polyadenylate polymerases from normal and cancer cells and ... · zyme, which indicates that the...

[CANCER RESEARCH 49, 2827-2833, June 1, 1989]

Review

Polyadenylate Polymerases from Normal and Cancer Cells and Their Potential Rolein Messenger RNA Processing: A Review1

Samson T. Jacob,2 Michael P. Terns,2 and Kathleen A. Maguire3

Department of Pharmacology and Cell and Molecular Biology Center, The Pennsylvania State University, College of Medicine, Hershey, Pennsylvania 17033

Abstract

Two structurally and immunologically distinct species of nuclear poh -

adenylate (poly(A)) polymerases have been characterized. One of theseenzymes is relatively absent in normal tissues but is predominant inprimary and transplanted tumors and transformed cell lines. The presenceof the tumor type enzyme in fetal liver, but not in regenerating liver,suggests that it is an oncofetal protein. Antibodies against the tumor-

type poly(A) polymerases are present in the sera of rats bearing tumorsand in some cancer patients. These antibodies are also found in the seraof rats fed hepatocarcinogen even before preneoplastic nodules werevisible, which suggests that elicitation of these antibodies is an earlyevent in neoplastic transformation. Autoantibodies against both liver-type and tumor-type poly(A) polymerase are also present in some rheu

matic autoimmune sera. Polyclonal antibodies against purified enzymefrom a rat hepatoma, which exhibit a single band upon immunoblotanalysis, were used in cell-free extracts to study the role of poly(A)polymerase in the 3'-end processing of pre-mRNA. These studies showed

that the antibodies blocked both endonucleolytic cleavage and poly(A)addition at the cleavage site and complex formation between factors inthe extract and pre-mRNA. Independent studies in other laboratories

have demonstrated that both the cleavage and poly(A) polymerase activities require the same component for their function. These observationssuggest that both cleavage and polyadenylation reactions are tightlycoupled in a functional complex.

I. Introduction

The vast majority of eukaryotic mRNAs contain poly(A)4tails at their 3' ends. Despite the significant progress made

towards the identification and partial characterization of a fewfactors involved in the cleavage/polyadenylation of pre-mRNAs, the mechanism by which a properly polyadenylatedmRNA 3' terminus is formed has not been fully elucidated.Current evidence indicates that RNA polymerase H-directedtranscription proceeds well beyond the poly(A) site and terminates over a heterogeneous DNA stretch at considerable distances from the eventual 3' end. The 3' terminus of mRNA

arises from a posttranscriptional reaction which involves anendonucleolytic cleavage of the larger precursor followed by theaddition of an adenosine tract, approximately 200-250 nucle-otides long, to this newly generated end (for extensive reviews,see refs. 1-3). It is presumed that the latter reaction is enzy-matically accomplished by poly(A) polymerase (polynucleotideadenyltransferase; ATP: polynucleotide adenyltransferase, EC2.7.7.19). This review focuses on the important characteristicsof poly(A) polymerase and the recent studies which have pro-

Received10/28/88;revised2/21/89;accepted2/27/89.Thecostsof publicationof thisarticleweredefrayedin partbythe payment

of pagecharges.Thisarticlemustthereforebe herebymarkedadvertisementinaccordancewith18U.S.C.Section1734solelyto indicatethisfact.

1Work carried out in the authors' laboratory was supported by USPHS Grants

CA 25078 and CA 31894, by the Pennsylvania Lupus Foundation, and by theDiagnostic Division of Allied Health and Scientific Products (to S. T. J.).

2 Present address: Department of Pharmacology and Molecular Biology, The

Chicago Medical School, N. Chicago, IL 60064.3 Present address: The Wistar Institute, Spruce St., Philadelphia, PA.4The abbreviations used are: poly(A), polyadenylate; RNP, ribonucleoprotein;

sn, small nuclear.

vided direct evidence for the role of this well characterizedenzyme in the 3' end processing of eukaryotic pre-mRNAs.

II. Summary of the Properties of Poly(A) Polymerase

For extensive discussions on the characteristics of poly(A)polymerase the reader is asked to refer to previous reviews (seeRefs. 4 and 5). These reviews have covered in great detailintracellular distribution, size and subunits, other physicalproperties, substrate, primer and divalent metal ion requirements, inhibitors, phosphorylation of enzyme, and nucleasesassociated with enzyme preparations. Only a brief descriptionof the properties of this enzyme from higher organisms that arerelevant to the present discussion is presented here.

Poly(A) polymerase appears to be a ubiquitous enzyme andis present in all organisms ranging from prokaryotes to eukar-yotes. In mammalian cells, the enzyme has been identified inthe nucleus (6), RNPs (7), mitochondria (8), microsomes (9),ribosomes (10), and postmitochondrial fractions (11). The nuclear enzyme is found exclusively in the extranucleolar fraction,which is consistent with the lack of polyadenylation of rRNAin the nucleolus. The nuclear enzyme occurs as chromatin-bound and free forms (12-16). Following solubilization of thebound enzyme, it attains the characteristics of the "free" en

zyme, which indicates that the two populations of the enzymerepresent the same polypeptide in two functional states. Thechromatin-bound enzyme is highly sensitive to low levels ofcordycepin triphosphate (3'-dATP) whereas the free form of

the enzyme is inhibited only by relatively high concentrationsof the ATP analogue (17). The differential sensitivity of the twopoly(A) polymerase populations to 3'-dATP is consistent with

similar responses of the initial polyadenylation and poly(A)extension reactions to cordycepin in vivo (18, 19). Based onthese data, we have concluded that the chromatin-bound andfree forms of the enzyme are responsible for the initial polyadenylation and poly(A) elongation reactions, respectively (17,20). More recent studies with a reconstitution system in vitrohave further supported this contention (see Section VIII).

Purification of enzymes consisting of a single subunit hasbeen accomplished in several animal systems (for review, seeRefs. 4, 5, 21, and 22). Poly(A) polymerase is assayed usuallyby monitoring the incorporation of labeled AMP into syntheticprimer. It is clear that purified poly(A) polymerase cannotcatalyze the entire reaction which leads to mRNA 3' end

formation. Poly(A) polymerase activity is only one of the activities required for this highly specific and complex reaction.

