Selection of transfected mammalian cells

Analysis of gene function frequently requires the formation of mammalian cell lines that contain the studied gene in a stably integrated form. Approximately one in 10(4) cells in a transfection will stably integrate DNA (the efficiency can vary depending on the cell type). Therefore, a dominant, selectable marker is used to permit isolation of stable transfectants. In the first part of this unit, the procedure for determining selection conditions and the resulting stable transfection is presented and the most commonly used selectable markers are discussed. The second protocol includes conditions for thirteen markers commonly used for selection of mammalian cells. A third protocols describes selection of transfected cells from the total population soon after transfection with plasmids that express both the gene of interest and a selection tag. Optimization of transfection conditions can be facilitated by a simple staining assay detailed in a support protocol.

High-throughput expression of fusion proteins

The expression of putative ORFs as fusion proteins can accelerate research greatly. The availability of an epitope tag allows the use of Western blotting as an efficient means to identify useful recombinant plasmids, which can then be used to study protein function. In addition, the epitope tag can be extremely useful in downstream applications such as protein purification, immunolocalization, and immunoprecipitation experiments. The preceding protocols should be applicable to a variety of expression vectors, and should be useful in the identification of functional plasmids. The protocols require no exotic equipment and can be adapted for use in high-throughput transfection assays, protein purification protocols, and immunolocalization studies.

The transcriptional response of yeast to saline stress

Adaptation to changes in extracellular salinity is a critical event for cell survival. Genome-wide DNA chip analysis has been used to analyze the transcriptional response of yeast cells to saline stress. About 7% of the genes encoded in the yeast genome are induced more than 5-fold after a mild and brief saline shock (0.4 m NaCl, 10 min). Interestingly, most responsive genes showed a very transient expression pattern, as mRNA levels dramatically declined after 20 min in the presence of stress. A quite similar set of genes increased expression in cells subjected to higher saline concentrations (0.8 m NaCl), although in this case the response was delayed. Therefore, our data show that cells respond to saline stress by inducing the expression of a very large number of genes and suggest that stress adaptation requires regulation of many cellular aspects. The transcriptional induction of most genes that are strongly responsive to salt stress was highly or fully dependent on the presence of the stress-activated mitogen-activated protein kinase Hog1, indicating that the Hog1-mediated signaling pathway plays a key role in global gene regulation under saline stress conditions.

In vivo selection of single-chain antibodies using a yeast two-hybrid system

The current methodology for screening libraries of single-chain fragments of immunoglobulin variable domains (sFvs) utilizes bacterial phage systems. We have developed a unique in vivo selection protocol combining a modified yeast two-hybrid assay with a novel prey vector expressing sFvs. The viability of the system is demonstrated with the screen of a sFv library cloned into a yeast two-hybrid prey vector for molecules that target the bait ATF-2, a member of the CREB/ATF family of transcriptional regulatory proteins. The isolated sFv was capable of recognizing ATF-2 in vitro on Western blots and in vivo in mammalian cells.

Genome-scale cloning and expression of individual open reading frames using topoisomerase I-mediated ligation

The in vitro cloning of DNA molecules traditionally uses PCR amplification or site-specific restriction endonucleases to generate linear DNA inserts with defined termini and requires DNA ligase to covalently join those inserts to vectors with the corresponding ends. We have used the properties of Vaccinia DNA topoisomerase I to develop a ligase-free technology for the covalent joining of DNA fragments to suitable plasmid vectors. This system is much more efficient than cloning methods that require ligase because the rapid DNA rejoining activity of Vaccinia topoisomerase I allows ligation in only 5 min at room temperature, whereas the enzyme's high substrate specificity ensures a low rate of vector-alone transformants. We have used this topoisomerase I-mediated cloning technology to develop a process for accelerated cloning and expression of individual ORFs. Its suitability for genome-scale molecular cloning and expression is demonstrated in this report.

Fusion of green fluorescent protein with the Zeocin-resistance marker allows visual screening and drug selection of transfected eukaryotic cells

Green fluorescent protein (GFP) and the Zeocin-resistance gene Sh ble (ZeoR) were fused together to generate a bifunctional protein for the identification and selection of transfected mammalian cells. Expression of this hybrid selectable marker, GFP-ZeoR, was visually detected and conferred Zeocin resistance in prokaryotes and eukaryotes. This selectable marker provides a way to determine transient transfection efficiencies in tissue culture cells using fluorescence microscopy. Expression of the GFP-ZeoR was also used to identify and select stable mammalian cell lines expressing a heterologous gene. Selection was efficient and GFP fluorescence provides an excellent, noninvasive technique to monitor the success of Zeocin selection during the development of the stable cell lines. This hybrid resistance gene combines the functional properties of the Zeocin-resistance marker and GFP and should be useful for combined selection and fluorescence in a variety of organisms.

Identification of a novel marker for primordial smooth muscle and its differential expression pattern in contractile vs noncontractile cells

The assembly of the vessel wall from its cellular and extracellular matrix components is an essential event in embryogenesis. Recently, we used the descending aorta of the embryonic quail to define the morphological events that initiate the formation of a multilayered vessel wall from a nascent endothelial cell tube (Hungerford, J.E., G.K. Owens, W.S. Argraves, and C.D. Little. 1996. Dev. Biol. 178:375-392). We generated an mAb, 1E12, that specifically labels smooth muscle cells from the early stages of development to adulthood. The goal of our present study was to characterize further the 1E12 antigen using both cytological and biochemical methods. The 1E12 antigen colocalizes with the actin cytoskeleton in smooth muscle cells grown on planar substrates in vitro; in contrast, embryonic vascular smooth muscle cells in situ contain 1E12 antigen that is distributed in threadlike filaments and in cytoplasmic rosette-like patterns. Initial biochemical analysis shows that the 1E12 mAb recognizes a protein, Mr = 100,000, in lysates of adult avian gizzard. An additional polypeptide band, Mr = 40,000, is also recognized in preparations of lysate, when stronger extraction conditions are used. We have identified the 100-kD polypeptide as smooth muscle alpha-actinin by tandem mass spectroscopy analysis. The 1E12 antibody is an IgM isotype. To prepare a more convenient 1E12 immunoreagent, we constructed a single chain antibody (sFv) using recombinant protein technology. The sFv recognizes a single 100-kD protein in gizzard lysates. Additionally, the recombinant antibody recognizes purified smooth muscle alpha-actinin. Our results suggest that the 1E12 antigen is a member of the alpha-actinin family of cytoskeletal proteins; furthermore, the onset of its expression defines a primordial cell restricted to the smooth muscle lineage.

The PerFect lipid optimizer kit for maximizing lipid-mediated transfection of eukaryotic cells

The transfection efficiencies of a panel of eight uniquely different lipid reagents has been evaluated with two other commercially available lipids for use in transfecting a diversity of eukaryotic cell lines. The PerFect lipids are available individually or together in an optimization panel format that can be tested in any given cell line, enabling one to evaluate the optimal lipid for transfecting each individual cell line. Our results demonstrate that no single lipid is optimal for plasmid transfection over a broad range of cell types, thus emphasizing the need for multiple unique lipid reagents and a simple format for testing their transfection efficiency on a given cell type.

