Isolation of a novel gene underlying batten disease, CLN3

Cell, Vol. 82, 949-957, September22, 1995,Copyright© 1995by Cell Press

Isolation of a Novel Gene Underlying Batten Disease, CLN3 The International Batten Disease Consortium*

Summary Batten disease (also known as juvenile neuronal ceroid lipofuscinosis) is a recessively inherited neurodegenerative disorder of childhood characterized by progressive loss of vision, seizures, and psychomotor disturbances. The Batten disease gene, CLN3, maps to chromosome 16p12.1. The so-called 56 chromosome haplotype defined by alleles at the D'16S299 and D16S298 loci is shared by 73% of Batten disease chromosomes. Exon amplification of a cosmid containing D16S298 has yielded a candidate gene that is disrupted by a 1 kb genomic deletion in all patients carrying the 56 chromosome. Two separate deletions and a point mutation altering a splice site in three unrelated families have confirmed the candidate as the CLN3 gene. The disease gene encodes a novel 438 amino acid protein of unknown function.

*The International Batten Disease Consortium is comprised of the following groups:

Group 1: Terry J. Lerner, ~,2t Rose-Mary N. Boustany, 3 John W. Anderson, ~ Kenneth L. D'Arigo, ~ Karen Schlumpf, ~ Alan J. Buckler, ~,2 James F. Gusella, ~,4 and Jonathan L. Haines 1,2 ~Molecular Neurogenetics Unit Massachusetts General Hospital Charlestown, Massachusetts 02129 2Department of Neurology 4Department of Genetics Harvard Medical School Boston, Massachusetts 02114 3Division of Pediatric Neurology Duke University Medical Center Durham, North Carolina 27710

Group 2: Gabriel Kremmidiotis, Ingrid L. Lensink, Grant R. Sutherland, and David F. Callen Department of Cytogenetics and Molecular Genetics Women and Children's Hospital Adelaide SA 5006 Australia

tCorrespondence should be addressed to Terry Lerner.

Introduction

The neuronal ceroid lipofuscinoses (NCLs) are a group of inherited neurodegenerative disorders characterized by the accumulation of autofluorescent lipopigments (ceroid and lipofuscin) in neurons and other cell types (Dyken, 1988). At least five subtypes are recognized, based on age of onset, clinicopathological features, and chromosomal location. Inheritance is autosomal recessive for the childhood onset forms, which include infantile (CLN1; HaltiaSantavuori disease, MIM256730), classical late infantile (CLN2; Jansky-Bielschowsky disease, MIM204500), juvenile (CLN3; Batten or Spielmeyer-Vogt-SjSgren disease, MIM304200), and Finnish-variant late infantile (CLN5; MIM256731). The primary biochemical defects in these disorders are not known. Batten disease, the juvenile onset form of NCL, is the most common neurodegenerative disorder of childhood. Its incidence is estimated at up to 1 in 25,000 births (Zeman, 1974), with an increased prevalence in the northern European populations. Clinical onset begins with visual

Group 3: Peter E. M. Taschner, 1 Nanneke de Vos, 1 Gert-Jan B. van Ommen, 1 and Martijn H. Breuning~,2 1Department of Human Genetics Leiden University 2333 AL Leiden The Netherlands 2Department of Clinical Genetics Erasmus University 3015 GE Rotterdam The Netherlands Group 4: Norman A. Doggett, Linda J. Meincke, Zi-Ying Liu, Lynne A. Goodwin, and Judith G. Tesmer Life Sciences Division and Center for Human Genome Studies Los Alamos National Laboratory Los Alamos, New Mexico 87545 Group 5: Hannah M. Mitchison, 1 Angela M. O'Rawe, 1 Patricia B. Munroe, ~ Irma E. J~irvel~i, ~,2 R. Mark Gardiner, ~ and Sara E. Mole 1 1Department of Pediatrics University College London Medical School The Rayne Institute London WC1 E6JJ England 2Present address: National Public Health Institute SF-00300 Helsinki Finland

Cell 950

failure between the age of 5 and 10 years. Seizures and mental deterioration ensue with relentless decline to death, usually in the second or third decade. Diagnostic criteria include the presence in many cell types of inclusions that appear as so-called fingerprint profiles on electron microscopy (Wisniewski et al., 1988). The major protein component of these abnormal deposits is subunit 9 of mitochondrial ATPase (Palmer et al., 1992), although the genetic defect does not lie in a gene encoding this 75 amino acid protein (Dyer and Walker, 1993; Yan et al., 1994). The loci for several NCL subtypes have been mapped by linkage analysis to different chromosomal regions. The gene for Batten disease, CLN3, was assigned to human chromosome 16 by demonstration of linkage to the haptoglobin locus (Eiberg et al., 1989). A collaborative effort has allowed progressive refinement of the genetic localization of CLN3 (Gardiner et al., 1990; Callen et al., 1991; Yan et al., 1993; Mitchison et al., 1993; Lerner et al., 1994) with the aim of isolating the defective gene. Analysis of recombinations using a high resolution map of microsatellite marker loci has defined D16S288 and D16S383 as flanking markers and has thereby confined CLN3 to an interval of 2.1 cM in 16p12.1-11.2 (Mitchison et al., 1994). Within this candidate region, four microsatellite loci, D16S288, D16S299, D16S298, and SPN, display strong linkage disequilibrium with CLN3 (Mitchison et al., 1993, 1994, 1995; Lerner et al., 1994), and analysis of haplotypes on disease chromosomes suggests that CLN3 lies closest to D16S299 and D16S298 (see Figure 1). A core 56 chromosome haplotype, based on the 5 allele at D16S299 and the 6 allele at D16S298, is present on 73% of disease chromosomes and on only 3% of control nondisease chromosomes. The 6 chromosome allele at D16S298 is present on 96% of Finnish Batten disease chromosomes, suggesting that CLN3 is close to this marker (Mitchison et al., 1995). This is supported by the finding that two unusual patients have apparent deletions of D 16S298: a Moroccan patient (NCL39.3) from a consantel

guinous relationship homozygous for the absence of D16S298 as part of a cytogenetically undetectable microdeletion (Taschner et al., 1995a) and a Finnish patient (L199Pa) hemizygous for D16S298 (Mitchison et al., 1995; this paper). We now report the identification of the Batten disease gene, defined by cDNA (clone 2-3), from a cosmid containing D16S298. The gene is disrupted by a small (1.02 kb) deletion on all Batten disease chromosomes with the 56 chromosome haplotype and by the independent deletions in the Finnish and Moroccan patients mentioned above. In another unrelated Batten disease family, we have identified a point mutation that segregates with the disease and alters an invariant base of a 5' donor splice site. The Batten disease gene, CLN3, possesses an open reading frame of 1314 bp encoding a predicted 438 amino acid protein product with no significant similarity to previously described proteins. The discovery of this disease gene will permit direct genetic diagnosis of Batten disease and represents an initial step to understanding the pathogenesis of this disorder. Results Isolation and Characterization of cDNA2-3 Previously, we described a cosmid walk from D16S48 in Los Alamos contig c343.1 (Stallings et al., 1992) toward D16S299, yielding the 185 kb contig cNL/343.1 (Taschner et al., 1995a) (Figure 1). Cosmid NL11A, the most distal cosmid in this contig, was found to contain the D16S298 locus and to encompass the region deleted in the Moroccan patient NCL39.3. These findings implicate NL11A as the most likely site of CLN3. Thus, in this study, we targeted this cosmid for saturation with candidate Batten disease gene transcripts. A 180 bp exon isolated from N L l l A by exon amplification (Buckler et al., 1991; Church et al., 1994; Lerner et al., 1995) was used to screen a fetal brain cDNA library (Stratagene), yielding a 1.7 kb cDNA (clone 2-3). The reten - . ~

