Myelin proteolipid protein gene structure and its regulation of expression in normal and jimpy mutant mice

.J. Mol. Biol. (1988)

199, 587-596

Myelin Proteolipiq Protein Gene Structure and its Regulation of Expression in Normal and Jimpy Mutant Mice Kazuhiro

Ikenaka’, Teiichi Furuichi2, Yasuno Iwasakil, Akira Moriguchi’ Hideyuki Okanol and Katsuhiko Mikoshiba1*2 1Division

of Regulation of Macromolecular Institute for Protein Research Osaka University, 3-2 Yamada-oka, Suita Osaka, 565, Japan

Function

2Division of Behaviour and Neurobiology Department of Biological Regulation National Institute for Basic Biology Aichi, 444, Japan (Received 20 August 1987) The mouse proteolipid protein (PLP) gene was cloned int’o the 3, bacteriophage Charon 4A. The organization and the nucleotide sequence of the exons of the mouse PLP gene were quite similar to those of their human counterparts, consisting of seven exons. The transcription of the PLP gene started from multiple sites. There was a unique sequence tandemly repeated four times, sharing homology with the herpes simplex virus DR2 sequence, upstream from the transcribed region. Expression of the myelin basic protein (MBP) is also restricted to the oligodendrocytes in the central nervous system as is the PLP expression. Homology search against the mouse MBP gene revealed that several boxes in the $-flanking region of PLP show a high degree of homology with the sequence present in the MBP 5’-flanking region, possibly of importance in the concomitant expression of both genes in the central nervous system. PLP-mRNA in jimpy mutant mice does not contain exon 5 and its content is greatly reduced. We analyzed the jimpy PLP-mRNA and showed that the transcription initiated from the same sites as those in normal mice. Cloning and sequencing of the $-flanking region of the jimpy PLP gene revealed that there were no mutations in the promoter region of the jimpy PLP gene. Therefore, it is likely that a mutation, presumably existing within the jimpy PLP gene, caused the skipping of exon 5 and directly affected the mRNA level.

1. Introduction Myelination is a process in which an oligodendrocyte extends its processes, recognizes axons, and surrounds them with its membrane. Thus myelination is a good model with which to study neuron-glia interaction. Moreover, several hereditary mutant mouse strains are known that are incapable of forming a normal myelin structure in their central nervous systems; namely, shiverer, myelin dejkient, twitcher, jimpy and myelin synthesis dejkient. These mutants have yielded much genetic information on the processes or substances necessary for myelination. Both myelin basic 002%2836/88/040587-10

$03.00/O

587

protein (MBP)f and proteolipid protein (PLP) are the major proteins present in myelin of the central nervous system, and their syntheses concomitantly increase at the active myelination stage. Through studies of the above mutants, it has been possible to speculate functions of MBP and PLP; MBP is believed to play an important role in making a compact myelin structure by causing adhesion between the inner membranes of the oligodent Abbreviations PLP, protoolipid chromatography; pairs.

used: MBP, myelin basic protein; HPLC, high-pressure kb, lo3 bases or base-pairs;

0

1988 Academic

protein: liquid bp, base-

Press Limited

588

K. Ikenuka

drocytic processes, while PLP is thought to feature in the maturation of oligodendrocytes (Hudson et al., 1987). The molecular mechanisms of the deficit in some of these mutants have recently been elucidated. The shiverer mutant mouse has a deletion in the MBP gene (Kimura et al., 1985; Roach et al., 1985), and jimpy cannot synthesize myelin PLP as a result of its unusual mRNA structure (Nave et aE., 1986; Moriguchi et aZ., 1987). In order to understand the molecular basis for neuron-glia interaction, we began to study the regulation of MBP and PLP gene expression. Here we report the cloning of the mouse PLP gene and show that the nucleotide sequences in the 5’-flanking region (including the promoter region) and the exons of the PLP gene are almost completely conserved upon comparison of homology between mouse and human, while those of the introns are quite different. We also describe the results of analysis of the 5’-flanking region of the jimpy PLP gene.

2. Materials

and Methods

(a) Animals

and reagents

Thejimpy mutant mice (BGCBA-jp/Y) (Sidman et al., 1964) were maintained in our laboratory.. Restriction enzymes, phage T4 DNA ligase, DNA polymerase large fragment (Klenow fragment), and other enzymes were obtained from Boehringer-Mannheim, West Germany, unless otherwise specified. Radioactive compounds were obtained from Amersham, England. (b) Cloning

of

the mouse PLP gene

The 1 bacteriophage Charon 4A library constructed from partially EcoRI-digested mouse chromosomal DNA was provided by K. Shimada (Kumamoto Univ., Japan). Rat PLP cDNA clone (~27) (Milner et al., 1985) provided by J. C. Sutcliffe (Research Institute of Scripps Clinic, CA, U.S.A.) was 32P-labeled by nick translation. Plaques (1.5 x 106) of the 1, phage library were screened with a 32P-labeled cDNA clone. Nylon filters (Biodyne, Pall Ultrafine Filtration Co.) were prehybridized for at least 4 h at 42°C in 50% (v/v) formamide, 5x Denhardt’s solution (0.02% polyvinylpyrrolidone, 0.02% Ficoll, 0.02% (w/v) bovine serum albumin), 5 x SSC (1 x SSC is 0.15 M-NaCl, 0.015 M-sodium citrate, pH 7.0), 0.1 y. (w/v) SDS, 250 pg heat-denatured herring testis DNA/ml, and 50 mxl-sodium phosphate buffer (pH 7.0) and subsequently hybridized with the probe at lo5 cts/min per ml for 16 h. Filters were washed 3 times in 2 x SSC, 0.1 y. SDS at room temperature for 5 min and then 3 times in 0.1 x SSC. 0.1% SDS at 50°C for 15 min. They were exposed against Fuji RX film at -70°C overnight with intensifying screens. The DNA fragments were subcloned into plasmids by the method of Maniatis et al. (1982). (c) DNA sequencing DNA fragments were subcloned into Ml3 mp18 or mpl9, and the single-stranded DNA was sequenced by the dideoxy chain-termination method (Sanger et al., 1977).

