A profusion of controls

A Challenging Pie to Splice: Drugging the Spliceosome

Since its discovery in 1977, the study of alternative RNA splicing has revealed a plethora of mechanisms that had never before been documented in nature. Understanding these transitions and their outcome at the level of the cell and organism has become one of the great frontiers of modern chemical biology. Until 2007, this field remained in the hands of RNA biologists. However, the recent identification of natural product and synthetic modulators of RNA splicing has opened new access to this field, allowing for the first time a chemical-based interrogation of RNA splicing processes. Simultaneously, we have begun to understand the vital importance of splicing in disease, which offers a new platform for molecular discovery and therapy. As with many natural systems, gaining clear mechanistic detail at the molecular level is key towards understanding the operation of any biological machine. This minireview presents recent lessons learned in this emerging field of RNA splicing chemistry and chemical biology.

Simulation of the M13 life cycle I: Assembly of a genetically-structured deterministic chemical kinetic simulation

To expand the quantitative, systems level understanding and foster the expansion of the biotechnological applications of the filamentous bacteriophage M13, we have unified the accumulated quantitative information on M13 biology into a genetically-structured, experimentally-based computational simulation of the entire phage life cycle. The deterministic chemical kinetic simulation explicitly includes the molecular details of DNA replication, mRNA transcription, protein translation and particle assembly, as well as the competing protein-protein and protein-nucleic acid interactions that control the timing and extent of phage production. The simulation reproduces the holistic behavior of M13, closely matching experimentally reported values of the intracellular levels of phage species and the timing of events in the M13 life cycle. The computational model provides a quantitative description of phage biology, highlights gaps in the present understanding of M13, and offers a framework for exploring alternative mechanisms of regulation in the context of the complete M13 life cycle.

IL-18 translational inhibition restricts IFN-gamma expression in crescentic glomerulonephritis

BACKGROUND

Interleukin-18 (IL-18), a potent inducer of interferon gamma (IFN-gamma) production, is a cytokine involved in the cell-mediated immune response that is expressed during inflammatory and pathologic conditions. IFN-gamma plays a role in the development of some models of glomerulonephritis (GN); however, the role of IL-18 in the production of IFN-gamma during these pathologies has not been studied.

METHODS

Rat IL-18 cDNA was isolated and the regulation of IL-18 gene expression was studied. IFN-gamma and IL-18 expression were determined in anti-glomerular basement membrane (GBM) antibody (Ab)-induced GN. Recombinant active IL-18 (rIL-18) was used to further identify its effect on IFN-gamma production during this GN. Glomerular injury and levels of IFN-gamma were assayed in Wistar Kyoto (WKY) rats with anti-GBM GN in the presence or absence of rIL-18.

RESULTS

Rat IL-18, similar to the mouse clone, requires processing by the IL-1beta converting enzyme to become activated. A rat IL-18 5'-untranslated region (UTR) translational inhibitor was identified that strongly inhibited the synthesis of IL-18. This translational inhibitor with different lengths (180 and 130 bp) was highly expressed during GN and correlated with minimal IFN-gamma mRNA expression. Injection of recombinant active IL-18 in WKY rats with anti-GBM GN was associated with an increase of glomerular IFN-gamma levels, proliferating cell nuclear antigen (PCNA)-ED1+ cells, and PCNA-CD8+ cells, with worsening of glomerular injury.

CONCLUSION

These data suggest that the translational control of IL-18 expression by its 5'-UTR limits the production of IL-18, resulting in restricted expression of mRNA and protein IFN-gamma in this model of GN. Furthermore, it was suggested that possible IL-18/IFN-gamma induction of local proliferation of macrophages and CD8+ cells might be an important mechanism for amplifying CD8+-mediated macrophage-dependent GN.

Double-stranded RNA-dependent protein kinase, PKR, down-regulates CDC2/cyclin B1 and induces apoptosis in non-transformed but not in v-mos transformed cells

The interferon (IFN)-induced, double stranded RNA (dsRNA)-activated serine/threonine kinase, PKR, is a potent negative regulator of cell growth when overexpressed in yeast or mammalian cells. Paradoxically, while it can function as a tumor suppressor and inducer of apoptosis, it is overexpressed in a variety of human cancers. To resolve this enigma, we established cell-lines that overexpress PKR in non-transformed and in v-mos transformed CHO cells. Overexpression of PKR suppressed the proliferation of CHO cells by inducing a transient G0/G1 arrest, followed by a delayed G2/M arrest, which attenuated cell cycle progression. These effects were accompanied by early induction of p21/WAF-1 and delayed downregulation of CDC2 and cyclin B1. Induction of proapoptotic activity of the ectopic PKR paralleled the onset of G2/M arrest in CHO cells. However, while transiently inducing p21/WAF-1, PKR did not impose G2/M arrest or apoptosis in v-mos-transformed cells, nor was CDC2 or cyclin B1 down-regulated in those cells. These findings link the proapoptotic activity of PKR to the arrest of cell cycle at the G2/M phase. Consequently, the apoptotic activity of PKR could be counter-acted by an oncogene-like v-mos that overrides the G2/M arrest induced by PKR.

Domain organization, genomic structure, evolution, and regulation of expression of the aggrecan gene family

Proteoglycans are complex macromolecules, consisting of a polypeptide backbone to which are covalently attached one or more glycosaminoglycan chains. Molecular cloning has allowed identification of the genes encoding the core proteins of various proteoglycans, leading to a better understanding of the diversity of proteoglycan structure and function, as well as to the evolution of a classification of proteoglycans on the basis of emerging gene families that encode the different core proteins. One such family includes several proteoglycans that have been grouped with aggrecan, the large aggregating chondroitin sulfate proteoglycan of cartilage, based on a high number of sequence similarities within the N- and C-terminal domains. Thus far these proteoglycans include versican, neurocan, and brevican. It is now apparent that these proteins, as a group, are truly a gene family with shared structural motifs on the protein and nucleotide (mRNA) levels, and with nearly identical genomic organizations. Clearly a common ancestral origin is indicated for the members of the aggrecan family of proteoglycans. However, differing patterns of amplification and divergence have also occurred within certain exons across species and family members, leading to the class-characteristic protein motifs in the central carbohydrate-rich region exclusively. Thus the overall domain organization strongly suggests that sequence conservation in the terminal globular domains underlies common functions, whereas differences in the central portions of the genes account for functional specialization among the members of this gene family.

