Association of Poly(A) Polymerase with U1 RNA*

Previous studies (Stetler, D. A., and Jacob, S. T. (1984) J. Biol. Chem. 259, 7239-7244) have shown that poly(A) polymerase from adult rat liver (liver-type) is structurally and immunologically distinct from the corresponding rat hepatoma (tumor-type) enzyme. When hepatoma 7777 (McA-RH 7777) cells were labeled with [32P]inorganic phosphate, followed by immunoprecipitation with anti-hepatoma poly(A) polymerase antibodies and analysis of the RNAs in the immunoprecipitate, only one labeled small nuclear RNA corresponding to U1 RNA was found. Preimmune sera did not form a complex with U1 RNA. Hepatoma poly(A) polymerase antisera did not immunoprecipitate U1 RNA or any other small nuclear RNA from a cell line (H4-11-EC3) which does not contain the tumor-type poly(A) polymerase. Immunoblot analysis of hepatoma 7777 nuclear extract or purified poly(A) polymerase with anti-ribonucleoprotein antisera did not show any cross-reactivity of the latter sera with poly(A) polymerase. The major RNA immunoprecipitated from the hepatoma nuclear extracts using trimethyl cap (m3G) antisera corresponded to the RNA immunoprecipitated with poly(A) polymerase antisera. These data indicate that U1 RNA is closely associated with poly(A) polymerase and suggest the potential involvement of this RNA in the cleavage/polyadenylation of mRNA precursor.

In eukaryotes, specific endonucleolytic cleavage of pre-mRNAs and poly (A) addition at the cleavage site are required for the production of functional mRNAs (for reviews, see Refs. [1][2][3][4]. Perhaps the most convincing role for poly (A) is in the control of gene expression, as poly(A) addition at different sites on a single mRNA molecule can generate alternate forms of proteins (for review, see Ref. 5).
Considerable progress has been made in elucidating the signals for the cleavage and polyadenylation reactions. Two sequences in the pre-mRNA, the hexanucleotides AAUAAA or its modified form (Ref. 6 and references therein) which is located 10-30 nucleotides upstream of the poly(A) site and the GU-rich sequences downstream of the cleavage site ( 7 ) , are essential for accurate and efficient formation of 3' ends of mRNAs. Analysis of these reactions has shown that factors in the extract can interact selectively with the AAUAAA sequence within the mRNA precursor to form a precleavage * This work was supported by United States Public Health Service Grants CA 25078 and CA 31894 (to S. T. J.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. complex (8). Further studies in other laboratories have demonstrated that formation of the precleavage complex requires both the hexanucleotide sequence and the sequence(s1 downstream of the cleavage site (9)(10)(11)(12) and that the critical elements of the downstream region are either diffuse or redundant (13).
In contrast to the progress made in the elucidation of the nucleotide sequences obligatory to the endonucleolytic cleavage and poly(A) addition, the biochemical nature of these reactions is not fully understood. Since anti-Sm or anti-(Ul) RNP' antisera or antisera to the nuclear antigen La can inhibit site-specific polyadenylation of adenovirus L3 mRNA, the possible involvement of one or more small nuclear RNAs (snRNAs) in mRNA polyadenylation has been suggested. Probing in vitro polyadenylation reactions with antibodies specific for snRNPs has shown that a factor with the properties of an Sm snRNP is associated with the AAUAAA polyadenylation signal (8). Subsequent studies (14) have shown that pretreatment of the nuclear extract with micrococcal nuclease can inhibit cleavage and polyadenylation but could be restored by the addition of purified Escherichia coli RNA. Partial cleavage of U1, U2, or U4 RNAs did not inhibit polyadenylation (15). Since none of these treatments results in complete degradation of snRNAs, it remains a possibility that other regions of the snRNA(s) may play a role in the cleavage and/or polyadenylation (15).
Our laboratory has been involved in the characterization of poly(A) polymerases (16)(17)(18). Antibodies raised against nuclear poly(A) polymerase from a rat hepatoma have been shown to form a distinct immune complex in immunoblot analysis using relatively crude nuclear extract or a purified enzyme preparation (19). Addition of these antibodies to an in vitro polyadenylation system has resulted in inhibition of polyadenylation of adenovirus L3 mRNA whereas control serum had no effect indicating a direct role of poly(A) polymerase in this post-transcriptional reaction.' To determine whether poly(A) polymerase is associated with any snRNA, 32P-labeled nuclear extracts derived from H4 or 7777 hepatoma cells were immunoprecipitated with anti-hepatoma poly(A) polymerase antibodies, and the immune complex was analyzed for snRNAs. The only snRNA in the complex was U1 RNA. Preparation of Nuclear Extract-Cells were seeded at a concentration rate that allowed them to attain confluency in 5-6 days. At 75% confluency, the cells were grown in RPMI 1640 phosphate-free medium for 4 h. The cells were then grown in RPMI 1640 phosphatefree media containing [32P]inorganic phosphate (100 pCi/dish) for 16 h. The cells were washed with PBS twice and collected by scraping with rubber cell policeman, and nuclear extract was prepared essen-

