Cloning, Expression, and Chromosomal Localization of the Mouse ...

THEJOURNALOF BIOLOGICAL CHEMISTRY

Vol. 268,No. 28, Iasue of October 5, pp. 21036-21043,1993 Printed in U.S. A.

Cloning, Expression, and Chromosomal Localization of the Mouse Meprin /3 Subunit* (Received for publication, May 3, 1993, and in revised form, June 21,1993)

Carlos M. Gorbea$#, Petra Marchand$, Weiping Jiang$, Neal G. Copelandll, Debra J. Gilbertll, Nancy A. Jenkinsll, and JudithS . Bondall From the $Department of Biological Chemistry, Pennsylvania State University College of Medicine, Hershey, Pennsylvania 17033 and the lMammalian Genetics Laboratory, Advanced BwScience Laboratories, Im.-Basic Research Program, Natwnul Cancer Institute-Frederick Cancer Research and Development Center, Frederick, Marylnnd 21 702

Meprins are plasma membrane homo- or hetero-oli- ICR and BALB/c mice and the rat enzyme from Spraguegomeric metalloendopeptidases that contain glycosy- Dawley strains (Beynon et al., 1981; Kenny and Ingram, 1987; lated a and/or @ subunits. This paper reports the clon- Kounnas et al., 1991). These enzymes are capable of hydroing andsequencing of the mouse kidney @ subunit. The lyzing a great variety of peptides and proteins such as bradyprimary translation product is composed of 704 amino kinin, insulin B chain,glucagon, transforming growth factoracids which includes a transient signal sequence of 20 a, and azocasein (Butler et al., 1987; Choudry and Kenny, amino acids at the NH2 terminus. The proteasedomain 1991; Wolz et al., 1991). Meprin A (sometimes referred to as (Asn-63 to Leu-260) contains the putativezinc-binding endopeptidase-2 in rats)contains two types of subunits, a and motif characteristic of metalloendopeptidases of the p, which are highly glycosylated and linked by disulfide “astacin family.” The COOH terminus containsan epi- bridges. Molecular weight estimates for the subunits, from dermal growth factor-like domain, a potential mem- mobility when subjected to sodium dodecyl sulfate-polyacrylbrane-spanning domain, and an additional 26 amino amide gel electrophoresis, have ranged from 72,000 to 110,000. acids. The B subunit hasan overall 42% identity to the a subunit, however,a 56-amino acidsegment near the The a subunits from mouse and rat have been cloned and COOH terminus of a is missing in B, and the putative sequenced and shown to be 87% identical (Jiang et al., 1992; transmembrane and cytoplasmic domain’s of the sub- Corbeil et al., 1992).A human intestinalform of the a subunit units share no significant sequence similarity. NH2- of meprin A (referred to as PPH2)also has over 80% identity terminalanalyses of detergent-solubilized mature with mouse meprin a (Dumermuth et al., 1991).’ The strucforms revealed that, unlike a, the prosequence (Leu- tural gene for the mouse a subunithas beenlocalized to 2 1to Lys-62) isnot removed from the @ subunit. North- chromosome 17 near the major histocompatibility complex ern blot analysis revealed a 2.5-kilobase message for (Jiang et al., 1993). the @ subunit in the kidney and intestine of C57BL/6 The expression of the a subunit is tissue-specific and differs and C3H/Hemice. The gene for the @ subunit was in mice and rats (Jiang et al., 1993). In those mouse strains localized to mouse chromosome 18. These studies in- that express a, the mRNA was found only in the kidney; no dicate that a and @ probably derived from a common message waspresent in intestine, brain, heart, skeletal muscle, ancestral gene, but have evolved so that their genes liver, lung, or spleen. In Sprague-Dawley rats, it is expressed are on two different chromosomes, and their tissue- in both kidney and intestine.Thus, expression of the a specific expression and post-translational processing subunit is highly specific and differs in thetwo species. differ. From NHz-terminal analyses of mouse andrat meprin subunits, the a and /3 subunits were knownto be evolutionarily related (Jiang et al., 1992). Furthermore, sequencing of the protease domain of the ratp subunit revealed a 55% identity Kidney brush border membranes of some strains of mice with mouse a, and 54% identity with human PPH2 (Johnson and ratscontain a high concentration of a metalloendopepti- and Hersh, 1992). The reported rat p sequence, deduced from dase called meprin A (EC 3.4.24.18) (Kounnas et al., 1991; a full-length cDNA, however, differed from rat and mouse a Jiang et al., 1992). The mouse enzyme has been isolated from in that neither an epidermal growth factor (EGF)*-like domain nor a hydrophobic domain was found near the COOH * This work was supported by the National Institutes of Health terminus. The COOH-terminal hydrophobic sequence is Grant DK 19691 (to J. S. B.) and by the National Cancer Institute, thought to be the transmembrane domain for a. Johnson and Department of Health and Human Services, under contract N01CO-74101 with Advanced Bioscience Laboratories, Inc. (to N. A. J.). Hersh (1992) suggested that the ratp subunit is anchored to The costs of publication of this article were defrayed in part by the the membrane by an NHz-terminal hydrophobic domain, the payment of page charges. This articlemusttherefore be hereby equivalent of which is removed in the processing of the a marked “advertisement” in accordance with 18 U.S.C. Section 1734 subunit. solely to indicate this fact. Inbred mouse strains are particularly useful for the eluciThe nucleotide sequence(s) reported in thispaper h m been submitted to the GenBankTM/EMBLData Bank with accession number(s) dation of meprin structuresand expression because some mouse strains (such as C3H and CBA mice) do not express L15193. 3 Present address: Dept. of Biochemistry, University of Utah, Salt the a subunit in the kidney. These mice do not, therefore,

Lake City, UT 84132. llTo whom correspondence and reprint requests should be addressed Dept. of Biological Chemistry, Penn StateUniversity College of Medicine, Hershey PA 17033. Tel: 717-531-8586; Fax: 717-5317072.

E. E. Sterchi, personal communication. The abbreviations used are: EGF, epidermal growth factor; PCR, polymerase chain reaction; kb, kilobase($; cM, centiMorgans; RACE, rapid amplification of cDNA ends.

