Proc. Natl. Acad. Sci. USAVol. 81, pp. 5026-5030, August 1984Biochemistry

Human renin gene: Structure and sequence analysis(recombinant DNA/aspartyl protease/hypertension/homology to pepsin gene/multiple transcriptional promoters)

PETER M. HOBART*, MICHAEL FOGLIANO*, BARBRA A. O'CONNOR*, IDA M. SCHAEFERt,AND JOHN M. CHIRGWINt*Molecular Genetics Research, Pfizer Central Research, Groton, CT 06340; and tDepartment of Anatomy and Neurobiology, Washington University School ofMedicine, St. Louis, MO 63110

Communicated by Max Tishler, April 23, 1984

ABSTRACT The complete protein precursor of humankidney renin has been determined from the sequence of clonedgenomic DNA. The gene spans 12 kilobases of DNA and is in-terrupted by eight intervening sequences. The nine regions(exons) encoding the protein were mapped with a mouse renincDNA probe, synthetic oligonucleotide probes, and by hybrid-ization of genomic restriction fragments to a 1600-nucleotidehuman kidney mRNA. The predicted 403-amino acid pre-prorenin consists of mature renin and a 66-residue amino-ter-minal prepropeptide. The DNA sequence 5' to the first exonindicates the location of a transcriptional promoter (T-A-T-A-A-A) for a mRNA encoding preprorenin. An additional tran-scriptional promoter site is located within the first intron,which, if used, would express a shortened nonsecreted pro-renin. The structure of the human renin gene is similar to thatof human pepsinogen, a closely related aspartyl protease en-zyme. This observation suggests that renin and pepsinogenhave a common evolutionary origin.

Renin is an endocrine hormone catalyzing the first step in acascade of factors that modulate arteriole blood pressure. Ithydrolyzes a single peptide bond in the circulating globulin,angiotensinogen, releasing the amino-terminal decapeptideangiotensin I. The absolute specificity of renin is in contrastto other aspartyl proteases, which act on a broad range ofsubstrates (1). The characterization of this specificity is ofconsiderable pharmaceutical and medical interest. However,detailed biochemical analysis of the protein has been limitedbecause the hormone is produced in such small amounts byits known physiological source, the juxtaglomerular cells ofthe kidney cortex. Renin purified from human tissue hasshown variations in molecular weight and amino acid com-position (2-4).We report here the isolation and sequence analysis of the

human renin gene.

EXPERIMENTAL PROCEDURESPlaque Screening. A library of bacteriophage Charon 4A

containing human fetal DNA (5) was grown in Escherichiacoli strain LE392 (6). Nitrocellulose filter (Schleicher &Schuell) replicas of -300,000 plaques were incubated in 2xNaCl/Cit (lx NaCl/Cit is 0.15 M NaCl/0.015 M Na cit-rate)/0.1% NaDodSO4/5x Denhardt's solution (lx Den-hardt's solution is 0.02% polyvinylpyrrolidone/Ficoll (Phar-macia)/bovine serum albumin) at 55°C overnight and thenhybridized (106 cpm per filter) to a mouse renin probe [la-beled to greater than 108 cpm/,g ofDNA by nick-translation(7) in the presence of [a-32P]dATP and dCTP (New EnglandNuclear)] at 55°C for 24 hr in 2x NaCl/Cit/0.1% Na-DodSO4/1x Denhardt's solution/10% dextran sulfate (Phar-macia). Filters were washed at 50°C in 0.1x NaCl/Cit/0.1%

NaDodSO4 prior to autoradiography. Positive plaques werepurified by standard methods (8). Cloning procedures wereaccording to National Institutes of Health guidelines.

Hybridization to Filter-Bound mRNA. RNA was purified(9) from human kidneys (obtained immediately postmortemor from surgical nephrectomy) and the poly(A)+ mRNA wasfractionated using poly(U)-Sepharose (Bethesda ResearchLaboratories) column chromatography (10). RNA dot hy-bridization assays involved spotting 1-10 jug ofRNA directlyonto activated diazobenzyloxymethyl paper (Schleicher &Schuell). Papers were prehybridized (14-16 hr) and then hy-bridized (24-28 hr) in 50% formamide as described (27).DNA Sequencing Methods. Genomic DNA fragments were

subcloned into pUC9, pUC13 (11), or pUR222 (12) plasmidderivatives of pBR322 for mapping and sequencing. DNArestriction sites were labeled at their 5' termini with [-32P]ATP and T4 polynucleotide kinase (P-L Biochemicals)and at their 3' termini using either a DNA polymerase(Klenow fragment; New England Nuclear) fill-in reactionand [a-32P]dNTP or a terminal nucleotidyltransferase (P-LBiochemicals) reaction with cordycepin [a-32P]triphosphate.DNA sequencing procedures were according to Maxam andGilbert (13).

