Characterization of the Human Inosine-5' -monophosphate Dehydrogenase Type II Gene*

activity and mRNA levels are induced up to IS-fold upon mitogenic or antigenic stimulation of human peripheral blood T lymphocytes. This increase in IMPDH activity is required for cellular proliferation and has been associ ated with malignant transformation. We have cloned the human IMPDH type II gene and show that it contains 14 exons and is approximately 5.8 kilobases in length. Ex ons vary in size from 49 to 207 base pairs and introns from 73 to 1065 base pairs. The transcription start site was mapped to a position 50 nucleotides upstream ofthe translation initiation site. The 5' -flanking region con sisting of 463 base pairs upstream of the translation initiation site confers induced transcription and differ ential regulation upon a chloramphenicol acetyltrans ferase reporter gene when transfected into Jurkat T cells and human peripheral blood T lymphocytes, re spectively. DNase I footprinting analysis using Jurkat T cell nuclear extract identified four protected regions in the promoter which coincide with consensus transcrip tion factor binding sites for the nuclear factors AP2, ATF, CREB, Egr-L, Nm23, and Spl. These findings sug gest that several of these nuclear factors may play a critical role in the regulation of IMPDH type II gene expression during T lymphocyte activation. at the point extended reverse transcriptase at 42°C for 30 ethanol-precipitated using 20 I'-gof carrier tRNA. Following precipita the was resuspended in 6 1'-1 of gel loading dye, and 3 1'-1 were analyzed on a denaturing 8 Murea, 6% polyacrylamide gel. The gel was and autoradiographed overnight at hybridization

Inosine-5'-monophosphate dehydrogenase (IMPDH) activity and mRNA levels are induced up to IS-fold upon mitogenic or antigenic stimulation of human peripheral blood T lymphocytes. This increase in IMPDH activity is required for cellular proliferation and has been associated with malignant transformation. We have cloned the human IMPDH type II gene and show that it contains 14 exons and is approximately 5.8 kilobases in length. Exons vary in size from 49 to 207 base pairs and introns from 73 to 1065 base pairs. The transcription start site was mapped to a position 50 nucleotides upstream of the translation initiation site. The 5' -flanking region consisting of 463 base pairs upstream of the translation initiation site confers induced transcription and differential regulation upon a chloramphenicol acetyltransferase reporter gene when transfected into Jurkat T cells and human peripheral blood T lymphocytes, respectively. DNase I footprinting analysis using Jurkat T cell nuclear extract identified four protected regions in the promoter which coincide with consensus transcription factor binding sites for the nuclear factors AP2, ATF, CREB, Egr-L, Nm23, and Spl. These findings suggest that several of these nuclear factors may play a critical role in the regulation of IMPDH type II gene expression during T lymphocyte activation.
Inosine-5'-monophosphate dehydrogenase (IMPDH,l EC 1.1. 1.205) is positioned at the branch point of adenine and guanine nucleotide biosynthesis from IMP and constitutes the rate-limiting enzyme in the de novo synthesis of guanine nucleotides. The enzyme catalyses the NAD-dependent oxidation of IMP to XMP and is responsible for maintaining cellular guanine deoxy-and ribonucleotide pools required for DNA and RNA biosynthesis, respectively. Enzyme activity varies with the cell cycle, exhibiting maximal activity in S phase (1).
Total cellular IMPDH activity is accounted for by the expression of two distinct genes, IMPDH type I located on chromosome 7 and IMPDH type II located on chromosome 3 (2,3). The human IMPDH type I and type II cDNAs have been isolated * This work was supported by National Institutes of Health Grants CA64192 and CA34085. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
The close correlation between elevated IMPDH activity and cell proliferation and the observation of high activity in neoplastic cells have linked IMPDH activity to malignant transformation (8,9). This association has led to the search for and development of several inhibitors with demonstrated antineoplastic and immunosuppressive potential (10,11). Such inhibitors of IMPDH activity have been demonstrated to inhibit cell proliferation and induce cellular differentiation as a consequence of the reduction of guanine nucleotide levels (12)(13)(14)(15)(16).
