1": 111 31131 1115111111 1111111‘311"1~.11"‘%111"1111’ 1-1 1 . I 11111111111 11 11111113 11111 '.‘”1 .1111111“~11..1 "‘L 3131333133131“ 31' 331,133 3393111, 1' '11' '“131333333 3333313 113.31 "11 1-111 1311 '33 1'133333 111311 31333 11.1311... 1111 11111111 1311 111 111 1"""'133133'3""' 1 ' "Mi" 333L [31'31131 111333 111.1 1'" .1 ‘ < IU '13' 1 '1‘ ‘1 1' ":1 11111151811111; 1"11‘3 1'1 .113 1111 '11::1,11.1111111313133313.11 '11" 11111111"1.1111111111111111E111;1111111 "" "1" ‘1""11 1 "11111 - "1111.11""9‘11111111111111211121111, 1'11 '1 3.1111 1‘11 ' 1111111111111111. "1111111111. 1“ 111111 111 111.11. 1' l l H i '111 (1131'13333111333313333 v. 1'1 3.3133? .9 33133 ' 111"'1'11""31'13! "t 313L111 1:351 ' ' "'th'” ""§"':'u' ' " i'E'L' 1"11'N. .311 31133.“ 3311:11333333 3 "'1 :3'11111'1 1119551133"11'0'§"fr'34 '.'.2' . 1 3333 1'91" 333333333 . ."' -11 3 .313: 3 .333 £33 :33333: 3 .z 1". .. '21 1:3; : 1’ 3" Liking? ‘1'. ' "V ' 1.21. ,- .13 1 2' .3 3.11111313i1-3131 3333331 33.331333333333333 3.33 3 3 33, 3333333333 ' "" I"""1‘ ' '3-3g3 ' 311:2 --'. 3111'11’311-‘1‘5 ' 3 .... 3. 11 . 3 #311115. . 1 '11'1' ' 1131331! 1'11. 111 "1113713“3331'1'111‘ 3" 3" 331'31 I' 3 2111133 "'33 :‘1'.' 1‘111 . ' 3 .. 3. 33.31.333'3‘ 3 3 3;. 3313333. 333 33333 i \ 33333‘3v3 3'33. I 3:3 3 I: ‘, 33333333313333: . 3-3 01 " 1‘1- 3 ‘ 6:: 338111: 1" '1 1'11"" '3111313 .13 131 1 ‘1' 13 331333333 1' 3'13'3‘3'. '313 "1'1 3'13 1 V11 5 "3"" '31:; 1.1.311'3 ,1; 11111 3.1. 1 .3 31 ’ - {1:31:7‘ 1311311131313 1313331333 3 33 ‘3."‘13131.313131’11 "'1'1'1'1. 11111 111111.11 "'" "~"“'1'111'1':"'1'1'E 1" "111 1111I 1131131 "'"'1"' "1'1 ' L11'.'."3.'11‘.1,"'1'1'1"3»'513".".' -" " "1'11" 31.311'11"'11'."‘.333 331.333.3311. 1111 '.'-1'" ~. 3 313 3333333 3 3 "'1 .1 "'"""""' "" 33133 33333 133' . . A .'1 13:}33l3'3 11311113131. 1, :11'131‘1-231333'31‘13'3313-3 13 1.1.13 1313131333 ‘3". 11.1"1'1'113 '1'1' '11"'1' 11"'3 "' '1'1'1'1"11"" 'B ‘13 111"" 333 1 11" "1'13 111133 131'- 1' 11"1 11111111111111 '.L' 111111'111IL'.'1""'" IHESIS ‘ " “- g" \‘i. W4 ' 1-". "I“: -.§a. &l% i Pf??g.§",u, . 1‘4 * .i I.‘ t“- M. 9 '.‘ .‘ I. _ w». A “aw v‘ This is to certify that the thesis entitled STUDIES ON MESSENGER RNA METHYLATION presented by Sally Ann Camper has been accepted towards fulfillment of the requirements for Ph. D. degree in Biochem stry @3131: M. @8de ajor professor Date 7/22/83 0-7 639 MS U is an Affirmative Action/Equal Opportunity Institution PV1531_J RETURNING MATERIALS: . Place in book drop to LJBRARJES remove this checkout from .—,— your record. FINES will be charged if book is returned after the date stamped below. (L14 a 32. S” 7 STUDIES ON MESSENGER RNA METHYLATION By Sally Ann Camper A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Biochemistry 1983 ABSTRACT STUDIES ON MESSENGER RNA METHYLATION By Sally Ann Camper The major goal of my research was to understand the role of methylation in mRNA metabolism by focusing on the characterization of specific gene products. Initial characterization of two abundant bo- vine pituitary mRNAs was approached through analysis of their respec- tive cDNA clones. A cDNA clone for prolactin mRNA was shown to be nearly full length by R-loop mapping. A growth hormone cDNA clone was subcloned into M13 and sequenced by the dideoxy chain terminator method. Determination of gene structure and sequence often provides information important for studying the mRNA products of the genes. A partial genomic clone for prolactin was obtained by screening a bovine genomic library, and portions of the prolactin gene were restriction mapped and subcloned. The primary sequence of the 5‘ flanking region, 5' untranslated region and a portion of the first intervening sequence was determined. Comparison of the bovine prolactin S'fflanking region with a published sequence for rat prolactin revealed several regions of potential regulatory significance. Another phase of my studies involved the use of the methylation inhibitor S-tubercidinylhomocysteine to study the role of methylation in mRNA metabolism. S-tubercidinylhomocysteine was used to perturb mRNA nethylation in primary pituitary cultures. Radioactive labeling studies indicated that cytoplasmic mRNA produced in the presence of S- tubercidinylhomocysteine was undermethylated at the 2'-Q-methyl posi- Sally Ann Camper tion in the cap and deficient in N6-methyladenosine. S-tubercidinyl- homocysteine was also found to be extremely potent in HeLa cells for inhibition of N6-methyladenosine methylation internally and at the cap site. The function of methylation in processing was probed by measur- ing the rate of cytoplasmic appearance of HeLa mRNA in the presence and absence of S-tubercidinylhomocysteine. The time of cytoplasmic appearance of undermethylated mRNA appeared to be delayed compared to normal mRNA. To test whether methylation affects mRNA stability, turnover of HeLa mRNA was measured. The half-life of the mRNA popula- tion was unaltered in cells treated with the inhibitor, compared to control cells. These studies on methylation of total mRNA have pro- vided a framework for future examination of specific mRNAs in HeLa cells. In memory of Carolyn R. Camper ii ACKNOWLEDGEMENTS The following people deserve special thanks for helpful dis- cussions, encouragement, assisting with specific experiments, and providing reagents: Dr. Robert Albers, Mr. David Ayers, Dr. James Coward, Ms. Karen Friderici, Dr. Ray Goodwin, Dr. Sarah Horowitz, Dr. Hsing-Jien Kung, Mr. Jeff Lee, Dr. Dennis Luck, Mr. Robert Lyons, Ms. Pat Maroney, Dr. Timothy Nilsen, Dr. John Nilson, Dr. Fritz Rottman, Dr. David Walker, Mr. Richard Noychik, and Ms. Yvonne Yao. In addition, I would like to mention Mr. Cairns, Dr. Burmeister and Dr. White, who particularly stimulated my interest in science. Most importantly, I want to acknowledge my family and friends for their support and encouragement. TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . vi LIST OF FIGURES . . . . . . . . . . vii LIST OF ABBREVIATIONS . . . . . . . . viii BACKGROUND Function of post-transcriptional modifications of RNA molecules . . . . . . . . 1 I. Transfer RNA . . . . . . . . 1 II. Ribosomal RNA . . . . . . . 6 III. Small nuclear RNA . . . . . . 9 IV. Messenger RNA . . . . . . . 10 References . . . . . . . . 20 PART ONE Partial characterization of the bovine prolactin gene . . . . . . . . . 25 Abstract . . . . . . . . . 26 Introduction . . . . . . . . 26 Methods . . . . . . . . . 28 Results . . . . . . . . . 30 Discussion . . . . . . . . 48 References . . . . . . . . 55 iv PART TWO Sequence of bovine growth hormone mRNA . . Abstract . . Introduction . Methods . . Results . . Discussion . References PART THREE Effect of STH on pituitary mRNA . . Abstract . . Introduction . Methods . . Results . . Discussion . References PART FOUR Effect of STH on HeLa cell mRNA half-life and cytoplasmic appearance Abstract . Introduction . Methods . . Results . Discussion . References . 58 59 59 6O 61 65 68 69 7O 70 71 76 9O 96 97 98 99 104 105 114 121 LIST OF TABLES Page Background . . . . . . . . . . . . 