doi: 1 o. 1053/ejpn.2000.0439 available online at http://www.idealibrary.com on I | l ~ l" European Journal of Paediatric Neurology 2001 ; 5(Suppl. A): 73-79

O R I G I N A L A R T I C L E

Tripeptidyl-peptidase I in neuronal ceroid lipofuscinoses and other lysosomal storage disorders

KRYSTYNA E WISNIEWSKI, ELIZABETH KIDA, MARIUSZ WALUS, PETER WUJEK, WOJCIECH KACZMARSKI, ADAM A GOLABEK Department of Pathological Neurobiology, New York State Institute for Basic Research in Developmental Disabilities, New York, USA

The classic late infantile form of neuronal ceroid lipofuscinosis (CLN2, cLINCL) is associated with mutations in the gene encoding tripeptidyl-peptidase I (TPP-I), a lysosomal aminopeptidase that cleaves off tripeptides from the free N- termini of oligopeptides. To date over 30 different mutations and 14 polymorphisms associated with CLN2 disease process have been identified. In the present study, we analysed the molecular basis of 15 different mutations of TPP-I by using immunocytochemistry, immunofluorescence, Western blotting, enzymatic assay and subcellular fractionation. In addition, we studied the expression of TPP-I in other lysosomal storage disorders such as CLN1, CLN3, muccopolysaccharidoses and GM 1 and GM 2 gangliosidoses. Our study shows that TPP-I is absent or appears in very small amounts not only in cLINCL subjects with mutations producing severely truncated protein, but also in individuals with missense point mutations, which correlates with loss of TPP-I activity. Of interest, small amounts of TPP-I were detected in lysosomal fraction from fibroblasts from cLINCL subject with protracted form. This observation suggests that the presence of small amounts of TPP-I in lysosomes is able to delay significantly CLN2 disease process. We also show that TPP-I immunoreactivity is increased in the brain tissue of CLN1 and CLN3 subjects, stronger in glial cells and macrophages than neurons. Less prominent increase of TPP-I staining was found in muccopoly- saccharidoses and GM 1 and GM 2 gangliosidoses. These data suggest that TPP-I participates in lysosomal turnover of proteins in pathological conditions associated with cell/tissue injury.

Keywords: CLN2-TPP-I. Mutations. Lysosomal storage disorders.

Introduction

Mutations in the gene coding for tripeptidyl- peptidase I (TPP-I) are associated with the classic- late infantile form of neuronal ceroid lipofuscinosis (cLINCL, CLN2, Jansky-Bielschowsky disease). 1,2,3 This autosomal recessive lysosomal storage dis- order starts at the age of 2 4 years with seizures and dementia, followed by visual loss, cerebellar, pyramidal and extrapyramidal signs, leading to death in the second decade of life. Rare, atypical cases with protracted course of the disease process also have been described. 4,s The CLN2 disease

process is associated with accumulation of auto- fluorescent storage material visualized by electron microscopy as curvilinear profiles in lysosomes in various tissues and organs; however, only the central nervous system shows severe damage, with neuronal loss and gliosis. 6

The enzyme mutated in cLINCL, TPP-I, is an aminopeptidase with acidic pH optimum, 7 local- ized to lysosomes. 8 The deduced amino acid sequence of TPP-I consists of 563 amino acid residues and includes a 16-amino acid signal sequence and a 179-amino acid prodomain, which are removed during the maturation process of the enzyme. 1,9 TPP-I is an exopeptidase, which cleaves

Correspondence: Dr Krystyna E Wisniewski, NYS Institute for Basic Research, Department of Pathological Neurobiology, 1050 Forest Hill Road, Staten Island, NY 10314, USA

1090-3798/01/05/A073+7 $35.00 © 2001 European Paediatric Neurology Society

74 Original article: KE Wisniewski et al.

off tripeptides from the free N-termini of oligo- peptides. TPP-I cleaves in vitro peptide hormones such as angiotensin II, glucagon, 7 or substance p10 as well as synthetic amyloid-]/ peptide 1-42 and 1-28, TM and collagen. 7,11 Most probably, TPP-I also hydrolyses subunit c of mitochondrial ATP synthase, s,l° a proteolipid that accumulates in all types of NCLs except for the infantile form. 12 However, natural substrates of TPP-I are still unknown.