Oligonucleotide Synthesis. DNA oligomer probes were pre-pared on a Genetic Design (Watertown, MA) automatedDNA synthesizer using a modification (14) of the phosphora-midate procedure of Caruthers (15). Oligomers were purifiedby acrylamide gel electrophoresis.

Peptide Numbering Convention. To clarify the relationshipof this predicted renin sequence to other aspartyl proteases(16, 17), the first amino acid of mature renin is designated 1.References to the predicted mouse submandibular prepro-renin (18) and human prepepsinogen (19) sequences use theirrespective numbering systems.

RESULTSIsolation of the Human Renin Gene. A cloned mouse sub-

mandibular gland renin cDNA probe (20) was used to screen300,000 plaques of a X bacteriophage library of human geno-mic DNA. Ten positive plaques were purified to homogene-ity and shown by restriction endonuclease mapping to repre-sent six nonidentical, overlapping DNA fragments spanning32.5 kilobases (kb) of the genome (Fig. 1A). Two cloned frag-ments, designated XH6 (18.9 kb) and XH10 (20.2 kb) wereselected for further analysis.

Location of Renin Exons. Southern blot analysis of AH10DNA with the mouse submandibular renin probe localizedthe cross-hybridizing region to a 2.3-kb EcoRI/Pst I (XH10-EP2.3) fragment. DNA sequence analysis revealed the fourcarboxyl-terminal exons (Fig. 1 B and C) encoding aminoacids 165-337 of mature human renin (Fig. 2). Since thismouse probe (20) carries only the 3' 57% of the full-length

Abbreviations: NaCI/Cit, standard saline citrate (0.15 M NaCI/0.015 M Na citrate); kb, kilobase(s).

5026

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement"in accordance with 18 U.S.C. §1734 solely to indicate this fact.

Dow

nloa

ded

by g

uest

on

July

8, 2

021

Proc. Natl. Acad. Sci, USA 81 (1984) 5027

A

1 kb

B , ' p5I1HS3. - pn 1U0-4.4pH 10-EP2.3

pH sx0.8 pH W.8 pH 10-R1.3 pH 10.BEO.6 pH iOX1.81 kb - _n -

pH 6.1.2 pH 640.3 pH 10-RO.8 pH 10-XI.6

2 3 4 5

(B) (C) (D)

H X B E HB Xa I m mI

Pow. Ps._o _P-o P S

6 7 8 9

(E) (F) (G) (H)EH XE X

P K S P_ ._

FIG. 1. Restriction endonuclease mapping and DNA sequencing strategy for cloned genomic DNA fragments of human renin. (A) Restric-tion map of six fragments isolated from the human renin gene and cloned into the EcoRI site of Charon 4A phage. (B) Subcloned fragments fromthe XH6 and XH10 phage fragments shown in A. All subcloned fragments shown were tested for ability to hybridize to human kidney mRNA.

, Fragment that hybridized specifically to a 1600-base mRNA; '-, fragment that did not hybridize; , fragment containing one ormore Alu-type repetitive sequences that hybridized to multiple size classes of mRNA. (C) Strategy for sequencing. Arrows indicate directionand approximate length of sequence determined. Restriction sites used for sequencing are not necessarily shown. Exons are numbered 1through 9. Introns are designated (A) through (H). Restriction enzymes used are abbreviated as follows: B, BamHI; E, EcoRI; H, HindlI; K,Kpn I; P, Pst I; S, Sac I; X, Xba I.

mRNA [excluding poly(A)] (18), the positions of the remain-ing amino-terminal exons were determined from the abilityof subcloned genomic DNAs to hybridi4e tp oligonucleotideprobes and/or human kidney mRNA.Mapping with Oligonucleotide Probes. A 20-base oligonu-

cleotide (Fig. 3 A and C), synthesized complementary to nu-cleotides 359-379 (18) of mouse submandibular renin cDNA(a region encoding an amino acid sequence that is highly con-served in several aspartyl proteases) specifically hybridizedto a 2.0-kb BamHI and a 1.2-kb Pst I fragment of XH10. Inaddition, the 20-mer hybridized to a 0.6-kb BamHI/EcoRIfragment of the XH10-E2.8 subclone (not shown). This frag-ment contained exon 3, encoding amino acids 18-58. A sec-ond probe (21-mer) whose sequence was predicted from ami-no acid residues -63 to -57 of the signal peptide for humanpreprorenin (30) hybridized to an 800-base Xba I fragment inthe H6-H3.5 subclone (Fig. 3 B and D). This fragment con-tained the signal peptide exon and 5' flanking DNA sequence.