An open reading frame element mediates posttranscriptional regulation of tropoelastin and responsiveness to transforming growth factor beta1

Elastin, an extracellular component of arteries, lung, and skin, is produced during fetal and neonatal growth. We reported previously that the cessation of elastin production is controlled by a posttranscriptional mechanism. Although tropoelastin pre-mRNA is transcribed at the same rate in neonates and adults, marked instability of the fully processed transcript bars protein production in mature tissue. Using RNase protection, we identified a 10-nucleotide sequence in tropoelastin mRNA near the 5' end of the sequences coded by exon 30 that interacts specifically with a developmentally regulated cytosolic 50-kDa protein. Binding activity increased as tropoelastin expression dropped, being low in neonatal fibroblasts and high in adult cells, and treatment with transforming growth factor beta1 (TGF-beta1), which stimulates tropoelastin expression by stabilizing its mRNA, reduced mRNA-binding activity. No other region of tropoelastin mRNA interacted with cellular proteins, and no binding activity was detected in nuclear extracts. The ability of the exon-30 element to control mRNA decay and responsiveness to TGF-beta1 was assessed by three distinct functional assays: (i) insertion of exon 30 into a heterologous gene conferred increased reporter activity after exposure to TGF-beta1; (ii) addition of excess exon 30 RNA slowed tropoelastin mRNA decay in an in vitro polysome degradation assay; and (iii) a mutant tropoelastin cDNA lacking exon 30, compared to wild-type cDNA, produced a stable transcript whose levels were not affected by TGF-beta1. These findings demonstrate that posttranscriptional regulation of elastin production in mature tissue is conferred by a specific element within the open reading frame of tropoelastin mRNA.

A cognate dopamine transporter-like activity endogenously expressed in a COS-7 kidney-derived cell line

The activity of the dopamine transporter is an important mechanism for the maintenance of normal dopaminergic homeostasis by rapidly removing dopamine from the synaptic cleft. In kidney-derived COS-7, COS-1 and HEK-293 but not in other mammalian cell lines (CHO, Y1, Ltk-), we have characterized a putative functional dopamine transporter displaying a high affinity (Km approximately 250 nM) and a low capacity (approximately 0.1 pmol/10(5) cells/min) for [3H]dopamine uptake. Uptake displayed a pharmacological profile clearly indicative of the neuronal dopamine transporter. Estimated Ki values of numerous substrates and inhibitors for the COS-dopamine transporter and the cloned human neuronal transporter (human dopamine transporter) correlate well with the exception of a few notable compounds, including the endogenous neurotransmitter dopamine, the dopamine transporter inhibitor GBR 12,909 and the dopaminergic agonist apomorphine. As with native neuronal and cloned dopamine transporters, the uptake velocity was sodium-sensitive and reduced by phorbol ester pre-treatment. Two mRNA species of 3.8 and 4.0 kb in COS-7 cells were revealed by Northern blot analysis similar in size to that seen in native neuronal tissue. A reverse-transcribed PCR analysis confirmed the existence of a processed dopamine transporter. However, no immunoreactive proteins of expected dopamine transporter molecular size or [3H]WIN 35,428 binding activity were detected. A partial cDNA of 1.3 kb, isolated from a COS-1 cDNA library and encoding transmembrane domains 1-6, displayed a deduced amino acid sequence homology of approximately 96% to the human dopamine transporter. Taken together, the data suggest the existence of a non-neuronal endogenous high affinity dopamine uptake system sharing strong functional and molecular homology to that of the cloned neuronal dopamine transporter.

Regulation of rat pyruvate carboxylase gene expression by alternate promoters during development, in genetically obese rats and in insulin-secreting cells. Multiple transcripts with 5'-end heterogeneity modulate translation

A previous study on the gene structure of rat pyruvate carboxylase revealed that two tissue-specific promoters are responsible for the production of multiple transcripts with 5'-end heterogeneity (Jitrapakdee, S., Booker, G. W., Cassady, A. I., and Wallace, J. C. (1997) J. Biol. Chem. 272, 20522-20530). Here we report transcription and translation regulation of pyruvate carboxylase (PC) expression during development and in genetically obese rats. The abundance of PC mRNAs was low in fetal liver but increased by 2-4-fold within 7 days after birth, concomitant with an 8-fold increase in the amount of immunoreactive PC and its activity and then decreased during the weaning period. Reverse transcriptase polymerase chain reaction analysis indicated that the proximal promoter was activated during the suckling period and reduced in activity at weaning. In genetically obese Zucker rats, adipose PC was 4-5-fold increased, concomitant with a 5-6-fold increase in mRNA level. Reverse transcriptase-polymerase chain reaction analysis also showed that the proximal promoter was activated in the hyperlipogenic condition. Conversely, transcription of the proximal promoter was not detectable in various liver cell lines, suggesting that this promoter was not functional under cell culture conditions. In rat pancreatic islets and insulinoma cells, only transcripts D and E, generated from the distal promoter of the PC gene, were expressed. Glucose increased PC transcripts from the distal promoter when the insulinoma cells were maintained in 10 mM glucose. We conclude that the proximal promoter of the rat PC gene plays a major role in gluconeogenesis and lipogenesis, whereas the distal promoter is necessary for anaplerosis. In vitro translation and in vivo polysome profile analysis indicated that transcripts C and E were translated with similar translational efficiencies that are substantially greater than that of transcript D, suggesting that 5'-untranslated regions play a role in translational control.

Autoimmunity to tropomyosin isoforms in ulcerative colitis (UC) patients and unaffected relatives

Autoimmunity to cytoskeletal protein tropomyosin (TM) has been demonstrated in UC. However, the TM isoforms involved in this IgG-mediated autoimmune response in UC and the possible presence of serum IgG antibodies against TM (hTMs IgG) in unaffected UC relatives are unknown. The aim of this study was to investigate the human TM (hTM) isoforms recognized by serum IgG from UC and to explore whether hTM IgG antibodies are present in healthy UC relatives. We studied 33 UC patients with 58 unaffected relatives, 31 Crohn's disease (CD) patients with 31 unaffected relatives and 20 controls (C). Serum IgG against four recombinant hTM isoforms (hTM1, 2, 3, 5) were tested by ELISA. p-ANCA were tested by ELISA and immunofluorescence. Serum hTM1 and hTM5 IgG were higher in UC patients than in CD and C (P<0.005). Among UC patients 52% were seropositive for hTM1 and 64% for hTM5 (P<0.001 versus CD and C). In UC, hTM5 IgG were higher in p-ANCA+ than in ANCA- patients (P=0.04). In UC relatives hTM1 IgG were higher than in CD relatives and C (P<0.01). UC relatives were more frequently seropositive for hTM1 than hTM5 IgG (P=0.001). while probands were more frequently seropositive for hTM5 IgG (P=0.008). We conclude that autoimmunity to hTMI and hTM5 is a feature of UC, while hTM1 IgG differentiate UC relatives from controls. A genetic susceptibility to immune recognition of hTM isoforms in UC is suggested.