By immunocytochemistry we showed recently that TPP-I is distributed widely in various tissues and organs, although it shows a predilection for specific types of cells (Kida et al., in press). In the developing brain, endothelial cells, choroid plexus, microglial cells and ependyma show TPP-I immu- nostaining distinctly earlier than neurons. The adult pattern of TPP-I immunoreactivity in the brain appears at around the age of 2 years.

Although over 30 different mutations and 14 polymorphisms have been identified in the CLN2 gene, three mutations: R208X and IVS5-1G--~C/A are the most often and account for around 60% of cLINCL chromosomes examined. 2,13 Most muta- tions result in loss or very low levels of TPP-I activity as measured by using Ala-Ala-Phe-amino- methylcoumarin, 14,1s fluorescein isothiocyanate- conjugated haemoglobin. 16 or glycine-fenylo- alanine-fenyloalanine-leucine-7-amino-4-trifluoro- methyl-coumarin TM as substrate. Absence of TPP-I activity is easily explainable in patients with genetic defects leading to severe truncation of TPP-I protein. However, it is still unknown how other types of mutations such as missense point mutations affect biological activity and protein levels of TPP-I. Western blot analysis performed in the past indicated that TPP-I is undetectable or present in very small amounts in CLN2 patients. 17 However, because mutational analysis was not done in these patients, correlation between genetic defect, enzymatic activity, and protein levels of TPP-I was not possible. Furthermore, clinicopatho- logic/molecular genetic correlation was not yet done. Thus we addressed these issues in the present manuscript. We also examined immuno- reactivity to TPP-I in other NCL forms as well as in some other storage disorders.

Material and methods

Material

Post-mortem, frozen or formalin-fixed and paraffin embedded brain tissue sections were collected

from seven subjects with CLN2, five subjects with CLN3, three subjects with CLN1, three subjects with mucopolysaccharidosis type I (MPS I), one subject with MPS IIIA, three cases with GM1 and GM2 gangliosidosis, and eight age-matched controls.

In addition, primary cultures of skin fibroblasts from three CLN2, three controls and three CLN3 subjects as well as transformed lymphoblasts from 10 CLN2, one control and one CLN3 individual also were examined. Material analysed was obtained through the Cell and Tissue Culture Repository at the Institute for Basic Research (IBR), National Neurological Research Specimen Bank (Los Angeles) and IBR Brain Bank. Mutational analysis in all CLN2 subjects included to the present study was performed in the past. 2,13

Immunocytochemistry and immunofluorescence

For immunocytochemistry, 6#m-thick, formalin- fixed and paraffin-embedded tissue sections were deparaffinized and extensively washed. Endo- genous peroxidase activity was blocked with 0.33% hydrogen peroxide in methanol and non- specific binding sites were blocked with 10% fetal calf serum (FCS) in phosphate-buffered saline (PBS) for 30m in at room temperature. Incubation with primary antibodies, either monoclonal anti- body (mAb) 8C4 to TPP-I, produced by us against recombinant TPP-I, as described (Kida et al., in press), (tissue culture supernatant diluted 1:400) or polyclonal antibody (pAb) to subunit c of mito- chondrial ATPase (IBR) (diluted 1:1000) lasted overnight at 4°C. Afterwards, sections were extensively washed in PBS, incubated with bioti- nylated, species-specific secondary antibodies (Amersham) and then with Extravidin-peroxidase conjugate (Sigma), both for 1 h at room tempera- ture at dilution 1:200. Reaction product was visualized by using diaminobenzidine in the presence of hydrogen peroxide. Sections were lightly counterstained with haematoxylin.

For immunofluorescence, cells were fixed with methanol for 20 min and washed in PBS. Non- specific binding sites were blocked with 10% FCS in PBS for l h. After incubation with primary antibodies, mAb 8C4 diluted 1:100 in 10% FCS in PBS for 12h at 4°C, cells were washed in PBS/ 0.05% Tween-20 and incubated for I h at room temperature with the secondary, species-specific antibodies Cy3-conjugated (Jackson Laboratories), diluted 1:600. The cover slips were mounted with

Original article: TPP-I in lyososornal disorders 75

Vectashield (Vector) and viewed with a Nikon Eclipse E600, laser-scanning confocal microscope, equipped with objective Planfluor 100/1.25 oil.