The proximal regulatory element of the interferon-gamma promoter mediates selective expression in T cells

Interferon-gamma (IFN-gamma) is produced by natural killer cells and certain subsets of T cells, but the basis for its selective expression is unknown. Within the region between -108 and -40 base pairs of the IFN-gamma promoter are two conserved and essential regulatory elements, which confer activation-specific expression in T cells. This report describes studies indicating that the most proximal of these two regulatory elements is an important determinant of its restricted expression. The proximal element is a composite site that binds members of the CREB/ATF, AP-1, and octamer families of transcription factors. Jun is essential for activation-induced transcription and binds preferably as a heterodimer with ATF-2. In contrast, CREB appears to dampen transcription from this element. The CpG dinucleotide in this element is selectively methylated in Th2 T cells and other cells that do not express IFN-gamma, and methylation markedly reduces transcription factor binding. As a target for DNA methylation and for binding of transcription factors that mediate or impede transcription, this element appears to play a central role in controlling IFN-gamma expression.

Selective isolation of transiently transfected cells from a mammalian cell population with vectors expressing a membrane anchored single-chain antibody

We present here a novel technology for the rapid selection of transiently transfected cells from total populations in culture. This system utilizes recombinant antibody technology to produce a 'molecular hook' by displaying a hapten-binding single-chain antibody (sFv) on the surface of transfected cells. Mammalian cell lines from several origins were transiently transfected with a plasmid (pHook-1) that encodes an sFv fused with a transmembrane anchor and found to express and display the functional hapten-binding sFv on their membranes. Transfected cells were selected from total populations in culture by virtue of their ability to bind to hapten-coated magnetic beads. Some cell lines were able to display sFv sufficient for selection as early as 2 h post-transfection. SK-BR-3 human breast carcinoma cells were co-transfected with pHook-1 and pCR31acZ (expresses beta-galactosidase), selected, and assayed for beta-galactosidase activity. The positive correlation between sFv and beta-galactosidase expression in these cells (95% of selected cells also expressed beta-galactosidase activity) suggests that pHook-1 will be useful in isolating cells co-expressing an exogenous gene of interest. Another vector was constructed in which a gene of interest may be expressed from the same plasmid as the sFv 'hook'. This construct (pHook-2) allows the selection of a homogenous population of cells expressing exogenous genes without co-transfection or the generation of stable transfectants. In experiments where the lacZ gene was co-expressed with the sFv 'hook' from this single plasmid, 100% of 293 human kidney cells and 100% of SK-BR-3 cells selected with antigen-coated magnetic beads stained positively for beta-galactosidase activity. We propose that this system will be a valuable tool for studying the acute and chronic effects of the expression of a variety of wild type and mutant proteins.

Signal transduction and the regulation of cell growth

The need to understand and characterize critical steps controlling mitogenesis is a problem being addressed in many laboratories today. Genetic damage in cancer cells is a universal characteristic, and the search for the targets of this damage has led to the study of numerous complex signal transduction pathways which are involved in regulating cellular proliferation and differentiation. One important signal transduction cascade involved in controlling mitogenesis is the mitogen-activated protein kinase pathway. It is now clear that both tyrosine kinase growth factor receptors and G protein-coupled receptors regulate this pathway. Recent work from many groups has resulted in the initial characterization of the events that lead to mitogen-activated protein kinase activation by these different receptors. This review will contain a brief background of G proteins and their potential role in controlling cell growth.

Production of a recombinant bovine enterokinase catalytic subunit in the methylotrophic yeast Pichia pastoris

We describe the heterologous expression of a 26.3 kD protein containing the catalytic domain of bovine enterokinase (EKL) in the methylotrophic yeast Pichia pastoris. A highly active protein is secreted and glycosylated, and it has the native amino-terminus of EKL. The cDNA encoding EKL was cloned with the KEX2 protease cleavage site following the alpha mating factor prepro secretion signal from Saccharomyces cerevisiae. The secreted EKL was easily purified from the few native proteins found in the P. pastoris fermentation supernatant, using ion exchange and affinity chromatography. The yield of the purified EKL was 6.3 mg per liter of fermentation culture. This is significantly higher than previous reports of expressions in E. coli and COS cells. The ability of this highly specific protease to cleave immediately after the carboxyl-terminal residue of the (Asp)4-Lys recognition sequence allows regeneration of native amino-terminal residues of recombinant proteins. Its application is demonstrated by the removal of thioredoxin (TrxA), and polyhistidine fusion partners from proteins of interest.

Upstream stimulatory factor, a basic-helix-loop-helix-zipper protein, regulates the activity of the alpha-glycoprotein hormone subunit gene in pituitary cells

In an effort to determine whether basic-helix-loop-helix (bHLH) proteins are important in pituitary-specific expression of the alpha-glycoprotein hormone subunit gene, we examined the effect of the dominant negative HLH protein, Id, on the activity of the alpha-subunit gene promoter in pituitary cells. Id over-expression reduces the expression of alpha-subunit reporter genes in either alpha T3-1 gonadotrope-derived or alpha TSH thyrotrope-derived cells. A deletion fragment containing nucleotides from -131 to +44 of the human alpha-subunit promoter is inhibited to a similar degree as a -244 to +44 fragment in alpha T3-1 cells. Nuclear proteins in alpha T3-1 cells bind two potential bHLH protein binding sites (E-boxes, alpha EB1 and alpha EB2) present in this fragment but not to mutations that specifically alter only these sequences. An antibody-specific for upstream stimulatory factor, a widely expressed bHLH-leucine zipper protein, is able to inhibit factor binding to the alpha EB2 sequences but not the alpha EB1 site. Mutating the alpha EB1 element of the alpha-subunit promoter decreases basal activity of this promoter to about 42% of control levels in alpha T3-1 cells. A mutation that abolishes upstream stimulatory factor binding, either alone or in combination with the alpha EB1 mutation, reduces basal activity of the promoter to approximately 21% of control levels in alpha T3-1 cells and abolishes the decrease in promoter activity seen when Id is overexpressed. These results demonstrate that the bHLH family of proteins are important regulators of alpha-subunit gene expression in pituitary cells.

Identification of a second ATF/CREB-like element in the herpes simplex virus type 1 (HSV-1) latency-associated transcript (LAT) promoter

The herpes simplex virus type-1 (HSV-1) latency-associated transcript (LAT) promoter (LP) has been shown to function in a cell type-specific manner. We have constructed an extensive series of PCR deletion mutations of the LP from nucleotides +1 to -348 to delineate the specific sequences involved in the cell type-specific activity of the HSV-1 LP. This series of 5' LP deletion constructs has been transiently transfected into both C1300 (neuronal) and L929 (nonneuronal) cells. When nucleotides -75 to -83 were added to nucleotides +1 to -74, a three- to fourfold C1300-specific increase in promoter activity was observed. In addition, when sequences upstream of nucleotide -211 were added to nucleotides +1 to -211, a second threefold increase in promoter activity was observed in C1300 cells. To begin to understand the biochemical basis for these observations, we have examined the interaction of a segment of the HSV-1 LP (nucleotides -54 to -134) with factors present in neuronal and nonneuronal nuclear extracts. This region of the LP contains the sequence most proximal to the transcriptional start site demonstrated to be involved in cell type-specificity (nucleotides -75 to -83). By coupling the functional studies with electrophoretic mobility shift (EMS), oligonucleotide competition EMS, and antibody supershift EMS analyses, we have demonstrated that members of the activating transcription factor (ATF)/cyclic-AMP response element binding protein (CREB) transcription factor family interact with nucleotides -75 to -83 of the HSV-1 LP. The identification of a novel ATF/CREB-like element in the HSV-1 LP may facilitate the understanding of neuronal factors which regulate LAT expression during HSV-1 infection. These studies may ultimately provide additional insight concerning the role of HSV-1 LAT in the regulation of viral latency and reactivation.

