The endoplasmic reticulum-Golgi intermediate compartment.

The recent identification of an endoplasmic reticulum-Golgi intermediate compartment has added to the complexity of the structural and functional organization of the early secretory pathway. Protein sorting along the endoplasmic reticulum-Golgi pathway depends on different signals and mechanisms, some of which guarantee recycling from various levels of the Golgi apparatus to biosynthetically earlier compartments.

The lectin ERGIC-53 is a cargo transport receptor for glycoproteins.

Soluble secretory proteins are transported from the endoplasmic reticulum (ER) to the ER-Golgi intermediate compartment (ERGIC) in vesicles coated with COP-II coat proteins. The sorting of secretory cargo into these vesicles is thought to involve transmembrane cargo-receptor proteins. Here we show that a cathepsin-Z-related glycoprotein binds to the recycling, mannose-specific membrane lectin ERGIC-53. Binding occurs in the ER, is carbohydrate- and calcium-ion-dependent and is affected by untrimmed glucose residues. Binding does not, however, require oligomerization of ERGIC-53, although oligomerization is required for exit of ERGIC-53 from the ER. Dissociation of ERGIC-53 occurs in the ERGIC and is delayed if ERGIC-53 is mislocalized to the ER. These results strongly indicate that ERGIC-53 may function as a receptor facilitating ER-to-ERGIC transport of soluble glycoprotein cargo.

Targeting of protein ERGIC-53 to the ER/ERGIC/cis-Golgi recycling pathway.

ERGIC-53 is a lectin-type membrane protein that continuously recycles between the ER, ER-Golgi intermediate compartment (ERGIC) and the cis-Golgi. To identify the targeting signals that mediate this recycling, N-glycosylated and myc-tagged variants of ERGIC-53 were constructed. By monitoring endoglycosidase H resistance, we measured the loss from the ER-ERGIC-cis-Golgi cycle of ERGIC-53. A domain exchange approach with the plasma membrane reporter protein CD4 showed that the transmembrane and the lumenal domains are not sufficient, while the cytoplasmic domain of ERGIC-53 is required and sufficient for pre-medial-Golgi localization. However, the ERGIC-53 cytoplasmic domain on CD4 lead to increased ER-staining by immunofluorescence microscopy indicating that this domain alone cannot provide for unbiased recycling through the ER-ERGIC-cis-Golgi compartments. Complete progress through the ER-ERGIC-cis-Golgi recycling pathway requires the cytoplasmic domain acting together with the lumenal domain of ERGIC-53. Dissection of the cytoplasmic domain revealed a COOH-terminal di-lysine ER-retrieval signal, KKFF, and an RSQQE targeting determinant adjacent to the transmembrane domain. Surprisingly, the two COOH-terminal phenylalanines influence the targeting. They reduce the ER-retrieval capacity of the di-lysine signal and modulate the RSQQE determinant.

Nocodazole, a microtubule-active drug, interferes with apical protein delivery in cultured intestinal epithelial cells (Caco-2).

The polarized delivery of membrane proteins to the cell surface and the initial secretion of lysosomal proteins into the culture medium were studied in the polarized human intestinal adenocarcinoma cell line Caco-2 in the presence or absence of the microtubule-active drug nocodazole. The appearance of newly synthesized proteins at the plasma membrane was measured by their sensitivity to proteases added either to the apical or the basolateral surface of cells grown on nitrocellulose filters. Nocodazole was found to reduce the delivery to the cell surface of an apical membrane protein, aminopeptidase N, and to lead to its partial missorting to the basolateral surface, whereas the drug had no influence on the delivery of a basolateral 120-kD membrane protein defined by a monoclonal antibody. Furthermore, nocodazole selectively blocked the apical secretion of two lysosomal proteins, cathepsin D and acid alpha-glucosidase, whereas the drug had no influence on their basolateral secretion. These results suggest that in Caco-2 cells an intact microtubular network is important for the transport of newly synthesized proteins to the apical cell surface.

Retention of p63 in an ER-Golgi intermediate compartment depends on the presence of all three of its domains and on its ability to form oligomers.