21035

The 6 Subunit of Mouse Meprin

21036

have meprin A which is characterized by a high azocaseinase activity. They do, however, have meprin B, which is a homooligomer of /3 subunits (Gorbea et al., 1991). Papain-purified or membrane-associated meprin B has little or no activity against bradykinin or azocasein and a distinct peptide bond specificity toward the insulin B chain compared to meprin A (Kounnas et al., 1991). Meprin B has been found to have latent activity toward azocasein which can be activated in vitro by trypsin-like enzymes. In order to further understand the structure and expression of mouse meprins, the mouse /3 subunit was cloned and sequenced, and the expression and gene localization were investigated. The results herein indicate that the mouse meprin /3 subunit has a similar sequence and domain structure compared to a, including EGF-like and transmembrane domains near the COOH terminus. The NH2-terminal hydrophobic domain is removed in themature form of the enzyme, and therefore this sequence cannot anchor the subunit to the /3 subunit is expressed membrane as suggested for the rat. The in the kidney of mice whether or not they express a, and in the intestine of both types of mouse strains, unlike a. In addition, the structural gene for the /3 subunit ( M e p - l p )was localized to chromosome 18 unlike the structuralgene for the a subunit which is located on chromosome 17. EXPERIMENTALPROCEDURES

cDNA

c J

-

“++” ++” e e

1

2004

~~

~

-

___)

-”

t

” 1

FIG. 1. Cloning and sequencing strategy of mouse meprin B subunit cDNA. DNAs encoding residues 65-345 and 98-345 were obtained by the PCR using the Clontech or Berger mouse library cDNAs as templates, and either oligo-1 or oligo-2, in combination with oligo-3 as primers (see Table I). The resulting PCR products were cloned into pBluescript KS+ or SK- vectors and sequenced using the primers for the vectors KS, SK, T3, and T7. The restriction enzyme AccI was used to construct subclones for sequencing. The 5’end of the cDNA encoding amino acids 1-76 was obtained using the 5’-RACE system; the sequence for amino acid residues 338-704 was obtained through the 3’-RACE procedure (see “Experimental Procedures”). The PCR products of the 5’-RACE and 3’-RACE reactions were cloned into pAMP 1vector and sequenced using the primers T7 and SP6. Subclones were generated using KpnI, PstI, XbaI, and StuI restriction sites. Arrows indicate the extent and direction of the sequencing information obtained. Arrows originating a t sites other than restriction sites represent sequence obtained by extension from synthetic oligonucleotides.

Animals-Adult male mice (strains of C57BL/6 and C3H/He) were obtained from Dr. Francis Gwazdauskas of the Department of Dairy Science at Virginia Polytechnic Institute and State University or purchased directly from Jackson Laboratories. Male ICR frozen mouse kidneys were obtained from Rockland Farms. NH2-termimlAmino Acid Sequence Analysis of Detergent-solubilized Meprin Subunits-Renal brush border membrane fractions were prepared from kidneys of ICR, C57BL/6, and C3H/He mice as previously described (Butler and Bond, 1988). Meprin was solubilized from the brush border membranes by incubation with 100 mM n- performed in a finalvolume of 100 p1 and contained 2.5 pg of library octyl-@-D-glucopyranosidefor 2 h a t 4 “C,followed by centrifugation DNA, 100 pmol of each primer, and 2 mM MgClp. Amplification on for 1 h at 100,000 X g. For some preparations, the resulting superna- the GeneAmp PCR System 9600 (Perkin Elmer-Cetus) was performed tant fractions were applied to a concanavalin A-Sepharose column as described by Jiang et al. (1993). The amplified DNA fragments (Pharmacia LKB Biotechnologies Inc.) equilibrated and washed with were of the predicted size (843 and 744 base pairs using oligo-1 or Tris-buffered saline containing100 mM n-octyl-@-D-glucopyranoside, oligo-2 in combination with oligo-3, respectively). The amplified DNA and thesubunits were eluted with buffer containing 1M a-D-methylfragments were cloned into pBluescript KS+ and SK- (Stratagene). mannopyranoside. The partially purified samples were boiled with The 5’-end of the meprin p subunit cDNA wasobtained by the RACE 2% SDS and 5% @-mercaptoethanol,subjected to SDS-polyacrylprocedure of Frohman et al. (1988) using the 5’-RACE System from amide gel electrophoresis according to the method of Laemmli and Favre (1973) on 10% polyacrylamide gels, and transferred to polyvi- Life Technology Inc. (Fig. 1). Total RNA (1pg) from C3H/He mouse kidney was used for the reverse transcriptase reaction using oligo-4 nylidene difluoride membranes. Portions of the Coomassie-stained (Table I) as primer. The resulting cDNA was “tailed” with deoxycymembranes containing individual subunits were excised for direct tidine using terminal deoxynucleotide transferase and amplified by NHp-terminal sequence analyses which were performed in the Protein “anchor” primer. Chemistry Laboratory, Department of Molecular Biology and Phar- the PCR using gene-specificoligo-5 (Table I) and an A portion from the first round of the PCR was amplified with a macology, at Washington University in St. Louis. Phosphutidylinositol-Glycan Anchor Determinution-Renal brush “nested gene-specific primer oligo-6 (Table I), and a “universal border membrane-enriched fractions from C3H/He mouse kidneys amplification” primer. The products from the second round of the were incubated at 37 “C for 90 min in the absence or presence of PCR were isolated from agarose gels and cloned into pAmpTMlusing phosphatidylinositol-specific phospholipase C (ICN Biochemicals). the CloneAmpTMSystem (Life Technology Inc.). The 3’-end of the After centrifugation at 4 ‘C for 1 h a t 100,000 X g, the resulting meprin p subunit cDNA was obtained by the 3’-RACE procedure precipitate and supernatant fractions were assayed for meprin and (Life Technology Inc.; Fig.1).cDNA was synthesized from total RNA alkalinephosphatase activity. Meprin activity was assayed using (1pg) from C3H/He mouse kidney using a polyadenylation sequenceazocasein as substrate as described by Reckelhoff et al. (1985). One specific “adapter” primer and was amplified by the PCR using oligounit of activity is defined as an increase in absorbance at 340 nm of 7 (Table I) and a universal amplification primer. A portion from the first round of the PCR was amplified with a nested gene-specific 0.001/min, which is equivalent to the solubilization of 1.1 pgof azocasein/min. Alkaline phosphatase activity was assayed at 37 “C primer oligo-8 (Table I), and theuniversal amplification primer. The with 1 mM p-nitrophenylphosphate in 1 M Tris-HC1 buffer, pH 9.0, products from the second round of thePCR were cloned into pAmpTMlas described for the 5’-RACE product. containing 10 mM MgC12; the absorbance change a t 410 nmwas Sequencing Strategy-Plasmids containing the @ subunit cDNA monitored. Cloning Strategy-A A p t 10 CD-1 mouse kidney cDNA library was were sequenced using the Sanger dideoxy termination method with obtained from Clontech, and a XZAPI1 C57BL/6 androgen-induced SequenaseTM(U. S. Biochemical Corp.). The primers T3, T7, SK, mouse kidney cDNA library was a gift from Dr. Franklin G. Berger KS, and SP6 (Stratagene) and two gene-specific primers were used of the Department of Biology, University of South Carolina at Colum- as sequencing primers. Subclones were constructed by self-ligation of the original plasmids which had been digested with either AccI, KpnI, bia. Meprin @ subunit cDNA fragments encoding residues 65-345 were amplified from the Berger and Clontech cDNA libraries using PstI, XbaI, StuI, and SnuBI, or StuI, and SmaI. A modification of the Gene Amp kit with AmpliTaqTMDNA polymerase (Perkin Elmer- the double-stranded DNA sequencing method (Chen and Seeburg, Cetus; Fig. 1and Table I). Typically, the amplification reactions were 1985) was followed as described by Jiang et al. (1992). Both strands

