Proc. Nati. Acad. Sci. USAVol. 84, pp. 2233-2237, April 1987Biochemistry

Cloning of the bovine 215-kDa cation-independent mannose6-phosphate receptor

(cDNA clone/DNA sequencing/repeating domains)

PETER LOBEL*, NANCY M. DAHMS*, JAMES BREITMEYER*t, JOHN M. CHIRGWINf§, AND STUART KORNFELD**Departments of Medicine and Biological Chemistry, and tDepartment of Anatomy and Neurobiology, Washington University School of Medicine,St. Louis, MO 63110

Contributed by Stuart Kornfeld, December 24, 1986

ABSTRACT Four overlapping cDNA clones encoding apartial sequence of the cation-independent 215-kDa mannose6-phosphate receptor have been identified by screening a fetalcalf liver cDNA library with oligonucleotide probes. RNAhybridization analysis showed that the length of the mRNAis -9.5 kilobases. Sequence analysis demonstrated that theclones consist of 4647 contiguous nucleotides and contain anopen reading frame coding for a polypeptide of 1461 aminoacids, which we estimate represents >75% of the primarystructure of the receptor. The deduced amino acid sequenceindicates that the receptor has a carboxyl-terminal cytoplasmicdomain of 163 amino acids that is rich in acidic residues, a23-amino acid transmembrane segment, and an extracellulardomain containing at least eight homologous repeats of %145amino acids. One of the repeats contains an additional 43-residue segment that is similar to the type II repeat offibronectin. Each repeat contains a highly conserved 13-aminoacid unit bordered by cysteine residues that may be functionallyimportant.

Lysosomal enzymes are delivered to lysosomes by twopathways that require the recognition of phosphomannosylresidues by specific receptors (reviewed in ref. 1). In thebiosynthetic pathway, mannose residues on newly synthe-sized lysosomal enzymes are phosphorylated in the Golgicomplex. Acquisition of this marker allows binding to man-nose 6-phosphate (Man-6-P) receptors. The receptor-lyso-somal enzyme complex leaves the Golgi and is transported toan acidic prelysosomal compartment. In the endocytic path-way, phosphorylated lysosomal enzymes bind to Man-6-Preceptors on the cell surface and are transported to an acidicendosomal compartment. The low pH of the acidic compart-ment promotes dissociation of lysosomal enzyme and recep-tor. The receptors recycle to the Golgi or cell surface whilethe lysosomal enzymes are packaged into lysosomes. Thedegree of overlap between the two pathways is not known.Two different Man-6-P receptors have been implicated in

the targeting of lysosomal enzymes. The cation-independent(CI) Man-6-P receptor has an apparent molecular weight of215,000 and does not require divalent cations for ligandbinding (2). The cation-dependent Man-6-P receptor has anapparent molecular weight of 46,000 and requires divalentcations for ligand binding (3). Both show similar, but notidentical, binding specificities toward various types of phos-phorylated oligosaccharides (4). It is not clear if the receptorshave distinct functions. While both receptors have beenpurified to homogeneity, little is known about their primarystructures. We now report the isolation and sequencing ofcDNA clones encoding part of the CI Man-6-P receptor. Thenucleotide sequence reveals that the receptor is a transmem-

brane protein with an extracytoplasmic repeating domainstructure.

METHODS

Protein Sequencing. CI Man-6-P receptor was purified frombovine liver acetone powder by affinity chromatography asdescribed (3). The receptor was reduced with 2-mercapto-ethanol, alkylated with iodoacetamide, and hydrolyzed withtrypsin (5). The tryptic peptides were fractionated on acolumn of LiChrosorb RP-8, 5 ,um (Unimetrics-Knauer,Anaheim, CA) using a 0-50% gradient of acetonitrile in 0.1%trifluoroacetic acid. Peptides were sequenced by automatedEdman degradation using an Applied Biosystems model 470Aamino acid sequencer.

Library Screening. A fetal calf liver cDNA library, kindlyprovided by Peter Rotwein (Washington University), wasconstructed as described (6). The library consists of 100,000cDNA molecules of >500 base pairs inserted into the Sst Isite of pUC19 by homopolymer tailing. Nitrocellulose filterreplicas of the library were screened by colony hybridization(7) with oligonucleotide probes. The probes were synthesizedon an Applied Biosystems model 380A oligonucleotide syn-thesizer and labeled at the 5' end with 32p using [y-32P]ATPand polynucleotide kinase (8). Hybridization and washingconditions were as described (9). Positive clones were puri-fied by sequential screening at low colony density. PlasmidDNA was isolated and characterized by restriction endonu-clease mapping. Restriction fragments were labeled by nick-translation (10) and used to rescreen the library.