Three N-terminal variants of the AE2 Cl-/HCO3- exchanger are encoded by mRNAs transcribed from alternative promoters

Multiple AE2 Cl-/HCO3- exchanger mRNAs have been identified in rat. To determine the genetic basis for these mRNAs and whether they encode different variants of the exchanger, we used both rapid amplification of cDNA ends and S1 nuclease protection protocols and examined the organization of the gene. mRNAs encoding three N-terminal variants of AE2 (AE2a, AE2b, and AE2c) were identified and shown to be transcribed from alternative promoters. The AE2a transcription unit consists of 23 exons, with exons 1 and 2 containing 5'-untranslated sequence and the first 17 codons. The first exon of AE2b is located in intron 2; it contains 5'-untranslated sequence and an alternative 3-amino acid N-terminal coding sequence and is spliced to exon 3. The first exon of AE2c is located in intron 5; it consists of 5'-untranslated sequence and is spliced to exon 6, which contains the translation initiation codon corresponding to Met-200 of AE2a. Northern analysis shows that AE2a is expressed in all tissues, AE2b exhibits a more restricted distribution with highest levels in stomach, and AE2c is expressed only in stomach. Thus, the use of alternative promoters leads to the production of three N-terminal variants of AE2 that exhibit tissue-specific patterns of expression.

Modulation of thermogenic response to parenteral amino acid infusion in surgical stress

Thermogenic response to parenteral infusion of amino acid mixtures in rats undergoing surgical stress was investigated with the aid of a small animal indirect calorimeter. Male Wistar rats (n = 12) were laparotomized and received intestinal abrasion along with construction of a parenteral route. After 48 h, nutrient-induced thermogenesis (NIT) produced by amino acid mixtures was examined and compared with that of control rats (n = 12) without surgical stress. The NIT values of the leucine-enriched solution were greater than those of the control solution in both rats with and without surgical stress. The NIT generated by the leucine-enriched solution in rats with surgical stress was significantly higher than that in rats without. NIT values for 10 different kinds of single amino acid solutions with identical concentrations was then examined in 60 rats. The NIT values of the leucine and the glycine solutions were relatively higher than those of other solutions. However, when puromycin was injected intraperitoneally into 12 rats just before measurement of the NIT in response to amino acid mixtures, no significant differences were found in NIT values between the mixtures. In summary, surgical stress increased the thermogenic response to the leucine-enriched solution, indicating that utilization of leucine may be augmented under surgical stress. In fact, leucine itself may be a thermogenic amino acid. Inhibition of protein synthesis prevented the increase in thermogenic response induced by the leucine-enriched solution. We conclude that thermogenic responses to parenteral amino acid mixtures may differ depending not only upon the compositions of amino acids but also upon the host conditions.

Molecular analysis of the multiple Golf alpha subunit mRNAs in the rat brain

The alpha subunit of GTP-binding protein Golf (G alpha olf), identified in the olfactory epithelium, in which it is supposed to mediate odorant-generated adenylyl cyclase activations, is much more expressed in the striatum than G alpha s, the classical stimulatory G protein alpha subunit (Hervé et al., J Neurosci., 13 (1993) 2237-2248). Four species of G alpha olf messenger RNA (mRNA) were observed by Northern blot in the rat striatum. Analysis of striatal G alpha olf cDNA clones demonstrated that despite extensive variations in their 5' and 3' untranslated regions, these four G alpha olf mRNAs encode the same G alpha olf polypeptide. One of the four mRNA species, which was selectively observed in the striatum, is generated by a novel promoter whose activity was barely detectable in the olfactory epithelium. Surprisingly, this promoter generates an unexpectedly high proportion of transcripts in which the first intron is unspliced. The retention of intron was found to abolish the translation of G alpha olf mRNA in the reticulocyte lysate system, suggesting that it could be involved in the down regulation of G alpha olf expression in the striatum. Furthermore, a new polyadenylation site with a non canonical sequence, AATACA, was found to be responsible of the two shortest forms of G alpha olf mRNA. In conclusion, we have shown that the G alpha olf proteins present in the striatum and the olfactory epithelium are identical and that multiple variations in the untranslated parts of the mRNAs could affect G alpha olf expression.

Developmental regulation of elastin production. Expression of tropoelastin pre-mRNA persists after down-regulation of steady-state mRNA levels

To assess the mechanisms controlling the developmental regulation of tropoelastin expression in vivo, we developed a reverse-transcription-polymerase chain reaction (RT-PCR) assay to detect tropoelastin pre-mRNA as an indicator of ongoing transcription in intact tissue. RNA was isolated from mid-fetal (early-elastogenic), neonatal (peak tropoelastin expression), and adult (very low tropoelastin expression) rat lungs and reverse transcribed, and the cDNA was amplified with intron specific primers. A weak hybridization signal for tropoelastin pre-mRNA was seen in mid-fetal samples, and paralleling the increase in steady-state mRNA levels, a strong signal for pre-mRNA was detected in neonatal samples, indicating transcriptional regulation. Stimulation of fetal lung tropoelastin expression by maternal administration of dexamethasone also led to an increase in pre-mRNA levels. However, signal for tropoelastin pre-mRNA in adult samples was equal to that detected in neonatal samples, even though mRNA levels had dropped about 80-fold. Persistence of tropoelastin transcription in adult tissue was also seen in cell culture models and was verified by nuclear runoff assay. In addition, an RT-PCR assay for alpha 1 (I) procollagen pre-mRNA accurately revealed the known transcriptional regulation of this gene. Our results demonstrate that the induction and maintenance of elastogenesis is controlled by a transcriptional mechanism, whereas, the cessation of tropoelastin expression is controlled by a post-transcriptional mechanism.

The 3' untranslated region of bovine preprolactin contains a transferable non-poly(A) mRNA sequence that prolongs translation

Preprolactin transcripts, synthesized in vitro, were actively translated for a prolonged period when injected into Xenopus oocytes. As a result, prolactin continued to be secreted into the media for up to 6 days after injection of the transcript. To investigate the role of the preprolactin 3' untranslated sequence in stabilizing transcripts, it was fused to coding regions derived from signal recognition particle receptor alpha-subunit or preproinsulin receptor. The translational half-life of the chimeric RNA was increased for both coding regions, suggesting that a sequence within the preprolactin 3' untranslated region that prolongs translation is transferable. Deletion mutagenesis of this untranslated region demonstrated that a sequence of 98 nucleotides immediately following the prolactin stop codon was sufficient to prolong translation of RNAs injected into Xenopus oocytes.