$299

$298

$48

150 kb

125 kb

Cy21Bl I Cy302GI2 Cy85D3

NLI IA i ....................................................... D16S298

i "............................................

i

I

eDNA2-3

Figure 1. Schematic Representation of the CLN3 Candidate Region on Chromosome 16p12.1 Adapted from Jfirvelfi et al. (1995a, 1995b). The positionsof selected DNA microsatellites used for linkageand haplotypeanalysisare indicated. Individual cosmids (NL11A and NL6OD3)of cosmid contig CNL/343A, which contains D16S298and D16S48, and cosmid contigC182,whichcontainsD16S299,are indicated. Three yeast artificial, chromosomes (Cy21B11, Cy302G12, and Cy85D3)that form part of a 980 kb contig spanningthe candidate region are indicated by horizontal lines. Pstl restriction fragments of NL11A that could be ordered are shown. The genomic extent of cDNA2-3 is shown below the map, with the arrow indicating the direction of transcription.

Isolation of Batten Disease Gene 951

1 2

3456

78

Kb 9.57.44.4

m

2.4

m

1.35 -

Figure 2. Northern Blot Analysisof cDNA2-3 cDNA2-3 was hybridized to a Northern blot (CIontech) containing 2 I~gof poly(A)÷RNA per lanefrom heart (lane 1), brain (lane2), placenta (lane 3), lung (lane 4), liver (lane 5), skeletal muscle (lane 6), kidney (lane 7), and pancreas(lane8). The filter was washed at 50°C in 0.1 x SSC and exposed for 24 hr. An - 1.7 kb transcript can be seen in all the tissues tested. suits of Southern blot and polymerase chain reaction (PCR) analyses of genomic and cosmid DNAs showed that the 1.7 kb cDNA is contained in NL11A and that transcription proceeds toward D16S299 (Figure 1). PCR amplification of individual Pstl fragments of NL11A using the D16S298 microsatellite primers placed D16S298 on a 1.3 kb Pstl fragment previously shown to be contained within the deletion of the Moroccan patient (Taschner et al., 1995a). This fragment is not detected by cDNA2-3, but consists of intron sequences mapping between bases 1193 and 1194 (see Figure 3) of the cDNA (H. M. M., unpublished data). Thus, cDNA2-3 spans the D16S298 locus and overlaps with the deletion found in the Moroccan patient. Northern blot analysis using cDNA2-3 as probe revealed a 1.7 kb transcript in poly(A) mRNA isolated from a wide variety of human tissues, including heart, brain, placenta, lung, liver, skeletal muscle, kidney, and pancreas (Figure 2), consistent with the cDNA clone being nearly full length. The transcript is not detected by Northern blot analysis in cultured lymphoblasts and in fibroblasts, but is detectable by reverse transcription PCR (RT-PCR) analysis of poly(A) mRNA isolated from such cell lines. A so-called zoo blot containing genomic DNAs from several animal species showed that this gene is conserved in mammals. Strong signals were obtained from mouse, sheep, cow, and pig (data not shown).

Sequence Analysis of cDNA2-3 Figure 3 shows the sequence of the cDNA that spans 1689 bp and has a 47 base poly(A) tail. A predicted open reading frame of 1314 bp begins with a potential initiator ATG codon at base 138 and ends with a TGA termination codon at base 1452. An in-frame stop codon is located 36 bases upstream of the initiator site, and a consensus polyadenyl-

ation site is located at base 1666. The predicted product of the cDNA is a protein of 438 amino acids with a molecular mass of 48 kDa. The cDNA sequence was compared against the GenBank and dbEST databases using BLASTN (Altschul et al., 1990) and FASTA (Pearson and Lipman, 1988) sequence alignment algorithms. These searches revealed no significant similarities to genes of known function. However, near identity (>95% similarity) was found to 13 expressed sequence tags (ESTs) (F11432, F12401, T74504, T08995,

R12998, Z42735, T47968, D20292, T47969, T97772, F09095, F10019, and T61330) isolated from five independent cDNA libraries (infant brain, fetal spleen, fetal liver and spleen, adult liver, and promyelocyte cell line HL60). Three pairs of ESTs (F11432/F09095, F12401/F10019, and T47968/T47969) are 5' and 3' sequences of three cDNA clones. Five ESTs (Fl1432, F12401, T74504, T08995, and R 12998), all isolated from a normalized infant brain cDNA library (Soares et al., 1994), are missing bases 184-262 of cDNA2-3 (Figure 3). If the same initiator ATG is used, this transcript is expected to produce a truncated protein of only 27 amino acids and, thus, is unlikely to be the result of normal RNA splicing. The physiological significance of this variant is unclear, since its relative abundance may be exaggerated by preparation of the normalized cDNA library. The predicted protein sequence was compared against the SwissProt database using BLASTP and Smith-Waterman (Smith and Waterman, 1981) sequence alignment algorithms and against the predicted translation products of GenBank database using TBLASTN. In all these cases, no significant similarities were found to known proteins. A search of the BLOCKS database (version 8.0; Henikoff and Henikoff, 1994) for motifs found only single blocks of homology for any group of proteins, and this could be attributed to chance. A search for protein motifs in the CLN3 protein using the ProSite Database (version 12.2) revealed pattern matches for four N-glycosylation sites, two glycosaminoglycan attachment sites, two cAMP- and cGMP-dependent protein kinase phosphorylation sites, six protein kinase C phosphorylation sites, eight casein kinase II phosphorylation sites, and 12 N-myristoylation sites (Figure 3). Hydropathy calculations (Kyte and Doolittie, 1982) predicted five hydrophobic regions that may be potential membrane-spanning regions at amino acids 3861, 93-233, 278-310, 345-399, and 408-438.

