid-to late-S phase (1, 6, 13, 44, 45, 46, 49), and this induction commences with the onset of DNA synthesis (16, 44). In contrast, the genetically distinct mitochondrial form of the enzyme shows no variation in activity throughout the cell cycle (1, 23). By treating tissue culture cells with RNA or protein synthesis inhibitors, the increases in cytoplasmic TK enzyme activities were shown to be dependent upon both RNA and protein synthesis 104 (16, 34). Addition of DNA synthesis inhibitors does not affect the induction of TR enzyme activity, but a certain proportion of the genome must be replicated for efficient termination of TK activity (3, 16, 19). Another method of induction of TX activity is by infection with papovaviruses such as simian virus 40 (SV40) (20, 21, 22). As in normal cell cycle progression, TK induction in infected cells coincides with the onset of DNA synthesis. Use of various temperature sensitive mutants of SV40 has shown that the A gene product, or large T-antigen, is responsible for the induction of TX (38). Large T-antigen is also necessary for efficient lytic viral replication and is required for the initiation and maintenance of transformation (31, 37, 47). The induction of TX by serum growth factors and by SV40 infection appears to occur at multiple levels. We have previously demonstrated (44) that increases in TK enzyme activity appear to be paralleled, at least temporally, by an increase in steady state levels of TK mRNA in both systems. By using nuclear run-on transcription assays, we have shown that there is a 6- to 7—fold increase in TK transcription in serum stimulated simian CV-l cells at the 61/5 phase border (43). A smaller 3-3.5-fold increase in transcription has also been found at the beginning of S phase in SV40 infected CV-l 105 cells (42). Other studies have shown that the chicken, hamster and mouse TK genes are regulated at a post- transcriptional level (12,13,26,27). TK regulation was found to be independent of the promoter used to transcribe the gene. Finally, induction of TR enzyme activity depends upon protein synthesis and may also be dependent upon the constant presence of serum growth factors (16, 34). Multiple levels of regulation also have been found for other S phase specific genes. The replication- dependent histone genes are regulated by both transcriptional and post-transcriptional mechanisms (2, 5, 7, 8, 29, 40). The dihydrofolate reductase (DHFR) gene, another DNA synthesis enzyme, shows a pattern of I transcriptional induction similar to the TK gene (10. 17). Post-transcriptional control of DHFR has also been reported (28). Thymidylate synthetase is another DNA synthesis enzyme where both transcriptional and post- transcriptional control has been reported (15, 35). The purpose of our present study is to more thoroughly investigate the nature of post-transcriptional regulation of TK expression. We report a very stable TK message with approximate half-lives of 8—9 hours in the GO/G1 phase, 20 hours in S phase, and 9-10 hours in the G2 phases of the cell cycle, as well as an extremely long half-life of 32 hours in SV40 infected cells. The 106 sequences responsible for this stability have been localized to the 3’ end of the mRNA, and the regulation of TR termination as cells progress through the cell cycle also has been examined. MATERIALS AND METHODS ' W Simian CV-1 (African Green Monkey Kidney) cells were grown to confluence in 1X Dulbecco’s Modified Eagle’s Medium containing 5% fetal calf serum (FCS) and 5% calf serum (CS). For synchronization, the medium was removed after the cells reached confluence and was replaced by medium containing 0.1% FCS + CS. To obtain synchrony, cells were allowed to incubate for 48 hours to cause a uniform arrest in 60/61“ For viral infections, SV40 was added at this time at a multiplicity of infection (MOI) of 15. Infections were done by removing medium, infecting for 1 hour with a concentrated viral stock, and then replacing the original 0.1% serum-containing medium. For serum stimulations, new medium containing 5% FCS + 5% CS was added. At various times after induction with either SV40 or serum, plates were harvested for RNA, protein or DNA synthesis analysis. Rat3 TK- cells (43) were grown in 1X DME with 10% CS. Serum stimulations were performed as described for CV-l cells. —_—'—— ' "”’ 107 DNA pulse labelling To obtain a profile of the cell cycle, the specific activity of DNA was determined in a separate experiment by pulse labelling One 100 mm plate of cells per time point with 1 uci of [3H] deoxythymidine (61 c1/ mole) and 4 x 10"7 M uridine for 1 hour. TCA precipitates were prepared as described (44) and the number of incorporated counts was determined by counting for 3[H] in a scintillation counter. One 100 ul aliquot of cells in 1x PBS was counted on a hemacytometer. Specific activities were calculated as counts incorporated/number of cells. WM Total RNA was prepared from tissue culture cells as follows. Cells were washed twice with phosphate buffered saline (PBS) without calcium and magnesium. Lysis buffer (100 mM Tris 7.5, 12 mM EDTA {Ethylenedinitrilo}- Tetraacetic Acid, 150 mM NaCl, 1% Sodium Dodecyl Sulfate) containing 200 ug/ml of proteinase K was added to each sample in a volume of 2 mls/100mm plate. DNA in the cell lysate was sheared by passage through a 22-gauge needle, transferred to a tube, and fresh proteinase K was added to 100 ug/ml. This solution was incubated at 37°C for 45 minutes, and then extracted twice with 50:50 v/v Phenol:Chloroform. Sodium Acetate was then added to 0.3M and the solution was ethanol precipitated. Samples were spun in a Sorvall RC-2 centrifuge at 10,000 RPM for 20 *n '-—-. 108 minutes, the ethanol poured off and pellets allowed to air dry. Pellets were resuspended in 400 ul of RNase-free 10 mM TRIS (7.5), 1 mM EDTA ((Ethylenedinitrilo}-Tetraacetic Acid) (1X TE) and transferred to eppendorf tubes. A solution containing 4 ul of 1M MgClZ, 5 ul of vanadyl ribonucleoside complex (VRC) and 1 ul of a 1 mg/ml solution of Pancreatic DNase was added, and tubes were incubated at 37° c for 30 minutes. Next, 16 ul of 0.5 M EDTA and 20 ul of 20% Sodium Acetate was added, this mixture was extracted twice with 50:50 v/v Phenol:Chloroform and then ethanol precipitated at -70°C. RNA was next pelleted in a microcentrifuge at 4°C for 25 minutes and pellets were dried in a vacuum pump dessicator. Pellets were resuspended in 150 ul of 20% Sodium Acetate and spun for 10 minutes in a microcentrifuge at 4°C. Supernatants were discarded and the remaining pellets were resuspended in 100 ul of 1X TE and then ethanol precipitated after addition of 10 ul 20% Sodium Acetate. For determination of optical density, samples were spun down at 4°C in a microcentrifuge for 25 minutes, drained, dryed and resuspended in 200 ul of 1X TE. One-fifth of each sample was diluted into 1 ml of 1X TE and the optical density was monitored at 260 nm. WW Northern blotting and TK enzyme extraction and TK assays were performed as described previously (44). All 109 32P-labelled internal SmaI-Bam HI blots were probed with a TK fragment from p5’TKcDNA (Fig. 1). Construction of Tk 3’ end deletion mutants Plasmid pS’chDNA is a human cDNA clone capable of expressing TK after transfection into eukaryotic cells and was constructed as described previously (43). The plasmid contains the 5' genomic promoter region attached to the TR cDNA and 3’ untranslated region in a vector containing the SV40 polyadenylation signal. Figure 1 shows this plasmid, as well as two gross deletions extending from the SV40 polyAdenylation recognition site (leaving this site intact) to a Hind III site (p5’TKcDNAAH) and to a Bam H1 site (p5’TKcDNAAB), both of which are contained in the rather long 662bp 3' untranslated region of the TK mRNA. Also shown is a construction which deletes the SV40 polyAdenylation recognition signal and 60 base pairs of the 3’ untranslated region up to the Bam H1 site (p5’TKcDNAAA+). Transfections of these constructs into Rat3 cells were performed as described (43). NuQleiQ_AQiQ.flxhridiaetigu§ Hybridizations to RNA filters and post-hybridization washes were performed as previously described (44). The final wash step with 0.1XSSC-0.1%SDS at 68°C was deleted. 110 E1gpp§_1 Functional TK cDNAs. Each TK cDNA construct contains the genomic TK promoter region cloned into an EcoRl (E) to Sma1 (8) site. The body of our TK cDNA is represented by the large open box. The 662 base pair 3’ untranslated region is represented by a line and contains two restriction sites used for a gross mutational analysis, Hind III (H) and Bam H1 (B). The smaller open box contains the SV40 polyadenylation recognition sequence. pS’TKcDNA is the full length mini-gene. pS'TKcDNAAB deletes from the poly-A site up to the Bam H1 site, leaving the poly-A site intact. PS’TKcDNAAH deletes up to the Hind III site and also leaves an intact poly-A sequence. p5'TKcDNAcA+ deletes both poly-A and 60 base pairs up to the Bam H1 site. 111 a .1 33208.50 m :04. 208:. be m 3.280.er ”ml .208.er rIL ... F Uh f I U)- In q U)— -r “.1 l E R U G I F 112 RESULTS Determination of TR mRNA half-life in CV-1 cells. Previously, we have reported on the induction of TX mRNA and enzyme activities in CV-1 cells following either serum-induction or SV40-infection. A 10-20 fold induction was seen in both systems (for both RNA steady state levels and enzyme activity) which peaked by 12-24 hours in serum induced cells and by 24-36 hours in SV40 infected cells. I have also characterized the mechanism of this induction as being partially at the transcriptional level in both systems. At the Gl/S phase border a 6—7-fold transcriptional induction occurs at 8-9 hours post-serum stimulation, and a 3-3.5-fold induction occurs at 14-16 hours post SV40 infection (42). Since the levels of transcriptional induction did not seem to account for the high level of message produced, I now have characterized the extent of the stability of the messenger RNA during various phases of the cell cycle as this was a likely candidate for post transcriptional control. The results of this experiment are shown in figure 2. Stability of the TK message was determined by the addition of 5 ug/ml of Actinomycin D to CV-1 cells in either the arrested (Go/G1) state, at 10-12 hours post-serum- induction or at 15-22 hours post-SV40-infection when cells are in S phase. The decay of the TR message was then followed by preparing total RNA at various time points 113 following addition of drug and analyzing constant amounts of total RNA from each time point on denaturing agarose- formaldehyde gels. Each time course was run twice, and the relative values of all these experiments can be seen in table 1. Samples of our autoradiograms are shown in figure 2. All of our autoradiograms were traced by densitometer and approximate half-lives determined by linear regression analysis. Plots of the decay curves are shown in figure 3. For all of these phases, the half- lives are rather long. In arrested GO/G1 phase cells (fig. 2a), TK message half-life is 8-9 hours, while in serum induced S phase cells (fig. 2b), the half-life extends to 20 hours. TK in the SV40-infected cells (fig. 2c) shows an even longer half life of 32 hours. A human c-myc probe was used as a positive control for the inhibition of transcription, since it has a half-life of 30 minutes. c-myc was not detected by 1 hour following addition of drug in any of these studies (data not shown). Charactepization of Gg Phase in Serum Induced CV-1 sells In order to characterize the stability of TK mRNA during the G2 phase of our CV-l cells, the timing of the S and G2 phases of the cell cycle in our CV-l cells needed to be more finely characterized. Figure 4 contains the results of pulse-labelling experiments using [3H]thymidine uptake as an indication of the timing of DNA synthesis. 114 E1gppe_1 TK mRNA Stability in CV-1 Cells. The rate of decay of TR message was determined by the addition of 5 ug/ml Actinomycin D to CV-l cells which were either: (a) non-induced or in G1 phase, (b) Serum-stimulated and treated with the drug at 12 hours or S phase, (c) infected with SV40 (MOI = 15) with drug added at 15 hours, and (d) serum-stimulated with drug added at 20 hours or G2 phase. Total RNA was prepared from cells harvested after the number of hours indicated on the abcissa, and constant amounts (30 ug for part a, 10 ug for parts b, c and d) were run in each lane on denaturing agarose-formaldehyde gels. The figure shows autoradiograms of gels blotted to nitrocellulose and probed with an internal SmaI - Bam HI TK fragment. FIGURE 2 115 ‘ ' .‘ - ' J . 0 . L ‘1..- 1.1.1 '. ' 3.9“... _'.“.M w 1.- 0 12 14 1618 20 24 36 48 C. O . ' \ . ‘ iii 31: . :1 . ‘1 _ 1 b” ' '.' '1 C. n “U!“ . 015 25 35 45 60 ...-1w 0 15 20 25 30 40 116 TABLE 1 GO/Gl S-Phase SV40-S 62 HRS Exp 1 Exp2 HRS Exp 1 Exp2 HRS Expl Epo HRS Expl Epo 1.0* 1.0 o 0 1.0* 1.0* 0 1.0* 1.0 0 1.0 1.0 2 0.74 12 4.8 6.3 15 7.9 * 15 * 14.8 4 0 77 14 4.5 22 25.8 20 8.8 28.0 6 0 62 16 3.8 25 6.5 22 6.4 8 0.71 18 3.9 3 9 28 17.2 24 6.2 9 0.57 20 4.1 35 6.3 13 3 25 14 0 12 0.51 0.64 24 3.6 2.1 45 3.3 11 4 26 5.3 15 0.34 35 1.6 60 2.3 28 4.9 16 0.49 36 2.1 30 4.7 10 0 20 0.31 48 2.0 32 1.7 34 1.9 40 l 6 44 1.2 50 0.9 Table 1. Relative values representing the amount of TK RNA seen at various times post addition of Actinomycin D. Asterisks indicate the time point in each experiment where the drug was added. 