Biochemistry 5 ' control regions of the apolipoprotein ( a ) gene and members of the related plasminogen gene family [ apolipoprotein / atherosderosis / lipoprotein ( a ) I

Elevated blood levels of apolipoprotein(a), the component of lipoprotein(a) that distinguishes it from low density lipoprotein, are a major risk factor for atherosclerosis. The apolipoprotein(a) gene is highly similar to the pasminogen gene and to at least four other genes or pseudogenes. The 5' untranslated and flanking sequences of these six genes contain extensive regions of near identity and share sequence elements involved in the initiation of transcription and translation. About 1000 base pairs of flanking DNA of each gene are sufficient to promote transcription in cultured hepatocytes. The apolipoprotein(a) gene promoter contains functional interleukin 6-responsive elements, consistent with the reported acute-phase response of apolipoprotein(a). Flanking genomic fragments ofthe apoliprotein(a) gene from two individuals with vastly different plasma apolipoprotein(a) concentrations have sequence differences that are reflected in differences in the rate of in vitro transcription. Over the past decade, high plasma levels of lipoprotein(a) [Lp(a)] have been established as one ofthe major risk factors for atherosclerosis and its major manifestations: myocardial infarction, restenosis following angioplasty and bypass surgery, and stroke (for reviews, see refs. 1-3). Two independent studies conclude that 27% of premature myocardial infarctions in males may be due to elevated Lp(a) (4, 5). Unlike other lipoproteins, individual variation in Lp(a) concentration is almost entirely due to inheritance (6). Lp(a) resembles low density lipoprotein (LDL) in its content of cholesterol, phospholipids, and apolipoprotein B-100. It is distinguished from LDL by the large glycoprotein apolipoprotein(a) [apo(a)] that is attached to the particle by disulfide linkage to apolipoprotein B-100. Cloning and sequencing of the cDNA of human apo(a) revealed an unexpected homology to plasminogen (7). The 5' regions of the two cDNAs have nearly 100% identity, whereas other domains of the plasminogen and apo(a) have between 75% and 94% local sequence identity. An unusual feature of the apo(a) cDNA is the large number of times that one domain is repeated. This domain is homologous to kringle four of plasminogen and is variously repeated from about 15 to 40 times in individual apo(a) genes, resulting in a large number of molecular weight isoforms of the protein (7-9). It is likely that the apo(a) gene arose from a duplication of the plasminogen gene, perhaps recently during mammalian evolution, with regions at the 5' end undergoing subsequent gene conversion. Plasma Lp(a) concentration varies widely in the human population, ranging from <0.1 to >200 mg/dl. Since blood levels of Lp(a) are a major contributor to morbidity, it is important to understand its regulation. Metabolic and genetic evidence shows that Lp(a) synthesis, rather than catabolism, is the key determinant for plasma concentration (10, 11) and that it is the apo(a) component of the lipoprotein particle that is the most important determinant of its final concentration (1, 6, 8). Studies in human populations show a general inverse relationship between the size ofthe apo(a) protein and plasma concentration (cf. refs. 1, 12, and 13). Protein translation or particle assembly might in some way account for this relationship. However, even within a given apo(a) isoform size class, a great deal of variation in plasma concentration exists, emphasizing the importance ofdistinct regulatory elements in the gene. This is supported by studies in humans and cynomolgous monkeys which show that hepatic apo(a) mRNA levels are related to Lp(a) concentration (14-16). To begin to understand the control of apo(a) gene expression, we present here the sequence of the flanking regions of the apo(a) gene and closely related homologues and demonstrate the activity of these regions in transcription assays.t MATERIALS AND METHODS Cloning of 5' Flanking Region of