The Common Mutation Is A Small Deletion To screen for possible deletions, insertions, and other chromosomal rearrangements associated with the clone 2-3 gene, we scanned conventional Southern blots of restriction-digested DNA from unrelated Batten disease patients. Figure 4a shows the results of hybridizing a panel of Pstl-digested patient DNAs with PCR probes P1-P3 and P2-P5, representing the 5' and 3' halves of the cDNA, respectively. When the P1-P3 fragment was used as probe, affected individuals homozygous for the D16S299/ D16S298 56 chromosome haplotype (Figure 4a, lanes 1-3) displayed the loss of a 3.8 kb Pstl fragment and the gain of a novel 2.8 kbp fragment. When the P2-P5 frag-

Cell 952

i 61

CCCCTAGAC~GCCGGAGC~GGACCGGC~GGGCG~A~C~ACC~GCA GACC~A~CC~CCG~GGAGCCCCCTT~GACAC~TA~ACCC~GACCC~GG

121 1

GAC~CT~A~GA~GGAGGC~CAGGC~GCGGCGGCGCTT~GGA~CG M G G C A G S R R R F ## %

181 16

AG~GA~AGACCG~CCGGAGCCCCGGC~CC~GACCA~AGG~GCGCA~ G E E T V P E P R L P L L D H Q G

S

D

A

S

H

E

W

2 4 1 G G ~ G ~ C G C G G ~ G C ~ C ~ G G C C q ~ F T G C ~ C ~ C ~ A ~ G ~ A 56 K N A V G F W L L G L C N N F S Y V V M # .......... 301

~C~AG~CCGCCCACGACA~C~AGCCAC~GAGGACA~GGG~CCAGAGCCA~

361 76

TGGACCCAGGCCCAACGCCGATCCCCC ACAAC AGCTCATC ACGATTTGACTGCAACTCTG D P G P T P I P H N S S S R F D C N S V ___% . . . . . ^

421 96

T C T C T A C G G C T G C q~3T G C T C C T G G C G G AC A T C C T C C C C AC AC TC G TC A T C A ~ C P G T ~ G S T A A V L 5 A D I L P T L V I K L L A

481 116

CTCCTCT~fGG C Cq?TC A C C TG C "CGCCCTAC A G C C C C C G G G I ? F C T C G T C A G T G G G A T q ? r G T G P L G 5 H L L P Y S P R V L V S G I C A %

541 136

C'I~C'IX~G,~-G C ' ~ GTCCTGGT'IX;CC~ ' I ~

601 156

T G G T C T T C G C T A G C A T C T C ATC AGG CC T T G G G G A G G T C A C C T T C C T C T C C C T C A C T G C C T V F A S I 8 S G L G E v T F L S L T A F

661 176

T C T A C C C C A G G G C C G T G ATC T C C T O G TG G T C C TC A G G G ACqX3G G G G A G C T G G G C T G C T G G Y P R A V I S W W 5 S G T G G A G L L G ...~ ...... #

721 196

C~CCCTG A L

781 216

CCA'I~C'IN3GGTATCCC'I~CCCTGC'I~C'I'GGCC A ~ TATTTCT'I~T~C'I~ACA'I~TCCTG

841 236

A G G C C C A G G A C C C T G G A G G G G A A G A A G A A G C A G AG A G C G C A G C C C G G C A G C C C C T C A T A A A Q D P G G E E E A E S A A R Q P L I R

56

~ ~ ~ ~ . ~ i ~ ~ , ~ ~ • ~ ~ ~__Q_~__~ v

A

M

G

S

L

F

V

L

V

A

F

S

A'FI'C'IX3TGG GGACCAGCC'I~TG~ H

S

v

G

T

S

L

C

T C C T A C C T G G GCC TC AC C C AGG C C G G C C T C T C C C C T C A G C A G A C C C T G C T G T S Y L G L T Q A G L S P Q Q T L L S

G

I

P

A

L

L

L

A

S

Y

F

L

L

L

T

S

P

256901 G A A C C GEA G GAC C CPC G GEA G TSC G AKA G C PC A ~ C T CSC A GSC T C SC A G S CCTC L F C C q^' TLC G GRG A AEA G GRT T %4

961 276 1021 296

1081 316

1141 336

G G AC AGTAq'FC A A G G G T C T G C TG T G G T AC A'VFGTTCCCT'£GG T C G T A G T T T A C T ~ C C G T V F K G L L w Y I V P L V V V Y F A E AGTAT~ Y F

A ~ T A A C C AG G G A C q ' T T ~ AAC T C C T C ~ T G G I N Q G L F E L L F F w

AAC A C T T C C C T G AG TC N T S L S H

ment was used as probe, no difference was detected between controls and the homozygous 56 chromosome haplotype affected patients. Analysis of 148 control chromosomes, including seven with the 56 chromosome hapIotype, revealed no alterations. The affected individuals bearing a 56 chromosome also displayed altered fragments with Hindlll and Pvull digestion (data not shown), suggestive of a small ( - 1000 bp) genomic deletion of the 56 chromosome. Figure 4b illustrates the Mendelian inheritance of this deletion in a two-generation Batten disease family segregating the 56 chromosome haplotype. To determine the effect of this genomic deletion on the clone 2-3 transcript, we performed PCR amplification of RT cDNA from patients homozygous for the 56 chromosome haplotype. The P1-P3 primer pair yielded a novel product that was - 2 0 0 bp smaller than that predicted from the cDNA sequence and that was found in all non56 chromosome normal controls (Figure 4c). RT-PCR amplification with the P2-P5 primer pair yielded identical - 800 bp products in affected patients and controls. DNA sequence analysis of the P1-P3 product from five homozygous 56 chromosome patients showed in all cases a 217 bp deletion, from base 598 to base 814 of the cDNA (see Figure 3). The DNA sequence of the RT cDNA from four control individuals revealed no evidence of deletion, matching the cDNA2-3 sequence. Deletion of these 217 bases of coding sequence produces a frameshift, generating a TAA termination codon 84 bp downstream of the deletion junction. The predicted translation product is a truncated protein of 181 amino acids consisting of the first 153 residues of the protein followed by 28 novel amino acids before the stop codon. DNA sequence analysis of the genomic fragment containing this deletion from a 56 chromosome homozygous patient revealed the loss of 1.02 kb of genomic sequence (Figure 5a). The intron sequence immediately 5' to the deletion is 91% homologous to bases 84-290 of the AluSx family consensus sequence in the 5' to 3' orientation. The A-rich sequence at the 3' end of this Alu sequence includes a (GA)n repeat within the 5' deleted segment. The sequence at the 3' portion of the deleted region is 87% homologous to bases 1-290 of the Alu-Sx sequence, also in the 5'to 3' orientation, and contains a (GA)o repeat within the A-rich sequence of the 3' tail. Included in this deletion are 217 bp of the open reading frame (base pairs 598814; see Figure 3), corresponding to two exons.