et al. (d) Preparation

of

oligonucleotides

Oligonucleotides were synthesized using the Beckman System 1 E Plus DNA synthesizer by the p-cyanoethyl phosphoramidite method. The protected oligonucleotides were removed from the resin and subsequently deprotected by 2Syb aqueous ammonia treatment. They were purified by 7 M-urea/polyacrylamide gel electrophoresis followed by column chromatography on DE52 (Pharmacia). The purity was examined by HPLC (column: Radial Pak C18, Waters). (e) Primer

extension

A synthetic oligonucleotide primer (primer 1 in Fig. 3) was 5’-end-labeled with [y-‘*P]ATP (5000 Ci/mmol) using T4 polynucleotide kinase and was annealed to a total mouse brain RNA prepared by the guanidinium/cesium chloride method (Maniatis et al., 1982), in 20 mi\l-Tris . HCl buffer containing 0.1 M-NaCl and 0.1 mM-EDTA (pH 7.5). The primer was extended with avian myeloblastosis virus reverse transcriptase (Life Science), in 50 mmTris . HCl buffer containing 6-m&lMgCl,, 40 mM-KCl, 1 mM-dithiothreitol, and 1 mm-dNTPs (PH 8% at 42°C for 30 min. The same 32P-labeled primer was also annealed to a single-stranded recombinant Ml3 phage, containing a Scar-Hind111 fragment (Fig. 6(c)), and the sequencing ladder was prepared by the dideoxy termination method (Sanger et al., 1977). The extended products were run on a 6% (w/v) denaturing polyacrylamide gel next to the sequencing ladder, and were exposed against Kodak XAR film with an intensifier screen at -70°C. (f) Ml3

T4 DNA polymerase primer

extension

T4 DNA polymerase primer extension was performed by the method of Hu & Davidson (1986). Briefly the total mouse brain RNA (25 or 5 pg) was hybridized to a singlestranded HindIII-ScaI fragment subcloned into Ml 3 mp18 (0.1 pg), in 50~1 of hybridization buffer (80% formamide, 40 mM-NaPipes, 0.4 M-Nacl, 1 mmEDTA, pH 6.4) at 37°C for 16 h, after heating at 80°C for 10 min. The hybrid was precipitated with ethanol, rinsed with 70% (v/v) ethanol, dried and taken up in 9 ~1 of T4 DNA polymerase buffer (67 mM-Tris. HCl, 16.6 ITIM-aITIUIOniUIn 10 mM-2-mercaptoethanol. sulfate, 6.7 mM-MgCl,, 6.7 PM-EDTA, 0.17 mg bovine serum albumin/ml, 0.2 mg heat denatured herring testis DNA/ml, pH 8.8). A 5’.endlabeled primer (1 x 10’ cts/min per pg) (primer 2 in Fig. 2(a)) was added to the mixture. which was then heated at 65°C for 10 min, followed by incubation at 37°C for 1 h: 4.5 units of T4 DNA polymerase (Takara,
RNase

mapping

RNase mapping was essentially performed according to the method of Melton et al. (1984). Briefly, the HindIII&a1 fragment in Fig. 6(c) was subcloned into pSP65, and 32P-labeled antisense RNA was generated by the action zf SP6 RNA polymerase in the presence of [c~-~‘P]UTP (600Ci/mmol). The RNA probe (2x lo6 cts/min) was hybridized to a total mouse brain RNA and unhybridized

Characterization

I K > I lATG 1 II ’ B

Bg

II i-----T

ITGA fr P

589

of the Mouse Proteolipid Protein

m r----T

I I S

I I PS

I(A)n

I

I 0.5

kb

Figure 1. Fragments of the rat PLP cDKA clone, p27 (Milner et al. 19&j), used as probes. Fragments I. IT and III, indicated in the Figure, were nick translated and used as probes to screen the mouse PLP genomic clones. B. Han~H1; Bg, QZII: P, PatI; S, &UT. A bold bar indicates the coding region.

RNA was digested with RNase A (100 ng) and RNase T, (20 units) at 17°C for 30 min. The RNase-resistant fragment was analyzed on a denaturing 6% (W/V) polyacrylamide gel.

3. Results (a) Cloning of the mwusePLP gene The mouse genomic library (1.5 x lo6 plaques) was screened with the 32P-labeled rat PLP-cDNA probe (~27) (Milner et al., 1985), and seven candidates were selected. The candidates were next screened using various parts of the rat PLP-cDNA as probes (illustrated in Fig. 1). In general, two patterns of hybridization were obtained; (a) one that hybridized only with the 5’-portion (base pairs l-418, probe I) of p27 and (b) the other, which hybridized with all portions (probe I, II and III) of the cDNA. As representatives, phage numbers 23 (pattern a) and 38 (pattern b) were used in the following experiments. Various parts of the phage DNA were subcloned into plasmids (mainly pSP64 or pSP65). The plasmids were digested with proper restriction enzymes and were subjected to an agarose gel electrophoresis (0.7 %, w/v). The samples were alkaline transferred to a nylon membrane filter (Zeta Probe, Bio-Rad) (Reed & Mann, 1985) and fragments carrying an exon (or exons) were detected by hybridization with 32P-labeled rat PLP-cDNA. 32P-labeled RNA was prepared from the plasmids carrying these fragments by the action of SP6 RNA polymerase and more detailed analyses on the location and the length of the exons were performed by RNase mapping. Finally, the DNA sequence of the region that should contain an exon was determined and the exact location of the exon was determined through comparison with the DNA sequence of the rat PLP cDNA (Milner et al., 1985) (Figs 2 and 3). Thus, ;1 23 was found to carry exon 1, whereas ;1 38 carries exon 2 through exon 7. Since 2 23 and A 38 did not contain an overlapping DNA region, our clones did not contain some portions of the first intron of the mouse PLP gene. Absence of an exon within this region was confirmed by sequencing the PLP-mRNA by primer extension. Tn this experiment, an oligonucleotide primer complementary to a portion of the