1 1. Distribution of methylated nucleosides in RNA . 2 Part One . . . . . . . . . . . . . 25 1. Genomic blots . . . . . . . . . 36 2. Restriction digests and southern blots of APro6 . 38 3. Mapping exons in APro6 . . . . . . . 43 Part Two . . . . . . . . . . . . . 58 1. Summary of sequence differences in bovine growth hormone mRNA . . . . . . . . . . 66 Part Three . . . . . . . . . . . . 69 1. Effect of STH on total and cytoplasmic RNA synthesis and methylation . . . . . . 77 2. Distribution of methyl label in poly(A)+ RNA . . 78 3. Effect of STH on methylation in an extended labeling time . . . . . . . . . 80 4. Distribution of methyl label in poly(A)+ RNA . . 82 5. Analysis of distribution of methylation by HPLC . 83 6. Molar distribution of caps . . . . . . 85 7. Incorporation of radioactive amino acid into protein . . . . . . . . . . 87 8. Rate of accumulation of 3H-uridine into total RNA 89 Part Four 0 O O O O O O O O O O 0 97 1. Percent inhibition of mRNA methylation by STH . . 110 vi LIST OF FIGURES Background . . . . . . . Part Part Part 1. tRNA and rRNA processing pathways one 0 O I O O O I O O R—loop of pBPRL72 with polysomal poly(A)+ RNA .Genomic titration . . . . APro6 restriction map . . . Restriction map of pBPRL72 . . Partial sequence of the bovine prolactin gene and comparison with rat prolactin Two 0 O O O O O O O O 1. 2. Nucleotide sequence and predicted amino acid Sequencing strategy . . . . sequence of bovine growth hormone mRNA Four . . . . . . . . 1. 2. Structure of STH and site of action STH dose curve for inhibition of poly(A)+ RNA methylation . . . . . . Method for separation of modified and unmodified nucleosides, and cap structures . Turnover of poly(A)+ RNA . . . Time of cytoplasmic appearance of RNA vii Page 25 32 34 4o 42 47 58 63 64 97 103 106 108 113 116 LIST OF ABBREVIATIONS S-tubercidinylhomocysteine N5-methyladenosine high performance liquid chromatography transfer RNA messenger RNA ribosomal RNA small nuclear RNA 5-methyluridine (ribothymidine) I-methyladenosine 2-methylguanosine 2 ' -Q-met hyl guanosi ne 2'-Qfmethyladenosine 2' -9_-methyl uri di ne 2'-Qfmethylcytidine 2‘1Q-methylpseudouridine N2, NZ-dimethylguanosine 1-methylguanosine 7-methylguanosine 3-methyl cytosine N6, N6-dimethyladenosine 3-methyluridine 5-methylcytosine Newcastle disease virus Vesicular stomatitis virus viii SAH SAM ASV TRH cDNA EDTA kb bp 305 TCA S-adenosylhomocysteine S-adenosylmethionine Avian sarcoma virus Thyrotropin releasing hormone complementary DNA Ethylenediamine tetraacetic acid kilobase base pairs Sodium dodecyl sulphate Trichloroacetic acid ix BACKGROUND FUNCTION OF POST TRANSCRIPTIONAL MODIFICATIONS OF RNA MOLECULES Eukaryotic ribosomal, transfer, messenger and small nuclear RNAs all contain post-transcriptional modifications, the nature of which has been the subject of study for many years. Much still remains to be defined concerning the time sequence of these events and their role in the function and metabolism of specific RNA molecules. I. Transfer RNA Transfer RNAs are the most highly modified RNAs. The primary transcripts undergo 5' and 3' terminal trimming, and 3' terminal CCA addition. Numerous base and sugar modifications take place and some tRNAs undergo splicing to remove an intervening sequence. The methyl modifications include eight different base modifications and, in lesser abundance, 2'-_Q-methylation of ribose (Table 1). Overall there are approximately seven methylations per molecule. These post-trans- criptional methylations have been proposed to be important for more efficient and accurate recognition of the tRNA. For instance, methyl- ation could affect aminoacyl tRNA synthetase interactions, and wobble recognition might be amplified or restricted by modifications in the anticodon. Modifications may involve tRNA interaction with ribosomes including A and P site binding, translocation and termination (Kersten, 1982). In support of this idea it is interesting to note that two modifications, which arise from transglycosylation rather Table 1 Distribution 91 methylated nucleosides 193M 5355 tRNA 185 285 m5C 17 Am 27 Am 30 sz 14 um 25 Gm 25 m5u 13 Gm 17 Cm 20 m1A 12 Cm 17 Um 15 Gm 9 m62A 5 m1A 1.5 mzzG 8 m6A 2.5 m5A 1.5 mlG 8 m7G 2.5 m5C 1.5 m7G 7 Vm 2.5 m3u 1.5 Um 4 Om 1.5 m3C 3 Am 3 Cm 2 The percentage of the total (3H)-methionine label that is found in each modified base or 2'-Q—methylated ribose is tabulated above for mammalian tRNA and 18 and 285 rRNA (Munns and Liszewski, 1980). than methylation (Q base and Y base), alter the efficiency with which charged tRNAs are utilized. Hypomodification of Y base results in increased usage whereas the opposite is true for Q base. It has been proposed that preferential utilization of hypomodified tRNAs could facilitate synthesis of specific proteins in tumor cells (Smith et al., 1983). Microinjection of a gene for tRNAtyr into frog oocytes has facilitated analysis of the order and intracellular location of tRNA post-transcriptional events (Melton et al., 1980). Most of the base and sugar modifications occur in the nucleus and take place in a specific order correlating with other post-transcriptional processing steps such as the 5', 3' terminal trimming, 3' CCA addition and splicing (Figure 1). Since methylation events appear to take place in an obligatory order, the possibility exists that they facilitate recognition by enzymes responsible for other processing reactions. Yeast mutants have been particularly useful in sorting out the function of some specific tRNA modifications and their relationship to other tRNA post-transcriptional processing events. m5U, a modifica- tion that occurs prior to intervening sequence removal, appears not to be required for tRNA splicing or transport. In addition, these mutants appear to have no obvious physiological defect other than the absence of m5U (Hopper et al., 1982). Tissues undergoing rapid development have altered activity of tRNA methyltransferases and changes in tRNA methylation have been observed in malignant tumors (Kersten, 1982). The significance of these changes is not yet clear. A mammary adenocarcinoma deficient in Figure 1 tRNA and rRNA processing pathways The processing pathway for (A) tRNAtyr microinjected into frog oocytes is shown (Melton et al., 1981). The length of each transcript is given in nucleotides and base modifications are indicated as follows 5-methylcytosine (O), pseudouridine (V), mlA (v), and dihydro- uridine (0). The major rRNA pathways in (B) HeLa cells and (C) L-cells are shown with the sedimentation value for each intermediate (Lewin, 1980). mmfi D mmN ”WU mNn HTUI mwnnHoU.IUI.»I_UmO. .IIIIIIUIQIIU m3 * nHHUl-Ulllnul wn¢ .0. H ouzuwk mad 0 mom H“ a H mm. mNnnHUInTa InUmON “HUIUIIIU mew a “HIDIIIIHUIIII mnv .m <0 . 63.823 Ea 2 .335... mm a hm .mlfiHUi a V0. Amino. a 00. ”Inning. .n .n .