Location of Exons by Using Kidney mRNA Blots. Sub-cloned genomic DNA fragments were labeled by nick-trans-lation and hybridized to filter-immobilized human kidneypoly(A)+ RNA (dot blots). In this assay, positive hybridiza-tion indicated the presence of coding sequence in the geno-mic DNA fragment. Initially, large restriction fragmentswere tested, and those that proved positive were successive-ly subdivided and rehybridized to the mRNA (Fig. 1B). Inaddition to XH10-EP2.3 (described above), the XH10-E3.4and -E2.8 fragments were positive in this assay. The XH10-EO.9 fragment was negative. Hybridization to kidney reninmRNA was localized to Rsa I fragments of 800 and 400 basesin the XH10-E2.8 subclone and to a 1600-base EcoRI/Xba Ifragment in the XH10-E3.4 subclone. These three fragmentsencoded exons 2, 3, 4, and 5 based on subsequent DNA se-quence analysis. When this strategy for the location of exonswas applied to genomic fragments 5' to exon 2, extensivehybridization to human Alu-like repetitive sequences (22)was seen (unpublished work). The presence of these Alu-like

sequences necessitated the use of the synthetic 21-mer probedescribed above.

Prediction of mRNA Sequence. The boundaries of exonswere positioned by searching the genomic DNA sequencefor consensus splice junction (G-T/A-G) sequences (23) andby comparing the predicted spliced mRNA sequence withthat of mouse submandibular renin (17). When this predictedmRNA sequence was compared with the recently publishedhuman kidney renin cDNA sequences (21, 24), only thesplice site of intron H had to be revised.

DISCUSSIONRestriction mapping and partial seqtence analysis of clonedhuman genomic DNA fragments have revealed the structureof the human renin gene (Fig. 1). The amino acid sequence ofhuman preprorenin was deduced from the DNA sequence(Fig. 2). This precursor contains a signal peptide consistingof an initiating methionine followed by several charged resi-dues and a hydrophobic leucine-rich region. The actual siteof signal peptide cleavage has not been experimentally deter-mined for any renin. Structural homologies to other signalpeptides predict that cleavage could occur after glycine-(-49), cysteine-(-47), or glycine-(-44), resulting in a signalpeptide of 18, 20, or 23 amino acids. The assignment of theamino terminus of the mature human renin enzyme to leu-cine-(+1) is based on sequence homology with mouse sub-mandibular renin whose amino terminus has been deter-mined experimentally (16, 18). Activation of the human andmouse prohormones, therefore, occurs by removal of theprecursor peptide at paired basic residues [lysine-(-2)/ly-sine-(-1) in human renin], a common mechanism in theprocessing of hormone precursors. Other paired basic resi-dues [lysine-(-35)/arginine-(-34), lysine-(-30)/arginine-(-29)] are present in the propeptide region and may repre-sent alternative processing sites. The resulting mature hu-man renin peptide is 337 amino acids long with a calculatedmolecular weight of 36,858. It is 68% homologous in amino

C

(A)

H X x

K S SP S

Biochemistry: Hobart et aL

Dow

nloa

ded

by g

uest

on

July

8, 2

021

5028 Biochemistry: Hobart et al.

-66Met Asp Gly Trp Arg Arg Met

AACCTCAGTGGATCTCAGAGAGAGCCCCAGACTGAGGGMGC ATG GAT GGA TGG AGA AGG ATG-50

Pro Arg Trp Gly Leu Leu Leu Leu Leu Trp Gly Ser Cys Thr Phe Gly Leu ProCCT CGC TGG GGA CTG CTG CTG CTG CTC TGG GGC TCC TGT ACC TTT GGT CTC CCG

-34Thr Asp Thr Thr Thr Phe Lys Arg INTRON AACA GAC ACC ACC ACC TTT AAA CG GTAATTGGTMC::: (4.Okb):::ACMGAAGTAACTC

TTATAAATGCTCCAGAGGCCCTCAGTGACAGAGGTGATTTCCAGGTGGCTGGGCTAACGTTAAAGGTGGTT-33 -30Ile Phe Leu Lys Arg Met Pro

GACAGCACTTTTCTATTTTTGCTTCCTCCACCCTGGGCCAG GATC TTC CTC MG AGA ATG CCC-20 -10

Ser Ile Arg Glu Ser Leu Lys Glu Arg Gly Val Asp Met Ala Arg Leu Gly ProTCA ATC CGA GAA AGC CTG AAG GAA CGA GGT GTG GAC ATG GCC AGG CTT GGT CCC

-1 1 10Glu Trp Ser Gln Pro Met Lys Arg Leu Thr Leu Gly Asn Thr Thr Ser Ser ValGAG TGG AGC CAA CCC ATG AAG AGG CTG ACA CTT GGC AAC ACC ACC TCC TCC GTG

17 18Ile Leu Thr Asn Tyr Met Asp INTRON B ThrATC CTC ACC AAC TAC ATG GAC GTGAGTGCCTGG::: (0.6kb):::TTACCCCCACAG ACC

20 30Gln Tyr Tyr Gly Glu Ile Gly Ile Gly Thr Pro Pro Gln Thr Phe Lys Val ValCAG TAC TAT GGC GAG ATT GGC ATC GGC ACC CCA CCC CAG ACC TTC AAA GTC GTC