The type II membrane protein p63 is a resident protein of a membrane network interposed between rough ER and Golgi apparatus. To study the retention of p63, mutant forms were expressed in COS cells and the intracellular distribution determined by immunofluorescence microscopy. Investigation of chimeric constructs between p63 and the plasma membrane protein dipeptidylpeptidase IV showed that protein sequences from all three domains of the p63 protein are required to achieve complete intracellular retention. Mutational analysis of the 106-amino acid cytoplasmic tail of p63 revealed that the NH2-terminal 23 amino acids are necessary for retention. When p63 was solubilized with Triton X-100 and subjected to centrifugation at 100,000 g, it formed large, insoluble oligomers, particularly at neutral pH and below. A comparison of the behavior of wildtype and mutant p63 proteins in this assay revealed a perfect correlation between the formation of large oligomers and correct intracellular retention. These results suggest that self-association may be a major mechanism by which p63 is retained between the rough ER and the Golgi apparatus.

The isolated ER-Golgi intermediate compartment exhibits properties that are different from ER and cis-Golgi.

A procedure has been established in Vero cells for the isolation of an intermediate compartment involved in protein transport from the ER to the Golgi apparatus. The two-step subcellular fractionation procedure consists of Percoll followed by Metrizamide gradient centrifugation. Using the previously characterized p53 as a marker protein, the average enrichment factor of the intermediate compartment was 41. The purified fraction displayed a unique polypeptide pattern. It was largely separated from the rough ER proteins ribophorin I, ribophorin II, BIP, and protein disulfide isomerase, as well as from the putative cis-Golgi marker N-acetylglucosamine-1-phosphodiester-alpha-N-acetylglucosaminidase, the second of the two enzymes generating the lysosomal targeting signal mannose-6-phosphate. The first enzyme, N-acetylglucosaminylphosphotransferase, for which previous biochemical evidence had suggested both a pre- and a cis-Golgi localization in other cell types, cofractionated with the cis-Golgi rather than the intermediate compartment in Vero cells. The results suggest that the intermediate compartment defined by p53 has unique properties and does not exhibit typical features of rough ER and cis-Golgi.

Mistargeting of the lectin ERGIC-53 to the endoplasmic reticulum of HeLa cells impairs the secretion of a lysosomal enzyme.

ERGIC-53, a homo-oligomeric recycling protein associated with the ER-Golgi intermediate compartment (ERGIC), has properties of a mannose-selective lectin in vitro, suggesting that it may function as a transport receptor for glycoproteins in the early secretory pathway. To investigate if ERGIC-53 is involved in glycoprotein secretion, a mutant form of this protein was generated that is incapable of leaving the ER. If expressed in HeLa cells in a tetracycline-inducible manner, this mutant accumulated in the ER and retained the endogenous ERGIC-53 in this compartment, thus preventing its recycling. Mistargeting of ERGIC-53 to the ER did not alter the gross morphology of the early secretory pathway, including the distribution of beta'-COP. However, it impaired the secretion of one major glycoprotein, identified as the precursor of the lysosomal enzyme cathepsin C, while overexpression of wild-type ERGIC-53 had no effect on glycoprotein secretion. Transport of two other lysosomal enzymes and three post-Golgi membrane glycoproteins was unaffected by inactivating the recycling of ERGIC-53. The results suggest that the recycling of ERGIC-53 is required for efficient intracellular transport of a small subset of glycoproteins, but it does not appear to be essential for the majority of glycoproteins.

The Emp24 complex recruits a specific cargo molecule into endoplasmic reticulum-derived vesicles.

Members of the yeast p24 family, including Emp24p and Erv25p, form a heteromeric complex required for the efficient transport of selected proteins from the endoplasmic reticulum (ER) to the Golgi apparatus. The specific functions and sites of action of this complex are unknown. We show that Emp24p is directly required for efficient packaging of a lumenal cargo protein, Gas1p, into ER-derived vesicles. Emp24p and Erv25p can be directly cross-linked to Gas1p in ER-derived vesicles. Gap1p, which was not affected by emp24 mutation, was not cross-linked. These results suggest that the Emp24 complex acts as a cargo receptor in vesicle biogenesis from the ER.

(Na+ + K+)-ATPase and plasma membrane polarity of intestinal epithelial cells: presence of a brush border antigen in the distal large intestine that is immunologically related to beta subunit.