< I—methyladenosine (mlA) methyltransferase contains significant amounts of tRNA deficient in mlA. The molecular weight of mlA deficient tRNA is consistent with it being an unspliced precursor but it isrun:a substrate for tRNA processing enzymes. The mature cytoplasmic tRNA is somewhat deficient in mlA also (Salas and Leboy, 1983). These obser- vations imply that this particular base methylation may be important for efficient 5', 3' terminal processing or splicing of nuclear tRNA . precursors, but not absolutely required. II. Ribosomal RNA Ribosomal RNA is transcribed as a 455 precursor containing 185, 5.85 and 285 rRNA sequences separated by spacer regions as well as a 5' leader sequence. The processing pathway seems to be slightly different in different cells but generally to occur in an ordered manner (Lewin, 1980). The pathway for rRNA maturation in HeLa cells and L cells is illustrated in Figure 1. Ribosomal RNA is highly methylated, containing ribose and base methylations with the ribose methylations being the most abundant (Table 1). The 285 molecule contains 74 methylations per molecule and the 185 approximately 43 (Maden and Salim, 1974). The ribose methylations occur rapidly on nascent or newly synthesized transcripts, but the base methylations occur later in the maturation (Liau and Hurlbert, 1975). Although the methyl groups have been mapped to specific locations, they all do not occur within the same primary sequence (Maden and Salim, 1974), indi- cating either an incredible number of different methyltransferase enzymes or a type of recognition that does not involve primary sequence. The function of rRNA methylation is still open to debate. 2"{}- methylaticnihas been proposed to protect the rRNA from degradation because it inhibits cleavage by enzymes requiring cyclic phospho- diester intermediates. Base methylation may be important for tertiary structure necessary for ribosome function; this has been suggested for m6 2A located in a sequence near the 3'end of 185 rRNA (Gourse and Gerbi, 1980). Since it is clear that none of the spacer regions are methylated - all the methylation occuring on the 455 precursor are conserved in mature RNA molecules - it is possible that methylation is important for prOper processing. Experimental proof of any of these theories is fragmentary thus far. In experiments where HeLa cells were subjected to methionine starvation, 45$ rRNA was synthesized but was approximately 80% defi- cient in methylation. The undermethylated rRNA could be cleaved to produce 325, but no further, and the smaller cleavage product appar- ently was not stable (Vaughan et al., 1967). Similar results were obtained when ethionine, an inhibitor of methylation but not protein synthesis, was used in HeLa cells (Wolf and Schlessinger, 1977). These studies imply that methylation, although not essential for cleavageeof 45$ rRNA is important for cleavage of 325 rRNA and for stability of the RNA since the undermethylated species do 225 accum- ulate. However it is possible that perturbation of other aspects of metabolism by methionine starvation or treatment with ethionine are responsible for some of these effects on rRNA. For example, if a rapidly turning over protein were required for rRNA maturation, inhi- bition of protein synthesis by methionine starvation could block processing. Also, a protein with ethionine incorporated instead of methionine could exhibit impaired function resulting in the apparent processing block. In this regard, a study involving a BHK cell line temperature sensitive for processing rRNA from 323 to 28$ becomes particularly interesting. The rate of appearance of 28S rRNA in the cytoplasm is reduced 95% at the nonpermissive temperature and a 36S rRNA species is found, representing a shift to a less predominant pathway (Toniolo and Basilico, 1976). Because the methylation of 328 rRNA is unaltered at the nonpermissive temperature, it seems unlikely that the lesion is due to a methylation defect. The synthesis of small nuclear RNAs, which may be important in rRNA processing, also does not appear to be impaired. The defect appears to be in the activity of enzymes required for the synthesis of polyamines - ornithine decarboxylase and S-adenosylmethionine decarboxylase. These enzymes are subject to temperature sensitive synthesis or accelerated degradation (Levin and Clark, 1979). While this study does not rule out an important function for methylation in rRNA processing, it establishes the importance of polyamines in rRNA processing and under- lines the dangers of utilizing agents to study RNA metabolism which also inhibit protein synthesis. The methylation of rRNA in these ts BHK' cell mutants appears to be normal at the elevated temperature (Levin and Clark, 1979). Thus, 32$ rRNA whose processing is blocked appears not to have antaccel- erated turnover in the case where it is methylated (Toniolo et al., 1973), but is wasted in a case where it is undermethylated (Wolf and Schlessinger, 1977). These observations correlate with a role for methylation in nuclear stability of rRNA. A role for methylation in nuclear events in rRNA metabolism, such as processing or stability does not rule out an additional role for methylation in rRNA function in the cytoplasm. Chick embryo fibro- blasts treated with cycloleucine, an inhibitor of S-adenosylmethionine synthetase (Lombardini et al., 1973), have undermethylated rRNA in the cytoplasm which does not accumulate in ribosomes at the normal rate (Dimock and Stoltzfus, 1979). The nature of the undermethylation was not carefully examined, but did include a reduction in 2'-_Q-methyl- adenosine. Perhaps methylation of rRNA is necessary for production of functional ribosomes,ialthough a direct inhibition of protein syn- thesis by cycloleucine can not be ruled out. III. Small nuclear RNA Small nuclear RNA (snRNA) is a class of RNA molecules found in the nucleus of eukaryotic cells, ranging in size from about 75 to 200 nucleotides in length, which is generally very abundant (104—106 copies per cell) and stable. The genes for snRNAs are apparently transcribed by RNA polymerase II which is also responsible for mRNA transcription, but they do not contain the “Hogness box“ usually associated with genes coding for mRNAs and are not interrupted by intervening sequences (Roop et al., 1981). UV inactivation studies indicate that the transcription unit may be 5 kb (Elicieri, 1979), leaving open the possibility that snRNAs are cleavage products of larger RNA precursors. The transcription start site for U1 snRNA has been mapped to approximately 183 nucleotides upstream from the 5' end of the mature snRNA (Murphy et al., 1982). Therefore, the primary 10 transcripts must undergo a processing step to trim the 5‘ terminal extension. Most snRNAs contain a methylated cap structure where 2,2,7- trimethylguanosine is coupled in a 554? phOSphodiester linkage at the first nucleotide of the chain. Small nuclear RNAs also contain all four 2'-_(_l-methylated nucleosides positioned in the two nucleosides adjacent to the cap (m2’2’73Gppmemep...), similar to mRNA