THE JOURNAL OF BI~L~G~~L. CHEMISTRY 0 1990 by The American Society for Biochemistry and Molecular Biology, Inc

Vol. 265, No. 9, Issue of March 25, pp. 5008-5013, 1990 Printed in U.S. A.

The cDNA Sequence of Mouse LAMP-2 EVIDENCE FOR TWO CLASSES OF LYSOSOMAL MEMBRANE GLYCOPROTEINS*

(Received for publication, November 17, 1989)

Ying Cha, Steven M. Holland& and J. Thomas August From the Department of Pharmacolofry and Molecular Sciences, The Johns Hopkins University School of Medicine, Baltimore, h&yland 2j205

We describe the isolation and sequencing of a cDNA encoding the mouse lysosomal membrane glycoprotein mLAMP-2 and the sequence differences that distin- guish this molecule from the LAMP-l class of proteins. An oligonucleotide probe corresponding to the NH,- terminal amino acid sequence of purified mLAMP-2 was synthesized by the polymerase chain reaction and used to screen several cDNA libraries. cDNA clones with an insert of 1,700 nucleotides were identified and sequenced. The deduced amino acid sequence of mLAMP-2 comprises a signal sequence of 25 residues and a 390-amino acid polypeptide (Mr 43,017) with the following putative domains: a large intraluminal region (residues l-354) with 17 N-linked glycosyla- tion sites (Asn-X-Ser/Thr), a hydrophobic transmem- brane-spanning region of 24 residues (355-378), and a COOH-terminal cytoplasmic tail of 12 residues (379- 390). When this sequence is compared with those of other lysosomal membrane glycoproteins, it is appar- ent that mouse LAMP-2 and human LAMP-2 form one homology class (LAMP-2) that is separated from the LAMP- 1 class of proteins. The sequence differences in these two classes provide a basis for comparing the structure of the proteins with their biochemical and biological properties.

Glycoproteins localized primarily to the limiting membrane of lysosomes of mouse (Chen et al., 1985a and 1985b), rat (Lewis et al., 1985; Barriocanal et al., 1986), chicken (Lippin- cott-Schwartz and Fambrough, 1986 and 1987), and human cells (Carlsson et al., 1988; Mane et al., 1989) have been identified recently by use of polyclonal and monoclonal anti- bodies. These molecules appear to represent immunodomi- nant components of the lysosomal membrane and have proved to be highly useful immunological markers in studies of lyso- somal biogenesis and function, and in differentiating lyso- somes from other organelles. It is clear that these membrane glycoproteins are delivered to lysosomes by a pathway that is independent of the mannose 6-phosphate-directed targeting of lysosomal hydrolases (Dahms et al., 1989), implying that

*This work was supported by Office of Naval Research Grant N00014-85-K-0703 and by National Institutes of Health Research Training Grants 5 T32 GM 07626 (to Y. C.) and T32 AI 07291 (to S. M. H.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “advertisemetzt” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

The nucleotide sequence(s) reported in thispaper has been submitted to the GenBankTM/EMBL Data Bank with accession number(s) JO528 7.

$ Current address: NIH-NIAID, Bldg. 4, Rm. 328, Bethesda, MD 20892.

other mechanisms are involved in the biogenesis of the mem- brane of this organelle (Griffiths et al., 1988; Geuze et al., 1988; Arterburn et al., 1989). Although the molecules are present predominantly in the lysosomal membrane, they have also been identified as cell surface components bearing un- usual oligosaccharides that are selectively associated with some neoplastic cells (Dennis et al., 1987; Fukuda, 1985). It is also of interest that the surface expression of the human lysosomal membrane glycoproteins is markedly increased on thrombin-activated platelets (Febbraio et al., 1989). The pres- ence of these molecules on the cell surface is also inducible in blood monocytes after exposure to the lysosomotropic reagent methylamine HCl (Mane et al., 1989).

The most thoroughly characterized lysosomal membrane glycoproteins are those with a molecular mass of loo-120 kDa: mouse LAMP-l and LAMP-2 (mLAMP’-1 and mLAMP-2, Chen et al., 1985a, 1986); rat lgp120, lgpll0, and LIMP III (Lewis et al., 1985; Barriocanal et al., 1986); chicken LEPlOO (Lippincott-Schwartz and Fambrough, 1986); and human LAMP-l and LAMP-2 (hLAMP-1 and hLAMP-2, Carlsson et al., 1988; Mane et al., 1989). Each has a core protein of about 40 kDa and is synthesized as a glycosylated precursor molecule of approximately 90 kDa containing 17- 20 N-linked, high mannose oligosaccharides that are subse- quently processed to the complex oligosaccharides of the mature glycoproteins. cDNAs encoding mLAMP-1 (Chen et al., 1988), hLAMP-1 (Viitala et al., 1988; Fukuda et al., 1988), hLAMP-2 (Fukuda et al., 1988), LEPlOO (Fambrough et al., 1988), and lgp120 (Howe et at, 1988) show sequence homol- ogy, indicating that they represent the same or related pro- teins in different species. The deduced amino acid sequences of these molecules contain a cleavage signal peptide of 18-28 amino acids and a mature protein of 380-396 amino acids (M, - 42,000). The predicted structure of the molecules consists of a large intraluminal domain containing 17-20 N-linked glycosylation sites, with a transmembrane region of 24 amino acids near the COOH terminus and a cytoplasmic tail of 11 or 12 residues.