The size of the enzyme, as displayed after gel electrophoresisunder denaturing conditions, ranges from 37,000 to 60,000depending upon the source of the enzyme (see Refs. 4 and 5).The enzyme can occur as a dimer or tetramer (see Section III).Very high molecular weights (300,000) have been reportedunder certain conditions (see Section IV). The enzyme hasdisplayed heterogeneity upon chromatography on phosphocel-lulose (23, 24) or carboxymethyl cellulose (25) columns. Al-

2827

on April 2, 2021. © 1989 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

http://cancerres.aacrjournals.org/

POLY(A) POLYMERASES AND POTENTIAL ROLE IN mRNA PROCESSING

though multiple forms of the enzyme may correspond to structurally distinct species of the enzyme, such display of heterogeneity is, in most cases, the result of posttranslationalmodification of the enzyme such as phosphorylation which isknown to alter its Chromatographie elution characteristics (25).

The enzyme activity can also undergo rapid changes in response to a variety of physiological and pathological stimuli(4). The earliest and perhaps the most rapid change in theenzyme activity occurs in rabbit heart following treatment withnoradrenaline or dibutyryl cyclic AMP (26, 27). The nuclearenzyme can also be stimulated by glucocorticoid (28), progesterone (16), and estradiol (29). The nuclear poly(A) polymeraseactivity can vary with age of the animal. Thus, the enzymeactivities in the brain and liver nuclei derived from 25-day-oldrats are 3- and 26-fold greater than the corresponding enzymeactivities from 30-day-old animals (30). Mixing experimentshave shown that the reduced poly(A) polymerase activity intissues from the older rats is not due to high levels of inhibitors(30). The differences in the enzyme activity appear to be tissueand perhaps species specific (31). Poly(A) polymerase activitycan also respond to mitogens (32) and is subject to diurnalvariations (33). The enzyme is also known to respond to aminoacid supply (34). Alterations in poly(A) polymerase are notconfined to the mammalian enzyme, as wheat poly(A) polymerase has been shown to be stimulated by the plant hormone,gibberellic acid, a process that is dependent on de novo proteinsynthesis (35). Poly(A) polymerase activity can also be modulated by actin and tubulin (36). The poly(A) polymerase fromcalf thymus appears to be closely associated with a poly(A)-specific endoribonuclease (36). The enzyme activities in thiscomplex are strongly inhibited by G-actin and tubulin dimerwhereas the polymerized forms of this protein are less effective.

Changes in poly(A) polymerase activity were detected byassaying the solubilized or partially purified enzymes usingexogenous primers. Since solubilized enzymes generally losetheir specificity for primers (17,20), the physiological relevanceof the alterations in the enzyme is not evident. Further studiesusing reconstituted cell-free systems are required to understandthe exact implications of these interesting findings. The adventof such in vitro systems for studying polyadenylation and therecent success in reconstituting partially purified poly(A) polymerase with other components in the polyadenylation ofcertain specific mRNA precursors (see Section VIII) have provided further impetus for exploring the potential alterations ofthe polyadenylation reaction by posttranslationally modifiedpoly(A) polymerase.

III. Two Distinct Nuclear Species of Poly(A) Polymerase: Demonstration of a Liver-type and Tumor-type Enzyme

More than a decade ago, we (6) showed that the nuclearenzyme from rat hepatoma is distinct from the correspondingrat liver enzyme with respect to molecular weight and aminoacid composition. Subsequent studies demonstrated that thetwo enzymes differ in zinc content (37) and degree of phosphorylation (23). The hepatoma enzyme is much more phospho-rylated than the liver enzyme (23). Since phosphorylation ofpoly(A) polymerase causes a dramatic activation of the enzyme,it is likely that the higher activity of the hepatoma enzyme is,in part, due to its high degree of phosphorylation.

Later studies have shown that rat liver nuclei contain apredominant M, 38,000 enzyme and a minor M, 48,000 species.The M, 48,000 enzyme constitutes approximately 1% of thetotal liver nuclear enzyme. The corresponding enzyme from the

rat tumor, Morris hepatoma 3924A, is composed of a singlespecies of M, 48,000 enzyme which is identical to the minorliver nuclear poly(A) polymerase in every respect (38). Gelfiltration studies indicate that the major liver nuclear enzymeis a tetramer of A/r 38,000 polypeptide whereas the hepatomaenzyme is a dimer of a M, 48,000 polypeptide (38). Fractiona-tion of liver poly(A) polymerase by Sephacryl S-200 resulted incomplete loss of activity which could be restored by incubationwith exogenous N,-type protein kinase (38). The latter kinasewas initially associated with poly(A) polymerase but could beseparated and purified following extensive Chromatographiefractionations (39). The rapid loss of the liver enzyme activityby dephosphorylation is consistent with the relatively reducedlevel of phosphorylation in the liver enzyme relative to thehepatoma enzyme (38).

IV. Structural and Immunologie-ai Differences between the Liver-type and Tumor-type Enzyme

Cleavage of liver (MT 38,000) and hepatoma (Mr 48,000)poly(A) polymerases with cyanogen bromide followed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis andsilver staining showed that the peptide maps of the two enzymesare distinct. The peptide map of the tumor enzyme was identicalwith that of minor (M, 48,000) liver enzyme. It is not evidentwhy the liver nuclei contain a small proportion (


RNA was translated or when preimmune serum was used forprecipitation.

The presence of structurally and immunologically distinctnuclear poly(A) polymerase in transplanted hepatoma raisedthe possibility that this enzyme may be produced at an earlystage of tumorigenesis. To test this possibility, primary hepa-tomas were induced in male Fischer 344 rats by feeding anazodye diet (42). Nuclear poly(A) polymerase purified from theprimary hepatomas was identical to the corresponding enzymefrom the transplanted hepatoma with respect to molecularweight and immunoreactivity with the anti-hepatoma poly(A)polymerase antibodies (42). The production of a distinctpoly(A) polymerase in the early stages of neoplastic processsuggests that this enzyme may be involved in some way intumor development. The enzymatic activity of this protein maynot be crucial in tumorigenesis. Alternatively, the tumor-typepoly(A) polymerase in concert with other factor(s) could poly-adenylate a distinct class of mRNAs that are required for thetransformation of normal cells to cancer cells.