and inter- nally, similar to rRNA and tRNA. Other post-transcriptional modifica- tions include the base methylations m6A and sz (Choi et al., 1982). The function of snRNAs is still controversial. One snRNA (U1) has been postulated to be involved in alignment of intervening sequences for Cleavage of mRNA precursors (Murray and Holiday, 1979), possibly functioning as a coenzyme like the RNA factor'in the tRNA processing enzyme RNAse P (Stark et al., 1978). It has been suggested that methylation could be a way of distinguishing between RNA strands for cleavage (Murray and Holiday, 1979). The existence of a snRNA identical to U1 but lacking the cap and first six nucleotides (Lerner et al., 1980) casts doubt on that theory. Another snRNA (U3) may be involved in rRNA processing since it has been found in the nucleolus bound to 328 and 28S pre rRNA (Choi et al., 1982). Definitive evi- dence for involvement of these small nuclear RNA molecules in RNA processing is still not available. The function of post-transcrip- tional modifications of snRNAs is even more obscure. IV. Messenger RNA Messenger RNA, the least methylated of any of the RNA groups, is subject to several post-transcriptional processing events. Many mRNAs 11 undergo cleavage and splicing to remove intervening sequences as well as 3' terminal polyadenylation. The 5' end contains a cap - a 7- methylguanosine in a 554? phosphodiester linkage with the first nucleotide of the mRNA. 2'-Q-methylation occurs only on the nucleo- tides adjacent to the cap hag. m7GpppNKm)pN”(m)pruJ. Inter- nally, the predominant methylation is m6A, but in BHK cells m5C can be detected (Dubin and Taylor, 1975). Knowledge of the relationship of methylation to other post-transcriptional processing events is crucial to understanding the regulation of gene expression, since splicing and polyadenylation have each been shown to be involved in regulating the level of expression of various genes (Nevins and Wilson, 1981; Darnell, 1982). mRNA of higher eukaryotes is more methylated than that of lower eukaryotes. Cap structures that do not Contain 2'-Q-methylated ribose moieties (cap zero mRNA, m7GppprNpruJ are the predominant cap form in yeast and dictostelium, comprise a small fraction of caps in droso- phila and are normally not found at all in animal cells (Lewin, 1980). Internal m5A has not been detected in slime mold, yeast, and some plant mRNAs (Banerjee, 1980). The increase in methyl content of mRNA with an increase in complexity of the organism may argue for a regula- tory role for mRNA methylation. The order of post-transcriptional processing events for mRNA tran- scripts in the nucleus is beginning to be resolved. Capping is appar- ently a very early event in processing, occuring while the transcripts are still nascent chains (Babich et al., 1980). 2'-Q-methylation of the N' position of the cap and m6A addition are also nuclear events, 12 but 2'-_Q-methylation of the N" position is catalyzed by a different cytoplasmic enzyme after incorporation of the mRNA into polysomes (Perry, 1981). The enzymology of the guanyltransferase and methyl- transferases has been reviewed recently (Banerjee, 1980). Some evidence has suggested that initiation of transcription could be dependent on addition of the m7G cap structure (Jove and Manley, 1982, Chen-Kiang et al., 1982), but recent studies using an in 31532 transcription system have shown that transcripts can be initiated and elongated to at least 20 nucleotides without the occur— ence of capping or 2'-_O_-methylation (Coppola et al., 1983). While transcription is apparently separable from capping and methylation, RNA polymerase II may exist in a complex with capping and methylation enzymes and thereby be stimulated in an allosteric fashion by SAM and its analogs (Furuichi, 1978; Wertheimer et al., 1980). Polyadenylation probably occurs before splicing for a variety of HnRNA transcripts because examination of northern blots containing nuclear RNA that had been fractionated on oligo(dT)-cellulose reveals a large proportion of precursors in the poly(A) containing fraction (Schibler et al., 1978; Lai et al., 1978; Tilghman et al., 1978). Polyadenylation does not appear to be required for splicing, however (Zeevi et al., 1981). Evidence has begun to accumulate from jg_yjtrg_ transcription studies that implies that transcription proceeds beyond the polyadenylation site (Darnell, 1982). Perhaps poly(A) addition occurs following an endonucleolytic cleavage event which may not always be precise (Sasavage et al., 1982; Ahmed et al., 1982). In cases where mRNA undergoes differential splicing, selection of a 13 poly(A) site may then restrict the mRNA to a specific splicing pathway (Amara et al., 1982; Early et al., 1980). Transport to the cytoplasm apparently does not require polyadenylation, but non-adenylated RNA appears to be less stable (Darnell, 1982). It is not known whether methylation is coupled in any way to polyadenylation. Transription of vesicular stomatitis virus (VSV) and Newcastle disease virus (NDV) 1.911112 results in large 3' terminal poly(A) (Rose et al., 1977; Weiss and Bratt, 1974). In VSV, large poly(A) was generated by in _v_i_tr_o transcription in the presence of S- adenosylhomocysteine. Cap methylation was also inhibited, suggesting a possible link between polyadenylation, ribose 2'-9_-methylation and/or m7G methylation. mRNA from Novikoff hepatoma cells and bovine pituitary cells deficient in ribose 2'-Q-methylation did not exhibit an increased size when examined on sucrose gradients (Kaehler, 1973; S. Camper, unpublished results) indicating that the link between methylation and polyadenylation may be related to 5' m7G methylation or may be peculiar to some viral systems. Since the discovery that non-viral genes were interrupted by intervening sequences (Jeffreys and Flavell, 1977), it has been of interest to determine whether internal methylation could play a role in intervening sequence removal. Perhaps mRNA methylation works like a restriction-modification system in bacteria to designate regions to be cut or protected. Examination of intron-exon splice junctions has revealed a consensus sequence (Lerner et al., 1980) but homology with this sequence does not appear to be adequate for splicing. N6-methylation of adenosine results in a destabilization of A-U 14 base pairs (Engel and von Hippel, 1974, 1978) although they can still be formed. Thus, internal m6A may be important in establishing the appropriate secondary structure for recognition by splicing enzymes. Precedence for an RNA processing enzyme being able to recognize secon- dary structure comes from the bacterial rRNA processing enzyme RNAse III (Robertson, 1982). As mentioned previously, the role of methylation in mRNA splicing is still open to debate. Unlike the internal 2'-Q-methylnucleosides of rRNA, internal m6A methylation of mRNA does occur within a consen- sus sequence. Approximately 75% of the m6A is found as G-mGA-C and the remainder as A-m6A-C