The p Subunit of Mouse Meprin

21037

TABLEI Oligonucleotidesequences of the mouse meprin j3 subunit The oligonucleotides used for the PCR amplification of the protease domain (oligo-1, -2, and -3) were designed based on known peptide sequences. The peptide sequence for oligo-1 was part of the NH2-terminal sequence of a papain-purified preparation of meprin j3 activated with trypsin (Jiang et al., 1992). The peptide sequence for oligo-2 was part of peptide j31 in Jiang et al. (1992) obtained by NH2-terminal sequence analysis of a cyanogen bromide peptide fragment from meprin A. The peptide sequence for oligo-3 was part of peptides & and B, in Kounnas etal. (1991), or peptide j32 in Jiang etal. (1992) derived by cyanogen bromide fragmentation of purified meprin A and B. The rat meprin j3 subunit cDNA sequences were used to design the oligonucleotides except in three instances (underlined) in which the amino acids differed (Johnson and Hersh, 1992). Degenerate codons were used for the different amino acids. The oligonucleotides used in the 5’- and 3’RACE procedures (oligo-4, -5, -6, -7, and -8)were designed based on the determined cDNA seauence of the B subunit. Oligonucleotides

Sequences

Nucleotide no. in cDNA

Oligo-1 (sense) Oligo-2 (sense) Oligo-3 (antisense) Oligo-4 (antisense) Oligo-5 (antisense) Oligo-6 (antisense) Oligo-7 (sense) Oligo-8 (sense)

CCCGAATTCATCATTGGAGAC(C,A)ACAA(A,G)AGA(G,T)(A,G)(T,C,G)CCACATACC GGGGAATTCGCTATCGCCTTAAAACGT(G,A)CATT CCCGAATTCATAAAA(T,C)TTCAAG(C,T)ACTGAAACCC TCCAATGGACAAGGCTCCTGCTTC GCGAGATATCCCAAGGCTTGAAGTCAATGCACG CAUCAUCAUCAUGAATTCCAAGCTGTCCTCTAGAACATATGG GCGCTGCAGAGCATTTTGAATGAGGGGGCCACG CAUCAUCAUCAUGATATCTTGTACCCCAAGAGAGGGTTTCAG

242-256 323-346 1043-1066 436-456 339-361 260-283 983-1006 1028-1051

were sequenced at least twice (Fig. 1). Analysis of Sequencing Data-DNA sequences were analyzed using the program McMolly (Soft Gene, Berlin). The deduced amino acid sequence was searched with PROSITE to find consensus sequences and patternsin the j3 subunit (Bairoch, 1990). The currentdatabases (GB, SP, and PIR) were searched with the j3 subunit cDNA and protein sequences using the BLAST system (Altschul et al., 1990). Northern Hybridization-Total RNA was isolated from tissues of both C57BL/6 and C3H/He mice using a modification of the acid guanidinium thiocyanate-phenol-chloroform extraction method (Chomczynski and Sacchi, 1987). RNA was prepared from kidney, small intestine, spleen, heart, brain, testis, lung, and liver (Jiang et al., 1993). The cDNA probe for the mouse j3 subunit was labeled with [cu-SzP]dCTP(50 pCk NewEngland Nuclear) according to themethod described by Jiang et a1 (1993). More than 60% of the radioisotope was incorporated into labeled fragments, usually corresponding to 5 X 10’ counts/min. The labeled fragment was used directly, without further purification, in hybridization experiments. RNA hybridization experiments were performed essentially as described previously (Jiang etal., 1992), including the modifications described by Jiang et al. (1993). Membranes were exposed to x-ray film at -70 “C for fluorography for 24-96 h. Interspecific Backcross Mapping-Interspecific backcross progeny were generated by mating (C57BL/6 X Mus spretus)F1 females and C57BL/6 males as described (Copeland and Jenkins, 1991). A total of205N1 micewere used to map the Mep-lj3 locus (see text for details). DNA isolation, restriction enzyme digestion, agarose gel electrophoresis, Southern blot transfer, and hybridization were performed essentially as described (Jenkins et al., 1982). All blots were prepared with Zetabind nylon membrane (AMF-Cuno). The j3 subunit probe, a 744-base pair clone, was labeled with [a-3zP]dCTPusing a nick translation labeling kit (Boehringer Mannheim); washing was at a final stringency of 0.8 X SSCP, 0.1% SDS, 65 “C. A fragment of 11.0-kilobases (kb) was detected in SphI-digestedC57BL/6 DNA and a fragment of 7.8 kb was detected in SphI-digested M. spretus DNA. The presence or absence of the 7.8-kbM. spretus-specific SphI fragment was followedin backcross mice. A description of the probes and restriction fragment length polymorphisms for the loci linked to Mep-lj3 including N cadherin (Ncad), prealbumin (Palb), adenomatosis polyposis coli (Apc), and fibroblast growth factor, acidic (Fgfa) has been reported previously (Miyatani et al., 1992; Justice et al., 1992). Recombination distances were calculated as described (Green, 1981) using the computer program SPRETUS MADNESS. Gene order was determined by minimizing the number of recombination events required to explain the allele distribution patterns. RESULTS

The cDNA sequence of the /3 subunit contains an open reading frame that starts with a Met residue, which could be the initiation site of translation, andcodes for 704 amino acid residues (Fig. 2). The sequence obtained by the 5’-RACE procedure extends beyond the first Met residue and contains

a vertebrateconsensus sequence for ribosome binding (Kozak, 1991). The first 23 amino acids contain a hydrophobic sequence that could represent a signal sequence, as is present in most members of the “astacin family” of metalloendopeptidases (Jiang et aL, 1992). The protease domain of the /3 subunitcan be assigned by alignment with otherastacin family members to start at Asn-63 and end at Leu-260; this domain is composed of 198 amino acid residues. It contains the putative zinc-binding pentapeptide motif, HEXXH in the %amino acid consensus sequence that is characteristic of metalloendopeptidases of the astacin family (Fig. 2; Dumermuth et al., 1991). The subunit has a stretch of amino acids from Ser-261 to Ala-610 that has no identifiable functional domain. There are six potential NHZ-glycosylation sites in this section, and three additional sites the in protease domain. The subunit also contains an EGF-like domain (Cys-611 to Cys-646), followed by a hydrophobic sequence (Thr-655 to Val-678), a short hydrophilic sequence (Arg-682 to Arg-690), and 14 additional COOH-terminal amino acids. The deduced amino acid sequence of the @ subunit contains several potential phosphorylation sites, two of which are in the COOH-terminal cytoplasmic domain. One is a potential protein kinase C phosphorylation site at Thr-681 (inconsensus sequence TRR); the other is Thr-693 (in the consensus sequence R X X T ) , a potential site for calmodulin kinase I1 phosphorylation (Kennelly and Krebs, 1991). Alignment of the mouse /3 subunit with the a subunit reveals an overall identity in the amino acid sequences of 42% (Fig. 3). There is, however, very little homology in the NH2terminus (first 37 amino acids of p), or in the COOH-terminal sequences after the EGF-like domain (after Glu-647 of p), although both subunits contain hydrophobic sequences in this region. In addition, asegment spanning from Arg-628 to Tyr683 in the a subunit is missing in the B subunit. A hydropathy analysis of the mouse /3 subunit identified a NHderminal sequence (Trp-8 to Ala-23) and a COOH-terminal sequence (Thr-655 to Val-678) as themost hydrophobic regions of the subunit (Fig. 4). For rat p the NH2-terminal sequence was proposed as thetransmembrane domain because no hydrophobic sequence was found in the COOH terminus (Johnson andHersh, 1992). To determine whether the hydrophobic NHZ-terminal sequence is present on the mature mouse ,f3 subunit, detergent-purified meprin was prepared from C3H/He, ICR, and C57BL/6 mice and the NH, termini of the subunits were sequenced (Fig. 5 ) . The detergent-purified (3 subunit from all three mouse strains was found to start