V. The Tumor-type Poly(A) Polymerase Is an Oncofetal Antigen

The expression of fetal antigens in tumor has been welldocumented (43-45). To investigate whether the hepatomapoly(A) polymerase is related to the fetal enzyme, poly(A)polymerase was purified from fetal livers derived from pups at18 days of gestation (46). Since the quantity of tissue waslimited, the enzyme was only partially purified. When thisenzyme was subjected to Western blot analysis, only one bandcorresponding to M, 48,000 was visible. There was no immunecomplex formation when anti-hepatoma RNA polymerase Iantibodies were used for the immunoblot analysis. The adultliver-type poly(A) polymerase (M, 38,000) was either absent orpresent in minute quantity in the fetal liver (46). The immu-nological data were consistent with the structural data whichshowed that the CNBr cleavage patterns of the fetal enzymeand the hepatoma enzyme were identical. Further studies haveshown that fetal poly(A) polymerase is not characteristic ofactively growing tissue, because poly(A) polymerase from regenerating liver (18 h posthepatectomy) did not contain tumor-type poly(A) polymerase; rather, partial hepatectomy caused a3-fold increase in the level of liver-type poly(A) polymerase.5These observations coupled with the cell-free translation datahave concluded that tumor poly(A) polymerase gene is inactivein normal tissues postnatally but is reexpressed in tumors byreactivation at an early stage in carcinogenesis.

VI. Detection of Anti-Tumor-type Poly(A) Polymerase Antibodies in the Sera of Tumor-bearing Rats and Cancer Patients

The immunological difference between adult or regeneratingliver poly(A) polymerase and the corresponding hepatoma enzyme raised the possibility that the tumor enzyme may beimmunogenic to the host. The tumor poly(A) polymerase maybe released by tumor necrosis which may then lead to elicitationof antibodies against the tumor enzyme in the sera of the tumor-bearing hosts. To test this possibility, sera from rats bearingtransplanted Morris hepatomas with different growth rates (44,7288C, 5123D, 28A, 9618A, 7787, 3924A) and mammaryadenocarcinoma R3230AC were screened for anti-hepatomapoly(A) polymerase antibodies by immunoblot analysis (47). Adistinct immune complex was observed when sera from all ratsbearing tumors for more than 7 weeks were used for analysis.

"'D. A. Stetler and S. T. Jacob, unpublished data.

Since rats carrying Morris hepatoma 3924A usually die priorto the minimal period required for significant immune response,antibodies were not found in the sera of these animals despitethe high sensitivity of the technique used for Western blotanalysis. Anti-tumor poly(A) polymerase antibodies were alsodetected in the sera of azo dye-fed rats (42). Interestingly, serafrom rats fed the carcinogen for just 21 weeks contained antibodies against the hepatoma enzyme (42). The antibody titerof sera from rats that were fed carcinogens for 30 weeks wasidentical to that from rats bearing the transplanted Morrishepatoma 3924A. Poly(A) polymerase from adult rat liver didnot react with the sera from carcinogen-fed rats, which indicatethe specificity of the antibodies against the tumor enzyme. Sinceantibodies against RNA polymerase I or II were not detectedin these sera, the elicitation of anti-poly(A) polymerase antibodies in these sera do not appear to be a random occurrence.The detection of the latter antibodies prior to the formation ofpreneoplastic nodules suggests that tumor-type poly(A) polym

erase is produced at an early stage of carcinogenesis and mayvery well be a unique tumor marker. Further studies are requiredto validate this possibility.

The presence of anti-poly(A) polymerase antibodies in thesera of tumor-bearing rats provided an impetus to examine thesera of cancer patients for these antibodies. Studies in ourlaboratory (47) have demonstrated that sera from patients withacute lymphocytic leukemia and Wilms' tumor contain anti-

poly(A) polymerase antibodies. Since the presenting antigen forthese studies was purified rat hepatoma poly(A) polymerase, itappears that poly(A) polymerases from rat and human are, atleast, antigenically and probably structurally, identical. Themost significant observation was a dramatic increase (as muchas 15-fold) in the antibody titer following relapse of a patientwith acute lymphocytic leukemia. Although this is an encouraging finding, it is important to screen several sera beforemeaningful conclusions can be made. One noteworthy findingwas the elicitation of antibodies against tumor poly(A) polymerase in polycythemia vera, a condition marked by an abnormalnumber of erythrocytes with a potential relationship to myelo-cytic leukemia. The specificity of the anti-tumor poly(A) polymerase antibodies in the human cancer sera was evident by theabsence of such antibodies in the sera of (a) healthy individuals,(Â¿>)cancer patients in remission for several months, (c) hypertensive patients, or (d) patients with certain inflammatorydiseases such as Crohn's disease, Guillain-BarrÃ©syndrome or

postneonatal meningitis (47). However, certain rheumatic autoimmune sera also contain anti-tumor-type poly(A) polymerase antibodies (see Section VII).

VII. Autoantibodies against Nuclear Poly(A) Polymerases inRheumatic Autoimmune Diseases

The presence of circulating autoantibodies against a varietyof nuclear proteins is one of the major characteristics of rheumatic autoimmune diseases, which has been used in the clinicaldiagnosis of these diseases (48). Recent studies in our laboratoryhave shown that poly(A) polymerase is an autoantigen to whichantibodies are produced by patients with a variety of rheumaticautoimmune diseases such as systemic lupus erythematosus,rheumatoid arthritis, SjÃ¶gren's syndrome, and scleroderma

(49). Sera from 53 patients, 26 with systemic lupus erythematosus, 8 with rheumatoid arthritis, 9 with SjÃ¶gren's syndrome,

and 10 with scleroderma were screened for anti-poly(A) polymerase antibodies using purified liver and hepatoma poly(A)polymerases as the presenting antigens. Interestingly, some

2829




patients produced antibodies against liver poly(A) polymerasewhich has not been an effective antigen in rabbits (41). Thepresence of anti-liver poly(A) polymerase antibodies in these

sera is consistent with the capacity of the autoimmune patientsto elicit antibodies against highly conserved molecules (see Ref.48). Sixty and 74% of the sera used for screening containedantibodies against liver poly(A) polymerase and tumor poly(A)polymerase, respectively, and 25% of the sera exhibited antibodies exclusively against the tumor poly(A) polymerase. IgGcontaining anti-liver and anti-hepatoma poly(A) polymeraseantibodies inhibited liver and hepatoma poly(A) polymerase,respectively, whereas IgG against liver and tumor enzymesinhibited both enzymes. These observations confirmed the specificity of the antibodies present in the autoimmune sera. Unlikethe very early appearance of poly(A) polymerase antibodies inthe hepatocarcinogen-fed rats well before preneoplastic noduleswere detected, these antibodies are produced only after theonset of the rheumatic diseases.