Regulation of IMPDH activity during cellular growth and differentiation has been largely attributed to changes in the expression of the IMPDH type II gene. The type II 2.3-kb mRNA transcript is the predominant species in neoplastic cells and is selectively up-regulated in replicating cells (17)(18)(19). Conversely, when neoplastic cells are induced to differentiate, the enhanced levels of the type II transcript and total cellular activity are down-regulated (13)(14)(15). In contrast, the 3.5-kb type I transcript remains constitutively expressed during cell proliferation and induction of cell differentiation (4,(17)(18)(19). The modulation of cellular IMPDH activity during cell growth and differentiation suggests a critically important role for the regulation ofIMPDH type II gene expression in the progression of normal cell development. In order to assess the molecular mechanisms governing the expression of IMPDH type II in quiescent, replicating, and differentiating cells, we have cloned the IMPDH type II gene and characterized the gene and its 5' -flanking region.

MATERlALS AND METHODS
Cloning ofIMPDH Type II Genomic DNA-A human genomic library established in A-FIX by partial BglI digestion of human leukocyte DNA followed by ligation into the phage DNA vector was provided by Dr. J. Lowe (University of Michigan, Ann Arbor). Approximately 7 x 10 5 plaques were screened using full-length IMPDH type II cDNA,provided by Dr. F. Collart (Argonne National Laboratory, Argonne, IL). The cDNA was labeled with [a_ 3 2P]dCTP (3000 Ci/mmol; Amersham Corp.) using DNA polymerase large fragment (New England Biolabs Inc.), Unincorporated nucleotides were removed by ethanol precipitation. Hybridization was performed in 2 x SSC, 1%SDS, 10%dextran sulfate, and 50% formamide at 42°C for 24 h after which the filters were washed with 1 x SSC, 0.1% SDS at 65°C. Two positive phage clones of approximately 15 kb were identified, plaque-purified, and characterized by restriction endonuclease mapping. Sad fragments from a single insert were subcloned into the vector pGEM7Zf\+) and analyzed by Southern blotting. Positive clones containing 4.8-and 6.8-kb inserts were further analyzed by restriction mapping and sequenced according to the Sanger dideoxy chain termination method (20) using eDNA, intron, and vector primers.
Plasmid Constructs-pBSCAT was derived from the pBS vector by cloning the chloramphenicol acetyltransferase (CAT) eDNA into the BamHI site. Employing a 5' probe spanning nucleotides 90-126 of the IMPDH type II cDNA (5), a 1536-bp EcoRI fragment containing the 5' 6808 This is an Open Access article under the CC BY license. region of the gene was identified within the 4.8-kb Sac! fragment. This fragment was subcloned into pGEM7Ztl:+) and sequenced on both strands. The fragment spanning bp -463 to + 1073 relative to the A (+ 1) of the translation start site was reisolated using EcoRl, cloned into the Sall site of the vector pBSCAT, and designated pBS1536(5'~3')CAT. A 466-bp genomic fragment containing the 5'flanking region extending from EcoRl to a Ncol site at the ATG initiation codon of the first exon (bp -463 to +3; Fig. 1) was excised from the 1536-bp EcoRl pGEM7Ztl: +) construct using the vector Pstl and genomic Ncol sites and subcloned into pBSCAT in the 3'~5' orientation relative to the CAT reporter gene (designated pBS466(3'~5')CAT). The identical fragment was subcloned into the Smal site of pGEM7Ztl:+ ), removed by a XbaIlHindlU restriction digest, and subcloned into the identical sites of pBSCAT to obtain the 5'~3' orientation (designated pBS466(5'~3')CAT).
Primer Extension Analysis-Primer extension analysis was performed with the Primer Extension System (Prornega, Madison, WI) employing 5 I'-g of poly(A +) Jurkat T cell RNA and the reverse complementary oligonucleotide 5'-GTT GAA GAG CTG CTG TGC TGT GAG TC-3' that anneals to the region extending from +49 to +75 relative to the translation initiation site (5). Briefly, 100 ng of oligonucleotide primer was end-labeled with [y_ 32 p IATP using T4 polynucleotide kinase. Approximately 1 ng of labeled primer was added to 5 I'-g of RNA in the presence of 2 x primer extension buffer. The primer was annealed to the RNA at 58°C for 20 min followed by cooling to room temperature for 30 min. The annealed primer was extended with avian myeloblastoma virus reverse transcriptase at 42°C for 30 min and ethanol-precipitated using 20 I'-g of carrier tRNA. Following precipitation, the sample was resuspended in 6 1' -1 of gel loading dye, and 3 1' -1 were analyzed on a denaturing 8 Murea, 6% polyacrylamide gel. The gel was fixed, dried, and autoradiographed overnight at -70°C.