Omission of the primary antibodies was used as a control of the method.

Cell culture

Skin fibroblast cultures were maintained at 37°C in a humidified atmosphere with 5% C O 2 in Dulbecco's modified Eagle medium with 10% FCS, 2 mM glutamine and antibiotics. Epstein- Barr virus-transformed lymphoblasts were grown in RPMI 1640 medium supplemented with 10% FCS, 2 mM glutamine and antibiotics.

SDS-PAGE and Western blotting

Brain tissue samples were homogenized in 1% SDS in 50 mM Tris, pH 7.4 and protease inhibitor cocktail (Complete, Roche). Cells were washed in PBS and lysed in a buffer containing 50 mM Tris, pH 7.4, 1% Triton X-100 and protease inhibitor cocktail (Complete, Roche). The protein content was measured by using a BCA method and BSA as a standard (Pierce). Brain tissue homogenates and cell lysates were solubilized in Tricine sample buffer (BioRad), and 30-80 #g of proteins per lane were loaded onto 10% Tris/Tricine PAGE. Electrophoretically separated proteins were electrotransferred onto nitrocellulose membranes according to Towbin et al. 18 Membranes were subsequently blocked with 5% non-fat dry milk in PBS with 0.05% Tween-20 (PBST), incubated over- night with mAbs 8C4 to TPP-I, diluted 1:400 (tissue culture supernatant), washed extensively in PBST buffer, incubated with peroxidase-conjugated secondary antibodies diluted 1:5000 (Amersham), and developed by using the Enhanced Chemi- luminescence Kit (ECL, Amersham).

Subcellular fractionation

Cells were washed extensively in cold PBS, scraped if grown attached, and homogenized in homo- genization buffer (HB): 50 mM Tris, pH 7.5, 0.25 M sucrose with proteases inhibitors (Complete, Roche) by passing it 20 times through ball-bearing homogenizer with clearance of 51 #m. The homo- genate was first centrifuged at low speed (10 rain at 400 x g) to remove nuclei and unbroken cells and to obtain a post-nuclear supernatant (PNS); PNS was layered over discontinuous gradients consisting of

1.2 ml cushion of 10 x HB and 9.5 ml of an 18% Percoll solution in HB. Gradient was centrifuged for 30min at 67000 x g. The cushion and 11 frac- tions were collected from the top of the tube. Sub- cellular fractions were assayed for organelle-specific enzymes: fl-hexosaminidase-lysosomal marker and ~-mannosidase II-Golgi apparatus marker and ER by using Western blotting with pAb to calreticulin (Affinity Bioreagents), as described. 19

TPP-I activity

Brain tissue and cultured cells were homogenized in 0.1% T-100, 20mM ammonium formate, pH 3.5 and protein concentration was determined by using BCA assay and BSA as a standard. As a TPP-I substrate we have used Phe-Phe-Ala-AMC (Sigma). Reaction mixture contained 10#g of protein and 0.1mM substrate in a total volume of 100#1 of 0.1M sodium acetate, pH 4.5. Reaction was carried out at 37°C for 30min and was terminated by addition of 100#1 of 10% SDS. Liberated 7-amino-4-methylcoumarin was measured fluorometrically (excitation 370 nm, emission 460nm, both slits set at 5nm), after alkalizing the solution by adding 2 ml of 0.1 M Tris-HC1, pH 9.0. Standard solution of AMC was used to calculate specific activity expressed in nmol * min -1 * mg protein -1.

Results

The pattern of brain tissue damage in all seven cLINCL subjects examined was similar to that associated typically with CLN2 disease process. Neuronal loss was the most prominent in cortical mantle and the cerebellum, and was accompanied by gliosis and scattered macrophages. Severe accumulation of subunit c of mitochondrial ATPase was present in various types of cells. However, the degree of brain tissue damage was distinctly less intense in cLINCL subject demon- strating a protracted course of the disease (Table 1, case no. 4) than in other cLINCL subjects examined with typical course (Fig. la,b).

In normal brain, TPP-I immunoreactivity is present in all types of cells. The reaction product is visible as coarse granular material localized mostly to cell somata (Fig. lc). In contrast to control brain tissues, TPP-I immunoreactivity was either absent or weak in the brain tissue of all cLINCL subjects studied at present (Fig. ld, Table 1). None of the cLINCL cases examined demonstrated TPP-I

76

Table 1 Clinicopathologic and molecular genetic correlation

Original article: KE Wisniewski et al.