117 F1gppg_1 Quantitation of the results of the densitometer scan of the autoradiograms in figure 2. The area under the curves was calculated, and plotted with respect to time. The line of best fit and the half—lives were calculated by linear regression analysis (IBM SigmaPlot). RELATIVE AR E A 118 GO/Gl Phase 30 3 Phase 20" 30 ° 62 Phase 20' 0 1 1 l ? L 1 1 1 IO 20 30 40 so 60 70 60 HOURS Figure 3 119 F1gppg_1 Measurement of DNA synthesis in serum- stimulated CV-1 cells. CV-l cells were arrested and then serum-stimulated by addition of fresh 1X DME + 5% calf serum and 5% fetal calf serum. At the indicated times (hours indicated on the abcissa), cells were labelled with [3H] Thymidine for one hour, and the counts per minute incorporated per cell were determined by trichloroacetic acid precipitation of a given number of cells. 120 50 -(3 ‘- _), and Hydroxyurea treated at t = 12 hours (Ar—"¢l ). TK activiy was determined as described previously (44). r ~w-w 8.69:. 851986»... 6.16 Umztamo Eamm 01.0 .8250 0.10 i we «6 mm 6... on em mu m. ... o. m N _ a _ _ _ - a _ a _ _ .1 — — _ fi _. _ _ _ fl — _ fl. q 01 l / o \o .. . .. I I . / \7. 1.. \V\\ 04 lo, \ / QI-lu0\ I, D l // 0 .\ O: \\ ./ \nfl“ . L . \ l 2 W J l \\\\q 1 Ill». . .\hr/Q\ I o i / \u N. m_ ON ¢N (S.Lan) 3wsz3 >11 BALLV'IBH FIGURE 6 128 Figpre 7 Late S/G2 Phase TK mRNA Levels. Total RNA was made as described in Materials and Methods with the same cell populations from figure 5. An equal amount (15 ug) of RNA from each time point was run on denaturing agarose-formaldehyde gels and blotted to nitrocellulose as described (44). All blots were probed with a 32P-labelled SmaI-8am HI internal TK fragment. (A) Serum stimulated control, (B) Serum depleted at t = 20 hours, and (C) Hydroxyurea treated (0.1 ug/ml) at t = 12 hours. 129 A. 0 12 1618 20 22 24 26 28 32 36 4044 50 B. o 12 20 22 24 26 28 3032 34 36 44 50 C. -.-.... 0 12 20 25 30 35 45 55 FIGURE 7 - enema-r . 130 Figpre 8 Densitometer Analysis of Late S/G2 Phase TK mRNA Levels. The autoradiograms in figure 6 were traced with a densitometer, the area under each curve was calculated, and the value at t = 0 was set to 1.0 and plotted vs. time (hours). Control (D-n-dfl ), Serum Depleted at t = 20 (CD-*0) , and Hydroxyurea treated (Ar-no.) values are shown. mmDOI DI \ \ /./fi—./ //0 \ .\ \ l . IIOI/ 0 AN\ \\ 0111 D/ ’0 \z \ m\ l 131 . . '2’ .\’ \ z \ \ \ J 4\ .382... 85.086»... 4.16 i uozamo Eatmm 010 .9250 03.0 N. 0.. ON ¢N mu mm on (VBHV) \7’Ntlu‘l >11 BAIIV'IEH FIGURE 8 132 densitometer scanning. The control TK enzyme activity and RNA levels appear to be somewhat similar. By 24 hours the enzyme activity has increased 12 fold and the mRNA level 20 fold over the level seen at t=0. These levels then fall, and a second peak occurs at 28 hours (for mRNA) and at 32 hours (for enzyme) post-serum-stimulation. Each induction begins during the highest measurable rate of DNA synthesis, and each peak occurs just after. In CV- - . 1 cells which were transferred to serum depleted media at t=20 hours post stimulation, TK enzyme levels never attain the high levels seen in our controls. A 7-7.5 fold induction of activity occurs by 20 hours, and this level falls to 5.5-fold by t=50 hours. In contrast, TK mRNA induction is similar to the control TK mRNA levels. Again, two peaks of induction are seen at approximately the same times, and attain only slightly lower levels of mRNA with a very similar kinetic pattern. The addition of hydroxyurea (0.1 ug/ml) to these cells inhibits DNA synthesis by 97% (data not shown) and has a rather extreme effect on both TK enzyme activity and mRNA levels. Figure 6 shows that TK enzyme levels are slightly reduced at 24 hours from control levels, however the activity level keeps rising even up to 50 hours post-serum-induction, where it (fig. 7 and 8) attains extremely high levels. The initial induction is similar to the control, 133 however no decrease is seen. TK steady state mRNA levels also keep rising until a 30-fold level is seen by 50 hours. DISCUSSION Our present study demonstrates that the mechanisms involved in the regulation of the steady state levels of TR mRNA are dependent upon each phase of the cell cycle in serum-induced CV-l cells. Throughout the entire cell cycle, TK mRNA appears to have an extended stability. No previous determination of G1 phase TK message half-life exists. However, the S phase levels that we see are similar to those seen with the human TK gene (48). TR mRNA is less stable in our simian cells when Actinomycin D is added at the start of G2 phase, while the human TK levels remain high in aphidicolin- treated HeLa cells (48). In contrast, confluent growth- arrested cells which are mitogenically stimulated to enter S phase by SV40-infection show an even greater stability. Our previous results show a temporal dependence on the Gl/S phase interface for transcriptional activation of TX in serum-induced CV-1 cells and later results suggest a smaller activation at this same phase in SV40 infected cells. The molecular basis for these cell cycle dependent regulatory controls has yet to be determined. One piece of evidence now exists for the mediation of the extended 134 stability of TK mRNA. My deletion analysis at the 3' end of the gene shows that the only absolutely necessary sequence in this area is a functional polyadenylation recognition site. Removal of a large portion of the 3' untranslated region had no effect on stability of the TK message. However, when the polyadenylation site was removed, the TK message was destabilized and had a half- 1ife of less than one hour during S phase in actinomycin D treated Rat3 cells transfected and stably selected for this construct. This is in general agreement with Greenberg (11), who found that poly-adenylated mRNAs turn over approximately once per cell generation time. Also histone genes, which are one of the only sets of mammalian mRNAs which lack polyA sequences, turn over much more rapidly than other types of RNA. The replicative histone variants all have half-lives of only 10-30 minutes (11, 40). The poly-A site has also been shown to be important for the mediation of mRNA stability of the globin gene of Xenopus Oocytes, and many viral genes from Adenovirus (14, 25). Since the poly-A site used in our constructs was the SV40 polyadenylation site, it remains to be seen whether the same situation is true 1p y1yg with the endogenous TK poly-A site intact. Whether the poly- A site in our constructs requires the presence of other TK sequences for the mediation of its stability has yet to be determined, however, it is entirely possible that this 135 site is not the only site necessary for a stable message. Further characterization of the cell cycle in CV-l cells has allowed us to more finely characterize TK mRNA and enzyme levels. By taking time points at two to four hour intervals at late times after serum induction, we were able to observe two peaks in TX mRNA levels, both of which occur just after DNA synthesis levels peak in the first and possibly second cell cycles following serum induction. The second peak in TK mRNA induction may be due to a second transcriptional induction of TX as the cells traverse G1 into S phase for a second round of DNA replication. Since there is a relatively short time span between the peaks of DNA synthesis, not enough time is allotted for much measurable degradation of TR mRNA. This fact probably accounts for the high levels of TX steady state mRNA levels previously observed at late times post serum stimulation (44), and allows us to now state that the regulation of the thymidine kinase gene does appear to be dependent upon multiple cellular controls. The two peaks of TK mRNA levels are also observed with cells that are placed in serum depleted medium during the G phase of the first cell cycle. Since the 2 level of induction is only slightly decreased, it is questionable as to whether serum removal has any real effect on TK mRNA levels or its regulation in cells that have been mitogenically stimulated for long periods of 136 time by the addition of serum growth factors. In contrast, TK enzyme levels appear to be greatly affected by removal of serum, as control levels (Fig 6) are never reached. However, a steady low level of activity is maintained in our cells, in contrast to Johnson (16) who found a rapid decrease in activity by six hours following serum removal from 3T6 cells. It does appear, however, that a constant presence of serum growth factors is required for cells to attain high levels of TR activity, and this effect appears to be a post- transcriptional one. What significance this result has for the 1p vivo regulation of TX remains to be seen. The most striking result presented here involves the use of the DNA synthesis inhibitor, hydroxyurea. E Inhibition of DNA synthesis by 97% does not prevent induction of either TK enzyme or TK mRNA levels in serum stimulated CV-l cells, but the termination of this induction is greatly affected. The initial induction ) itself appears not to be directly influenced by DNA X synthesis. Rather, it is probably transcriptionally J induced in late Gl by some Gl-specific mechanism which is somehow inducible by serum growth factors. The decrease of TX mRNA and enzyme activity appears to be mediated by a different mechanism. It is known that hydroxyurea directly inhibits Ribonucleoside Diphosphate Reductase (RDP Reductase) activity in the cell, and that 137 cellular dNTP pools are then rapidly depleted. It is also known that dTTP is specifically required in the cell as an allosteric effector for the reduction of GDP to dGDP, and the dGTP generated from dGDP is an allosteric effector for the reduction of ADP to dADP (18). It is therefore possible that TK activity is induced to a high level in hydroxyurea treated cells to produce more dTTP in an attempt to compensate for the loss of RDP reductase function. This cellular induction of TX activity could possibly be used to maintain cellular dNTP pools at a specific level. It is already well known that in a variety of systems a regulatory feedback mechanism exists in which TK activity drops upon the accumulation of its end product, TTP (33, 36, 39). It would therefore seem that since intracellular pools are depleted in hydroxyurea treated cells, the normal regulatory "feedback" mechanism may never become operational. This theory also has significance for our SV40-infected cells. It is known that levels of TR mRNA and enzyme activity are typically much higher in SV40-infected cells than in serum- stimulated cells. SV40-infected cells present a condition which demands rapid DNA replication. If intracellular dNTP pools become extremely depleted during this process, TK activity may be induced to an extremely high extent to help compensate for this situation. The induction may involve an even longer increase in the half-life of the new 138 message, and this may account for the greater half-life seen in our SV40-infected cells. It is also possible that if replication of SV40 DNA continues until cellular lysis occurs, a prolonged half-life of messenger RNAs involved in DNA synthesis may also be a necessary result. Obviously, these ideas require a more in depth analysis. However, we can state that the thymidine kinase gene is regulated at multiple levels throughout the cell cycle. Both a transcriptional control is operational in Gl/S, while a lengthy half life of the message is seen in all phases. However, the exact mechanism regulating serum dependence of the TR enzyme is not known and must be more fully investigated. The dependence of TX messenger RNA and enzyme activity shut-off on cellular DNA synthesis should also prove to be an interesting area of study. ACKNOWLEDGMENTS I would like to thank Richard Mehigh for help in the preparation of figures and Moriko Ito for helpful discussions. 140 REFERENCES (1) (2) (3) (4) (5) (6) (7 V (8 v (9) (10) (11) Adler, R. and B.R. McAuslan. (1974). Expression of Thymidine Kinase Variants is a Function of the Replicative State of Cells. Cell 1 : 113—117. Alterman, R.B., S. Ganguly, D.H. Schulze, W.F. Marzluff, C.L. Schildkraut and A.I. Skoultchi. (1984). Cell Cycle Regulation of Mouse H3 Histone mRNA Metabolism. Mol. Cell. Biol. 1 : 123-132. Bello, L.J. (1974). Regulation of Thymidine Kinase Synthesis in Human Cells. Exp. Cell Res. 82 : 263-274. Bradshaw, H.D. and P.L. Deininger. (1984). Human Thymidine Kinase Gene: Molecular Cloning and Nucleotide Sequence of a cDNA Expressible in Mammalian Cells. Mol. Cell. Biol. 5 : 2316- 2320. Capasso, 0., G.C. Bleecker and N. Heintz. (1987). Sequences Controlling Histone H4 mRNA Abundance. EMBO J. 6 : 1825-1831. 'Coppock, D.L. and A.B. Pardee. (1985). Regulation of Thymidine Kinase Activity in the Cell Cycle by a Labile Protein. J. Cell. Physiol. 111 : 269-274. 1 Dailey, L., S.M. Hanly, R.G. Roeder and N. Heintz. (1986). Distinct Transcription Factors Bind Specifically to Two Regions of the Human Histone H4 Promoter. Proc. Natl. Acad. Sci. USA 81 : 7241-7245. DeLisle, A.J., R.A. Graves, W.F. Marzluff and L.F. Johnson. (1983). Regulation of Histone mRNA Production and Stability in Serum-Stimulated Mouse 3T6 Fibroblasts. Mol. Cell. Biol. 1 : 1920-1929. Eker, P. (1965). Activities of Thymidine Kinase and Thymidine Deoxyribonucleotide Phosphatase During Growth of Cells in Tissue Culture. J. Biol. Chem. 119 : 2607-2611. Farnham, P.J. and R.T. Schimke. (1985). Transcriptional Regulation of Mouse Dihydrofolate Reductase in the Cell Cycle. J. Biol. Chem. ggg : 7675-7680. Greenberg, J.R. (1972). High Stability of Messenger RNA in Growing Cultured Cells. Nature 119 : 102-104. (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) 141 Groudine, M. and C. Casimir. (1984). Post- Transcriptional Regulation of the Chicken Thymidine Kinase Gene. Nuc. Acids Res. 12 : 1427-1446. , Hofbauer, R., E. Mullner, C. Seiser and E. Wintersberger. (1987). Cell Cycle Regulated Synthesis of Stable Mouse Thymidine Kinase mRNA is Mediated by a Sequence Within the cDNA. Nuc. Acids Res. 12 : 741-753. Huez, G., G. Marbaix, E. Hubert, M. Leclercq, U. Nudel, H. Soreq, R. Salomon, B. Lebleu, M. Revel and U.Z. Littauer. (1974). Role of the Polyadenylate Segment in the Translation of Globin Messenger RNA in Xenopus Oocytes. Proc. Natl. Acad. SCi. USA 11 : 3143-3146. Jenh, C.