Nucleotide Sequencing. Restriction fragments were sub-cloned into bacteriophage M13mp18 and -mpl9 (11) andsequenced by the dideoxynucleotide chain-termination meth-od (12), using both a universal sequencing primer andoligonucleotide primers specific for the CI Man-6-P receptor.The sequence was determined across all restriction sites, andthe entire sequence of the coding region was determined onboth strands. When necessary, we used 7-deaza-dGTP tosequence through GC-rich regions (13). In four instanceswhere we could not choose among several possible se-quences, the following strategy was used: oligonucleotidescorresponding to each of the possible sequences were syn-thesized, each having the same theoretical melting temper-ature. The oligonucleotides were 5'-end labeled with 32P andhybridized to restriction fragments immobilized on nitrocel-lulose. The complexes were then sequentially washed atincreasing temperatures and autoradiographed. In all cases,one oligonucleotide dissociated at the theoretical tempera-

Abbreviations: Man-6-P, mannose 6-phosphate; CI, cation-indepen-dent.tPresent address: Dana-Farber Cancer Institute, Division of TumorImmunology, Boston, MA 02115.§Present address: Department of Medicine, University of Texas, SanAntonio, TX 78284.

2233

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.

Proc. Natl. Acad. Sci. USA 84 (1987)

ture, while the other oligonucleotides dissociated at lowertemperatures.RNA Analysis. Polyadenylylated RNA was isolated by

homogenization offresh calf liver in guanidine isothiocyanateand sequential precipitation from guanidine hydrochloride(14), followed by chromatography on oligo(dT)-cellulose(15). RNA was electrophoresed through a 1% agarose gelcontaining 2.2 M formaldehyde (8). RNA standards (Bethes-da Research Laboratories) were in an adjacent lane. The gelwas stained with ethidium bromide, and the RNA wastransferred electrophoretically to a cationized nylon mem-brane (GeneScreenPlus, DuPont). The membrane was hy-bridized to probe in 50% (vol/vol) formamide as recommend-ed by the supplier.Computer Analysis. DNA sequence data were analyzed

using programs from Compugene (St. Louis, MO). Proteinsequence homology searches, computer-assisted alignments,and dot matrix analysis were performed using programsdescribed (16, 17).

RESULTSProtein Sequencing. Tryptic peptides from S-carboxami-

domethylated purified CI Man-6-P receptor were separatedby HPLC. Eight isolated peaks from two independent prep-arations were analyzed by amino-terminal amino acid se-quencing. One sample (T2) contained two peptides while theother samples were homogeneous (Fig. 1A).

Isolation and Characterization ofcDNA Clones. We used thesequence information and codon usage frequencies in mam-malian proteins (19) to design three oligonucleotide probes,OP1, OP2, and OP3 (Fig. 1B). These probes were used toscreen a plasmid library of fetal calf liver cDNA. One clone,pBR1, reacted with all three probes (Fig. 1C). Nucleotidesequencing revealed that pBR1 contained an open readingframe coding for four of the tryptic peptides. We rescreenedthe library with a nick-translated probe derived from pBR1(nucleotides 101-1768) to obtain clone pBR2 (Fig. 1C).Similarly, we used a fragment of pBR2 (nucleotides1756-3241) to obtain clones pBR3 and pBR4 (Fig. 1C).The four clones contain 4647 contiguous nucleotides con-

sisting of an open reading frame of 4383 nucleotides thatcodes for 1461 amino acids followed by 264 nucleotides of3'-untranslated sequence (Fig. 2). Eight of the nine peptidesequences (Fig. LA) are found in the nucleotide-derivedsequence. These peptide sequences contain 114 identifiableresidues, 113 of which are identical in the sequence predictedfrom the cDNA clones. The disagreement is at residue 19 ofT6 (Fig. LA) where the experimentally determined residue islysine, while the residue predicted from the nucleotide

1 AACGGGAGGGCTGAVGCCTTCATCATCCGCrCGTCTGCAAT CGTT!ACCCAGGGACAC1 H G R A D A F I I R F V C N D D V Y P G T_ T13133 GCGCTGGCCTGTGTACCTTCTCCGGTAGATTGCCAAGTCACAGACCCCGCCGGGAACGAGTATC45 A L A C V P S P V D C Q V T D P A G NETY