The previously produced monoclonal antibody IEC 1/48 against cultured rat intestinal crypt cells (Quaroni, A., and K. J. Isselbacher. 1981. J. Natl. Cancer Inst. 67:1353-1362) was extensively characterized and found to be directed against the beta subunit of (Na+ + K+)-ATPase as assessed by immunological and enzymatic criteria. Under nondenaturing conditions the antibody precipitated the alpha-beta enzyme complex (98,000 and 48,000 Mr). This probe, together with the monoclonal antibody C 62.4 against the alpha subunit (Kashgarian, M., D. Biemesderfer, M. Caplan, and B. Forbush. 1985. Kidney Int. 28:899-913), was used to localize (Na+ + K+)-ATPase in epithelial cells along the rat intestinal tract by immunofluorescence and immunoelectron microscopy. Both antibodies exclusively labeled the basolateral membrane of small intestine and proximal colon epithelial cells. However, in the distal colon, IEC 1/48, but not C 62.4, also labeled the brush border membrane. The cross-reacting beta-subunit-like antigen on the apical cell pole was tightly associated with isolated brush borders but was apparently devoid of (Na+ + K+)-ATPase activity. Subcellular fractionation of colonocytes in conjunction with limited proteolysis and surface radioiodination of intestinal segments suggested that the cross-reacting antigen in the brush border may be very similar to the beta subunit. The results support the notion that in the small intestine and proximal colon the enzyme subunits are exclusively targeted to the basolateral membrane while in the distal colon nonassembled beta subunit or a beta-subunit-like protein is also transported to the apical cell pole.

Expression and intracellular transport of microvillus membrane hydrolases in human intestinal epithelial cells.

A panel of monoclonal antibodies was produced against purified microvillus membranes of human small intestinal enterocytes. By means of these probes three disaccharidases (sucrase-isomaltase, lactase-phlorizin hydrolase, and maltase-glucoamylase) and four peptidases (aminopeptidase N, dipeptidylpeptidase IV, angiotension I-converting enzyme, and p-aminobenzoic acid peptide hydrolase) were successfully identified as individual entities by SDS PAGE and localized in the microvillus border of the enterocytes by immunofluorescence microscopy. The antibodies were used to study the expression of small intestinal hydrolases in the colonic adenocarcinoma cell line Caco 2. This cell line was found to express sucrase-isomaltase, lactase-phlorizin hydrolase, aminopeptidase N, and dipeptidylpeptidase IV, but not the other three enzymes. Pulse-chase studies with [35S]methionine and analysis by subunit-specific monoclonal antibodies revealed that sucrase-isomaltase was synthesized and persisted as a single-chain protein comprising both subunits. Similarly, lactase-phlorizin hydrolase was synthesized as a large precursor about twice the size of the lactase subunits found in the human intestine. Aminopeptidase N and dipeptidylpeptidase IV, known to be dimeric enzymes in most mammals, were synthesized as monomers. Transport from the rough endoplasmic reticulum to the trans-Golgi apparatus was considerably faster for the peptidases than for the disaccharidases, as probed by endoglycosidase H sensitivity. These results suggest that the major disaccharidases share a common biosynthetic mechanism that differs from that for peptidases. Furthermore, the data indicate that the transport of microvillus membrane proteins to and through the Golgi apparatus is a selective process that may be mediated by transport receptors.

Naturally occurring mutations in intestinal sucrase-isomaltase provide evidence for the existence of an intracellular sorting signal in the isomaltase subunit.

Mutations in the sucrase-isomaltase gene can lead to the synthesis of transport-incompetent or functionally altered enzyme in congenital sucrase-isomaltase deficiency (CSID) (Naim, H. Y., J. Roth, E. Sterchi, M. Lentze, P. Milla, J. Schmitz, and H. P. Hauri. J. Clin. Invest. 82:667-679). In this paper we have characterized two novel mutant phenotypes of CSID at the subcellular and protein levels. The first phenotype revealed a sucrase-isomaltase protein that is synthesized as a single chain, mannose-rich polypeptide precursor (pro-SI) and is electrophoretically indistinguishable from pro-SI in normal controls. By contrast to normal controls, however, pro-SI does not undergo terminal glycosylation in the Golgi apparatus. Subcellular localization of pro-SI by immunoelectron microscopy revealed unusual labeling of the molecule in the basolateral membrane and no labeling in the brush border membrane thus indicating that pro-SI is missorted to the basolateral membrane. Mapping of biosynthetically labeled pro-SI with four epitope- and conformation-specific monoclonal antibodies suggested that conformational and/or structural alterations in the pro-SI protein have prevented posttranslational processing of the carbohydrate chains of the mannose-rich precursor and have lead to its missorting to the basolateral membrane. The second phenotype revealed two variants of pro-SI precursors that differ in their content of mannose-rich oligosaccharides. Conversion of these forms to a complex glycosylated polypeptide occurs at a slow rate and is incomplete. Unlike its counterpart in normal controls, pro-SI in this phenotype is intracellularly cleaved. This cleavage produces an isomaltase-like subunit that is transport competent and is correctly sorted to the brush border membrane since it could be localized in the brush border membrane by anti-isomaltase mAb. The sucrase subunit is not transported to the cell surface and is most likely degraded intracellularly. We conclude that structural features in the isomaltase region of pro-SI are required for transport and sorting of the sucrase-isomaltase complex.

