Prothymosin alpha is not found in yeast.

According to published accounts, prothymosin alpha exhibits high evolutionary conservation from yeast to man (Makarova, T., Grebenshikov, N., Egorov, C., Vartapetian, A., and Bogdanov, A. FEBS Lett. 257, 247-250, 1989). We report here our failure to find evidence for prothymosin alpha in yeast using three biochemical approaches: hybridization of yeast mRNA and genomic DNA with human prothymosin alpha coding region probes, performance of the polymerase chain reaction with yeast genomic template DNA and three sets of primers recognizing human prothymosin alpha coding region sequences, and isolation of yeast proteins essentially as described in the publication above. A survey of the Saccharomyces cerevisiae complete genome database using the program BLASTp verified our findings: there is no prothymosin alpha-homologue in yeast. Furthermore, DNA representing organisms from bacteria to amphibians also failed to hybridize with the same probes. Therefore, the presence of a prothymosin alpha gene in animals other than mammals is highly unlikely.

A high-risk lesion for invasive breast cancer, ductal carcinoma in situ, exhibits frequent overexpression of retinoid X receptor.

The development of prevention strategies for breast cancer will require a molecular map of carcinogenesis. We have investigated gene expression patterns in premalignant and early carcinomatous human breast lesions that confer to the patient varying risks for developing invasive breast cancer. The relative expression levels of one of the retinoid receptors, retinoid X receptor (RXR), was determined by in situ hybridization to 58 biopsy specimens; RXR mRNA grain density over each lesion was compared to that over the normal ductal/lobular units in each section. Overexpression of RXR mRNA was observed in 66% of noncomedo ductal carcinoma in situ (DCIS), which confer a >8-fold increase in breast cancer risk, and 88% of comedo DCIS lesions, which are associated with a yet higher risk. In contrast, only 8% of lesions that confer little or no increase in breast cancer risk overexpressed RXR mRNA (P = 0.0008). Limited in situ hybridization data using retinoic acid receptor (RAR) riboprobes showed overexpression of RAR alpha, but not RAR beta or -gamma, in only a modest percentage (36%) of cases, suggesting that all members of the retinoid receptor superfamily are not similarly regulated. Immunohistochemistry performed on 52 DCIS specimens for alpha, beta, and gamma isoforms of RXR confirmed its overexpression at the protein level and implicate RXR alpha as the predominant overexpressed form. The data indicate that RXR overexpression is associated with an increased risk for the development of invasive breast cancer in human breast lesions and suggest the hypothesis that it is causally involved in breast oncogenesis. The implications for retinoid chemoprevention are discussed.

Site-directed mutation of Nm23-H1. Mutations lacking motility suppressive capacity upon transfection are deficient in histidine-dependent protein phosphotransferase pathways in vitro.

We previously compared the structure and motility suppressive capacity of nm23-H1 by transfection of wild type and site-directed mutant forms into breast carcinoma cells. Wild type nm23-H1 and an nm23-H1(S44A) (serine 44 to alanine) mutant suppressed motility, whereas the nm23-H1(P96S), nm23-H1(S120G), and to a lesser extent, nm23-H1(S120A) mutant forms failed to do so. In the present study wild type and mutant recombinant Nm23-H1 proteins have been produced, purified, and assayed for phosphorylation and phosphotransfer activities. We report the first association of Nm23-H1 mutations lacking motility suppressive capacity with decreased in vitro activity in histidine-dependent protein phosphotransferase assays. Nm23-H1(P96S), a Drosophila developmental mutation homolog, exhibited normal autophosphorylation and nucleoside-diphosphate kinase (NDPK) characteristics but deficient phosphotransfer activity in three histidine protein kinase assays, using succinic thiokinase, Nm23-H2, and GST-Nm23-H1 as substrates. Nm23-H1(S120G), found in advanced human neuroblastomas, exhibited deficient activity in several histidine-dependent protein phosphotransfer reactions, including histidine autophosphorylation, downstream phosphorylation on serines, and slightly decreased histidine protein kinase activity; significant NDPK activity was observed. The Nm23-H1(S120A) mutant was deficient in only histidine-dependent serine autophosphorylation. Nm23-H1 and Nm23-H1(S44A) exhibited normal activity in all assays conducted. Based on this correlation, we hypothesize that a histidine-dependent protein phosphotransfer activity of Nm23-H1 may be responsible for its biological suppressive effects.