Escape from the antiproliferative effect of transforming growth factor-beta 1 in LLC-PK1 renal epithelial cells

Transforming growth factor-beta 1 (TGF-beta 1) usually inhibits proliferation of epithelial cells. We find that LLC-PK1 renal tubular epithelial cells develop rapid in vitro resistance to the inhibitory effects of TGF-beta 1 and subsequently proliferate in response to TGF-beta 1. This unique response to TGF-beta 1 is not observed in another renal tubular epithelial cell line (MDCK cells). The proliferative response to TGF-beta 1 is additive to that produced by other growth factors. The proliferative response to TGF-beta 1 occurs despite an effect of TGF-beta 1 to suppress epidermal growth factor stimulated c-myc mRNA as determined by Northern analyses. These results suggest that LLC-PK1 cells develop rapid resistance to TGF-beta 1 inhibition of proliferation in vitro and that this resistance occurs despite continued suppression of c-myc mRNA.

Ras mutations in human melanoma: a marker of malignant progression

In this study we address whether there is an association between ras mutations and disease progression in malignant melanoma. DNA was extracted from 100 paraffin-embedded melanomas and sequences around the 12th, 13th and 61st codons of N-, H-, and K-ras were amplified using the polymerase chain reaction and probed for single base pair mutations using synthetic oligonucleotide probes. Thirty-six melanomas contained mutations, which in 25 cases (69%) occurred at the 61st codon of N-ras. The results from dot blot hybridizations were confirmed by subcloning and sequencing the polymerase chain reaction products from two tumors. No ras mutations were found in Clark's level I melanomas, whereas 19% of level II and 45% of the more advanced primary tumors contained ras mutations (Chi squared test: p < 0.05). The median Breslow thickness of primary melanomas with ras mutations was 0.72 mm, significantly thicker than the 0.42 mm of melanomas without mutations (Mann-Whitney U test, p = 0.042). Ras mutations were found more frequently in primary tumors from continuously exposed skin (56%) than tumors from intermittently or non-sun exposed sites (21%). Fifty percent of locally recurrent and 47% of metastatic melanomas had ras mutations. We conclude that ras mutations occur in a subset of melanomas from sun-exposed skin as a feature of tumor progression.

Transactivation by the human T-cell leukemia virus Tax protein is mediated through enhanced binding of activating transcription factor-2 (ATF-2) ATF-2 response and cAMP element-binding protein (CREB)

The human T-cell leukemia virus type I (HTLV-I)-encoded transcriptional activator protein Tax is strongly implicated in HTLV-I pathogenesis. Tax regulates HTLV-I gene expression through three 21-base pair (bp) repeat enhancer elements located in the transcriptional control region of the virus. Tax does not bind these elements directly, but mediates transactivation through the cellular transcription factors that recognize a cAMP response element (CRE)-like sequence centered within each of the 21-bp repeats. In this report, we identify activating transcription factor-2 (ATF-2) and CRE-binding protein (CREB) as the principal T-cell proteins that bind the three 21-bp repeats in vitro. Purified Tax protein augments the level of RNA synthesis induced by ATF-2 and CREB in a cell-free transcription assay, providing evidence that Tax cooperates with these cellular proteins to activate HTLV-I transcription. Furthermore, Tax dramatically increases the binding of both the T-cell-derived and recombinant forms of ATF-2 and CREB to each of the 21-bp repeats. The target sequences for this enhancement reside within the DNA binding/dimerization domains of these proteins. These data suggest that Tax transactivates HTLV-I gene expression by increasing the number of bound ATF-2 and CREB molecules at the viral promoter.

Transforming growth factor-beta 1 regulation of signal transduction in two renal epithelial cell lines

The present studies examine the effect of transforming growth factor-beta 1 (TGF-beta 1) on signal transduction pathways in two cultured renal epithelial cell lines. TGF-beta 1 promotes basal and agonist-stimulated adenylate cyclase activity in LLC-PK1 but not MDCK cell membranes. TGF-beta 1 stimulation of LLC-PK1 membrane adenylate cyclase activity occurs quickly and can be attenuated by pertussis toxin pretreatment. Both TGF-beta 1 and adenosine 3',5'-cyclic monophosphate (cAMP) exert comparable effects on [3H]thymidine uptake in LLC-PK1 cells, suggesting that TGF-beta 1 regulation of adenylate cyclase activity potentially plays a role in mediating biological responses to TGF-beta 1. The activities of protein kinase C and phospholipase A are not affected by TGF-beta 1 in either LLC-PK1 or MDCK cells. Both TGF-beta 1 and epidermal growth factor (EGF) increase expression and induce the appearance of new forms of the cAMP response element binding protein (CREB) in LLC-PK1 cells. These effects of TGF-beta 1 and EGF on CREB appear to be specific since neither TGF-beta 1 nor EGF alters expression of an activating transcription factor in LLC-PK1 cells. The effect of TGF-beta 1 and EGF to alter expression of CREB does not affect CREB binding to its regulatory element in LLC-PK1 cell lysates. These results suggest that some of the biological effects of TGF-beta 1 may be attributed to stimulation of adenylate cyclase activity and cAMP formation as well as to enhanced expression and/or modification of the CREB transcription factor in LLC-PK1 cells.

Helix-loop-helix proteins are present and differentially expressed in different cell lines from the anterior pituitary

The basic-helix-loop-helix (bHLH) class of transcriptional activators, important in the establishment of many different cell lineages, share two important properties: the ability to heterodimerize with other members of this family and to bind DNA containing the loose consensus sequence CANNTG. This study takes advantage of these shared characteristics to begin to address whether or not bHLH proteins are present in pituitary cells. Gel-shift and Southwestern assays using an oligonucleotide containing a bHLH binding consensus sequence demonstrate that pituitary-specific proteins are present in extracts from adult pituitary tissue and pituitary cell lines and bind specifically to this sequence. Pituitary extracts were also found to contain several factors which interact with Id protein, a negative regulator of bHLH activity, in Far-Western assays of protein-protein interactions. Finally, messenger RNA for Id is present in pituitary cell lines but is absent in adult pituitary tissue. Together, these studies indicate that bHLH proteins are present in pituitary cells and their levels are differentially regulated in the separate cell types.

Antagonist-occupied human progesterone B-receptors activate transcription without binding to progesterone response elements and are dominantly inhibited by A-receptors

When antagonist-occupied steroid receptors have agonist-like effects, the clinical consequences are grave. We present evidence that human progesterone B-receptors (hPRB) when occupied by progesterone antagonists, inappropriately activate transcription by an unusual mechanism that does not require the canonical progesterone response element (PRE). In HeLa cells cotransfected with a PRE-tk-chloramphenicol acetyltransferase reporter and a hPRB expression vector, strong transcription is seen not only when receptors are activated by the agonist R5020, but also in the presence of the three antiprogestins, RU486, ZK112993, and ZK98299. Human PRB occupied by ZK98299 do not bind to a PRE, suggesting that the transcriptional stimulation is independent of DNA binding. Indeed, a tk-chloramphenicol acetyltransferase promoter-reporter lacking the PRE loses transcriptional activation by the agonist, but retains transactivation by the three antagonists. The PRE-independent antagonist-induced transcription requires that hPRB have an intact DNA-binding domain, but hPR target gene specificity is not required, because a hPRB mutant that binds an estrogen response element still activates transcription. It appears that antagonist-occupied hPR activate transcription without binding to a PRE, perhaps by interacting with tethering proteins instead. Even a gene that is not a normal progesterone target could be aberrantly activated. Human cells contain equimolar amounts of hPRB and the N-terminally truncated natural isotype, hPRA. Unlike hPRB, hPRA are not transcriptionally activated by progesterone antagonists. We, therefore, tested the effects of antagonists when the two receptor isotypes are coexpressed and found that A-receptors can annul the inappropriate transcription by B-receptors. Thus, when both receptor forms are present, the hPRA phenotype is dominant. Moreover, pure hPRB/hPRA heterodimers, produced by fos/jun leucine zipper domain-hPR chimeras, also have the inactive transcriptional phenotype of hPRA. Our studies suggest not only that the two hPR isotypes are functionally quite different, but also that some of the agonist-like transcriptional effects of antagonist-occupied B-receptors proceed through novel mechanisms.