Selective reentry of recycling cell surface glycoproteins to the biosynthetic pathway in human hepatocarcinoma HepG2 cells.

Return of cell surface glycoproteins to compartments of the secretory pathway has been examined in HepG2 cells comparing return to the trans-Golgi network (TGN), the trans/medial- and cis-Golgi. Transport to these sites was studied by example of the transferrin receptor (TfR) and the serine peptidase dipeptidylpeptidase IV (DPPIV) after labeling these proteins with the N-hydroxysulfosuccinimide ester of biotin on the cell surface. This experimental design allowed to distinguish between glycoproteins that return to these biosynthetic compartments from the cell surface and newly synthesized glycoproteins that pass these compartments during biosynthesis en route to the surface. Reentry to the TGN was measured in that surface glycoproteins were desialylated with neuraminidase and were monitored for resialylation during recycling. Return to the trans-Golgi was traced measuring the transfer of [3H]fucose residues to recycling surface proteins by fucosyltransferases. To study return to the cis-Golgi, surface proteins were metabolically labeled in the presence of the mannosidase I inhibitor deoxymannojirimycin (dMM). As a result surface proteins retained N-glycans of the oligomannosidic type. Return to the site of mannosidase I in the medial/cis-Golgi was measured monitoring conversion of these glycans to those of the complex type after washout of dMM. Our data demonstrate that DPPIV does return from the cell surface not only to the TGN, but also to the trans-Golgi thus linking the endocytic to the secretory pathway. In contrast, no reentry to sites of mannosidase I could be detected indicating that the early secretory pathway is not or is only at insignificant rates accessible to recycling DPPIV. In contrast to DPPIV, TfR was very efficiently sorted from endosomes to the cell surface and did not return to the TGN or to other biosynthetic compartments in detectable amounts, indicating that individual surface proteins are subject to different sorting mechanisms or sorting efficiencies during recycling.

Dissection of the asynchronous transport of intestinal microvillar hydrolases to the cell surface.

Novel subcellular fractionation procedures and pulse-chase techniques were used to study the intracellular transport of the microvillar membrane hydrolases sucrase-isomaltase and dipeptidylpeptidase IV in the differentiated colon adenocarcinoma cell line Caco-2. The overall rate of transport to the cell surface was two fold faster for dipeptidylpeptidase IV than for sucrase-isomaltase, while no significant differences were observed in transport rates from the site of complex glycosylation to the brush border. The delayed arrival of sucrase-isomaltase in the compartment where complex glycosylation occurs was only in part due to exit from the endoplasmic reticulum. A major slow-down could be ascribed to maturation in and transit of this enzyme through the Golgi apparatus. These results suggest that the observed asynchronism is due to more than one rate-limiting step along the rough endoplasmic reticulum to trans-Golgi pathway.

Identification, by a monoclonal antibody, of a 53-kD protein associated with a tubulo-vesicular compartment at the cis-side of the Golgi apparatus.

Purified Golgi membranes of the human intestinal adenocarcinoma cell line Caco-2 were used as an antigen to produce a monoclonal antibody, G1/93, which specifically labels a tubulovesicular compartment near the cis side of the Golgi apparatus, including the first cis-cisterna itself, as visualized by single and double immunoelectron microscopy with antibodies against galactosyltransferase. The antigen recognized by G1/93 was identified as a protein with a subunit size of 53 kD. Pulse-chase experiments revealed that the 53-kD protein dimerizes immediately after synthesis followed by formation of oligomers of approximately 310 kD, probably homohexamers. The protein has a transmembrane topology with only a short cytoplasmic segment as assessed by protease protection experiments. Glycosidase digestion studies indicated that the protein is probably not glycosylated. The unique subcellular distribution of the G1/93 antigen in close vicinity to the cis-Golgi is in line with the notion that this protein may delineate the biosynthetic transport pathway from the endoplasmic reticulum to the Golgi apparatus. Moreover, G1/93 is a useful marker to identify the cis side of the Golgi apparatus in a variety of human cells.

