Overexpression, biophysical characterization, and crystallization of ribonuclease I from Escherichia coli, a broad-specificity enzyme in the RNase T2 family.

We have constructed a strain that overproduces ribonuclease I of Escherichia coli and we have purified large quantities of the enzyme. Data from fluorescence, CD, and DSC measurements showed that it was a very stable protein. The conformation energy determined from urea and guanidine hydrochloride denaturation experiments was 11.5 kcal mol(-1) at pH 7.5. Thermal denaturation studies indicated that it had a T(m) of 64 degrees C at pH 4.0. RNase I belongs to the RNase T2/S-RNase group of endoribonucleases, but near the amino terminus it has an unusually long hydrophilic segment. Part of this was removed in the deletion construct, RNase I Delta(26-38). We have obtained crystals of both RNase I and of an enzyme-G2'p5'G complex in the P2(1) space group and oligonucleotide complexes with both wild type and mutant enzymes. The current study lays the groundwork for extensive investigation into the structure, function, and physical properties of this widely distributed group of ribonucleases.

Genomic organization and promoter analysis of the murine Id3 gene.

Overexpression of the helix-loop-helix motif-containing transcription inhibitor Id3 has been shown to repress muscle-specific gene expression. Consistent with its putative negative regulatory role in the myogenic process, Id3 is highly expressed in proliferating myoblasts but down regulated when myoblasts are induced to differentiate. To investigate how Id3 expression may be transcriptionally regulated, we isolated a mouse Id3 genomic DNA fragment and characterized its organization and promoter activity. Comparison of the Id3 gene from human and mouse demonstrated a conserved exon-intron organization in which the first intron interrupts the C-terminal protein coding region and the second intron interrupts the 3' untranslated region at analogous positions in the two species. Sequence analysis of the 5'-flanking region revealed an unexpected mouse strain-specific genetic polymorphism due to a single base substitution. Deletion analysis revealed that as little as 180 base pairs of the mouse Id3 promoter upstream of the transcription start site is sufficient for a high level of gene expression in proliferating C2C12 myoblasts. In particular, the region between the nucleotide position -180 and -34 appeared to be crucial for maximal reporter gene activity and interacted specifically with C2C12 nuclear proteins. Finally, we showed that, despite the creation of a putative transcription factor-binding site, the genetic polymorphism observed did not affect Id3 promoter activity in proliferating C2C12 cells.

Involvement of p27(kip1) and cyclin D3 in the regulation of cdk2 activity during skeletal muscle differentiation.

Terminal myogenic differentiation involves an irreversible transition from a proliferative state to a post-mitotic quiescent state. We showed here that in addition to the previously reported down regulation of G(1)-related cyclin-associated kinase activities, this transition was also accompanied by an extensive reorganization of the cyclin-cdk complexes, including a dramatic shift of cdk2 from cyclin A to cyclin D3. Moreover, the inhibition of cdk activity also correlated with an increase in the expression of the p27(kip1) cdk inhibitor and in its association with the cyclin-cdk2 complexes. Since depletion of p27 substantially reduced the cdk inhibitor activity present in differentiated muscle cells, we believe that the increase in p27 expression along with the reorganization of the cyclin-cdk2 complexes may play an important role in the inhibition of cdk2 activity during the differentiation process.

Furosemide stimulates K transport in HCD57 erythroid cells.

We examined the influence of serum and furosemide on K movement and cell volume in HCD57 cells, a murine erythroleukemia cell line, which require erythropoietin (EPO) for survival. We found that maintenance of cell volume depends on the concentration of serum in the culture medium. In isotonic medium containing 20% serum, HCD57 cells maintain their steady-state volume. In contrast, the cells shrink progressively as medium serum content is reduced. In serum-free medium, raising external K to 75 mm prevents cell shrinkage and a further increase in K to 145 mm results in swelling, revealing a role for K permeability in the regulation of cell volume. Of particular interest has been a serendipitous finding with furosemide. Below an external K concentration of 2.1 +/- 0.3 mm in medium containing 2% serum, furosemide inhibits K uptake, probably stemming from its well known inhibitory action on KCl cotransport. However, above that K concentration, furosemide stimulates K uptake in a dose-dependent manner. Moreover, furosemide potentiates cell shrinkage induced by serum withdrawal. These findings suggest that the transport machinery mediating cellular shrinkage, once primed by serum depletion, becomes receptive to a second stimulus.