40 50Phe Asp Thr Gly Ser Ser Asn Val Trp Val Pro Ser Ser Lys Cys Ser Arg LeuTTT GAC ACT GGT TCG TCC AAT GTT TGG GTG CCC TCC TCC AAG TGC AGC CGT CTC

58 59 60Tyr Thr Ala Cys INTRON C Val Tyr His LysTAC ACT GCC TGTG GTGAGACCTAAG::: (0.6kb)::TCCCCCTGCCAG TG TAT CAC AAG

70 80Leu Phe Asp Ala Ser Asp Ser Ser Ser Tyr Lys His Asn Gly Thr Glu Leu ThrCTC TTC GAT GCT TCG GAT TCC TC$ AGC TAC AAG CAC AAT GGA ACA GAA CTC ACC

90 98Leu Arg Tyr Ser Thr Gly Thr Val Ser Gly Phe Leu Ser Gln Asp Ile Ile ThrCTC CGC TAT TCA ACA GGG ACA GTC AGT GGC TTT CTC AGC CAG GAC ATC ATC ACC

99 100INTRON D Val Gly Gly Ile Thr Val Thr Gln

GTAAGTTGGGCC::(0.9kb):::TTCCTCCCACAG GTG GGT GGA ATC ACG GTG ACA CAG110 120

Met Phe Gly Glu Val Thr Glu Met Pro Ala Leu Pro Phe Met Leu Ala Glu PheATG TTT GGA GAG GTC ACG GAG ATG CCC GCC TTA CCC TTC ATG CTG GCC GAG TTT

130 140Asp Gly Val Val Gly Met Gly Phe Ile Glu Gln Ala Ile Gly Arg Val Thr ProGAT GGG GTT GTG GGC ATG GGC TTC ATT GMA CAG GCC ATT GGC AGG GTC ACC CCT

Proc. Natl. Acad. Sci. USA 81 (1984)

150 160Ile Phe Asp Asn Ile Ile Ser Gln Gly Val Leu Lys Glu Asp Val Phe Ser PheATC TTC GAC MC ATC ATC TCC CAA GGG GTG CTA AAA GAG GAC GTC TTC TCT TTC

164 165Tyr Tyr Asn Arg INTRON E Asn Ser Gln SerTAC TAC AAC AG GTGGGGACTGGG:: (2.4kb):::TCCCCCTGCCAG GMT TCC CAA TCG

170 180Leu Gly Gly Gln Ile Val Leu Gly Gly Ser Asp Pro Gln His Tyr Glu Gly AsnCTG GGA GGA CAG ATT GTG CTG GGA GGC AGC GAC CCC CAG CAT TAC GAA GGG AAT

1 90 200 204Phe His Tyr Ile Asn Leu Ile Lys Thr Gly Val Trp Gln Ile Gln Met Lys GlyTTC CAC TAT ATC AAC CTC ATC MG ACT GGT GTC TGG CAG ATT CAA ATG MG GGG

205 210INTRON F Val Ser Val Gly Ser Ser Thr Leu

GTCAGAAATCCT:: (0.3kb)::: GCCTCCCCCAAG GTG TCT GTG GGG TCA TCC ACC TTG220 230

Leu Cys Glu Asp Gly Cys Leu Ala Leu Val Asp Thr Gly Ala Ser Tyr Ile SerCTC TGT GAA GAC GGC TGC CTG GCA TTG GTA GAC ACC GGT GCA TCC TAC ATC TCA

240Gly Ser Thr Ser Ser Ile Glu Lys Leu Met Glu Ala Leu Gly Ala Lys Lys ArgGGT TCT ACC AGC TCC ATA GAG AAG CTC ATG GAG GCC TTG GGA GCC MG MG AGG

250 251 252Leu Phe Asp INTRON G Tyr Val Val Lys CysCTG TTT GAT GTAAGAAGCCAAA:: (0.2kb)::CCCCCACCCCAG TAT GTC GTG AAG TGT

260 270Asn Glu Gly Pro Thr Leu Pro Asp Ile Ser Phe His Leu Gly Gly Lys Glu TyrAAC GAG GGC CCT ACA CTC CCC GAC ATC TCT TTC CAC CTG GGA GGC MA GAA TAC

280 284Thr Leu Thr Ser Ala Asp Tyr Val Phe Gln INTRON HACG CTC ACC AGC GCG GAC TAT GTA TTT CAG GTGAGGTTCGAG::(0.6kb):::CCTTCC

285 290 300Glu Ser Tyr Ser Ser Lys Lys Leu Cys Thr Leu Ala Ile His Ala Met

TGCCAG GAA TCC TAC AGT AGT MA MG CTG TGC ACA CTG GCC ATC CAC GCC ATG310

Asp Ile Pro Pro Pro Thr Gly Pro Thr Trp Ala Leu Gly Ala Thr Phe Ile ArgGAT ATC CCG CCA CCC ACT GGA CCC ACC TGG GCC CTG GGG GCC ACC TTC ATC CGA