A multifunctional prokaryotic protein expression system: overproduction, affinity purification, and selective detection

A series of plasmid vectors, pRSET A, B, and C, have been developed for high-level protein expression in prokaryotes and have been characterized. Based upon the T7 RNA polymerase-driven pET system, the pRSET vectors encode recombinant proteins as fusions with a multifunctional leader peptide containing a hexahistidyl sequence for purification on Ni(2+)-affinity resins, a tyrosine residue for radioiodination, and an enterokinase proteolytic cleavage site for leader peptide removal. Monoclonal antibodies (MAbs) to two epitopes on the leader peptide, which also contains amino acids 1-12 of the T7 gene 10 major capsid protein, were developed and provide for universal immunological detection of pRSET-expressed fusion proteins. Subcloning of protein-encoding DNA is facilitated by an 11-site polylinker which is offset for all three ribosomal reading frames, and an f1(+) origin of DNA replication permits single-stranded DNA synthesis for site-directed mutagenesis protocols. Representative fusion proteins overexpressed in Escherichia coli were successfully purified under both denaturing and nondenaturing conditions by single-step Ni2+ affinity chromatography. Purification was independent of recombinant protein solubility in sonicated or freeze-thawed E. coli lysates. Isolation of MAbs for selective recognition of either of two leader peptide epitopes was demonstrated by immunoprecipitation, but this selectivity was less evident under conditions for Western blotting. In combining the utility of T7 RNA polymerase-directed expression with several recent advances in protein purification and detection, the pRSET vectors will serve as a powerful resource for a variety of studies in protein biochemistry.

Structural determinants outside of the leucine zipper influence the interactions of CREB and ATF-2: interaction of CREB with ATF-2 blocks E1a-ATF-2 complex formation

Dimerization of leucine zipper-containing proteins has been associated characteristically with the formation of a coiled-coil structure between two compatible leucine zipper motifs. In the present study we demonstrate the association of the leucine zipper of cAMP response element-binding protein (CREB) with a zinc finger motif of ATF-2. The association of the CREB leucine zipper with the ATF-2 zinc finger is stabilized if the ATF-2 leucine zipper is intact, implying that the preferred interactive structure of ATF-2 juxtaposes the amino-terminal zinc finger motif of this protein with the carboxy-terminal leucine zipper of this same protein. Furthermore, we demonstrate that the association of the CREB leucine zipper with the ATF-2 zinc finger in vitro blocks the association of the adenoviral E1a protein with ATF-2. Similarly, overexpression of full-length CREB, or a truncated version of this protein corresponding to the carboxy-terminal 74 amino acids that make up the DNA-binding and dimerization domains, can block the ATF-2-mediated transcriptional stimulation by E1a in vivo. Mutation of the ATF-2 zinc finger motif stimulates DNA binding of this protein, and abolishes interactions with E1a and CREB proteins. These results demonstrate that the structural conformation of ATF-2 is critical for DNA binding and protein-protein interactions and, further, that leucine zippers can mediate protein-protein interactions with structural motifs other than leucine zippers.

Modification of DNA topoisomerase II activity via direct interactions with the cyclic adenosine-3',5'-monophosphate response element-binding protein and related transcription factors

DNA topoisomerase II (topo II) is an essential nuclear enzyme which catalyzes the interconversions of various forms of DNA. As predicted from the human topo II cDNA, the enzyme contains a potential leucine zipper protein dimerization motif. We therefore tested whether topo II could enter protein-protein interactions with other better characterized leucine zipper-containing proteins and determined if these interactions could modify topo II enzymatic activity in vitro. By far Western analyses, a large C-terminal fragment of human topo II was shown to interact with the DNA binding and dimerization regions of either cAMP response element binding protein (CREB) or the activating transcription factor-2. The C-terminal topo II fragment also interacted with full-length c-Jun, but not with full-length c-Fos. Using CREB as a prototype, the effect of this interaction on various topo II catalytic activities was assessed in vitro. CREB, at a 1- to 10-fold molar excess relative to topo II, inhibited site-specific DNA cleavage activity on a 242-base pair fragment of the human alpha-glycoprotein hormone subunit gene promoter. Very high CREB concentrations (400-fold excess) apparently inhibited topo II DNA relaxation activity, but this result was likely a direct effect of CREB on the topology of the DNA substrate. More interestingly, a 10-fold molar excess of CREB stimulated topo II decatenation activity, the essential function of this enzyme in cell division. This stimulatory effect could also be elicited by c-Jun, which interacts with topo II, but not by c-Fos, which does not bind topo II in our in vitro assay. Since similar amounts of CREB reduced the abundance of topo II DNA cleavage products from the human alpha-CG promoter yet also stimulated decatenation activity, it can be concluded that either: 1) CREB stimulated the religation rate of topo II; or 2) CREB directed topo II to a new cleavage site present on the decatenation substrate but not present on the limited alpha-CG promoter. The structural requirements for topo II protein-protein interactions were also investigated. Site-directed mutations which destroyed the putative topo II leucine zipper did not disrupt topo II protein-protein interactions. Since the putative leucine zipper in topo II does not appear to mediate protein-protein interactions, we propose that an alternate as yet uncharacterized structure is involved in the association of topo II with itself and other regulatory proteins.

Activating transcription factor-2 DNA-binding activity is stimulated by phosphorylation catalyzed by p42 and p54 microtubule-associated protein kinases

Recent studies have detailed the ability of activating transcription factor-2 (ATF-2) to mediate adenoviral E1a stimulation of gene expression; however, an endogenous regulator for the transcriptional activity of this protein has not been described. To characterize the regulation of ATF-2 activity, we have expressed full-length and truncated peptides corresponding to various regions of the ATF-2 protein in bacteria and the baculovirus insect cell system. Bacterially expressed truncated (350-505) but not full-length ATF-2, was able to bind a consensus cAMP response element-containing oligonucleotide, suggesting the N-terminal moiety may serve as a negative regulator of DNA-binding activity. In contrast, the full-length ATF-2 protein expressed in Spodoptera frugiperda (Sf9) cells using a recombinant baculovirus was fully competent to bind DNA. Protein phosphatase 2A reversed the DNA-binding activity by dephosphorylating the ATF-2 polypeptide. Microtubule-associated protein kinase catalyzed the phosphorylation and stimulated the DNA-binding activity of bacterially expressed full-length ATF-2. Phosphopeptide mapping of phosphorylated ATF-2 proteins identified a single peptide in the N-terminal moiety of ATF-2 phosphorylated by p42 or p54 microtubule-associated protein kinase. Therefore, we propose that phosphorylation of this regulatory site is sufficient to induce an allosteric structural change in the ATF-2 protein, which allows dimerization and subsequent DNA binding.