Figure 2. Restriction maps of the mouse PLP genomic clones, I 23 and I 38. Mouse chromosomal DNA was partially digested with EcoRI and ligated into a Charon 4A 1 phagr vector. 1 23 and 138 hybridized to a rat PLP cC)KA probe and were shown to be thp mouse PLP genomic clones. Open squares represent exons (I to VII). BarnHI( EcoRI(E), HindIII(H), and PstI(P) sites are indicated. The BumHI fragment, shown by (H). was also cloned from the jimpy genomic library.

second exon was synthesized, hybridized to a total mouse brain RNA and extended by reverse transcriptase in the presence of a proper dideoxynucleotide (data not shown). DNA sequence of the mouse PLP cDNA (Hudson el al., 1987) was also consistent with our result. (b) Determination of the transcription initiation

site

Primer extension analysis was carried out to determine the transcription initiation site of the mouse PLP gene. An oligonucleotide primer (primer l), shown in Figure 3, was 5’-end-labeled with 32P, annealed to a total mouse brain RNA (Fig. 4, lane I, 10 fig; lane II, 50 pg) and extended by reverse transcriptase. The same primer was also utilized to generate a sequencing ladder, which was run next to the extended products on a 6% (w/v) sequencing gel. The multiple bands in Figure 4, suggested that there were multiple transcription initiation sites in the mouse PLP gene; five major sites, indicated by arrow heads, forming three clusters. There were also other minor primerextended products. When RNA was omitted from the reaction mixture, there were no extended products (data not shown). To confirm whether the multiple bands observed on the gel truly reflected multiple transcription initiation sites, they were once more ma.pped by primer extension using T4 DNA polymerase. In this experiment, a HindIII-ScaI fragment containing the first exon (seeFig. 6(c)) was subcloned into Ml3 mp18, and the single-stranded DNA, capable of hybridizing with the PLP-mRNA, was prepared. An oligonucleotide primer (primer 2 in Fig. 3) was annealed to the single-stranded DNA, which had already been hybridized with the total mouse brain RNA. The primer was then extended by T4 DNA polymerase. The primer extension reaction is blocked by the 5’ end of the PLP-mRNA, since T4 DNA polymerase does not catalyze strand displacement synthesis (Masamune & Richardson, 1971). Five extended products were observed, as in the case of the usual primer extension, indicating that