Association of Poly(A) Polymerase
with Ul RNA tially as described (20), except that the cells were homogenized in 2 ml or 2 volumes of solution A. Sera-Antinuclear antibodies reference human sera against U1 RNP and snRNP were obtained from t.he immunology branch of the Center for Disease Control, Atlanta, GA. Rabbit antisera raised against hepatoma poly(A) polymerase were used in these studies. The rabbit antisera were purified on DEAE Affi-Gel blue as specified by the manufacturer's recommendations. The flow-through fraction was saturated with ammonium sulfate and dialyzed against 0.02 M Tris-HCl, pH 8.0 and stored at -20 "C.
Immunoblot Analysis-Aliquots of nuclear extract or hepatoma poly(A) polymerase purified essentially to homogeneity (16,19) were precipitated with equal volumes of acetone on ice for 1 h. The pellets were washed once with 100 pl of acetone and dried under vacuum in a Speedvac for 5 min. The proteins were dissolved in SDS gel sample buffer, heated for 10 min at 90 "C and subjected to electrophoresis on 10% SDS-polyacrylamide gel as described (21), and transferred to nitrocellulose sheets (Hybond C, Amersham Corp.) by electrophoresis at room temperature at 100 mA for 13-15 h in a buffer containing 25 mM Tris-HC1, pH 8.3, 0.15 M glycine, and 10% methanol. Antigenic proteins were probed with different antisera as described (22) with some modifications using a Vector ABC staining kit. Marker lines were cut and stained with Amido Black. Nitrocellulose sheets were incubated with PBS containing 5% dry milk powder for 1-2 h at room temperature to block the nonspecific binding. The sheets were washed with PBS for 5 min, and the sheets were then incubated with anti-poly(A) polymerase antisera (1:25 dilution), human control sera, or anti-U1 antisera (1:lOO) in PBS-milk for 12-15 h. The blots were washed five times with PBS and incubated with bioadenylated goat antirabbit IgG (1:400 dilution) in PBS-milk for 2 h at room temperature. After five washes with PBS, the sheets were incubated with avidin horseradish peroxidase DH reagents from the Vectastain ABC kit for 30 min at room temperature. The blots were washed once for 5 min with PBS containing 0.1% Triton X-100 and four times with PBS. The blots were then incubated with 100 ml of PBS solution containing 60 mg of 4-chloro-1-naphthol and 60 p1 of 30% hydrogen peroxide.
Analysis of Immunoprecipitated RNA-The radiolabeled nuclear extracts were incubated for 30 min with 0.5 volumes of 10% Pansorbin solution (Calbiochem) in NET-2 (150 mM NaCl, 50 mM Tris-HC1, pH 7.5,0.05% Nonidet P-40) at 0 "C (23) and centrifuged to remove Pansorbin. The immunoprecipitation was carried out with 10-20 p1 of various antibodies coated onto 3 mg of protein A-Sepharose (Sigma) as described (23). The antibody-coated protein A-Sepharose beads were incubated at room temperature with 300 pl of nuclear extract (obtained from approximately 1.42 X 10' hepatoma cells) for 2 h. The beads were suspended in 100 pl of NET after three washes with NET-2 buffer (500 p1 each time), 50 pl of 2 M sodium acetate containing 500 pg/ml carrier tRNA and 350 p1 of 8 M guanidine HCl in 50 mM Tris-HC1, pH 7.5, 10 mM EDTA. The samples were mixed well and centrifuged for 5 min to remove Sepharose beads. RNA was extracted with 300 pl of chilled absolute ethanol as described (24) and subjected to electrophoresis on 10% polyacrylamide, 7 M urea gels and detected by autoradiography.