The p Subunit of Mouse Meprin

21038 * 1

A

G

C

T

T

G

C

A

G

C

T

T

T

C

A

T

C

P

G

G

C

C

A

C

A

G

T

A

T

U

;

A

T

G

C

C

M D A R H Q P W F L V F A T F L L A S G L P a P E

1

n 106 26

A A G T T T G T C A A A G A C A T A G A T G G A G G A A T T G A C C A A G A C A T

K

F

V

K

D

I

D

G

G

I

D

Q

D

I

F

D

I

N

Q

G

L

G

L

D

L

F

E

G

D

I

K

L

E

A

N

211 61 316 96 421 131 52 166

T6C A C G T G C T G A C C G G G A T G T A C G T C A T T A T A G T T T ~ C A ~ T T C A G C C A ~ G C A C A A C T T ~ C A T C T A ~ T ~ C A G T G T ~ ~ T T C C C T G

S

R

A

D

R

D

D

Y

V

I

I

V

W

D

R

I

Q

P

G

K

E

H

N

F

N

I

Y

N

D

S

V

S

D

S

L

A A A A A A A

631 201

A A T G T C C C A T A T G A C T A C A C C T C A G T A A T G C A C T A C A G T A C G T G

N

V

P

Y

D

Y

T

S

V

M

H

Y

S

K

T

A

F

Q

N

G

T

E

S

T

I

V

T

R

I

P

R

F

E

D

V

A A A A A A A

736 236

A T T G G C C A A C G A A T G G A C T T C A G T G A C T A T G A C C T T C T G A T G

I

G

Q

R

M

D

F

S

D

Y

D

L

L

K

L

N

Q

L

Y

N

C

T

S

S

g

S

F

M

D

S

C

D

F

E

L

A A A A A A A

841 271 94 6 306

G A A A A T A T C T G T G G C A T G A T C C A A A G T T C G G G G G A T A G T G C A G

E

N

I

C

G

M

I

Q

S

S

G

D

S

A

D

W

Q

R

V

S

Q

V

L

S

G

P

E

S

D

H

S

K

M

G

Q

T G C A A A G A C T C T G G C T T C T T C A T G C A T T T ~ ~ C T A G C A T T C A G

C

K

D

S

G

F

F

M

H

F

N

T

S

I

L

N

E

G

A

T

A

M

L

L

S

R

L

L

Y

P

K

R

G

F

Q

A A A A A A A

1051 341

T G C T T G G A G T T T T A T C T G T A C A A C A G T G G A A G T G G A A A T G G

G

L

E

F

X

L

Y

1156 366

A

G

A

C

A

A

A

R

Q

I

K

E

1261 401

A C G T C A T C A G G T G G T C T G T C T A T C G A T G A C A T C A A T C T C T C A G

T

S

S

G

G

V

L

P

S

N T

S A

I

I

G ~

G

D

S ~

S

D

G ~

W

I

N ~

Q

N

D

Q

L

N

I

Y

T

R

E

Y

T

T

G

Q

Q

G

G

C

A

?

T

G

G

G

A

G

A

G

A

G

G

T

A

C

C

L

H

Y

V

T

L

Q

V

T

L

S

E

T

R

C

P

H

H

K

I

K

W

F

H

R

I

V

Q

V

N

F

F

E

T

V

L

C

G

Q

T

T

L

I

L

G

R

L

Q

G

G

G

C

P

G

A

A

C

T

T

G

Q

AAAAAAA

A A A A A A A

1366 446

G A C A U \ T C T G T A T A C A G T C C T C C A T T T T A T T C T T C T A A A G C T G

1471 481

ATCTCCGGTGCCAATGATGATCAATTACAGTGGCCATGTC~~CAA~ACAAT~CACT~GGAT~TCCT~CATC~CA~AT~

1576 516

C A G C G G A G T A T A A C C A C A C A G A C C C A A C G A T G A C C A G T G A T A Q R S I T T D P T M T S D N G S Y F W D R P S K V G V T D V F P N G T

D

I

T

S

S

G

V

A

Y

N

S

D

P

D

P

Q

F

L

Y

Q

S

W

S

P

K

C

G

P

Y

W

A

Q

F

Q

Q

A

I

T

Y

M

M

T

D

L

L

L

R

D

Y

Q

S

N

AAAAAAA

1681 551

T

P

N

D

V

I

G

R

I

Q

Y

R

F

M

H

F

L

N

AAAAAAA

CAGTTTAGCAGAGGTATGC~C~TCTTCAT

Q F S R G I G Y G T T V F I T R E R L K S R E F I K G D D I Y I L L T

1786 G T T G A A G A C A T A T C T C A C C P C A A T T C T A C A T C A ~ T G T C C 586 V E D I S H L N S T S A V P D P V P T L A V H N A - C - S J - Y - V - 5 : _ L N - G - E AAAAAAA

1891 621

A

I

T -C-

C

Y

T -V-

G

T

G

Q -D- E

T

T

G

T

C

C

-R. A -E- 5: 3 - C

A

A

A

G -G-

A

T

G

E -D- H

T -W_

~

G

T

~

G

T

X JL si -K- B -C-

G C A A G T G P E K R G S T R D

C

L

1996 656 2101 691 2206

FIG.2. cDNA and deduced aminoacid sequence of the mouse meprin fl subunit. The single-underlined sequences were confirmed by NH2-terminal analyses of peptides prepared from purified preparations of meprin A and B. The boxed region is the 24-amino acid consensus sequence for the astaoin family of metalloendopeptidases. The sequences underlined by carets identify potential NHp-glycosylation sites. The brackets after Lys-62 and Leu-260 indicate the beginning and end, respectively, of the p subunit protease domain. The asterisk identifies the initiation codon for translation. The triangle represents the potential cleavage site for signal peptidase. The dashed underline indicates an epidermal growth factor-like domain. The double-underlined sequences were identified as hydrophobic regions.