-H., P.K. Geyer and L.F. Johnson. (1985). Control of Thymidylate Synthase mRNA Content and Gene Transcription in an Overproducing Mouse Cell Line. Mol. Cell. Biol. 2 : 2527- 2532. Johnson, L.F., L.G. Rao and A.J. Muench. (1982). Regulation of Thymidine Kinase Enzyme Level in Serum Stimulated Mouse 3T6 Fibroblasts. Exp. Cell Res. 122 : 79-85. Kaufman, R.J. and P.A. Sharp. (1983). Growth Dependent Expression of Dihydrofolate Reductase mRNA from Modular cDNA Genes. Mol. Cell. Biol. 2 : 1598- 1608. Kit, S. (1976). Thymidine Kinase, DNA Synthesis and Cancer. Mol. Cell. Biochem. 11 : 161-182. Kit, S., D.R. Dubbs and J.L. Melnick. (1966). Effects of Inhibitors of Protein and Nucleic Acid Synthesis on Thymidine Kinase (deT Kinase). Induction by Papovavirus SV40. Fed. Proceed. 22 : 777. Kit, 8., D.R. Dubbs, L.J. Piekarski, R.A. deTorres and J.L Melnick. (1966). Acquisition of Enzyme Function By Mouse Kidney Cells Abortively Infected with Papovavirus SV40. Proc. Natl. Acad. Sci. USA 22 : 463-470. Kit, S., D.R. Dubbs, R.A. deTorres and J.L. Melnick. (1965). Enhanced Thymidine Kinase Activity Following Infection of Green Monkey Kidney Cells by Simian Adenoviruses, Simian Papovavirus SV40 and an Adenovirus-SV40 "Hybrid". Virology 21 : 453-457. Kit, S., D.R. Dubbs, P.M. Frearson and J.L. Melnick. (1964). Enzyme Induction in SV40- Infected Green Monkey Kidney Cultures. Virology 22 : 69-83. (23) (24) (25) (26) (27) (28) (29) (30) (31) (32) (33) (34) 142 Kit, S. and W.C. Leung. (1974). Submitochondrial Localization and Characterization of TK Molecular Forms in Parental and Kinase Deficient HeLa Cells. Biochem. Genet. 11 : 231-247. Lewin, B. Gene Expression 2: Eucapyotic Chromosomes (1980). John Wiley & Sons, Inc. New York, NCYO Lewin, B. Genes (1983). John Wiley & Sons, Inc. New York, N.Y. Lewis, J.A. (1986). Structure and Expression of the Chinese Hamster Thymidine Kinase Gene. Mol. Cell. Biol. g : 1998-2010. Lewis, J.A. and D.A. Matkovich. (1986). Genetic Determinants of Growth Phase-Dependent and Adenovirus 5-Responsive Expression of the Chinese Hamster Thymidine Kinase Gene are Contained Within Thymidine Kinase mRNA Sequences. Mol. Cell. Biol. 2 : 2262-2266. Leys, E.J., G.L. Grouse and R.E. Kellems. (1984). Dihydrofolate Reductase Gene Expression in Cultured Mouse Cells Is Regulated by Transcript Stabilization in the Nucleus. J. Cell Biol. 22 : 180-187. Littlefield, J.W. (1964). Studies on Thymidine Kinase in Cultured Mouse Fibroblasts. Biochim. Biophys. Acta 22 : 14—22. Luscher, B. and D. Schumperli. (1987). RNA 3’ Processing Regulates Histone mRNA Levels in a Mammalian Cell Cycle Mutant. A Processing Factor Becomes Limiting in Gl-Arrested Cells. EMBO J. g : 1721-1726. Martin, R.G. and J.Y. Chou. (1975). Simian Virus 40 Functions Required for the Establishment and Maintenance of the Transformed State. J. Virol. 12 : 599. Merrill, G.F., R.M. Harland, M. Groudine and S.L. McKnight. (1984). Genetic and Physical Analysis of the Chicken TK Gene. Mol. Cell. Biol. 2 : 1769-1776. Metabolic Pathways "Nucleic Acids, Protein Synthesis and CoEnzymes". D.M. Greenberg, ed. pp.103-108. Mittermayer, C., R. Busselmann and V. Bremerskov. (1968). Initiation of DNA Synthesis in a System of Synchronized L-Cells: Rhythmicity of Thymidine Kinase Activity. Eur. J. Biochem. 2 : 487-489. (35) (36) (37) (38) (39) (40) (41) N vv (43 (44) (45) (46) (47) 143 Navalgund, L.G., C. Rossana, A.J. Muench and L.F. Johnson. (1980). Cell Cycle Regulation of Thymidylate Synthetase Gene Expression in Cultured Mouse Fibroblasts. J. Biol. Chem. 255 : 7386-7390. Okazaki, R. and A. Kornberg. (1964). Deoxythymidine Kinase of Escherichia coli. J. Biol. Chem. 112 : 275-284. Osborn, M. and K. Weber. (1975). Simian Virus 40 Gene A Function and Maintainence of Transformation. J. Virol. 1; : 636. Postel, E.H. and A.J. Levine. (1976). The Requirement of Simian Virus 40 Gene A Product for the Stimulation of Cellular Thymidine Kinase Activity After Viral Infection. Virology 11 : 206—215. Protein Metabolism and Biological Function (1970). Rutgers University Press. New Brunswick, N.J. P.C. Bianchi and R. Hilf, eds. pp. 141-156. Sittman, D.B., R.A. Graves and W.F. Marzluff. (1983). Histone mRNA Concentrations are Regulated at the Level of Transcription and mRNA Degradation. Proc. Natl. Acad. Sci. USA gg : 1849-1853. Sive, H.L., N. Heintz and R.G. Roeder. (1984). Regulation of Human Histone Gene Expression During the HeLa Cell Cycle Requires Protein Synthesis. Mol. Cell. Biol. 1 : 2723-2734. Stewart, C.J.--Unpublished Data. Stewart, C.J., M. Ito and S.E. Conrad. (1987). Evidence for Transcriptional and Post- transcriptional Control of the Cellular Thymidine Kinase Gene. Mol. Cell. Biol. 1 : 1156-1163. Stuart, P. M. Ito, C.J. Stewart and S.E. Conrad. (1985). Induction of Cellular Thymidine Kinase Occurs at the mRNA Level. Mol. Cell. Biol. § : 1490-1497. Stubblefield, E. and G.C. Mueller. (1965). Thymidine Kinase Activity in Synchronized HeLa Cell Cultures. Biochem. Biophys. Res. Comm. 29 : 535-538. Stubblefield, E. and S. Murphree. (1967). Synchronized Mammalian Cell Cultures. II. Thymidine Kinase Activity in Colcemid Synchronized Fibroblasts. Exp. Cell Res. 1g : 652-656. Tooze, J- (1981). 2NA.Tum2r_yiru§e§;_The_Mglsgular Biologx_2£_Tumor_yiruse§- Second Edition. Cold Spring Harbor Laboratory. 144 (48) Traina-Dorge, V.L., S. Caradonna and P.L. Deininger. (1987). Cell Cycle Regulation of the Human Thymidine Kinase Gene. Nuc. Acids Res. --in press. (49) Yang, H.C. and A.B. Pardee. (1986). Insulin-like Growth Factor I Regulation of Transcription and Replicating Enzyme Induction Necessary for DNA Synthesis. J. Cell. Physiol. 111 : 410- 416. ' 145 CHAPTER IV (ARTICLE) Sequence Analysis of the Human Thymidine Kinase Gene Promoter Region bY Christine J. Stewart and Susan E. Conrad 146 ABSTRACT The gene for the cytoplasmic thymidine kinase (TK) enzyme has been shown to be regulated in a growth phase dependent manner. Since part of this regulation is known to be at the level of transcription, the promoter of the human TK gene has been analyzed. We have located the human TK promoter in a 900 bp EcoRI fragment of human DNA. The 900 base pair fragment was sequenced and the promoter sequence localized to a 550 bp EcoRl-Smal fragment which was subcloned onto the body of a TKcDNA for further 1n vitro and in vivo analysis of its regulation. Nuclease Sl mapping experiments were performed and two transcription start sites were mapped in human 293 cells. Computer sequence analysis has localized a 40 bp homology in the hamster and human TK promoter regions. It is possible that this area may be a thymidine kinase gene specific regulatory element. 147 INTRODUCTION Knowledge of how a gene such as TX is regulated during the cell cycle is essential for understanding the process of cell growth on a molecular level. Localizing and sequencing the promoter region of the thymidine kinase gene became important in order to obtain a better understanding of transcriptional regulation of this gene. By comparing the thymidine kinase gene promoter sequence to various other cell cycle regulated genes we hoped to find novel gene specific sequences which may be important for directing the timing and extent of transcription in the cell cycle. Many consensus sequences have been identified as important for transcription of RNA polymerase II transcribed genes. One of the most well known elements is an A-T rich region, or "TATA" box, typically found at -25 to -30 nucleotides from the transcriptional start site. Mutation analysis of this sequence in many genes has shown that it is important for the correct initiation of transcription (16, 33). An example of some of the Pol II- transcribed genes which do not have a TATA homology are the polyoma virus late genes (31). These late genes 148 exhibit multiple start sites for transcription (31). Another DNA sequence that is necessary for maintaining a correct or efficient rate of transcription is the CCAAT box. This element typically resides at -70 to -90 nucleotides from the transcription start site, and has the consensus sequence 5'-C5GPyCAA:C'I‘--3’ (2). Mutations in this element in a variety of genes, such as Rabbit B-globin (9) and Xenopus hsp70 gene (3) are known to reduce transcription of these genes. CAAT boxes function in either orientation, as evidenced by studies on the thymidine kinase gene of herpes simplex virus (15, 27). Inverted and/or tandem CAAT boxes have been found in a variety of cell cycle regulated histone genes (5, 18, 29), and more recently in the human thymidine kinase gene promoter (23). The CAAT box may be functioning either independently or in conjunction with other elements (3, 17). ' Another important promoter element is the G/C rich sequence. The consensus sequence for this element is T GGG -GGGGCGGAAT- function in either orientation (10). This element is 5’ 3, (22) and like the CCAAT box, it can also common to a host of viral and cellular genes and is typically found in multiple copies in the promoter region. 149 This element has also been shown to be necessary for an efficient rate of transcription, and is known to bind the transcription factor Spl (10, 12). One of the most potentially interesting elements to be found in the promoter region of any gene, and specifically of the TR gene, would be a gene—specific or gene family-specific sequence such as an enhancer. Finding a sequence with these properties would help us to identify specific transcriptional controls within a particular gene or gene family and may explain how genes such as TK are regulated in the cell cycle. An enhancer can be distinguished from other promoter elements by its ability to function upstream, within or downstream of a gene, as well as to function in an orientation independent manner. Enhancers are gig-acting and usually function to stimulate transcription from either heterologous or homologous promoters (1, 13). Gene—specific regulatory elements which exhibit some or all of the properties of enhancers are often found in groups of genes which are regulated in a similar fashion. For example, cAMP regulated genes contain a well conserved sequence (7, 8, 14, 34) and this element is quite similar to a consensus element of the cAMP responsive genes in procaryotes (11). In order to more fully explore these possibilities 150 with the human TK gene, the promoter region of the human TK gene was isolated from a lambda library, characterized by restriction mapping and sequenced. Sl nuclease mapping experiments were performed and the transcriptional start site was mapped. A computer dot-matrix analysis was performed in order to compare the human TK promoter sequence with a variety of other sequences. A 5' non coding element was identified which is highly conserved between the hamster and human thymidine kinase genes. MATERIALS AND METHODS 21a§m1g§: Figure 1 from Appendix 1 shows a restriction map of a variety of lambda clones which produced a TK+ phenotype upon transfection of two or more clones into Rat3 TK- cells. Part D of this figure shows fragments which hybridized to the human TK cDNA clone pHuTK-cDNA7 (30). The 5’ most fragment to which this cDNA hybridized was a 900 base pair EcoRl-EcoRl fragment. This fragment was isolated and subcloned into the EcoRl site of pBR322 to give pBRTKpro. grgparation of DNA Fragments and Seggencing Two fragments were obtained by double digestion of pBRTKpro with EcoRI and Smal (Bethesda Research Laboratories), and 151 the Smal site was mapped to approximately 550 base pairs from the 5’ most (relative to the gene) EcoRl site. For labellings, the DNA was first cut with either EcoRl or Smal, and then labelled at either the 5' end with T4 DNA 32P-ATP or at the 3' end with the Klenow kinase and gamma- fragment of DNA polymerase I and a-BZP-ATP (New England Nuclear). A second cut was then made, and fragments isolated by agarose gel electrophoresis. Sequencing of the 900 base pair fragment was performed by the Maxam- Gilbert procedure (25). Preparation of Tota1 RNA Total RNA was prepared from 293 cells via the procedure described in chapter 3. 81 Nuclease Analysis For 81 analysis a 258 bp Hinfl fragment was isolated from pS'TKcDNA, which is a plasmid containing the 550 base pair EcoRl-Smal fragment (found by sequence analysis to encode the TK genomic promoter region) subcloned onto the body of the TK cDNA. Figure 1 shows the map of this plasmid as well as the relevant restriction sites. The Hinfl fragment was isolated by gel electrophoresis and 5’ end labelled with T4 DNA Kinase and gamma-32P-ATP. 10 ng of labelled DNA was ethanol precipitated with 15 ug of RNA, resuspended in 10 ul of hybridization buffer (80% formamide, 0.4M NaCl, 0.04 M Pipes pH 6.4, and lmM EDTA). This mixture was heated to 152 W Figgre 1 Structure of pS'TKcDNA. pS’TKcDNA expresses the TK mRNA from the human genomic TK promoter, but utilizes an SV40 polyadenylation signal. Symbols: (ID HuTK sequences, a::::) SV40 sequences, (-—-—) pBR322 sequences.r Arrows denote the Hinfl sites used in the $1 mapping analysis. 