320 330Lys Phe Tyr Thr Glu Phe Asp Arg Arg Asn Asn Arg Ile Gly Phe Ala Leu AlaMG TTC TAC ACA GAG TTT GAT CGG CGT AAC AAC CGC ATT GGC TTC GCC TTG GCC337Arg OPCGC TGA GGCCCTCTGCCACCCAGGCAGGCCCTGCCTTCAGCCCTGGCCCAGAGCTGGMCACTCTCTG

AGATGCCCCTCTGCCTGGGCTTATGCCCTCAGATGGAGACATTGGATGTGGAGCTCCTGCTGGATGCGTGC

CGTTGCATCTGGGTTCACTAGGGTTAGMCAGAGGGAGGGGCTGCGTGATCATGTGTGGACAGGAMTGTGA

FIG. 2. Composite nucleotide and predicted amino acid sequences of the human renin gene. Predicted 5' and published (21) 3' termini of thetranscribed region. An additional 68 bases of 5' and 70 bases of 3' flanking regions are shown. The approximate size of the intron is given inparentheses. The T-A-T-A-A-A sequence in the 5' flanking region and the one in intron A proximal to exon 2 are overlined and underlined. TheA-A-T-A-A-A polyadenylylation signal sequence in the 3' untranslated region is underlined.

acid sequence to the 338-residue mouse enzyme. The twocoding nucleotide sequences are 77% homologous. Purifiedmouse submandibular renin may be cleaved into a 288-aminoacid A peptide and 48-amino acid B peptide at paired argi-nine residues (16, 18). There is currently no published evi-dence that human renin is cleaved into a two-peptide en-zyme. Moreover, recent biosynthetic studies in the mousesuggest that the two-chain form of submandibular renin maynot be an obligatory intermediate in the activation pathwayof the enzyme (25, 26).Human renin, which is reported to be a glycoprotein (4),

has two presumptive N-glycosylation sites, Asn-Thr-Thr(positions 5-7) and Asn-Gly-Thr (positions 75-77). Thesesites are located in exons 2 and 4, respectively, placing themnear the amino terminus of the mature renin enzyme and onseparate domains from the catalytically important asparticacid residues. The mouse submandibular renin, a nonserumprotein, lacks any recognizable glycosylation acceptor sites(16, 18).The positions of aspartates-38 and -223 (residues 32 and

215 in the pepsin numbering) and the conservation of flank-ing residues between aspartyl proteases indicate that theyare the two acidic groups directly involved in the renin cata-lytic activity (17). From the gene sequence, it is clear thatthese aspartic acid residues lie in the middle of separate ex-ons, suggesting that these domains may contribute to thestructure of the active site.

Alternative Exon-Intron Junctions. It is possible to predictsplice junctions for all eight introns using homologies to thesplice site consensus sequence G-T/A-G (23) and to the se-

quence of mouse submandibular renin cDNA. By this meth-od, exon junctions with introns D, E, and F were unambigu-ous. There are, however, alternative splice sites for intronsA, B, C, G, and H that retain reasonable homology to themouse sequence (Fig. 4). The junctions shown in Fig. 4 cor-respond to those found in the human kidney cDNA sequence(21, 24). Alternative splice sites could result in significantchanges to the translated sequence. For example, the alter-native sites for introns A and C would eliminate basic resi-dues in exon 2 and a cysteine residue in exon 3, respectively.Such changes could affect both the processing and the struc-ture of the human renin precursor peptide.Polymorphism. The only difference between the sequence

of human renin mRNA predicted from the gene (Fig. 2) andthe sequence derived from the complete cDNA clone (21)occurs at the splice site of exons 5 and 6, where both cDNAshave three additional amino acids. Our sequence of thissplice junction is unambiguous and involves sequence deter-mination of both DNA strands. Neither alternative splicesites nor the appropriate extra codons in this region havebeen found. These three amino acids (Asp-Ser-Glu) do notoccur in mouse submandibular renin. Since Southern blotsof genomic DNA (unpublished work; ref. 24) support only asingle-copy human renin gene, located on chromosome 1(27), two reasonable explanations for the extra three aminoacids can be offered. Either they are carried on a nine-baseexon that was not detectable in the 2.8-kb intron E by thescreening methods employed here or there is a coding se-quence polymorphism in humans, resulting in two slightlydifferent renins. If this latter explanation is correct, we have

Dow

nloa

ded

by g

uest

on

July

8, 2

021

Proc. NatL. Acad. Sci USA 81 (1984) 5029

* + ^+

Trp Val Pro Ser Thr Lys Cys5'TGG GTG CCC TCC ACC AAG TGC

ACC CAC GGG AGG TGG TTC AC 5'x

5'TGG GTG CCC TCC TCC AAG TGCTrp Val Pro Ser Ser Lys Cys

Mouse Submandibular ReninMouse Coding Region

20 Base Oligonucleotide Probe

Human Coding SequenceHuman Renin Peptide Sequence

B

Trp Arg Arg Met Pro Arg Trp

5'TGG AGG AGG ATG CCT CGC TGG

Human Signal Peptide

Predicted Coding Sequence

ACC TCC TCC TAC GGA GCG ACC 5' 21 Base Oligonucleotide Probex

5'TGG AGA AGG ATG CCT CGC TGG Human Coding Sequence

xH6-H3.5 XH1O-E2.8r -- r

XH6-H3.5 XHI1O-E2.81 1 [

Std. E Xb R E Xb R Std. E Xb R E

0.3-

0.56 -

Xb R

1 ... .M j

PW,;

:., .;..:t .~~~~ ~~ ~ ~~~~~~~~............:

....