RNase A Protection Assay-RNase A protection analysis was performed using the pGEM7Ztl: + )-466bp construct that extends 463 bp upstream of and includes the translation initiation site. The vector was linearized at the 5' end of the insert with the restriction enzyme HindUl, and transcribed in vitro from the T7 RNA polymerase promoter to generate a [ 32pICTP-labeled antisense RNA transcript. The assay was performed using the Ambion ribonuclease protection assay kit (Ambion, Austin, TX). Five I'-gofpoly(A+) Jurkat T, HL60, and Raji cell RNA were precipitated with 5 x 10 5 cpm of probe and resuspended in 20 1' -1 of hybridization buffer. Samples were denatured at 95°C for 3 min and then incubated at 42°C for 18 h. Following hybridization, 200 1' -1 of RNase digestion buffer containing RNase A and RNase T1 were added, and the samples were incubated at 37°C for 30 min. RNA was subsequently precipitated, resuspended in gel loading buffer, and separated on a 8 M urea, 6% polyacrylamide gel. The gel was fixed, dried, and autoradiographed overnight at -70°C.
Isolation of Peripheral Blood T Lymphocytes-Buffy coats from normal donors were obtained from the American Red Cross, and the mononuclear cells were isolated by density gradient centrifugation using Histopaque 1077 (Sigma) (21). Cells at the interface were removed, washed with PBS, and resuspended in RPMl 1640 medium containing 10% heat-inactivated fetal calf serum. Monocytes were depleted by culture dish adherence and B lymphocytes by negative selection using an anti-CD20 antibody (Coulter Corp., Hialeah, FL). Flow cytometric analysis of the isolated T lymphocytes with an anti-Cfrz" marker revealed a greater than 95% enrichment of CD2+ T cells. Cellular [3Hlthymidine incorporation into DNA from resting and activated cells was determined as a measure of proliferative activity.
Transient Transfections-ln order to study the promoter activity of the lMPDH type U 5'-flanking sequence, each CAT-reporter construct (30 I'-g) was transfected into 1 x 10 7 exponentially growing Jurkat T cells or 2 X 10 7 isolated peripheral blood T lymphocytes prestimulated according to the protocol of Park et al. (22). A f3-actin-f3-galactosidase (pf3Ac-lacZ)construct was used in Jurkat T cells to determine transfection efficiency (23). Jurkat T lymphoblasts and peripheral blood T lymphocytes were maintained in RPMl 1640 medium supplemented with 2 mM L-glutamine, 100 units/ml penicillin, 100 I'-g/ml streptomycin, and 10% fetal calf serum for the Jurkat T cells and 10% heatinactivated fetal calf serum for the T lymphocytes, respectively (Hyclone Labs, Logan, UT). Cells were cultured at 37°C in a humidified atmosphere in the presence of 5% CO 2, Electroporations were performed at room temperature in the presence of culture medium using a Bio-Rad Gene Pulser with settings of 250 Vl960 I'-F for Jurkat T cells and 350 V/960 I'-F for T lymphocytes. Transfected Jurkat T cells were cultured for 48 h, harvested, washed three times with phosphate-buffered saline (PBS), resuspended in 150 1' -1 of 0.25 M Tris-HCl (pH 8.0), and extracted with three cycles of freeze-thawing. Transfected periph-eral blood T lymphocytes were incubated for 18 h prior to phorbol 12-myristate 13-acetate (PMA) (10 ng/ml) and ionomycin (125 ng/ml) treatment. The cells were maintained in culture for 48 h and then processed as described above for Jurkat T cells. Cleared supernatants obtained after spinning extracts at 16,000 x g for 10 min were used for protein and f3-galactosidase activity assays. Aliquots were heated to 60°C for 10 min followed by centrifugation at 16,000 x g for 10 min. The supernatants were assayed with 0.1 I'-Ci of [ 14Clchloramphenicol and 25 I'-gof n-butyryl-CoA for 1-6 h, extracted with xylenes according to the Promega CAT enzyme assay system, and analyzed by liquid scintillation counting. Protein concentrations were determined with the Bio-Rad protein assay according to the method of Bradford (24).