It was of interest to note that poly(A), the product of thepoly(A) polymerase reaction, is also a major autoantigen (50).There are reports indicating inhibition of polyadenylation ofadenovirus mRNA by sera classified as anti-sm, anti-(Ui)RNP,and anti-La (51, 52). Although these sera might contain antibodies against the endonuclease responsible for the specificcleavage of mRNA precursor, it is plausible that these seraaffect the polyadenylation reaction via direct inhibition ofpoly(A) polymerase.

Antibodies against the liver-type poly(A) polymerase wererecovered from the autoimmune sera by using an affinity purification procedure. These antibodies were used to study tissueand species distribution of the liver-type enzyme.6 These studiesshowed that the liver-type enzyme is present in rat brain, kidney,spleen, and heart whereas the tumor-type enzyme is predominant in a variety of tumors and transformed liver cell lines. Therat liver-type poly(A) polymerase was also found in pig andhuman tissues and cell lines.

VIII. Occurrence of Poly(A) Polymerase as a Complex and ItsFunctional Implications

Studies on the polyadenylation reactions in vivo as well as incell-free systems suggest that the endonuclease and poly(A)polymerase activities may act in a coupled fashion wherebycleavage occurs at the poly(A) site and addition of adenylatetracts rapidly follows cleavage. This apparent linkage betweencleavage and polyadenylation implies a physical association ofthe two enzymes involved in these reactions and has led to theconcept that poly(A) polymerase may exist as part of a largemulticomponent enzyme complex. The existence of a poly(A)polymerase complex could account for the efficiency of 3'

editing. Interaction with such a multicomponent poly(A) polymerase complex could induce conformational changes in thepre-mRNA substrate and the resultant RNA structures wouldbe readily and coordinately acted upon by intrinsic processingactivities of this enzyme complex. Alternatively, selection ofcleavage at the poly(A) site and poly(A) addition could beperformed by separate factors which individually act on mRNAprecursors.

Elucidation of the mechanisms by which poly(A) polymerasemediates the processing of eukaryotic pre-mRNAs requiresidentification and characterization of the molecules with whichthe enzyme specifically interacts. Despite efforts to elucidate

' J. Hengst and S. T. Jacob, unpublished data.

the catalytic components required for 3' end processing, pro

gress in the positive identification of the trans factors involvedin these reactions has been relatively slow. Hashimoto andSteitz (53) showed that a factor with the properties of an smâ€”snRNP associates with the hexanucleotide sequence upstreamof the poly(A) site. A M, 64,000 polypeptide which can interactspecifically with the AAUAAA sequence has been identified byUV cross-linking (54). This factor appears to be a componentof the cleavage machinery. Zarkower and Wickens (55) havedemonstrated the formation of a stable precleavage complexbetween a factor present in HeLa nuclear extracts and theAAUAAA sequence. This hexanucleotide is located 10-30 nu-cleotides upstream of the poly(A) site of almost all eukaryoticniRNAs (56) and is implicated in the cleavage/polyadenylationreactions (57-61). Zarkower and Wickens have postulated thatthis factor may indeed be poly(A) polymerase. Further studiesare needed to prove that poly(A) polymerase can bind to thehexanucleotide sequence.

A gel mobility shift analysis of in vitro polyadenylation reactions has been used for the detection of factors involved inthe cleavage and polyadenylation processes. These studies analyzed the RNA/protein complexes produced after addition ofpre-mRNA precursors to HeLa nuclear extracts under conditions favorable for 3' end-processing. Electrophoretic separa

tion on nondenaturing gels of the complexes formed withHSV1-TK (62), Adenovirus L3 (63, 64), herpes simplex virus(65), and SV40 late (66) pre-mRNA substrates revealed common characteristics. Similar complex formation has also beenobserved following glycerol (67) or sucrose density gradient(68) fractionation of the product formed with Adenovirus L3mRNA. Distinct complexes were formed with pre-mRNAs overtime and the association of these factors correlated with therequirements essential for the in vitro reactions. Humphrey etal. (64) have reported that mRNA precursors containing theAdenovirus L3 polyadenylation site can form specific complexes with HeLa nuclear extract and that the complex protectsa 67-nucleotide region of the mRNA from RNase Ti digestion.This protected fragment contains both the AAUAAA- and GU-rich downstream cis element. On the basis of their findings,Humphrey et al. (64) have proposed that the assembly ofspecific complexes represents an essential step during the 3'

end processing of mRNA.Recent studies in our laboratory (69) suggest that newly

synthesized poly(A) polymerase is associated with other poly-peptides. Polyclonal antibodies raised against highly purifiedtumor-type poly(A) polymerase were used to immunoprecipi-tate poly(A) polymerase from [35S]methionine-labeled McA-

RH 7777 cells. In addition to immunoprecipitating the expectedpoly(A) polymerase molecule, at least 4 other labeled proteinswith approximate molecular weights of 70,000, 35,000, 30,000,and 23,000 were also consistently immunoprecipitated (69).The specificity of the anti-poly(A) polymerase antibodies wasdemonstrated by immunoblot analysis of the 7777 nuclearextract. A single band corresponding to MT48,000 tumor-typeenzyme was observed by this analysis. Further fractionation ofthe enzyme showed that the complex consisting of poly(A)polymerase and other polypeptides could be dissociated; onlyone 35S-labeled band corresponding to poly(A) polymerase was

visible after the fractionation. No polypeptide was precipitatedfollowing reaction with preimmune sera. The observation thatan antibody directed against a single purified protein can precipitate additional components implies that nascent poly(A)polymerase is associated with four distinct polypeptides as acomplex. It is conceivable these polypeptides may have a role

2830




in the polyadenylation of pre-mRNAs.Another factor that appears to be part of the polyadenylation

complex is small nuclear RNA and perhaps its associatedproteins (51-53, 68, 70, 72, 73). Ui and U4 snRNAs have beenimplicated in the cleavage/polyadenylation reactions (52, 53,71, 74). Interestingly, histone mRNA precursors that are notpolyadenylated require U7-RNA for the cleavage reaction whichgenerates the correct 3' terminus of these transcripts (75, 76).