265 AGGACCTTACCTGAGCGGTGTCACGCCTCTCCCGTACATTCCCGGCTGCCACGGCACCGCTC89 R T F Y L S V C T P L P Y I P G C H G T A

ASampleTi

T2ab

T3

T4

T5

T6T7

SequenceA E G D Y C E V R D P R

A G F T A A Y S E KA A C A V K P Q E V Q M V X G T

F V C N D D V Y P G T P K

F L H Q D I D S S L G I R D T F F Q

I Y K G P Q D C S E R

V A G P P I L N P I A N E V Y L N F K

H G N L Y N L I P L G L X D T V V

T8 A T L I T F L C D R D A G V G

B** ** ** * * *

OP1 GCCGGCTTCACCGCCGCCTACTCNGAGGTGCAGATGGTGA G F T A A Y S E V Q M V

* * * * * * *OP2 TTCGTGTGCAACGACGACGTGTACCCCGGCACCCCCAAG

F V C N D D V Y P G T P K* * * * *

OP3 CACGGCAACCTGTACAACCTGATCCCCCTGGGCCTGH G N L Y N L I P L G L

C

0 1

-OP2 -OP3 -OP1

KILOBASES2 3 4 5

BR1BR2

BR3BR4

FIG. 1. Amino acid sequence of tryptic peptides, nucleotidesequences of oligonucleotide probes, and cDNA clones for the CIMan-6-P receptor. (A) Tryptic peptides. Phenylthiohydantoin deriv-atives were identified by HPLC following each cycle of the aminoacid sequencer (18). X, no residue was detected. Samples T1-T5 andT6-T8 were from independent preparations. (B) Oligonucleotideprobes. The nucleotide sequences are displayed above their concep-tual translation products. Mismatches to the actual cDNA sequenceare marked by asterisks above the probe sequence. OP1, OP2, andOP3 were designed based on the sequences of T2a, T3, and T7,respectively. Mismatches in the penultimate three codons ofOP1 aredue to our mistakenly attributing sequence of peptide T2b to T2a.The actual oligonucleotide probes were complementary to thesequences shown. (C) cDNA clones.

sequence is glutamic acid (Fig. 2). Furthermore, all eightpeptides are preceded by a potential tryptic cleavage site. Nopotential initiation codon is found upstream of the sequencecoding for the first tryptic peptide, nor is there a poly(A)

:CCCAAGTTCCTGCACCAGGACATCGACTCTAGCCTGGGGATCCGGGACACTTTCTTCGAGTTTGAAACCP K F L H Q D I D S S L G I R D T F FE F E T

I1 T41r.ATCGTGGCCTCGGCAAGGCCAGGAACCGTGGACTCGGTTGCACGTCNCATGACGCGCGAAGD L S G L S K A R K P W T A V D T F D E G K K

rGTGGGGTGCTGCCTGGTGACGGAAGACAGCAAGTTGAACCTAGGCGTCGTGCAGATCAGTCCTCAGGTGV G C C L V T E D S K L N L G V V Q I S P Q V

397 GGCGCCAACGGGTCCCTGAGCCTCGTCTACGTCAACGGGGACAAGTGCAAGAACCAGCGTTTCT133 G A In G S L S L V Y V 3 G D K C K N Q R F

529 GAGTATGTGTTTCTCTGGAGAACCGTGGAAGCCTGTCCCGTCGTGCGTGCGGAAGGAGACTAC.177 E Y V F L V R T V E A C P V V R A E G D V

661 GSGCGGGCCGGCGAATACACCTATTACTTCCGCGTCTGCGGAGAGCTGACATCCGGCGTCTGCC221 V R A G E Y T Y Y F R V C G E L V S G V C

793 GTGGCAGGTCTGTITAATCAG;AAGCTGACCTACGAAATGGGGTGCTGAAUGATGAACTACACCG265 V A G L F N Q K L T Y E k G V L K N V VTl

TCCACCAGGATA&ACCTCGAGTGTGCCCACACAACGGGCTCCCCGACCTTTCACTCCAGAACGACTGTS T R I N L E C A H T T G S P T F Q L Q N D C

uGCGAGGTGAGAGACCCAAGGCACGGCAACCTGTATAACCTGATACCTCTTGGTCTGAACGACACTGSCC E V R D P RIH G N L Y N L I P L G L N D V V-T1 1 T7CAAccAGTGACAAGTC~tAAGGTCATcTcATCATGCCAGGAAAAGCGGGGCCCCACCGATTCAAA"P T S D K S K V I S S C Q E K R G P Q G F Q K

l;GGGCGACACCTGCCACAAGGTGTACCAGCGTTCCACCACC&C TTTCTACTGCGACCGCAGCACGG G D T C H K V Y Q R S T T I FF V C D R S T

FIG. 2. (Figure continues on the opposite page.)

925 CAGGCGCCCGTGTTTCTCCAGACGTCCGATTGCTCCTACCTGTTTGAGTGGCGCACGCAGTACGCCTGCCCGCCCTACGACCTGACCGAGTGTTCGTTCCAAAAACG:GCT=GACCTACGACCT309 Q A P V F L Q E T S D C S Y L F E V R T Q Y A C P P Y D L T E C S F Q K R G W G N L R P

1057 CTCGTCT CTGTCACGGGCACAGGGTCCACCGAGCACTACCTCATCA&CGSGTGCAAGTCCCTGTCCCCGCAGGCTGGCTCAGATCCGTGCCCTCCGGAGCGGC353 L V S V E V Q R Q L G G C H G H R V H R A L P H Q R V Q V P V P A G W L R S V P S G G G

I-

2234 Biochemistry: Lobel et al.

Biochemistry: Lobel et al. Proc. Nati. Acad. Sci. USA 84 (1987) 2235

1189CCCTAGAGTAGTATGGCAACGTCGAAGA397 R V S A G R F Q A V N L G R V R D S F Q V S Q G L T L L K Y V D G D L C F D Q I R K K S