DNase I Footprinting-A probe covering the 466-bp promoter region was prepared by Ncol digestion of the pBS466(5'~3')CATconstruct, fill-in labeling with DNA polymerase large fragment, and digestion with HindU!. The resulting fragments were separated on a 6% nondenaturing polyacrylamide gel, and the probe was excised and eluted by the "crush and soak" method (27). DNase I footprinting was performed according to Blake et al. (26). Ten ng op 2P-labeled DNA were incubated with 120 and 240 I'-gof Jurkat T cell nuclear extract in the presence of 15 I'-gofpoly(dl-dC), 6.1% glycerol, 0.07 mM EDTA, 0.07 mM EGTA, 7.2 mM HEPES, pH 7.9, 39 mM KCl, 7.5 mM MgC12' and 0.7 mM DTT. The binding reactions were performed at room temperature for 30 min after which CaCl 2 (2 mM final concentration) was added, and the probe was digested with DNase I (Worthington) at room temperature for 3 min. Digestions were terminated by the addition of 2 volumes of 100 mv Tris, pH 8.0,20 mMEDTA, 0.1% SDS, 100 ug/ml proteinase K, and 100 I'-g/ml glycogen. After an incubation at 37°C for 20 min, the samples were extracted with phenol/chloroform, precipitated with ethanol, resuspended in formamide loading dye, and analyzed on a 8 M urea, 6% polyacrylamide sequencing gel.

RESULTS
Isolation of Type II Genomic DNA Clones-Genomic clones were isolated by screening a human genomic A-FIX library with a 2.0-kb full-length IMPDH type II eDNA. Digestion of a single genomic clone with Sac! resulted in fragments of 6.8, 4.8, 2.9, 1.5, and 0.8 kb that were subcloned into the vector pGEM7Ztr + ) and analyzed by Southern blotting. The 4.8-and 6.8-kb fragments hybridized to the IMPDH type II eDNA. Sequence analysis revealed that the 4.8-and 6.8-kb fragments contain the entire coding region of the type II gene as previously published by Collart and Huberman (5).
Structure ofthe IMPDH Type II Gene-Deoxyribonucleic acid primers were used to sequence and map the exon-intron junctions of the type II gene encompassed in both the 4.8-and 6.8-kb Sac! fragments. Exons 1 through 5 were located in the 4.8-kb Sac! fragment and exons 6 through 14 in the 6.8-kb Sac! fragment. The following inconsistencies were found with respect to the published cDNA sequence: a single base mismatch (bp 2; G~C), the absence of 9 nucleotides (GTCTCTGCG) at the 5' terminus of the eDNA, and 3 erroneous cytosine residues (CCC~AAA) at the 3' terminus of the eDNA. In addition, two consecutive cytosine residues at positions 608-609 are replaced by a single thymidine base at position 608 and a guanine base between bp 613 and 614 of the eDNA. These substitutions result in the conversion of arginine 110 and serine 111 to alanine and glycine residues, respectively (5). To confirm that the 4.8-and 6.8-kb Sac! fragments are contiguous, human genomic DNA was subjected to polymerase chain reaction anal-  Table I). All exon-intron boundaries contained the canonical splice acceptor (GT) and donor (AG) consensus sequences ( Table I).
Mapping of the Transcription Start Sites-The transcription start site of the IMPDH type II gene was identified by primer extension analysis and confirmed by RNase A protection analysis. A single major primer extended product was identified at a position 50 bp 5' to the A (+ 1) of the translation initiation codon ( Fig. 2A). RNase A protection analysis was performed using the pBS466CAT genomic clone that contains the 463-bp 5'-flanking region and the 3-bp translation start site of the gene. A complementary RNA transcript which extends 463 nucleotides 5' of the translation start site was synthesized and hybridized to Jurkat T cell poly(A +) mRNA. RNase A digestion resulted in a single protected fragment of approximately 59 bp in the T cell lines Jurkat and HL60 and the B cell line Raji (Fig.  2B). This is in close agreement with the location of the transcription start site identified by primer extension analysis.