In this study, we have cloned mLAMP-2 cDNA by use of NHn-terminal sequence of the protein and an oligonucleotide probe synthesized by the polymerase chain reaction (PCR) (Lee et al., 1988). The primary structure of the mLAMP-2 protein deduced from the sequence of this cDNA is highly similar to the analogous human molecule hLAMP-2, and these two molecules form a second class of glycoproteins which is distinct from the other characterized lysosomal mem- brane glycoproteins. The sequence differences between these

1 The abbreviations used are: mLAMP, mouse LAMP; hLAMP, human LAMP; PCR, polymerase chain reaction; SDS, sodium dode- cyl sulfate; hp, base pair(s).

5008

by guest on April 19, 2018

http://ww

w.jbc.org/

Dow

nloaded from

cDNA of Mouse LAMP-2

two classes of proteins substantiate their distinct immunolog- ical identities and may have functional relevance.

EXPERIMENTAL PROCEDURES

Purification of r&AMP-L-Membrane proteins were extracted from the livers of Swiss-Webster mice (Rockland Inc.) as described previously (Hughes and August, 1982) with the following modifica- tions. The insoluble fraction of a crude cell homogenate was extracted sequentially with 1 M KI in 20 mM NaP04 (pH 7.8) and 1 M guanidine hydrochloride in 20 mM NaP04 (pH 7.8) before the fraction was treated with detergent to solubilize integral membrane proteins. The detergent-soluble extract was passed sequentially through columns containing bovine serum albumin and normal rat serum coupled to Sepharose and then was chromatographed on a monoclonal antibody (ABL-93) affinity column as described by Chen et al. (1988). The protein was eluted stepwise with 100 mM diethylamine (pH 10.5) and 100 mM diethylamine (pH 11.5), each containing0.5% octyl glucoside. Fractions were monitored by SDS-polyacrylamide gel electrophoresis with silver staining of protein bands. Fractions containing mLAMP- 2 were pooled and concentrated by negative pressure dialysis against 20 mM NaPO* (pH 7.6) with 0.2% octyl glucoside.

Amino Acid Sequence Analysis-The NH*-terminal sequence of purified mLAMP-2 (1 nmol) was determined in duplicate by auto- mated Edman degradation and amino acid analysis in a gas-phase sequenator and phenylthiohydantoin-derivatized analyzer (Applied Biosystems, Inc.) (Hewick et al., 1981). Tryptic peptides of mLAMP- 2 for sequence analysis were isolated as follows.* The purified protein (80 fig) was deglycosylated with trifluoromethanesulfonic acid (Edge et al., 1981), denatured with 8 M urea in 0.4 M ammonium bicarbonate, reduced in 10 mM dithiothreitol, alkylated with 30 mM iodoacetamide, and digested for 20 h at 37 “C with 5% (w/w) trypsin (Boehringer Mannheim) (Stone and Williams, 1988). The peptide fragments were separated by reverse-phase high performance liquid chromatography on a Vydac Cls column (4.6 mm X 25 cm).

Oligonucleotide Synthesis-Oligonucleotides for the PCR were pre- pared in a DNA synthesizer (Applied Biosystems model 380A) and purified on 20% polyacrylamide gel containing 3.5 M urea (Atkinson and Smith, 1984). The oligonucleotides were further purified on a Cis Sep-Pak cartridge (Waters Associates).

cDNA Libraries-The CD-1 mouse liver cDNA library in XgtlO was a gift from David Weng (The Johns Hopkins University School of Medicine). The mouse embryo Balb/c 3T3 and CD-1 mouse mac- rophage cDNA libraries in hgtll were purchased from Clontech Laboratories, Inc.