in a diverse group of organisms including maize and HeLa cells (Nichols and Welder, 1981; Wei and Moss, 1977), indicating a conservation of sequence recognition. Many laboratories attemptedixidetermine whether internal methylation was conserved during the maturation of mRNA (Lavi and Shatkin, 1975; Sommer et al., 1978; Chen-Kiang et al., 1979) in an analogous fashion to the conser- vation of internal methylation in ribosomal RNA maturation. The results were ambiguous, but recently it was clearly shown, for SV4O late RNA, that methylation occurs both within and outside~of exons (Aloni et al., 1979). This result, however, does not preclude m6A from involvement in recognition for splicing. In fact, some bacterial restriction enzymes recognize sequences distant from their cleavage sites (Roberts, 1980). Internal methylation may serve a different function for different mRNAs. Viruses have many examples of differential processing and in some cases production of viral RNA requires that some RNA transcripts 15 remain unspliced. The location of m6A, although not determined precisely, has been shown to roughly correlate with the location of splice junctions of some viral transcripts (Beemon and Keith, 1977; Canaani et al., 1979). Viral proteins may regulate the extent of splicing primary viral transcripts (Leis et al., 1980). There are cases of differential processing of non-viral transcripts including calcitonin (Amara et al., 1982), growth hormone (DeNoto et al., 1981), prolactin (Taylor et al., 1981), and immunoglobulin (Early et al., 1980), but it is not yet known how commonly this occurs or whether methylation is involved in the selection of the Splice site. It has been clearly demonstrated that capping and m7G methyl ation are important features for efficient translation of mRNA on eukaryotic ribosomes (Both et al., 1975; Filipowicz, 1978; Kozak, 1978). Many viral systems have developed elaborate mechanisms to take advantage of the preference for translation of capped mRNA (Revel and Groner, 1978; Banerjee, 1980). 2'-9_-methylation probably enhances binding of mRNA to ribosomes but its influence is less dramatic than m7G methylation (Mithukrishnan, 1978). Capping seems to be required for mRNA stability (Furuichi et al., 1977; Green et al., 1983). Perhaps other methylations such as cytoplasmic 2'-9_-methylation and internal m5A methyl ation play a role in cytoplasmic stability. There are several examples of situations where a specific mRNA appears to be preferentially stabilized or destabilized (Guyette et al., 1979; Heintz et al., 1983; Stiles et al., 1976; Bastos et al., 1977; Chung et al., 1981; Greenberg, 1972; Wiskocil, 1980). The mechanism by which this is accomplished remains 16 a mystery. Viral and host mRNA late in Herpes virus infection do not contain m5A and late mRNAs have a shorter half-life (Bartkoski and Roizman, 1978). Alteration of methylation could be the cause of the different half-lives, but this hypothesis must be proven. There is no good evidence for an effect of internal methylation on translation. Heavily alkylated brome mosaic virus RNAs apparently are not pre- maturely terminated when translated _i_p_v_it_r_~g (Fraenkel—Conrat and Singer, 1980). SV40 RNA is known to be hypermethylated in interferon treated cells and is not translated in cells under these conditions. However, the nechanism is unlikely to be via overmethylation of the SV40 messages because ifl_11trg translation of the mRNA from interferon- treated SV40 infected cells did not appear different from control SV40 infected cells (Kahana et al., 1981). Although a func- tion for internal m5A in translation is not ruled out, it seems unlikely. One approach to dissect the importance of methylation in RNA metabolism is to perturb nethylation using an inhibitor and observe the consequences for mRNA metabolism. Several groups have utilized cycloleucine in this manner. Cycloleucine inhibits S-adenosylmethio- nine (SAM) synthesis (Lombardini et al., 1973) and should therefore be a general inhibitor of SAM dependent reactions. In rapidly growing cells SAM is partitioned 2:1 in favor of decarboxylation to produce polyamines rather than as a "ethyl donor fbr transmethylation reac- tions (Hoffman and Clark, 1983L Thus, reagents that deplete SAM necessarily impair polyamine biosynthesis. Polyamines have been shown to be important for rRNA synthesis (Levin and Clark, 1979), therefore 17 experiments involving the use of cycloleucine to inhibit methylation and study mRNA processing must be cautiously interpreted. Bachellerie et al. (1978) showed that pulse-labeled heterogeneous nuclear RNA in cycloleucine treated cells was transiently of higher molecular weight than that of normal cells, and a role for methylation in efficient cleavage and splicing was suggested. Since a large portion of the HnRNA turns over in the nucleus (Salditt-Georgieff et al., 1981) and apparently normal RNA transcripts can be processed aberrantly (Dolan et al., 1983), the nature of the transient high molecular weight HnRNA requires further investigation. Perhaps cyclo- leucine is causing an increase in production of aberrant RNA tran- scripts which are then degraded in the nucleus. Stoltzfus and Dane (1982), using cycloleucine, have shown that when avian sarcoma virus (ASV) RNA is 90% undermethylated, splicing of the viral transcripts is inhibited. These results imply a role for m6A methylation in splicing. We have used the methylation inhibitor S-tubercidinylhomocysteine (STH) to probe the function of mRNA methylation. STH is the 7-deaza analog of SAH and inhibits methylation competitively. It is stable in cells since it is not a substrate for SAH hydrolase (Chiag et al., 1977; Crooks et al., 1979). STH has been shown to inhibit methylation of tRNA (Crooks et al., unpublished results; Chang and Coward, 1975), catecholamines (Coward et al., 1974), dopamine (Michelot et al., 1977) and mRNA (Kaehler et al., 1977). STH does not cause significant inhibition of rRNA methylation (Rottman et al., 1979). DNA synthesis in lymphocytes is inhibited by STH (Chang and Coward, 1975). 