Exon

1 and

5'-Flanking

Region

cagctggttc

tatctgcatt

cttctaaaac

aaagaatgct aggcttttga tttgggagga

ttttttgctt ttcagacccc ttcaagaacc

gaaaagagag

gaagaaatta

ttttaaatga

ctttttttct

tccccattgt

gtttccagtg

ccaggaagag

-473

acagaggaaa cttctcatca cctccattta

caaasctttt Hind111 ggaaaggttc gggctactat atttacaccc

catggtcaag ttcacatgac ctaattcaca

ggcaacgagc

agtgagagtt

gggtgcggtg

tgtttggtag

tatagtaagt

tttacatgct

cagacccagg

tatgacacat

cttcctgatt

tatttaaagc

aaaatgaaat

ttaaatggac tctagagaag

ccaaggatca ctttaggggg

-373 -273 -173

agaaagaaaa

aaaacaattg

ggagtgaaaa

ggcataaaga

gaagatggag

cccttaaaga

agggagtatc

ccaaaggagg

cwgaggagaa

wgwggaw

agaggaggw

gqaaacqagc

ctgtctcttt

aagggggttg

gctgtcaatc

+l ag*GCCgTTTCATT

AcAAAGATA&

TUXGAGAGAA

AAAGTAAAGG

Primer ACAGAAGMG

2 GAGACTI’AG

AACAGGCTAC

MTTGGAGTC 5CCTCAG

AGAGTGCCM TCTCACGGTT

AGACATGGgt TPlcPBr _____.._

Primer

Exon

Hind111

r CA

tggggacaag

-73

GGAGAAGAGG

+20

AGACCAGGAT CCTTCCAGCT BamHI

GAGCMAGTC

AGCCGCMAA

CAGACTAGCC

+128

gagttcaaaa

actccagcat

caaagatgca

ggcacaggag

ttcaactttg

gggctttggg

+166

MetG

1

2

cacactctgt

gcttggtaac

atgggctgct

tggcccagca

gtctagtgtg

agtggatgag

ttacctcgta

tgcgctacct

gactttctcc

ttcttcttcc

1yLeuL ccagGCTTGT

euGluCysCy TAGAGTGTTG

sAlaArgCys TGCTAGATGT

LeuValGlyA CTGGTAGGGG

laProPheA1 CCCCCTTTGC

aSerLeuVa1 TTCCCTGGTG

AlaThrGlyL GCCACTGGAT

euCysPhePh TGTGTTTCTT

eGlyValAla TGGAGTGGCA

LeuPheCysG CTGTTCTGTG

1yCysGlyHi GATGTGGACA tctcaccact

sGluAlaLeu TGAAGCTCTC cagcta

ThrGlyThrG ACTGGTACAG

luLysLeuI1 AAAAGCl'AAT

eGluThrTyr TGAGACCTAT

PheSerLysA TTCTCCAAAA

snTyrGlnAs ACTACCAGGA

pTyrGluTyr CTATGAGTAT

LeuIleAsnV CXCATTAATG

a Tgtaagtacg

tgctctctga

ctggattttc

tgtctgtcca

tgcagGATTC

ATGCITTCCC

nTyrValIle GTATGTCATC

TyrGlyThrA TATGGMCTG

1aSerPhePh

ataaggttat

CCTClTTCTT

ePheLeuTyr CTTCCTTTAT

GlyAlaLeuL GGGGCCCTCC

euLeuAlaG1 TGCTGGCI'GA

UGlyPheTyr

ThrThrGlyA

nIlePheGly GATCTTTGGC

hrThrIleCy CCACCATCI'G

sGlyLysGly CGGCAAGGGC

LeuSerAlaT

ACCACCGGCG

laValArgG1 CTGTCAGGCA

AspTyrLysT

GGGCTTCTAC

hrValThrG1 CGGTAACAGG -7-Em oAspLys

yGlyGlnLys GGGCCAGAAG

GlyArgGlyS

nHisGlnAla ACATCAAGCT

HisSerLeuG CA-

1uArgValCy

GGGAGGGGTT

erArgGlyG1 CCAGAGGCCA

AGCGGGTGTG

sHisCysLeu TCATTGTTTG

GlyLysTrpL GGAAMTGGC

euGlyHisPr TAGGACATCC

CGACAAGgtg

atcatcctca

ggattttgtg

gcaataataa

ggggtggggg

acaattggga

gtgagtctgt

agcctgatcc

ccacccaagg

ttgggtcctc

tgtatgctga

tttttaacca

ctccatgtca

attgttttag

PheValGlyI TlTGTGGGCA

leThrTyrA1 TCACCTATGC

aLeuThrVa1 CCTGACTGTT

ValTrpLeuL GTATGGCTCC

euValPheA1 TGGTGTWGC

aCysSerAla CTGCTCGGCT

ValProValT GTACCTGTGT

yrIleTyrPh

eAsnThx.Trp CMTACCTGG

ThrThrCysG ACCACCTGTC

lnSerIleA1 AGTCTATTGC

aPheProSer CTTCCCTAGC

LysThrSerA MGACCPCTG

laSerIleG1 CCAGTATAGG

ySerLeuCys

ACATTTACTT

CAGTCTCTGC

AlaAspAlaA GCTGATGCCA

rgMerTyrG GAATGTATGg

tgagttgaat

gtggga

gccttatgaa

gtttactctg

gctgctttta

tgtatcttag

1yValLeuPr GTGTTCTCCC

ATGGAATGCT

PheProGlyL TTCCCTGGCA

Excm

l?xon

3 1IleH

isAlaPheG1

GACTACAAGA

5 ggcctcta

ysValCysG1 AGGTTTGTGG

OTrpAsnAla

CPGAGCGCAA-

ySerAsnLeu CTCCAACCTT

LeuSerIleC CTGTCCATCT

ysLysThrA1 GCAAAACAGC

aGlu TGAGgtaagt

g atgcgaag

agatgctttt

taaaaggata

gattggctag

acatggagg

ttttctgttc

taagaaataa

ttctctctca

tacatcttct

PheGl tgcagTXCA

nMerThrPhe

ttgtgcttgc

AATGACCI’TC

HisLeuPheI CACCTGTTTA

1eAlaAlaPh TTGCPGCGTT

eValGlyAla TGTGGGTGCT

AlaAlaThrL GCGGCCACAC

euValSerLe TAGTTTCCCT

u Ggtaagttat

tttaagataa

tattagaaaa

gaagtggtcc

agggatagca

ttaggccgaa

agaCtagCag

agagactcct

Le uThrPheMet ccctacagCT CACCTTCATG TTGI'CTAATA GCAAGGm GAMGGAGAG TClTGCAGTG GGGATGCCl'G AGMGGfGAC ACTTCCAACT

IleAlaAlaT ATTGCTGCCA AACCACACAG

hrTyrAsnPh CTTACAACTT

eAlaValLeu CGCCGTCCTT TGTGTTTTAA TCTCCCTCTT

LysLeuMetG

1yArgGlyTh GCCGAGGCAC GCCACTGATT

rLysPhe*** CAAGTTCTGA GGcccTcTTc lYX?l'CCACAG

AGTCMATAC ATATAGCAlT GAC%ATTTM

ATT’ TGCCCAAATC

ATGCACMM

CCMGMCTC

GCCCTAACK

AG0ccMCCl-T

TTATAGCl'U AGiATAGAAA

TTAGGMGAG GGMCTAGCT

MCAGTGTT

GCMGATACT -G AGMAGACCT CCCATGAMT ~~MTM CACATAATAG TfX2CAGATl.l’ AAAGMMGG CTGATCGAAT GAATTTACM ACCMTCMG ACACACACAC GATACTTTTA

TMGGTTTTATCC GGAGGATGAT TA TTGGGMT AT~~AUU* ATTCI’GC CAG CAt3GCTTATC MGGAAGGM lTCCT@AQG GMGACACTT ATCMGGAAG At!ACACACAC GMACTTTTA

tgagtgaatt

gttctat

Exon

Exon

6

7

cttaccttct GCTCCCATAG

tttctctgtt MACTCCCCT

TI’AClTGATG

AGTATAACAA CTAGMATGG

CAGGCTCCTG AGACCAGGAT CATCAGTAGG TAGCAGAGCA AGTGGCMTG

TCATCTGTCC CAAGGAGCTG TCGGGTMAT AATGGTGCTT GAGTTGTATC ATTGTACGGT ~cn~Tm=rcr ATATATAGGT CCTMAGMC AMGA,iAGAA AAMMGCCC AGAGATATM CATGTMGM TCTCACATGC CAACATAAAG tcgtgtcctt