Identification of multiple promoters within the N-ras proto-oncogene

N-ras possesses a 'housekeeping' promoter, being G + C-rich and devoid of a TATA-box. Transcription initiates at a number of locations within this gene, a phenomena that is generally attributed to the absence of a TATA-box. In this report we investigate the possibility that multiple promoters, which could potentially contribute to the observed 5' end heterogeneity, exist within the murine N-ras gene. The 5' region of the gene was subdivided into several fragments, each corresponding to a region in which one or more transcription initiation site(s) had been mapped, and the ability of each fragment to express a reporter gene was assessed. Promoter activity was found associated with three independent, non-overlapping fragments, two of which were located entirely within transcribed regions of the gene. We found that these intragenic promoters were able to express the N-ras gene itself, as well as the reporter gene. In addition, we found that the activity of an intragenic promoter fragment was dependent upon the presence of regions encompassing initiation sites, and that a small fragment (approximately 40 bp) encompassing several initiation sites possessed promoter activity. These data support the existence of an 'initiator' element within the N-ras gene. Overall, our results demonstrate that multiple promoters reside within N-ras and suggest that they may play a role in generating the observed mRNA 5' end heterogeneity. The identification of multiple promoters within N-ras may have important implications regarding the regulation of expression of this gene in normal and malignant tissues. In addition, since a number of other genes with housekeeping promoters also initiate transcription at multiple locations, it is possible that the utilization of multiple promoters may represent a common feature of this class of genes.

Structural features of the human gene for muscle-specific enolase. Differential splicing in the 5'-untranslated sequence generates two forms of mRNA

We report here the isolation and characterization of the human gene for the beta or muscle-specific isoform of the glycolytic enzyme enolase. The nucleotide sequence analysis revealed structural features, such as organization as 11 coding exons, the first exon consisting of an untranslated sequence and hence resembling sequences of the other two members of the gene family, the alpha and gamma enolase genes. The beta enolase locus spans about 6 kbp genomic DNA. Sequences matching the consensus sequence for muscle-specific regulatory factors are present in the 5'-flanking region and within the first intron. A combination of primer extension, S1 nuclease protection and RNA-sequencing experiments indicates that the gene has a unique transcriptional start site, 26 bp downstream of a TATA-like box; the differential usage of two donor sites within the untranslated exon I generates two alternatively spliced transcripts. The existence of the two mRNA, differing from one another in the presence or absence of a 42-nucleotide fragment in the leader sequence, was confirmed by cloning the corresponding cDNA using the rapid amplification of cDNA ends strategy. Secondary-structure predictions indicated that the leader sequences of the spliced forms could form hairpin structures with different free energies of formation, suggesting translational control.

Two different promoters direct expression of two distinct forms of mRNAs of human platelet-activating factor receptor

The human platelet-activating factor (PAF) receptor gene exists as a single copy on chromosome 1. We identified two 5'-noncoding exons, each of which has distinct transcriptional initiation sites. These exons are alternatively spliced to a common splice acceptor site on a third exon that contains the total open reading frame to yield two different species of functional mRNA (Transcript 1 and 2). Transcript 1 has consensus sequences for transcription factor NF-kappa B and Sp-1, and the Initiator (Inr) sequence homologous to the murine terminal deoxynucleotidyltransferase gene. Transcript 2 also contains consensus sequences for transcription factor AP-1, AP-2, and Sp-1. Transcripts 1 and 2 were both detected in heart, lung, spleen, and kidney, whereas only Transcript 1 was found in peripheral leukocytes, a differentiated human eosinophilic cell line (EoL-1 cells), and brain. Existence of distinct promoters was thus suggested to play a role in the regulatory control of PAF receptor gene expression in different human tissues and cells.

A human somatostatin receptor (SSTR3), located on chromosome 22, displays preferential affinity for somatostatin-14 like peptides

We report here on the cloning of a human intronless gene encoding a member of the G-protein linked somatostatin (SST) receptor subfamily, termed SSTR3. Based on the deduced amino acid sequence, this gene encodes a 418 amino acid protein displaying sequence similarity, particularly within putative transmembrane domains, with the recently cloned human SSTR1 (62%), SSTR2 (64%) and SSTR4 (58%) receptors. Membranes prepared from COS-7 cells transiently expressing the human SSTR3 gene bound [125I]Leu8,D-Trp22,Tyr25 SST-28 in a saturable manner with high affinity (approximately 200 pM) and with rank order of potency (D-Trp8 SST-14 > SST-14 > SMS-201-995 > SST-28) indicative of a somatostatin-14 selective receptor. The pharmacological profile of the expressed human SSTR3 receptor is similar but not identical to that reported for the rat homolog [(1992) J. Biol. Chem. 267, 20422] where the peptide selectivity is SST-28 > or = SST-14 >>> SMS-201-995. Northern blot analysis reveals the presence of an SSTR3 mRNA species of approximately 5 kb in various regions of the monkey brain, including the frontal cortex, cerebellum, medulla, amygdala, with little or no SSTR3 mRNA detectable in brain regions such as the striatum, hippocampus, and olfactory tubercle. The SSTR3 receptor gene maps to human chromosome 22. The existence of at least four distinct human genes encoding somatostatin-14 selective receptors with diverse pharmacological specificities may help to account for some of the multiple biological actions of somatostatin under normal and pathological conditions.

Serotonin receptor cDNA cloned from Lymnaea stagnalis

Serotonin (5-HT) is a major neurotransmitter that influences various behaviors, neuronal plasticity, learning, and memory in molluscs. Although the physiology of 5-HT transmission in molluscs is well studied, little is known about the pharmacology and diversity of the 5-HT receptor system. Based on the high homology of genes coding for guanine nucleotide-binding protein (G protein)-coupled receptors, we have cloned a gene for the Lymnaea stagnalis 5-HT (5HTlym) receptor. The putative receptor protein, 509 amino acids long, has highest homology with the Drosophila 5-HT receptors and mammalian 5HT1 receptors. As revealed by RNA blot-hybridization analysis, two mRNA species of 2.3 and 3.2 kb are detected in the central nervous system of Lymnaea. Transient expression of 5HTlym in COS-7 cells showed saturable [3H]lysergic acid diethylamide binding with an estimated dissociation constant of 0.9 nM. The 5HTlym receptor exhibited a mixed 5HT-like pharmacology that cannot be precisely categorized with existing mammalian classification nomenclature. However, the 5HTlym receptor does display some characteristics that have been attributed to the putative mammalian vascular 5HT1-like receptor.