Polymerase Chain Reaction-cDNA corresponding to the NH*- terminal sequence of mLAMP-2 was synthesized by the PCR (Lee et al., 1988). The reaction conditions were as follows. PCR buffer (50 mM KCl, 10 mM Tris-HCl (pH 8.3), and 0.01% gelatin); sense and antisense primers (1 pM each); 1 pg of template DNA (prepared from cDNA libraries by the method of Davis et al., 1980); 2 mM MgCl2; dCTP, dCTP, dATP, and TTP (200 pM each); and 2.5 units of Taq polymerase (Cetus Corp.) were mixed in a total reaction volume of 100 ~1 and overlaid with mineral oil. The reaction was run for 40 cycles of denaturation (2 min at 94 “C), annealing (2 min at 52 “C), and polymerization (3 min at 72 “C). To confirm the identity of the product, an aliquot of the PCR product was analyzed by electropho- resis on a 4% agarose gel (1% agarose (Bethesda Research Labora- tories) and 3% NuSieve agarose (FMC BioProducts)). The product was visualized by ethidium bromide staining before being capillary- blotted to Zeta-Probe membranes (Bio-Rad) and hybridized to the ““P-labeled internal probe oligonucleotide (Wood et al., 1985). Pre- hybridization and hybridization buffers contained 0.5% SDS to reduce nonspecific binding.

A 32P-labeled PCR product was prepared by end labeling the primers with [-y-32P]ATP (Du Pont-New England Nuclear) using T4 polynucleotide kinase (New England BioLabs) (Maniatis et al., 1982). The specific 32P-labeled mLAMP-2 sequence was purified on a 12% polyacrylamide gel and eluted on a Gel/X extractor (Genex Corp.) using 0.5 M ammonium acetate containing 1 mM EDTA (pH 8.0).

cDNA Cloning and Sequencing-Phage XgtlO containing mouse liver cDNA was plated at 5 x lo4 plaque-forming units/dish and incubated for 6 h at 37 ‘C. The phage DNA was transferred in duplicate to Hybond-N filters (Amersham Corp.), denatured and neutralized, and then hybridized to the s*P-labeled PCR product

‘Arterburn, L. M., Earles, B. J., and August, J. T. (1990) J. Biol. Chem. 265, in press.

116- 97-

68- ": 0

X

IF 43-

25-

FIG. 1. Purification of mLAMP-2. mLAMP-2 purified from mouse liver was analyzed by 10% SDS-polyacrylamide gel electro- phoresis and silver stained. Molecular weight standards were: p- galactosidase, 116,000; phosphorylase B, 97,400; bovine serum albu- min, 68,000; ovalbumin, 43,000; and chymotrypsin, 25,000.

A.

1 10 20 30 NH*-LIVXLTDSKGTXLYAEWEHXFTITYETXNO

Sense primer lnlernal probe Anllrense pnmr

B.

Sense primer (28 bp) 5'GGAATTCC CTA ACA GAT TCA AAA GGA AC" =TC C C C G C

G G G G T T T T

Antisense primer (27 bp) "E TGT TTC ATA TGT TAT TGT AAA3 PstI G C G C G C G

C G A G A A' A

Internal orobe 121 bo1 "CTA TAT GCA GAA TGG GAA ATG3' . . I

TC C C G G G G T T

FIG. 2. NH*-terminal amino acid sequence of mLAMP-2 and structure of the PCR oligonucleotides. A, the NH&erminal sequence of mLAMP-2, with unidentified amino acids (X) at posi- tions 4, 12,20, and 28. The regions of the sequence used to construct the sense and antisense PCR primers and internal probe are under- lined. B, the sense and antisense primers used in PCR and the internal probe used to test the products of PCR are oligonucleotides with degeneracies of 2048,1536, and 256, respectively. A restriction enzyme site was present at the 5’ end of the sense (EcoRI) and antisense (PstI) primers. Two sets of sense primers, one containing four codons for serine (as shown) and another with two codons (AGT, AGC), were used in the initial PCR assays (such as the conditions for Fig. 3). It was later found that sense primers containing AGT/C could be omitted without affecting synthesis of the specific product.

by guest on April 19, 2018

http://ww

w.jbc.org/

Dow

nloaded from

5010 cDNA of Mouse LAMP-2

bp 672 -

603.

310- 271/261-

234-

194-

116-

72-

1 2 3 4 5 6 7 6 9 IO II 12 13 I4 I5 16 17 I 2 3 4 5 6 7 6 9 IO II I2 I3 1415 16 17

3

FIG. 3. PCR product from cDNA libraries. Reactions using 0.5 pM each of the sense primers were performed as described under “Experimental Procedures.” A, aliquots of PCR products (10 ~1 in each lane except 20 ~1 for lane 9) from mouse 3T3 (lanes 3-7), macrophage (lanes 8-12), and liver (lanes 13-17) cDNA libraries were separated in a 4% agarose gel and stained with ethidium bromide. Lane 1 contains the DNA molecular weight standards as shown (Bethesda Research Laboratories). Lane 2 is a PCR control using Xgtll as template DNA. The concentration of MgCl, in each reaction is as follows: 1 mM (lanes 2, 3, 8, and 13), 1.5 mM (lanes 4, 9, and 14), 2 mM (lanes 5, 20, and 15), 2.5 mM (lanes 6, II, and Z6), and 3 mM (lanes 7, 12, and 17). B, autoradiogram of the Southern blot of the same gel as shown in A, obtained by hybridization with 256 rig/ml ?labeled internal probe, as described under “Experimental Procedures.”