Antibodies Raised against Poly(A) Polymerase from Morris Hepatoma 3924A React with the Corresponding Enzyme from Hepatoma 7777 Cells, but Not with the Enzyme from the Cell
Line H4-Previous studies in our laboratory have demonstrated that nuclear poly(A) polymerase from Morris hepatoma 3924A (a transplanted solid tumor) is structurally distinct from the corresponding rat liver enzyme and that antibodies raised against the hepatoma enzyme do not react with the liver enzyme. It was, therefore, crucial to determine whether the anti-hepatoma poly(A) polymerase antibodies will react with the enzyme from the 7777 hepatoma cell line used for labeling the RNA but not with the enzyme from a cell line that exhibits properties of normal hepatocytes (25). To test this possibility, the proteins in the nuclear extracts from the cell line H4 and hepatoma 7777 cells were separated on 10% SDS-polyacrylamide gels, transferred onto nitrocellulose paper, and probed with antisera raised against hepatoma 3924A poly(A) polymerase (for details of the immuno-blot analysis, see "Materials and Methods"). The cell line derived from another Morris hepatoma instead of the hepatoma 3924A was selected for the present studies, as a suitable cell line from the latter solid tumor was not available. The nuclear extracts from the hepatoma 7777 cells (Fig. 1, lane 2 ) formed a distinct immune complex corresponding to a molecular weight of 48,000. A similar complex was also formed with highly purified poly(A) polymerase from hepatoma 3924A used as the control (lane 3 ) . Under these conditions, none of the proteins from the cell line H4 reacted with the antibodies (lane 1 ). The preimmune control sera did not react with either tumor enzyme (lanes 4-6). These data demonstrated the specificity of the anti-poly(A) polymerase antibodies and allowed their use in immunoprecipitation using the hepatoma 7777 cell line.

Anti-poly(A) Polymerase Antibodies Precipitate U l RNA-
Previous studies in other laboratories have indirectly suggested a potential role of small nuclear RNAs (26) H4 cells (lanes 1 and 4 ) , hepatoma 7777 cells (lanes 2 and 5 ) , highly purified tumor-type poly (A) polymerase (lanes 3 and 6 ) , and marker proteins (lane M ) were fractionated on 10% SDS-polyacrylamide gel, transferred to nitrocellulose paper, and probed with either antihepatoma poly (A) polymerase antisera (lanes 1, 2, and 3 ) or preimmune sera (lanes 4, 5, and 6  incubation with the antibodies raised against purified poly (A) polymerase from the hepatoma 3924A. RNA was extracted and analyzed on a high percentage gel to visualize only the low molecular weight RNAs as described under "Materials and Methods." Only one 32P-labeled band corresponding to a size of 172 nucleotides was precipitated (Fig. 2, lane 2) whereas no detectable RNA was observed when preimmune sera were used for precipitation (lane 3 ) . To determine the nature of the RNA associated with poly(A) polymerase, human RNP antisera that are known to react only with U1 snRNA (27) were used as a control. As anticipated, the human sera precipitated only U1 RNA which corresponded exactly to the RNA contained in the immune complex with anti-poly(A) polymerase antibodies (lane 1 ). Since all snRNAs contain a 5' cap (m3G) structure, antibodies against this structure should immunoprecipitate the RNA associated with poly(A) polymerase. To test this possibility, the nuclear extract from 32P-  labeling the RNA and immunoprecipitation. If U1 RNA from the H4 cell line is associated with the liver-type poly (A) polymerase, antibodies against the tumor enzyme should not precipitate U1 RNA from this cell line. No detectable RNA was observed following analysis of the immunoprecipitate whereas the anti-RNP sera precipitated U1 RNA (Fig. 3, lane  2). Another sera, human Sm antisera that are known to react with U1, U2, U4, U5, and U6 snRNAs precipitated all these  (lanes 2 and 4 ) were fractionated on 10% SDS-polyacrylamide gel and transferred to nitrocellulose paper as described under "Materials and Methods." The nitrocellulose blots were probed with anti-RNP antisera (lanes 1 and 2 ) or with normal human sera (lanes 3 and 4 ) . Lane M corresponds to markers ( a = 97,000; b = 67,000; c = 43,000; e = 20,000; and f = 14,000 daltons).
2 and 4 ) . Several proteins in nuclear extract reacted with both sera, suggesting nonspecific binding. It can be concluded from these observations that poly(A) polymerase is tightly associated with U1 RNA. The size of the RNA, determined by plotting distance migrated in millimeters uersus log (nucleotides) of 4x174 DNA markers is 172 nucleotides which corresponds to the length of U l a RNA (27). Although cleavage of RNAs does not appear to inhibit addition of poly(A) to mRNA (14,15), the lack of complete degradation of the snRNA under these conditions does not preclude the potential role of any one of these molecules in the polyadenylation reaction (15). The association of U-type snRNP with the 50 S polyadenylation complex, identified by density gradient fractionation (28), is consistent with a possible involvement of snRNA in mRNA polyadenylation. Although our studies have not directly demonstrated the role of U l a snRNA in the 3' end processing of adenovirus L3 mRNA, the specific association of this RNA with poly(A) polymerase suggests that it is involved in some aspect of cleavage/polyadenylation. Since anti-poly(A) polymerase antibodies can inhibit polyadenylation of adenovirus L3 mRNA,* U l a RNA is probably associated with a functional poly(A) polymerase in the polyadenylation complex.