Regulated association of microtubule-associated protein 2 (MAP2) with Src and Grb2: evidence for MAP2 as a scaffolding protein.

Microtubule-associated protein 2 (MAP2) and tau, which is involved in Alzheimer's disease, are major cytoskeletal proteins in neurons. These proteins are involved in microtubule assembly and stability. To further characterize MAP2, we took a strategy of identifying potential MAP2 binding partners. The low molecular weight MAP2c protein has 11 PXXP motifs that are conserved across species, and these PXXP motifs could be potential ligands for Src homology 3 (SH3) domains. We tested for MAP2 interaction with SH3 domain-containing proteins. All neuronal MAP2 isoforms bound specifically to the SH3 domains of c-Src and Grb2 in an in vitro glutathione S-transferase-SH3 pull-down assay. Interactions between endogenous proteins were confirmed by co-immunoprecipitation using brain lysate. All three proteins were also found co-expressed in neuronal cell bodies and dendrites. Surprisingly, the SH3 domain-binding site was mapped to the microtubule-binding domain that contains no PXXP motif. Src bound primarily the soluble, non-microtubule-associated MAP2c in vitro. This specific MAP2/SH3 domain interaction was inhibited by phosphorylation of MAP2c by the mitogen-activated protein kinase extracellular signal-regulated kinase 2 but not by protein kinase A. This phosphorylation-regulated association of MAP2 with proteins of intracellular signal transduction pathways suggests a possible link between cellular signaling and neuronal cytoskeleton, with MAP2 perhaps acting as a molecular scaffold upon which cytoskeleton-modifying proteins assemble and dissociate in response to neuronal activity.

Evidence of intramolecular regulation of the Dictyostelium discoideum 34 000 Da F-actin-bundling protein.

Intramolecular interaction within the Ca(2+)-regulated 34 kDa actin-bundling protein from Dictyostelium discoideum was found to contribute to the regulation of its actin-binding activity. Recombinant N-terminally truncated proteins aa77-295, 124-295, and 139-295 bound actin at > or = 2:1 stoichiometry, which is 5-fold greater than the intact protein aa1-295 as assessed by cosedimentation with F-actin. These proteins also have enhanced cross-linking activity as assessed by viscometry and electron microscopy. All truncated 34 kDa proteins failed to bind (45)Ca(2+) on blots and displayed Ca(2+)-insensitive binding with actin, although most proteins possessed intact putative EF-hand Ca(2+)-binding motifs. An intramolecular interaction within the 34 kDa protein was inferred from direct demonstrations of domain-domain interaction among the truncated 34 kDa proteins both in the presence and absence of actin. The intramolecular interaction between interaction zone 1 (aa71-123) and interaction zone 2 (aa193-254) is proposed to maintain the N-terminal inhibitory region (aa1-76) in close proximity with the strong actin-binding site (aa193-254) in order to modulate the interaction of the intact protein with actin filaments.

Three distinct F-actin binding sites in the Dictyostelium discoideum 34,000 dalton actin bundling protein.

The Dictyostelium 34 kDa protein is an actin bundling protein composed of 295 amino acids. However, the region(s) of the molecule that bind actin filaments is (are) unknown. Studies of the cosedimentation of 125I-34 kDa protein and F-actin show that the 34 kDa protein binds to F-actin with positive cooperativity and Hill coefficients of 1.9 and 3.0, for filaments 4.9 microm and 0.6 microm, respectively. The Hill coefficient is larger for short filaments that are more efficiently bundled than long filaments, suggesting that one of the binding sites is used in interfilament contacts or contributes to filament orientation within the bundle. Three distinct actin binding sites were identified using a synthetic peptide, protein truncations, and a novel epitope library screening method. The ability to bind actin was assessed by 125I-F-actin overlays under denaturing and nondenaturing conditions, cosedimentation, viscometry, and pyrene-labeled actin disassembly. The three actin binding domains were identified as amino acids 1-123, 193-254, and 279-295. The 62 amino acid domain (193-254) can cosediment with F-actin. The estimated Kapp obtained by the disassembly of pyrene-labeled actin was 0.11 microM and 2.7 microM for the amino acids 1-123 and 279-295, respectively. These results identify three distinct regions of the 34 kDa protein that may contribute to the positive cooperative formation of F-actin bundles.

Interactive risk factors for treatment adherence in a chronic psychotic disorders population.