(Wood et al., 1985). Alternatively, the phage DNA was hybridized to a 700-base uair (ba) cDNA arobe labeled with la-3”PldCTP bv ran- dom primed extension (Feinberg and Vogelstein,‘l983)‘using standard conditions (Maniatis et al., 1982).

Putative positive clones were selected and plaque purified, and the phage DNA was extracted and digested with EcoRI (Maniatis et al., 1982). The cDNA insert was purified by agarose gel electrophoresis and then extracted from the agarose by Geneclean (Bio 101, Inc.) before it was inserted into the M13mp8 phage vector. Single-stranded DNA was prepared for sequencing as described in the Bethesda Research Laboratories manual for Ml3 cloning and sequencing. The nucleotide sequence of both DNA strands was determined by the dideoxy chain termination method (Sanger et al., 1977) using the modified T7 DNA polymerase (Sequenase, United States Biochemical Corp.). Sequencing was initiated with the Ml3 universal primer, and overlapping sequences were obtained with sequentially constructed oligonucleotide primers (17 bp) derived from the 3’ end of obtained sequences.

Protein Sequence Analysis-The hydrophobicity analysis and ho- mology search for the deduced protein sequence were conducted using GENmenu (Devereux et al., 1984; Gribskov et al., 1986).

Other Methods-Polyacrylamide gel electrophoresis (Laemmli, 1970), ‘2SI-labeling of purified protein by chloramine-T procedure (Hunter, 1967), and silver staining of polyacrylamide gels (Wray et al., 1981) were performed as described.

RESULTS AND DISCUSSION

PCR Analysis of cDNA Libraries Utilizing the NH?-terminal Amino Acid Sequence of mLAMP-2-The strategy used to isolate mLAMP-2 cDNA was to obtain the NH*-terminal amino acid sequence of purified mLAMP-2 and to use the polymerase chain reaction to select the corresponding mLAMP-2 cDNA.

PCR was carried out at MgCl* concentrations of l-3 mM with cDNA libraries prepared from mRNA of mouse embryo Balb/c 3T3, CD-l mouse macrophage, and CD-l mouse liver cells. The reaction with the liver cDNA library generated the expected 83-bp oligonucleotide (Fig. 3A). The specificity of this product was confirmed by hybridization to the “‘P-labeled internal probe (Fig. 3B). Other PCR products visible after ethidium bromide staining did not hybridize to the internal probe. Much less of the 83-bp product was present in reactions with the mouse embryo 3T3 and macrophage cDNA libraries. The specific product from liver library was not detected at 1 mM MgCl*, nor was it present under any tested conditions when the primers contained only the preferred codons (data not shown).

mLAMP-2 was purified by monoclonal antibody affinity chromatography from an extract of mouse livers. The purified mLAMP-2 contained a single protein of >90% purity as assessed by SDS-polyacrylamide gel electrophoresis and silver staining (Fig. 1) or by autoradiography of the ‘251-labeled purified protein (data not shown).

cDNA Cloning and Sequence Analysis-The 83-bp PCR product labeled with 32P was used as a probe to isolate the appropriate cDNA clones from the same mouse liver cDNA library in XgtlO. Three positive clones were detected and plaque purified in the initial screening. Each contained a 700- bp cDNA insert, and one was subcloned and sequenced. The

The NHP-terminal amino acid sequence of the first 30 residues of the purified protein was determined by automatic

“In the interest of minimizing primer degeneracy, the third nu- cleotide of the codon for threonine was not included in the sense

Edman degradation (Fig. 24). Unidentified residues at posi- primer; thus, the actual distance between primers was 28 bp.

tions 4, 12, 20, and 28 were later identified as asparagine, cysteine, asparagine, and threonine, respectively, from the nucleotide sequence of the cloned cDNA (see below).