ACGCTCAGCAATACCGCTGGTACCAGATGCTGTACCAGGCTGGCGTCTTTGCCTCCCGCT CT'PCTCTCCGCTGCTGTCGCATCCGTTTCACCTGOGCCCTGGCCCTGCTGCA~CCTCA A

Q

Q

Y

R

W

Y

Q

M

L

Y

Q

A

G

V

F

A

S %

L

R

C

C

R

I

R

F

T

w

A

L

A

L

L

Q

S

R

S

L

N

1201 356

ACCTGGTGTTCCTGCTGGCAGACGTGTGGTTCGGCTTTCTGCC AAGCATCTACCTCGTCT L V F L L A D V W F G F L P S I Y L V F

1261 376

T C C T G A q ~ C A T ' £ C T G T A T G A G G G G C T C C T G G G A G G C G C A G C C T A C G T GA A C A C C T T C C A C A L I I 5 Y E G L L G G A A Y V N T F H N

1321 396

AC A T C G C C C "~3G A G A C C AGTGAq~3AG C A C C G GG AG Tq'TGCA A T G G C G G C C A C C T G C A T C T I A L E T S D E H R E F A M A A T C I S

1381 416

CqhGAC A C A C TG G GG A T C T C C C T G T C G G G G C T C C TG G C T'CTGC C T C T G C A T G ACT~fCCTCT D T L G I S L S G L L A L P L H D F L C

1441 436

G C C A G C q ~ T C C T G A T A C T C G G G A T C C T C A G G A C G C A G G T C AC ATTC ACCq~GTGGGC AG A G Q L S

1501

G G A C A G G T C A G A C A C C C A G G C C C A C C C C A G A G A C C C T C C AqTGAACTGTG'CTCCCAGCCTT

1561

C C C G G C A G G T C T G G G A G T A G G G A A G G G C T G AAG C C T T G T T ~ C qTfGCA G G G G G G C C A G C C

1621

AT~fGTCT C C C A C T ~ G G G AG T ~ T T C

C T G G C A TC A T G C C T T C q~ A A T A J ~ A T G C C G A T ~ $$$$S$

1681

TG~DCC A

G

T

G

~

Figure 3. The NucleotideSequenceof cDNA2-3 The predictedprotein is shown below the DNA sequence, assuming that translationbeginsat the first in-framemethionineof the longopen readingframe.FourpotentialN-linkedglycosylationsitesare indicated by a broken line at residues49, 71, 85, and 310 (Bause, 1983). Two potentialglycosaminoglycansites are indicatedby the dotted lines at residues162 and 186 (Bourdonet al., 1987).PotentialN-myristoylation sites are indicatedby a numbersign (Grand, 1989). Serineand threonine residuesthat are potentiallyphosphorylatedby cAMP-and cGMPdependentprotein kinases(indicatedby a percent sign) (Glasset al., 1986), by protein kinaseC (indicatedby an asterisk)(Woodgettet al., 1986), or by casein kinase II (indicatedby a caret) (Pinna, 1990) are shown. The polyadenylationsite at base 1666 is indicatedby a dollar

Screening for the 56 Chromosome Deletion PCR amplification of genomic DNA with primers F2 and P3 flanking the cDNA deletion produces a 3.5 kb product from normal chromosomes and a 2.5 kb product from the chromosomes with the 1.02 kb deletion described above. sign. cDNA sequencesdeletedin the 56 chromosomedeletion(bases 598-814) and the L199Pa deletion (bases 928-1193) are shown in boxes. The positions of forward PCR primers used in these studies are as follows:P1, bases39-58; F2, bases552-569; F4, bases 676695; F5, bases778-797; P2, bases860-879;/=9, bases888-907; GF1, bases1470-1487. The positionsof the reverseprimersare as follows: R1, bases 658-637; P3, bases 880-861; R5, bases 1246-1229; R3, bases 1612-1595; GR1, bases 1661-1643; P5, bases 1669-1651.

Isolation of Batten DiseaseGene 953

1

2

3

4

I

I

I

I

5 I

1

2

I

I

7' '°'~':- ~%

kb

3

4

I

l

We tested for the presence of the 1.02 kb deletion associated with the D16S299/D16S298 56 chromosome haplotype in 81 unrelated Batten patients representing 24 haplotypes and originating from 16 countries (Table 1). Of these, 46 were homozygous for the 56 chromosome haplotype, 24 were heterozygous for the 56 chromosome haplotype, and 11 did not carry the 56 chromosome haplotype on either chromosome. In all 70 patients with an affected 56 chromosome, the 2.5 kb fragment was detected, and in all 46 homozygotes for this haplotype, no normal size product was produced. Smaller numbers of chromosomes bearing closely related haplotypes (66, 36, 46, 57, and 55) also carried this deletion, suggesting that these chromosomes most probably derived from the 56 chromosome haplotype by mutation of the polymorphic marker or recombination. Additional affected chromosomes bearing the 66 and 46 chromosome haplotypes apparently possess mutations independent of the 56 chromosome, as they do not carry this deletion. Thus, the 1.02 kb genomic deletion of the clone 2-3 gene associated with the 56 chromosome haplotype is the most common mutation in Batten disease, accounting for 81% of disease chromosomes tested to date.

5 I

' • m

& 1.50

--

1.10

--

0•65

--

' -

'

-

b

kb

....

3,80

--

2.80

--

1,95

-

~D i- •-

-

Other Mutations Disrupting the Gene C 1

2

3

4

5

6

7

8

I

I

i

i

i

i

i

I

9 i

10

11

12

I

i

bp

--

800

--

600

--

800

Figure 4, Analysisof cDNA2-3 in Batten DiseasePatients (a) Southernblot analysisof cDNA2-3 in Battendisease patients.Genomic DNA was digestedwith Pstl, subjectedto agarosegel electrophoresis, and transferred to Hybond N÷ membrane (Amersham). Lanes 1-3, patients homozygousfor the 56 chromosomealleles;lane 4, unaffectedcontrol; lane 5, patient heterozygousfor the 56 chromosome alleles(see text for data on patient L199Pa).The left panelwas probed with a radiolabeledPCR fragment amplified from cDNA2-3 using primersP1-P3, correspondingto the 5' half of the cDNA. The right panel was probed with a radiolabeledPCR fragment amplified from cDNA2-3 using primers P2-P5, correspondingto the 3' half of the cDNA. (b) Southernblot analysisof cDNA2-3 in a familywith an affectedchild homozygousfor the 56 chromosomehaplotype.Southern blots were prepared as described above and hybridizedwith the P1-P3 probe• The affected individual is shown by the closed symbol and carriers by the half-closedsymbols.Progenyare shown by diamonds,and the birth order of some individualshas been changed for confidentiality. The chromosomessegregating in this pedigree have been distinguishedby extensivetypingwith polymorphicmarkersin 16p12.1-11.2 (Lerner et al., 1994).

Haplotype analysis of Finnish patient L199Pa revealed one 56 chromosome and one @null chromosome, exhibiting absence of any D16S298 allele (for clinical details see Experimental Procedures). Southern blot analysis of this patient (see Figure 4a, lane 5) revealed two alterations: the 1.02 kb deletion typical of the 56 chromosome haplotype and a second deletion present on the chromosome missing D16S298 that results in the formation of a novel 1.5 kbjunction fragment. This junction fragment combines sequences from an upstream 1.1 kb Pstl fragment detected by the cDNA probe and from a Pstl fragment 3' to D16S298 that contains only intron sequence. PCR analysis of patient DNA using the i ntron primer IntR 14 and cDNA primer F9 confirmed an - 3 kb deletion, including the entire 1.3 kb Pstl fragment containing D16S298 (Figure 5b). RT cDNA from this second mutant allele was selectively amplified using primer R5 and primer FS, which is deleted on the 56 chromosome. The amplified product revealed the absence of 266 bp of coding sequence between bases 928 and 1193 of the cDNA (see Figure 3), generating a TGA termination codon 84 bp downstream of the deletion junction. The predicted translation product is a truncated protein of 291 amino acids consisting of the first 263 amino acids of the protein followed by 26 novel amino acids before the stop codon. Partial DNA sequence analysis of the genomic fragment containing this - 3 kb deletion has confirmed the loss of bases 928-1193 of the cDNA. (c) RNA analysis of cDNA2-3 in Batten disease patients. RT cDNA was preparedfrom cytoplasmicRNA and amplifiedusingeitherprimer pair P1-P3 (top)or P2-P5 (bottom).The PCR productswere fractionated on 1.5o/oagarose gels, transferred to Hybond N÷ (Amersham) membranes, and hybridized with the corresponding probe• Lanes 1-6, RNA from peripheral blood lymphocytesof control individuals; lane 7, water PCR control; lanes 8 and 10-12, RNA from fibroblasts of patienthomozygousfor the 56 chromosomehaplotype;lane9, RNA from fibroblastsof control individual•