153 EcoR1 'H Sma1 San w .H 1 l p5'TKcDNA BamH1 PoWA FIGURE 1 154 90°C for 3 minutes and immediately transferred to 60°C and incubated for 12 hours. The sample was then diluted to 200 ul with 1x 81 salts (0.28 M NaCl, 0.05 M Sodium Acetate pH 4.6, 4.5mM ZnSo4) and aliquoted in equal volumes into 3 separate tubes. 81 nuclease (Pharmacia) was added at 50, 100, and 175 units in each of the three tubes, and incubated at 37°C for 15 minutes. The samples were then extracted with an equal volume of phenol/chloroform (50:50 v/v) and precipitated with ethanol. Samples were resuspended in loading buffer and run on an 8% denaturing polyacrylamide sequencing gel. RESULTS To study the human thymidine kinase promoter it was first necessary to isolate the 5' end of the gene. This was done by localizing the 5'-most area of lambda clones which were known to hybridize to a nearly full length TK cDNA (see appendix 1). A 900 base pair EcoRl fragment was isolated and subcloned into the EcoRI site of pBR322 and pSp64. By digesting with SmaI, this fragment could be broken down into two fragments of unequal length for isolation and labelling. The nucleotide sequences of two 3’ 5 I . l fragments 5 EcoRl-SmaI (550 base pairs) and SmaI- 155 EcoR13’ (350 base pairs) were determined. Figure 2 shows the sequence of the 550 base pair fragment as well as part of the sequence of the 350 base pair fragment. This 350- bp fragment contains the junction between the first exon and intron as determined by comparison with the published cDNA sequence (4). There are several interesting features of this sequence. There are two 6 base pair A-T rich sequences (TTTAAA), one at -25 and the other at -248. These have been denoted in figure 2 by a box placed around each element. The hexanucleotide GGGCGG is found seven times-- ' I I - I twice in the 5 -GGGCGG-3 position, twice inverted 3 - GGCGGG-S', twice on the opposite strand in the inverted position (CCGCCC), and once in the 5’---CCCGCC-3’ position on the opposite strand (GC rich elements have.been signified in figure 2 by dots placed overhead). Four of these elements reside within a 27 bp inverted repeat element; one copy of each repeat is found on each side of the first A-T rich homology (TTTAAA) (shown in fig. 2 with a line below and arrows above each 27 bp repeat). In studies on the SV40 promoter and the HSV TK promoter only the GGGCGG hexanucleotide sequence (or its inverse) has been shown to bind the Spl transcription factor, however the other GC rich elements seen in this promoter have not been shown in 156 Figgre 2 Sequence of the human TK promoter region. I I A 550 bp 5 EcoRl-Smal3 fragment was sequenced via the Maxam—Gilbert technique using base specific chemical cleavages (25). Symbols: (°) indicates G/C rich hexanucleotides, (*) the 40 bp hamster-human homology, «J A-T rich TATA sequences, (———9 over sequence) the.9 bp inverted CCAAT sequence, and (-—-+ over and under) the 27 bp inverted repeats. Base pair changes or differences noted from other published sequences (4, 24) are indicatai by placing the appropriate nucleotide below the nucleotide found in this analysis. X denotes nucleotides not found in the other two sequences. I 5 C TAAATCTA ATAAATQ QCC CX X CCCCC ********** ********* CTACCTCTGC AGACATCTTC C CCATCATGGC GTCTACAGCC CCTCGCTCCG cccdiijggfl x CCCCGATCAG CCACGTACCA x TCCCGGATTC CTCCCACGAG GGGCCCTGAT TGGCCCCATG 157 TAACTCCGCC CCAGCCCCTT *i-ki-k‘k TTCCAAGGAA CCTTGCTTGG GCATGGGCGT GCGTCCCTCT :rrr-rr-—-—w+. CTTGGTEGE gnaw TCGCCCTGAT TTCCAGGCCC GGGGCGGGCT GCGGCCAAAT X ...... , CCGGCGGGGC CGGCTCGTCA ’IAAAECGGTC GGCCCGG GAA CCAUGGGCTT AITGCGCGAC ' T TTCGTGAACT TCCCGGAGGC * **** **‘k* AGTCCCTCCC TGCAATCCAC GAAACCCACA CCAGACACAT c GTTTATATGG chcAc cccc c QCGGCTCAQA CCAGCCCCCA T TCCCAGTCCC TCGGCGCACG CTCCCCGCCA GTCAGCGGCC TTCCCCACCA cccccchEfi GCCCTTCCAG ACTACTCCCG GCA LTGA—GCLCEAHMMQAQIQMMQI ch cgg gcd ccA Agg ggg ggc 959 11g 919 crcccccccc CACCCCTCCG CGTCGCTCGC GATGACGTGG TCAGTCGT....3 FIGURE 2 158 other systems to bind to Spl (12). . The other arrows in figure 2 have been placed above a 9 base pair inverted repeat found by Kreidberg and Kelley (24). This sequence contains the canonical CCAAT box in an inverted orientation and has been recently shown to be important in the binding of cell cycle regulated transcription factors (23). There are several base pair differences in this sequence compared to the two published sequences. These have been denoted by placing the 'published alternate base pair below the particular nucleotide in question. An "X" has been placed where a nucleotide which I found was not reported in either published sequence, and nucleotides contained in the published sequence(s) but not in this sequence are shown placed below the position in which they were found by the other researcher(s). / total RNA from human 293 cells was isolated and analyzed using an 81 nuclease procedure. The probe was a Hinfl fragment which begins 73 base pairs 3’ of the first ATG and extends to a site approximately 185 bp 5’ (upstream) of the first ATG of the cDNA sequence. Figure 3 shows that the RNA-protected two fragments that differ in size ’I To correctly identify the 5’ end of the TK mRNA, / by 7-8 base pairs with similar intensity: this 159 £1gpr§_1 Autoradiogram of an 81 nuclease experiment run on an 8% polyacrylimide gel. For size comparison the cognate sequenced TK promoter fragment is shown. The sequence position of the first cap site is denoted by an arrow. FIGURE 3 161 indicated two equally efficient start sites for the mRNA. In order to search for sequences that might mediate- cell cycle regulation, the human TK promoter sequence was compared to other known promoter sequences of cell cycle- regulated genes and other thymidine kinase genes using the GENEPRO computer program (IBM). The most significant homology was found using a Dot Matrix program in a comparison of the human TK gene promoter to the hamster TK promoter. Figure 4 shows that an extremely homologous 40 base pair region was found at a location 400 base pairs upstream from the first ATG in the human sequence, and 232 base pairs upstream from the first ATG in the hamster sequence. This sequence is shown in figure 2, with an asterisk placed above each homologous nucleotide. The computer search turned up other close approximations (80% homology) of this sequence in 22 cases, but none were contained within the promoter regions of any of these genes. Most were contained in internal sequences. The only gene that had any type of similar function was the Rat PEP Carboxykinase (GTP) gene. None of the genes that are regulated in a manner similar to TK, such as DHFR or thymidilate synthetase contained this sequence. Nine out of the 22 sequences found came from eucaryotic viruses, but the significance of any of this is unclear. Unfortunately, 162 Figure 4 Dot matrix comparison of the human TK promoter and the hamster TK promoter regions. The 5' most sequences for each gene begin in the lower left corner. This program compares every 10 base pair stretch in one sequence to every 10 base pair stretch in the other and places a dot on the page in an area signifying a match based on 8 out of 10 possible base pairs matching correctly. 163 600 :st . . ' ' ' .r 500 » - ‘ * 400 ’ 300 >- Human 200!- . . . " 100 b ‘ l l A A A L A A 50 100 150 200 250 300 350 400 450 Hamster FIGURE 4 164 comparisons to the mouse and chicken TK gene promoters were not possible as these sequences have not yet been published. DISCUSSION We have performed a DNA sequence analysis of a 550 base pair genomic region found closely associated with the first exon of the human TK gene. By comparison with the published cDNA sequence (4) it was possible to map the 3’ end of this sequence to the first exon of the TR cDNA. This area was subcloned onto the body of our TKcDNA via a common SmaI site, and an internal Hinfl fragment was used to perform an S1 analysis. The start site of transcription was mapped, and two separate sites were identified. The intensity of these bands on the autoradiogram indicated that each site is used to an equal extent in human 293 cells. This same 550bp region has since been defined as containing the functional promoter region by a deletion mutational analysis. Only 83 bp 5’ to the cap site are required for the efficient expression of the gene, as assayed by transfection of a mini-gene 165 construct into Ltk- cells and assaying for growth in selective HAT (hypoxanthine-amethopterin-thymidine) media (24). A published analysis of TK promoter mutations and their transformation efficiencies indicates that the A-T rich region nearest the CAP site is not absolutely required for expression, but deletions which remove it seem to reduce promoter function (24). This region is not capable of directing expression in the abscence of other upstream promoter sequences. Similarly, mutation or removal of one or both 9 bp sequences containing the inverted CAAT sequence dramatically decreases the efficiency of TK+ transformation. This result has also been seen with various globin genes (6, 9) as well as the HSV TK gene (26). Recently, it was shown that both inverted CAAT sites in the human TK promoter form complexes with nuclear DNA-binding proteins (23). The nature of the complexes changes dramatically, as seen by gel mobility shift analysis, as the cells approach 8 phase. Between 9 and 18 hours the mobility of bands shifted to other areas of the gel. This result correlates well with our previous data (see chapter two) on the timing of the transcriptional increase of the TK gene in the cell cycle. However, it is not yet known whether the CAAT box binding observed by Knight et. al. (23) is truly 166 resulting in a transcriptional increase. The factor which binds to the TR CCAAT sequence may be similar to the CAAT binding transcription factor (CTF) which has been purified from HeLa cell nuclear extracts (21). CTF was shown to be indistinguishable from nuclear factor 1 (NF-1), and is known to bind the CAAT sequence of the hsp70 gene (28). The importance of the promoter region in the regulation of cell cycle stage-dependent expression of the thymidine kinase gene has not yet been fully explored. Experiments performed by Travali et. al. (32) clearly show that the promoter region is necessary for cell cycle ,dependent regulation of TR mRNA levels. When the TR promoter was linked to the bacterial CAT gene, it was expressed at the highest levels in S phase. An HSP-70 promoter from Dros0phila linked to the body of the TK cDNA did not show cell cycle dependent expression. Heat shock of cells transfected with this construct showed high levels of TK mRNA, even in G cells. Likewise the 0 calcyclin promoter linked to the TK cDNA showed its highest expression in serum deprived cells. Also, a construct carrying the SV40 promoter linked to the TK cDNA was shown to be cell cycle regulated. Further experiments in our laboratory have shown that there is an induction of TR mRNA levels with our SV40 promoter-TKcDNA construct 167 between 0-2 hours post serum stimulation, but TK enzyme activity remains S phase dependent (20). It remains to be seen if our cells are inducing the SV40 promoter in early G1 via some cellular mechanism or if our TKcDNA plays a role in this induction. The experiments of Travali (32), as well others(19, 20, chapter 3) make it clear that a translational and/or post-translational control is/are present which specifies cell cycle dependent expression of TR enzyme activity. What has also become very clear is that the TK promoter region is definitely involved in the mediation of cell cycle dependent expression of TH mRNA steady state levels. Future work should focus on trying to define area(s) of the promoter, such as the 40 bp hamster-human TK promoter homology, which may mediate cell cycle dependent expression. 168 REFERENCES (l) Banerji, J., L. Olson and W. Schaffner. (1983). A Lymphocyte-Specific Cellular Enhancer is Located Downstream of the Joining Region in Immunoglobulin Heavy Chain Genes. Cell 11 : 729-740. . (2) Benoist, C., K. O’Hare, R. Brethnach, and P. Chambon. (1980). The Ovalbumin Gene - Sequence of Putative Control Regions. Nuc. Acids Res. 8 : 127-142. (3) Bienz, M. and H.R.B. Pelham. (1986). Heat Shock Regulatory Elements Function as an Inducible Enhancer in the Xenopus hsp70 Gene when Linked to a Heterologous Promoter. Cell 15 : 753-760. (4) Bradshaw, H.D. and P.L. Deininger. (1984). Human Thymidine Kinase Gene: Molecular Cloning and Nucleotide Sequence of a cDNA Expressible in Mammalian Cells. Mol. Cell. Biol. 1 : 2316- 2320. (5) Busslinger, M., R. Portmann, J.C. Irminger and M.L. Birnstiel. (1980). Ubiquitous and Gene - Specific Regulatory 5’ Sequences in a Sea Urchin Histone DNA Clone Coding for Histone Protein Variants. Nuc. Acids Res. 8 : 957-977. ' (6) Charnay, P., P. Mellom, and T. Maniatis. (1985). Linker Scanning Mutagenesis of the 5’ Flanking Region of the Mouse fi-Major-Globin Gene: Sequence Requirements for Transcription in Erythroid and Non-Erythroid Cells. Mol. Cell. Biol. 5 : 1498-1511. (7) Comb, M., N.C. Birnberg, A. Seasholtz, E. Herbert (8) (9) (10) (11) (12) (13) (14) (15) 169 and H.M. Goodman. (1986). A Cyclic AMP - and Phorbol Ester - Inducible DNA Element. Nature 323 : 353-356. Cooke, N.E. and J.D. Baxter. (1982). Structural AnalySls of the Prolactin Gene Suggests a Separate Origin for it's 5’ End. Nature 297 : 603-606. Dierks, P., A. vanOoyen, M.D. Cochran, C. Dobkin, J. Reiser, and C. Weissmann. (1983). Three Regions Upstream From the Cap Site are Required for Efficient and Accurate Transcription of the Rabbit fi-Globin Gene in Mouse 3T6 Cells. Cell 3; : 695-706. Dynan, W.S. and R. Tijan. (1983). The Promoter Specific Transcription Factor Spl Binds to Upstream Sequences in the SV40 Early Promoter. Cell 15 : 79-87. Ebright, R.H. (1982). in "Molecular Structure and Biological Function". Griffin, J. and Duax, W., editors, Elsevier/North Holland, New York: 91- 102. ’ Gidoni, D., W.S. Dynan and R. Tijan. (1984). Multiple Specific Contacts Between a Mammalian Transcription Factor and Its Cognate Promoter. Nature 312 : 409-413. Gillies, S.D., S.L. Morrison, V.T. Oi, and S. Tonegawa. (1983). A Tissue - Specific Transcription Enhancer Element is Located in the Major Intron of a Rearranged Immunoglobulin Heavy Chain Gene. Cell 1; : 717-728. Goodwin, R.G., C.L. Moneman, F.H. Rottman and J.H. Nilson. (1983). Characterization and Nucleotide Sequence of the Gene for the Common a-subunit of the Bovine Pituitary Glycoprotein Hormones. Nuc. Acids Res. 11 : 6873-6882. Graves, B.J., P.E. Johnson, and S.L. McKnight. (1986). Homologous Recognition of a Promoter Domain Common to the MSV LTR and the HSV tk Gene. Cell 11 : 565-576. .