Oligonucleotides used as sense and antisense primers for the PCR were designed to correspond to regions within the NHP-terminal amino acid sequence with least coding degen- eracy. Each primer contained a restriction endonuclease site at the 5’ end and all the possible codons for each amino acid (Fig. 2B). An oligonucleotide used as an internal probe to test the reaction product was also synthesized from a region of protein sequence between those used for the primers. On the basis of the g-amino acid distance between the sequence used for primers and the combined size of the primers (55 bp), the size of an appropriate PCR product was calculated to be 83 bp.”

by guest on April 19, 2018

http://ww

w.jbc.org/

Dow

nloaded from

FIG. 4. The nucleotide sequence and deduced amino acid sequence of mLAMP-2. The nucleotide and amino acid residues of clone RL3 are numbered at the left of each line. The amino acid residues of the predicted signal peptide are labeled -1 to -25. The leucine in the square box is the first residue from the NH1-terminal sequence of the purified mLAMP-2 protein. Potential glycosyla- tion sites of N-glycans (Asn-X-Thr/Ser) are marked as - - -CHO- - -. Cysteine residues are circled. The NH*-terminal protein sequence and the tryptic peptide sequence obtained from the purified mLAMP-2 are labeled by a solid under- line. The putative transmembrane por- tion is underlined by a single broken line, and the consensus sequence for polyad- enylation signal is underlined by double broken lines. The nucleotide sequences in italic letters are EcoRI linkers added to the cDNA in the process of cDNA library preparation.

cDNA of Mouse LAMP-2 5011

‘AATTCC‘ GCTTTCGCTTCCGCAGGGCT‘~‘*““*T*c‘TT“~T*‘T~“T~~“T“T~‘~~T“~cT‘*“‘*T~*~‘ 85 AT‘ KC CTC XT CC‘ ‘TT AAA G‘C ‘CA AA‘ CTC ATC CT‘ ATC TTT CT‘ TTC CTA “A ‘CC ‘TT CA‘ ‘CA TX TCC

-25 Met cys Le” ser Pro “al Ly5 my *la l.ys Le” 11e I&” 1le Phe I,eu Ph.3 Le” GlY Ala “al ‘ln ser Am Ala

I---C”‘---,

385 ‘AA ‘CA TCT CAT TAT TCA ATT CAT ‘AC ATC GT‘ CTT XC 76 ‘1” Ala ser His *yr ser 11e HIS Asp 11e “al lx” ser

910 CAA CTT A“ CT‘ AAC ARC AGC CAA ATT AA‘ TAT CTT ‘AC 251 ‘1” k” AL9 Le” *5n *sn ser ‘ln Il.? Lys *yr IA” Asp

I---cm---,

985 CT‘ AA‘ GM ‘T‘ RAT GTC TAC AT‘ TAT TTG ‘CT AAT GGC TCA ‘CT T-K Ax ATT *cc AAC AA‘ Ax CTT AGC TTC 276 Le” Lys GlU va1 *sn “al Tyr Met Tyr Le” Ala ASn GlY Ser *la Phe *s,, Ile Ser Asn Lys Asn Le” Ser Phe

r*c IUC ACT *G= G*= *Gc *CA G=* === CC= a= cc= *y* *sn Thr ser ASP se= ThZ “al Phe Pro ‘ly Ala

WC AM. GTT cc* TTG ‘AT ‘TC RTC TTT AA‘ T‘C RAT Phe Lys “al Pro Leu ASP “al Ile Pha Lys 0 Cys 11s.n

TAT TG‘ ‘CT ATT CAC CT‘ CAA ‘CT TTT ‘TC CAR ART *yr TTp Gly Ile H1S Leu Cl” Ala me va1 ‘133 *=n

I--

CAA ACT ccc *cc ACT ‘TG ‘CA ccc ATC ATT CAC ACC ‘ln ‘rhr em ‘chr ‘rhr Val *la PTO Ile 11-z His Thr

CTT AK TTT ‘CT GT‘ AAA ART ‘AA AAA C“ TTC TAT LBU Ile Phe *la Val ‘ys Asn ‘lu LYS Arg Phe Tyr

I---C”‘---, I---CHO---I I---cm---I

1060 KG ‘AT ‘CC CCT CT‘ “A AGT XT TAT AT‘ TGC Ax !+A?3 GAG CA‘ ‘T‘ CTT TCT ‘TG TCT AGA ‘CG TTT CA‘ ATC 0 301 *rp ASP Al.2 Pro l&?u ‘ly ser ser *yr “et cys *sn Lys ‘1” Gl” Val Le” SeT va1 ser *rg w

1135 AAC *cc ITT AAC CTR AA‘ GT‘ CAR CCT TTT ART GT‘ ACA AM. “A CA‘ TAT TCT ACA ‘CT CAR ‘AC KC 0 A‘T ‘CR 326 Asr ThT PW “al ‘1” Pro Phe A\sn “al Thr Lys Gly Gln Tyr Ser Phr Ala Gl” ASP Cys Ser Ala

I---CHO---I

FIG. 5. Hydropathy plot of mLAMP-2. Hydropathy values were determined using the algorithm of Kyte and Doolittle (1982) with a window of n = 11. Potential N-linked glycosylation sites (NXT/S) and the predicted membrane-spanning region (box) are shown below the hydropathy plot. The hydrophobic regions are represented by positive u&es.