Variants of the 5'-untranslated sequence of human growth hormone receptor mRNA

The human growth hormone receptor (GHR) gene was proposed to contain multiple 5'-noncoding exons (Leung et al., 1987). The exact number and structure of these exons are unknown. As a first step in investigating this point more closely, we decided to clone alternative 5'-noncoding sequences of human liver GHR mRNA. The ligation-mediated single-sided polymerase chain reaction (PCR) was applied for selective amplification of 5'-terminal sequences of human liver GHR cDNA. PCR products were cloned and sequenced. Eight different sequence variants diverging in the 5'-untranslated regions beginning 12 base pairs upstream from the initiating ATG codon were found. One variant seems to represent unspliced or partially spliced GHR mRNA. The remaining variants probably correspond to multiple alternatively spliced forms of GHR mRNA. Homologs for three of these variants were found among previously published 5'-noncoding sequences of GHR cDNA obtained from other species by conventional cDNA cloning. Most of the cloned human liver GHR cDNA variants contain one or more ATG preceding the main GHR open reading frame start of translation. Thus, the GHR genes appeared to be a striking example of a very complex transcription unit.

Cloning of two different 5' untranslated exons of bovine acidic fibroblast growth factor by the single strand ligation to single-stranded cDNA methodology

In an attempt to characterize the 5' UTR of the aFGF mRNAs we used the new anchored PCR methodology, single strand ligation to ss-cDNAs (SLIC). In bovine brain and retina, two kinds of aFGF cDNA clones were isolated. They contained two alternative exons located 34 bp upstream to the translation initiation codon ATG. Taking into account the number of clones specific for each exon, the two mRNAs are expressed with the same ratio in both tissues. One of these bovine 5' UTR exons (136 bp) showed 81% identity to a human 5' UTR exon, the second one (323 bp) was 70% identical to the second human 5' UTR exon with a central region of 90 nucleotides showing 41% identity. The conservation of the splicing positions for these 5' UTR alternate exons in both bovine and human species, suggests that the overall structure of the aFGF gene is conserved in mammals. Furthermore, the conservation of the nucleotide sequences and of the localization of these 5' UTR exons suggests that these non-coding regions may be involved in the control of aFGF gene expression.

Identification of a second promoter which drives the expression of gamma-glutamyl transpeptidase in rat kidney and epididymis

In rat, gamma-glutamyl transpeptidase (GGT) is encoded by multiple mRNAs (mRNAI, mRNAII, mRNAIII, and mRNAIV) that differ only in their 5' untranslated regions and are transcribed from a single-copy gene. Using oligonucleotides designed from the 5' untranslated sequences of the GGT mRNAII and mRNAIII, we amplified a 3.4-kb genomic sequence which contains the promoter region for mRNAII. The sequence flanking the two initiation start sites for mRNAII contains consensus motifs for several potential regulatory proteins and a TATA-like element at the expected position 26 bp upstream from the predominant start site. The sequence from positions -528 to +72 associated with the chloramphenicol acetyltransferase (CAT) reporter gene drives a promoter activity in LLC-PK1, a pig kidney cell line. Deletion analysis revealed that the region from nucleotides -528 to -322 mediates an activation of the promoter activity, whereas the sequence from -322 to -114 has a negative effect. Furthermore, the structural organization of the 5' end of the GGT gene reveals that the GGT mRNAIII is transcribed from a third promoter located upstream from the promoter II on the GGT gene. By Northern blot analysis, the promoter II was found to be expressed only in the kidney and in the epididymis. We also identified two new mRNA species which are expressed in the H5 hepatoma cells. Therefore, the GGT gene expression reveals a strong tissue- or cell-specific pattern which is based on the transcription of several mRNA species from multiple promoters.

The translocation (6;9), associated with a specific subtype of acute myeloid leukemia, results in the fusion of two genes, dek and can, and the expression of a chimeric, leukemia-specific dek-can mRNA

The translocation (6;9) is associated with a specific subtype of acute myeloid leukemia (AML). Previously, it was found that breakpoints on chromosome 9 are clustered in one of the introns of a large gene named Cain (can). cDNA probes derived from the 3' part of can detect an aberrant, leukemia-specific 5.5-kb transcript in bone marrow cells from t(6;9) AML patients. cDNA cloning of this mRNA revealed that it is a fusion of sequences encoded on chromosome 6 and 3' can. A novel gene on chromosome 6 which was named dek was isolated. In dek the t(6;9) breakpoints also occur in one intron. As a result the dek-can fusion gene, present in t(6;9) AML, encodes an invariable dek-can transcript. Sequence analysis of the dek-can cDNA showed that dek and can are merged without disruption of the original open reading frames and therefore the fusion mRNA encodes a chimeric DEK-CAN protein of 165 kDa. The predicted DEK and CAN proteins have molecular masses of 43 and 220 kDa, respectively. Sequence comparison with the EMBL data base failed to show consistent homology with any known protein sequences.

The translocation (6;9), associated with a specific subtype of acute myeloid leukemia, results in the fusion of two genes, dek and can, and the expression of a chimeric, leukemia-specific dek-can mRNA

The translocation (6;9) is associated with a specific subtype of acute myeloid leukemia (AML). Previously, it was found that breakpoints on chromosome 9 are clustered in one of the introns of a large gene named Cain (can). cDNA probes derived from the 3' part of can detect an aberrant, leukemia-specific 5.5-kb transcript in bone marrow cells from t(6;9) AML patients. cDNA cloning of this mRNA revealed that it is a fusion of sequences encoded on chromosome 6 and 3' can. A novel gene on chromosome 6 which was named dek was isolated. In dek the t(6;9) breakpoints also occur in one intron. As a result the dek-can fusion gene, present in t(6;9) AML, encodes an invariable dek-can transcript. Sequence analysis of the dek-can cDNA showed that dek and can are merged without disruption of the original open reading frames and therefore the fusion mRNA encodes a chimeric DEK-CAN protein of 165 kDa. The predicted DEK and CAN proteins have molecular masses of 43 and 220 kDa, respectively. Sequence comparison with the EMBL data base failed to show consistent homology with any known protein sequences.