- (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) 170 Grosschedl, R. and H.L. Birnstiel. (1980). Identification of Regulatory Sequences in the Prelude Sequences of an H2A Histone Gene by the Study of Specific Deletion Mutants in vivo. Proc. Natl. Acad. Sci. USA 11 : 1432-1436. Grosveld, G.C., E. deBoer, C.K. Shewmaker, and R.A. Flavell. (1982). DNA Sequences Necessary for Transcription of the Rabbit fl-Globin Gene in vivo. Nature 225 : 120-126. Harvey, R.P., A.J. Robins, and J.R.E. Wells. (1982). Independently Evolving Chicken Histone H28 Genes: Identification of a Ubiquitous H28 - Specific 5’ Element. Nuc. Acids Res. 19 : 7851-7863. Heintz, N. - Unpublished data. Ito, M. - Unpublished data. Jones,-K.A., J.T. Kadonaga, P.J. Rosenfeld, T.J. Kelley and R. Tijan. (1987). A Cellular DNA - Binding Protein that Activates Eukaryotic Transcription and DNA Replication. Cell 18 : 79-89. Kadonaga, J.T. and R. Tijan. (1986). Affinity purification of Sequence - Specific DNA Binding Proteins. Proc. Natl. Acad. Sci. USA 8; : 5889-5893. Knight, G.B., J.M. Gudas, and A.B. Pardee. (1987). Cell Cycle Specific Interaction of Nuclear DNA Binding Proteins with a CCAAT Sequence from the Human Thymidine Kinase Gene. Proc. Natl. Acad. Sci. USA--in press. Kreidberg, J.A. and T.J. Kelley. (1986). Genetic Analysis of the Human Thymidine Kinase Gene Promoter. Mol. Cell. Biol. 6 : 2903-2909. Maxam, A.M. and W. Gilbert. (1980). Sequencing End Labelled DNA with Base - Specific Chemical Cleavages. Meth. Enzymol. g; : 499-560. (26) (27) (28) (29) (30) (31) (32) (33) 171 McKnight, S.L., E.R. Gavis, R. Kingsbury and R. Axel. (1981). Analysis of Transcriptional Regulatory Sequences of the HSV Thymidine Kinase Gene: Identification of an Upstream Control Region. Cell 25 : 385-398. McKnight, S.L., R.C. Kingsbury, A. Spence, and M. Smith. (1984). The Distal Transcription Signals of the_Herpesvirus tk Gene Share a Common Hexanucleotide Control Sequence. Cell 11 : 253-262. Morgan, W.D., G.T. Williams, R.I. Morimoto, J. Greene, R.E. Kingston and R. Tijan. (1987). Two Transcriptional Activators, CCAAT-Box- Binding Transcription Factor and Heat Shock Transcription Factor, Interact with a Human hsp70 Gene Promoter. Mol. Cell. Biol. 1 : 1129-1137. Sierra, F., G. Stein, and J. Stein. (1983). Structure and 1g vitgo Transcription of a Human H4 Histone Gene. Nuc. Acids. Res. 11 : 7069-7086. Stuart, P., M. Ito, C.J. Stewart and S.E. Conrad. (1985). Induction of Cellular Thymidine Kinase Occurs at the mRNA Level. Mol. Cell. Biol. § : 1490-1497. Tooze, J. DNA Tumor Vigusgs : The Molecular Biology .9; Tumor Viruses. Second Edition, Revised. (1981). by Cold Spring Harbor Laboratories. Travali, S., R.E. Lipson, D. Jaskulski, E. Lauset, and R. Baserga. (1988). Role of the Promoter in the Regulation of the Thymidine Kinase Gene. Mol. Cell. Biol. 8 : 1551-1557. Wasylyk, B., R. Derbyshire, A. Guy, D. Molko, A. Roget, R. Teoule and P. Chambon. (1980). Specific 1p vitro Transcription of Conalbumin Gene is Drastically Decreased by Single - Point Mutation in T-A-T-A box Homology Sequence. Proc. Natl. Acad. Sci. USA 11 : 7024-7028. 172 (34) Wynshaw-Boris, A., T.G. Lugo, J.M. Short, R.E.K. Fournier, and F.W. Hanson. (1984). Identification of a cAMP Regulatory Region in the Gene for Rat Cytosolic Phosphoenolpyruvate Carboxykinase (GTP). J. Biol. Chem. £52 : 12161-12169. 173 SUMMARY AND CONCLUSIONS The regulation of the cell cycle dependent expression of the thymidine kinase gene was examined. The experiments in Appendix 1 showed us that thymidine kinase (TK) enzyme and mRNA levels are regulated in a cell cycle- dependent manner in CV-l cells after mitogenic stimulation by addition of serum growth factors or by infection with SV40 virus. In both cases, TK induction paralleled the entryof CV-l cells into S phase. However, TK mRNA typically accumulated to a higher level in the virus infected cells. Most of the enzyme induction was accounted for by increases in the level of mRNA, and this indicated that the control of TK in the cell cycle might be at the level of transcription of this gene. The experiments described in Chapters 2 and 3 show that both a transcriptional and at least one type of post- transcriptional control are involved in the cell cycle regulation of TK. The experiments in Chapter 2 describe nuclear run-on experiments which show a very low level of transcriptional induction in both systems. A transient six fold increase in transcription was seen at the Gl/S 174 phase border in serum-stimulated cells, and a 2.0-3.5-fold increase has been noted at the beginning of S phase in SV40 infected cells. It is possible that the increase is lower in SV40 infected cells because the timing of infection of each cell within each population may be highly variable. This might show up in our assays as a slow but steady increase in transcription even though the actual rate per cell may be much higher. This seems likely as this slow steady type of induction is observed with the mRNA steady state level as observed on agarose gels. The experiments presented in Chapter 3 begin to focus on possible mechanisms of post-transcriptional regulation of TK. Specifically, mRNA half-lives were examined and were found to be quite long throughout the cell cycle. The 20 hour half-life seen in S phase helps to explain the high levels of mRNA steady state levels seen at later times in the cell cycle. Transcription takes place throughout approximately a six hour period of time, so the 10-20-fold increase in levels of mRNA seen later in S phase and into G2 are probably a result of the message accumulating over this period of time and remaining intact due to its lengthy half-life. ' Our highest measurable levels of TK mRNA occur very 175 late in the cell cyle and seem to depend upon DNA synthesis for efficient termination of transcription. It is not known how this dependence occurs, and this suggests a possible area of research to pursue. It is known that TK enzyme operates via a feedback mechanism due to the accumulation of its end product TTP. It is possible that intracellular pools of TTP may function to regulate TK mRNA by stimulating transcription when levels fall below a certain limit, and by shutting off expression when TTP levels reach another specific level. Another possible area of research to pursue would involve the examination of serum-dependence on TK enzyme expression. Our results, as well as others, have shown that the removal of serum growth factors from tissue culture cells has a rather negative effect on TK enzyme activity. We have shown that this is not related to mRNA levels. Therefore, it is possible that serum growth factors induce some type of post-translational ( modification of the thymidine kinase enzyme which allows it to become active. Removal of serum may abolish this activity. Finally, the results presented in Chapter 4 will allow future work to progress on the characterization of gene-specific regulatory elements which are involved in (J 176 transcriptional control of the TR gene, as well as cell cycle control. The promoter elements and homologies found may be useful in a molecular characterization of sequence elements necessary for cell cycle dependent expression. APPENDIX I MOLECULAR AND CELLULAa BIOLoov. June 1985. p. 1490-1497 0270-7306/857061490-08502.00/0 Copyright 0 1985. American Society for Microbiology 177 Vol. 5. No. 6 Induction of Cellular Thymidine Kinase Occurs at the mRNA Level PHILIP STUART.l MORIKO ITO.2 CHRISTINE ST l'iWAR'I‘.2 AND SUSAN E. CONRAD" Department of Microbiology. Michigan State University. East Lansing. Michigan 48824 .2 and Department of Molecular Biology. University of C altfornia. Berkeley, California 94720I Received 16 November 1984/Accepted 26 February 1985 The thymidine kinase (TK) gene has been Isolated from human genomic DNA. The gene was passaged twice by transfection of LTK" cells with human chromosomal DNA, and genomic libraries were made in A Charon 30 from a second-round TK" transformant. When the library was screened with a human Alu probe, seven overlapping A clones from the human TK locus were obtained. None of the seven contained a functional TK gene as judged by transfection analysis, but several combinations of clones gave rise to 'I‘K+ colonies when cotransl'ected Into TK’ cells. A functional cDNA clone encoding the human TK gene has also been isolated. Using this cDNA clone as a probe In restriction enzyme/blot hybridization analyses, we have mapped the coding sequences and direction of transcription of the gene. We have also used a single-copy subclone from within the coding region to monitor steady-state levels of TK mRNA in serum-stimulated and simian virus 40-infected simian CVl tissue culture cells. Our results indicate that the previously reported increase in TK enzyme levels seen after either treatment is paralleled by an equivalent increase in the steady-state levels of 'I‘K mRNA. in the case of simian virus 40-infected cells, the induction was delayed by 8 to 12 h. which is the length of time after infection required for early viral protein synthesis. in both cases, induction of TK mRNA coincides with the onset of DNA synthesis. but viraily infected cells ultimately accumulate more TK mRNA than do serum-stim- ulated cells. Thymidine kinase (TK) is an enzyme in the pyrimidine salvage pathway that catalyzes the phosphorylation of thy- midine to dTMP. Many mammalian cells. including human HeLa cells. contain both cytoplasmic and mitochondrial forms of the enzyme (3). but the cytoplasmic form alone concerns as here. The regulation of the synthesis of TK is interesting because it is typical of that seen for many enzymes involved in DNA metabolism. TK activity is closely linked to the growth state of the cell. being present in rapidly growing but not in resting cells (13). In synchronized popu- lations of cells. the activity is low in resting or 01 phase cells. but increases dramatically 10 to 20 h after the cells are released from arrest by serum stimulation. in parallel with the onset of DNA synthesis and entry into S phase. This induction is not absolutely dependent upon DNA synthesis (13). but does require both RNA and protein syntheses, suggesting that induction may be at the level of transcrip- tion. TK can also be induced by infection of resting cells with papovaviruses such as simian virus 40 (SV40) and polyoma (16. 17). and the viral genes required for this induction are the large 'I‘ antigens (30). Whether viral induc- tion occurs by the same or a different mechanism(s) as serum induction is a question that remains to be answered. , The TK gene provides a useful model system for carrying out a molecular analysis of genes that are cell cycle regulated and induced by viral infection. First. TK shows a great increase in activity (10- to 20—fold) after both serum and viral induction. Moreover. TK enzyme assays are both sensitive and easily performed. We can genetically select both for (hypoxanthinc-aminopterin-thymidine media) and against (bromodeoxyuridine media) the TK’ phenotype. and many TK‘ cell lines exist. it has been shown that LTK" cells transfected to a TK ' phenotype with heterologous (human. rat. or hamster) chromosomal DNA containing a functional T K gene exhibit normal cell cycle regulation of the gene (31). Recent experiments with a cloned human gene (5) indicate ' Corresponding author. that the sequences required for cell cycle regulation are . closely linked to the gene and function after transfection into TK‘ cells. Thus. this system should offer the chance to dissect the sequences involved in cell cycle-specific gene regulation. The mechanisms by which the expression of cell cycle-dc- pendent genes is controlled. and by which the papovaviruses override these controls. remain obscure. although many of the initial observations were made more than 15 years ago. T o a large extent this is because molecular probes for these genes and their transcripts have not been available. in this paper we report the isolation of both the human chromo- somal TK locus and a functional human TK cDNA clone. isolation of the chromosomal locus has previously been reported by several investigators l5. I9, 22). and our map- ping is essentially in agreement with their data. In addition. we have used the cDNA clone to map mRNA coding sequences and the direction of transcription within the locus. We have also used a subclone from within the coding sequence of the gene to monitor TK mRNA levels in serum-stimulated and SV40-infected simian CV1 cells. Our results show that the induction of TK enzyme activity is paralleled by an increase in mRNA levels. indicating that induction may be at the level of transcription. MATERIALS AND METHODS DNA transfections. Human genomic DNA was prepared from an SV40-transformed cell line. GM638. DNA transfec- tions were done according to the method of Wigler et al. (35). Briefly. 