deduced amino acid sequence (a total of 209 amino acid residues) contained the NHz-terminal amino acid sequence of the purified mLAMP-2 protein, thus confirming the identity of the clone. Although this clone encoded the NH2 terminus, it did not contain full-length cDNA of mLAMP-2 correspond- ing to the core polypeptide of 43 kDa which is synthesized in the presence of tunicamycin (Chen et al., 1986). Therefore, the 700-bp cDNA was labeled with 32P and used as a probe to rescreen the same liver cDNA library for a clone containing an insert of >1200 bp. Two positive clones (RL3 and RL7) containing inserts of approximately 1700 bp were obtained, and each was sequenced. Both contained a 5’ sequence iden- tical to those of the 700-bp cDNA clones selected by the PCR product.

The complete nucleotide sequence of clone RL3 (Fig. 4) begins with an 84-bp 5’-untranslated region followed by a single open reading frame of 1245 bp, a termination codon after nucleotide 1329, and a 3’-untranslated sequence of 385 nucleotides. A putative polyadenylation signal AATAAA (nu- cleotides 1680-1685), is present 21 nucleotides proximal to a poly(A) sequence (Nevins, 1983).

The nucleotide sequences of the two clones were identical

by guest on April 19, 2018

http://ww

w.jbc.org/

Dow

nloaded from

cDNA of Mouse LAMP-2

FIG. 6. Comparison of lysosomal membrane glycoprotein sequences. The sequences of the LAMP-1 molecules mLAMP-1 (m), hLAMP-1 (h), lgp120 (r), and LEPlOO (c) and of the LAMP-2 molecules hLAMP-2 (h) and mLAMP-2 (m) were compared by the Align program of GENmenu. Amino acid residues are numbered at the end of each line. Conserved amino acid residues within each class are boxed, identical residues among all six proteins are shadowed and gaps are indicated by dashes.

except for 2 residues. At nucleotide 949, clone RL7 contained a T residue instead of C, resulting in a substitution of phen- ylalanine for leucine at amino acid residue 264; and at nucleo- tide 1135, RL7 contained G instead of A, resulting in a substitution of aspartic acid for asparagine at residue 326. A tryptic fragment of the purified mLAMP-2, corresponding to amino acids 322-331, contained asparagine at residue 326, in agreement with the deduced amino acid sequence of clone RL3. The significance of these single-base differences at positions 949 and 1135 between the two mLAMP-2 clones is unclear. They may be the result of DNA mismatch during cellular proliferation (Modrich, 1987) or a transcriptional modification.

Predicted Structure of mLAMP-e--The amino acid se- quence of mLAMP-2, predicted from the cDNA sequence and starting at the amino terminus as determined by protein sequencing, consists of 390 amino acids with a calculated M, of 43,017. The predicted mass is in close agreement with the apparent molecular mass of the mLAMP-2 core polypeptide (43 kDa). The deduced amino acid sequence matches both the NH*-terminal sequences (30 residues) determined from the purified mLAMP-2 protein (disregarding the unidentified res- idues) and residues 322-331 of a tryptic peptide. In addition to the sequence corresponding to the mature protein, the cDNA encodes a presumed NH*-terminal signal peptide, as expected for an integral membrane protein. This 25-amino

acid sequence contains the common pattern for a signal sequence (Von Heijne, 1983): an initiating methionine, a central hydrophobic region, and alanine at the -1 position.

The sequence of the molecule and its hydropathy values define three putative domains (Fig. 5): 1) the NH*-terminal intraluminal sequence from residues 1 to 354; 2) a hydropho- bic transmembrane region of 24 residues (residues 355-378) lacking charged amino acids or glycosylation sites; and 3) a COOH-terminal cytoplasmic tail of approximately I2 amino acid residues. The 17 potential N-glycosylation sites (Asn-X- Ser/Thr) all reside in the putative intraluminal domain.

Homology to Other Lysosomal Membrane Glycoproteins- Comparison of the sequences of mLAMP-2 with those re- ported for other lysosomal membrane glycoproteins shows that each of these molecules falls into one of the two classes, with mLAMP-2 and hLAMP-2 in one class (LAMP-2) and mLAMP-1,4 hLAMP-1, lgp120, and LEPlOO in another (LAMP-l) (Fig. 6). These data thus confirm the immunolog- ical and biochemical evidence (Chen et al., 1985a; Mane et al., 1989) of at least two classes of lysosomal membrane glycopro- teins in mouse and human cells.