18 In vitrg_studies have revealed a differential sensitivity of viral enzymes to STH. NDV and vaccinia virus (guanine-7-)methyl- transferases were inhibited by STH with Ki of 0.4 uM and 140 uM, respectively (Pugh et al., 1977; Pugh and Borchardt, 1982). Vaccinia mRNA (nucleoside-ZY) methyltransferase was inhibited jg_yitrg with a Ki of 1.2 uM (Pugh and Borchardt, 1982. Kaehler et.al (1979) found STH not to inhibit m7G in Novikoff cells. This may be a reflection of the instability of the uncapped transcripts, or a difference in the Khfls of the enzymes. The mamnmlian m7G-methyltransferase is reported to be more like the vaccinia enzyme (Banerjee, 1980). STH was shown to inhibit 2'-9_-methylation and m6A methylation in Novikoff cells (Kaehler et al., 1977). Undermethylated mRNA could be found in the cytoplasm on polysomes indicating that 2'-Q-methylation and m6A methylation were not strictly required for translation. Similarly, in cycloleucine treated cells, incorporation of mRNA into polysomes appeared normal (Dimock and Stoltzfus, 1979L These studies suggested that the selection of mRNA for translation is not influenced very much by 2'-Q-methylation. As polyamines have been implicated to be important in rRNA processing (Levin and Clark, 1979), it was crucial to determine whether STH perturbed polyamine biosynthesis in cells. Spermine and spermidine synthases, thought to be the rate limiting enzymes in polyamine biosynthesis, were both shown to be insensitive to inhibi- tion by STH. In extracts of rat ventral prostate, 1 mM STH resulted in only 17 and 19% inhibition of these enzymes whereas SAH at the same 19 concentration inhibited by 7 and 47% respectively (Hibasami et al., 1980). We have utilized STH to further explore the role of methylation in mRNA metabolism, using HeLa cells as a model system. We have also begun to develop a system for analysis of the post-transcriptional processing of two specific non-viral mRNAs, bovine pituitary prolactin and growth hormone. 20 REFERENCES Ahmed, C.M.I., Chanda, R., Stow, N., and Zain, 8.5. (1982) Gene g, 297-301. Aloni, Y., Dhar, R., and Khoury, G. (1979) g_. Virol. 32, 52-60. Amara, S.G., Jonas, V., Rosenfeld, M.G., Ong, E.S., and Evans, R.M. (1982) Nature 298, 240-244. Babich, A., Nevins, J.R., and Darnell, J.E. (1980) Nature 287, 246- 248. Bachellerie, J.-P., Amalric, F., and Caboche, M. (1978) Nucl. Acids Res. _5_, 2927-2943. Banerjee, A.K. (1980) Microbiological Rev. E, 175-205. Bartkoski, M.J. and Roizman, B., (1978) Virol. 85, 146-156. Bastos, R.N., Volloch, Z., and Aviv, A. (1977) JiMol. Biol. 110, 191- 203. Beemon, K.L. and Keith, J.M. (1977) g_._ Mol. Biol. 113, 165-179. Both, G.W., Furuichi, Y., Muthukrishnan, S., and Shatkin, A.J. (1975) Cell 5, 185-195. Canaani, D., Kahana, C., Lavi, S., and Groner, Y. (1979) Nucl. Acids Res. 6, 2879-2899. Chang, C.-D. and Coward, J.K. (1975) Mol. Pharmacology 1;, 701-707. Chen-Kiang, S., Nevins, J.R., and Darnell, J.E. (1979) g; Mol. Biol. 135, 733-752. Chen-Kiang, S., Wolgemuth, D.J., Hsu, M.-T., and Darnell, J.E. (1982) Cell Q, 575-584. Chiang, P.K., Richards, H.H., and Cantoni, G.L. (1977) flo_l_._ Pharmacology _1_§, 939-947. Choi, Y.C., Reddy, R., and Busch, H. (1982) In Biochemistry 9_f__§_: adenosylmethionine and Related Comgounds (E. Usdin, R.T. Borchardt, and C.R. Creveling, eds.) Macmillan Press Ltd., England, 313-320. 21 Chung, S., Landfear, S.M., Blumberg, D.D., Cohen, N.S., and Lodish, H.F. (1981) Cell 24, 785-797. Coppola, J.A., Field, A.S., and Luse, D.J. (1983) Proc. Natl. Acad. Sci. _8_0, 1251-1255. Coward, J.K., Bussolotti, D.L., and Chang, C.D. (1974) 9; Med. Chem. 11, 1286-1289. Crooks, P.A., Dreyer, R.N., and Coward, J.K. (1979) Biochem. 33, 2601- 2609. Crooks, P.A., Kaehler, M., Rottman, F.M., and Coward, J.K., unpublished results. Darnell, J.E. (1982) Nature 297, 365-371. DeNoto, F.M., Moore, 0.0., and Goodman, H.M. (1981) Nucl. Acids Res. 3, 3719-3730. Dimock, K. and Stoltzfus, C.M. (1979) J_. Biol. Chem. 254, 5591-5594. Dolan, M., Dodgson, J.B., and Engel, J.D. (1983) 3; Biol. Chem. 258, 3983-3990. Dubin, D.T. and Taylor, R.H. (1975) Nucl. Acids Res. 2, 1653-1668. Early, P., Rogers, J., Davis, M., Cal ame, K., Bond, M., Wall, R., and Hood, L. (1980) Cell 23, 313-319. Elicieri, G.L. (1979) Nature 279, 80-81. Engel , J.D. and von Hippel, P. (1974) Biochem 33, 4143-4158. Engel, J.D. and von Hippel, P. (1978) J_. Biol. Chem. 253, 927-934. Filipowicz, W. (1978) FEBS Lett. 3, 1-11. Fraenkel-Conrat, H. and Singer, B. (1980) Proc. Natl. Acad. Sci. Z_7_, 1983-1985. Furuichi, Y. (1978) Proc. Natl. Acad. Sci. 25, 1086-1090. Furuichi, Y., LaFinandra, A., and Shatkin, A.J. (1977) Nature 266, 235-239. Gourse, R.L. and Gerbi, S.A. (1980) _._]_._ Mol. Biol. 140, 321-340. Green, M.R., Maniatis, T., and Melton, D.A. (1983) Cell 32, 681-694. Greenberg, J.R. (1972) Nature 240, 102-104. 22 Guyette, W.A., Matuski, R.J., and Rosen, J.M. (1979) Cell 31, 1013- 1023. Heintz, N., Sive, H.L., and Roeder, R.G. (1983) Molec. and Cell. Biol. 3, 539-550. Hibasami, H., Borchardt, R.T., Chen, S.Y., Coward, J.K., and Pegg, A.E. (1980) Biochem 3:381, 419-428. Hoffman, J.L. and Clark, R.S. (1983) Fed. Proc. Am. Soc. Expt. Biol. 42, 1767. Hopper, A.K., Furukawa, A.H., Pham, H.D., and Martin, N.C. (1982) Cell 23, 543-550. ' Jeffreys, A.J. and Flavell, R.A. (1977) Cell _1_2, 1097-1108. Jove, R. and Manley, J.L. (1982) Proc. Natl. Acad. Sci. Z_9_, 5842-5846. Kaehler, M.M. (1978) Ph.D. dissertation, 102. Kaehler, M., Coward, J., and Rottman, F. (1977) Biochem. 33, 5770-5775. Kaehl er, M., Coward, J., and Rottman, F.M. (1979) Nucl. Acids Res. 3, 1161-1175. Kahana, C., Yakobson, E., Revel, M., and Groner, Y. (1981) Viral. 112, 109-118. Kersten, H. (1982) In Biochemistry of S-adenosylmethionine and Related Compounds (E. Usdin, R.T. Bofihardt, and C.R. Creveling, eds.) Macmillan Press Ltd., England, 357-369. Kozak, M. (1978) Cell 33, 1109-1123. Lai, C.—J., Dhar, R., and Khoury, G. (1978) Cell fl, 971-982. Lavi, S. and Shatkin, A. (1975) Proc. Natl. Acad. Sci. 2, 2012-2016. Leis, J., Scheible, P., and Smith, R.E. (1980) _J_._ Virol. 33, 722-731. Lerner, M.R., Boyle, J.A., Mount, S.M., Wolin, S.L., and Steitz, J.A. (1980) Nature 283, 220-224. Levin, E.G., Clark, J.L. (1979) _J_._ Cell. Physiol. 101, 361-368. Lewin, B. (1980) In Gene Expression 2 John Wiley and Sons, N.Y. Liau, M.C. and Hurlbert, R.B. (1975) Biochem. fl, 127-133. 23 Lombardini, J.B., Chou, T-C., and Talalay, P. (1973) Biochem. 3.333, 43-57. Maden, B.E.H., and Salim, M. (1974) J_. Mol. Biol. 33, 133-164. Melton, D.A., DeRobertis, E.M., and Cortese, R. (1980) Nature 284, 143-148. Michelot, R.J., Lesko, N., Stout, R.W., and Coward, J.K. (1977) Mol. Pharmacol. _1_3, 368-373. Munns, T.W. and Liszewski, M.K. (1980) Progr. 3n Nucl. Acids Res. 213, 109-165. Murphy, J.T., Burgess, R.R., Dahlberg, J.E., and Lund, E. (1982) Cell 29, 265-274. Murray, V. and Holliday, R. (1979) FEBS Lett. 106, 5-7. Muthukrishnan, S., Moss, 8., Cooper, J.A., and Maxwell, E. S. (1978) J_._ Biol. Chem. 253, 1710-1715. Nevins, J.R. and Wilson, M.C. (1981) Nature 233, 113-118. Nichols, J.L., and Welder, L. (1981) Plant Science Letters 23, 75-81. Perry, R.P. (1981) J.Cell. Biol. 33, 28s-38s. Pugh, C.S.G. and Borchardt, R.T. (1982) Biochem. 21_, 1535-1541. Pugh, C.S.G., Borchardt, R.T., and Stone, H.O. (1977) Biochem. 33, 3928-3932. Revel, M. and Groner, Y. (1978) Ann. Rev. Biochem. 33, 1079-1126. Roberts, R.J. (1980) Nucl. Acids Res. 3, r63-r78. Robertson, H.D. (1982) Cell 33, 669-672. Roop, D.R., Kristo, P., Stumph, W.E., Tsai, M.J. and O'Malley, B.W. (1981) Cell 23, 671-680. Rose, J.K., Lodish, H.F., and Brock, M.L. (1977) J. Virol. _3, 683- 693. Rottman, F.M., Kaehler, M., and Coward, J. (1979) In Transmethylation (E. Usdin, R.T. Borchardt, and C.R. Creveling, eds.) Elsevier North Holland, N.Y., 361-371. Salas, C.E., and Leboy, P.S. (1983) Fed. Proc. Am. Soc. Expt. Biol. 52, 1885. 