10 to 20 pg of human DNA was added as a CaPO, precipitate to 10“ LTK' cells. and TK' colonies were se- lected in hypoxanthine-uminopterine-thymidinc media. After 2 to 3 weeks colonies were picked and expanded into cell lines. Construction of genomic libraries. Recombinant phage libraries were constructed and screened as described previ- ously (25). Genomic DNA from secondary transformant cell 1490 VOL. 5. 1985 line B4 was partially digested with Saa3A. and 15- to 20-kilobase (kb) fragments were purified by centrifugation through 5 to 20% sucrose gradients. Charon 30 DNA was digested with BamHI. and the arms were purified away from the internal fragments. The human DNA was ligated to the A arms. packaged in vitro. and used to infect Escherichia coli K802. Approximately 10" phage from this unamplified li- brary were screened by the plaque hybridization method of Benton and Davis (2). using Blur DNA labeled by nick translation as a radioactive probe. Nucleic acid hybridizations. Hybridizations to DNA filters were performed at 68°C under aqueous conditions in 6x SSC (SSC is 0.15 M NaCl plus 0.015 M sodium citrate). 10x Denhardt solution—0.1% sodium dodecyl sulfate (SDS) with approximately 10 to 20 ng of 32P-labeled probe per ml. Hybridizations were done for 15 to 24 h. For northern blots. hybridizations were done in 50% formamide-3x SSC-5x Denhardt solution-50 mM sodium phosphate (pi-l 6.8)—5% dextran sulfate—50 pg of denatured sheared salmon s rm DNA per ml—0.l% SDS-approximately 10" cpm of 3‘P-la- beled probe per ml. Hybridizations were carried out for 15 to 20 h at 42°C. . After hybridization. all filters were rinsed once at room temperature in 2x SSC-0. 1% SDS and then washed two to three times in the same solution at 68°C. In cases where the background was still high after such treatment. an additional wash in 0.1x SSC—0.1% SDS was added. Viral infection of tissue culture cells. Simian CV1 (African green monkey kidney) cells were grown to confluence in media containing 5% calf serum-5% fetal calf serum. Cells were allowed to remain at confluence for 48 h and were then infected with SV40 at a multiplicity of infection (MOI) of 5. Infections were done by removing media from cells. infect- ing for 1 h with a concentrated viral stock. and then replacing the original media. Mock-infected cells were treated for l h with serum-free media. At various times alter infection. plates were harvested for RNA. protein. or DNA synthesis analyses. Preparation of poly(A*) RNA and northern analysis. Poly- adenylated [polytA')] RNA was prepared from tissue cul- ture cells as follows. Cells were washed once with phos- phate-bufi'ered saline (PBS) without calcium and magne- sium. They were then lysed on the plate with 1 ml of lysis buffer (0.5 M NaCl. 10 mM Tris [pl-I 7.5]. 1 mM EDTA. 1% SDS, 200 pg of proteinase K per ml) per 100-mm plate. The lysed cells were scraped from the plate. and cellular DNA was sheared by passage three times through a 21-gauge needle. Fresh proteinase K was added to 100 tag/ml, and the solution was incubated at 37°C for 30 to 60 min. A small amount (~0.1 ml of packed volume per 100-mm plate) of solid oligodeoxythymidylate-cellulose was added. and the RNA was bound in bulk by shaking at room temperature for l h. The mix was then loaded into a small column and washed with 20 column volumes of loading buffer (0.5 M NaCl, 10 mM Tris. 1 mM EDTA. 0.2% SDS) followed by 20 column volumes of the same buffer with 0.1 M NaCl. RNA was then eluted with 2 column volumes of 10 mM Tris-l mM EDTA and ethanol precipitated after the addition of sodium acetate and tRNA carrier. Precipitated RNA was suspended in gel sample buffer (50% formamide, 1x running buffer. 2.2 M formaldehyde). heated to 60°C for 5 min. and run on 1.2% agarose gels containing 2.2 M formaldehyde. Running buffer was 20 mM MOPS (morpholineprOpanesulfonic acid). pH 7-1 mM EDTA—5 mM sodium acetate. Gels were run in 1x buffer plus 2.2 M formaldehyde. After electrophoresis. gels were 178 INDUCTION 0F CELLULAR TK 1491 soaked once briefly in water and then in 20x SSC for 30 min. RNA was transferred to nitrocellulose filters in 20x SSC. TK extraction. Cells to be assayed for TK activity were harvested by a modification of the method of Johnson et al. (13). Two IOO-mm plates of confluent cells were washed with cold PBS and taken up with rubber policemen in 1 ml of PBS per plate. The cells were pooled. pelleted. and suspended in 200 pl of NonidetP-40 (NP40) reagent (50 mM Tris-hydro- chloride [pH 8.0], 3.6 mM B-mercaptoethanol. 0.5% NP40). The lysed suspension was vortexed. the nuclei were pellet- ed. and the supematants were frozen and stored at -70°C in two loo-pl aliquots. TK assay. TK activity was determined by a modification of the method of Ives et al. (11) and Johnson et al. (13). Either 5. 10. or 20 ul of the thawed cell extract (brought to a 20-p.l volume. if necessary. with NP40 reagent) was added to 60 pl of reaction buffer to yield a final concentration of 50 mM Tris-hydrochloride (pH 8.0). 15 mM NaF. 3.6 mM B-mer- captoethanol. 5 mM ATP. 2.5 mM MgCl2. 0.08 mM unla- beled thymidine. and 50 pCi of _ [’Hjthymidine (specific activity. 20 Ci/mmol) per ml. The reaction mix was incu- bated at 37°C for 10. 20. or 30 min and stopped by immersing for 2 to 3 min in a boiling-water bath. Control reactions without ATP were included at zero time and 30 min. Ten- or ZO-nl samples of the reactions were spotted in duplicate on Whatman DEBl anion-exchange filter paper as follows: +ATP. t = 30 min: -ATP. t = 30 min; —ATP. t = zero: and for a total count of [3HIthymidine available: +ATP. t = 30 min. without washing. The filters were dried. washed twice in 1 mM ammonium formate and once in methanol. and dried. The disks were then placed in scintillation vials. and dTMP product was eluted by adding 1 ml of 0.1 M HCI~0.2 M KCl and shaking for 20 to 30 min. A lO-ml portion of liquid scintillation fluid was added per vial. gently shaken for 2 to 4 h. and counted for "H. TK activities are expressed in nanomoles of deoxythymi- dine converted to dTMP per minute per microgram of extract protein: TK units = [(percent conversion)(6.6 nmol of deoxythy- midine in 80 pl of reaction mixture)l/[(reaction time) (micrograms of protein in reaction)! Percent conversion = “counts per minute of +ATP at t = 30 min - counts per minute of -ATP at t = zero)/(counts per minute of unwashed control - backgroundll X 100 The protein concentrations in the extracts were deter- mined by both the Bradford (4) and the Lowry (23) protein assays. using a standard curve of bovine serum albumin in each case. In the 8- to 40-ug range of protein. NP40 (nonionic detergent) did not interfere with the Lowry reac- tion. The amount of enzyme activity observed was directly proportional to both the concentration of the enzyme and the elapsed time of reaction. up to conditions converting 50% of the substrate to product. ' DNA pulse-labeling. To determine the specific activity of DNA. confluent cells on loo-mm plates were labeled in 1 ml of media with 1 pCi of [JHIdeoxythymidine and 4 x 10‘7 M' uridine for 1 h. Labeled cells were then washed with PBS. trypsinized in 2 ml. spun down. and suspended in 0.5 ml of PBS. Cells were counted on a hemacytometer at this point. 'l‘wo IOU-pl aliquots were trichloroacetic acid precipitated onto fiber glass filters with 5% trichloroacetic acid. washed with ethanol, dried. and counted for 3H in 10 ml of aqueous liquid scintillation floor. The remaining 300 pl of the PBS suspension of cells was assayed for DNA content by the 1 7 9 I492 STUART ET AL. MOL. CELL. BIOL. A. 1);. 4am“) 321‘;— jHBJKD "’L - a): f ‘ :igfibfil‘bl 1o" ""9"” ' “who. ' attention 3 E Shiv i t Ill. l‘ fit”! it “5% H 5mm 56 (am—— ' I s It I I II I II I r I'III II IIII II III I r I H c. l’ei' l>03l W q“. "0 D i——-—l H H H CDNA E. o—a-rl—ar. 0?,- o? Xho I- Hindlll IA M w“? v ‘ 5; ‘7' Direction of Human ‘l’K Transcription "(b H FIG. 1. Restriction map of the human TK locus. (A) Seven overlapping clones from the human TK locus were mapped by restriction enzyme/blot hybridization analysis. The relative positions of these seven clones are shown. Cotransfection of clones I + 11. 1 + 9. and 11 + 12 gives rise to TK’ colonies. (8) Restriction enzyme map of the human TK locus. Enzymes used were: E. Emil“: S. Sstl: l-i. Ilindlll: K. Kpnl; B. BamH1: X. lel. The 1.4-kb Xhol-EcoRl fragment used as a probe in northern analyses is underlined. (C) Fragments within the TK locus that hybridize to a human Alu repeat probe (Blur-8). Numbers below the line indicate the minimum number of Alu repeats within that fragment. (D) Fragments within the human TK locus that hybridize to the human cDNA clone pHuTKcDNA7. (E) Two fragments from within the cDNA clone were used to map the direction of transcription of the TK gene. The 5' probe was a 1.03-kb leloliindlll fragment. and the 3' probe was a 0.52-kb Xle-Ht’ndlll fragment. The Xhol sites in both cases were located in the vector DNA. diphenylamine colorimetric assay (7). The cells were incu— bated for at least 30 min in 0.1% SDS-0.1 mg of proteinase K per ml at 37°C. ethanol precipitated. and suspended in TE before the assay was performed. Radiolabellng of proteins. Cells were labeled in methio- nine-free media with [”SJmethionine at time points every 4 h postinfection. After a 45-min incubation period. cells were washed twice in ice-cold PBS and once in T-antigen wash bufl'er (0.137 M NaCl. 20 mM Tris-hydrochloride [pl-i 9.0]. 1x Ca2*-Mg+ salts). Cells were lysed on the plate by the addition of 1 ml of extraction buffer (Toantigen wash, 10% glycerol. 1% NP40. 1 mM phenylmethylsulfonyl fluoride) and incubated for 20 min at 4°C. Cells were scraped from the plate Cand centrifuged. and the supematants were stored at -70° . Immunopreclpitatlon. The lysate (1 ml) was added to 80 pl of a 50% suspension of protein A-Sepharose. Anti-T-antigen antisera were added. and this mixture was kept on ice and vortexed every 5 min for a 30-min period. The Sepharose- antibody—protein complex was centrifuged in a microcentri- fuge. and the pellet was washed once with PBS. twice with wash buffer (0.5 M LlClz. 100 mM Tris-hydrochloride [pH 6.8)). twice with 1% deoxycholate—l% NP40 in Tris-buffered saline. and finally twice in PBS. The final pellet was dried, and 50 pl ol' protein sample buffer was added. SDS-polyacrylamide gel electrophoresis. Samples were heated to 100°C for 3 min and spun in a microcentrifuge. The supernatants were resolved by SDS-polyacrylamide gel elec- trophoresis on a 10% polyacrylamide gel by the procedure of Laemmli ( 18). The gel was stained. destained. fluoro- graphed. dried. and exposed to X-ray film at -70°C. RESULTS Molecular cloning and sequence organization of the human TK gene. The human TK gene was isolated by using an experimental protocol similar to one used by other investi- gators (5. 22). Briefly. total human DNA from a transformed cell line. GM638. was used to transfect murine LTK’ cells to a TK’ phenotype. Genomic DNA was prepared from cell lines derived from these primary transformants and used to again transfect LTK' cells to a TK” phenotype. Genomic DNA was prepared from cell lines derived from these secondary transformants and examined by blot-hybridiza- tion. The radioactive probe used to detect the human DNA in these cell lines was Blur-8. a cloned member of the human Alu family of repeats (12). DNAs from a total of five difl'ercnt secondary transformants derived t'rom the two different primary transformants were digested with either BamI-Il or EeoRl, and the pattern of hybridization to the Blur-8 probe was examined. All five cell lines contained one to four Alu-hybridizing fragments ranging in length from 5.000 to 20.000 base pairs (data not shown). Identical patterns of hybridization were seen to Blur-ll. another cloned Alu repeat. but not to a pBR322 control. Cell line B4. which contains one strong Ala-hybridizing BamHI fragment of approximately 15 kb. was chosen as a source of DNA for molecular cloning of the TK gene. B4 DNA was partially digested with Sau3A and cloned into A Charon 30 as described in Materials and Methods. A library of approximately 10" phage was screened with the Blur-8 probe. and seven overlapping phage from the human TK locus were isolated. Endonuclease cleavage maps of the seven A Charon 30-human clones were determined. and the combination gives a restriction enzyme map of a 40-kb stretch of DNA. This map (except that portion unique to A clone 10). along with the extent and length of each human TK clone insert. is shown in Fig. I. That it is an accurate representation of the human TK gene within the secondary transformant B4 genome was confirmed by Southern blot analysis of B4-LTK’ genomic DNA (data not shown). Many similarities exist between this map of a twice-transfected GM638 TK gene and the published restriction endonuclease maps of other human TK gene isolates: the HeLa 'l‘K gene VOL. 5. 1985 (5. 