Cell 954

Figure 5. Genomic Deletions of CLN3

a

'4,"'~'''~'~ "''~'''~'~'~. ~'~.~'~

~,.,~.~.., .o, .~,o,~.~,~.~,~.o.~,~,.,~.o.~.., .~,~'~

I~

Im.I

~,

.41

.,q

12

RI

P3

(a) Genomic deletion of 1.02 kb in disease chromosomes bearing the 56 chromosome haplotype. The complete sequence of the deletion is known, but only the sequences bordering the deletion are shown. The deletion covers two exons and the flanking intronic sequence and leads to the deletion of 217 bp of coding sequence. Positions of primers F2, R1, and P3, used to delineate the deletion, are indicated. Hatched boxes represent exons. The deletion breakpoints are shown by the arrows, and deleted sequences are shown in italics. The positions of flanking Alu-Sx sequences are indicated. (b) Schematic representation of the genomic deletions of the CLN3 gene. Position of primers used to delineate the deletions are indicated.

b ls~o ,I

7 °

CLN3

t

I

I

I

Pstl

HIndEII

Pstl

Pstl

Hind[ P3 f9 F4

PZ

I BarnHI

IntR14 GR]

F5 "5 6" deletion NCL39,3deletion L199Padeletion

We previously described a homozygous deletion of the D16S298 locus in a Batten patient of Moroccan origin (NCL39.3). Although the size of the deletion was not determined, it did include the 1.3 kb Pstl fragment containing D16S298 (Taschner et al., 1995a), which has proved to Table 1. Deletion Analysis of Batten Disease Patients PCR Amplification Product 3.5 kb

Number of Chromosomes"

+

-

116

-

+

0

+

-

4

-

+

7

36

+

-

4

-

+

0

46

+

Haplotype

D16S199/D16S298

2.5 kb

56 66

65 67 57 55 Other haplotypes

2

-

+

+

-

1

1

-

+

0

+

-

1

-

+

0

+

-

2

-

+

0

+

-

1

-

+

0

+

-

-

+

1 22

Genomic PCR was carried out using primer pair F2-P3 at bases 552 and 880 of the cDNA2-3 sequence. PCR amplification was carried out as described in Experimental Procedures. a Total number of chromosomes was 162.

be within an intron of the clone 2-3 candidate gene. PCR amplification of genomic DNA with primers F2 and R3 yielded a 1.1 kb fragment instead of the expected - 7 kb fragment (Figu re 5b). Additional PCR amplifications using nested primers on either the 5' (F4-R3) or 3' (GF1-GR1) sides gave no product, suggesting a deletion in the Moroccan patient of - 6 kb that starts between F2 and F4 and ends between GF1 and R3. Single-stranded conformation polymorphism (Orita et al., 1989) was performed to scan the CLN3 gene for further mutations. Patient L198Pa (for clinical details see Experimental Procedures), who is heterozygous with one 56 chromosome and one 76 chromosome (D16S299/D16S298), exhibited a mobility shift in a 73 bp exon corresponding to bases 598-670 of the cDNA (data not shown). Nucleotide sequence analysis showed a G to C transition at +1 of the splice donor site following the exon. Analysis of the parents of patient L198Pa showed the father (haplotype 76•46) to be a heterozygous carrier of this mutation. Discussion

The results of haplotype analysis of 142 families suggested that the major Batten disease mutation occurred on an ancestral chromosome bearing the D 16S299/D 16S298 56 chromosome haplotype (Mitchison ~pt al., 1994). The data presented here provide compelling evidence that this mutation is a 1.02 kb deletion that implicates cDNA2-3 as the product of CLN3. This deletion involves the 3' end of two Alu-Sx elements and the following (GA)o sequence and may therefore have arisen by recombination involving

Isolation of Batten DiseaseGene 955

bordering Alu sequences, a mechanism for which other examples exist in human disease (e.g., Rudiger et al., 1995). The deletion mutation is found on all affected 56 chromosomes examined to date and on several chromosomes with related haplotypes, accounting for 81% of Batten disease chromosomes. Studies are in progress to characterize the mutations of the clone 2-3 gene found on the minority of disease chromosomes that do not carry the 1.02 kb deletion. With the notable exceptions of patients L199Pa and NCL39.3, Southern blot and long-range PCR analyses of patients with chromosomes lacking the 1.02 kb deletion have failed to reveal additional genomic rearrangements. These results suggest that these affected chromosomes most likely carry point mutations, small deletions, or regulatory mutations of CLN3, e.g., the altered splice site in patient L198Pa. The independent deletions in L199Pa and NCL39.3, both of which encompass the D16S298 microsatellite locus, provide the strongest confirmatory evidence that cDNA2-3 is the product of CLN3. Ultimate proof, however, that cDNA2-3 represents CLN3 will require the reversal of the disease phenotype in a cell or animal model system or the generation of knockout animals that exhibit a Batten disease-like phenotype. No homology has been found at the nucleotide or amino acid level with any previously identified gene, indicating that CLN3 encodes a novel protein of (as yet) unknown function. Diverse approaches may now be used to explore the normal physiological role of the CLN3 protein. For example, the conservation of coding sequences across species should allow the identification of homologous sequences and target conserved domains of functional significance. The presence of several potential phosphorylation sites suggests that the protein may undergo phosphorylation as a prerequisite for binding an additional protein(s). The PSORT program (version 6.3; Nakai and Kanehisa, 1992) for prediction of protein localization sites indicates that the CLN3 protein may be a type IIIb membrane protein (Heijne and Gavel, 1988), a possibility supported by hydropathy calculations that suggest the presence of several hydrophobic domains and by numerous potential N-glycosylation and N-myristoylation sites. The deletions identified to date are predicted to remove >100 amino acids from the C-terminal portion of the CLN3 protein, suggesting that its normal function would be severely compromised in the disease. However, it is also conceivable that the disease phenotype may involve abnormal accumulation of truncated protein products rather than or in addition to direct loss of protein function. The CLN3 gene is expressed not only in the brain, the site of massive neuronal cell death in Batten patients, but also in a wide range of tissues. Consistent with this, inclusion bodies have been found in many Batten disease tissues in addition to the brain. Palmer et al. (1992) demonstrated the abnormal accumulation of subunit 9 of mitochondrial ATPase in these inclusions, but experiments mapping the subunit 9 genes P1 and P2 to chromosomes 17 and 12, respectively (Dyer and Walker, 1993), and P3 to chromosome 2 (Yan et al., 1994) excluded these genes as the site of the Batten disease defect. It will now be of interest to