G-

ttg

ATTMTCTCT

TCCCCAGCTG GAGGAGAGCA

TGCCTCC!TGC TATTTGMA GATAGCATGT ATTCACTCAT TCTTI’GAGCA TCCCCCMGT A~~~ACTTG CTGTTAGCTG CTGTAGGAMXAA~ MMGTTATT GAGACTTGTT

AGCTGGOAGA

CCTACGGCM GTCAGGATCA T MGhCTGTGA CPIWTCCCC

ACACMXTAC GMACTCATC

AGCAGGTGAT TAGTCTACCAT CTMTATFAT TTUGTTUT TAACTl%ZGG MGMM;uu GTAGCTGTTT CTTGGGCCM CTATA’IVXTT AMGAGAGTT TTACMMTG

acctcaataa

caggtactta

GTGAMTMT’

CCTACAGTGT cTCXG%GAC

CMGTCGCAG GMCGGAMG

AGGAGTGAAA

ACTCTTACCT ACGePOTCAA TTAGGGAGCA

AGCMGGATC

TCTUCTCCC CCTMGGAGC

@ZXXXWT TCCAGTGW

CACCCCTATC TAGACGWCT TGAMACAAG TTGIMAACT t23xmu~rr MGCTTCCAT TGMAGMAG Ghh&GhMGA CAT’lWhGAT TATILTMGGT ThTlTTGTTT MTChACTGC TAT?TTATAA

TTCTCCTTGA ATUGTTGGT CGCTAGTTCT TGATTAGTTT AAAGAGAAAC AGGCTTGhTA SA~W~TTTA AGAGCTCTGG AMGAMGM -W MAUAGCAG TGGTTTMTA M-C MTTMAGAA

attatagtta

gctcgag

Fig. 3.

AAACTCATGG

v AGTACCTCTT GCTCTAATTG AGAGAGCAGA ACTGAAMCA

TM'CGCCCTT ATAGACMAG TMCAGCTAC GGATGCTAM

GGCCCCTGGC CATA’ITTCCC TGGGCAGMA MTTTGMCC CGCTTACUT AGCCGCAGAG AGAMGhAhG S AT-l’CCCCMA GATGCACMG TMCAGATAA ‘ITCATTTCTG &‘&#&hTTA

TTAG TCATTT A TTTXCMG AATGTCCCTG GAAGMMTA AGAMGMAG -AG

CATCTCCTGA

CA%iACAhcC GGTGTM'lTi'

-C MMGM%C CTClT%ffiCc GB m;FGc;mc GAAA'I'GGC~

CT-ATA MlTGTl’TAT GCATCACAM TCMT‘I’TGM AGMGTWY AGGAUGGCA MAGNW%h B AGTGAGGAGA CCMCCATTC ATAGhGMTT AU&-

ATATCTTCAC CTcTTGTTTG MTATTTGM SAA-T GAATTl’GMA AGAMGMAG 6B CFGwjTTcM GCTMMGGA CCTTAAAMA TTGhTi-l’TM Caatcaaaca

Characterization

of the Mouse Proteolipid

t

Protein

591

GCT AAT

*

CTf TCT TCT CT TT

TTA CfT

t

Figure 4. Primer extension analysis of the mouse PLP-mRNA. A synthetic oligonucleotide primer (primer 1 in Fig. 3) was annealed to total mouse brain RNA (lane I, 10 pg; lane II, 50 pg) and was extended by AMV reverse transcriptase. The extended products were run next to the sequencing ladder on a denaturing 6% polyacrylamide gel, and the arrow heads indicate the extended products. The nucleotide sequences are shown on the right-hand side of the gel and the underlinings indicate the nucleotides to which the primer extended. there mouse to be Figure

are indeed multiple initiation sites in the PLP gene (Fig. 5(a)). The sites were mapped identical or very close to the sites shown in 4. However, when the most frequently used

initiation site was mapped, there was a consistently small difference (9 nucleotides) between the initiation sites mapped by the two methods (see Figs 3, 4 and 5(b)). The nucleotide sequence of this region of

Figure 3. Nucleotide sequence of the mouse PLP gene. The nucleotide sequence of the exons and their flanking regions are shown. Bold capital lett,ers represent the nucleotide sequences in exons and small letters show the sequences of introns and 5’-flanking region. The 4 tandemly repeated sequences are surrounded by boxes, and the TATA-like and CAAT-like sequences are indicated by dotted underlines. The transcription initiation sites determined by primer extension analysis (primer 1) and T4 DXA polymerase primer extension (primer 2) are shown by arrows from closed circles (+) (see Fig. 4). The most frequently used initiation sites mapped by the 2 methods did not coincide with each other, and both sit,es are indicated by arrows from open circles (@-+, primer extension; O-+, T4 DNA polymerase primer extension). Numbers indicate the relative position from the most upstream transcription initiation site. A putative splice site of DM20-mRNA is underlined in exon 3. The polyadenylation site of 3.2 kb PLP-mRNA is indicated by an arrow head in exon 7. and its possible polyadenylation signal by a double line.

592

K. Ikenaka

et al.