22) and the placental TK gene (19). A cDNA clone able to express TK activity (see below) was labeled with 32P and hybridized to a blot of a gel containing restriction enzyme digests of A clones l, 9. and 11. The resulting pattern of cDNA hybridization. shown schematically in Fig. 1. indi- cates a minimum of three intervening sequences in the human TK gene and. assuming that pHuTK-c7 contains nearly all of the mature TK mRNA sequence, a maximum gene length of approximately 14.5 kb. The seven A clones from the human TK locus were transfected individually and in pairwise combinations 9 + 11. l + 9. l + 11. and 11 + 12 onto Rat-3 TK‘ cells (29). and TK’ transfectants were selected. Although no one clone contains the intact gene (i.e.. gives rise to TK’ colonies). overlapping phage l + 9. 1 + 11. and 11 + 12 do give rise to TK‘ colonies. We interpret these results to show that the overlapping clones can recombine during transfection to generate an intact gene. as has been suggested by others (22). One interesting aspect of our results is that A phage II + 12, which according to our restriction mapping contain little or no overlapping sequence. do give rise to TIC colonies upon cotransfection. We suggest that this may be due to recombination between different Alu repeats within the large intron in the TK gene. These results indicate that we have cloned the entire human TK locus on overlapping clones. Isolation of a functional human TK cDNA clone. A human cDNA library was screened for the presence of cDNA clones homologous to the TK gene. The library used was a gift of H. Okayama and P. Berg; it had been constructed by using poly(A *) RNA from log-phase GM639 cells. an SV40- transformed human cell line. by a method favoring full- length cDNA copies of mRNAs (28). In addition. the cDNAs were linked to an expression vector containing. along with the pBR322 replication origin and B-lactamase gene. several SV40 transcriptional control sequences, the early gene pro- moter. a late gene splice donor and acceptor, and a polyad- enylation signal, which allow expression of the cDNA insert when the clones are introduced into mammalian cells (Okayama and Berg, personal communication). In a screening of 2.4 x 10" bacterial colonies. over 200 positive clones were found. Of these. 20 were picked. and of these 16 were successfully isolated. Two clones containing the longest cDNA inserts were tested for function by trans- fection into Rat-2 TK‘ cells. and both were able to stably transform the cells to a TK' phenotype at a frequency equal to or greater than that of a chicken TK gene clone (29) used as a control. Since the 1.500-basc pair size of the cDNA insert of pHuTK-c7 correlates well with the known TK mRNA size. and since the clone expresses TK activity in mammalian cells at a high level. we believe this cDNA isolate must contain nearly all, if not all. of the sequence of the mature TK mRNA. The known transcriptional polarity of the cDNA insert with respect to its expression vector allowed the direction of transcription of the human TK gene to be readily deter- mined. and these results are also illustrated in Fig. 1. Since this work was initiated, the DNA sequence of a similar human TK cDNA clone has been reported (6). Viral induction of simian TK. Infection of confluent mam- malian tissue culture cells by the papovaviruses SV40 and polyoma causes several changes in cell metabolism. includ- ing the induction of cell DNA synthesis and many of the enzymes involved in DNA metabolism. We have studied the induction of TK activity at a molecular level. using a DNA probe from within the TK coding region. A subclone con- 1 8 O INDUCTION OF CELLULAR TK 1493 A') 222.22 8') 223.25 °s~.' s a: or r. a 2: f E‘f -."_- ~ ’ ':-‘.,.' '- " ‘ 52’ It” 0 TK Probe SV40 Probe FIG. 2. Northern blot analysis of SV40-infected CV1 cells. Con- fluent plates of CV1 cells were infected with wild-type SV40 at an MOI of 5 at zero time. Poly(A’) mRNA was isolated from cells at 12-h intervals after infection. RNA from equal numbers of cells was electrophoresed on 1.2% formaldehyde gels and transferred to nitrocellulose paper. Duplicate northern blots were hybridized to (A) pHuTK 1.4 probe and (B) SV40 probe. taining the 1.4-kb XhoI-EcoRl fragment underlined in Fig. l (pHuTK 1.4) was used as a probe for northern blot analysis. Confluent simian CV1 cells were infected with SV40 virus at an MOI of 5. and poly(A‘) mRNA was prepared at various times from 0 to 48 h postinfection. RNA from equal numbers of cells at each time point was electrophoresed on a 1.2% formaldehyde-agarose gel. transferred to nitrocellu- lose filters. and hybridized with 32P-labeled pHuTK 1.4 probe as described in Materials and Methods. The results of such an analysis are shown in Fig. 2A. The length of the TK mRNA is approximately 1.5 kb. as has been previously reported for human cells (21). The mRNA is barely detect- able at zero time. and the first increase is seen at 24 h. Figure 28 shows a northern blot run in parallel hybridized with an SV40 probe. SV40 early and late mRNAs are apparent by 12 and 24 h postinfection. respectively. Densitometer tracings of the autoradiogram in Fig. 2A are plotted in Fig. 3 and show a final level of TK mRNA induction of approximately lS-fold at 48 h. In control experiments. mock-infected cells showed no induction within the 48-h time period (data not shown). Also shown in Fig. 3 is a plot of TK enzyme activities from cells infected in parallel with those used for the northern analysis. As can be seen. both the time course and the extent of enzyme induction are similar to that of the mRNA. Also plotted is [’l-Ilthymidine incorporation into DNA. indicating the onset of S phase by 24 h postinfection. It has previously been shown (10) that this incorporation represents both viral and cellular DNA synthesis occurring after infection of confluent monkey cells with SV40. Since it has been reported that the viral protein T antigen is responsible for the observed TK induction (30). we I494 STUART ET AL. 25< t 2 24 30 48 Hours Post lntoction FIG. 3. SV40 induction of TK in simian CV1 cells. Confluent plates of CV1 cells were infected with SV40 at an MOI of 5 at zero time. At 12-h intervals after infection. samples were taken for TK enzyme assays. TK mRNA analysis. and pulse-labeling of total DNA. TK enzyme and mRNA analyses are plotted using the scale on the left; DNA labeling is plotted using the scale on the right. TK enzyme activity is expressed as follows: 1 unit = 1 nmol of deoxythymidine converted to dTMP/min per pg of protein. (0) TK units x 10". TK mRNA levels were estimated from densitometer tracings of the northern gel shown in Fig. 2A. (A) Relative TK mRNA levels. DNA was labeled in vivo for 60 min with 1 (LG of [’Hlthymidinc per ml. Counts incorporated were determined by using acid precipitations. and DNA concentrations were detemtined by a DNA colorimetric assay (diphenylamine reaction). (0) Specific activity of DNA (counts per minute per microgram). The first increase in specific activity seen at 24 h postinfection indicates the onset of S phase. Although DNA synthesis seems to precede TK induction. this is deceiving since the DNA synthesis measurement is done by pulse-labeling and the TK mRNA and enzyme levels are cumulative. determined the earliest detectable TK induction relative to the presence of T antigen. A more extensive time course was performed. with time points taken every 4 h postinfection. Duplicate RNA blots were done. and one hybridized to pHuTK 1.4 probe and one hybridized to SV40 probe (Fig. 4). Figure 4A shows that TK mRNA induction may begin by 16 h and has clearly begun by 20 h. Shown in Fig. 4B is the same blot rehybridizcd with a c-myc probe, which shows approximately equal amounts of RNA at each time point except 0 and 8 h. The RNA sample at 8 h has apparently been lost during preparation. This film was scanned by a densitometer. and the amount of 'I‘K mRNA relative to c-myc mRNA was determined. These results are shown in Fig. 5. It has previously been reported that the levels of c-myc mRNA increase very quickly within 2 h of serum stimulation of mouse fibroblasts (8, 15). The levels of c-myc mRNA also appear to increase between 0 and 4 h after 181 MOL. CELL. BIOL. infection of CV1 cells with SV40 in this experiment. After the 4-h time point. the c-myc mRNA level is relatively constant over the course of this experiment. allowing us to say that the increase in TK mRNA levels is not a result of a general increase in the level of cellular mRNA. Figure 4C shows a parallel northern blot hybridized with SV40 probe. In this experiment. early SV40 mRNA appears by 12 h and late mRNA appears by to h. In other experiments. SV40 early mRNA has been detected at 8 h postinfection. Also shown in lane 1 is poly(A’) mRNA from C087 cells. a h o 82 Sf vofieuu m oovo§28§ l..- .“.‘. Al 'F .3 W, -. ':. " . ‘- 1‘ '.' .2 . ‘- . (I ”LU‘Q' ”’Q’; ' I. 75.1 n l "I "* ' I t'u‘l' '1" ‘ . .‘l’ I I h “ . . r - - . ' "4gl. Feufiffi: '1. ‘ E. {at rf‘iiiflfiz ”is: '- ‘ '1 . v: " " ‘ ‘ ' 'I t“- '1'3" v' 'I I .-_». _,.-.'l 5';- if} ‘ A; ..‘{“‘\'7.?‘_35: vat-#23,: 56 a 'I ’-| - g. ._ W“ s . ' . _.,. SV40 PROBE FIG. 4. ’I‘wenty-four-hour time course of SV40-infected CV1 cells. Coniluent CV1 cells were infected with SV40 as described in the legend to Fig. 3. Plates were harvested at 4-h intervals for mRNA and T-antigen analyses. Also included as a control are RNA and protein from COS7 cells. which constitutively express the SV40 early genes. (A) Hybridization with pHuTK 1.4 probe. (B) Rehy- bridization of (A) with a c-myc probe. (C) Duplicate blot of (A) hybridized with SV40 probe. (D) Acrylamide gel of immunoprecipi- tated proteins pulse-labeled for 60 min with (”Slmethionine. VOL. 5. 1985 simian cell line transformed by SV40. This cell line contains only the SV40 early mRNA. To look directly at SV40 T antigen. SDS-polyacrylamide gels have been run. Cells infected in parallel with those analyzed in Fig. 4 were labeled with (”Slmethionine at 4-h intervals after infection. The proteins were immunoprecipi- tated with anti-T-antigen antibody and electrophoresed on a 10% polyacrylamide gel. The results (Fig. 40) show that large T antigen first appears at 8 postinfection, approxi- mately 8 to 12 h before the increase in TK. mRNA levels. Since one of our long-term goals is to determine the mechanism by which SV40 induces arrested cells to re-enter the cell cycle. we would like to determine whether the same events occur during 01 in virally infeCted and serum-stimu- lated cells. To begin to approach this question. we have compared TK and c-myc mRNA levels in cells induced in parallel by either scrum stimulation or viral infection. Con- fluent CV1 cells were treated either by the addition of fresh media containing 10% serum or by infection with SV40 virus (MOI = S). Poly(A*) RNA was isolated at various times (0. 6. 12. 24. 36. and 48 h) after treatment and analyzed for the levels of TK and c-myc transcripts. Since c-myc mRNA levels increase soon after serum stimulation of mouse fibroblasts. we have also hybridized with a B-Z microglobulin probe as an additional internal control. (A human B-Z microglobulin clone was a gift of Hsiu-Ching Chang.) The results of this experiment are shown in Fig. 6. To analyze this data. we have compared the levels of TK and c-myc mRNA to the B-2 microglobulin internal control. In serum-stimulated cells. TK mRNA levels are significantly increased by 12 h after treatment and decrease somewhat by 48 h. The earliest time of the induction (~12 h) coincides with the onset of DNA synthesis in these cells. and the time course of mRNA induction parallels that of enzyme induc- tion (data not shown). In virally infected cells. the level of 'I'K mRNA does not significantly increase until 24 h postin- fection. but again the increase coincides with the onset of DNA synthesis. We believe that this delay is due to the 8- to 12-h time period required for the virus to infect the cells and express the early viral proteins (Fig. 4). Once this viral induction of TK begins. the magnitude of the response exceeds that seen in serum-stimulated cells. c-mvc 7K hrs. post inlectlon Tch-myc ratio £4 24 .90 _”/\._’_\__-~ to .23 _‘J_\__—~=. 12 .13 g 5,... 4 < 0.1 .m... o < 0.1 Janna... cos 7 < 0.1 FIG. 5. Densitometer tracing of autoradiogram in Fig. 4B. The ratio of TK/c-myc RNA is shown. 182 INDUCTION OF CELLULAR TK 1495 SERUM SV40 CID N V ‘0 0 CD 01 V (D co '- N to st '- N 6') st _ . .r..,m_}..,,. ' “". _ - :v. a» . i:§1§"-'“ «:35 £3ui‘.‘ % .§ \ myc—> ...-“E a". . Ik-O 4“.“ . --. fl-2-> ~'-..' ~‘... FIG. 6. Comparison of serum and SV40 induction of CV1 cells. Confluent CV1 cells were induced to reenter the cell cycle either by addition of fresh media containing 10% serum or by infection with SV40 (MOI = 5). At various times (6. 12. 24. 36. and 48 h) after treatment. poly(A’) RNA was prepared. and RNA from eqtutl numbers of cells was electrophoresed on a denaturing agarose gel as previously described. The RNA was transferred to nitrocellulose filter paper and hybridized with a mixture of TK and c-myr probes. The same filter was later rehybridizcd with a human B-2 microglobu- lin probe. The picture above shows the two lilms superimposed on one another. so that the three RNAs can'be viewed together. The level of c-myc mRNA appears to increase approxi- mately threefold by 6 h after both serum stimulation and viral infection. In the case of serum-stimulated cells this level remains relatively constant. although it may decrease somewhat by the 48-h time point. In the case of the SV40. infected cells we see a dramatic disappearance of the c-myc mRNA at the 36- and 48-h time points. DISCUSSION We report the molecular cloning of the human TK gene and its use in monitoring TK mRNA levels in both serum- stimulated and virally infected simian CV1 cells. This gene has been isolated by others (4. 