1321ACACTCCTAGGACGGCAGGATCGCCGTCTACCGGAGTGGAAACCTCGCCCGCCGCGGGTLiACAGGC441 T T I N F T C S K S S V N S R F A F I S A V K D C K TY F S V P T A A A C A V K S N V S

485 D D C Q V T N P AT GENL F D L S S L S GRAGFTAAY K L V T L S V C G D N E

529 N C A N G V G A C F G Q T R I S V G K A S K R L T T V D Q V L Q L V Y K G G S P C F S K

1717 Ac~cG~TcAAc~Ac~TcTTcGc~~~G~c~c~A~ GAGG.CCTTTCCTGAAACCGC573 T G L S T K S V I S F V C R P K V G F T N R F A L I S L D K R T C T L F F S V S T F L A

1849 Tc~cGc~CATCcGGGAc~ACCCCTGC~.CCATACACCA CAACGCTA617 C Z Q T T K C S V NR S L I D L S P L I S R T G G Y K A Y D K S K D D G S D T S P D

I131 C Q F L S PAS G L A C F AG T A V C K V FT D G FF1 D IG REI TAG FF1 L

IAK VT L FEK S S T P C L A D R S F IN T T S L I T F S C K R G V S A G T P K LT6

L RT S V C D F V F E VE T P L VC P DK9 V K T D G C s L T DK Q L Y T S F3L S L S

K S T F K V T R G PEBT Y S V G V C T A AA G L D E GG C K D G A V C L L S G S K GA S

F G R L AS M K L D Y REBQ D K A V I L STYA N G D T C P P K T E D GE P C V F P F V F

E G K S Y E K C V V K S R AE LVC A T TAN Y DRI DE EVWG F C KSB S T SE8 R T S V I

I F K C D K D AD V G R P Q V F E V R G C K V T F K VK T K V V C P P K K M K C K F V

Q KSH R T Y D L R L L S S L T G SVWS F VEHN G AS T T I N L CQ K I1 TEKG PFQ D C SKE

_!tAS V C K K S T S G K V Q V L G L V S T Q K L D V V D D R V I V T Y S K G S Y C G D rN

1189 Q D G

A S A V I E LY C A K T V GER P S FT R F D V D S C T YEB F S W D S R A A C AT K P

QIVN G TIl T N P AN GER SFPS L G D ITY F KERF S A S G D V EYE G D R T I Y

L S SI T G S SS P A C S G A S IC Q R K ASN D QS F S R K V G T SN~ TRY Y V

D L D V V F T S S SK C G K D KYT K STV S S T I FFS C D FL V K D G IF K F SEB

D C Q Y L F S VE8T S AT C F L G AG F D EE I A G D D A Q ESB K G L S EERS Q A

V G ATV L S L L L VA L TA C L L T L L L Y K K K 2ZE EV MS R L T N C C R R S AN

SJY K Y SKXV N K K K K A DEN KYK L KI P PPE K QKN E A

R A A D T L S A LE G D EQ D SER D ZST L T L PET KTVR P P G N A P G A E GG P

F L FR K A P P P LERAD D R V G L V R G K P A RR GE F R AA AT F I ST FEHD

D L L V end

4489GGGAGGGGCCCTTCGTAGCGGCGCCGTTCCCGCTCCTACTC4621 GC&AGGCWGATAGC

FIG. 2. Partial nucleotide sequence and deduced amino acid sequence of the bovine CI Man-6-P receptor. The numbers on the left refer tothe nucleotide sequence (upper sequence) and the amino acid sequence (lower sequence, underlined). Sequences corresponding to the trypticpeptides are underlined and named as in Fig. lA. Potential sites for N-linked glycosylation are boxed. A potential transmembrane region isunderscored by a solid bar. The nucleotide sequence is a composite of sequences determined from the overlapping clones shown in Fig. 1C:nucleotides 1-2220 were determined from pBR1; 1700-3590 from pBR2; and 3270-4641 from pBR3 and pBR4. Nucleotide 3455 is cytosine inpBR3 and pBR4 and thymine in pBR2, which changes the amino acid at position 1052 from threonine to methionine. The sequence of nucleotides4203-4222, 4246-4265, 4540-4559, and 4565-4584 were confirmed by hybridization with oligonucleotides.

sequence or a canonical polyadenylylation signal at the 3' end

of the cDNA. These observations show that we have cloned

an internal fragment of the CI Man-6-P receptor mRNA.

The deduced 1461-amino acid sequence contalns a stretch

of 23 hydrophobic residues (positions 1276-1298) that is

followed directly by the basic sequence Lys-Lys-Glu-Arg-

Arg-Glu (Fig. 2). This is likely to be a transmembrane region,

as the CI Man-6-P receptor spans the membrane, and this

sequence is similar to the transmembrane sequences of

numerous membrane proteins (20). There are 163 amino acids

on the carboxyl side of the proposed transmembrane region.