.....

Xba- 0.8

Rsal0.25

FIG. 3. Mapping of cloned genomic DNA fragments with oligonucleotide probes. Oligonucleotides were chemically synthesized and endlabeled for use as probes. (A and B) Rationale for choosing the two oligonucleotide sequences. (C and D) Ethidium bromide-stained gels ofrestriction endonuclease-digested cloned genomic DNAs (Left) and the corresponding autoradiographs (Right) of filters blotted and hybridizedto the respective oligonucleotide labeled probe. Hybridization was for 48 hr at 50°C in 4x NaCV/Cit/lOx Denhardt's solution containing tRNAat 2 mg/ml and glycine at 8 mg/ml. Filters were washed (six times, 10 min each) at room temperature in 2x NaCl/Cit. (A and C) Results ofhybridization of the 20-mer to XH10 DNA. *, Amino acid residues conserved in all known aspartyl proteases; +, amino acid residues conservedin mammalian aspartyl proteases; X, mismatch of the oligonucleotide probe relative to the determined human renin gene sequence. The lanes ofthe agarose gel and corresponding filter autoradiograph are as follows: Std., HindIII-digested X DNA molecular weight standard; E, EcoRIdigest; P, Pst I digest; E/P, EcoRI/Pst I digest. (B and D) Results of hybridization of the 21-mer to XH6-H3.5 and XH1O-E2.8DNAs. *, Aminoacid residues conserved between human and mouse preprorenins; X, mismatch of the oligonucleotide probe relative to the determined humanrenin gene sequence. The lanes of the agarose gel and corresponding filter autoradiograph are as follows: Std., HindIII-digested X DNA andHae III-digested 4X174 DNA molecular weight standards; E, EcoRI digest; Xb, Xba I digest; R, Rsa I digest.

sequenced a renin structural allele different from that de-scribed by others.Regulatory Regions of the Renin mRNA. DNA sequencing

5' to the region encoding the signal peptide indicated a Hog-ness box ("TATAAA") sequence at -73 nucleotides fromthe initiation codon. Comparing this human renin sequencewith the 5' untranslated region of the human pepsinogengene (19), the mouse submandibular renin cDNA (18), andthe human kidney cDNA (21) predicts that initiation of tran-scription could occur 28 bases 3' to the T-A-T-A-A-A se-quence. If this is correct, the mRNA will possess a 45-nucle-otide 5' untranslated region.

Partial sequence analysis of the first renin intron (A) indi-cates that there are at least three additional T-A-T-A-A-Asequences in the gene. One such T-A-T-A-A-A sequence ex-ists 110 bases before the 3' end of intron A. If this presump-tive Hogness box were functional, transcription would giverise to a mRNA encoding a 48-residue-foreshortened nonse-creted prorenin molecule beginning with methionine-28 inexon 2. This would support the evidence showing that neuro-blastoma cells from the rat have an intracellular renin (28).There is a standard polyadenylylation signal sequence (A-

A-U-A-A-A) 184 nucleotides after the translation termina-tion UGA codon. Thus, the calculated size of the mRNA is1448 nucleotides plus poly(A) tail. This is consistent with theestimated length of the kidney message (1550-1600 bases)

using denaturing agarose gel electrophoresis (27).Comparison of Human Renin and Pepsin Genes. The struc-

tures of human renin and human pepsinogen genes are re-markably similar (19). Both genes are split by eight interven-ing sequences, all occurring in the protein coding region. Inaddition, the size of the corresponding exons and the loca-tion of introns in the expressed (mRNA) gene sequence forrenin and pepsinogen are nearly congruent (Table 1). Themajor exception is exon 1, which encodes a longer renin pre-propeptide (46 residues versus 33 residues in human pepsin-ogen). However, this exon does not contribute to the struc-ture of the mature enzyme. There is also 66% homology be-tween the amino acid sequences of exon 3, which includesthe first catalytic aspartyl residues for both human renin andhuman pepsinogen. This conservation of sequence (Table 1)is exceptional relative to the rest of the enzyme and certainlydifferentiates it from exon 7 (33%), which carries the othercatalytic aspartic acid residue.The similarities in the structure of the human renin and

human pepsinogen genes support the view that the mammali-an aspartyl protease family of enzymes (which includes chy-mosin and cathepsin D) share a common protein structureessential for activity (17). This conserved structure is appar-ently built from peptide domains of similar size that are en-coded by individual exons. However, the substrate specific-ities of these aspartyl proteases differ widely. For example,

A

DStd. E E/P P

C

23.1 -

9.4-

6.6-

4.4-

2.3-

2.0-

23.1-

9.4 -

6.6 -

4.4

EcoRI- 2.0

PstI

- 1.2

2.3 -2.0 i:

1.3 -

1.1

0.9 -

0.6 -

Biochemistry: Hobart et al.