determine whether the CLN3 protein or fragments thereof also accumulate in the disorder. Similarly, various biochemical approaches have suggested that Batten disease involves perturbations in several metabolic pathways, including, for example, lipid peroxidation (Siakotos et al., 1988), metabolism of dolichol-linked oligosaccharides (Hall and Patrick, 1985), and lysosomal proteinase activity (Wolfe et al., 1987). Whether these diverse biochemical phenotypes are the result of the primary gene defect or are secondary effects of the disease process can now be examined. Because of the slow progression of symptoms in Batten disease and its similarity to other NCL subtypes and neurologic disorders, diagnosis is often missed or delayed. Current diagnostic protocols call for examination of skin biopsies for hallmark fingerprint profiles in inclusion bodies, a technically demanding procedure. Since the demonstration of linkage disequilibrium, carrier detection by haplotype analysis has been possible (Taschner et al., 1995b). The direct PCR assay for the 56 chromosome deletion described in this paper will imp.rove the reliability of the diagnosis for the majority of Batten disease patients and provide families with the opportunity for prenatal and carrier testing. The identification of the Batten disease gene is the initial step in understanding the pathology underlying this complex disorder. This study provides the basis for analyzing the role of the CLN3 protein in both normal and disease cells and provides a starting point for the design of rational therapies. Moreover, what we learn from studying the CLN3 protein may reveal the underlying cause of the other ceroid lipofuscinoses and provide insights into the mechanisms involved in other neurodegenerative disorders.

Experimental Procedures Patients and Cell Lines

Patients with Batten disease were identified through contacts with organizationsof volunteerparentsand through clinical referrals.Diagnoses were confirmed using standard criteria (Boustanyet al., 1988; Santavuori, 1988).The establishmentof lymphoblastoidcell lineswas previously described (Anderson and Gusella, 1984). Finnish patient L199Pa had a normal birth and early childhood.At age 6.5, he was referred to Dr. P. Santavuoriat the Universityof Helsinki Clinic and Children's Hospital becauseof failing vision. Electroretinogramwas abolished, and visual evoked potential was abnormal with delayed latency. Slight motor clumsinessand muscularhypotoniawere found. Vacuolated lymphocytes were positive on repeated examinations. From age 11, he had generalizedepileptic seizures that were well controlled by sodium valproate-clonazepam. An MRI at age 16 showedslight central, cortical, and cerebellaratrophy. The patient is still able to walk independently,but jumping has becomedifficult. He has finished school and is working in a daycare center. Finnish patientL198Pahad an uneventfulbirth and earlychildhood. Since the age of 7, she has experiencedprogressivevisual failure. At age 9, she showed an abnormal MRI. Vacuolated lymphocytes were repeatedlyobserved,and electron microscopyof a rectal biopsy specimenshowedinclusionstypical for Battendisease.She has been on sodiumvalproatemedicationsince the age of 9, when she experienced her only seizure. Recentexaminationat the age of 13 showed that her motor status is good but that her mental decline has been relativelyfast. The Moroccan patient has been previouslydescribed(Taschneret al., 1995a).

Cell 956

DNA Electrophoresis and Hybridization DNA extraction, restriction digests, electrophoresis, Southern blotting, hybridization, and washing were performed by standard methods (Sambrook et al., 1989). cDNA Screening and Characterization Exon amplification was carried out using the pSPL3 vector as described by Church et al. (1994). A human fetal brain cDNA library in Z7_.APII (Stratagene) was screened by standard methods using exon probes, cDNA clones and trapped exons were sequenced manually (Sanger et al., 1977) with Sequenase T7 DNA polymerase (United States Biochemicals). RNA Procedures Cytoplasmic RNA was isolated by standard methods (Sambrook et al., 1989) or using RNazol (Biogenesis). RNA was reverse transcribed using oligo(dT) or random hexamer primers and Superscript reverse transcriptase (GIBCO). Portions of the cDNA were amplified using primer sets described in the text. Direct sequencing of PCR products was carried out as described previously (McClatchey et al., 1992) or by purification with Qiagel (Qiagen) followed by sequencing with an Applied Biosystems (ABI) 373A automated sequencer. PCR products were subcloned using the TA cloning kit (Invitrogen). PCR PCR was carried out using Taq polymerase, following the recommendations of the manufacturer. The assay for the 56 chromosome deletion was carried out on 100 ng of genomic DNA using primers F2 and P3 in a reaction including 0.2 p.M of each primer, 0.2 mM of each dNTP, 1.5 mM MgCI2, and 0.5-1 I~1 of AmpliTaq (Perkins-Elmer). In one laboratory, the reaction was supplemented with 5 U of TaqExtender (Stratagene), which was found to enhance the amplification. Annealing temperatures ranging between 55°C and 62°C were used successfully. Samples were fractionated on an 0.8% agarose gel. Genomic Sequencing Genomic DNA from a normal control and the somatic cell hybrid CY101, which carries a single copy of chromosome 16 derived from a patient homezygous for the 56 chromosome hapletype, was PCR amplified with primers P1-P3. The resulting PCR products were digested with Taql. A 1.5 kb fragment was detected in the control, and a 0.5 kb fragment was detected in CY101, These two fragments were subcloned into pUC19 and sequenced with an ABI 373A automated sequencer, In an independent study, the sequence spanning the 56 chromosome deletion was generated by PCR sequencing of the subcloned 3.8 kb Pstl fragment using an AB1373A automated sequencer. Single-Stranded Conformation Polymorphism Analysis Intron primers IntF6 (5'-GGAGCCTCTATGAGCTGATACTG-3') and IntR12 (5'-GGAACATTCAGGAGGACCTAGG-3') were used to amplify patient DNA for single-stranded conformation polymorphism analysis (Orita et al., 1989). Each sample ( - 0 . 5 p.I) was loaded onto 12,5% homogeneous gels, and these were electrophoresed for - 2 0 0 vh at 15°C in a PhastSystem apparatus (Pharmacia). Gels were subsequently stained and fixed using the method suggested bythe manufacturer. The patient sample exhibiting a band shift was sequenced using the Taq dye deoxy terminator cycle sequencing kit and analyzed using the ABI 373A sequencer. Acknowledgments We are extremely grateful to the Batten disease families and their physicians for supporting and participating in this research effort. We would especially like to thank Dr. J. Alfred Rider (Children's Brain Disease Foundation) for his continuing encouragement and support of Batten disease research. We gratefully acknowledge the contributions to this work of Mary Anne Anderson, Heather MacFarlane, Patricia Crawford, Barbara Jenkins, Nicole Lawrence, Kathleen MacCormack, Richard Lee, and Isaac Cheng (Massachusetts General Hospital) for technical assistance; Jingmei Liu (Los Alamos National Laboratory) for technical assistance and Gouchun Xie (Los Alamos National Laboratory) for assisting in the analysis of the DNA and protein