Structural and Functional Analysis of the IMPDH Type II Promoter-The 1536-bp EcoRI fragment located within the 4.8-kb Sad fragment was sequenced on both strands and found to contain 463 bp of sequence 5' to the translation initiation site, and downstream sequence extending to bp + 1073 in exon 4 (Fig. 3). The downstream EcoRI site corresponds with the EcoRI site located at bp 334 in the eDNA (5). A computerassisted transcription factor data base search (Genetics Computer Group, Madison, WI) ofthe 1536-bp genomic sequence for putative nuclear protein binding sites revealed a cluster of consensus motifs for DNA binding proteins in the 5' -flanking region of the gene. As shown in Fig. 3, a TATA box is located at bp -74 relative to the translation initiation site (28). Two putative activator protein-2 (AP2) binding sites are located at positions -133 and -163 (29), and three putative binding sites for Sp1 are located at positions -137, -149, and -167 (30)(31)(32). In addition, the sequence from bp -89 to -94 and -114 to -121 contain consensus cAMP response element binding protein (eRE) binding sites (33,34). The more upstream cAMP response element overlaps with a putative activating transcription factor (ATF) binding site (35). Potentially important consensus binding sites for the early response gene Nm23 at bp -246 (36) and the early growth response 1 gene (Egr-l) at bp -163 (37-39) were identified. The latter site overlaps with both an AP2 and a Sp1 binding site. The importance, abundance, and overlapping nature of the putative transcription factor binding sites suggest a complex regulation of IMPDH type II gene transcription.
To evaluate the functional significance of the putative promoter region in the regulation of IMPDH type II expression, CAT reporter plasmids were constructed and transiently transfected into exponentially growing Jurkat T cells. The constructs were derived from pBSCAT and contained the 1536-bp genomic EcoRI fragment (position -463 to + 1073) and the 466-bp 5' EcoRIINcoI fragment (position -463 to +3) in 5'~3' and 3'~5' orientations. Fig. 4 demonstrates that all constructs containing the IMPDH type II 5' region have promoter activity. The 1536-bp construct exhibited 200-fold higher activity than did the pBSCAT vector alone. The pBS466(5'~3')CAT construct manifested 70-fold and the pBS466(3'~5')CAT construct 10-fold increased activity over the vector alone, demonstrating that the 463-bp DNA fragment immediately upstream of the human IMPDH type II gene's initiation codon functions as a promoter upon transfection into Jurkat T cells, with substantially less activity in the reverse orientation.
Transfection of Peripheral Blood T Lymphocytes-IMPDH type II mRNA levels and activity are strongly induced upon activation of peripheral blood T lymphocytes (40). To determine whether the 466-bp promoter fragment contains the elements necessary for proliferation-dependent transcriptional regulation of the IMPDH type II gene, pBSCAT, pBS466(5'~3')CAT and pBS500dCK-CAT (41) constructs were transfected into peripheral blood T lymphocytes prestimulated according to the protocol of Park et al. (22). Following transfection, the T lymphocytes were maintained in culture medium or stimulated for 48 h with PMA and ionomycin. As shown in Fig. 5, PMA and ionomycin stimulation induced pBS466(5'~3')CATexpression 6.5-fold over that of the nonstimulated cells. Expression from a control construct containing the 500-bp core promoter of the human deoxycytidine kinase gene (pBS500dCK-CAT) (41) was not increased upon stimulation of T lymphocytes with PMA and ionomycin, a finding consistent with the lack of proliferation-related up-regulation of dCK expression (42,43). These data demonstrate that the 466-bp promoter fragment contains the elements required for at least a portion of the proliferationdependent induction of IMPDH type II expression.