The predicted sequence also contains 11 consensus se-

quences for N-linked glycosylation (Asn-Xaa-Ser/Thr), nine

of which are on the amino-terminal side of the transmem-

brane region. The asparagines at positions 217 and 1107 (Fig.

2) are probably glycosylated since no residue was detected by

amino acid sequencing at cycle 13 of T7 and cycle 14 of T2

(Fig. lA).

Blot Hybridization Studies. To estimate the size of the CI

Man-6-P receptor mRNA, polyadenylylated RNA from calf

F T

N p

1981661

2113705

2245749

2377793

2509837

2641Sa1

2773925

29G5969

30371013

3169 AAGACAG1057

3301 C&GGAGG1101 Q E

T2b-3433 GAGATCC1145 K I

;TGV

AGQ

3565

36971233 K T A

38291277

39611321

4093 AAGTCGGTGJ1365 K S V

4225 CCGCTGCGG41409 F L R

4357 GACAGCGAC(1453 D S D

TTCTACATCAACATCTGCCAGCCGCTCAACCCCATGCACGGGTTGGCCTGCCCCGCCGGCACGGCCGTGTGCJUGGTTCCCGTGGACGGCCCCCCGATAGATATTGGCCGAGTGGCAGGACCTCCGATCCTC

CACTGTAAGCGGGGCGTCuAG

;TCTGCTGTCATCGAGCTGACCTGTCsCCAAGAC.AGTGGGGCGGCCTTCGTT CAGCTGGGACTCACGAGCGGCCTGCGCCGTCsAAGCCT

C-NACATGGCGACCTGGATrxTGGTGTTCACCTCGTCCTCCAACsTG TCTGTGTCCTCCACCAIL..CACTGTGACCCCCTGGTGAAGGACGGGATCCCCGAGTTCAGCCAC

GAGACTGCCGACTGCCAGTACCTCTTCTCCTGGCACACCTCTGCCGTGTGC CAGAGCGCAGCCAGGCG

MG

Proc. Natl. Acad. Sci. USA 84 (1987)

A B

es 9'-9.5-7.5

-4.4

-2.4

-1.4

FIG. 3. Hybridization analysis of polyadenylylated RNA fromcalf liver. RNA (10 ,ug) was electrophoresed through a 1% agarose gelcontaining formaldehyde and electrophoretically transferred to acationized nylon membrane. The membrane was hybridized to eithera nick-translated probe made from a restriction fragment correspond-ing to nucleotides 1756-3241 (lane A) or an end-labeled oligonucle-otide complementary to nucleotides 333-382 (lane B). Numbers onthe right indicate the size in kilobases of the RNA standards.

liver was electrophoresed on a denaturing agarose gel andtransferred to a cationized nylon membrane. Duplicate sam-ples were hybridized to two probes made from differentregions of the cDNA sequence. Both probes hybridized to asingle mRNA species that is =9.5 kilobases in length (Fig. 3).Assuming the molecular weight of the deglycosylated proteinto be 205,000 (21), the protein would be encoded by an openreading frame of =6 kilobases. This suggests that the mRNAcontains -3.5 kilobases of untranslated sequence.

Similarity with Other Proteins. A search of the NationalBiomedical Research Foundation protein sequence data-base$ revealed that CI Man-6-P receptor contains a 43-residue sequence that is similar to sequences found infibronectin, factor XII, and a bovine seminal fluid protein(Fig. 4). Over 40% of the residues in this segment of thereceptor are identical with residues in each of the three otherproteins. Fibronectin and the bovine seminal fluid proteincontain two copies of this repeat arranged in tandem, whilefactor XII and the CI Man-6-P receptor contain one copy. Infibronectin this repeat, termed the type II repeat, constitutespart of a collagen binding domain (25). We also compared theCI Man-6-P receptor to a number of cell surface receptors(26) and carbohydrate binding proteins (27) to determine ifany distant homologies existed. No significant relationshipswere detected.

Internal Homology. The CI Man-6-P receptor containsmultiple repeating units. This is illustrated by Fig. 5, wherethe predicted amino acid sequence is compared with itselfand displayed in the form of a dot matrix. The continuousdiagonal represents the exact match that occurs when thereceptor is compared with itself in perfect register; all otherdiagonals represent repeating units of sequence. The repeat-ing units have a periodicity of :145 amino acids. The mosthighly conserved region within the repeat consists ofa stretchof 13 amino acids that contains cysteines at both ends. The

SNational Biomedical Research Foundation (1986) Protein SequenceData Base of the Protein Identification Resource (Washington,DC), Release No. 10.0.