01X

AC

C

T

A G

/

C

1

P

T

/ /

A

/

C

/

T

/ /

/

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

Figure 5. Ml3 T4 DNA polymerase primer extension analysis of the mouse PLP-mRNA. (a) Total mouse brain RNA (lane I, 5 pg; lane II, 20 gg; lane 0, no RNA) was hybridized with the single-stranded genomic DNA (Scu-Hind111 fragment in Fig. 6(c)). A synthetic oligonucleotide primer (primer 2 in Fig. 3) was annealed to the hybrid and then extended by T4 DNA polymerase. The extended products (shown by arrow heads) were analyzed on a denaturing 8%) polyacrylamide gel. The length of the extended products can be estimated by the length marker shown on the left-hand side of the gel. (b) Precise mapping of the transcription starting sites. The extended products were run next to thr sequencing ladder on a 6% polyacrylamide gel. Nucleotide sequences are shown on the right-hand side of the gel and the underliningsindicate the nucleotidesto which the primer extended.

RNA is CUUUUCAUU, and therefore, rich in U residues. Perhaps this contributed to a loose double strand formation and T4 DNA polymerase did not terminate at the proper position. (c) PLP expression in jimpy mutant mice Expression of the PLP gene is impaired in the jimpy mutant mouse. Its mRNA does not contain the fifth exon (Moriguchi et al., 1987) and the level of PLP-mRNA is also greatly reduced (Dautigny et al., 1986; Gardinier et al., 1986). At present, the nature of the mutation at the nucleotide level is still unknown. To determine the transcription initiation site of the jimpy PLP gene, we attempted to analyze the jimpy PLP-mRNA by primer extension. We were able to detect the most frequently

used initiation site, which was conserved in the jimpy mice. However, we were unable to detect other extended products, probably due to their low abundance in the jimpy brains (data not shown). In order to increase the sensitivity, RNase protection analysis of the first exon of the ‘impy PLP gene was performed using uniformly 3!IP-labeled antisense RNA as a probe. A HindIII-ScaI fragment. containing the first exon and its flanking region (Fig. 6(c)), was subcloned into pSP65 and the antisense RNA was produced by the action of SP6 RNA polymerase in the presence of [~x-~*P]UTP The probe was hybridized to total brain RNA from normal mice (Fig. 6(a) and (b); lane I, 1 pg; lane II, 5 pg; lane III, 20 pg) or jimpy mice (lane J, 20 pg) and was digested with RNase T, and RNase A. The RNase-resistant products were analyzed by electro-

Characterization (01

of the Mouse Proteolipid

MPmlIIOJ I

I

Protein

593

(bi

I///

461

342 M

PXltIOJ

I27

127

\

(cl

r

Hindm

t

Figure 6. RKase mapping of the first exon of mouse PLP-mRNA. (a) A HindIII--8caI fragment (c) of the i phage 23 was subcloned into pSP65 and an antisense RNA probe was produced. The RKA probe was hybridized to total brain RKA from a normal mouse (lane I, 1 pg; lane II, 5 pg; lane III, 20 pg; lane 0, no RNA) or from ajl:mpy mouse (lane J, 20 /rg), and digested with RNases. The RNase-resistant fragments were analyzed by electrophoresis on a So/b sequencing gel. Lane P shows the degradation products of the probe and lane M shows molecular weight markers. (b) The gel used in (a) was over-exposed and the area of interest was magnified. RNase-resistant bands in lane J can be observed. (c) The HindIII-ScaI fragment used in the experiment is shown: Boxes represent. exon 1 of the mouse PLP gene and putative t,ransrription initiat,ion sites are indicated by arrows (g).

phoresis on a denaturing 6% (w/v) polyacrylamide gel. Figure 6(a) clearly shows that there were multiple protected products, and that there were no differences in the pattern between the RNaseresistant products from normal and the jimpy RNA (Fig. 6(b)). Thus the transcription initiation sites were completely conserved in the jimpy mice. The reduced mRNA level might be caused by a mutation in a region of DNA important for PLP transcription. We therefore tried to clone the 5’.flanking region of the jimpy PLP gene. In order to improve the efficiency of the molecular cloning,

1 he fragment of interest was enriched by the f allowing strategy. First, chromosomal DNA isolated from jimpy liver was digested wit#h EcoRI and fractionated by column chromatography on Sephacryl S-1000. A portion from each fraction was spotted onto a nitrocellulose filter membrane and fractions containing the PLP first exon were detected by hybridization with 32P-labeled wilda portion type 1.8 kb BamHI fragment, containing of PLP exon 1 and its upstream region (see Fig. 2). The hybridized fractions were pooled and then further digested with BamHI. The DNA fragments

594

K. Ikenaka

were subjected to agarose gel electrophoresis and a part of the gel was used to detect the band containing the jimpy 1.8 kb BamHI fragment by Southern transfer followed by hybridization with the wild-type 1.8 kb BamHI fragment. The corresponding area was excised from the agarose gel. The DNA fragments were electro-eluted, and ligated into pSP65 digested with BamHI followed by bacterial alkaline phosphatase treatment. The ligation mixture was used to transform Escherichih coZi HBlOl, and the resultin colonies (1500 l-8 kb colonies) were screened with the “P-labeled BamHI fragment. Three colonies hybridized with the probe and the plasmids were recovered from t’hem. The restriction map of one of them showed a perfect match with the wild-type clone, and thus, we successfully cloned the 1.8 kb BamHI fragment shown in Figure 2 from the jimpy chromosomal DNA. A 355 bp Hind111 fragment and a 269 bp HindIII-BamHI fragment (see Fig. 2), which contain the 5’-flanking region of the jimpy PLP gene, were subjected to DNA sequencing, but no mutations were found (data not shown).

4. Discussion (a) Structure

of the mouse PLP gene

In this paper, we report the molecular cloning and sequencing of the mouse PLP gene, which has been reported to be present as a single gene (Dautigny et al., 1986; Nave et al., 1986). Upon comparison with the sequence of mouse PLP-cDNA (Hudson et al., 1987), we showed that the PLP gene consists of seven exons. The organization of the exons, and the DNA sequences of the exons and the exon-intron junction points, are well conserved upon comparison of homology between mouse and human (Diehl et al., 1986), while sequences within the intron are quite different from each other.