sequences; and Keith Parker and Treasa Creavin (University College London Medical School) for technical assistance. We also thank Xandra O. Breakefield, Thomas Dooley, Marcy MacDonald, Jill Murrell, and Arthur C. Ley for their helpful discussions during the course of this project. This work was supported by National Institutes of Health (National Institute of Neurological Disorders and Stroke) grants NS32009 (T. J. L.), NS24279 (J. L. H., R.-M. N. B., and J. F. G.), NS30152 (M. H. B. and P. E. M. T.), and NS28722 (R. M. G.); by United States Department of Energy contracts W-7405-ENG-36 (N. A. D.) and DE-FG02-89-ER-60863 (G. R. S.); and by grants from the National Health and Medical Research Council of Australia, the Medical Research Council (United Kingdom), Action Research (United Kingdom), the Wellcome Trust (United Kingdom), the Research Trust for Metabolic Diseases in Children (United Kingdom), the Los Alamos National Laboratory, a Laboratory Directed Research and Development grant, and the Batten Disease Support and Research Association. A. M. O. is a Wellcome Trust research fellow. G. R. S. is an international research scholar of the Howard Hughes Medical Institute. Received July 7, 1995; revised July 31, 1995.

References Altschul S. F., Gish, W., Miller, W., Myers, E. W., and Lipman, D. J. (1990). BLAST: basic local alignment search tool. J. Mol. Biol. 215, 403-410. Anderson, M. A., and Gusella, J. F. (1984). Use of cyclosporin A in establishing Epstein-Barr virus-transformed human lymphoblastoid cell lines. In Vitro 20, 856-858. Bause, E. (1983). Structural requirements of N-glycosylation of proteins. Biochem. J. 209, 331-336. Bourdon, M. A., Krusius, T., Campbell, S., and Schwarz, N. B. (1967). Identification and synthesis of a recognition signal for the attachment of glycosaminoglycans to proteins. Proc. Natl. Acad. Sci. USA 84, 3194-3198. Boustany, R.-M., Alroy, J., and Kolodny, E. H. (1988). Clinical classification of neuronal ceroid lipofuscinosis subtypes. Am. J. Med. Genet. (Suppl.) 5, 47-58. Buckler, A. J., Chang, D. D., Graw, S. L., Brook, J. D., Haber, D. A., Sharp, P. A., and Housman, D. E. (1991). Exon amplification: a strategy to isolate mammalian genes based on RNA splicing. Proc. Natl. Acad. Sci. USA 88, 4005-4009. Callen, D. F., Baker, E., Lane, S., Nancarrow, J., Thompson, A., Whitmore, S. A., MacLennan, D. H., Berger, R., Cherif, D., J&rvel&, I., Peltonen, L., Sutherland, G. R., and Gardiner, R. M. (1991). Regional mapping of the Batten disease locus (CLN3) to human chromosome 16p12. Am. J. Hum. Genet. 49, 1372-1377. Church, D. M., Stotler, C. J., Rutter, J. L., Murrell, J. R., Trofatter, J. A., and Buckler, A. J. (1994). Isolation of genes from complex sources of mammalian genomic DNA using exon amplification. Nature Genet. 6, 98-105. Dyer, M. R., and Walker, J. E. (1993). Sequences of members of the human gene family for the c subunit of mitochondrial ATP synthase. Biochem J. 293, 51-64. Dyken, P. R. (1988). Reconsideration of the classification of the neuronal ceroid lipofuscinoses. Am. J. Med. Genet. (Suppl.) 5, 69-84. Eiberg, H., Gardiner, R. M., and Mohr, J. (1989). Batten disease (Spielmeyer-SjSgren disease) and haptoglobin (HP): indication of linkage and assignment to chromosome 16. Clin. Genet. 36, 217-218. Gardiner, R. M., Sandford, A., Deadman, M., Poulton, J., Cookson, W., Reeders, S., Jokiaho, I., Peltonen, L., Eiberg, H., and Julier, C. (1990). Batten disease (Spielmeyer-Vogt disease, juvenile onset neuronal ceroid lipofuscinoses) gene (CLN3) maps to human chromosome 16~ Genomics 8, 387-390. Glass, D. B., EI-Maghrabi, M. R., and Pilkis, S. J. (1986). Synthetic peptides corresponding to the site phosphorylated in 6-phosphofructo2-kinase/fructose-2,6,-biphosphatase as substrates of cyclic nucleotide-dependent protein kinases. J. Biol. Chem. 261, 2987-2993. Grand, R. J. A. (1989). Acylation of viral and eukaryotic proteins. Bio-