Dow

nloa

ded

by g

uest

on

July

8, 2

021

5030 Biochemistry: Hobart etaLP

TARN 6TIA6T Pyrimidine rich....NqA6 619-34 -33 f

ThrThrPheLysArq fIIPhoLuLysArgACCACCTTTAAAC6 6TAATT66TAACTCASS TCCTCCACCCTS66CCA6 SATCTTCCTCAASAGA (H)ACCACCTTTSAAC6 AATCCCACTCAA6AAA (N)ThrThrPhe61uArg IleProLeuLysLys

131 ~~~~~~~~~~~~~3117 Sa

AmnTyrketAip Thr6lnTyrTyr6I ySuAACTACATS66A 6T6A6T6CCT66CTCA6 CTCTTTTTACCCCCACA6ACCCABTACTATS6C6A6S(H)

1***t to * tf**IAACTACCT6AAT A6CCA6TACTAT66CGA6 (H)AsnTyrLeuA t irrfnTyrTyr6lvyIu

WTyrThrAl iCy

TACACTSCCTST6 6T6AGACCTAA6

TACCTTSCTT6TBTyrLeuAlaC s

121

59Y& Ter~i sLysLeu

TCCCCCT6CCAS TOTATCACAASCTC (H)SBATTCACABCCTC (N)61 lleHiiSerLeu122I

251 2r521VArtLeuPheAspTVyrA66CTGTTTGAT 6TAAAA6CCAAAS CCCCCCACCCCAG TATSTCST6AAS

A6ACTACAT6AA TAT6TT6T6A6CArgLeuies6lu TyrYalYV1Ser31531

254 285Tyr~aIPhe~in SluSerTyrSerSerTAT6TATTTCA6 6T6A66TTC6A6TC66CCC CCCCCTTCCT6CCAG 6AATCCTACA6TAST (H)

TAC6T6CTACA6 TATCCCAACA66AGA (N)TyrValLeu3ln T rProAsnArgArg

348 1

INTRON A

INTRON

INTRON C

INTRON 6

INTRON H

FIG. 4. Alternative human renin intron-exon junction sites. Ca-nonical intron-exon junction sequences are depicted at the top ofthe figure. The splice sites shown correspond to those determinedfrom the human renin cDNA sequence (21, 24). Alternative junc-tions are indicated by vertical arrows. H, human renin gene se-quence; M, mouse submandibular cDNA sequence. *, Homology inthe coding sequences. Amino acid numbers correspond to those inFig. 2 for the human sequence and to those of Panthier et al. (18) forthe mouse sequence.

pepsin cleaves at all internal aromatic residues whereas re-nin, the most specific of this group of enzymes, cleaves atonly one site in a single peptide substrate, angiotensinogen.Moreover, individual mammalian renins exhibit reduced ac-tivity with other mammalian angiotensinogens (29). Thus,specific amino acid side chains at the active site cleft of hu-man renin and species-specific structural features of angio-tensinogens (T. Blundell, personal communication) mustcontribute to the stringent specificity of the human renin-angiotensin interaction.

Table 1. Comparison of the human renin and pepsinogengene structures

Exon size* Percent homologyHuman Human Amino Nucleicrenin pepsinogen acid acid

Exon 1 33 19 18% (6/33) 27% (26/98)Exon 2 50 54 30% (16/54) 43% (69/162)Exon 3t 41 39 66% (27/41) 67% (82/123)Exon 4 40 40 30% (12/40) 55% (67/120)Exon 5 66 67 37% (25/67) 48% (96/201)Exon 6 40 39 25% (10/40) 41% (50/120)Exon 7t 47 48 33% (16/48) 51% (74/144)Exon 8 37 33 28% (12/33) 41% (46/111)Exon 9 53 49 36% (19/53) 51% (79/154)The nucleic acid and predicted amino acid sequences of human

renin and human pepsinogen (19) were compared for homologoussequence within the peptide coding region. Where correspondingexons of either renin or pepsinogen differed in size, the smaller exonwas gapped so as to align the remaining amino acid sequence withmaximum homology.*Expressed as number of amino acid residues.tExon containing catalytic aspartic acid residues.

We thank Dr. Glenn Andrews for preparation of oligonucleotideprobes, Dr. Tom Maniatis for providing us with the human genomicDNA library, and Drs. Steven Atlas and Edwin Clayton for theirhelp in obtaining human kidney tissue. We also appreciate the assist-ance of Beatrice L. Ralls in preparation of the manuscript. J.M.C. issupported in part by a Basil O'Connor Starter Grant from the Marchof Dimes and by the American Heart Association.