Site-directed mutagenesis of nm23-H1. Mutation of proline 96 or serine 120 abrogates its motility inhibitory activity upon transfection into human breast carcinoma cells.

We report the first correlation of Nm23 sequence and its tumor metastasis-suppressive capacity using site-directed mutagenesis and an in vitro tumor cell motility assay. MDA-MB-435 human breast carcinoma cells were transfected with a control expression vector (pCMVBamneo), the vector containing the wild type nm23-H1, or the nm23-H1 vector encoding mutations at the following amino acids: serine 44, a phosphorylation site; proline 96, the k-pn mutation in the Drosophila nm23 homolog that causes developmental defects; histidine 118, involved in Nm23's nucleoside diphosphate kinase activity; and serine 120, a site of mutation in human neuroblastomas and phosphorylation. The wild type nm23-H1 transfectants were 44-98% less motile to serum and 86-99% less motile to autotaxin than control vector transfectants. The proline 96 k-pn, serine 120 to glycine, and to a lesser extent serine 120 to alanine mutant nm23-H1-transfected cell lines exhibited motility levels at or above the control transfectants, indicating that these mutations can abrogate the motility-suppressive phenotype of nm23-H1. No effect was observed on cellular proliferation, nor were the serine 44 to alanine nm23-H1 mutant transfectants motile, demonstrating the specificity of the data. The data identify the first structural motifs of nm23-H1 that influence its metastasis suppressive effect and suggest complex biochemical associations or activities in the Nm23 suppressive pathway.

Cloning, chromosomal localization, and tissue expression of autotaxin from human teratocarcinoma cells.

Autotaxin, a potent human tumor cell motility-stimulating exophosphodiesterase, was isolated and cloned from the human teratocarcinoma cell line NTera2D1. The deduced amino acid sequence for the teratocarcinoma autotaxin has 94% identity to the melanoma-derived protein, 90% identity to rat brain phosphodiesterase I/nucleotide pyrophosphatase (PD-I alpha), and 44% identity to the plasma cell membrane marker PC-I. Utilizing polymerase chain reaction screening of the CEPH YAC library, we localized the autotaxin gene to human chromosome 8q23-24. Northern blot analysis of relative mRNA from multiple human tissues revealed that autotaxin mRNA steady state expression is most abundant in brain, placenta, ovary, and small intestine.

Overexpression of cyclin D mRNA distinguishes invasive and in situ breast carcinomas from non-malignant lesions.

The elucidation of molecular alterations that occur during human breast cancer progression may contribute to the development of preventative strategies. Using in situ hybridizations on a cohort of 94 biopsy lesions, quantitatively increased cyclin D mRNA expression levels were observed in only 18% of benign lesions, which confer no or slightly increased breast cancer risk, and 18% of premalignant atypical ductal hyperplasias, which confer a four to fivefold increase in breast cancer risk. The transition to carcinoma was accompanied by frequent cyclin D mRNA overexpression in 76% of low-grade ductal carcinomas in situ, 87% of higher grade comedo ductal carcinomas in situ and 83% of infiltrating ductal breast carcinomas. The data identify a molecular event that may separate benign and premalignant human breast lesions from any form of breast carcinoma.

GAG triplets as splice acceptors of last resort. An unusual form of alternative splicing in prothymosin alpha pre-mRNA.

Prothymosin alpha pre-mRNAs are alternatively spliced as a consequence of adjacent AG acceptor couplets at the intron 2/exon 3 boundary of the only expressed human prothymosin alpha gene. These acceptors are found in a unique sequence motif, GAGGAG, located immediately 3' to a consensus polypyrimidine tract. The frequency with which each acceptor is utilized appears to be invariant in all human cells and tissues examined; two mRNA transcripts in the ratio of 9:1, shorter form: longer form, have been observed in every case. Production of the shorter mRNA violates two consensus rules for splice site selection: (1) the preferred AG dinucleotide is the second, rather than the first, following the polypyrimidine stretch; and (2) it lies in a potentially unfavorable, purine-rich region. The poor performance of the first AG dinucleotide cannot be explained by its position relative to other splicing signals; in mutant prothymosin alpha gene, this AG couplet promoted efficient splicing in vivo when preceded by a C residue or followed by a GAA triplet. The GAGGAG motif in prothymosin alpha genes has been retained by the African monkey, Colobus. Because the monkey's ancestors and our own diverged some 30 million years ago, the data suggest that the ambiguity in splice-site selection confers a selective advantage.