Role of the cyclic adenosine 3',5'-monophosphate response element in efficient expression of the rat thyrotropin receptor promoter

The "minimal" promoter region of the TSH receptor gene, -195 to -39 basepairs (bp), exhibits basal promoter activity, thyroid specificity, and negative regulation by TSH via its cAMP signal. In FRT thyroid cells and by comparison to pTRCAT5'-199, 5'-deletion mutants of chloramphenicol acetyltransferase (CAT) constructs from -199 to -150 bp of the minimal promoter decrease basal CAT activity by 50%, whereas continued deletion to -146 bp increases activity more than 4-fold. Continued deletion to -131 bp results in basal activity less than that of the -199 bp construct. An octameric cAMP response element (CRE)-like sequence, TGAGGTCA, is within -146 to -131 bp and starts at -139 bp. Its mutation to a consensus CRE (TGACGTCA) or AP1 (TGAGTCA) site or mutation of several residues flanking its 3'-terminus can improve promoter activity as much as 8-fold compared to pTRCAT5'-199. A nonpalindromic mutation to CGAGGACA decreases basal promoter activity to the level of the 199-bp minimal promoter. The CRE-like sequence between -139 and -132 bp is a constitutive enhancer of promoter activity in FRT thyroid cells, since, ligated to a simian virus-40-promoter-driven CAT gene, it increases CAT activity in the absence of forskolin in proportion to copy number and independent of direction or position. It can, however, function as a cAMP-responsive CRE, as evidenced by the fact that forskolin increases the activity of the same simian virus-40-promoter-driven CAT gene constructs in Buffalo rat liver (BRL) cells. DNAase-I footprinting shows that the CRE region is protected by a purified binding region peptide of the CRE-binding protein, activating transcription factor-2, and recombinant AP1 (human c-jun) as well as by BRL, FRT, and FRTL-5 rat thyroid cell nuclear extracts. Gel mobility shift analyses show that multiple CRE-binding proteins in the BRL, FRT, and FRTL-5 cell nuclear extracts form complexes with the CRE-like site, that one of these is CRE-binding protein, and that all form complexes with mutant sequences of the CRE-like site in a manner that exactly parallels their effects on constitutive enhancer function in FRT thyroid cells. We show, therefore, that the CRE-like site in the minimal TSH receptor promoter functions as a constitutive enhancer of promoter activity in FRT thyroid cells yet is a cAMP-responsive CRE.(ABSTRACT TRUNCATED AT 400 WORDS)

Structure/function relationships of CREB/ATF proteins

Signal-transduction pathways converge ultimately at the level of transcriptional activation to produce specific patterns of gene expression in response to environmental stimuli. The initiation of transcription mediated by these signaling pathways is regulated by the coordinate expression and/or activation of specific transcription factors that bind to the control regions of genes. Specific insights into the mechanisms underlying transcriptional activation have recently arisen from studies of the structure and functions of these transcription factors. The CREB/ATF family of transcriptional transactivating proteins has only recently been discovered and appears to provide a link between the regulation of gene expression in response to activators of cellular signaling pathways and the regulation of gene expression by viral transactivating proteins. In addition, these proteins may be involved in the normal regulation of growth and differentiation. Understanding the nature and importance of the role(s) of these proteins in the normal regulation of growth and differentiation will have profound influences on the understanding of the aberrant regulation of these processes during oncogenesis.

Overexpression and purification of transcriptionally competent CREB from a recombinant baculovirus

Signal transduction and viral stimulatory pathways converge ultimately at the level of transcriptional activation to influence the expression of a variety of cellular genes in response to environmental stimuli and developmental signals. Recent studies have implicated the cyclic AMP-responsive element-binding protein (CREB) to be involved in mediating transcriptional activation in response to multiple varied stimuli, including (1) stimulation of the protein kinase A signal transduction pathway; (2) membrane depolarization and increases in intracellular calcium; and (3) viral induced gene expression. In order to study the structure and functional mechanisms of CREB actions in these systems, full-length CREB-327 was expressed in Spodoptera frugiperda (Sf9) cells with the baculovirus expression vector system. The expressed CREB, which is phosphorylated and localized in the nucleus, is capable of enhancing the transcription of a reporter gene containing the CRE sequence in a cell-free transcription assay. Approximately 12.5 mg of purified CREB per liter of infected Sf9 cell culture can be obtained. These large amounts of purified protein will facilitate studies of the structure and functions of this important transcriptional regulatory protein.

HBV X protein alters the DNA binding specificity of CREB and ATF-2 by protein-protein interactions

The hepatitis B virus (HBV) X gene product trans-activates viral and cellular genes. The X protein (pX) does not bind independently to nucleic acids. The data presented here demonstrate that pX entered into a protein-protein complex with the cellular transcriptional factors CREB and ATF-2 and altered their DNA binding specificities. Although CREB and ATF-2 alone did not bind to the HBV enhancer element, a pX-CREB or pX-ATF-2 complex did bind to the HBV enhancer. Thus, the ability of pX to interact with cellular factors broadened the DNA binding specificity of these regulatory proteins and provides a mechanism for pX to participate in transcriptional regulation. This strategy of altered binding specificity may modify the repertoire of genes that can be regulated by transcriptional factors during viral infection.

Identification of multiple nuclear factors that interact with cyclic adenosine 3',5'-monophosphate response element-binding protein and activating transcription factor-2 by protein-protein interactions

Understanding the nature and importance of protein-protein interactions in the mechanisms of eukaryotic gene expression is essential to understanding the normal and aberrant regulation of gene transcription. Using 125I-labeled cAMP response element-binding protein (CREB) and activating transcription factor-2 (ATF-2) recombinant peptides to probe Western blots of HeLa nuclear extracts, we have identified multiple separate nuclear factors that form specific protein-protein interactions with these leucine zipper-containing transcriptional regulatory proteins. The interaction is specific because preincubation of blots with cold homologous protein blocks the binding of labeled protein, whereas preincubation of blots with cold heterologous protein has no effect on labeled protein interactions. Although these studies focus on two specific transactivators, CREB and ATF-2, the approach is of general use for the study of other leucine zipper-containing mammalian transcription factors. Furthermore, in addition to allowing the detection of protein-protein interactions of CREB and ATF-2 with nuclear factors, we have used this strategy to isolate cDNA clones expressing these nuclear proteins. We demonstrate that CREB will form heterodimers with ATF-1, but not ATF-2, Jun, Fos, or C/EBP whereas, ATF-2 will form heterodimers with Jun and Fos, but not with C/EBP or ATF-1. This strategy, therefore, allows a systematic approach to identifying, characterizing, and cloning proteins involved in the control of eukaryotic transcriptional regulation. The identification and characterization of the components of eukaryotic transcription complexes will allow studies that address the molecular mechanisms of normal and abnormal control of cellular gene expression.

Cyclic-AMP-responsive transcriptional activation of CREB-327 involves interdependent phosphorylated subdomains

Cyclic AMP-regulated gene expression is mediated by specific phosphoproteins (CREBs) which bind to cAMP-responsive elements of gene promoters. By analyzing the transactivation activities and phosphorylations in vivo of deletion and point mutated chimeric fusion proteins of the placental CREB-327, in which the DNA-binding domain is replaced by the heterologous binding-domain of the yeast transcription factor GAL4, we localized the cAMP-responsive and phosphorylated domain to a minimal-essential sequence module of 46 amino acids (residues 92-137). This serine-rich, multiply-phosphorylated sequence consists of at least three interdependent subdomains required for transcriptional activation. Although phosphorylation of serine-119 by cyclic AMP-dependent protein kinase A is necessary for transcriptional activation, such activation requires both a phosphorylated heptadecapeptide domain located ten residues amino terminal to the serine-119 and an eleven-residue domain carboxyl terminal to the serine-119. Deletion of these two domains does not impair phosphorylation of serine-119. Further, deletion of the carboxyl-terminal domain does not alter phosphorylation of the heptadecapeptide domain. We propose that akin to the phosphorylation-dependent activation of enzymes, the transcriptional transactivation functions of CREB-327 involve a phosphorylation-dependent allosteric conformational mechanism.