1. Foltmann, B. & Pedersen, V. B. (1977) in Acid Proteases,Structure, Function, and Biology, ed. Tang, J. (Plenum, NewYork), pp. 3-22.

2. Galen, F.-X., Devaux, C., Guyenne, T., Menard, J. & Corvol,P. (1979) J. Biol. Chem. 254, 4848-4855.

3. Yokosawa, H., Holladay, L. A., Inagami, T., Haas, E. & Mur-akami, K. (1980) J. Biol. Chem. 255, 3498-3502.

4. Slater, E. E. & Strout, H. V. (1981) J. Biol. Chem. 256, 8164-8171.

5. Lawn, R. M., Fritsch, E. F., Parker, R. C., Blake, G. & Man-iatis, T. (1978) Cell 15, 1157-1174.

6. Woo, S. L. C. (1979) Methods Enzymol. 68, 389-395.7. Rigby, P., Dieckmann, M., Rhodes, C. & Berg, P. (1977) J.

Mol. Biol. 113, 237-251.8. Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) Molecular

Cloning: A Laboratory Manual (Cold Spring Harbor Labora-tory, Cold Spring Harbor, NY).

9. Chirgwin, J. M., Przybyla, A. E., MacDonald, R. J. & Rutter,W. J. (1979) Biochemistry 24, 5294-5299.

10. Deeley, R. G., Gordon, J. I., Burns, A. T. H., Mullinix,K. P., Bina-Stein, M. & Goldberger, R. F. (1977) J. Biol.Chem. 252, 8310-8319.

11. Vieira, J. & Messing, J. (1982) Gene 19, 259-268.12. Ruther, U., Koenen, M., Otto, K. & Muller-Hill, B. (1981) Nu-

cleic Acids Res. 9, 4087-4098.13. Maxam, A. M. & Gilbert, W. (1980) Methods Enzymol. 65,

499-560.14. Adams, S. P., Kavka, K. S., Wykes, E. J., Holder, S. B. &

Galluppi, G. R. (1983) J. Am. Chem. Soc. 105, 661-663.15. Caruthers, M. H. & Beaucage, S. L. (1981) Tetrahedron Lett.,

1859-1862.16. Misono, K. S., Chang, J. J. & Inagami, T. (1982) Proc. NatI.

Acad. Sci. USA 79, 4858-4862.17. Blundell, T., Sibanda, B. L. & Pearl, L. (1983) Nature (Lon-

don) 304, 273-275.18. Panthier, J.-J., Foote, S., Chambraud, B., Strosberg, A. D.,

Corvol, P. & Rougeon, F. (1982) Nature (London) 298, 90-92.19. Sogawa, K., Fujii-Kuriyama, Y., Mizukami, Y., Ichihara, Y.

& Takahashi, K. (1983) J. Biol. Chem. 258, 5306-5311.20. Chirgwin, J. M., Schaefer, I. M., Diaz, J. A. & Lalley, P. A.

(1984) Somatic Cell Mol. Genet., in press.21. Imai, T., Miyazaki, H., Hirose, S., Hori, H., Hayashi, T., Ka-

geyama, R., Ohkubo, H., Nakanishi, S. & Murakami, K.(1983) Proc. NatI. Acad. Sci. USA 80, 7405-7409.

22. Deininger, P. L., Jolly, D. J., Rubin, C. M., Friedmann, T. &Schmid, C. W. (1981) J. Mol. Biol. 151, 17-33.

23. Sharp, P. A. (1981) Cell 23, 643-646.24. Soubrier, F., Panthier, J.-J., Corvol, P. & Rougeon, F. (1983)

Nucleic Acids Res. 11, 7181-7190.25. Catanzaro, D. F., Mullins, J. J. & Morris, B. J. (1983) J. Biol.

Chem. 258, 7364-7368.26. Pratt, R. E., Ouellette, A. J. & Dzau, V. J. (1983) Proc. Natl.

Acad. Sci. USA 80, 6809-6813.27. Chirgwin, J. M., Schaefer, I. M., Rotwein, P. S., Piccini, N.,

Gross, K. W. & Naylor, S. L. (1984) Somatic Cell Mol. Genet.10, 415-421.

28. Okamura, T., Clemens, D. L. & Inagami, T. (1981) Proc. Natl.Acad. Sci. USA 78, 6940-6943.

29. Tewksbury, D. A., Dart, R. A. & Travis, J. (1981) Biochem.Biophys. Res. Commun. 99, 1311-1315.

30. Miyazaki, T., Imai, T., Hirose, S., Hori, H., Hayashi, R., Ka-geyama, R., Ohkubo, H., Nakanishi, S. & Morakami, K.(1983) Jpn. J. Hypertension 6, 19 (abstr.).

Proc. NatL Acad Sci. USA 81 (1984)

Dow

nloa

ded

by g

uest

on

July

8, 2

021