The LAMP-2 proteins are highly similar, with 63% identity in overall amino acid sequence between the mouse and human forms (Fig. 6). All cysteines are conserved, as are 11 of the potential N-glycosylation sites. The protein sequence homol- ogy is particularly extensive at the COOH terminus, with 81 matches in the last 100 residues. In the transmembrane domain and cytoplasmic tail, the homology is 91%. The close similarity between mouse and human LAMP-2 suggests that they represent species variants of the same molecule.

The LAMP-l molecules are likewise closely related and contain extensive sequence homology in the transmembrane domain (83%) and in cytoplasmic tail, where the four mole- cules have identical sequence. All cysteines are conserved, as are seven potential N-glycosylation sites. The proline-rich hinge-like region in the middle of the intraluminal domain is the least homologous portion of the sequence, and several gaps must be introduced to align these regions.

The LAMP-l and LAMP-2 molecules appear to have evolved from a common ancestor because there is a -20% identity in amino acid sequence among the six proteins that have been analyzed. All cysteines are conserved, suggesting that the two classes of molecules have the same disulfide bond structure; in proteins with disulfide bonds the cysteine resi- dues are among the most highly conserved during evolution (Doolittle, 1986). It is noteworthy, however, that these two classes of molecules differ considerably in the COOH-terminal cytoplasmic domain, with identity in only 5 of 12 residues; in contrast, the intraclass molecules of LAMP-l and LAMP-2 contain virtually identical sequences. These differences may have functional significance to the cytoplasmic associations of the molecules.

The homologies between LAMP-l and LAMP-2 proteins suggest that these molecules share some structural aspects that may be important in their common role as lysosomal membrane components. The major difference between the two molecules (besides the sequence) is that LAMP-2 appears to be present at a considerably lower concentration. Other- wise, the molecules have similar biochemical properties, mem- brane localization, biosynthetic processing, tissue or cell

4 The deduced protein sequence reported previously for mLAMP- 1 was truncated at the NH2 terminus (2 residues missing) as compared with the amino acid sequence obtained from the purified mLAMP-1 protein, The sequence shown here is from a clone that was isolated recently from a mouse macrophage cDNA library and contains the full-length mLAMP-1 sequence for the mature protein and a partial sequence of the putative signal peptide.

by guest on April 19, 2018

http://ww

w.jbc.org/

Dow

nloaded from

expression, and biosynthesis and transport to lysosomes in I- cells (Chen et aZ., 1986; Sandoval et al., 1989; Mane et al., 1989). However, since LAMP-l and LAMP-2 differ at the molecular level, they may perform different functions in ly- sosomes.

Acknowledgments-We wish to acknowledge Dr. R. F. Ambinder for the use of a C&us DNA thermal cycler; L. M. Arterburn for the preparation of tryptic peptides; B. J. Earles and F. G. Guarnieri for technical assistance; Dr. M. Strand, L. M. Arterburn, and E. J. Wolffe for critical reading of the manuscript; Dr. D. McClellan for editorial assistance; and G. Will for word processing and editing.

REFERENCES

Arterburn, L. M., McClellan, D. A., and August, J. T. (1990) in Intracellular Trafficking of Proteins (Steer, C. J., and Hanover, J. A., eds) Cambridge University Press, Cambridge, United Kingdom, in press

Atkinson, T., and Smith, M. (1984) in O&onucleotide Synthesis (Gait, M. J., ed) pp. 35-81, IRL Press, Washy, D. C.

Barriocanal. J. G., Bonifacino, J. S., Yuan, L.. and Sandoval, I. V. (1986) J. Viol. ehem. 261, 1’6755516763

Carlsson, S. R., Roth, J., Piller, F., and Fukuda, M. (1988) J. Biol. Chem. 263,18911-18919

Chen, J. W., Murphy, T. L., Willingham, MC., Pastan, I., and August, J. T. (1985a) J. Cell Biol. 101.85-95

Chen, J. W., Pan, W., D’Souza; M. P., and August, J. T. (1985b) Arch. Biochem. Biophys. 239,574-586

Chen, J. W., Chen. G. L.. D’Souza. M. P., Murnhv, T. L., and August, J. T. (1986) B~o~hem. Sot. Sym& 51, b7-lf2 -'

I

Chen. J. W.. Cha. Y.. Yuksel. K. U.. Gracv. R. W.. and Auzust. J. T. (1988) J. biol. Che’m. 263; 8754&758 ” ‘ ’ -

Dahms, N. M., Lobel, P., and Kornfeld, S. (1989) J. Bio2. Chem. 264, 12115-12118

Davis, R. W., Botstein, D., and Roth, J. R. (1980) Aduanced Bacterial Genetics, pp. 80-83,106-108, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

Dennis, J. W., Laferti, S., Waghorne, C., Breitman, M. L., and Kerbel, R. S. (1987) Science 236, 582-585

Devereux, J., Haeberli, P., and Smithies, 0. (1984) Nucleic Acids Res. 12,387-395

Doolittle. R. F. (1986) Oi Urfs and Orfs. Universitv Science Books. Mill Valley, CA ’ ’ ’ -

Edge, A. S. B., Faltynek, C. R., Hof, L., Reichert, L. E., Jr., and Weber, P. (1981) Anal. B~ochem. 118,131-137

Fambrough, D. M., Takeyasu, K., Lippincott-Schwa~z, J., and Siegel, N. R. (1988) J. CeEZ Biol. 106.61-67

Febbraio; M., ‘August, J. T., and-Silverstein, R. L. (1989) J. Cell Biol. 109, 254 (abstr.)