the Apo(a) Gene and Related Members of the Plasminogen Gene Family. Electrophoresis, blot hybridization, and other standard molecular biology procedures essentially followed the protocols of Sambrook et al. (17). Genomic libraries in bacteriophage A (A-ZAP II and Gigapack Gold II; Stratagene) were constructed from human genomic DNA digested to completion with EcoRI and screened with the probe pcr A (see below). In some cases, the EcoRI-digested DNA was size-selected by fractionation by 1% agarose gel electrophoresis. Positive clones were isolated as pBluescript plasmids by the in vivo excision procedures detailed by the manufacturer and sequenced to completion by the Sanger dideoxy method using a Sequenase kit (United States Biochemical) and oligonucleotide primers derived from either vector sequences or from previous sequence analysis of the cloned inserts. In some cases, nested deletion sets were produced by the exonuclease III mung bean system (Stratagene). Oligonucleotide Primer Sequences. pcr 6, GCCTGTTGGAAAGCTTG, and pcr 7, AGTAGAAGAACCACTTC, were used to generate the genomic DNA 483-bp probe designated pcr A. ACCACATGGCTTTGC, the reverse complement of sequence in the first kringle region of apo(a) mRNA, was used to determine the mRNA start site by reverse transcriptase extension. Human liver poly(A)+ RNA was isolated by the guanidium isothiocyanate procedure (18). Abbreviations: apo(a), apolipoprotein(a); LDL, low density lipoprotein; Lp(a), lipoprotein(a). tThe sequences reported in this paper have been deposited in the GenBank data base [accession nos. L07899 for the human apo(a) gene exon 1, L07900 for apo(a)rg-B, and L07901 for apo(a)rg-C]. 1369 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. Proc. Natl. Acad. Sci. USA 90 (1993) Luciferase Reporter Gene Constructs. To assay promoter activity, we synthesized a series of reporter gene constructs (pLuc A-D), comprising the 5' flanking regions of the apo(a) gene and its homologues inserted into the polylinker region of pGeneLight-2 (Promega), a promoterless vector containing the firefly luciferase gene. The genomic inserts were isolated from modified pBluescript plasmids of the original clones from which genomic fragments 3' to the desired flanking regions had been removed by restriction enzyme digestion. The flanking region inserts were isolated from these modified plasmids by digestion with restriction enzymes appropriate for directional ligation into the pGeneLight-2 polylinker. The inserts contained 1.4 kb of DNA upstream of a Bal I site located 12 nucleotides before the translation initiation ATG for the Al, A2, and B fragments, 1.07 kb upstream of this site in the D fragments, and 3.0 kb upstream of a HincHI site at -54 for the C fragment. Transfection and Assay of Reporter Gene Constructs. Plasmids for transfection were purified by two cesium chloride density gradient ultracentrifugations or by Qiagen (Studio City, CA) columns. For transient transfections with luciferase construct plasmids, HepG2 cells were seeded at 0.75 x 106 cells per 6-cm dish and maintained in Dulbecco's modified Eagle's medium (DMEM) (GIBCO/BRL) supplemented with 10% fetal calf serum. Transfections were performed when the cells attained 40-60% confluence. For each dish, 10 ,g of test plasmid DNA and 5 ,ug of pSV-j3galactosidase control plasmid (Promega) were mixed with 1.5 ml of Opti-Mem medium (GIBCO/BRL) in a polystyrene tube, and 1.5 ml of Opti-Mem containing 30 ,ug of Lipofectin (GIBCO/BRL) was added. Complexes were allowed to form for 30 min at room temperature and the lipofectin/DNA mix was added to the dishes after washing the monolayers three times in Pucks saline A (GIBCO/BRL). After 16 hr the medium was replaced by 5 ml ofDMEM containing 10% fetal calf serum, and incubation was continued for 48 hr before assay of cellular luciferase and ,B-galactosidase activity. Monolayers were washed five times with ice-cold phosphate-buffered saline (PBS), scraped into 1 ml of PBS, and pelleted by centrifugation at 4°C for 10 sec in a Microfuge. The pellet was resuspended in 100 ,ul of 0.1 M potassium