NPRci (871-913)BSFP (66-109)FACT12 (25-70)FIBN (326-371)

300-

uLJ

600zLU

a 900LUo2

1200-

-- ---

\ \.\\''x' \*''';.\\ " \'\\ Ns

'\\"X as a'\ \ N

v \\ sb"~~\-" \. "%act-\"\\"~ ~~~~" \"

300 600 900 1200RESIDUE NUMBER

FIG. 5. Dot matrix analysis of the amino acid sequence of the CIMan-6-P receptor. The amino acid sequence is presented on both thevertical and horizontal axes. Each 30-residue segment is comparedwith all other 30-residue segments. A homology score is calculatedfor each comparison using the mutation data scoring matrix (17). Adot is placed at the position given by the midpoints of the twosegments when the comparison score is .30.

partial sequence contains eight of these repeats (R2-R9, Fig.6) plus a ninth imperfect repeat (R1, Fig. 6). The periodicspacing of the repeated units is also shown in Fig. 6. Thesegment homologous to the fibronectin type II repeat islocated between R6 and R7; if this is deleted from thereceptor sequence, then R6 and R7 are separated by 156amino acids. The similar length and overall homology of theeight 145-amino acid repeats suggests that they arose fromduplications of a single ancestral sequence and may formfunctional domains.

DISCUSSIONWe have isolated and sequenced cDNA clones coding formost of the CI Man-6-P receptor. Analysis of the sequencesuggests that the receptor contains a cytoplasmic, membranespanning, and extracytoplasmic region. The sequence on theamino-terminal side of the membrane spanning region con-tains eight repeats of =145 amino acids and a single 43-aminoacid long segment that is similar to the type II repeat infibronectin.How does the partial sequence fit with what is known about

the structure and function of the receptor? The cDNAsequence indicates that the protein component of the CIMan-6-P receptor is at least 160 kDa. Assuming no post-translational proteolytic cleavage, this is consistent with theapparent molecular weight of 215,000 reported by a numberof investigators (2, 3, 28, 29) and inconsistent with reportsthat the receptor is composed of smaller (<50,000) subunits(30, 31). There is only one apparent membrane-spanningregion in the sequence. The region on the carboxyl-terminalside of the membrane spanning domain is 163 amino acidslong, yielding a molecular weight of 18,000. The region on theamino-terminal side of the membrane spanning region con-tains at least 1275 amino acids, giving a minimum molecularweight of 140,000. We have inferred that at least two N-linkedoligosaccharides are within this region, thus this sequence is

GEPCVFP FVFN KSYEE VVE SRA--RLWCATTANYDRDHEiGFCCVF P FI YGGK KYE T CT I GS MW-N SWC S L S PN YD K DRAWKYC

CHF PFQYHRQLYHKECTEHKIG UPGPQPWCATTPN FDQDQRWGYC

FTYNRTFYSCTTEGRQDGHLWCSTTSNEQDQKYS FC

FIG. 4. Alignment of the amino acid sequences of the bovine CI Man-6-P receptor (MPRci), bovine seminal fluid protein [BSFP (22)], humanfactor XII [FACT12 (23)], and human fibronectin [FIBN (24)]. Residues that are identical in at least three proteins are boxed.

2236 Biochemistry: Lobel et al.

Wm:.

Proc. Natl. Acad. Sci. USA 84 (1987) 2237

residue sequence separationRi 36-48 RDTFFEFETALAC 139

R2 176-188 CEYVFLWRTVEAC

R3 320-332 CSYLFEWRTQYAC

R4 465-477 CEYTFSWPTAAAC

R5 605-617 CTLFFSWHTPLAC

R6 754-766 CDFVFEWETPLVC

R7 946-958

R8 1084-1096

R9 1237-1249

CONSENSUS

CEVTFEWKTKVVC

CTYHFSWDSRAAC

CQYLFSWHTSAVC

Cey FsW T aaCe lv

143

144

139

148

199

137

152

FIG. 6. Alignment of the highly conserved repeat. A consensussequence is determined for repeats R2-R9. The consensus residue isuppercase if it is identical in seven out of eight sequences, islowercase if it is identical in three to six out of eight sequences."Separation" refers to the distance in residues between adjacentrepeats R(N) and R(N + 1).

extracytoplasmic. These assignments are supported by pro-teolysis experiments that demonstrate that the bulk of thereceptor is extracytoplasmic and that the carboxyl-terminalend is cytoplasmic (32, 33).The CI Man-6-P receptor mediates the transport of phos-

phorylated lysosomal enzymes from the Golgi complex andthe cell surface to lysosomes. While ligand binding anddissociation clearly involve extracytoplasmic determinants,transfer between compartments may involve determinantslocated throughout the receptor. Goldstein et al. (26) havecompared different receptors that mediate endocytosis (theasialoglycoprotein, low-density lipoprotein, and transferrinreceptors) and suggest that cytoplasmic cysteines and clus-ters of acidic residues may be involved in receptor internal-ization and targeting. The CI Man-6-P receptor differs fromthese endocytic receptors in that its primary role is to shuttleligands between intracellular compartments. It is related tothe other receptors in that it also travels between a prelyso-somal compartment and the cell surface. While the cytoplas-mic domain of the CI Man-6-P receptor is larger than that ofthe endocytic receptors, it is similar in that it containscysteines (residues 1313 and 1314) and clusters of acidicresidues (residues 1330-1339, 1379-1386, and 1452-1457). Itis also rich in prolines, as is the cytoplasmic domain of theasialoglycoprotein receptors (34). The fundamental questionsof what determinants are responsible for targeting and themechanism by which targeting occurs remain open.The similar size and overall homology of the 145-amino