This study identified the unique and primary contributions of several concurrent risk factors for poor adherence to treatment recommendations in a clinic population of individuals with chronic psychotic disorders, i.e. 48% had DSM-IV diagnoses of schizoaffective disorder, 38% had schizophrenia, paranoid type, 12% had schizophrenia, undifferentiated type, and 2% had affective disorder with psychotic features. The target cohort consisted of 87 consecutive admissions to a continuing day treatment program. As part of a services-oriented quality assurance program, clinical staff completed rating scales for all patients. These included the BASIS-32 rating scale, which consisted of the following five subscales: psychosis; depression/anxiety; impulsive/addictive behavior; relation to self and others; and daily living and role functioning, and the Working Alliance Inventory-short form (therapist version), which consisted of the following three subscales: goal; task; and bond. These data were used to identify risk factors that weaken a patient's adherence to medication and non-medication treatment during the first 2 weeks of treatment in the clinic. Medication treatment consisted of both typical and atypical neuroleptic medications, with most patients being on multiple medications. Correlational analyses suggested that many of the risk factor variables were significantly associated with poor treatment adherence. Regression analyses suggested that the degree of psychoticism was most strongly associated with poor adherence to medication treatment and that difficulties relating to self and others were the strongest predictor of poor adherence to non-medication treatment. A large-sample services research design such as this can begin to determine patterns of associations between previous identified risk factors and poor treatment adherence in individuals with chronic psychotic disorders.

Cloning, expression, and characterization of a novel guanylate-binding protein, GBP3 in murine erythroid progenitor cells.

We report the molecular cloning of a novel guanylate-binding protein (GBP), termed mouse GBP3 (mGBP3) in Friend virus-induced mouse erythroid progenitor (FVA) cells. The 71-kDa mGBP3 belongs to a family of known GBPs that contain the first two consensus motifs, GXXXXGK(S/T) and DXXG, but lack the third element, (N/T)KXD, found in typical GTP-binding proteins. Recombinant mGBP3 protein, expressed using a baculovirus expression system, binds to agarose-immobilized guanine nucleotides (GTP, GDP and GMP). Moreover, mGBP3 has been found to have an intrinsic GTPase activity with K(m) and Vmax values of 77 +/- 4 microM and 21 +/- 0.5 pmol min-1 microgram-1 of protein, respectively. The mGBP3 is distinct from the other GBPs, in that it does not have an isoprenylation/methylation motif CAAX at the carboxyl terminus. The mGBP3 appears to be localized in the cytosol based on immunofluorescence staining. Although the mGBP3 transcript is expressed to a varying degree in numerous mouse tissues, the message is most abundant in FVA cells. The mGBP3 transcript increases in FVA cells undergoing differentiation to a maximum within a few hours and then decreases to an undetectable level by 24 h. These results, taken together, suggest that mGBP3 is a novel member of a family of guanylate-binding proteins, which plays a role in the erythroid differentiation. The nucleotide sequence reported in this paper has been submitted to the GenBank with accession number U44731.

Overexpression, purification, and characterization of recombinant Dictyostelium discoideum calcium-regulated 34,000-dalton F-actin bundling protein from Escherichia coli.

The Dictyostelium discoideum 34-kDa protein is an F-actin bundling protein that demonstrates diverse distributions in the cell during cell shape changes and cell movement. The protein is expressed at a very low level in the amoeba, just 0.4% of the total cell protein. This presents a challenging problem when purifying sufficient protein for structural and biochemical studies. The purification procedure is lengthy and yields only a few milligrams of protein. An alternative protein expression system, that of the bacterial T7 expression system, was used to produce large quantities of recombinant 34-kDa protein (r34-kDa). The soluble r34-kDa protein constitutes up to a quarter of the total bacterial protein, and was purified to homogeneity by a modification of the purification procedure for the native D. discoideum 34-kDa protein (N34-kDa). The r34-kDa possesses all the same functional characteristics as the N34-kDa protein with respect to its interactions with F-actin in vitro: it bound to and cross-linked F-actin, mediated F-actin bundle formation, directly bound calcium, and demonstrated calcium-sensitive F-actin binding activities.

Glucocorticoid effects on the skeletal muscle differentiation program: analysis of clonal proliferation, morphological differentiation and the expression of muscle-specific and regulatory genes.