Isolation of Batten Disease Gene 957

chem. J. 258, 625-638. Hall, N. A., and Patrick, A. D. (1985). Dolichol and phosphorylated dolichol content of tissues in ceroid lipofuscinosis. J. Inherited Metab. Dis. 8, 178-183. Heijne, G. V., and Gavel, Y. (1988). Topogenic signals in integral membrane proteins. Eur. J. Biochem. 174, 671-678. Henikoff, S., and Henikoff, J. G. (1994). Protein family classification based on searching a database of blocks. Genomics 19, 97-107. J&rvel~, I. E., Mitchison, H. M., Callen, D. F., Lerner, T. J., Doggett, N. A., Taschner, P. E. M., Gardiner, R. M., and Mole, S. E. (1995a)~ Physical map of the region containing the gene for Batten disease (CLN3). Am. J. Meal. Genet. 57, 316-319. J&rvel&, I. E., Mitchison, H. M., O'Rawe, A. M., Taschner, P. E. M., de Vos, N., Lerner, T. J., D'Arigo, K. L., Callen, D. F., Thompson, A. D., Knight, M., Marrone, B. L., Mundt, M. O., Meincke, L., Breuning, M. H., Gardiner, R. M., Doggett, N. A., and Mole, S. E. (1995b). YAC and cosmid contigs spanning the Batten disease (CLN3) region at 16p12.1-p11.2. Genomics, in press. Kyte, J,, and Doolittle, R. F. (1982). A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157, 105-132. Lerner, T. J., Boustany, R.-M. N., MacCormack, K., Gleitsman, J., Schlumpf, K., Gusella, J. F., Breakefield, X. O., and Haines, J. L. (1994). Linkage disequilibrium between the juvenile NCL gene (CLN3) and marker loci on chromosome 16p12.1. Am. J. Hum. Genet. 54, 88-94. Lerner, T. J., D'Arigo, K. L., Haines J. L., Doggett, N. A., Taschner, P. E. M., de Vos, N., Buckler, A. J., and the Batten Disease Consortium (1995). Isolation of genes from the Batten candidate region using exon amplification. Am. J. Med. Genet. 57, 320-323. McClatchey, A. I., Van den Bergh, P., Pericak-Vance, M. A., Raskind, W., Verellen, C., McKenna-Yasek, D., Rao, K., Haines, J. L., Bird, T., Brown, R. H., and Gusella, J. F. (1992). Temperature-sensitive mutations in the UI-IV cytoplasmic loop region of the skeletal muscle sodium channel gene in paramyotonia congenita. Cell 68, 769-774. Mitchison, H. M, Thompson, A. D., Mulley, J. C., Kozman, H. M., Richards, R. I., Callen, D. F., Stallings, R. L., Doggett, N. A., Attwood, J., McKay, T. R,, Sutherland, G. R., and Gardiner, R. M. (1993). Fine genetic mapping of the Batten disease locus (CLN3) by haplotype analysis and demonstration of allelic association with chromosome 16p microsatellite loci. Genomics 16, 455-460. Mitchison, H. M., Taschner, P. E. M., O'Flawe, A. M., de Vos, N., Phillips, H. A., Thompson, A. D., Kozman, H. M., Haines, J. L., Schlumpf, K., D'ArJgo, K., Boustany, R.-M. N., Callen, D. F., Breuning, M. H., Gardiner, R. M., Mole, S. E., and Lerner, T. J. (1994). Genetic mapping of the Batten disease locus (CLN3) to the interval D16S288D16S383 by analysis of haplotypes and allelic association. Genomics 22, 465-468. Mitchison, H. M., O'Rawe, A. M., Taschner, P. E. M., Sandkuijl, L. A., Santavuori, P., de Vos, N., Breuning, M. H., Mole, S. E., Gardiner, R. M., and J~.rvel~, I. E. (1995). Batten disease gene, CLN3: linkage disequilibrium mapping in the Finnish population and analysis of European haplotypes. Am. J. Hum. Genet. 56, 654-662. Nakai, K., and Kanehisa, M. (1992). A knowledge base for predicting protein localization sites in eukaryotic cells. Genomics 14, 897-911. Orita, M., Suzuki, Y., Sekiya, T., and Hayashi, K. (1989). Rapid and sensitive detection of point mutations and DNA polymorphisms using the polymerase chain reaction. Genomics 5, 874-879. Palmer, D. N., Fearnley, I. M., Walker, J. E., Hall, N. A., Lake, B. D., Wolfe, L. S., Haltia, M., Martinus, R. D., and Jolly, R. D. (1992). Mitochondrial ATP synthase subunit c storage in ceroid-lipofuscinoses (Batten disease). Am. J. Med. Genet. 42, 561-567. Patschinsky, T., Hunter, T., Esch, F. S., Cooper, J. A., and Sefton, B. M. (1982). Analysis of the sequence of amino acids surrounding sites of tyrosine phosphorylation. Proc. Natl. Acad. Sci. USA 79, 973977. Pearson, W. R., and Lipman, D. J. (1988). Improved tools for biological sequence comparison. Proc. Natl. Acad. Sci. USA 85, 2444-2448. Pinna, L. A. (1990). Casein kinase 2: an eminence grise in cellular

regulation? Biochim. Biophys. Acta 1054, 267-284. Rudiger, N. S., Gregersen, N., and Kielland-Brandt, M C. (1995). One short well-conserved region of Alu sequences is involved in human gene rearrangements and has homology with prokaryotic 7.. Nucl. Acids Res 23, 256-60. SambrookJ., Fritsch, E. F., and Maniatis, T. (1989). Molecular Cloning: A Laboratory Manual, Second Edition (Cold Spring Harbor, New York: Cold Spring Harbor Laboratory Press). Sanger, T., Nicklen, S., and Coulson, A. R. (1977). DNA sequencing with chain termination inhibitors. Proc. Natl, Acad. Sci. USA 74, 54635467. Santavuori, P. (1988). Neuronal ceroid-lipofuscinoses in childhood. Brain Dev. 10, 80-83. Siakotos, A. N., Bray, R., Dratz, E., van Kuijk, F., Sevanian, A., and Koppang, N. (1988). 4-hydroxynonenal: a specific indicator for canine neuronal-retinal ceroidosis. Am. J. Med. Genet. (Suppl.) 5, 171-181. Smith, T. F,, and Waterman, M. S. (1981). Indentification of a common molecular subsequences. J. Mol. Biol. 147, 195-197. Soares, M. B., Bonaldo, M. D. F., Jelene, P., Su, L., Lawton, L., and Efstratiadis, A. (1994). Construction and characterization of a normalized cDNA library. Proc. Natl. Acad. Sci. USA 91, 9228-9232. Stallings R, L., Doggett, N. A., Callen, D., Apostolou, S., Chen, L. Z., Nancarrow, J. K., Whitmore, S. A, Harris, P., Mitchison, H., Breuning, M., Saris, J., Fickett, J., Cinkosky, M., Torney, D. C., Hildebrandt, C. E., and Moyzis, R. K. (1992). Evaluation of a cosmid contig physical map of human chromosome 16. Genomics 13, 1031-1039. Taschner, P. E. M., de Vos, N., Thompson, A. D., Callen, D. F., Doggett, N. A., Mole, S. E., Dooley, T. P., Barth, P. G., and Breuning, M. H. (1995a). Chromosome 16 microdeletion in a patient with juvenile neuronal ceroid lipofuscinosis (Batten disease). Am. J. Hum. Genet. 58, 663-666. Taschner, P. E. M., de Vos, N., Post, J. G., Meijers-Heijboer, E. J., Hofman, I., Loonen, M. C. B., Pinckers, A. J. L. G., BleekerWagemakers, E. M., Gardiner, R. M., and Breuning, M. H. (1996b). Carrier detection of Batten disease (juvenile neuronal ceroid liposfuscinosis). Am. J. Med. Genet. 57, 333-337. Wisniewski, K. E., Rapin, I., and Heaney-Kieras. J. (1988). Clinicopathological variability in the childhood neuronal ceroid lipofuscinoses and new observations on glycoprotein abnormalities. Am. J. Med. Genet. (Suppl.) 5, 27-46. Wolfe, L. S., Ivy, G. O., and Witkop, C. J. (1987). Dolichols, lysosomal membrane turnover, and relationships to the accumulation of ceroid and lipofuscin in inherited diseases, Alzheimer's disease and aging. Chem. Soc, Crit. 27, 79-84. Woodgett, J. R., Gould, K. L., and Hunter, T. (1986). Substrate specificity of protein kinase C: use of synthetic peptides corresponding to physiological sites as probes for substrate recognition requirements. Eur. J. Biochem. 161, 177-184. Yan, W., Boustany, R.-M. N., Konradi, C., Ozelius, L., Lerner, T., Trofatter, J. A., Julier, C., Breakefield, X. O., Gusella, J. F., and Haines, J. L. (1993). Localization of juvenile, but not late-infantile, neuronal ceroid lipofuscinosis on chromosome 16. Am. J. Hum. Genet. 52, 8995. Yan, W. L., Lerner, T. J., Haines, J. L., and Gusella, J. F. (1994). Sequence analysis and mapping of a novel human mitochondrial ATP synthase subunit 9 cDNA (ATP5G3). Genomics 24, 375-377. Zeman, W. (1974). Studies in the neuronal ceroid lipofuscinosis. J. Neuropathol. Exp. Neurol. 33, 1-12. GenBank Accession Number The accession number for the sequence reported in this paper is U32680.