acid long repeats suggests that they arose from duplication ofa single unit. The high degree of conservation between the13-amino acid long repeats (Fig. 6) indicates that there issome selective pressure that maintains this sequence. The145-amino acid long repeats may be functional domains, andthe highly conserved 13-amino acid long repeats may form a

critical component within these domains. In support of this,we have cloned and sequenced a cDNA encoding the com-

plete 46-kDa cation-dependent Man-6-P receptor (unpub-lished observations): the entire predicted extracytoplasmicdomain of this receptor is homologous to the eight 145-aminoacid repeats of the CI Man-6-P receptor, and it contains theconserved 13-amino acid segment. Thus, it is likely that theseconserved sequences are important for a function common toboth receptors, probably ligand binding.

We thank Gregory A. Grant and Mark Frazier of the WashingtonUniversity Protein Facility and Peter S. Rotwein for their help. P.L.

and N.M.D. are supported by Fellowships PF-2777 and PF-2873 fromthe American Cancer Society. This investigation was supported byGrants CA08759 and HL33933 from the U.S. Public Health Service,Grant 83-1271 from the American Heart Association, and a MonsantoCompany/Washington University Biomedical Research Grant.

1. von Figura, K. & Hasilik, A. (1986) Annu. Rev. Biochem. 55,167-193.

2. Sahagian, G. G., Distler, J. D. & Jourdian, G. W. (1981) Proc.Natl. Acad. Sci. USA 78, 4289-4293.

3. Hoflack, B. & Kornfeld, S. (1985) J. Biol. Chem. 260,12008-12014.

4. Hoflack, B., Fujimoto, K. & Kornfeld, S. (1987) J. Biol.Chem. 262, 123-129.

5. Ziberstein, A., Snider, M. D., Porter, M. & Lodish, H. F.(1980) Cell 21, 417-421.

6. Dixit, V. M., Hennessy, S. W., Grant, G. A., Rotwein, P. &Frazier, W. A. (1986) Proc. Nati. Acad. Sci. USA 83,5449-5453.

7. Hanahan, D. & Meselson, M. (1983) Methods Enzymol. 100,333-342.

8. Maniatis, T., Fritsch, E. F. & Sambrook, J. (1982) MolecularCloning: A Laboratory Manual (Cold Spring Harbor Labora-tory, Cold Spring Harbor, NY).

9. Ullrich, A., Berman, C. H., Dull, T. J., Gray, A. & Lee, J. M.(1984) EMBO J. 3, 361-364.

10. Rigby, P. W. J., Dieckmann, M., Rhodes, C. & Berg, P. (1977)J. Mol. Biol. 113, 237-251.

11. Yanisch-Perron, C., Vieira, J. & Messing, J. (1985) Gene 33,103-119.

12. Sanger, F., Nicklen, S. & Coulson, A. R. (1977) Proc. Natl.Acad. Sci. USA 74, 5463-5467.

13. Mizusawa, S., Nishimura, S. & Seela, F. (1986) Nucleic AcidsRes. 14, 1319-1324.

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

15. Aviv, H. & Leder, P. (1972) Proc. Natl. Acad. Sci. USA 69,1408-1412.

16. Lipman, D. J. & Pearson, W. R. (1985) Science 227,1435-1441.

17. Dayhoff, M. O., Barker, W. C. & Hunt, L. T. (1983) MethodsEnzymol. 91, 524-545.

18. Grant, G. A., Sacchettini, J. C. & Welgus, H. G. (1982) Bio-chemistry 22, 354-358.

19. Grantham, R., Gautier, C., Gouy, M., Jacobzone, M. &Mercier, R. (1981) Nucleic Acids Res. 9, r43-r74.

20. Sabatini, D. D., Kreibich, G., Morimoto, T. & Adesnik, M.(1982) J. Cell Biol. 92, 1-22.

21. Sahagian, G. G. & Neufeld, E. F. (1983) J. Biol. Chem. 258,7121-7128.

22. Esch, F. S., Ling, N. C., Bohlen, P., Ying, S. Y. & Guillemin,R. (1983) Biochem. Biophys. Res. Commun. 113, 861-867.

23. McMullen, B. A. & Fujikawa, K. (1985) J. Biol. Chem. 260,5328-5341.

24. Kornblihtt, A. R., Umezawa, K., Vibe-Pedersen, K. &Baralle, F. E. (1985) EMBO J. 4, 1755-1759.

25. Hynes, R. (1985) Annu. Rev. Cell Biol. 1, 119-142.26. Goldstein, J. L., Brown, M. S., Anderson, R. G. W., Russell,

D. W. & Schneider, W. J. (1985) Annu. Rev. Cell Biol. 1, 1-39.27. Drickamer, K., Dordal, M. S. & Reynolds, L. (1986) J. Biol.