We examined the effect of glucocorticoids on the proliferation and differentiation of skeletal muscle cells using the C2C12 cell line. We found that treatment with glucocorticoids enhanced muscle cell differentiation but had only minor effects on the clonal growth rate of C2C12 cells. The stimulatory effect of glucocorticoids on myogenic differentiation was reflected in the increased expression of muscle-specific genes, creatine kinase (CK) and acetylcholine receptor gamma subunit (AChR). Dexamethasone had no effect on CK and AChR mRNA stability and enhanced transcription from a CAT reporter genes containing the 3.3kb 5' flanking region of the murine CK gene (-3300MCK-CAT). Since dexamethasone did not affect the expression levels of the myogenic regulatory genes such as myoD and myogenin, the enhancement of muscle-specific transcription might reflect an increase in the functional activity of the regulatory proteins. Other possible mechanisms involved in the differentiation-enhancing effect of glucocorticoids are discussed.

Inhibition of muscle-specific gene expression by Id3: requirement of the C-terminal region of the protein for stable expression and function.

We have examined the role of an Id-like protein, Id3 (also known as HLH462), in the regulation of muscle-specific gene expression. Id proteins are believed to block expression of muscle-specific genes by preventing the dimerization between ubiquitous bHLH proteins (E proteins) and myogenic bHLH proteins such as MyoD. Consistent with its putative role as an inhibitor of differentiation, Id3 mRNA was detected in proliferating skeletal muscle cells, was further induced by basic fibroblast growth factor (bFGF) and was down-regulated in differentiated muscle cultures. Overexpression of Id3 efficiently inhibited the MyoD-mediated activation of the muscle-specific creatine kinase (MCK) reporter gene. Deletion analysis indicated that the C-terminal 15 amino acids of Id3 are critical for the full inhibitory activity while deleting up to 42 residues from the C-terminus of the related protein, Id2, did not affect its ability to inhibit the MCK reporter gene. Chimeric protein containing the N-terminal region of Id3 and the C-terminus of Id2 was also non-functional in transfected cells. In contrast, wild-type Id3, the C-terminal mutants, and the Id3/Id2 chimera could all interact with the E-protein E47in vitro. Additional studies indicated that truncation of the Id3 C-terminus might have adversely affected the expression level of the mutant proteins but the Id3/Id2 chimera was stably expressed. Taken together, our results revealed a more complex requirement for the expression and proper function of the Id family proteins than was hitherto expected.

Modulation of muscle creatine kinase promoter activity by the inducible orphan nuclear receptor TIS1.

TIS1, an inducible orphan nuclear receptor, was originally isolated as a tumour-promoter-inducible gene in mouse 3T3 cells and later shown to be induced by growth factors and other extracellular stimuli. We show here that TIS1 mRNA was expressed in proliferating C2C12 mouse skeletal muscle cells out that the level of TIS1 expression increased during muscle differentiation. Overexpression of TIS1 transactivated muscle creatine kinase (MCK) reporter genes containing as little as 80 bp of the proximal 5' flanking region. In contrast, a promoterless TIS1 construct and a frameshift mutant TIS1 construct were unable to transactivate the MCK reporter gene. Moreover, the effect exerted by TIS1 appeared to be selective for the MCK promoter. Treatment of C2C12 cells with forskolin, which is known to induce TIS1 expression, also stimulated MCK reporter gene activity. Interestingly, in vitro translated TIS1 protein failed to bind to the MCK promoter region, suggesting that the transactivation effect of TIS1 may be mediated without direct interaction of the protein with the MCK promoter DNA. Collectively, these results suggest that changing levels of TIS1 may help to modulate the expression of MCK, and perhaps other muscle-specific genes, in response to physiological changes.

Physical and functional interactions between the transcriptional inhibitors Id3 and ITF-2b. Evidence toward a novel mechanism regulating muscle-specific gene expression.

We have used an interaction cloning strategy to identify an inhibitory isoform of the ITF-2 transcription factor, ITF-2b, that interacts with the transcriptional inhibitor Id3/HLH462. The interaction was confirmed in vitro, and inside intact myogenic C2C12 cells. As expected, overexpression of either Id3/HLH462 or ITF-2b effectively inhibited the activation of the muscle-specific creatine kinase promoter by the myogenic transcription factor MyoD. However, when overexpressed simultaneously, ITF-2b and Id3/HLH462 counteracted each other's inhibitory effect to produce a reduced overall inhibition. Moreover, while ITF-2b inhibited the creatine kinase promoter, it acted as a weak transactivator on an artificial promoter consisting of three tandem copies of the consensus myogenic factor DNA binding site. Further investigation indicated that the ITF-2b/MyoD heterodimer bound to its specific DNA binding site in vitro, and the DNA binding was effectively blocked by Id3/HLH462. Additional analysis revealed the presence of transcripts for both the activating (ITF-2a) and inhibitory (ITF-2b) isoforms in differentiating C2C12 cultures, suggesting that both isoforms might participate in regulating the differentiation process. Taken together, this study reveals a more complex pattern of regulatory interactions involving the helix-loop-helix proteins than was previously anticipated.