Phoenix mutagenesis: one-step reassembly of multiply cleaved plasmids with mixtures of mutant and wild-type fragments.

A method for site-specific mutagenesis which can be used with any gene in any vector has been devised. This method has been named "phoenix mutagenesis." Upon selection of the mutation site, the plasmid bearing the gene in question is cleaved with one or two restriction enzymes that generate fragments with random staggered ends. The enzymes are chosen to maximize the protrusions while minimizing the size of the fragment to be mutated. Ligation of the fragments reconstitutes the original plasmid. Mutations are obtained by allowing mutant fragments, added in 10-fold molar excess, to compete with their wild-type counterparts for a place in the reassembled vector. In a test system with the restriction enzyme PflMI, which cleaves an 8-kb vector bearing the prothymosin alpha gene into four fragments, the efficiency of reassembly exceeded that of sealing linearized molecules. The recovery of mutated molecules was 20-50%. With Bsa1, which also cleaves the plasmid into four fragments, both the efficiency of reassembly and the production of mutants were reduced. To expedite the use of phoenix mutagenesis, a simple technique for generating mutant fragments in tandem polymerase chain reactions is presented. In the first step, a mutated subfragment is synthesized from the region of the mutation to one terminus; in the second step, the complete fragment is made with the aid of an oligomer targeted to the other terminus and one strand of the product of the first reaction. The complementary strand is present, but does not interfere with the amplification process.

Phosphorylation of human and bovine prothymosin alpha in vivo.

Prothymosin alpha is post-translationally modified. When human myeloma cells were metabolically labeled with [32P]orthophosphoric acid, they synthesized [32P]prothymosin alpha. The incorporated radioactivity was resistant to DNase and RNases A, T1, and T2, but could be completely removed by alkaline phosphatase. No evidence was found for an RNA adduct as postulated by Vartapetian et al. [Vartapetian, A., Makarova, T., Koonin, E. V., Agol, V. I., & Bogdanov, A. (1988) FEBS Lett. 232, 35-38]. Thin-layer electrophoresis of partially hydrolyzed [32P]prothymosin alpha indicated that serine residues were phosphorylated. Analysis of peptides derived from bovine prothymosin alpha and human [32P]prothymosin alpha by treatment with endoproteinase Lys-C revealed that the amino-terminal 14-mer, with serine residues at positions 1, 8, and 9, was phosphorylated at a single position. Approximately 2% of the peptide in each case contained phosphate. Further digestion of the phosphopeptide with Asp-N followed by C18 reversed-phase column chromatography produced two peptides: a phosphate-free 9-mer containing amino acids 6-14 and a labeled peptide migrating slightly faster than the N-terminal 5-mer derived from the unmodified 14-mer. Positive identification of the phosphorylated amino acid was obtained by colliding the 14-residue phosphopeptide with helium in the mass spectrometer and finding phosphate only in a nested set of phosphorylated fragments composed of the first three, four, and five amino acids. The results prove that prothymosin alpha contains N-terminal acetylserine phosphate. In a synchronized population of human myeloma cells, phosphorylation occurred throughout the cell cycle. Furthermore, prothymosin alpha appeared to be stable, with a half-life slightly shorter than the generation time. Although prothymosin alpha is known to be essential for cell division, the constancy of both the amount of the protein and the degree of its phosphorylation suggests that prothymosin alpha does not directly govern mitosis.

The human prothymosin alpha gene family contains several processed pseudogenes lacking deleterious lesions.

The six members of the human prothymosin alpha gene family have been cloned and sequenced. One gene (PTMA) contains introns and appears to be the source of all isolated prothymosin alpha cDNAs. The remaining five genes are processed pseudogenes. Four of them have consensus TATA elements upstream of sequences nearly identical to the transcriptional start region of the intron-containing gene. Those four genes also contain open reading frames coding for proteins closely related to prothymosin alpha. In two of the pseudogenes, PTMAP2 and 5, the encoded proteins differ from the product of the parental gene at only two and four locations, respectively. The fifth pseudogene (PTMAP1) encodes a different protein owing to an upstream translational initiation start site and multiple deletions and insertions. Because the potential for expression exists in this system, a search for pseudogenomic transcripts was undertaken using the polymerase chain reaction to amplify reverse transcripts of mRNAs from many human tissues and bulk DNA from several human cDNA libraries. Evidence for pseudogenomic transcripts was not obtained. Therefore, we conclude that the human prothymosin alpha gene family contains only one functional gene.