Recently, our laboratory has been able to demonstrate a closeassociation of UrRNA in the polyadenylation complex byimmunoprecipitating extract derived from 32P-labeled 7777cells with anti-hepatoma poly(A) polymerase antibodies followed by extraction of RNA and gel electrophoresis underdenaturing conditions. The only RNA immunoprecipitated under these conditions corresponded to Ui-RNA (74). Thesestudies do not establish that Ui is directly involved in the 3'

end processing of pre-mRNA. It is plausible that other sn-RNAs/RNPs play a more direct role in the cleavage/polyadenylation reaction (see below).

Studies in our laboratory (77) have shown that anti-poly(A)polymerase antibodies raised against highly purified M, 48,000poly(A) polymerase can inhibit polyadenylation of AdenovirusL3 pre-mRNA. Surprisingly, the same antibodies also inhibitthe cleavage reaction when it was coupled or uncoupled withpolyadenylation. The anti-poly(A) polymerase antibodies couldalso prevent formation of specific complexes between pre-mRNA and components of nuclear extract. These studies haveconcluded that previously characterized poly(A) polymerase isthe enzyme responsible for the addition of poly(A) tract at thecorrect cleavage site and that coupling of cleavage and polyadenylation appears to result from the potential requirement ofpoly(A) polymerase for the cleavage reaction. These observations have prompted us to conclude that poly(A) polymerase isassociated with the endonuclease and probably with other essential components in a functional complex.

Another interesting observation (77) has been with regard toMg2+-dependent and Mn2+-dependent poly(A) polymerase ac

tivities which correspond to chromatin-bound and free form ofthe enzyme (also see Refs. 12 and 17). In the presence of Mg2+,200-250-nucleotide-long poly(A) were added at the poly(A) siteof Adenovirus L3 pre-mRNA whereas as much as 400-800adenylic acid residues were added to the presence of Mn2+ (77).Both Mg2+- and Mn2+-activated poly(A) polymerases were in

hibited by anti-poly(A) polymerase antibodies which supportsthe previously held notion (see Refs. 13 and 17) that the initialaddition of poly(A) tract and poly(A) elongation reactions aredirected by the same enzyme. Although the bound enzyme hasbeen designated as "chromatin bound," it is probably associated

with the complex involved in the cleavage/polyadenylation (78).Takagaki et al. (78) have succeeded in fractionating HeLa

nuclear extract into fractions containing poly(A) polymeraseand cleavage activity. Efficient and accurate cleavage of threedifferent pre-mRNAs required reconstitution of the two fractions. Gilmartin et al. (72) have observed that an snRNP isrequired for cleavage at the poly(A) site, although the exactnature of the snRNP has not been investigated. However,snRNP was not essential for poly(A) addition to the cleavagesite (79). Very recently, Christofori and Keller (73) have separated and purified three factors that are required for accuratecleavage and polyadenylation of Adenovirus L3 mRNA. Oneof the factors is a poly(A) polymerase with a molecular weightof approximately 50,000-60,000 which is in general agreementwith the molecular weight (48,000) of the enzyme purified inour laboratory (38). The second factor is a cleavage factor with

a native molecular weight in the range of 70,000-120,000. Thethird component is a 200,000 factor that is needed for thecleavage activity and also confers specificity to the poly(A)polymerase activity. Further purification has revealed that theA/r 200,000 cleavage factor activity is associated with Un-snRNP, which suggest that Un-RNA may be involved in the 3'

end processing of pre-mRNA. The conclusion that both thecleavage and polyadenylation activities require the same component is consistent with our observation that anti-poly(A)polymerase antibodies can inhibit both cleavage and polyadenylation and further suggests that both these reactions aretightly coupled.

IX. Future Directions

Although significant advances were made earlier on the en-zymology of poly(A) polymerase, characterization of the factorscontrolling cleavage at the poly(A) site and subsequent polyadenylation has progressed at a relatively slow pace. Recentsuccess in the fractionation of poly(A) polymerase from theendonuclease and their reconstitution into a functional complex(73, 78) should facilitate further studies on the mechanism bywhich these two enzymes direct polyadenylation of specificmRNAs. The role of UrRNA that has been found associatedwith poly(A) polymerase in a functional complex should bedetermined. For example, what has prevented detection of UrRNA/RNP in the complex by other techniques? Is it involvedat some stages of the processing reaction in an indirect mannerand if so, at what stage of 3' end processing is this RNA

required? Is it essential for both endonucleolytic cleavage andpoly(A) addition? Is Ui-RNA simply involved in facilitating thefunction of Uii-snRNA associated with the cleavage factor? Dothe proteins associated with Un- or Ui-RNA play a role inmRNA 3' end processing?

The nature and functions of the four polypeptides associatedwith nascent poly(A) polymerase should be determined. Arethese proteins related to endonuclease and other cleavage factors. Is any one of these polypeptides related to proteins associated with snRNAs?

Another interesting development is the discovery of twostructurally and immunologically distinct species of nuclearpoly(A) polymerase. Since the tumor-type enzyme is found infetal liver, but not in regenerating liver, it seems unlikely thatthis enzyme is required for normal cellular proliferation. Itwould be of considerable interest to determine the exact func-tion(s) of these two enzymes. Does the tumor-type enzyme playa key role in tumorigenesis? Could it be a viable marker forcertain tumors? Do these enzymes polyadenylate specificclasses of mRNA, and if so, what is the factor(s) regulating ashift from the liver-type to the tumor-type poly(A) polymerase?Are liver-type and tumor-type poly(A) polymerases recognizedby specific cif-acting elements on mRNA precursors? Is thetumor-type poly(A) polymerase gene related to an oncogene?