The p Subunit of Mouse Meprin beta alpha

MDARHQPWFL~ATFLLASGLPAPEKFVKDIDGGIDQDIFDINQ 44 ** I * I I I I III*II MARRLGRSSSFAIMLWIQPACLLSLIFSAHIAAVSIK~LLNGSDHDTDVGEQKDIFEINL 60 GLGLDLFEGDIKLEANGKNSIIGDHKRWPHTIPWLEDSLEMNAKGVILNAFERYRLKTC * I1 I I I l l I *I II III*I I l I * l I I I II I l l *lII*l

104

alpha

AAGLNLFQGDILL-PRTRNAMRDPSSRWKLPIPYILADNLELNAKGAILHAFEMFRLKSC

119

beta

IDFKPWSGEANYISVFKGSGCWSSVGNIHAGKQELSIGTNCDRIATVQHEFLHALGFWHE *IIII* II II I I I I I I *I I I *Ill II II* l l * l l l l l l * l I VDFKPYEGESSYIIFQKLSGCWSMIGDOQVG_QNISIGEGCDFKATIEHEILHALGFFHE

164

QSRADRDDYVIIVWDRIQPGKEHNFNIYNC-SVSDSLNVPYDYTSVMHYSKTAFQ NGTE I l l I I I I I I I I1 I I I I I I I I ***I I I I I I I I * I I I l*-l * QSRTDRDDYVNIWWDQIITDYEHNFNTYDDNTITD-LNTPYDYESLMHYGPFSFWKNESI

222

beta

alpha bet a alpha beta alpha

FIG. 3. Alignment of the amino acid sequences of the mouse meprin a and @ subunit. The upper sequence is the mouse meprin @ subunit, and the lower sequence is for the a subunit. Identical residues are indicatedby lines; similar residues are indicated by asterisks. For optimal alignment, several gaps were inserted into the sequences. Alignment was aided by the use of the program McMolly (Soft Gene, Berlin).

beta alpha bet a alpha beta alpha beta alpha bet a alpha beta alpha bet a alpha

178

237

282 STIVTRIPRFEDVIGQRMDFSDYDLLKLNQLYNCTSSLSFMDSCDFELENICGMIQSSGD I1 I*II I * I l l I l l II**II *Ill1 ***I I I I I I * I I I I I * I 97 PTITTKIPEFNTIIGQLPDFSAIDLIRLNRMYNCTATHTLLDHCDFEKTNVCGMIQGT~ 2

SADWQRVSQVLSGPESDHSKMGQCKDSGFFMHFNTSILNEGATAMLESRLLYF'KRGFQCL 342 lI* * I l l I*Il IIII* I I*IIII*IIIII Ill III 356 DADWAHGDSSQPE-QVDHTLVEQCKGAGYFMFFNTSLGARGEAALLESRILYPKRKQQCL EFYLYNSGSGNDQLNIYTREYTTGQQGGVLTLQRQIKEVPIGSWQLHYVTLQVTKKFRW I* *Ill I * * * I I I * Ill* IIII I QFFYKMTGSPADRFEWWRRDDNAGKVRQLAKIQTFQGDSDHNWXIAHVTLNEEKKFRYV

402 416

458 FEGLRG-PGTSSGGLSIDDINLSETRCPHHIWHIQNFTQILG-GQDTSVYSPPFYSSK I I *I I I I I I I * * I l l I * I l I I *I I I * * l I l I I I I II I FLGTKGDPGNSSGGIYLDDITLTETPCPAGVWTIRNISQILENTVKGDKLV-SPRFYNSE 476 513 GYAFQI-YMDLRYSTN-VGIYFHLISGANDDQLQWPCPWQQA~LLDQNPDIRQRM lI* I I * *I *I I l l I I II I I1 II I I * I I l I I*II G Y G V G V T L Y P N G R I T S N S G L L G L T F H L Y S G D N D A I L E ~ ~ N R Q A I M T I L D Q E A D T535 ~~ FNQRSITT DPTMTSDNGSYFWDRPSKVGVTDVFPNGTQFSRGIGYGTTVFITRERLKS I I *I I* II *III * Ill *IIIIIIIIl I SLTLMFTTSKNQTSSAINGSVIWDRPSKVGVYDKDCDC_F_RSLDWGWGQAISHQLLKR

571

REFIKGDDIYILLTVEDISHLNSTSAVPDPVPTLA I I * I I I * I** I**III I I* RNFLKGDSLIIFVDFKDLTHLNRTEVPASARSTMPRGLLLQGQESPALGESSRKAMLEES

616

592

652

VHNACSEWCQNGGICWQDGRAECKCPA 635 I Ill I l l I I I*I LPSSLGQRXPSRQKRSVENTGPMEDHNWPQYFRDPCDPNPCQNEGTCVNVKGMASCRCVS 712

beta

GEDWWYMGKRCEKRGSTRDTVIIAVSSTVTWAVMLIITLVSVYCTRRKYRKKARANTAA *** * * * I * * * * I * I * I I I I **I I I1

695

alpha

GHAFFYAGERCQAM-HVHGSLLGLL_IGCIAGLIFLTFVTGKLRQ

7

beta

MTLENQHAF

704

phob

3.0

1.l

0

-1,s

-3.0 phi 1 141

21039

422

s63

Amlno acid residue FIG. 4. Hydropathy plot of the deduced amino acid sequence of the @ subunit of meprin. Hydrophobic regions appear above the line and hydrophilic regions below the line. The plot was generated using a window size of 11 amino acids according to the method of Kyte and Doolittle (1982). Two segments were identified as hydrophobic regions; one is at the NH2 terminus (residues Trp-8 to Ala23), the other is close to the COOH terminus (Thr-655 to Val-678).

at Leu-21 (in the sequence LPAPEK.. .). Furthermore, the bond between Gly-20 and Leu-21 is a predicted cleavage site for the signal peptidase (von Heijne, 1983). Thus, the first20 amino acid residues (Met-1 to Gly-20) likely correspond to a transientNHz-terminal signal peptide, andthis sequence cannot represent the transmembrane domain in the mouse. To determine whether the /3 subunit is attached to kidney brush border membranes via a phosphatidylinositol-glycan anchor, membranes were prepared from C3H/He kidneys and treated with phosphatidylinositol-specific phospholipase C (Table 11).The results indicated that meprin was not released from the membranes with the phospholipase treatment whereas alkaline phosphatase,which is attached by the phosphatidylinositol-glycan anchor, was released. Thus, there is no evidence for a phosphatidylinositol linkage for the @ subunit. These data, along with the evidence above indicating that the NHz-terminalhydrophobic domain is removed from the mature /3 subunit, supportthe contention that theCOOHterminal hydrophobic domain is the membrane anchor for the mouse @ subunit. The mouse a subunit has Asn-78 at the NHz terminus in the sequence NAMRDP. . ., which is the beginning of the protease domain, whether it is released from membranes with