24 Salditt-Georgieff, M., Harpold, M.M., Wilson, M.C., and Darnell, J.E. (1981) Mole. and Cell. Biol. 3, 179-187. Sasavage, N.L., Smith, M., Gillam, S., Woychik, R.P., and Rottman, F.M. (1982) Proc. Natl. Acad. Sci. Q, 223-227. Schibler, U., Marcu, K.B., and Perry, R.P. (1978) Cell _1_3, 1495-1509. Smith, D.W.E., Hatfield, D.L., Mushinski, J.F., and McNamara, A.L. (1983) Fed. Proc. Am. Soc. Expt. Biol. 52, 2237. Sommer, S., Lavi, U., and Darnell, J.E. (1978) 33Mol. Biol. 124, 487- 499. Stark, B.C., Kole, R., Bowman, E.J., and Altman, S. (1978) Proc. Natl. Acad. Sci. E, 3717-3721. Stiles, C.D., Lee, K.-L., and Kenney, F.T. (1976) Proc. Natl. Acad. Sci.7_3, 2634-2638. Stoltzfus, C.M. and Dane, R.W. (1982) J. Virol. 42, 918-931. Taylor, W.L., Collier, K.J., Weith, H.L., and Dixon, J.E. (1981) Biochem. Biophys. Res. Commun. 102, 1071-1077. Tilghman, S.M., Curtis, P.J., Tiemeier, D.C., Leder, P., and Weissman, C. (1978) Proc. Natl. Acad. Sci. _7_5_, 1309-1313. Toniolo, 0., Basilico, C. (1976) Biochim. Biophys. Acta 425, 409-418. Toniolo, D., Meiss, H.K., and Basilico, C. (1973) Proc. Natl. Acad. Sci. _7_(_), 1273-1277. Vaughan, M.H., Soeiro, R., Warner, J.R., and Darnell, J.E. (1967) Biochem. 5_8, 1527-1534. Wei, C.M. and Moss, B. (1977) Biochem. _1_6_, 1672-1676. Weiss, S.R. and Bratt, M.R. (1974) g_. Virol. 33, 1220-1230. Wertheimer, A.M., Chen, S.-Y., Borchardt, R.T., and Furuichi, Y. (1980) 3._Biol. Chem. 255, 5924-5930. Wiskocil, R., Bensky, P., Dower, W., Goldberger, R.F., Gordon, J.I., and Deeley, R.G. (1980) Proc. Natl. Acad. Sci. 21, 4474-4478. Wolf, S.F. and Schlessinger, D. (1977) Biochem. gs, 2783-2791. Zeevi, M., Nevins, J.R., and Darnell, J.E. (1981) Cell 26, 39-46. PART ONE PARTIAL CHARACTERIZATION OF THE BOVINE PROLACTIN GENE 25 26 ABSTRACT A bovine genomic library was screened with a nearly full-length prolactin cDNA clone. One positive prolactin genomic clone was res- triction mapped, portions were subcloned into the plasmid vector pUC9, and exons partially mapped. The data indicate that a portion of the 3' end of the gene is missing and that the gene contains a minimum of 4 exons. A restriction fragment containing the 5' end of the gene was subcloned into the single stranded bacteriophage M13 and sequenced. The 5' flanking region of bovine prolactin was compared to the pub- lished sequence of rat prolactin. A region spanning 247 nucleotides flanking the bovine prolactin gene was approximately 87% homologous to the 5' flanking region of the rat gene, defining a region of potential regulatory significance. INTRODUCTION Post-transcriptional modifications of mRNA have been a primary interest of our laboratory and until about 1977, studies had largely been focused on bulk populations of mRNA. The advent of cDNA cloning technology made it possible to embark on the characterization of methylation and processing of a specific host cell mRNA. We Chose the pituitary polypeptide hormone prolactin because it is a very abun- dant, regulated gene product. We selected bovine as a system to study due to the large size of the gland and to easy access to tissue from local slaughterhouses. A technique for maintaining monolayer cell cultures of bovine anterior pituitary glands was developed in our laboratory, and prolactin protein production in these cultures remains 27 high (Padmanabhan et al., 1982). A cDNA library was constructed from pituitary poly(A)+ RNA by the dG-dC tailing technique. Prolactin cDNA clones were obtained (Nilson et al., 1980), and the largest cDNA clone, pBPRL72, was sequenced (Sasavage et al., 1982a). We assumed that a very abundant protein would be well-represented in the mRNA population. In fact, cytoplasmic poly(A)+ RNA from bovine pituitaries is very rich in prolactin mRNA (60-70%, Nilson et al., 1979). However, nuclear transcription studies in our laboratory indicated that the transcription rate of prolactin gene is low (0.04 to CLO65%, H. Meisner, unpublished results). Attempts to characterize the nuclear precursors for prolactin met with great difficulty, probably due in part to the low transcription rate of the gene. Other studies which required labeling the RNA in cell culture, such as the methylation analysis of a specific mRNA, were also technically not feasible. Meanwhile, techniques are being developed inav 0:500 000o0 000000: 000000: 00cox 0050 00000 0000 0:000 0005 0000 0000 000o0 000o0 000000: 00.0000 Hm 302 Emfizom 050 300300 08.50.20.030 N «0000 39 Figure 3 APro6 Restriction Map The prolactin cDNA clone (pBPRL72) is shown oriented in the Sfl‘to 1? direction with potential intervening sequence borders marked. A and B refer to the small fragments used as probes to orient the gene (Table 3). Below is shown a preliminary restriction map of AProG with the following sites and their abbreviations: KpnI (K1), SmaI ($1), HindIII (H3), EcoRI (R1), BamHI (Bl), 89111 (82), and XorII (X2). The bovine DNA is depicted as a single line, and the vector Charon 28 as a double line. The locations of four exons have been tentatively assigned and are indicated with dark boxes. Under the map for AProG are the HindIII subclones in pUC9 pAPZ, pAP6 and pAP12. The pAP12 subclone has been additionally mapped for PstI (P1) and PvuII (PZL "(a r—4 T s: :2 s 0.1 3 o 94 0 m ’J ." . E - o' ," M "' 0' ' 8 a' 'l‘ a '2... E 0I wz',.o:_‘:__,.. n ‘I ”0.1:........ I .0 x.:.:.... 2 " its .1 E N 2 dual J-M I “a «La «E . '0 [z \PROG I Kl SI 1 Kl 40 0.5 Kb I I ’I I ’I I ..s ‘.‘! 0. 4‘ B A; \‘~ NOD an ‘E’é n I n E N a. N J: 8 E m a; La :1 c: N ‘H -3 ‘L n. N- to” a z 41 Figure 4 Restriction map of pBPRL72 Restriction sites predicted from the published sequence (Sasavage et al., 1982a). Fragments purified and used as probes are indicated below the restriction map'and the coordinates of each fragment are indicated in parentheses. Potential exon junctions predicted by com- parison to the rat prolactin gene (Cooke and Baxter, 1982) are indi- cated with hatched lines. 42 10 Z {I‘d ’b Jo,” +———-4 (832-907) (406-6|2) '7 (l77-3l6) (l-Zl?) . IOOpr Figure 4 43 AL 00.000000 :0 03.03 000.0000 000 00000 05 00 0000000000002 <20 00.00.00.200 000000 000 000-0 00 .000000 0000000 000000 000 00 .000 000-000 00 .000 000-000 00 00000000 0000 000000000 00000200000 00000000000 03.; 0000.00 0003 3000 0000:0000 000.0303 0.00503 .3 000.2000 000 0003200 0.5.0: 0» 0000000000000 000: 3000.00 00.50.0500... 0000: .2600 00000003 0.00 Em: 000030.030 0000: 000.00.000.00, 0.00: 000:. 30050-00000 0000 .00 000.00 00: 00.05000 530.050 00.00.000.000 0000 0000000 :0 05.0 000 - 0000 00000 .0000 000000: -0000000 00000000000 000000000 000 000; 00000000 00: 020 00000 w¢.O mv.O ¢©.O ¢0.0 v0.0 ¢0.0 ¢.H¥ ¢.H¥ v.H ¢.~ ¢.H¥ O.N O.N O.N O.N O.m O.m N.N ¢.¢« v.v« 0.0 ¢.¢ m.¢« O O H.O O.m* 0.x mt Anxv Nat HH¥ ¢H« NH« HH« «at HH fiat mmNHm ON ON ON ON ON ON ON ON Hzmzosz Hm Hm\m: m: Hm Hm\m: m: Hm m: ZOHHOHmHmmm 0000-00 0 0000-0000 0 0000-0000 0 ”00000 00000 3. 