Does posttranslational modification of poly(A) polymeraseaffect polyadenylation of pre-mRNAs. Finally, the kinetics ofthe cleavage/polyadenylation reaction should be examined ingreat detail. The various stages at which poly(A) polymeraseand other key components(s) participate in the processingshould be determined.

Acknowledgments

The authors thank Doris Lineweaver for her assistance in the preparation of this article.

2831




References

1. Birnsteil, M. I... Busslinger, M., and Striili. K. Transcription terminationand 3' processing: the end is in site! Cell, 41: 349-359, 1985.

2. Manley, J. L. Polyadenylation of mRNA precursors. Him him. Biophys. Acta,950:1-12,1988."

3. Nevins, J. R. The pathway of eukaryotic mRNA formation. Annu. Rev.Biochem.. 52:441-466, 1983.

4. Jacob, S. T.. and Rose, K. M. Poly(A) polymerase from eukaryotes. In: S. T.Jacob, (ed.). Enzymes of Nucleic Acid Synthesis and Modification, Vol. 2,pp. 135-157. Boca Raton, FL: CRC Press, Inc., 1983.

5. Edmonds, M. Poly(A) adding enzymes. In: P. D. Boyer (ed.). The Enzymes,Vol. 15, pp. 217-244. New York: Academic Press, 1982.

6. Rose. K. M.. and Jacob, S. T. Nuclear poly(A) polymerase from rat liver anda hepatoma: comparison of properties, molecular weights and amino acidcompositions. Eur. J. Biochem., 67: 11-21, 1976.

7. Niessing, J., and Sekeris, C. E. Nucleoside triphosphate polymerizing enzymeassociated with ribonucleoprotein particles containing DNA-like RNA. Nature (Lond.), 243:9-12, 1973.

8. Rose, K. M., Morris, H. P., and Jacob, S. T. Mitochondria! poly(A) polymerase from a poorly differentiated hepatoma: purification and characteristics.Biochemistry, 14: 1025-1032, 1975.

9. Rose, K. M., and Jacob, S. T. Poly(A) polymerase activity of rat livermicrosomes. J. Cell Biol., 67: 370, 1975.

10. Milchev, G. T., and Hadjilov, A. A. Association of poly(A) and poly(U)polymerase with cytoplasmic ribosomes. Eur. J. Biochem., 84: 113-121,1978.

11. Chung, C. W.. Mahler, H. R., and Enrione, M. Incorporation of adeninenucleotide into ribonucleic acid by cytoplasmic enzyme preparations of chickembryos. J. Biol. Chem., 235: 1448-1461, 1960.

12. Jacob, S. T., Roe, F. J., and Rose, K. M. Chromatin-bound and free formsof poly-(adenylic acid) polymerase in rat hepatic nuclei. Biochem. J., 153:733-735. 1976.

13. Rose, K. M., Roe, F. J., and Jacob, S. T. Two functional states of poly(A)polymerase in isolated nuclei. Biochim. Biophys. Acta, 478: 180-191, 1977.

14. Capone, G., and Drusiam. F. Sub-nuclear distribution of poly(A) polymeraseactivity in rat liver nuclei. Boll. Soc. Ital. Biol. Sper., 57: 246-252, 1981.

15. Yu, F. L. Rapid inhibition by cycloheximide of rat hepatic nuclear free andengaged poly(A) polymerase activities. Life Sci., 26: 11-17, 1980.

16. Orava, M. M.. Isomaa, V. V., and Janne, O. Nuclear poly(A) polymeraseactivities in the rabbit uterus. Regulation by progesterone administration andrelation to the activities of RNA polymerases and chromatin template. Eur.J. Biochem., 101:195-203, 1979.

17. Rose, K. M., Bell, L. E., and Jacob, S. T. Specific inhibition of chromatin-associated poly(A) synthesis in vitro by cordycepin 5'-triphosphate. Nature(Lond.), 267: 178-180, 1977.

18. Diez, .1.. and Brawerman, G. Elongation of the polyadenylate segment ofmessenger RNA in the cytoplasm of mammalian cells. Proc. Nati. Acad. Sci.USA, 77:4091-4095, 1974.

19. Mendecki, J., Lee, S. Y., and Brawerman, G. Characteristics of the poly-adenylic acid segment associated with messenger ribonucleic acid in mousesarcoma 180 ascites cells. Biochemistry, //: 792-798, 1972.

20. Rose, K. M., Bell, L. E., and Jacob, S. T. Selective inhibition of initialpolyadenylation reaction in isolated nuclei by cordycepin 5'-triphosphate.Biochim. Biophys. Acta, 475: 548-552, 1977.

21. Jacob, S. T., and Rose, K. M. RNA poly(A) polymerases and poly(A)polymerase from neoplastic tissues and cells. Methods Cancer Res., 14:220-241, 1978.

22. Edmonds, M., and Winters, M. A. Polyadenylate polymerases. Prog. NucÃ-.Acid Res., 17: 149-179, 1976.

23. Rose, K. M., and Jacob, S. T. Phosphorylation of poly(A) polymerase.Comparison between liver and hepatoma enzymes. J. Biol. Chem., 254:10256-10261, 1979.

24. Rose, K. M., and Jacob, S. T. Mechanism of enhanced polyadenylation ofmRNA by the phosphorylated poly(A) polymerase. Biochemistry, 19: 1472-1476, 1980.

25. Niessing, J. Three distinct forms of nuclear poly(A) polymerase. Eur. J.Biochem., 59: 127-135, 1975.

26. Casti, A., Corti, A., Reali, N., Mezzetti, G., Orlandini, G., and Caldarera, C.M. Modification of major aspects of myocardial ribonucleic acid metabolismas a response to adrenaline. Behaviour of polyadenylate polymerase andribonucleic acid polymerase, acetylation of histones and rate of synthesis ofpolyamines. Biochem. J., 168: 333-340, 1977.