Acute recruitment of prolactin-secreting cells is regulated posttranscriptionally

17 beta-estradiol (E2) stimulates the release of an activity from neurointermediate lobe (NIL) cells which increases the relative abundance of prolactin (PRL) secretors in cultures of anterior pituitary (AP) cells. In the present study, we sought to determine whether this NIL/E2 effect was due to recruitment of growth hormone (GH)-releasing cells into the PRL-secreting population and to define the mechanism regulating this induction of PRL secretors. AP cells from ovariectomized rats were cultured overnight, exposed to NIL/E2 treatment (or medium alone) for 3 h and then subjected to reverse hemolytic plaque assays for PRL and GH release. We found that exposure to NIL/E2 increased by 10% the proportion of AP cells that secreted PRL but did not influence the overall abundance of cells that released GH. A more critical analysis of these cultures revealed that all of the newly recruited PRL-secreting cells also released GH. This increment in the proportion of cells that released both PRL and GH concurrently was accompanied by an equivalent decrease in the fraction that secreted GH alone. Thus, it appeared that NIL/E2 treatment initiated PRL secretion by cells that previously released only GH. We then tested whether this induction of PRL secretors required the synthesis of proteins and/or RNA. We found that the protein synthesis inhibitor cycloheximide completely abolished the recruitment of PRL-releasing cells by NIL/E2 treatment, whereas the RNA synthesis inhibitor actinomycin D had no effect on this response. We conclude that NIL/E2 treatment induces PRL secretion by cells that formerly released only GH and that this induction is regulated posttranscriptionally.

In vivo analysis of plant RNA structure: soybean 18S ribosomal and ribulose-1,5-bisphosphate carboxylase small subunit RNAs

A method to investigate the structure of RNA molecules within intact plant tissues has been developed. The RNA structures are analyzed using dimethyl sulfate (DMS), which modifies substituents of adenine and cytosine residues within single-stranded regions of RNA molecules. Reactive sites are identified by primer extension analysis. Using this procedure, an analysis of the secondary structure of the cytoplasmic 18S ribosomal RNA in soybean seedling leaves has been completed. DMS modification data are in good agreement with the phylogenetic structure predicted for soybean 18S rRNA. However, there are a few notable exceptions where residues thought to be involved in double-stranded regions in all 18S rRNAs are strongly modified in soybean leaf samples. These data taken together with the phylogenetic structure suggest that alternate structures may exist in vivo. The further applicability of this technique is demonstrated by comparing the modification pattern obtained in vivo to that obtained in vitro for a particular mRNA molecule encoding the small subunit of ribulose-1,5-bisphosphate carboxylase. The results obtained are compared to a predicted minimum energy secondary structure. The data indicate that the conformation of RNA molecules within the cell may not be reflected in a structural analysis of purified mRNA molecules.

Binding of human glutaminyl-tRNA synthetase to a specific site of its mRNA

The human glutaminyl-tRNA synthetase is able to bind to its own mRNA. The enzyme contains two binding regions. One is located in the central section of the enzyme which includes its most hydrophilic portion with ten lysine residues in a block of 20 amino acids. This part of the enzyme binds unspecifically to all RNA sequences tested. A second binding region is located in that part of the enzyme which shows high degrees of sequence similarities with the bacterial and yeast glutaminyl-tRNA synthetases, and which is most likely responsible for the charging of tRNA with glutamine. This second RNA binding region specifically interacts with a site in the 3' noncoding region of the synthetase's mRNA. The binding site in the mRNA is characterized by an extended secondary structure that includes elements of the 'identity set' of nucleotides recognized by the enzyme when interacting with tRNA. We discuss possible physiological implications of the interaction between glutaminyl-tRNA synthetase and its mRNA.

The 5' untranslated regions of acetyl-coenzyme A carboxylase mRNA provide specific translational control in vitro

Acetyl-coenzyme A carboxylase (ACC) catalyzes the rate-limiting step in the biosynthesis of long-chain fatty acids. Transcription of the single-copy ACC gene from two independent promoters, together with the differential splicing of the transcripts, gives rise to mature ACC mRNA having the same open reading frame (ORF), but exhibiting heterogeneity in their 5' untranslated region (5'-UTR). Class 1 ACC mRNA are transcribed from the inducible promoter 1 and their 5'-end leading sequences are provided by exon 1. Class 2 ACC mRNA are transcribed from the constitutively expressing promoter 2 and their leading sequences are derived from exon 2. In order to understand the role of different 5' UTR of ACC mRNA we have synthesized in vitro transcripts with defined ACC mRNA 5' UTR and examined their relative translational efficiencies in rabbit reticulocyte lysates. The major translation product of both forms of ACC mRNA was initiated at the first AUG of the ORF. Class 1 transcripts had a 6-9-fold better translational efficiency than class 2 transcripts, based on the quantity of major peptide produced by a given amount of transcript. The poor translational efficiency of class 2 transcripts can be improved by the removal of sequences contributed by exon 2, suggesting that they play an inhibitory role in the translation of class 2 types of ACC mRNA. In addition to their higher translational efficiency, the class 1 transcripts can also initiate translation at in-frame non-AUG codons, located in exon 1, i.e. upstream to the starting AUG of the common ACC mRNA ORF. This results in novel ACC peptides with extended N termini. These observations are consistent with the hypothesis that the 5' UTR heterogeneity in the ACC mRNA may be involved in post-transcriptional control, at the level of translation, of the ACC gene expression.

Complex pattern of alternative splicing generates unusual diversity in the leader sequence of the chicken link protein mRNA

We report here the isolation of the 5' end and the promoter region of the gene for chicken cartilage link protein, and demonstrate extensive heterogeneity of the leader sequence arising from differential utilization of multiple splice sites within the 5'-most exon. The 500-base pairs (bp) exon 1 consists of solely untranslated sequence and is followed by an intron greater than 33 kilobase pairs (kb). Together, the five exons predict a gene size longer than 100 kb. Multiple transcription initiation sites were mapped 34, 46, 56, 66 and 76 bp downstream of a TATA-like motif. Sequence analysis revealed that in addition to the non-spliced variant, multiple mRNA species were generated by alternative splicing resulting in the exclusion of 92, 166, 170, 174 and 263 nucleotides (nt), respectively, from exon 1. Polymerase chain reaction confirmed the existence of various splice forms, and showed cell type- and developmental stage-specific expression for one group of them. Secondary structure predictions indicated that the leaders of the splice forms could form stable hairpin structures with different free energies of formation (up to delta G = -110 kcal/mol), suggesting translational control. The splice variant detected in the largest amount had the least stable predicted hairpin (delta G = -31.7 kcal/mol).

Diversity of coding strategies in influenza viruses

Influenza viruses have exploited a variety of strategies to increase their genome coding capacities. These include unspliced, spliced, alternatively spliced and bicistronic mRNAs, translation from overlapping reading frames and a coupled stop-start translation of tandem cistrons.