Characterization of a cAMP-regulated enhancer-binding protein

The transcription regulation of many hormone genes is modulated by intracellular second messengers such as cAMP. The cAMP response element binding protein, CREB, binds to the 8 base pair CRE enhancer, TGACGTCA, that is found in the 5'-flank of certain genes including those for somatostatin and the alpha-subunit of human chorionic gonadotropin. The recent characterization of CREB and CREB-related cDNA clones, combined with Southwesterns and Northern blot analyses, reveals a family of transcription factors that dimerize via a leucine zipper motif and bind to the CRE through positively charged basic regions. The CREB cDNA encoding a 327 residue protein is transcriptionally activated via phosphorylation by protein kinases, including the cAMP-dependent protein kinase-A.

Multiple adenosine 3',5'-cyclic [corrected] monophosphate response element DNA-binding proteins generated by gene diversification and alternative exon splicing

The protein sequence deduced from the open reading frame of a human placental cDNA encoding a cAMP-responsive enhancer (CRE)-binding protein (CREB-327) has structural features characteristic of several other transcriptional transactivator proteins including jun, fos, C/EBP, myc, and CRE-BP1. Results of Southwestern analysis of nuclear extracts from several different cell lines show that there are multiple CRE-binding proteins, which vary in size in cell lines derived from different tissues and animal species. To examine the molecular diversity of CREB-327 and related proteins at the nucleic acid level, we used labeled cDNAs from human placenta that encode two different CRE-binding proteins (CREB-327 and CRE-BP1) to probe Northern and Southern blots. Both probes hybridized to multiple fragments on Southern blots of genomic DNA from various species. Alternatively, when a human placental c-jun probe was hybridized to the same blot, a single fragment was detected in most cases, consistent with the intronless nature of the human c-jun gene. The CREB-327 probe hybridized to multiple mRNAs, derived from human placenta, ranging in size from 2-9 kilobases. In contrast, the CRE-BP1 probe identified a single 4-kilobase mRNA. Sequence analyses of several overlapping human genomic cosmid clones containing CREB-327 sequences in conjunction with polymerase chain reaction indicates that the CREB-327/341 cDNAs are composed of at least eight or nine exons, and analyses of human placental cDNAs provide direct evidence for at least one alternatively spliced exon. Analyses of mouse/hamster-human hybridoma DNAs by Southern blotting and polymerase chain reaction localizes the CREB-327/341 gene to human chromosome 2. The results indicate that there is a dichotomy of CREB-like proteins, those that are related by overall structure and DNA-binding specificity as well as those that are related by close similarities of primary sequences.

Characterization of a cyclic AMP regulatory element DNA-binding protein

Signal transduction pathways converge ultimately at the level of transcriptional activation to produce specific patterns of gene expression in response to environmental stimuli. The initiation of transcription mediated by these signaling pathways is regulated by the coordinate expression and/or activation of specific transcription factors that bind to the control regions of eukaryotic genes. Specific insights into the mechanisms underlying transcriptional activation have arisen from the studies of the structure and functions of eukaryotic transcription factors. One of these factors, a cyclic (c)AMP response element binding protein (CREB), has only recently been discovered and appears to play a key role in the regulation of gene expression in response to the activation of the cAMP-dependent protein kinase A. The transcription factors related to CREB, c-jun and c-fos, mediate transcriptional responses to activators of protein kinase C.

Distinct adenosine 3',5'-monophosphate and phorbol ester-responsive signal transduction pathways converge at the level of transcriptional activation by the interactions of DNA-binding proteins

Evidence is presented that distinct cellular signal transduction pathways involving cAMP-dependent protein kinase-A and phorbol ester-stimulated protein kinase-C coordinately modulate gene transcription through common as well as distinct cis-acting elements and DNA-binding proteins. When transfected and expressed in HeLa and placental JEG-3 cells, fusion reporter plasmids that differ only by a single base deletion or addition to interconvert the octameric cAMP-responsive element TGACGTCA (CRE) to form the heptameric phorbol ester-responsive element TGACTCA (TRE) are differentially regulated by cAMP and phorbol esters [12-O-tetradecanoyl phorbol-14-acetate (TPA)]. Transcription directed by the CRE is stimulated by cAMP and not TPA, although the basal expression mediated by this element in JEG-3 and HeLa cells is augmented by endogenous protein kinase-C activity. In contrast, TRE mediates transcriptional responses to both cAMP and TPA, and the two agents together give synergistic responses. Inhibition of cAMP-dependent protein kinase-A by expression of a minigene encoding a peptide inhibitor of A-kinase abolishes the response of TRE to cAMP alone as well as the cAMP-induced component of the synergistic response to treatment with both TPA and cAMP. Desensitization of the protein kinase-C dependent pathway by prolonged exposure of cells to phorbol esters eliminates the TPA-induced transcription by TRE and inhibits the TPA-induced component of the synergistic response to both cAMP and TPA. Therefore, both protein kinases, A and C, are involved in transcriptional activation by the TRE; the function of either kinase alone results in a moderate level of activity, but the combined results of both functionally stimulated kinases are synergistically positive. Electrophoretic mobility shift assays using whole extracts of JEG-3 cells indicate that a common factor(s) binds both TRE and CRE; however, another factor(s) that binds to the CRE will not bind to the TRE. Further, a latent regulatory enhancer element (URE) located upstream of the CRE's in the human alpha gonadotropin gene, although inactive when paired alone with the alpha 100 promoter, induces basal and stimulated transcriptional activity of both CRE and TRE on the average of 10- to 20-fold. The data support the existence of a gene regulatory network consisting of related cis-acting elements and DNA-binding proteins whose transcriptional activities are regulated by the convergent actions of protein kinases-C and -A.

Factors that determine cell-specific gene expression in pancreatic endocrine tumor cells

Phenotypically distinct islet tumor cell lines may recapitulate certain of the developmental pathways of normal islet cell differentiation by expressing a combinatorial set of positively and negatively acting DNA-binding proteins to allow for the programmed expression of genes encoding polypeptide hormones. The structure of one of these DNA-binding proteins, a cyclic AMP-responsive protein (CREB) that binds specific DNA regulatory elements in the somatostatin gene, has been deduced from the sequence of a cloned cDNA. The CREB protein contains a DNA-binding domain separate from a cAMP-dependent protein kinase A activation domain. Further characterizations of the genes encoding the DNA-binding proteins should help to elucidate the cellular processes involved in islet cell differentiation and the genesis of tumors.

Cyclic AMP-responsive DNA-binding protein: structure based on a cloned placental cDNA

Cyclic AMP (cAMP) is an intracellular second messenger that activates transcription of many cellular genes. A palindromic consensus DNA sequence, TGACGTCA, functions as a cAMP-responsive transcriptional enhancer (CRE). The CRE binds a cellular protein of 38 kD in placental JEG-3 cells. A placental lambda gt11 library was screened for expression of specific CRE-binding proteins with the CRE sequence as a radioactive probe. A cDNA encoding a protein of 326 amino acids with the binding properties of a specific CRE-binding protein (CREB) was isolated. The protein contains a COOH-terminal basic region adjacent to a sequence similar to the "leucine zipper" sequence believed to be involved in DNA binding and in protein-protein contacts in several other DNA-associated transcriptional proteins including the products of the c-myc, c-fos, and c-jun oncogenes and GCN4. The CREB protein also contains an NH2-terminal acidic region proposed to be a potential transcriptional activation domain. The putative DNA-binding domain of CREB is structurally similar to the corresponding domains in the phorbol ester-responsive c-jun protein and the yeast transcription factor GCN4.