27. Corti, A., Casti, A., Mezzetti, G., Reali, N., Orlandini, G., and Caldarera, C.M. Modifications of major aspects of myocardial ribonucleic acid metabolismas a response of noradrenaline. Action of the hormone on cytoplasmicprocessing of ribonucleic acid after reserpine treatment. Biochem. J., 168:341-345, 1977.

28. Jacob, S. T., Janne, O., and Rose, K. M. Corticosteroid-induced changes inRNA polymerases and poly(A) polymerase. In: Talwar, G. P. (ed.). Regulation of Growth and Differentiated Function in Eukaryote Cells, pp. 369-378. New York: Raven Press, 1975.

29. Orava, M. M., Isomaa, V. V., and Janne, O. Early changes in nucleoplasmicpoly(A) polymerase activity in immature rabbit uterus after estradici administration. Steroids, 36: 689-696, 1980.

30. Richter, R., and Schlimm. D. E. Age-associated changes in poly(adenylate)polymerase of rat liver and brain. Mech. Ageing Dev., 15: 217-225, 1981.

31. Mueller, W. E., /.han. R. K., Schroder, C. H., and Arendes. J. Age-dependent

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

64.

enzymatic poly(A) metabolism in quail oviduct. Gerontology, 25: 61-68,1979.Coleman, M. S., Hutton, J. J., and Bollimi. F. J. Terminal riboadenylatetransferase in human lymphocytes. Nature (Lond.), 248: 407-409, 1974.Geppinger, G., Broking, P., and Hardeland, R. Diurnal rhythmicity of nuclearand cytoplasmic polyadenylated ribonucleic acids and or polyadenylate-dependent polyadenylate polymerase in rat liver. Int. J. Chronobiol., 6: 23-30, 1979.Jacob, S. T., Rose, K. M., and Munro, H. N. Responses of polyadenylic acidpolymerases in rat liver nuclei and mitochondria to starvation and to refeed-ing with amino acids. Biochem. J., 158: 161-167, 1976.Berry, M., and Sachar, R. C. Hormonal regulation of poly(A) polymeraseactivity by gibberellic acid in embryo-less half-seeds of wheat. FEBS Lett.,132: 109-113, 1981.Schroder, H. C., Zahn, R. K., and Muller, W. E. Role of actin and tubulinin the regulation of poly(A) polymerase-endoribonuclease IV complex fromcalf thymus. J. Biol. Chem., 257: 2305-2309, 1982.Rose, K. M., Allen, M. S., Crawford, I. L., and Jacob, S. T. Functional roleof zinc in poly(A) synthesis catalyzed by nuclear poly(A) polymerase. Eur. J.Biochem., 88: 29-36, 1978.Stetler, D. A., and Jacob, S. T. Structurally and immunologically distinctpoly(A) polymerases in rat liver: occurrence of a tumor-type enzyme innormal liver. J. Biol. Chem., 259:7239-7244, 1984.Stetler, D. A., Seidel, B. L., and Jacob, S. T. Purification and characterizationof a nuclear protein kinase from rat liver and a hepatoma that is capable ofactivating poly(A) polymerase. J. Biol. Chem., 259: 14481-14485, 1984.Maguire, K. A., and Jacob, S. T. Cell-free synthesis of tumor-type poly(A)polymerase. Biochemistry, 25: 1515-1519, 1986.Jacob, S. T., and Rose, K. M. Phosphorylation and immunology and poly(A)polymerase. Adv. Enzyme Regul., 22:485-497, 1983.Stetler, D. A., Wrenshall, L. E., Park, Y., and Jacob, S. T. Induction of adistinct nuclear poly(A) polymerase in hepatocarcinogenesis. Carcinogenesis(Lond.), 5:1363-1366, 1984.Stonehill, E. H., and Bendich, A. Retrogenetic expression: the reappearanceof embryonal antigens in cancer cells. Nature (Lond.), 228:370-372, 1970.Sell, S., and Becker, F. F. a-Fetoprotein. J. Nati. Cancer Inst., 60: 19-26,1978.Stolbach, L. L., Melvin, J., Krant, M. J., and Fishman, W. H. Ectopieproduction of an alkaline phosphatase isoenzyme in patients with cancer. N.Engl. J. Med., 281: 757-762, 1969.Stetler, D. A., and Jacob, S. T. Structural and immunologie-ai identity of rat

hepatoma and fetal nuclear poly(A) polymerase. Carcinogenesis (Lond.), 6:259-262, 1985.Stetler, D. A., Rose, K. M., and Jacob, S. T. Anti-poly(A) polymeraseantibodies in sera of tumor-bearing rats and human cancer patients. Proc.Nati. Acad. Sci. USA, 78: 7732-7736, 1981.Tan, E. M. Autoantibodies to nuclear antigens (ANA): their immunobiologyand medicine. Adv. Immunol., 33: 167-240, 1982.Stetler, D. A., Reichlin, M., Berlin, C. M., and Jacob, S. T. Antibodiesagainst liver-type and tumor-type poly(A) polymerases in rheumatic autoimmune diseases. J. Clin. Immunol., 7: 24-28, 1987.Pillarisetty, R. J., and I alai. N. Clinical studies of antibodies binding poly-riboadenylic acid in systemic lupus erythematosus. Arth. Rheum., 19: 705-710, 1976.Moore, C. L., and Sharp, P. A. Site-specific polyadenylation in a cell-freereaction. Cell, 36: 581 -591, 1984.Moore, C. L., and Sharp, P. A. Accurate cleavage and polyadenylation ofexogenous RNA substrate. Cell, 41: 845-855, 1985.Hashimoto, C., and Steitz, J. A. A small nuclear ribonucleoprotein associateswith the AAUAAA polyadenylation signal in vitro. Cell, 45: 581-591, 1986.Wilusz, J., and Shenk, T. A 64 kd nuclear protein binds to RNA segmentsthat include that AAUAAA polyadenylation motif. Cell, 52: 221-228, 1988.Zarkower, D., and Wickens, M. Formation of mRNA 3' termini: stability