60

21040

The /3 Subunit of Mouse Meprin NAMRDPSSR

4

Mouse a LPAPEKFVKDI

Mouse p NSIIGDHKRDPH

FIG.5. Domain structures of mouse and rat meprin subunits and NHz-terminal sequences of mouse subunits. The domains are aligned relative to the protease domain (m). Both mouse subunits have a signal sequence (a) a t the NH, terminus, followed by a potential prosequence ( D . Close to the COOH terminus there is an EGF-like domain (N), followed by the putative transmembrane spanning domain (El), and a short cytoplasmic domain (0). Between the protease domain and theEGF-like domain is a large domain (0)of unknown function. This domain is conserved in both mouse subunits; however, it is 56 residues shorter in the /3 subunit than in the a subunit. The signal sequences, transmembrane, and cytoplasmic domains of the two mouse subunits share no significant sequence similarity. The rat /3 subunit has a domain structure similar to the mouse /3 subunit; however, no EGF-like and hydrophobic domains are present at the COOH terminus (Johnson and Hersh, 1992). The NH, termini of the mature mouse subunits, indicated by the arrows, were determined from detergentsolubilized subunits as described under “Experimental Procedures.” The NH, terminus for the a subunit was the same (NAMRDPSSR..) whether the enzyme was solubilized by papain, trypsin, or detergent. The /3 subunit NH, terminus for papain and detergent-solubilized preparations was LPAPEKFVKDI.. (see arrow). The cleavage site for trypsin in /3 subunits purified from C3H/He mouse kidney with papain (Kounnas et al., 1991) is indicated by the triunggb. TABLE I1 Effect of phospholipase C treatment on solubilization of meprin B activity Brush border membrane-enriched fractions were prepared from C3H/He mouse kidney. The membrane fractions were incubated at 37 “C for 90 min in the presence or absence of phosphatidylinositol-specific phospholipase C (PLC), centrifuged a t 4 “C for 1h at 100,000 X g, and the activities measured in the resulting precipitated and supernatant fractions. Units for meprin are expressed as increase in absorbance at 340 nm of 0.001/min; for alkaline phosphatase, UIM 4-nitro~henvl~hosphate/min. Meprin activity

PLC units added

0.73 0.06

0 2

In precipitated fraction

supernatant In fraction

Alkaline phosphatase activity % activity in

precipitate

6.33 6.00

papain or detergent (Fig. 5). Thus, both the pre- and prosequences of the CY subunit are removed from the mature subunit. By contrast, the mouse @ subunit NH2 terminus begins with the prosequence (LPAPEK.. .) in detergent-purified preparations. The @ prosequence is not removed by papain treatment butis removed bytrypsin treatment in uitro. Trypsin treatmentof mouse p results inthe NH2terminus of Asn63 (NSIIGD.. .), which is the beginning of the protease domain (Jiang etal., 1992; Figs.2 and 5). Trypsin also activates meprin B againstazocasein, indicating that removal of the 42 amino acid segment (Leu-21 to Lys-62) is critical for the proteolytic activity of this subunit against large substrates. Comparison of the domain structures of CY and p demonstrate that the signal sequence and the large functionally unidentified sequence after the protease domain in the @ subunitare smaller than in the CY subunit. The COOHterminal segment, after the hydrophobic sequence, is larger in the fi subunit. The reported rat P subunit domains are also compared to themouse domains in Fig. 5. The mouse and rat P subunits are 89% identical up to the EGF-like domain of the mouse subunit. When the cloned cDNA of the mouse @ subunit was used to probe total RNA isolated from several tissues of C57BL/6 and C3H/He mice, a message of approximately 2.5 kb was found in kidney and small intestine (Fig. 6). These mouse strains were chosen because they differ in their expression of the kidney CY subunit and, therefore, in meprin A activity. Similar amounts of the /3 subunit message were detected in the kidney of both mouse strains. The intestine of both strains also had the 2.5-kb message and smaller messages as well,

In precipitated fraction

94 978

In supernatant fraction

% activity in

precipitate

26.5 8

which were probably degradation products. The CY subunit is not expressed in mouse intestine, thus this is anotherdistinguishing feature of the p subunit. Nomessage for @ was detected in spleen (Fig. 6), or other tissues including brain, heart, testis,lung, and liver (not shown). As a control to verify the quality of the RNA preparations, the same blots were probed with the human P-actin cDNA after removal of the subunit probe (not shown). The tissues of both mouse strains contained similar amounts of actin message. Overall, the results indicatethat thelack of detection of p subunit mRNA in tissues other than kidney and intestine was not due to a significant difference in the quantity or quality of the RNA preparations, but can be attributed to alack of expression of the P subunit gene. The mouse chromosomal location of Mep-l@ was determined by interspecific backcross analysis using progeny derived from matings of [C57BL/6 X Mus spretus)F1X C57BL/ 61 mice. This interspecific backcross mapping panel has been typed for over 1100 loci that are well distributed among all the autosomes as well as the X chromosome (Copeland and Jenkins, 1991). The mapping results indicated that Mep-lP is located in the proximal region of mouse chromosome 18 linked to Ncad, Palb, Apc, and Fgfa (Fig. 7). Although 144 mice were analyzed for every marker and are shown in the segregation analysis, up to191 mice weretyped for some pairs of markers. Each locus was analyzed in pairwise combinations for recombination frequencies using the additional data. The ratios of the total number of mice exhibiting recombinant chromosomes to the total number of mice analyzed for each pair of loci and the most likely gene order are: centromere -

The /3 Subunit of Mouse Meprin

21041

mo

0. 0. 0. 0.

I O ”U I0 APC I O F&

mo

mu

3

61 57

i

3

18q12.1 18q12.1

FIG. 6. Northern blot of total RNA from small intestine, kidney, and spleen of C67BL/6 and C3H/He mice. Total RNA 6.7 was isolated as described under “Experimental Procedures.” RNA samples (20 p g ) were subjected to formaldehyde-gel electrophoresis in the presence of ethidium bromide (0.5 pg/ml), transferred to a w1QLz nylon membrane, and hybridized to a ”P-labeled mouse meprin B 4.2 APc subunit probe encoding Arg-98 to Tyr-345 (seeFig. 2). The membrane Fgfa Sq31.3q33.: was washed first a t 25 “C for 1 h, then at 60 “C for 45 min, and exposed to x-ray film a t -70 “C for fluorography. Lanes I , 3, and 5 contain RNA preparations from C57BL/6 mice. Lanes 2, 4, and 6 contain RNA preparations from C3H/He mice. Lanes I and 2 are RNAsamples from smallintestine. Lanes 3 and 4 containRNA FIG. 7. Chromosomal localization of the mouse meprin B preparations from kidney. Lanes 5 and 6 contain spleen RNA prepsubunit structural gene. Mep-1P maps in the proximal region of arations. Mouse strains were: C57BL/6 (C57) and C3H/He (C3H). mouse chromosome 18 by interspecific backcross analysis. The segThe ribosomal RNA markers (18 S and 28 S from mouse tissues) regation patterns of Mep-IB and flanking genes in 144 hackcross were determined on ethidium bromide-stainedgels. animals thatwere typed for all loci are shownat the topof the figure. Each column represents the chromosome identified in the backcross Ncad 3:161- Palb 1:151- Mep-18 12:178 - Apc 8:191progeny that wasinheritedfromthe(CS7RL/6 X M. s p r e t w ) F, Fgfa. The recombinationfrequencies(expressed as genetic parent. The block boxes represent the presence of a C.i7RL/6 allele, distances in centiMorgans (cM) & the standard error) are and white boxes represent the presence of a M. spretus allele. The Ncad - 1.9 & 1.1 - Palb - 0.7 & 0.7 - Mep-1/3 - 6.7 f 1.9 number of offspring inheriting each type of chromosome is listed a t Apc 4.2 f 1.5 Fgfa. The placement of Mep-18 on mouse the bottom of each column. A partial chromosome 18 linkage map showing the location of Mep-IB in relation to linked genes is shown chromosome 18 indicates that the gene is unlinked to the at the bottom of the figure. Recombination distances between l o c i in structural gene for thea subunit, which has been assigned to centimorgans to the felt of thechromosome,andthe areshown mouse chromosome 17 (Jiang etal.,1993). positions of loci in human chromosomes, where known, are shown to the right. References for the human map positions of l o c i mapped in DISCUSSION this study can be obtained from GDR (Genome Data Rase). a comWith the cloning and deduced amino acid sequence of the puterized database of human linkage information maintained hy The mouse 8 subunit, a modelfor isoforms of meprin A and meprin William H. Welch Medical Library of The Johns Hopkins University (Baltimore, MD).