Uroguanylin: cloning of preprouroguanylin cDNA, mRNA expression in the intestine and heart and isolation of uroguanylin and prouroguanylin from plasma.

Uroguanylin is a small peptide isolated from opossum urine that activates membrane guanylate cyclases. We report the isolation by molecular cloning of cDNAs encoding the 109 amino acid preprouroguanylin containing the active uroguanylin peptide at its C-terminus. Preprouroguanylin mRNAs of 1.2 kb were detected throughout the small and large intestine and in the atria and ventricles of heart, but not in kidney, stomach or liver. Transfection of COS-1 cells with the uroguanylin cDNA resulted in prouroguanylin secretion. Both uroguanylin and prouroguanylin were isolated from opossum plasma. Thus, uroguanylin is made by the intestine and heart and circulates as a bioactive form of uroguanylin and the inactive prouroguanylin.

Signal-transduction-pathway-specific desensitization of expression of orphan nuclear receptor TIS1.

Induction of many primary-response genes by a variety of stimuli occurs in a transient manner. The precise mechanism responsible for these transient kinetics is not completely understood. We report here that the orphan nuclear receptor, TIS1, which is a potent sequence-specific transcription factor, was transiently induced by the adrenergic agonist isoprenaline in the C2C12 skeletal-muscle cell line. Moreover, we showed that the rapid decline in mRNA level after peak induction was due in part to a specific desensitization of the isoprenaline-mediated induction pathway. Desensitization of the induction response presumably occurred at the level of the receptor, as agents that either bypass the adrenergic receptor or activate alternative signalling pathways were able to induce TIS1 expression in the desensitized cells. However, stimulation by agents that directly activate intracellular enzymes also resulted in the signal-transduction-pathway-specific desensitization of TIS1 inducibility. Our results suggest that the pathway-specific nature of the desensitization process may be important for directing an integrated response to multiple physiological stimuli.

Differential regulation of primary response gene expression in skeletal muscle cells through multiple signal transduction pathways.

One of the earliest cellular responses to growth factors is the rapid induction of primary response genes. One group of such genes was originally isolated as tetradecanoyl phorbol acetate (TPA) inducible sequences (TIS genes) from mouse 3T3 cells. Proteins encoded by the TIS genes include two transcription factors: TIS8 (also known as egr1/NGFIA/zif268) and TIS1 (also known as NGFIB/nur77/N10). We have examined the inducibility of these two genes in a skeletal muscle cell line in response to agents that have been reported to block muscle differentiation. We report here that basic fibroblast growth factor (bFGF) induced the expression of both TIS1 and TIS8 in mouse C2C12cells. Both genes were also inducible by TPA while forskolin which activates the cAMP-dependent pathway induced TIS1 but not TIS8. Down-regulation of protein kinase C (PKC) activity by TPA pretreatment repressed the bFGF induction of TIS1 but had little effect on the bFGF-stimulated expression of TIS8. Moreover, while both TPA and bFGF stimulated the hyperphosphorylation of c-RAF and the activity of MAP kinase, TPA pretreatment failed to block RAF phosphorylation or the stimulation of MAP kinase activity by bFGF. Induction of the two TIS genes in skeletal myoblasts therefore appeared to be dependent to different extents on the activation of protein kinase A (PKA), PKC and MAP kinase.

Rapid induction of mouse virus-like (VL30) element transcripts by erythropoietin in murine erythroid progenitor cells.