Identification of multiple transcription start sites in the human insulin-like growth factor-I gene

We have localized four transcription initiation sites in the human insulin-like growth factor-I (IGF-I) gene. Two transcription start sites were identified which result in a longer and shorter version of the leader derived from the known exon 1 of the IGF-I gene. Transcription starting at the upstream transcription initiation site results in a leader exon 1 of about 1155 nucleotides (nt), whereas transcription starting at the downstream initiation site results in a leader of about 240 nt. The majority of the transcripts initiate at the latter site. We further identified a region in the human IGF-I gene between exons 1 and 2, which shows a high degree of homology with the rat IGF-I leader exon 1B. By means of the polymerase chain reaction (PCR) we detected human IGF-I mRNAs containing this novel leader. The corresponding exon was designated exon 1B according to the rat IGF-I gene terminology. PCR and RNase protection analyses identified two transcription start sites within this alternative leader exon 1B. Transcription initiated at the most upstream start site results in a leader of about 750 nt, whereas transcription starting at the downstream site is heterogeneous, resulting in leaders of 65-75 nt long. No consensus TATA-box or AT-rich regions are present immediately upstream of all four transcription start sites identified, nor are these regions particularly GC-rich. The IGF-I gene is known to be expressed differentially in a tissue- and development-specific fashion. Differential activation of multiple promoters could very well play a crucial role in IGF-I gene regulation.

Cowpea mosaic virus middle component RNA contains a sequence that allows internal binding of ribosomes and that requires eukaryotic initiation factor 4F for optimal translation

Cowpea mosaic virus (CPMV) middle component RNA (M-RNA) encodes two proteins of 105 and 95 kDa, of which translation starts at nucleotide (nt) 161 and nt 512, respectively. In vitro translation of both proteins directed by T7 transcripts of M-RNA was stimulated fourfold by eukaryotic initiation factor 4F (eIF-4F), the cap-binding protein complex. The ratio of the synthesis of both proteins after translation was not influenced by eIF-4F or by any known eIF. Part of the CPMV 5' sequence was cloned downstream of the 5' untranslated region of ornithine decarboxylase (ODC); the latter untranslated sequence has a highly stable secondary structure, preventing efficient translation of ODC. Insertion of nt 161 to 512 of CPMV M-RNA upstream of the ODC initiation codon resulted in a marked increase in ODC translation, which indicates that the CPMV sequence contains an internal ribosome-binding site. The insertion conferred stimulation by eIF-4F on ODC translation, showing that eIF-4F is able to stimulate internal initiation.

Both the 5' untranslated region and the sequences surrounding the start site contribute to efficient initiation of translation in vitro

The role of RNA sequences in the 5' leader region between the cap site and initiating AUG in mediating translation was examined in vitro. Hybrid mRNAs were synthesized in which the cognate leader sequence was replaced with either optimized or compromised leader sequences, and translational efficiency was measured for six different coding regions. Translation was most efficient with a leader containing the 5' untranslated region from Xenopus beta-globin and an optimized initiation sequence. Compared with the cognate leaders, this hybrid was observed to increase translation of the various coding regions as much as 300-fold. The translational efficiencies of the different coding regions also varied substantially. In contrast to earlier suggestions that increased leader efficiency results from higher affinity of the leader for a limiting factor, our experiments suggest that increased translation from the beta-globin hybrid leader sequence results from more rapid initiation of translation.

Prolactin receptor gene expression in the rabbit: identification, characterization and tissue distribution of several prolactin receptor messenger RNAs encoding a unique precursor

The expression of the prolactin (PRL) receptor gene was studied in rabbit tissues by Northern blot and S1 mapping analysis of mRNA preparations. Rabbit mammary gland contained three major (10.5, 3.4, and 2.7 kb) and one minor (6.2 kb) prolactin receptor poly(A)+ RNA transcripts all of which contain the entire coding sequence of the long form of PRL receptor. Each of these mammary mRNAs hybridized equally well with cDNA sequences encoding either the NH2 terminal, middle, or COOH terminal part of the rabbit mammary PRL receptor. The four mRNAs differed only in their 5'- and 3'-untranslated regions. The 10.5 kb mammary transcript was further shown to represent a primary transcript of nuclear origin. Among the various rabbit tissues tested, male and female adrenals, mammary gland, ovaries, and jejunum contained the highest level of prolactin receptor mRNA. The prolactin receptor gene was also expressed at moderate to weak abundance in uterus, liver, kidney, pancreas, testis and seminal vesicles. No prolactin receptor mRNA species were detected in adult muscle, lung, total brain, placental cotyledons and spleen, and in thymus from young animals. In all the rabbit tissues examined, the same four PRL receptor poly(A)+ RNA transcripts identified in the mammary gland were expressed and no additional transcript(s) were detected. Variations in the relative proportion of the 10.5 kb transcript and the two smaller transcripts were observed, while the ratio of the 3.4 and 2.7 kb mRNAs remained unchanged. These findings ask for the role of these different transcripts generated in the rabbit, all of which encode the same long form of PRL receptor precursor but have heterogenous 5'- and 3'-untranslated regions. Moreover, they suggest that the various forms of PRL receptor mRNA originate through differential splicing of a single PRL receptor gene.

Both the 5' untranslated region and the sequences surrounding the start site contribute to efficient initiation of translation in vitro

The role of RNA sequences in the 5' leader region between the cap site and initiating AUG in mediating translation was examined in vitro. Hybrid mRNAs were synthesized in which the cognate leader sequence was replaced with either optimized or compromised leader sequences, and translational efficiency was measured for six different coding regions. Translation was most efficient with a leader containing the 5' untranslated region from Xenopus beta-globin and an optimized initiation sequence. Compared with the cognate leaders, this hybrid was observed to increase translation of the various coding regions as much as 300-fold. The translational efficiencies of the different coding regions also varied substantially. In contrast to earlier suggestions that increased leader efficiency results from higher affinity of the leader for a limiting factor, our experiments suggest that increased translation from the beta-globin hybrid leader sequence results from more rapid initiation of translation.

Cloning of the gene for a human dopamine D5 receptor with higher affinity for dopamine than D1

Dopamine receptors belong to a superfamily of receptors that exert their biological effects through guanine nucleotide-binding (G) proteins. Two main dopamine receptor subtypes have been identified, D1 and D2, which differ in their pharmacological and biochemical characteristics. D1 stimulates adenylyl cyclase activity, whereas D2 inhibits it. Both receptors are primary targets for drugs used to treat many psychomotor diseases, including Parkinson's disease and schizophrenia. Whereas the dopamine D1 receptor has been cloned, biochemical and behavioural data indicate that dopamine D1-like receptors exist which either are not linked to adenylyl cyclase or display different pharmacological activities. We report here the cloning of a gene encoding a 477-amino-acid protein with strong homology to the cloned D1 receptor. The receptor, called D5, binds drugs with a pharmacological profile similar to that of the cloned D1 receptor, but displays a 10-fold higher affinity for the endogenous agonist, dopamine. As with D1, the dopamine D5 receptor stimulates adenylyl cyclase activity. Northern blot and in situ hybridization analyses reveal that the receptor is neuron-specific, localized primarily within limbic regions of the brain; no messenger RNA was detected in kidney, liver, heart or parathyroid gland. The existence of a dopamine D1-like receptor with these characteristics had not been predicted and may represent an alternative pathway for dopamine-mediated events and regulation of D2 receptor activity.