Nuclear targeting of prothymosin alpha.

Prothymosin alpha is a highly acidic protein which lacks an amino-terminal signal peptide, yet was once thought to be a precursor for thymosin alpha 1, a putative peptide hormone secreted by the thymus. Here, two lines of evidence are presented that strongly implicate prothymosin alpha as a nuclear protein: 1) in COS cells transfected with the human prothymosin alpha gene copious amounts of prothymosin alpha were present in sealed nuclei obtained by treating these cells with cytochalasin B and enucleating them centrifugally. 2) Constructs in which human prothymosin alpha nucleic acid sequences were fused in-frame either near the amino terminus of the beta-galactosidase gene in pCH110 or at the carboxyl terminus, when expressed in COS cells, resulted in nuclear localization of the fusion protein; indirect immunofluorescence in situ was used as the assay. The basic cluster of amino acids at the carboxyl terminus of prothymosin alpha, TKKQKT, has been identified as part of the nuclear targeting signal, whereas the basic cluster of amino acids situated within the thymosin alpha 1 sequence at the amino terminus failed to effect nuclear transport.

Prothymosin alpha antisense oligomers inhibit myeloma cell division.

The function of prothymosin alpha has been investigated by using four different antisense oligodeoxyribonucleotides directed at selected regions of its mRNA. In every case, when synchronized human myeloma cells were released from stationary phase by incubation in fresh medium containing antisense oligomers, cell division was prevented or inhibited; sense oligomers and random antisense oligomers had no effect. A detailed analysis of synchronized cell populations indicated that sense-treated and untreated cells divided approximately 17 hr after growth initiation, whereas cells incubated with antisense oligomer 183, a 16-mer targeted 5 bases downstream of the initiation codon, entered mitosis approximately one cell division late. The failure to divide correlated directly with a deficit in prothymosin alpha and with the continued presence of intact intracellular antisense oligomers over a period of at least 24 hr. Because antisense oligomers had no effect either on the timing of the induction of prothymosin alpha mRNA upon growth stimulation or on mRNA levels seen throughout the cell cycle, we concluded that antisense DNA caused specific hybrid arrest of translation. Our data suggest that prothymosin alpha is required for cell division. However, there is no evidence that prothymosin alpha directly regulates mitosis.

Human prothymosin alpha: purification of a highly acidic nuclear protein by means of a phenol extraction.

Human prothymosin alpha, virtually alone among proteins, is recovered from the aqueous phase of phenol-extracted cell lysates prepared from human myeloma cells or COS cells that were transfected with the human prothymosin alpha gene. This observation forms the basis for purification of the protein to homogeneity in two steps--phenol extraction followed by electrophoresis in sodium dodecyl sulfate polyacrylamide gels to remove residual contaminants consisting chiefly of carbohydrate and RNA.

Mapping introns by exon excision.

A direct method for mapping introns has been devised. The technique makes use of a radioactive synthetic RNA transcript of the gene and a complementary, single-stranded DNA copy of mRNA-derived sequences. Upon hybridization of the cDNA to RNA and cleavage with ribonuclease H, only exonic RNA sequences are degraded. The surviving RNA fragments are the introns. Electrophoretic analysis in denaturing agarose gels reveals the number and size of the introns. The order of the introns is determined separately using unlabeled RNA transcripts; surviving RNA fragments are transferred to a solid support and the blot is probed sequentially with a nested set of genomic RNA transcripts of the opposite strand. Using the human prothymosin alpha gene as an example, four introns were identified which from 5' to 3' were 2.6, 0.47, 0.47, and 0.28 kb in size. From mapping and sequencing experiments the sizes are 2.6, 0.465, 0.459, and 0.295 kb, respectively. Similarly, the presence of two 300-bp insertions in a human prothymosin alpha pseudogene was established; the inserts were later identified as 295-bp Alu repetitive elements.

Isolation and partial sequencing of the human prothymosin alpha gene family. Evidence against export of the gene products.