I

1

-

-

-

-

-

B can be constructed (Fig. 8). Meprin A is defined as any form of meprin that contains the a subunit, and studies of kidney brush border membranes demonstrated that homo- to remove the prosequence and fold the new NH2-terminal and hetero-oligomeric forms of meprin A exist in uiuo (Gorbea residue into the active site in order to activate the enzyme known to have et al., 1991). Meprin B is a homo-oligomer of /3 subunits. against large substrates. Both subunits are Furthermore, there is some evidence indicating that the dimer EGF-like domains near the membrane, and these domains, as well as membrane-spanning domains andS-S bridges, might is the fundamental unit of rat, human, and mouse meprins, and for simplicity, dimeric forms of meprins are presented in beimportantforinteractions between thesubunits.Both Fig. 8. The model shows mature, membrane-bound forms of subunits are glycosylated (Kounnas et al., 1991). The reson is mousekidney meprins with the signal sequences removed, between the protease domain and the EGF-like domain and the COOH-terminal hydrophobic domains as the memshorter in 8 compared to a due to a span of 56 amino acids /3 (see Fig. 3). There are several possible reasons brane-spanning regions. Both a and (? subunits contain posi- not present in tively charged amino acids on the cytoplasmic side of the forthisfinding,includingthepossibilitythatan exon is membrane, which may function as stop-transfer sequences to missing in the 8 gene, or that there are differences in RNA anchor themolecule at thecell surface (Mauxion etal., 1989). splicing. The model represents a working hypothesis thatcan The NH, termini of a and /3 are shown todiffer: a,with the be tested in the future. prosequence removed, isin a conformation that isactive The /3 subunit contains many of the properties of other against proteins, while /3 has retained the prosequence. The members of the astacinfamily of metalloendopeptidases. The “family” is an old one in that it is predicted that the protease active conformation of the protease domain is proposed on the basis of the crystal structureof astacin, the prototypeof domain existedbefore vertebrates and invertebratesdiverged. the astacin family enzymes (Bode etal., 1992). It is proposed Thus far,15 members of this family have been identified from of the family that j3 requires a proteolysis event at the membrane (arrow) the DNA andproteindatabases.Members

21042

The p Subunit of Mouse Meprin

FIG. 8. Models for dimeric formsof mouse meprins. The 012, PZ,and a0 isoforms of mouse kidney meprins are depicted from left to right. The bulk of the subunits are in theextracellular space. The subunits are anchored on the plasma membrane by single transmembrane helices near the COOH termini of the subunits. Positively charged residues are indicated onthe cytoplasmic tails of the subunits. The NH2terminal portions of the subunits contain the protease domains with zinc indicated a t the active sites. An active conformation for the a subunit isindicated, on the basis of the astacin crystal structure;the NH, terminus interacts through a salt bridge with a Glu residue adjacent to the thirdcoordinating His (Bode et al., 1992). A conformation that is inactive toward proteins is indicated for the /3 subunit by the lack of interaction of the NH2-terminal peptide with the Glu residue near the active site. The arrows indicate the cleavage site for the prosequence of the P subunit. The disulfide bonds in theprotease domain and in the EGF-like domain (near themembrane in the extracellular space) are shown based on the determined positions in astacin and murine EGF (Jiang et al., 1992). The branched structures ( Y shaped) are potential NH2-glycosylation sites; the sites on the a subunits have been determined experimentally (Jiang et al., 1992). Small lines perpendicular to the backbone of the subunits indicate cysteine residues, some of which are proposed to form intra- or intersubunit disulfide bonds.

include the crayfish astacin, a nematode protein (cmOlal2), a frog protein, three fish enzymes, two sea urchin proteases, tolloid (a Drosophila developmental protein),human bone morphogenetic protein-1, and the subunits of mouse, rat, and human meprins (Jiangand Bond, 1992; Yasumasu et al., 1992). These metalloendoproteinases appear to be expressed in developing as well as matureorganisms and to have diverse functions. The meprins from mouse, rat, and human brush border membranes are the only members of the family that contain COOH-terminal hydrophobic sequences that appear to be responsible for anchoring the protein to membranes. All the other members of the family described thus far are secreted from cells. Of the known astacin family members, the mouse /3 subunit is most closely related to the rat/3 subunit. The reported rat /3 sequence differs significantly from the mouse sequence in the COOH terminus (Johnson and Hersh, 1992). This could be due to a species difference. However, it is more likely that there was an error in the sequence determination of the rat /3 cDNA. There is an 89% identity over the entirelength of the mouse and rat cDNA. The deduced amino acid sequences are also 89% identical before the mouse /3 subunit EGF-like domain. The differences in the COOH-terminal deduced protein sequences are caused by a reading frame shift rather than nucleotide changes. By inserting a single base into the published rat /3 cDNA sequence, the newly deduced amino acid sequence continues to exhibit the same degree of identity to the mouse sequence (89%), andincludes an EGF-like domain, a hydrophobic stretch, and a hydrophilic sequence. Furthermore, recent data indicate that the human form of the p subunit also contains an EGF-like and COOH-terminal hydrophobic domain.’ Although the mouse a and fi subunits probably derived from a common ancestral gene that duplicated and evolved, the