A novel marker glycoprotein for the microvillus membrane of surface colonocytes of rat large intestine and its presence in small-intestinal crypt cells.

Murine mAbs were produced against purified microvillus membranes of rat colonocytes in order to establish a marker protein for this membrane. The majority of antibodies binding to the colonic microvillus membrane recognized a single protein with a mean apparent Mr of 120 kD in both proximal and distal colon samples. The antigen is membrane bound as probed by phase-partitioning studies using Triton X-114 and by the sodium carbonate extraction procedure and is extensively glycosylated as assessed by endoglycosidase F digestion. Localization studies in adult rats by light and electron microscopy revealed the microvillus membrane of surface colonocytes as the principal site of the immunoreaction. The antigen was not detectable in kidney or liver by immunoprecipitation but was present in the small intestine, where it was predominantly confined to the apical membrane of crypt cells and much less to the microvillus membrane of differentiated enterocytes. During fetal development, the antigen appears first in the colon at day 15 and 1-2 d later in the small intestine. In both segments, it initially covers the whole luminal surface but an adult-like localization pattern develops soon after birth. The antibodies were also used to develop a radiometric assay for the quantification of the antigen in subcellular fractions of colonocytes in order to assess the validity of a previously developed method for the purification of colonic brush-border membranes (Stieger, B., A. Marxer, and H.P. Hauri. 1986. J. Membr. Biol. 91:19-31.). The results suggest that we have identified a valuable marker glycoprotein for the colonic microvillus membrane, which in adult rats may also serve as a marker for early differentiation of enterocyte progenitor cells in small-intestinal crypt cells.

The apical submembrane cytoskeleton participates in the organization of the apical pole in epithelial cells.

In a previous publication (Rodriguez, M.L., M. Brignoni, and P.J.I. Salas. 1994. J. Cell Sci. 107: 3145-3151), we described the existence of a terminal web-like structure in nonbrush border cells, which comprises a specifically apical cytokeratin, presumably cytokeratin 19. In the present study we confirmed the apical distribution of cytokeratin 19 and expanded that observation to other epithelial cells in tissue culture and in vivo. In tissue culture, subconfluent cell stocks under continuous treatment with two different 21-mer phosphorothioate oligodeoxy nucleotides that targeted cytokeratin 19 mRNA enabled us to obtain confluent monolayers with a partial (40-70%) and transitory reduction in this protein. The expression of other cytoskeletal proteins was undisturbed. This downregulation of cytokeratin 19 resulted in (a) decrease in the number of microvilli; (b) disorganization of the apical (but not lateral or basal) filamentous actin and abnormal apical microtubules; and (c) depletion or redistribution of apical membrane proteins as determined by differential apical-basolateral biotinylation. In fact, a subset of detergent-insoluble proteins was not expressed on the cell surface in cells with lower levels of cytokeratin 19. Apical proteins purified in the detergent phase of Triton X-114 (typically integral membrane proteins) and those differentially extracted in Triton X-100 at 37 degrees C or in n-octyl-beta-D-glycoside at 4 degrees C (representative of GPI-anchored proteins), appeared partially redistributed to the basolateral domain. A transmembrane apical protein, sucrase isomaltase, was found mispolarized in a subpopulation of the cells treated with antisense oligonucleotides, while the basolateral polarity of Na+-K+ATPase was not affected. Both sucrase isomaltase and alkaline phosphatase (a GPI-anchored protein) appeared partially depolarized in A19 treated CACO-2 monolayers as determined by differential biotinylation, affinity purification, and immunoblot. These results suggest that an apical submembrane cytoskeleton of intermediate filaments is expressed in a number of epithelia, including those without a brush border, although it may not be universal. In addition, these data indicate that this structure is involved in the organization of the apical region of the cytoplasm and the apical membrane.

Subdomain-specific localization of CLIMP-63 (p63) in the endoplasmic reticulum is mediated by its luminal alpha-helical segment.