Prothymosin alpha and thymosin alpha 1 are believed to be thymus-derived, hormone-like materials with immunomodulatory functions performed outside the cell. These functions are inconsistent with the existence of a full length cDNA clone that does not encode an amino-terminal signal peptide or several consecutive hydrophobic residues. A study of the prothymosin alpha mRNAs and genes was undertaken in search of evidence for secreted forms of the protein. Prothymosin alpha mRNA was localized exclusively on free, rather than membrane-bound, polysomes. Upon screening cosmid and plasmid libraries totaling 2 X 10(6) clones, a gene family consisting of six members was identified. Sequence information from the 5'-ends of all the genes indicated that none encodes an amino-terminal signal peptide. One of the genes, apparently by means of alternate splicing, gives rise to two prothymosin alpha mRNAs, one of which has an additional internal glutamic acid codon with respect to the other. Comparison of the translated nucleic acid sequences of the five remaining genes with those encoded in the mRNAs revealed 30-98% homology in the first 50 amino acids. These five genes appear to be processed genes and/or pseudogenes. The localization of prothymosin alpha mRNAs on free polysomes, together with the partial nucleotide sequences of the genes, strongly suggest an intracellular function for prothymosin alpha. Therefore, the possibility must be raised that prothymosin alpha and its peptide derivatives act as xenobiotics when introduced into assays of immune function.

mRNA decay rates in late-developing Dictyostelium discoideum cells are heterogeneous, and cyclic AMP does not act directly to stabilize cell-type-specific mRNAs.

We reevaluated the use of 32PO4 pulse-chases for analyzing mRNA decay rates in late-developing Dictyostelium cells. We found that completely effective PO4 chases could not be obtained in developing cells and that, as a consequence, the decay rates exhibited by some mRNAs were influenced by the rates at which they were transcribed. In developing cells disaggregated in the presence of cyclic AMP, the poly(A)+ mRNA population turned over with an apparent half-life of 4 h, individual mRNA decay rates were heterogeneous, and some prestalk and prespore mRNAs appeared to decay with biphasic kinetics. In cells disaggregated in the absence of cyclic AMP, all prestalk and prespore mRNAs decayed with biphasic kinetics. During the first 1 to 1.5 h after disaggregation in the absence of cyclic AMP, the cell-type-specific mRNAs were selectively degraded, decaying with half-lives of 20 to 30 min; thereafter, the residual prestalk and prespore mRNA molecules decayed at rates that were similar to those measured in the presence of cyclic AMP. This short-term labilization of cell-type-specific mRNAs was observed even for those species not requiring cyclic AMP for their accumulation in developing cells. The observation that cell-type specific mRNAs can decay at similar rates in disaggregated cells with or without cyclic AMP indicates that this compound does not act directly to stabilize prestalk and prespore mRNAs during development and that its primary role in the maintenance of cyclic-AMP-dependent mRNAs is likely to be transcriptional.

Determinants of mRNA stability in Dictyostelium discoideum amoebae: differences in poly(A) tail length, ribosome loading, and mRNA size cannot account for the heterogeneity of mRNA decay rates.

As an approach to understanding the structures and mechanisms which determine mRNA decay rates, we have cloned and begun to characterize cDNAs which encode mRNAs representative of the stability extremes in the poly(A)+ RNA population of Dictyostelium discoideum amoebae. The cDNA clones were identified in a screening procedure which was based on the occurrence of poly(A) shortening during mRNA aging. mRNA half-lives were determined by hybridization of poly(A)+ RNA, isolated from cells labeled in a 32PO4 pulse-chase, to dots of excess cloned DNA. Individual mRNAs decayed with unique first-order decay rates ranging from 0.9 to 9.6 h, indicating that the complex decay kinetics of total poly(A)+ RNA in D. discoideum amoebae reflect the sum of the decay rates of individual mRNAs. Using specific probes derived from these cDNA clones, we have compared the sizes, extents of ribosome loading, and poly(A) tail lengths of stable, moderately stable, and unstable mRNAs. We found (i) no correlation between mRNA size and decay rate; (ii) no significant difference in the number of ribosomes per unit length of stable versus unstable mRNAs, and (iii) a general inverse relationship between mRNA decay rates and poly(A) tail lengths. Collectively, these observations indicate that mRNA decay in D. discoideum amoebae cannot be explained in terms of random nucleolytic events. The possibility that specific 3'-structural determinants can confer mRNA instability is suggested by a comparison of the labeling and turnover kinetics of different actin mRNAs. A correlation was observed between the steady-state percentage of a given mRNA found in polysomes and its degree of instability; i.e., unstable mRNAs were more efficiently recruited into polysomes than stable mRNAs. Since stable mRNAs are, on average, "older" than unstable mRNAs, this correlation may reflect a translational role for mRNA modifications that change in a time-dependent manner. Our previous studies have demonstrated both a time-dependent shortening and a possible translational role for the 3' poly(A) tracts of mRNA. We suggest, therefore, that the observed differences in the translational efficiency of stable and unstable mRNAs may, in part, be attributable to differences in steady-state poly(A) tail lengths.