Structural determinants for transcriptional activation by cAMP-responsive DNA elements

A transcriptional cAMP-responsive enhancer element (CRE) consisting of the 8-base pair (bp) palindrome, 5' TGACGTCA 3', is found in several eukaryotic genes. We analyzed the effects on gene transcription of point mutations within the CRE, the influence of the bases surrounding the CRE, and the requirements for transcriptional synergism of tandemly repeated CREs. When inserted as an oligonucleotide with restriction enzyme linker sites, the 8-bp CRE itself is as active in conferring cAMP responsivity on an enhancerless chloramphenicol acetyltransferase reporter plasmid as is a single copy of the choriogonadotropin alpha (CG alpha), twice repeated 18-bp sequence containing the CRE. Point mutations in the first (T to A), fourth (C to G), or eighth (A to T) positions of the CRE, when contained within the CG alpha 18-bp sequence, each inhibited transcriptional activity greater than 90%. However, the identical eighth position A to T mutation occurs in the cAMP-responsive sequence of the vasoactive intestinal peptide (VIP) gene, and that mutant sequence in the context of the adjacent bases of the native VIP sequence is maximally cAMP responsive when inserted in the reporter plasmid. The substantially reduced activity of the core 8-bp CRE when synthesized as a cassette including the adjacent bases of the rat glucagon or bovine parathyroid hormone gene further emphasizes the restrictive influence of particular surrounding sequences. Active oligonucleotides containing the 8-bp palindrome and different but equally permissive contexts have comparable properties in transfected reporter genes and gel mobility-shift assays. The pair of tandemly repeated 18-bp elements containing the CRE in the CG alpha gene synergistically stimulate transcription either with paired native CREs or when one native CRE is paired with one mutant CRE, suggesting the presence of cooperative interactions. Tandem insertion of more than two 18-bp sequences, or insertion of additional sequences between the two CREs, inhibits transcription. These observations indicate that the contexts of the bases adjacent to CREs exert profound influences on the transcriptional activities mediated by the cAMP-responsive elements.

Cyclic AMP and phorbol ester-stimulated transcription mediated by similar DNA elements that bind distinct proteins

cAMP and phorbol esters mediate cellular metabolism by the activation of distinct signal transduction pathways consisting of a cascade of sequential protein phosphorylations. An important consequence of the activation of these pathways is the stimulation of gene transcription by way of interactions of specific proteins with DNA control elements. The 8-base-pair (bp) DNA consensus sequence TGACGTCA [cAMP response element (cAMP-RE)] has been shown to confer cAMP responsivity on transcription from various promoters, and the closely related 7-bp consensus sequence TGA-(C or G)TCA [phorbol 12-myristate 13-acetate response element (PMA-RE)] lends transcriptional responsiveness to phorbol esters. In the JEG-3 placental cell line we find that several variants of the cAMP-REs fused to a gonadotropin alpha promoter chloramphenicol acetyltransferase reporter gene mediate responsiveness to cAMP but not to phorbol esters. The PMA-RE is responsive to phorbol esters but also imparts submaximal sensitivity to cAMP in the JEG-3 cells and in the Hep G2 hepatoma cell line. The transcriptional activities of cAMP-RE and PMA-RE are markedly influenced by the composition of the neighboring bases, but different sequences are permissive for the activity of the cAMP-RE versus the PMA-RE. The two signaling agents together display a supraadditive effect on reporter genes containing active PMA-REs but not cAMP-REs. Gel-mobility-shift and UV cross-linking analyses show that distinct proteins bind to the two control elements. One protein of 38 kDa binds to the cAMP-RE and several proteins of 48-84 kDa bind to the PMA-RE.

Hypothalamic peptides affect the ratios of GH and PRL cells: role of cell division

In the present study, we attempted to ascertain whether the differential effects of hypothalamic peptides (GRF, LHRH, CRF, TRH) on the proportions of GH and PRL secreting cells in culture were dependent upon cell proliferation. This was accomplished by chronically exposing rat pituitary cells to selected hypothalamic peptides in the presence or absence of cytosine arabinoside (an inhibitor of cell proliferation) and then measuring hormone release by plaque assays. Our results suggest that the effects of these peptides on the percentages of GH and PRL cells are manifest via both proliferation-dependent and proliferation-independent mechanisms.

A cellular basis for functionally releasable pools in somatotropes

In an attempt to define the cellular basis for the phenomenon of releasable pools, we compared the effects of two growth hormone (GH)-releasing peptides which differentially influence the dynamics of GH release. Monodispersed anterior pituitary cells from neonatal male rats were subjected to reverse hemolytic plaque assays for GH in the presence or absence of GH-releasing peptide (GHRP-6, an enkephalin-like hexapeptide) and rat GH-releasing factor (GRF). GRF increased the rate of plaque formation (an index of the rate of hormone release) from almost all somatotropes, whereas GHRP-6 influenced only half of these cells. Analysis of plaque sizes (which provides a relative index of the cumulative amount of hormone released per cell) revealed that GRF produced a bimodal frequency distribution of plaque sizes, demonstrating that some somatotropes released more hormone than others after treatment with a maximal dose of this secretagogue. This pattern contrasted with those of untreated and GHRP-6 treated somatotropes which each produced unimodal frequency distributions that were skewed to the left (toward smaller plaques) and were virtually superimposable at the end of a 4 h incubation. However, GHRP-6 greatly accelerated the rate at which the final size distribution pattern was attained. Taken together, these results suggest that GHRP-6 causes the immediate release of a limited pool of GH which is present only in a discrete subpopulation of somatotropes that respond to GRF. This pool may be identical to that which is released over a more prolonged period under basal conditions. Moreover, GRF appears to access a more substantial pool of hormone which is not released by GHRP-6. This pool is present in a small minority of somatotropes but probably accounts for a larger portion of the GH released by pituitaries stimulated with GRF.

Existence of somatotrope subpopulations which are differentially responsive to insulin-like growth factor I and somatostatin

It is generally accepted that hypothalamic somatostatin and hepatic insulin-like growth factor I (IGF-I)/somatomedin-C act directly on the pituitary to inhibit GH release, but it is not known whether all somatotropes are responsive to these agents. In the present study, we used a reverse hemolytic plaque assay to compare the acute (8 h) effects of somatostatin and IGF-I on the release of GH from individual cells in 24-h cultures of male rat pituitaries. Treatment with these factors caused comparable dose-dependent decreases in both the rate of plaque formation and the percentage of cells which released GH. In 8-h incubations, maximal (10(-8) M) doses of IGF-I or somatostatin alone decreased the percentage of GH-releasing cells to approximately the same degree (from 34.4% in controls to 29.7% and 28.4%, respectively), yet the effects of these factors were additive when both agents were applied to the same cells (to 24.5%). When we analyzed the sizes of plaques (an index of the amount of hormone released per cell) which resulted from these treatments, we noted that somatostatin was a much greater suppressor (to 11% of control value) of GH release than IGF-I (60% of controls). Coincubation with 10(-8) M GH-releasing factor had no effect on the percentage of GH-releasing cells at 8 h but completely overrode the inhibitory effect of IGF-I on plaque size without affecting the somatostatin-induced decrease in this regard. Taken together, these data suggest that IGF-I and somatostatin act, at least in part, on separate subpopulations of rat somatotropes. Somatostatin is a much more effective inhibitor of total GH release than IGF-I and appears to affect most, if not all, somatotropes. In contrast, IGF-I acutely inhibits GH release (prevents plaque formation) from some somatotropes, but does not seem to affect the remaining GH cells.