0:88 00.00002 m 0—005 44 lished the orientation of the gene within the clone. The Sfidnost probe (A, in Figure 3) spanned at least 2 exons, hybridizing to a 4.5 kb HindIII fragment, and to the right arm of Charon 28, as well as to a 7 and a 1.4 kb EcoRI fragment. A probe from the mid-portion of the cDNA (B, in Figure 3) was predicted to span 2 exons, but only hybri- dized to a single band containing the right arm of Charon 28. This information established the orientation as 5' to 3' left to right in the Charon 28 vector (Figure 3) An identical blot containing HindIII and EcoR1 digests of AProB, and a double digest of pBPRL72 with PvuII and PstI as a control, was probed with an AluI/PstI fragment corresponding to 74 nucleotides at the 3' end of of the cDNA insert (Figure 4). This probe did hybridize to the control 632 bp PvuII/PstI fragment of pBPRL72, but it failed to hybridize to any of the bands from AProG indicating that the 3'-most region of the cDNA was not represented in the AProG genomic clone. To facilitate restriction mapping, a HindIII digest of APro6 was subcloned into the plasmid vector pUC9 (Vieira and Messing, 1982). The assignment of EcoRI sites and the BamHI site of AProG was confirmed by cutting two clones containing the 6 kb HindIII fragment in opposite orientations (pAPZ, pAP6) with EcoRI and BamHI (Figure 3). Since this region represents Shifianking DNA, it was not mapped further. A more detailed restriction map for the 4 kb HindIII fragment was constructed using the pAPlZ subclone (Figure 3). In addition to single digests, double digests with HindIII, PvuII, EcoRI, BglII, HincII and PstI were performed in several combinations to confirm the position of the restriction sites. Ordering of the fragments in pAPlZ was facilitated 45 by Southern blot analysis of some of the restriction digests (data not shown) which were probed with a S'PstI/HpaII fragment of pBPRL72 spanning nucleotides 1-217 (Figure 4L.The location of the Shanost exon was deduced to fall within an approximately (L8 kb BglII/HincII site in pAPlZ. Primary sequence gf_thgu§: flanking region The 5' most exon was mapped to within the approximately 800 bp BglII HincII fragment in the subclone pAP12. The 2 kb HincII fragment containing this section was gel purified, cut with BglII, and the resulting fragments subcloned into the BamHI-SmaI site of the M13 bacteriophage vectors mp8 and mp9 (Messing and Vieira, 1982). 'Fwo clones positive for pBPRL72 sequence were selected for sequence analy- sis. The orientation of the single-stranded clones is such that they are complementary. To date 400 base pairs of each clone has been sequenced by the dideoxy method (Sanger et al., 1977), but no overlap has yet been obtained. The preliminary sequence information generated has revealed that further digestion of the 2 kb HincII fragment with AluI should produce several smaller clones whose sequence would provide overlap and the necessary confirmation of the sequence from the opposite strand. The sequence of the S'flanking region, first exon and intervening se- quence junction was obtained from examination of multiple gels begin- ning at the BglII-BamHI junction and extending various distances downstream (rightward, Figure 3% The sequence of the messenger sense strand is shown in Figure 5, and is compared to the 5' portion of the prolactin cDNA (pBPRL72). Approximately 13 nucleotides at the 5' end 46 Sequence of Q1; §'_ portion g the bovine prolactin gene and comparison with rat prolactin A BglII-HincII restriction fragment from pAP12 (approximately 0&3 kb) was subcloned into the bacteriophage M13mp8 to provide a single stranded template. The sequence of the messenger sense of a portion of the bovine prolactin gene was determined by dideoxy sequencing and is shown in the figure (bold face). The region homologous to the canonical TATAAA sequence is underlined. The sequence of the bovine prolactin cDNA pBPRL72 (Sasavage, 1982a) is in lower case letters, juxtaposed with an homologous sequence in the gene. The points of divergence have been emphasized by elevating the lower case cDNA sequence. A portion of the rat prolactin gener(Cooke and Baxter, 1982) is given in upper case letters below the bovine sequence for purposes of comparison. The position of the rat prolactin cap site is indicated with an asterisk (*). The rat prolactin sequence has been aligned to allow maximum homology; points of mismatch are indicated by vertical lines. 47 n wuawfim H MUH:oe-:mv t .5... £4 .6 3 2.6 653.... iHuvHv Ho Ho: ¢.H zpwz umumnaocw mew; mHHmu .aHpcmscmmnam .zpm 2:.oom Lo 0 :H meson N Low :65» use mcwcowgpme cw uwozvme mwume :_ Lao; H umamazocw mum: mHHmo NoH x osH .cowuwucou comm Lou am Hm.o eofixo.H aofixm.fi 6.. monm.m Sofixm.fl <2“ +H '5 O. O 250 500 IOOO [STH] ,uM Figure 2 STH dose curve for inhibition _gipolyfl)+ RNA methylation 4 x 106 cells were treated with varying amounts of STH in methio- nine-free Joklik's MEM supplemented with 5% dialyzed horse serum and 20 mM sodium formate. After 45 minutes, actinomycin D was added to 40 ng/ml to inhibit the synthesis of rRNA. Cells were labeled for 4 hr with (3H-methyl)-methionine (0.15 mCi/ml, 12 Ci/mmole) (0-0) and (14C)-uridine (0.1 uCi/ml, 0.48 Ci/mmole) (o-o). Cytoplasmic poly(A)+ RNA was isolated as described in Methods. 107 At all concentrations of STH tested, there was no change in cell viability, based on trypan blue exclusion. In a separate experiment, cells were treated with 0 and 500 M STH for 10 hours, then proteins were pulse-labeled for 2 hours with (353)-methionine. Examination of the radioactive proteins on one dimensional polyacrylamide gels indi- cated that protein synthetic capability was unaltered by STH treatment (data not shown). @antitation fl cap distribution and inhibition 9_f_ methylation To determine the extent of inhibition of the various nucleoside residues, mRNA was digested with RNases T2 and P1 and alkaline phos- phatase, and the digestion products separated on HPLC. RNase T2 is a nonspecific ribonuclease that cleaves RNA with free 2' hydroxyl groups to produce nucleotides with 3' phosphates and cap cores (e.g. m7GpppN'mpN"mpr). Cleavage by nuclease P1 is not blocked by the presence of a methyl group on the 2' position of the ribose and thus reduces cap two structures to a core (m7Gppmep) and releases the 25- .Q-methylated nucleotide. After simultaneous digestion with T2 and P1, the cap cores and modified and unmodified nucleotides were dephos- phorylated with alkaline phosphatase» The HPLC separation employed was devel0ped for rapid, simple quantitation of modified nucleosides and cap structures in a single chromatographic step. All of the possible core cap structures, unmodified nucleosides, 2'-Q-methylated nucleo- sides, base methylated nucleosides including mSC, m°A, and m6 2A, can be cleanly separated (Figure 3% The cap structures are charged and retained on the anion exchange pre-column while the non-charged nucleosides are separated as they are 108 .3058... 5 63.23% mm REP-E .5086 mm: 5.53m .58.; fl 55:8 323 3226.. m moo :mech 5.52: 23 55:3 mmcmzuxm cot; x