DNase I Footprint Analysis of the 5' Region-The 5' -flanking region of the IMPDH type II gene contains several consensus transcription factor binding sites. In order to implicate nuclear factors in the regulation oflMPDH type II gene expression in T lymphocytes we examined in vitro nuclear protein binding to the 463-bp 5'-flanking region of the gene. DNase I footprint analysis of the coding strand using a probe spanning the entire promoter from -463 to +3 revealed four protected regions, designatedA-D, in the presence of Jurkat T cell nuclear extract (Fig. 6). The extended footprint covering region A (-79 to -101) corresponds with a consensus binding site (TGACGAA) for the CREB family of leucine zipper transcription factors (33,34) and is located immediately upstream ofthe proximal TATA box (28) (Fig. 3). Protected region B extending from bp -111 to -124 coincides with a recognition sequence (CTGACGTCAG) for the CREB/ATF nuclear protein family (28,(33)(34)(35). Footprinted region C located at bp -152 through -176 (AGCTC-CGCCCCCGC) contains overlapping consensus binding sites for the nuclear factors AP2 (29), Egr-1 (37,38), and Sp1 (30)(31)(32). The most distal footprint D located at position -252 to -281 is adjacent to a nuclear factor recognition sequence (GGGTGGG) for the early response gene Nm23 (36). DISCUSSION The association of increased IMPDH enzymatic activity with cellular proliferation and transformation has been known for 20 years, originating in observations on rat hepatoma cells (8)

T ABl. E I Exon -intr on organization of th e human [AfPDH typ e II gene
Exon a nd intron nucl eotid e seq uence an d size wer e det e rm in ed from single-st ra nded se q ue ncing. Th e size of intron 5 se pa ra ting th e 4.8-and 6.8-kb clon es wa s det ermin ed by seque nce a na lys is a nd con fir med by polym er ase chain reaction analys is of ge no mic DNA using primer s located in exon 5 a nd 6. Uppe r-case an d lower -case characte rs indi cate exon a nd intron se q ue nce. respectively. Th e ATG ini tia tion codo n a de nine is desi gn ated as + 1.

72
118 Exon Fi e. 2. Analysi s of th e trans cription initia tion s ite of th e IMPDH t ype II g ene. A . pri mer exte nsion product produced from -Iurkat T cell polytA " ) RNA . DNA se q ue nce was obta ine d from th e 1536 -bp EcoR I pr om ot e r fragm en t us ing th e identical oligon ucleo t ide used for pri mer exte ns ion ; lan es contain primer a lone; prim er an d yeast tRNA; primer . t RNA. a nd Ju rk a t T cell polyfA. " ) RNA; a nd <j >X 174 D, ' f lIae'" marke rs.

Sequ en ce of exon-in t ro n jun ctions
Th e transcription initiati on si te a nd seque nce surro u nd ing th e site is indica te d on t he left. B. ribonuclease pr ot ecti on assay of -Iu r ka t, H L60. an d Raji poly rA " ! RNA hybrid ized to th e 466-bp : 12 P-la beled RNA template exte nding 5 ' of th e t ranslatio n init ia t ion s ite , Lanes conta in pr obe plu s yeas t tRNA; pr obe plu s poly rA " ) RNA from -Iur kat, HL60. a nd Raji cells. resp ecti vel y; an d <t>X 174 DN Hini! markers . a nd resu lti ng in a n inten siv e se a rch for IMPDH in hibitors as pote nt ial a ntineop lastic agents . Su ch inhibitors have been dem onst ra t ed to resul t in inhibition of cell growth a nd th e in duction of cellular differentiation , in conjunction with inhibition of DNA sy nt hes is directly a t t r ibu table to t he depl et ion of guanine nu cleot ides 02-16). In a ddit ion, se ve ra l in hibitors ha ve been found to be useful a s im muno su ppre ssiv e a gent and to inhibit th e ac t iva tio n of T ly m phocytes ill vitro (44. 45 ). These obse rva t ions underscor e t he pote nt ia l importance of th i en zyma t ic a ctivi ty in modulatin g normal cell growth . Th e recent ide nt ifica t ion of two se pa ra te genes enco din g Ii\! PDH activity (4, 5 ) a nd th e dist in ct association of increases in type II IMPDH mRNA with neoplastic transformation in several cell types (18,19) have made it feasible to search for the molecular basis for the regulation of expression of the type II gene at different stages of cell development. In order to determine the structural basis for IMPDH gene regulation, we have cloned and characterized the type II IMPDH gene and its 5' -flanking sequence.