Regulation of mRNA stability and the poly(A) problem in Dictyostelium discoideum.

This paper reviews our studies of three aspects of post-transcriptional regulation in Dictyostelium discoideum: 1) the determinants of mRNA stability in vegetative amoebae; 2) the effects of disaggregation and cyclic AMP on the decay rates of cell-type-specific mRNAs in late developing cells; and 3) the cytoplasmic function of the 3' poly(A) tracts present on most mRNAs. We find that: 1) mRNA stability in vegetative amoebae is not dependent on mRNA size, ribosome loading, or poly(A) tract length, but may be determined by specific 3'-untranslated sequences within a given mRNA; 2) mRNA decay rates in late developing cells are heterogeneous, and cyclic AMP does not act directly to stabilize cell-type-specific mRNAs; and 3) poly(A) is most likely involved in the initiation of protein synthesis via an interaction with cytoplasmic poly(A)-binding proteins.

Increased rates of decay and reduced levels of accumulation of the major poly(A)-associated proteins of Dictyostelium during heat shock and development.

Two major polypeptide species, 31,000 Mr (p31) and 31,500 Mr (p31.5), are associated with the 3' poly(A) tails of Dictyostelium mRNAs. We have measured the accumulation of newly synthesized p31 and p31.5 and the decay of preexisting p31 and p31.5 during heat shock and early development. Only trace amounts of newly synthesized p31 and p31.5 accumulate at elevated temperatures, indicating that these polypeptides are not heat shock proteins. In addition, preexisting p31 and p31.5 are rapidly degraded in heat-shocked cells. This degradation is selective and occurs simultaneously with a sharp drop in the rate of translational initiation. Similarly, in early development, a time when the rate of translational initiation is also sharply reduced, only trace amounts of newly synthesized p31 and p31.5 accumulate and most of the preexisting p31 and p31.5 is rapidly degraded. When translational elongation is inhibited with cycloheximide, preexisting p31 and p31.5 remain stable. Therefore, a correlation seems to exist between the abundance and stability of these poly(A)-associated proteins and the rate of translational initiation. Our results are consistent with the proposed role of the poly(A)-protein complex in translation and do not support the findings of Schönfelder et al. [Schönfelder, M., Horsch, A. & Schmid, H.-P. (1985) Proc. Natl. Acad. Sci. USA 82, 6884-6888] that the 73,000 Mr HeLa cell poly(A)-binding protein and the major 73,000 Mr mammalian heat shock protein (i.e., hsp70) are identical.

Identification and characterization of developmentally regulated mRNP proteins of Dictyostelium discoideum.

The isolation of poly(A)+ polysomal and nonpolysomal RNPs by oligo(dT)-cellulose chromatography has led to the identification of more than 20 polypeptides that bind to the poly(A)+ mRNA in growing Dictyostelium cells. Most of these polypeptides were identified in experiments using short-wave UV light (254 nm) to crosslink specifically bound proteins to the RNA. Digestion of the RNPs with ribonucleases A and T1 prior to their application to oligo(dT)-cellulose permitted the isolation of the 3' poly(A)-protein complexes. In polysomal RNPs, two major polypeptides, with molecular weights of 31,000 (p31) and 31,500 (p31.5), are bound to poly(A). These proteins can also be purified from cytoplasmic extracts by affinity chromatography on poly(A)-Sepharose. Partial proteolytic digestion of p31 and p31.5 indicates that they are closely related. The UV-crosslinking experiments established that p31 and p31.5 bind to the non-poly(A) segments of mRNA as well. In nonpolysomal RNPs, p31 and a polypeptide with a molecular weight of 29,500 (p29.5) are the major species associated with poly(A). Partial proteolytic digestion of p29.5 indicates that it is closely related to p31 and p31.5. Only small amounts of p29.5 were observed in the polysomal poly(A)-protein complexes. Early in Dictyostelium development, when cellular translation activity is sharply reduced, most of the p29.5, p31 and p31.5 present is selectively degraded. These observations are consistent with a translational role for these proteins.