Feinberg, A. P., and Vogelstein, B. (1983) Anal. Biochem. 132,6-13 Fukuda. M. (19851 Biochim. BioDhvs. Acta 780.119-150 Fukuda[ M.,‘Viit&a, J., Mattes&J., and Car&son, S. R. (1988) J.

Biol. Chem. 263,18920-18928 Geuze, H. J., Stoorvogel, W., Strous, G. J., Slot, J. W., Bleekemolen,

J. E., and Mellman, I. (1988) J. Cell Biol. 107, 2491-2501 Gribskov, M., Burgess, R. R., and Devereux, J. (1986) Nucleic Acids

Res. 14,327-334 Griffiths, G., Hoflack, B., Simons, K., Mellman, I., and Kornfeld, S.

(1988) Cell 52, 329-341 Hewick, R. M., Hunkapiller, M. W., Hood, L. E., and Dreyer, W. J.

(1981) J. Biol. Chem. 256,7990-7997 Howe, C. L., Granger, B. L., Hull, M., Green, S. A., Gabel, C. A.,

Helen&, A., and Mellman, I. (1988) Proc. Natl. Acad. Sci. U. S. A. 85,7577-7581

Hughes, E. N., and August, J. T. (1982) J. BioZ. Chem. 257, 3970- 3977

Hunter, W. M. (1967) in Handbook of Experimental Immunology (Weir. D. M.. ed) DD. 608-692. F. A. Davis. Philadelahia

Kyte, J.; and Doolit&, R. F. (1982) J. Mol. Viol. 167,*105-132 Laemmli, U. K. (1970) Nature 227,680-685 Lee, C. C., Wu, X., Gibbs, R. A., Cook, R. G., Muzny, D. M., and

Caskey, C. T. (1988) Science 239, 1288-1291 Lewis, V., Green, S. A., Marsh, M., Vihko, P., Helenius, A.. and

Mellman, I. (1985) J. cell Biol.’ 100, 1839-1847 LiDDincott-Schwartz. J.. and Fambroueh. D. M. (1986) J. Cell Biol.

-l-02,1593-1605 ’ I

Lippincott-Schwartz, J., and Fambrough, D. M. (1987) Cell 49,669- 677

Mane, S. M., Marzella, K., Bainton, D. F., Holt, V. K., Cha, Y., Hildreth. J. E. K., and August, J. T. (1989) Arch. Biochem. B~ophys. 268,366-378

_ -

Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982) MoZec~~ur Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

Modrich, P. (1987) Anna. Rev. Biochem. 56, 435-466 Nevins, J. R., (1983) Annu. Reu. Biochem. Fi2,441-466 Sandoval, I., Chen, J. W., Yuan, L., and August, J. T. (1989) Arch.

Biochem. Biophys. 271,157-167 Sanaer. F.. Nicklen. S.. and Coulson. A. R. (1977) Proc. Natl. Acad.

Sii. b. s. A. 74,546&5467 Stone, K. L., and Williams, K. R. (1988) Macromolecukzr Sequencing

and Synthesis: Selected Methods and Am&ations. DP. 7-24. Alan R. Liss Inc., New York

_ . _-

Viitala, J., Carlsson, S. R., Siebert, P. D., and Fukuda, M. (1988) Proc. Natl. Acad. Sci. U. S. A. 85, 3743-3747

Von Heijne, G. (1983) Eur. J. B~~hem. 133, 17-21 Wood, W. I., Gitschier, J., Laskey, L. A., and Lawn, R. M. (1985)

Proc. Natl. Acad. Sci. U. S. A. 82, 1585-1588 Wray, W., Boulikas, T., Wray, V. P., and Hancock, R. (1981) Anal.

Biochem. 1 X8, 197-203

by guest on April 19, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Y Cha, S M Holland and J T Augustmembrane glycoproteins.

The cDNA sequence of mouse LAMP-2. Evidence for two classes of lysosomal

1990, 265:5008-5013.J. Biol. Chem. 

  http://www.jbc.org/content/265/9/5008Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/265/9/5008.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on April 19, 2018

http://ww

w.jbc.org/

Dow

nloaded from