Type II IMPDH is a relatively small gene of approximately 5.8 kb consisting of 14 exons varying in size from 49 to 207 bp. The transcription initiation site, as determined by both primer extension analysis and RNase protection, occurs 50 bp upstream of the ATG and 9 bp 5' to the 5' terminus of the published cDNA (5). The 5' -untranslated region of the cDNA is highly (70%) GC-rich. The 463-bp 5'-flanking region confers strong promoter activity on a CAT reporter gene when transfected into Jurkat T cells and peripheral blood T lymphocytes.
When T lymphocytes are stimulated with the pharmacological agents PMA and ionomycin, promoter activity was increased by about 6-fold; in contrast, promoter activity was unaffected when Jurkat T cells were stimulated under the same conditions (data not shown).
Recent studies by our group have shown that activation of peripheral blood T lymphocytes with the mitogens PMA and ionomycin results in a 10-and 15-fold induction oflMPDH type II mRNA expression and total cellular enzymatic activity, respectively (40). It has been suggested that the growth-regulated increase in IMPDH expression is due to a posttranscriptional nuclear processing event (46). However, our data suggest that a major transcriptional component is responsible for the up-regulation of IMPDH type II gene expression in activated peripheral blood T lymphocytes. Although these data are not conclusive, they strongly suggest that the 463-bp upstream o B A eTA G model sys te m to examine IMP DH expression as a function of T cell activatio n (40). We observed the induction of IMPDH typ e II mRNA wit hi n 6 h after stimula tion of T lymphocytes with PMA a nd ionornycin , as well as with phytohemagglutinin a nd a lloge neic mononu clear cell s, although maximum induction was observed at 24 h. In addition , IMP DH type II mRNA levels increased in resp onse to PMA alone and to calcium ionoph ore a lone, a lt hough t he combination of PMA and ionomycin was significa nt ly more potent than eit he r agent al one. irnilar reo suits were obtained for Egr-l express ion; th e combina tion of ph orb ol ester a nd ca lcium ionophore lead to higher express ion th an eit he r age nt a lone (49). Although th e evide nce implica ting Egr -l in IMPDH ty pe II expression is circums ta ntia l a t present, t he pr esen ce of a protected region (C) corresponding to th e Egr -I bind ing site in the IMPDH type II prom oter a nd th e F IG. 6. DNa s e I fo otprin t a na lysis o f t h e 5 '-n a n k i n g region of t h e IM P DH type II ge ne. A 466-bp DNA fra gm ent contai ning the IMPDH ty pe II 5'-flanking region wa s lab eled a t th e 3 ' end on the codi ng st ra nd and subjected to DNase I treatment in th e a bse nce of nuclear extract (0 ), or in th e pr esen ce of 120 ",g or 240 ,..g of .I urkat T cell nuclear extract. Th e position of four regions prot ected in th e presence of nuclear ext rac t (A -D ) a re ind ica te d on th e r ight . Th e loca tion of th e transcription sta rt site is indica te d by th e arrow . Ch OJ a.