and dissociation of a complex involving the AAUAAA sequence. EMBO J.,6:177-186, 1987.Proudfoot, N. J., and Brownlee, G. G. 3' non-coding region sequences ineukaryotic messenger RNA. Nature (Lond.), 263: 211-214, 1976.Fitzgerald, M., and Shenk, T. The sequence 5'-AAUAAA-3' forms part ofthe recognition site for polyadenylation of late SV40 mRNAs. Cell, 24: 251-260,1981.Montell, C., Fisher, E. F., Caruthers, M. H., and Berk, A. J. Inhibition ofRNA cleavage but not polyadenylation by a point mutation in mRNA 3'consensus sequence AAUAAA. Nature (Lond.), 305: 600-605, 1983.Wickens, M., and Stephenson, P. Role of the conserved AAUAAA sequence:four AAUAAA point mutants prevent messenger RNA 3' end formation.Science (Wash. DC), 226:1045-1051, 1984.Manley, J. L., Yu, H., and Ryner, L. RNA sequence containing hexanucleo-tide AAUAAA directs efficient mRNA polyadenylation in vitro. Mol. Cell.Biol., 5: 373-379, 1985.Zarkower, D., Stephenson, P., Sheets, M., and Wickens, M. The AAUAAAsequence is required both for cleavage and for polyadenylation of simianvirus 40 pre-mRNA in vitro. Mol. Cell. Biol., 6: 2317-2323, 1986.Zhang, F., and Cole, C. N. Identification of a complex associated withprocessing and polyadenylation in vitro of herpes simplex virus type 1thymidine kinase precursor RNA. Mol. Cell. Biol., 7: 3277-3286, 1987.Skolnik-David, H., Moore, C. L., and Sharp, P. A. Electrophoretic separationof polyadenylation-specific complexes. Genes Dev., /: 672-682, 1987.Humphrey, T., Christofori, G., Lucijanic, V., and Keller, W. Cleavage and

2832




polyadenylation of messenger RNA precursors in vitro occurs within large required for specific RNA cleavage at a poly(A) addition site. Genes Dev., 2:and specific 3' processing complexes. EMBO J., 6: 4159-4168, 1987. 578-587, 1988.

65. McLauchlan, J., Moore, C. L., Simpson, S., and Clement, J. B. Components 73. Christofori, G., and Keller, W. 3' cleavage and polyadenylation of mRNA

required for in vitro cleavage and polyadenylation of eukaryotic mRNA. precursors in vitro requires a poly(A) polymerase cleavage factor and aNucÃ-.Acids Res., 16: 5323-5344, 1988. snRNP. Cell, 54: 875-889, 1988.

66. Zarkower, D., and Wickens, M. Specific pre-cleavage and post-cleavage 74- Raju, V. S., and Jacob, S. T. Association of poly(A) polymerase with Ulcomplexes involved in the formation of SV40 late mRNA 3 ' termini in vitro. RNA J- BlÂ°'-Chem., 263: 11067-11070, 1988.EMBOJ 6-4185-4192 1987 **â€¢Strub, K., and Birnsteil, M. L. Genetic complementation in the Xenopus

67. Moore, C Lâ€žSkoÃ-nik-David,H., and Sharp, P. A. Sedimentation analysis Ooc5"e: cÂ°â€ž7pressiÂ°nÂ°fsea "rchin ^Ã•Ã•.ÃŽÃ•lV'I ^ Ã¼Â£?T0L PrÂ°C"of polyadenylation-specific complexes. Mol. Cell. Biol., 8: 226-233, 1988. â€ž Â«â„¢gof H3 Pre-mRNA '" Â«Â«oocyte. EMBO J., 5: 1675-682 1986.

., c t < Ã- il. n r- A KI r i.i i -p ice 76- Mowry, K. L., and Steitz, J. A. Identification of a human U7 snRNP asonÃ©68. Stefano, J. E., and Adams, D. E. Assembly ofa polyadenylation-specific 25S Qf^ factors Â¡nvo,ve(1m ^ y ^ ma,uration of histone prernessengerribonucleoprotein complex in varo. Mol. Cell. Biol. Â«.-2052-2062, 1988. RNA,S Science (Wash DQ 23g. 1682_1687, 1987.

69. Hengst, J., Georgoff, I., Isom, H., and Jacob, S. T. Association of newly â€ž Ten]s M p an(j Jacob s T Ro,e of , (A) poiymerase in the cleavagesynthesized poly(A) polymerase with four distinct polypeptides. J. Biol. and po]yadeny|ation of mRNA precursor. Mol. Cell. Biol., 9: 1435-1444,Chem., 263:19270-19273, 1988. I989

70. Sperry, A. O., and Berget, S. M. In vitro cleavage of the simian virus 40 early 78 Takagaki, Y., Ryner, L. C., Manley, J. L. Separation and characterization ofpolyadenylation site adjacent to a required downstream TG sequence. Mol. a poly(A) polymerase and a cleavage/specificity factor required for pre-Cell. Biol., 6:4734-4741, 1986. mRNA polyadenylation. Cell, 52: 731-742, 1988.

71. Berget, S. M. Are U4 small nuclear ribonucleoproteins involved in polyade- 79. McDevitt, M. A., Gilmartin, G. M., Reeves, W. H., and Nevins, J. R.nylation? Nature (Lond.), 309: 179-182, 1984. Multiple factors are required for poly(A) addition to a mRNA 3' end. Genes

72. Gilmartin, G. M., McDevitt, M. A., and Nevins, J. R. Multiple factors are Dev., 2: 588-597, 1988.

2833



1989;49:2827-2833. Cancer Res Samson T. Jacob, Michael P. Terns and Kathleen A. Maguire Their Potential Role in Messenger RNA Processing: A ReviewPolyadenylate Polymerases from Normal and Cancer Cells and

Updated version

http://cancerres.aacrjournals.org/content/49/11/2827

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/49/11/2827To request permission to re-use all or part of this article, use this link


http://cancerres.aacrjournals.org/content/49/11/2827http://cancerres.aacrjournals.org/cgi/alertsmailto:[email protected]://cancerres.aacrjournals.org/content/49/11/2827http://cancerres.aacrjournals.org/

polyadenylate polymerases from normal and cancer cells and ... · zyme, which indicates that the...

Documents