Chem. 261, 6878-6887.28. Steiner, A. W. & Rome, L. H. (1982) Arch. Biochem. Biophys.

214, 681-687.29. Fischer, H. D., Creek, K. E. & Sly, W. S. (1982) J. Biol.

Chem. 257, 9938-9943.30. Alvares, K. & Balasubramanian, A. S. (1983) Biochem.

Biophys. Res. Commun. 112, 398-406.31. Maler, T., Rosenblum, B. B. & Jourdian, G. W. (1985) Proc.

Natl. Acad. Sci. USA 82, 8379-8383.32. von Figura, K., Gieselmann, V. & Hasilik, A. (1985) Biochem.

J. 225, 543-547.33. Sahagian, G. G. & Steer, C. J. (1985) J. Biol. Chem. 260,

9838-9842.34. Drickamer, K., Mamon, J. F., Binns, G. & Leung, J. 0. (1984)

J. Biol. Chem. 259, 770-778.

Biochemistry: Lobel et al.

Proc. Natl. Acad. Sci. USA 84 (1987) 7523

Correction. In the article "Toxicity of folic acid analogs incultured human cells: A microtiter assay for the analysis ofdrug competition" by David S. Roos and Robert T. Schimke,which appeared in number 14, July 1987, ofProc. Natl. Acad.

Methotrexate (nM) )-0 50 100 150 200 250

*10- A

.. .....~~~ ~~~~~~~~~~~~~~~~~~~~~~.......

- 50-

X100-

Z 150-1 A|

BW301 U (nM)0 25 50 75 100 125

50-

Sci. USA (84, 4860-4864), the authors request that Figs. 3and 4 be reprinted. The quality of reproduction of thesefigures was poor in the published article. The two figures andtheir legends are shown below.

Folinic Acid (nM) - *-32 128 512 2000 8000

32

'-

Q),x 128-x

Z 512-

2000

Folinic Acid (nM) - )32 128 512 2000 8000

I I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

32-

100-

150- ... _..

FIG. 3. Two-dimensional multiwell assay of competition withMTX. XPA cells were grown in each well ofa 96-well microtiter platein the presence of increasing concentrations of MTX along thehorizontal axis plus increasing concentrations of either MTX (A) orBW301U (B) along the vertical axis. The isobolograms were derivedfrom densitometric scanning of microtiter plates after fixation andstaining with crystal violet. Decreasing densities of shading indicategrowth inhibition of 10%, 35%, 65%, or 90%. The convex pattern ofgrowth inhibition observed in mixtures of BW301U and MTXindicates that these two drugs are not direct competitors (seeDiscussion and Fig. 5).

Correction. In the article "Cloning of the bovine 215-kDacation-independent mannose 6-phosphate receptor" by PeterLobel, Nancy M. Dahms, James Breitmeyer, John M.Chirgwin, and Stuart Kornfeld, which appeared in number 8,April 1987, of Proc. Natl. Acad. Sci. USA (84, 2233-2237),the authors request that the following correction be noted. InFig. 2, on page 2234, the cytosine at position 1027 should bedeleted, and a cytosine should be inserted between nucleo-tides 1215 and 1216. The published sequence from position1021 to position 1220 TCGTTCCAAAAA ... GGCCC-CAAGCCGTGAA should read TCGTTCAAAAA . . . GGC-CCCAAGCCCGTGAA. The 1-base deletion and 1-base in-sertion changes the deduced amino acid sequence fromresidue 343 to residue 405.

FIG. 4. Folinic acid rescue of MTX and BW301U toxicity. XPAcells grown in the presence of increasing concentrations ofMTX (A)orBW301U (B) were supplemented with increasing concentrations offolinic acid. (Note that drug concentrations in this experiment varyon a logarithmic scale.) Folinic acid supplements were able tocompletely rescue cells from all concentrations ofMTX tested, whilethe toxicity of BW301U was only mildly affected by folate.

Correction. In the article "Purification and characterizationof yeast myristoyl CoA:protein N-myristoyltransferase" byDwight A. Towler, Steven P. Adams, Shad R. Eubanks,Derek S. Towery, Emily Jackson-Machelski, Luis Glaser,and Jeffrey I. Gordon, which appeared in number 9, May1987, of Proc. Natl. Acad. Sci. USA (84, 2708-2712), theauthors request that the following error be noted. Theyincorrectly referred to the peptide Gly-Ala-Arg-Ala-Ser-Val-Ser-Gly as corresponding to the amino-terminal sequence ofHTLV-III gag. The correct peptide, Gly-Ala-Arg-Ala-Ser-Val-Leu-Ser, has been synthesized. Its kinetic parametersare Km = 0.002 x 10-' M, Vm. = 34% of control.

128- >$

512-

2000 - -- ------ -

Biochemistry: Corrections

....F

Z,z"SzQ

cc