region contains at least a portion of th e regulatory elemen ts nece ssary for the proliferation-related expression of th e gene. While we have not ruled out a direct effect of PMA and ionomycin on gen e expression th at is ind ep endent of proliferation , t he la ck of an effect of th ese agents on promoter-mediated CAT expression in Jurkat T cells mak es this expla na tion less lik ely. It remains possibl e th at ot he r region s of th e ge ne outsid e of th e promoter region contribute to IMPDH typ e II up-regulation. Ind eed, th e 2-3-fold high er level of CAT activity found with th e 1536-bp construct containing a portion of the proximal codin g regio n of the ge ne in a ddit ion to th e 463-bp 5 '-fla nking region sugges t th e exis te nce of enha ncer a ctivity in this region . Several pot entially important regulatory sites in th e 463-bp 5 '-flan king region of th e gene a re protected from DNase I digestion in th e pr esenc e of Jurkat T cell nuclear ext ra ct and suggest functional releva nce for IMPDH typ e II expression. Th e pre sence of a con sen su s bind ing site (CGCCCCCGC) for th e transcription factor Egr-l (37, 38) (sy nonymous with Kr ox-24, NGFI-A, Zif26 8, and T IS-8 ) (47) a t bp -163 is particul arly notable. Thi s site has been s hown to bind a family of zinc finger prot ein s that are immedi ate ea rly response ge nes im portant in gr owt h regu la tion. Egr-l expression is rap idly and transiently induced by nerve growth factor in PC12 cell s (48 ) and by se ru m in fibrob lasts (37,39 ). Of particul ar relevance is th e observation that Egr-l is induc ed during th e GoIG l transition in th e cell cycle after mitogenic stimula tion of T lymphocyt es, as well as during G 1 as a se pa ra te eve nt medi ated by interleukin-2 (49). Expo sure ofT lymphocytes to a n Egr -I antisense oligonucleoti de block ed lymphocyt e a ctivation, st rongly sugg es ti ng t hat Egr-l is ess entia l for th e expression of down stream ge nes required for th e T lymphocyte pr oliferative resp onse. In pr evious st udies, we have employed a peripheral blood T lymphocyt e F IG. 5. Tra n sc ri ptiona l r egu lntion of th e IMPDH type II p r om oter c o n s t r u c t in r e sti ng and act ivated p e riph eral bl o od T lympho cy t e s. Peripheral blood T lymphocytes we re pr estimul a ted as described und er "Ma teria ls a nd Methods" a nd t ransiently transfected wit h th e cons t r ucts ind icated. Ce lls were mainta in ed in med iu m in the ab sen ce of st imula tion or t rea te d wit h 10 ng/m l PMA a nd 125 ng/ ml ionomycin a nd cont inue d in cu lt ure for 48 h . Ce lls ext rac ts we re assayed for CAT act ivity . S ti mu lated T lymphocytes ex hibite d a 2.4-fold high er I"Hl t hy midine inco r pora tio n th a n nonstimulated cells . Values a re th e mean of a sing le ex peri me nt perform ed in dupli cate a nd represe nt data obt ain ed from three ind ep enden t ex peri me nts . • . cont rol; fJiJ. PMA + ion omycin . requirement for increased IMPDH gene expression for T lymphocyte activation do suggest that this site could be of considerable functional importance. It should also be noted, however, that the putative Egr-l site overlaps with sites for the transcription factor AP2 (CCGCCCCCGC) (29) and Spl (GCTC-CGCCCC) (30,32). These overlapping sites offer the potential for more complex regulation based on transcription factor interactions. It has been observed, for example, that Egr-l can act as a repressor of Spl activity at the coincident binding site in the adenosine deaminase gene promoter (50).
A second region of interest in the IMPDH type II promoter is the DNase I protected region D that occurs in close proximity to the recognition sequence for the nuclear purine-binding transcription factor Nm23 (GGGTGGG) (36). Nm23, also known as PuF, was previously shown to bind to a nuclease hypersensitive element of the human c-myc PI promoter and directly regulate c-myc transcription (51). Nm23 is a nucleoside diphosphate kinase enzyme, the phosphorylation status of which appears to vary as a function of the metastatic potential of some cell types (52). The transcriptional regulatory function of this protein has been shown to be independent of its enzymatic activity (53). Recent studies of Nm23 expression in peripheral blood lymphocytes have demonstrated a strong correlation between Nm23 expression and proliferative activity, with levels of Nm23 protein being significantly higher in activated peripheral blood lymphocytes and in malignant proliferating lymphoid cells than in resting leukocytes (54). The increase in the level of Nm23 occurred as a relatively late event, suggesting a potential role for this protein in the late G 1 and early S phases of the cell cycle. Similarly, increased IMPDH expression has been demonstrated to be a requirement for continued cellular proliferation (12,16). The finding that the murine homologue of Nm23-H2 serves as a differentiation inhibiting factor in mouse myeloid leukemia cells further supports a role for Nm23 in cell proliferation (55). Whether or not Egr-l and Nm23 are integral to IMPDH type II expression will be determined by specific mutagenesis experiments. Delineation of their respective roles should provide further insight into the cascade of molecular events required for the ultimate synthesis of guanine nucleotides necessary for the initiation of DNA replication.