No. Diagnosis Mutation Tissue TPP-I activity TPP-I protein Immunocytochemistry available n M / m i n / m g (Western blot)

1 cLINCL Arg208X; BF; BP 0.036 (BF) ND (BF) ND Gln422His

2 cLINCL SJ3556G > C; BF; BP; L 0.013 (BF) ND ND Arg208X

3 cLINCL SJ3556G > C; BF; BP 0.045 (BF) ND + / - Gly77Arg

4 Protracted SJ3556G > C; BF; BP; F 0.040 (BF); ND* + / - LINCL Arg447His ND (F)

5 cLINCL Gln422His; BF; BP; L 0.019 (BF); ND* +(BP); ND (L) Gln422His 0.012 (L)

6 cLINCL Arg208X; BP; F ND (F) ND ND Arg208X

7 cLINCL SJ3556G>C; BP; BF; L 0.010 (BF) ND + / - (BP); ND (L) SJ4188A > G

8 cLINCL 3081-3091 del; L ND ND ND Arg208X

9 cLINCL SJ3556G > C; L 0.010 ND ND Gln422His

10 cLINCL 6151-6152del; L ND ND ND 6151-6152del

11 cLINCL SJ4396T > G; L ND ND ND Arg208X

12 cLINCL SJ3556G > C; F 0.001 ND ND 4288-4295del

13 cLINCL Arg208X; L ND ND ND Va1385Asp

14 cLINCL Arg208; L ND ND ND Gin422

15 cLINCL SJ3556G > A; L 0.10 ND ND SJ3556G > C

16 Controls - - BF; BP; L; F 1.476-2.223 (BF); ++ __ 1.265 (F); 1.258 (L)

17 JNCL 1.02 kb del BF; BP; L; F 5.176 (BF); 0.921- +++ +++ (CLN3) (homozygotes) 2.086 (F); 1.521 (L)

cLINCL classic late infantile neuronal ceroid lipofuscinosis; JNCL juvenile neuronal ceroid lipofuscinosis; SJ splice-site junction; del deletion; BF frozen brain tissue; BP paraffin-embedded brain tissue; L lymphoblasts; F fibroblasts; ND non-detectable. Intensity of immunostaining was assessed as follows: ND non-detectable; + / - marginal; +moderate; ++strong stain; +++very strong immunoreactivity; *detectable in subcellular fractions or by extraction (see text for details).

immunoreac t iv i ty in neurons. In one cLINCL case, h o m o z y g o u s toward Gln422His, n u m e r o u s reactive astrocytes showed mode ra t e TPP-I stain- ing (Fig. le). Howeve r , in this case, the reaction p roduc t was ubiqui tous ly dis t r ibuted in cell cyto- p l a sm and was devoid of a granular staining pat tern, which is typical for lysosomes. In three other cLINCL cases, only some reactive astro- cytes showed a weak, h o m o g e n o u s in nature, TPP-I immunoreac t iv i ty in the cerebral cortex (Table 1).

TPP-I activity was only marg ina l or nondetect- able in the brain tissues of all cLINCL cases examined (Table 1). Maximal values obtained for cLINCL group, 0.045 n m o l / m i n / m g , were dis- tinctly lower that those obtained for controls: 1.476- 2.223 n m o l / m i n / m g . Very low TPP-I activity also

was found in cLINCL case wi th prot rac ted course (0.040 n m o l / m i n / m g in frozen brain tissue, and non-detectable in fibroblast lysate). By rout ine Western blot analysis, TPP-I was undetectable in the brain tissue of all cLINCL subjects examined (Fig. 2a, Table 1). However , extraction of the brain tissue homogena te s wi th SDS or formic acid, a l lowed visual izat ion of TPP-I on immunob lo t s in one case, h o m o z y g o u s toward Gln422His, a l though, in minu te amoun t s (Fig. 2b).