(b) Transcription the mouse

initiation site of PLP gene

Transcription of the PLP gene was shown to start from multiple sites, by primer extension analysis (Fig. 4), T4 DNA polymerase primer extension (Fig. 5), and RNase mapping (Fig. 6). The exact position of the initiation sites are indicated in Figures 3 and 4, although the positions of the most frequently used sites detected by the two methods were inconsistent with each other; primer extension analysis predicted a C residue at position + 7, while T4 DNA polymerase primer extension predicted a G residue at + 16. (Note that the cap site is one nucleotide from the T4 DNA polymerase primer extension termination site.) There were no common sequences found upstream of each initiation site. A possible TATA and CAAT box were found at position -26 and - 10, respectively (indicated by dotted underlining in Fig. 3), although we cannot, find the TATA box corresponding to the CAAT

et al.

1: GTGGGGACAAG 2: GGGAGGAGAAG 3: GGGAGGAGGAG 4: AGGAGGAGGGA Commus: (GGGAGGAGaAG) n mamrnm=~

13132:

GGGAGGAGCG;

Figure 7. Homology between the 11bp repeat in the region of the mousePLP gene and the DR2 sequence of the herpes simplex virus genome. The nucleotidesequence of each repeated box (1 to 4) and the

upstream

consensus sequence of the 4 are given together nucleotide sequence of the DR2 box.

with

the

box. The unique feature of the sequence of the 5’-flanking region of the PLP gene is that it has four tandemly repeated 11 bp sequences (indicated as a box in Fig. 3). A homology search against GenBank revealed that the repeated sequence shared high degree of homology (9/11) with a herpes simplex virus repeated sequence (DR2), which is believed to be involved in the cleavage and packaging reactions of the virus (Varmuza & Smiley, 1985; Chou & Roizman, 1985) (Fig. 7). At present, however, the function of this sequence is not known.

(c) Homology with the mouseMBP gene The upstream sequence of the transcription initiation site, including the promoter sequence, is well known to play an important role in the regulation of gene expression. Recently, the nucleotide sequence of the 5’.flanking region of the mouse MBP gene has been determined, and the regions necessary for MBP transcription have been mapped (Miura et al., unpublished results). PLP and MBP are expressed specifically in the oligodendrocyte in the central nervous system and the developmental changes in their expression are also quite similar. When the nucleotide sequences of the 5’-flanking regions of both genes were compared, four homologous regions were found, as shown in Figure 8. The distance between each box and the transcription initiation site of the PLP gene were approximately the same as the distance between the corresponding homologous box and the initiation site of the MBP gene. One of them was homologous to the repeated sequence found upstream from the PLP gene. However, the overall hdmology was quite small in the region we compared the DNA sequence (715 bp and 572 bp upstream from the MBP and PLP transcription initiation sites, respectively), although it is possible that the element determining the tissue specificity does not lie in this region. Studies are currently underway to search for the elements determining the tissue specificity for both PLP and MBP gene expression.

Characterization

PLP: hmp:

-463

-292

-[AATc~CET/ . . . . . . . . {AATGCTTT v

w

-453

-293

. . . . . .

-151

of the Mouse

. . . . .

-63

+ACAATTGGGAJ+AGGGGAGGA/-+ACAAT~~~~AHAAGG~E~ -191

-49

Figure 8. Nucleotide

sequenceshomologousbetween

the 5’.flanking regions of the PLP and MBP genes. The nucleotide sequences of the 5’-flanking regions of the PLP and MBP genes were aligned, and the homologous sequences were searched. An open circle indicates a deletion in the sequence. The numbers indicate the relative position from the transcription initiation site. In the case of PLP. the most upstream starting point was c-hosen as + 1,

(d) Polyadenylation mousePLP

site of the gene

Northern blot analysis of mouse PLP-mRNA shows three speciesof PLP-mRNA in mouse brain; 3.2 kb, 2.4 kb and 1.6 kb in size (Dautigny et al., 1986; Gardinier et al., 1986; Nave et al., 1986). The 1.6 kb mRNA is present in a trace amount. These are known to be produced by alternate usage of polyadenylation sites. The DNA sequencesaround the polyadenylation sites for 2.4 kb and 1.6 kb mRNA were shown by Hudson et al. (1987)? who have cloned and sequenced the mouse PLP-cDNA. Their sequence, however, did not show the polyadenylation site for the most abundant 3.2 kb PLP-mRNA. Through homology with the 3’ end of rat 3.2 kb PLP-cDNA (~27) (Milner et al., 1985), we were able to deduce the polyadenylation site, as indicated by an arrow head in Figure 3. A possible polyadenylation signal is found 21 bp upstream from the polyadenylation site (indicated with a double underline in Fig. 3). (e) Structure

of mouse DM20-mRNA

DM20 is also a proteolipid found in the normal central nervous system together with PLP, and it has been shown to have a high degree of structural similarity to PLP (Lees & Brostoff, 1984). Recently, Morello et al. (1986) and Hudson et al. (1987) suggested that DM20-mRNA was produced by alternative splicing of the PLP-mRNA precursor and that a part of the PLP-mRNA was deleted to form DM20-mRSA. The 3’ end of the deleted portion corresponded in this study to the junction point of exons 3 and 4 of PLP-mRNA, while the 5’ end of the deletion started in the middle of exon 3. A sequence ACG/GTAAC was found in the exon 3 (underlined in Fig. 3) where the deletion apparently begins. It is homologous to the consensus5’ splice junction sequence (C or A)AG/GTRAG, where R represent’s purine and i shows the splicing site (Ohshima & Gotoh, 1987). Moreover, the reading frame

would

not

shift

even

if the

splicing

event

Proteolipid

Protein

595

were to occur between the putative 5’ splice junction of DM20 RNA and exon 4. An oligonucleotide complementary to the putative junction region of DM20 exons 3 and 4 (5’-TGCCCACAAACGTTGCGCTC-3’) was synthesized, 5’ end-labeled by 32P and used as a probe for Northern transfer analysis of total brain RNA from normal mouse, fractionated on formaldehyde/ 1.5o/o agarose gel. After autoradiography, bands showing the same mobility as the PLP-mRNA were detected (data not shown). Therefore, it is highly likely that an alternate usage of the splice donor sites is responsible for producing the DM20-mRNA. (f) Analysis of the jimpy PLP promoter region