The microtubule-binding integral 63 kD cytoskeleton-linking membrane protein (CLIMP-63; former name, p63) of the rough endoplasmic reticulum (ER) is excluded from the nuclear envelope. We studied the mechanism underlying this ER subdomain-specific localization by mutagenesis and structural analysis. Deleting the luminal but not cytosolic segment of CLIMP-63 abrogated subdomain-specific localization, as visualized by confocal microscopy in living cells and by immunoelectron microscopy using ultrathin cryosections. Photobleaching/recovery analysis revealed that the luminal segment determines restricted diffusion and immobility of the protein. The recombinant full-length luminal segment of CLIMP-63 formed alpha-helical 91-nm long rod-like structures as evident by circular dichroism spectroscopy and electron microscopy. In the analytical ultracentrifuge, the luminal segment sedimented at 25.7 S, indicating large complexes. The complexes most likely arose by electrostatic interactions of individual highly charged coiled coils. The findings indicate that the luminal segment of CLIMP-63 is necessary and sufficient for oligomerization into alpha-helical complexes that prevent nuclear envelope localization. Concentration of CLIMP-63 into patches may enhance microtubule binding on the cytosolic side and contribute to ER morphology by the formation of a protein scaffold in the lumen of the ER.

Sucrase-isomaltase deficiency in humans. Different mutations disrupt intracellular transport, processing, and function of an intestinal brush border enzyme.

Eight cases of congenital sucrase-isomaltase deficiency were studied at the subcellular and protein level with monoclonal antibodies against sucrase-isomaltase. At least three phenotypes were revealed: one in which sucrase-isomaltase protein accumulated intracellularly probably in the endoplasmic reticulum, as a membrane-associated high-mannose precursor, one in which the intracellular transport of the enzyme was apparently blocked in the Golgi apparatus, and one in which catalytically altered enzyme was transported to the cell surface. All patients expressed electrophoretically normal or near normal high-mannose sucrase-isomaltase. The results suggest that different, probably small, mutations in the sucrase-isomaltase gene lead to the synthesis of transport-incompetent or functionally altered enzyme which results in congenital sucrose intolerance.

Biogenesis of intestinal lactase-phlorizin hydrolase in adults with lactose intolerance. Evidence for reduced biosynthesis and slowed-down maturation in enterocytes.

Enzymatic activity, biosynthesis, and maturation of lactasephlorizin hydrolase (LPH) were investigated in adult volunteers with suspected lactose intolerance. Mean LPH activity in jejunal biopsy homogenates of these individuals was 31% compared to LPH-persistent individuals, and was accompanied by a reduced level of LPH-protein. Mean sucrase activity in individuals with low LPH was increased to 162% and was accompanied by an increase in sucrase-isomaltase (SI)-protein. Biosynthesis of LPH, SI, and aminopeptidase N (APN) was studied in organ culture of small intestinal biopsy specimens. In individuals with LPH restriction, the rate of synthesis of LPH was drastically decreased, reaching just 6% of the LPH-persistent group after 20 h of culture, while the rate of synthesis of SI appeared to be increased. In addition, maturation of pro-LPH to mature LPH occurred at a slower rate in LPH-restricted tissue. Immunoelectron microscopy revealed an accumulation of immunoreactive LPH in the Golgi region of enterocytes from LPH-restricted individuals and reduced labeling of microvillus membranes. Therefore, lactose intolerance in adults is mainly due to a decreased biosynthesis of LPH, either at the transcriptional or translational level. In addition, intracellular transport and maturation is retarded in some of the LPH-restricted individuals, and this leads to an accumulation of newly synthesized LPH in the Golgi and a failure of LPH to reach the microvillus membrane.

Characterization of cleavage and polyadenylation specificity factor and cloning of its 100-kilodalton subunit.

During the formation of the 3' ends of mRNA, the cleavage and polyadenylation specificity factor (CPSF) is required for 3' cleavage of the transcript as well as for subsequent polyadenylation. Using peptide sequences from a tryptic digest, we have cloned the 100-kDa subunit of CPSF. This subunit is a novel protein showing no homology to any known polypeptide in databases. Polyclonal antibodies against the C terminus of the protein inhibit the polyadenylation reaction. Polyclonal and monoclonal antibodies were used to characterize the composition of CPSF. Immunoprecipitations of CPSF from HeLa cell extracts and from labeled chromatographic fractions show the coprecipitation of all four subunits of 160, 100, 73, and 30 kDa. Proteins of 160 and 30 kDa that are specifically cross-linked to precursor RNA by UV irradiation were identified as CPSF subunits by immunoprecipitation. Immunofluorescent detection of CPSF in HeLa cells localized it in the nucleoplasm, excluding cytoplasm and nucleolar structures.