genes for the subunits are clearly on different chromosomes and are regulated differentially. The a subunit in mice seems to be highly regulated in that it has only been identified in kidney and is notfound in thekidney of some mouse strains. The expression of the /3 subunit is also tissue-specific, but is expressed in the kidney and small intestine of all mouse strains. Meprin B has only been identified thus far in mice (mouse intestine and C3H/Hekidney). Both (Y and /3 subunits appear to be expressed in intestine of rats and humans, and no tissues in these species have thus far been found to express /3 and not a. We have compared our interspecific map of chromosome 18 with a composite mouse linkage map that reports the map location of many uncloned mouse mutations (compiled by M. T. Davisson, T. H. Roderick, A. L. Hillyard, and D. P. Doolittle and provided from GBASE, a computerized database maintained at The Jackson Laboratory, Bar Harbor, ME). Mep-la mapped in a region of the composite map that lacks mouse mutations with a phenotype that might be expected for an alteration in the Mep-l/3 locus (data not shown). The proximal region of mouse chromosome 18 shares a region of homology with human chromosomes lop, 18q, and 5q (summarized in Fig. 7). In particular, Palb has been placed on human 18q12.1. The tight linkage between Palb and Mep-10 in mouse suggests that Mep-l/3 will reside on 18q, as well. The discovery of multiple forms of meprins and thetissuespecific expression of the subunits hasprovided an intriguing model to study regulation of cell surface proteinases. The inherited “deficiency of meprin activity” was originally discovered in inbred mouse strains derived from the Strong C stock such as C3H/He (Beynon and Bond, 1983). No abnormal phenotype is associated with the activity “deficiency,” e.g. no defects in kidney function or longevity were noted in the C stock mice. It is now clear that thedeficiency ofmeprin

The /3 Subunit of Mouse Meprin activity is due to a lack of expression of one subunit of meprin, the a subunit, which is responsible for the high azocaseinase activity. The /3 subunit, by contrast, is present in kidney of all mouse strains tested, and in intestine aswell, and thereis no deficiency of one isoform of meprin, meprin B. It is therefore possible that the a subunit has a very specialized function that can be compensated for and that the/3 subunit performs essential activities at themembrane. Meprin B can hydrolyze some small substrates prior to activation and is known to have latent azocaseinase activity (Kounnas et al., 1991).The present work establishes that latency toward proteins is associated with the prosequence that is present onthe mature form of the subunit in the membrane. When the prosequence is removed by trypsin-like enzymes, meprin B exhibits comparable azocaseinase activity to that seen with meprin A. This latent proteinaseactivity allows for a possible activation mechanism at thebrush border membrane surface. In fact, preliminary studies indicate that meprin B is not latent in the mouse intestine, but fully active toward azocasein, perhaps due to the high concentration of trypsin in the gastrointestinal tract.3 Thus,the forms of meprins might well be altered by the environment at the membrane, and this provides another level of regulation in addition to that at the level of transcription and processing. Acknowledgment-We thank M. Barnstead for excellent technical assistance. REFERENCES Altschul, S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J. (1990) J. Mol. Biol. 216,403-410 Bairoch, A. (1990) PROSITE: A Dictionary of Protein Sites and Patterns, 6th

C. M. Gorbea and J. S. Bond, unpublished data.

21043

release, University of Geneva, Geneva Switzerland Beynon, R. J., and Bond, J. S. (1983) Science 219,1351-1353 Beynon, R. J., Shannon, J. D., and Bond, J. S. (1981) Biochem. J. 199, 591598

Bode, W., Gomis-Ruth, F. X., Huber, R., Zwilling, R., and Stocker, W. (1992) Nature 368. 164-166 Butler, P. E., and Bond, J. S. (1988) J. Biol. Chem. 263,13419-13426 Butler, P. E., McKay, M. J., and Bond, J. S. (1987) Biochem. J. 241,229-235 Chen, E. T.. and Seebure. P. H. (1985) DNA 4,165-170 Chomczynski, P., and Sacchi, N. (1987) Anal. Biochem. 162, 156-159 Choudry, Y., and Kenny, A. J. (1991) Biochem. J. 280,57-60 Copeland, N. G., and Jenkins, N. A. (1991) Trends Genet. 7,113-118 Corbeil, D., Gaudoux, F., Wainwright, S., Ingram, J., Kenny, A. J., Boileau, G., and Crine, P. (1992) FEBS Letts. 309,203-208 Dumermuth, E., Sterchi, E. E., Jiang, W., Wolz, R. L., Bond, J. S., Flannery, A. V., and Beynon, R. J. (1991) J. Biol. Chem. 266,21381-21385 Frohman, M. A., Dush, M. K., and Martin, G. R. (1988) Proc. Natl. Acod. Sci. U. S. A. 86,8998-9002 Gorbea, C. M., Flannery, A. V., and Bond, J. S. (1991) Arch. Biochem. Biophys. 290,549-553

Green, E. L. (1981) in Genetics and Probability in Animal Breeding Experiments, pp. 77-113, Oxford University Press, New York Jenkins, N. A,, Copeland, N. G., Taylor, B. A,, and Lee, B. K. (1982) J. Virol. 43,26-36

Jiang, W., and Bond, J. S. (1992) FEBS Letts. 3 2 1 , 110-114 Jiang, W., Gorhea, C . M., Flannery, A. V., Beynon, R. J., Grant, G. A,, and Bond, J. S. (1992) J. Biol. Chem. 267,9185-9193 Jiang, W., Sadler, P. M., Jenkins, N. A., Gilbert, D. J., Copeland, N. G., and Bond, J. S. (1993) J. Biol. Chem. 288.10380-10385 Johnson, G. D., and Hersh, L. B. (1992) J. Biol. Chem. 287,13505-13512 Justice, M. J., Gilbert, D. J., Kinder, K. W., Vogelstein, B., Buchber A M., Ceci, J. D., Matsuda, Y., Chapman, V. M., Patriotis, C., Makris, A.,%sidhlis, P. N. Jenkins N. A., and Co land, N. G. (1992) Genomics 13, 1281-1288 Kennedy, P. J., ;nd Krebs, E. (1991) J. Biol. Chem. 266,15555-15558 Kenny A. J and In am J. (1987) Biochem. J. 246,525-524 Kyte, d., andDoolittg, R:F. (1982) J. Mol. Biol. 167, 105-132 Kounnas, M. Z., Wolz, R. L., Gorbea, C.M., and Bond, J. S. (1991) J. Biol. Chem. 266,17350-17357 Kozak, M. (1991) J. Cell Biol. 116,887-903 Laemmli, U. K., and Favre, M. (1973) J. Mol. Biol. 80,575-599 Mauxion, F., Sobczak, J., and Kress, M. (1989) Immunogenetics 29,397-401 Mivatani. S.. Cooeland. N. G.. Gilbert. D. J.. Jenkins. N. A.. and Takeichi. M. i1992) ProC. Nuti. A&. sci: s.A: 89, &443-844?- ' Reckelhoff, J. F., Bond, J. S., Beynon, R. J., Savarirayan, S., and David, C. S. (1985) Immunogenetics 22,617-623 von Heijne, G. (1983) Eur. J. Biochem. 133, 17-21 Wolz, R. L., Harris, R. B., and Bond, J. S. (1991) Biochemistry 30,8488-8493 Yasumasu, S., Yamada, K., Akasaka, K, Mitsunaga, K., Iuchi, I, Sbimada, H., and Yamagami, K. (1992) Deu. Biol. 153,250-258

b

u.

I~~~

~~

~

~~~~~~~~