14. 17). and our mapping results are largely in agreement with theirs. Although the restriction maps of the 5’ ends of the genels) are virtually identical. considerable differences are seen within the large intron and at the 3' end. Although three of four maps contain a Baml-ll site at the 3' end of the gene. the fourth map does not (4). These differences may reflect true polymorphisms in the gene or may be an artifact of the rounds of transfections used to isolate the genes. Our isolation of a functional cDNA clone has allowed us to determine the orientation of the gene. to estimate a minimum number of introns, and to localize the approximate positions of the 5' and 3' ends. It has been reported that papovavirus infection or serum stimulation of contact-inhibited cells increases the levels of TK enzyme activity by approximately 20-fold (ll. 12). Using 1496 STUART ET AL. a subclone from within the TK mRNA coding region. we have examined the steady-state levels of TK mRNA in resting and stimulated simian CV1 cells. Our results show that the length of the simian TK mRNA is approximately 1.5 kb. the same size as has been reported for the human mRNA. ln contact-inhibited CV1 cells. the level of the RNA is quite low. and in some experiments it is barely detectable. After infection with SV40 the levels of RNA increase. reaching a maximum of about 15- to 20-fold by 48 h postin- fection. The first induction detected occurs at between 16 and 20 h postinfection and coincides with the onset of DNA synthesis. It is also interesting to note that the first accumu- lation of TK mRNA is not detectable until 8 to 12 h after the appearance of T antigen. which has been implicated as the viral protein necessary for induction. That this time interval is somewhat longer than that reported previously is due to our ability to detect T antigen earlier than previous investi- gators. In the long term. we would like to determine whether T antigen is acting directly to induce the synthesis of TK. or whether T antigen acts indirectly by initiating other events during those 8 to 12 h which in turn induce TK activity. In the case of scrum-stimulated cells. induction of TR mRNA occurs by 12 h after treatment and again coincides with the onset of DNA synthesis. Thus it appears that the time interval between either serum or T-antigen stimulation of CV1 cells and the onset of DNA synthesis is approximately 8 to 12 h. The TK mRNA seen after induction seems to be identical in size to that seen in untreated cells within the limits of resolution of these gels. Of course. small changes in molecular weights or 5' and 3' ends would not be detected in these experiments. These results indicate that at least most of the induction of TR enzyme activity can be accounted for by increases in the steady-state levels of TK mRNA. These changes in mRNA levels may be due to control at several steps during RNA synthesis. including transcription. processing. and RNA stability. Experiments studying cell cycle regulation of dihydrofolate reductase (DH FR) indicate that the increase in DHFR mRNA levels seen during 8 phase is due to dill‘er- enccs in mRNA stability (20). Also interesting in this regard is the result that the DNA sequences required for DHFR regulation map to the 3' end of the gene (14). In experiments studying regulation of the chicken TK gene in differentiating muscle cells. it has been shown that the sequences required for regulation are localized within the body of the gene itself (26. 27). Thus it is possible that cell cycle-controlled genes such as TK and DHFR are not regulated (at least exclu- sively) at the level of transcription. Experiments are cur- rently in progress to measure the rates of TK transcription before and after SV40 infection and to map. the DNA sequences required for cell cycle regulation of TK. We have also presented preliminary evidence that c-myc mRNA levels increase two to threefold within 6 h of both serum stimulation and SV40 infection of resting CV1 cells. This result is surprising. since we have not been able to detect any early viral mRNA or protein synthesis by this time. Since the magnitude of induction is quite low. further experiments will be required to determine whether this effect occurs reproducibly. but if it does it implies that factors other than early viral gene expression are responsible for the increase in c-myc mRNA levels. This viral induction of c-myr- is not sufficient to cause the cells to progress to S phase. since the synthesis of T antigen at ~8 h postinfection is necessary for induction of both TK and DNA synthesis (30). One marked difference between the serum- and virus- stimulated cells is the fact that c-myc mRNA is absent in 183 MOL. CELL. BIOL. virally infected. cells at 36 and 48 h. We are currently investigating the mechanism of this shutoff. In summary. we have shown that TK mRNA levels increase dramatically in both serum-stimulated and SV40- infected CV1 cells. Although this induction is delayed by approximately 8 to 12 h in the SV40-infected cells. we believe that this delay is due to the fact that it takes 8 to 12 h for the virus to infect cells and express its early proteins. In both cases. TK induction parallels the entry of cells into S phase. but the TR mRNA accumulates to a higher level in the virus-infected cells. ACKNOWLEDGMENTS We thank Michael Botchan. in whose lab this work was initiated. for helpful advice and discussions. This work was supported by Public Health Service grant CA30490 to M. Botchan from the National Institutes of Health and by NRSA Fellowship CA06928—02 and Public Health Service grant CA37144-01 to S. Conrad from the National Institutes of Health. LITERATURE CITED 1. Armelln. H. A., M. C. S. Amelia. K. Kelly. 1'. Stewart. P. Leder. B. H. Cochran. and C. E. Stlla. 1984. Functional role for c-myc in mitogenic response to platelet-derived growth factor.‘ Nature (London) 310:655—660. 2. Benton, W. D., and R. W. Davis. 1977. Screening Agt recombi- nant clones by hybridization to single plaques in situ. Science 196:180. 3. Berk. A. J., and D. Clayton. 1973. A genetically distinct thymidine kinase in mammalian mitochondria. J. Biol. Chem. 248:2722-2729. 4. Bradford, M. M. 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-254. 5. Bradshaw, H. D. 1983. Molecular cloning and cell cycle-specific regulation ofa functional human thymidine kinase gene. Proc. Natl. Acad. Sci. U.S.A. 805588-5591. 6. Bradshaw, H. D., and P. L. Deininger. 1984. Human thymidine kinase gene: molecular cloning and nucleotide sequence of a cDNA expressible in mammalian cells. Mol. Cell. Biol. 4:23")- 2320. 7. Burton. K. S. 1956. A study of the conditions and mechanisms of the diphenylamine reaction for the colorimetric estimation of deoxyribonucleic acid. Biochem. J. 62:315—323. 8. Greenberg. M. E., and E. B. 2111'. 1984. Stimulation of 3T3 cells induces transcription of the c-fos prom-oncogene. Nature (Lon- don) 311:433-438. 9. Hanahan. D., and M. Maelson. 1980. Plasmid screening at high colony density. Gene 10:63-67. llatanalta. M.. and R. Dulbecco. 19m. Induction of DNA synthesis by SV40. Biochemistry 56:736—740. 11. Ives. D., J. Durham. and V. Tucker. 1969. Rapid determination of nucleoside kinase and nucleotidase activities with tritium-la- beled substrates. Anal. Biochem. 28:192-205. 12. Jellnelt. W.. T. Toomey. L. Leinland. C. Duncan. P. Biro. P. Choudany. S. Weksman. C. Rubin. C. Houck. P. Deininger. and C. Schmidt. 1980. Ubiquitous. interspersed repeated sequences in mammalian genomes. Proc. Natl. Acad. Sci. U.S.A. 77:1398- 1402. 13. Johnson. L., L. G. Rao. and A. Muench. 1982. Regulation of thymidine kinase enzyme levels in serum-stimulated mouse 3T6 fibroblasts. Exp. Cell Res. 138:79-85. 14. Kaufman. R. J., and P. A. Sharp. 1983. Growth-dependent expression of dihydrofolate reductase mRNA from modular cDNA genes. Mol. Cell. Biol. 3:1598-1608. 15. Kelly. K., B. H. Cochran. C. D. Stiles, and P. Feder. 1983. Cell-specific regulation of the c-myr' gene by lymphocyte mito- gens and platelet-derived growth factor. Cell 35:603-610. 16. Kit, 8. 1968. Viral-induced enzymes and the problem of viral oncogenesis. Adv. Cancer Res. 11:73—221. 17. Kit, 8.. D. R. Dubbs. P. M. Frearson. and J. Melnick. 1966. ll. 184 VOL. 5. 1985 .— IO N .° N ... . Lin. P. I".. M. Yam-I1 Enzyme induction in SV40-infected green monkey kidney cul- tures. Virology 9:369—8. Lsemm U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 1: . Lsu u. .sndY ..Ksn 983 Direct isolation of the functional]: human thyminine kinase gene witha asco smid shuttle vector. Proc.N tl.A A.cad Sci U..S.A 81:44-41 R. E. Ke Items. 1981. Control of dihydrofolate reductase messenger nbonucleic acid producti oIII. MoL Ce IL Biol. 1:961- 971. umi. P. D. Murphy. A. Egg, and F. H. Ruddle. 1982. Partial purification and characterization of the mRNA for human thymidine kinase and hypoxanthine/guanine phosphoribosyl transferase. Proc. Natl. Acad. Sci. U S A 79 290-4294. 22?LIII,P. F..S. Y. Zhao. sIIdF. H. Ruddle. 1983. Genomic cloning I: N a N O . Lowry.0 and preliminary characterization of the human thymidine kinase gene. Proc. NatL Acad. Sci .1). S. A. 80:6528-653 .11.. N. J. Rmeh mu R.J. Will. 1951. Protein measurements With the Folin phenol reagent. J. Biol. Chem. [93:265-275. . Lusky. M. .srt. sndM. Botc hen. 1981. Inhibition of SV40 replication in simian cells by specific pBR322 DNA sequences. Nat ture (London) 293:7 .E. F. FrltscthsI Sumhroolt. 1982. Molecular . J. cloning. a laboratory manual. Cold Spring Harbor Press. Cold SpringH a.rbor N. Y. . Merrill. G. P., R. M. Harland. M. Groudine. and S. L. Me- lNDUCTION OF CELLULAR TK 1497 984.0enetic and physical analysis of the chicken TK Knish!- gene. Mol. Cell. Biol. 4:61769-177. 27. Merrill .0. F. 8 Its N on I.» O u ... u .N u u 53.: . ...H P..Berg toull- -Iength cDNA. uniMoi. Cell. Biol. 2: 161—1 29 Pe . Pastel. E. H..und . Stubble usehlss. and s. L. McKnIght. 1984. TK enzyme expression in differentiating muscle cells is regulated through an internal segment of the cellular TK gene. Mol. Cell. Biol. 4:1777-1784 1.982 High 1e119iciency cloning of and9 M. . pstch. ngler .1980. lsolatiorl of the chicken thymidine0 kinase gene by plasmid rescue. Nature (London) 2.85: 207— A. J. Levine. 1976. The reqmrement of sImIan virus 40 gene A product for the stimulation of cellular thymidine kinase activity after viral infection. Virology 73:206-215. . J. deWet. and l. Schaffner. 1981. Cell cycledependent regulation of thymidine kinase activity introduced into LMTK‘ cells by DNA and chromatin-mediated gene transfer. Proc.N atl. Acad. Sci. U. S..A 78. 1119-1 12 3. Sun The bidirectional transferG of DNA and RNA to nitrocellulose or diaz nz-yl oxymethyl paper. Anal. Biochem. I C. A., C. IIeIIer. .:hT ymidine kinase activ- ity in synchronized Hela cell cultures. Biochem. Biophys. Res. Commun :5 5 To .W. C. 199.81 Normal rat cell lines deficient in nuclear thymidine kinase. Virology 113: 408-411 nger .5. Silversteln. and R. Axel. 1978. Biochemical transfer of single copy eucaryotic genes using total cellular DNA as donor. Cell 14:725—731 . M., A. APPENDIX I I ‘— 185 APPENDIX II TK Transcriptional Regulation in SV40 Infected Cells The final transcription experiment in Chapter 2 involved a nuclear run on analysis of RNA made just before and during DNA synthesis in SV40 infected CV-l cells. Just prior to the time that this experiment was done, we had subcloned our human TK cDNA clone from an Sp64 vector (Pharmacia) into a T7/T3 vector (Bethesda Research Laboratories) for the purpose of saving money on the cost of the RNA polymerase required. The new construct transcribed well, and the experiment was performed with this construct and the previously used pzmicroglobulin gene still in an Sp64 vector. Following publication of the chapter 2 manuscript, I was informed by Dr. David Shalloway at the Pennsylvania State University that a sequence had been found between the T7 promoter and its polylinker which specifically binds to 28s Ribosomal RNA. This sequence has not been found in Sp6 vectors, so I therefore have repeated the last experiment from Chapter 2 using my original constructs in the Sp6 vector. The experiment was performed with both high passage and low passage CV-l cells, yet the 186 results were quite reproducible. Figure 1 shows an autoradiogram of the experiment from high passage cells, and figure 2 shows a graphic representation of both experiments. It can be plainly seen that there is a 3- 3.5 fold transcriptional increase in SV40 infected cells by 14-16 hours post infection. g 187 Figure 1 Nuclear transription assays performed at one hour intervals prior to and during the onset of DNA synthesis in SV40 infected CV-l cells. Nuclei were prepared at the times indicated after viral infection and used for transcription assays as described in Chapter 2. 2.0 x 106 cpm of labelled RNA was added to each hybridization. This autoradiogram shows the results of hybridization to nitrocellulose filters containing cRNAs and cDNAs of interest. FIGURE 1 13 14 15 16 17 12 O .0 .00 ... .u .. e. l 189 Figure 2 Graphic analysis of the results in figure 1, as well as the results of the repeat experiment. Plotted are values of TK/fl-z microglobulin (as described in Chapter 2) 190 V p O 0.. . . . 5. 4 3 2 SSC mmqmmoz. ézoEEomqu. .12.? 4mm 20 IS IO HOURS POST INFECTION FIGURE 2 i ummmm\mmmmlmlmurmuwmmmmm