Liver tissue produces a potent lactogen that partially mimics the actions of prolactin

An in vitro bioassay (based on the detection of casein release from isolated mammary cells by reverse hemolytic plaque assay) was used to detect a lactogenic factor secreted by liver tissue from suckled rats. This material stimulates casein release in the absence of added PRL and is actually more potent than PRL in this regard. The substance(s) possesses the following additional characteristics: it is released from liver tissue of lactating but not virgin female or male rats; when tested together with PRL, its effects are additive rather than synergistic; and unlike PRL, it does not increase the proportion of mammary cells committed to casein release. These findings are consistent with the possibility that a Liver Lactogenic Factor may mediate, at least in part, the lactogenic component of PRL's action. This functional relationship may be similar to that which exists between GH and the somatomedins.

Hypothalamic factors differentially affect the proportions of cells that secrete growth hormone or prolactin

Hypothalmic factors have been shown to regulate acutely the synthesis and release of adenohypophyseal hormones, yet few studies have investigated the long term effects of these agents on adenohypophyseal cell types. In the present study, we assessed the chronic influence of selected hypothalamic factors on the relative proportions of GH- and PRL-secreting cells in pituitary cultures derived from 5-day-old rats. Primary cultures were established and incubated for 6 days in the presence or absence of 0.1-microM doses of GH-releasing factor, LHRH, CRF, TRH, or vasoactive intestinal polypeptide and then subjected to reverse hemolytic plaque assays for analysis of the percentages of cells that released GH or PRL. In cultures from males, GH-releasing factor and LHRH treatment caused an increase in the proportion of PRL secretors and a commensurate decrease in the GH population. CRF increased PRL cells without affecting the GH secretors, while TRH reduced the percentages of both cell types, and vasoactive intestinal polypeptide had no effect. Virtually identical results were obtained for cells isolated from females. These results demonstrate that hypothalamic factors have the capacity to induce differential effects on the proportions of GH- and PRL-secreting cells. Surprisingly, our findings also show that hypothalamic factors that do not normally influence the acute release of GH or PRL can exert a chronic effect on the proportion of cells that secrete these hormones.

A novel bioassay for lactogenic activity: demonstration that prolactin cells differ from one another in bio- and immuno-potencies of secreted hormone

A reverse hemolytic plaque assay for detecting casein release from individual mammary cells in culture was developed as a bioassay for PRL. Treatment with rat PRL caused dose-related increases in the percentage of mammary cells that released casein and the average size of casein plaques that formed. The assay exhibited exquisite sensitivity (156 fg rat PRL per assay slide) and could be used to evaluate the biopotency of PRL released from individual cells. By combining this "plaque bioassay" with a standard plaque assay for measuring the secretion of immunoreactive PRL, it was possible to compare the bio- and immuno-potencies of hormone released from the same pituitary cells. The results of three separate studies revealed major differences among PRL secretors in these potency estimates. Given the existence of PRL variants with different biological and immunological efficacies, these findings suggest that PRL cells differ from one another in the molecular form(s) of hormone released.

Capacity of individual somatotropes to release growth hormone varies according to sex: analysis by reverse hemolytic plaque assay

A reverse hemolytic plaque assay was used to investigate the mechanism(s) underlying sexual differences in GH release which evolve at puberty in rats. The percentages of GH-secreting cells in 24-h pituitary cultures from each sex were similar for pituitary donors up to 30 days of age (range = 38.9% to 41.7% of all cells in culture, n = 3 separate experiments) but decreased by day 50. The decrease was more striking for females (to 24.1 +/- 0.3% mean +/- SE) than for males (to 33.2 +/- 1.1%). However, owing to the greater increase in total pituitary cell number exhibited by female rats at this time, the absolute numbers of somatotropes recovered from male and female pituitaries were almost identical on 50 and 100 days of age. To assess the secretory capacities of individual somatotropes, we measured the sizes of plaques formed. In prepubertal rats (days 10-30), the plaque areas under basal conditions were comparable for males and females at each age studied, and treatment with GH-releasing factor increased plaque sizes to approximately the same degree (10-fold) for both sexes at each age. However, by day 100, plaques that formed under both basal and stimulated conditions were consistently larger (P less than 0.01) for male than for female donors. Taken together, our results demonstrate that sexual differences in GH release are attributable to the secretory capacities of individual somatotropes rather than to differences in the numbers of GH cells in pituitaries of male and female rats.

Estradiol induces a shift in cultured cells that release prolactin or growth hormone

The process by which estradiol induces an increase in the percentage of prolactin cells in the rat pituitary requires clarification. In this study the effect of estradiol treatment on growth hormone and prolactin cells in monolayer cultures of male rat pituitaries was assessed by use of reverse hemolytic plaque assays. Treatment of cultures with 17 beta-estradiol (10 nM) for 6 days elevated significantly (P less than 0.01) the percentage of all cells in culture that released prolactin without affecting the percentage of growth hormone secretors. However, by sequentially assaying single cells for growth hormone and prolactin, we found that estradiol caused a marked increase (P less than 0.01) in the proportion of individual cells that released both growth hormone and prolactin and a commensurate decrease in the proportion that released growth hormone alone. These results demonstrate that estradiol acts directly on cultures of male pituitary cells to shift the proportions of cells that released growth hormone, prolactin, or both hormones and indicate that estradiol may convert cells that release only growth hormone to those that release both growth hormone and prolactin.

Analysis by plaque assays of GH and prolactin release from individual cells in cultures of male pituitaries. Evidence for functional heterogeneity within rat mammotrope and somatotrope populations

In this study we evaluated the quantitative influence of GRF and TRH on the rate of hormone secretion from single cells in cultures of male pituitaries. To accomplish this, we dispersed pituitaries from male rats with trypsin and cultured them for 24 or 48 h. Reverse hemolytic plaque assays for GH and prolactin were then performed on retrypsinized cultures to identify individual cells that secreted these hormones. Mammotropes and somatotropes were found to comprise 31.4 +/- 1.8 and 32.2 +/- 0.9% (mean +/- SE, n = 3 experiments), respectively, of all cells in 24-hour cultures. Immunocytochemical staining of different batches of cells from the same dispersions corroborated the proportions of these two cell types. Differences in the rate of basal hormone secretion were observed within each of these cell populations as evidenced by the gradual appearance of prolactin and GH plaques over a 4-hour period when incubations were conducted in the absence of stimulatory secretagogues. Addition of increasing concentrations of GRF (1 X 10(-10) -1 X 10(-7) M) or TRH (1 X 10(-9) -1 X 10(-6) M) to these incubations resulted in dose-related increases in the rate of GH and prolactin plaque formation, respectively. Maximal plaque development by somatotropes could be induced within 30 min of administering large doses of GRF, indicating that most, if not all somatotropes are responsive to this secretagogue. In contrast, approximately one third of all mammotropes could not be stimulated to form plaques acutely when subjected to similar treatment with TRH. This observation suggests that mammotropes are heterogeneous with respect to TRH responsiveness.(ABSTRACT TRUNCATED AT 250 WORDS)