We examined the activation of genes induced by erythropoietin (Epo) in erythroid progenitor cells that were isolated from the spleens of mice infected with anemia-inducing strain of the Friend virus. These erythroid progenitor cells, termed FVA cells, undergo in vitro differentiation to erythrocytes under the influence of Epo within 2 to 3 days. We used a differential hybridization procedure to screen a cDNA library constructed from FVA cells that were treated with Epo 2U/mL in the absence of serum for 2 hours. Of 20,000 recombinant phages, 47 plaques hybridized preferentially to cDNA probe prepared from Epo-stimulated cells. We found at least three different Epo-responsive genes (ERGs) and one of them corresponds to the mouse virus-like (VL30) element, similar to already reported BVL-1. The induction of VL30, which was evident within 30 minutes after Epo exposure, reached a maximum by 1 hour and remained stable for up to 4 hours. The treatment of FVA cells with cycloheximide (CHX) 10 micrograms/mL, which in itself activates the expression of VL30 caused a superinduction of the Epo signal. Changes in intracellular Ca2+ concentrations, either raised by ionomycin or depleted by EGTA, had no effect on the Epo-induced VL30 expression. In addition, protein kinase C (PKC) inhibitors such as staurosporine (3 mumol/L) or H7 (20 mumol/L) and a tyrosine kinase inhibitor, genistein (200 mumol/L), did not inhibit the Epo-induced expression of VL30. TPA (100 ng/mL), a PKC agonist, did not induce VL30 expression. Although the physiologic role of VL30 in the differentiation of erythroid progenitor cells is not known, our findings demonstrate that VL30 is an early ERG, and may be a useful indicator of the initial molecular actions of Epo.

Involvement of tyrosine kinase and protein kinase C in platelet-activating-factor-induced c-fos gene expression in A-431 cells.

In A-431 cells, platelet-activating factor (PAF) induces the expression of c-fos and TIS-1 genes in both the absence and the presence of cycloheximide in a structurally specific and receptor-coupled manner. We have now investigated the molecular mechanisms of this response, particularly in relation to the role of protein kinases. Pretreatment of cells with genistein or methyl-2,5-dihydroxycinnamate (tyrosine kinase inhibitors) or staurosporine (a protein kinase C inhibitor) for 20 min abolished the c-fos expression induced by PAF. Interestingly, when genistein was added 90 s after addition of PAF, no inhibition was observed. Similarly, staurosporine did not inhibit c-fos expression when added 8 min after PAF addition to the cells. These inhibitions were dose-dependent (IC50 for staurosporine was 180 nM, and for genistein 50 microM). Simultaneous addition of PAF and phorbol 12-myristate 13-acetate (PMA) did not give a synergistic effect on c-fos expression. Pretreatment of cells with PMA had no effect on [3H]PAF binding, but abolished the PAF-induced gene expression. PAF-stimulated gene expression was desensitized if cells were pretreated with PAF. Interestingly, epidermal growth factor was able to stimulate c-fos expression in PAF-desensitized cells, and thus indicated involvement of distinct mechanisms for the two stimuli. Forskolin, an activator of adenylate cyclase, did not induce c-fos expression and had no effect on the PAF response. Exposure of cells to PAF for as little as 1 min, followed by its removal, was sufficient to activate the gene expression and demonstrated the rapidity and the exquisite nature of the signalling involved in this process. It is concluded that activation of PAF receptor (a proposed G-protein-coupled receptor) causes rapid production of signals which induce the expression of c-fos gene and that this is mediated via tyrosine kinase and protein kinase C.

Structure and expression of TIS21, a primary response gene induced by growth factors and tumor promoters.

The TIS21 gene is a primary response gene that is induced rapidly and transiently in 3T3 cells by the tumor promoter and mitogen tetradecanoyl phorbol acetate. The predicted open reading frame of the TIS21 cDNA encodes a protein of 158 amino acids with no obvious similarity to any known protein. Antiserum prepared to TIS21 recombinant protein produced in Escherichia coli precipitates a 17-kDa protein from Swiss 3T3 cells. The 2040-nucleotide 3'-untranslated region of the cDNA includes an unusual T18 sequence. The TIS21 gene has a single 1.4-kilobase intron which interrupts the open reading frame and is otherwise identical to the cDNA sequence. The 5'-flanking sequence of the TIS21 gene contains TATA and CAAT box-type sequences, three potential Sp1 sites, two putative cyclic AMP response elements, two potential AP1 binding elements, and an AP2 element. A possible Z-DNA structure of 29 AC repeats is present 660 nucleotides from the start of transcription. Expression from a luciferase reporter construct containing a 460-nucleotide fragment of the TIS21 promoter is induced by tetradecanoyl phorbol acetate, forskolin, epidermal growth factor, and serum, despite the absence of a consensus serum response element.