As an approach to identify the &s-acting element necessary for PLP expression or responsible for determining the tissue specificity of PLP expression, we analyzed the 5’-flanking region of the jimpy PLP gene, since it is well known t,hat PLP expression is reduced in the jimpy brain (Dautigny et aE., 1986; Gardinier et al., 1986). The transcription initiation sites of the jimpy PLP gene were conserved and there were no mutations within the DNA fragment sequenced, including the 5’-flanking region and a part of exon 1 (a 624-bp HindIIIBamHI fragment). Oligodendrocytes are known to degenerate in the jimpy brain and, therefore, it is possible that the reduced transcription of the PLP gene in jimpy is merely caused by a reduction in the number of cells which produce PLP. Studies on the localization and quantitation of PLP-mRNA in the jimpy brain by in-situ RNA hybridization, currently

in progress

in our laboratory,

should

clarify

this point. After this work had been submit,ted, cloning of the DM20 cDNA was reported (Nave, K.-A., Lai, C., Bloom, F. E. & Milner, R. J. (1987). Proc. Nat. dead. Sci., U.S.A. 84, 5665-5669.) Our result was completely consistent with their results. This study was performed with the aid of Special Coordination Funds from the Science and Technology Agency of the Japanese Government. This study was also supported by a grant from the National Center for Nervous, Mental and Muscular Disorders of the Ministry of Health and Welfare, and a Grant-in-Aid for Scientific Research from the Ministry of Education, Japan. The authors thank Dr J. Gregor Sutcliffe for providing us with the rat PLP-cDPU’A clone, ~27, and Dr Kazunori Shimada for providing the mouse genomic library for use in our study. We also thank Mr Kunitatsu Izumi for synthesizing the oligonucleotides and Mr Tetsushi Kagawa for his technical assistance. References Chou, J. & Roizman, B. (1985). Cell, 41, 803-811. Dautigny, A., Mattei, M.-G., Morello, D., Alliel, P. M., Pham-Dinh, D., Amar, L., Arnaud. D., Simon, D., Mattei, J.-F., Guenet, J.-L., JollBs, P. & Avner. P. (1986). *Vature (London), 321, 867-869.

596

K. Ikenuka

IXehl,

H.-J., Schaich, M., Budzinski, R-M. & Stoffel, W. (1986). Proc. Nut. Acod Sk., U.S.A. 83, 9807-9811. Gardinier, M. V., Macklin, W. B., Diniak, A. J. & Deininger, P. L. (1986). Mol. Cell. Biol. 6, 3755-3762. Hu, M. C-T. & Davidson, pi. (1986). Gene, 42, 21-29. Hudson, L. D., Berndt, J. A., Puckett, C!., Kozak, C. A-& Lazzarini, R. A. (1987). Proc. Nat. Acad. Sci.. U.S.A. 84, 1454-1458. Kimura, M., Inoko, H., Katsuki: M., Ando, A., Sato. T., Hirose, T., Takashima, H., Tnayama, S., Okano, H., Takamatsu, K., Mikoshiba, K., Tsukada, Y. & Watanabe, I. (1985). J. Neurochem. 44, 692-696. Lees, M. & Brostoff, S. W. (1984). In MyeZin, pp. 197224, Plenum Press, New York. Maniatis, T., Fritsch, ‘E. F. & Sambrook, J. (1982). Editors of Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, TJY. Masamune, Y. & Rechardson, C. C. ( 1971). J. Biol. Chem. 246, 2692-2701. Melton, D. A., Krieg, P. A., Rebagliati, M. R., Maniatis, T., Zinn, K. 6 Green, M. R. (1984). Nucl. Acids Res. 12, 70357056. Edited

et al. Milner, R. J., Lai, C.. Nave, K-L., Lenoir, D.. Ogata. J. & Sutcliffe. J. G. (1985). Cell, 42, 931-939. Morello, D., Dautigny, A., Pham-Dinh. D. & JollBs. P. (1986). EMBO J. 5, 3489-3493. Moriguchi, A.. Ikenaka, K.. Furuichi, T., Okano. H.. Iwasaki. Y. & Mikoshiba, K. (1987). Gene, 55. 333337. Nave. K-L., Lai, C.. Bloom, F. E. & Milner. R. J. (1986). Proc. Tat. Acad. Sci.. U.S.A. 83, 92649268. Ohshima, Y. & Gotoh, Y. (1987). .J. Mol. Biol. 195, 247. 259. Reed. K. C. & Mann, 1). A. (1985). Nucl. Acids Res. 20. 7207-722 1. Roach, A., Takahashi, N;., Pravtcheva, D., Ruddie, F. & Hood. L. (1985). Cell, 42, 149-155. Sanger, F., Kicklen, S. & Coulson, A. R. (1977). Proc. Nat. Acad. Sci., U.S.A. 74, 5463-5467. Sidman. R. T,., Dickie, M. M. & Appel, S. H. (1964). Science. 144, 309-311. Varmuza, S. L. & Smiley, J. R. (1985). Cell. 41; 793-802.

by K. Matsuhara