Different forms of soluble cytoplasmic mRNA binding proteins and particles in Xenopus laevis oocytes and embryos

To gain insight into the mechanisms involved in the formation of maternally stored mRNPs during Xenopus laevis development, we searched for soluble cytoplasmic proteins of the oocyte that are able to selectively bind mRNAs, using as substrate radiolabeled mRNA. In vitro mRNP assembly in solution was followed by UV-cross-linking and RNase digestion, resulting in covalent tagging of polypeptides by nucleotide transfer. Five polypeptides of approximately 54, 56 60, 70, and 100 kD (p54, p56, p60, p70, and p100) have been found to selectively bind mRNA and assemble into mRNPs. These polypeptides, which correspond to previously described native mRNP components, occur in three different particle classes of approximately 4.5S, approximately 6S, and approximately 15S, as also determined by their reactions with antibodies against p54 and p56. Whereas the approximately 4.5S class contains p42, p60, and p70, probably each in the form of individual molecules or small complexes, the approximately 6S particles appears to consist only of p54 and p56, which occur in a near-stoichiometric ratio suggestive of a heterodimer complex. The approximately 15S particles contain, in addition to p54 and p56, p60 and p100 and this is the single occurring form of RNA-binding p100. We have also observed changes in the in vitro mRNA binding properties of these polypeptides during oogenesis and early embryonic development, in relation to their phosphorylation state and to the activity of an approximately 15S particle-associated protein kinase, suggesting that these proteins are involved in the developmental translational regulation of maternal mRNAs.

Transcriptional antitermination in the bgl operon of E. coli is modulated by a specific RNA binding protein

Regulation of bgl operon expression in E. coli occurs by a mechanism involving antitermination of transcription at two termination sites within the operon. The bglG gene product is absolutely required for this process. Here we provide evidence that BglG is an RNA binding protein that recognizes a specific sequence located just upstream of each of the terminators. The sequence was delimited using a series of specific oligonucleotide probes. Mutational analysis of this sequence indicates that the protein requires a specific RNA secondary structure for recognition. We propose that BglG prevents transcription termination by binding to nascent RNA and blocking formation of the terminator structure.

Higher order structure of Balbiani ring premessenger RNP particles depends on certain RNase A sensitive sites

Specific premessenger ribonucleoprotein (RNP) particles, the Balbiani ring (BR) granules from Chironomus tentans salivary glands, were treated with RNase A to study the effect of RNA strand breaks on the higher order structure of the particles. Isolated, radioactively labeled BR granules, known to sediment at 300 S, were digested with RNase A and centrifuged in sucrose gradients. The fractionated particles were subsequently analyzed using electron microscopy and caesium chloride centrifugation. At a low RNase concentration, most of the 300 S particles disintegrated completely, and no metastable degradation products were observed. At intermediate RNase concentrations, no 300 S particles were left, but a minor fraction of the BR granules had unfolded and sedimented at 160 S. These granules could represent particles modified during the RNase treatment or represent a more slowly degrading subfraction of the particles. At a high RNase concentration, no RNP particles at all remained in the gradient. The rapid disintegration of the majority of the BR granules was investigated further by electrophoretic analysis of RNA in the remaining particles. During the RNase treatment BR granules, still sedimenting at 300 S, accumulated strand breaks; in fact, as many as 50 to 100 nicks in the 37 kb RNA could be tolerated. It was concluded from RNA analyses that the disintegration of the BR granules was not dependent on any single nick in the RNA, nor on the accumulation of a certain number of nicks, but rather on one or a few critical strand breaks. We propose that there are organizing sequences essential for particle integrity; once these sequences are nicked, the premessenger RNP particles are rapidly and completely degraded.

Expression of alpha- and beta-human chorionic gonadotropin subunits in cultured human cells

We surveyed several human cell lines for production of alpha- and beta-human chorionic gonadotropin (hCG) under a variety of conditions known to induce gene expression. alpha- and beta-hCG subunits were monitored in culture media by specific radioimmunoassays and were shown to be quite sensitive to serum refeeding and growth state of all cell types studied. The permanent line JEG-3 secreted both alpha- and beta-subunits whereas HeLa cells secreted only the alpha-subunit. Production of both subunits was augmented in these permanent cell lines, for each growth state, by pretreating cells with 5-azacytidine; in contrast, spontaneous beta-hCG production by normal human fibroblasts (four of six strains) was only rarely increased after 5-azacytidine treatment, and more often was suppressed by 30 to 40%. Three of five strains from inherited chromosomal breakage syndromes produced immunoassayable beta-hCG spontaneously, two of which increased secretion upon treatment with either UV or mitomycin C. Surprisingly, one normal cell strain of fetal origin was induced to secrete alpha-hCG, but not beta-hCG, after UV irradiation. JEG-3 and HeLa cells produced detectable cognate mRNA for alpha- or beta-hCG subunits or both by Northern and S1 nuclease protection analyses, whereas such transcripts from untransformed human fibroblasts were consistently below detectable levels. Quantitation of beta-hCG mRNA by RNA:RNA annealing kinetics indicates that even the fibroblast strain producing the highest secreted beta-hCG levels contained cognate mRNAs at only approximately 0.1 per cell. We conclude that hCG expression in human fibroblasts is strongly repressed at the transcriptional level, although a variety of conditions (growth state, serum refeeding, cell senescence, or DNA damage) can affect the level of "leaky" expression, at least in some responding fraction of cells.

The gene for human acidic fibroblast growth factor encodes two upstream exons alternatively spliced to the first coding exon

We have isolated two cDNA clones encoding human acidic fibroblast growth factor (aFGF) which represent the utilization of alternative upstream exons in aFGF mRNA. Isolation and sequence analysis of genomic clones spanning the first coding exon and each of the upstream sequences confirms that the divergent 5' sequences are separate exons, spliced alternatively to the first coding exon 34 nucleotides upstream of the initiator AUG codon. Restriction mapping of the genomic clones provides a minimum size estimate of 45 kilobase pairs for the aFGF locus.