In cul tured p r i m a r y fibroblast or t r ans fo rmed lymphoblas t s f rom cLINCL cases, TPP-I was undetectable by bo th immunof luorescence and laser-scanning confocal microscopy analysis and Western blot t ing (Fig. 2c), and its activity was either undetectable or only marg ina l (Table 1). Of interest, however , small amoun t of TPP-I was

Original article: TPP-I in lyososomal disorders 77

• ~ , g ,

v

#

°

t

i

, i

~q

@ 2 1 1 ; , @ 2

Fig. 1. Immunocytochemistry of cLINCL brain tissues. A. Severe damage of the cerebellar cortex in a cLINCL case, heterozygous toward Arg208X/GIn422His mutations B. Relatively well-preserved cerebellar cortex in cLINCL subject with protracted form, heterozygous toward splice-site mutation 3556G>C and Arg447His mutation. C. Granular pattern of TPP-I staining of cortical neurons in the control case. D. Lack of TPP-I immunoreactivity in a cLINCL subject heterozygous toward Arg208X/GIn422His. E. Absence of TPP-I immunostaining in neurons and homogenous immunoreactivity of some reactive astrocytes in a cLINCL subject, homozygous toward GIn422His mutation. A, B-pAb to subunit c of mitochondrial ATP synthase; C, D, E-mAbs 8C4 to TPP-I. A x 200; B, D, E x 400; C x 800.

found in lysosomes isolated from cultured fibroblasts from the cLINCL subject with pro- tracted form (Fig. 3) by using subcellular fractionation.

By immunocytochemistry, we observed a strik- ing increase in TPP-I immunoreactivity in reactive astrocytes in the cerebral cortex of CLN1 subjects, which was less intense in brain regions better preserved (not shown). Survived neurons showed a slight increase of TPP-I staining in comparison

with controls. Increased TPP-I immunoreactivity, more prominent in reactive astrocytes and macro- phages than neurons also was disclosed in CLN3 subjects (not shown). Of interest, although TPP-I levels and activity were increased up to three-fold in brain tissues from CLN3 subjects, CLN3 fibro- blasts and transformed lymphoblasts showed TPP- I levels and activity close to control values (Table 1). Slight to moderate increase in TPP-I immuno- reactivity also was found in other lysosomal

7 8

A k D a

66

4 6 _ _ m m

3 0 m

g

B

1 2 3 4

k D a

6 6

4 6 ~ ~

3 0

6 7 8

C

1 2 3 4

k D a

6 6 m

3 0 m

1 2 3 4 5 6 7 8 9

Fig. 2. Western blot analysis of TPP-I in brain tissue homogenates and lysates of cultured fibroblasts. A. Absence of TPP-I in brain tissue homogenates of cLINCL subjects (lanes 3-8) with various mutations in TPP-I gene (Table 1 cases No. 1-6), and high levels of TPP-I in brain homogenates from normal control cases (lanes 1, 2) and CLN3 subject (lane 9). B. Small levels of TPP-I extracted from 40/~g of brain tissue homogenates from cLINCL case homozygous toward GIn422His mutation by using SDS (lane 1) or formic acid (lane 3) in comparison with control sample extracted by using SDS (lane 2) or formic acid (lane 4). C. TPP-I is present in comparable amounts in lysates from control (lanes 1-3) and CLN3 fibroblasts (lane 4-6) and is absent in lysates from cLINCL fibroblasts (lanes 7-9, Table 1, cases no. 4, 6, 12).

storage disorders examined, more prominent in MPS than GM cases.

The results obtained are summarized in Table 1.

Discussion

Our study shows that TPP-I is absent or appears in very small amounts not only in cLINCL subjects with mutations producing severely truncated pro- tein, but also in individuals with missense point

Original article: KE Wisniewski et al.

A k D a

66

46

30

B i~;. iii:ii.iii~i/~ ~ ~i:ii::?:iZ2 ~!!~:ii~:~iil./ ~ ~/~@i/: ii ( : ~ ~

A l E

46

30

1 2 3 4 5 6 7 8 9 10 1112

Fig. 3. Subcellular fractionation of control (A) and cLINCL (B) fibroblasts. Mature TPP-I protein is present in bottom, lysosomal fraction from control fibroblasts (lane 12, arrow) in much higher amounts than in cLINCL subject with protracted form (arrow). Please note the presence of proenzyme and mature enzyme in top fractions (lanes 1-4) enriched in the endoplasmic reticulum and the Golgi apparatus structures in control fibroblasts, and only proenzyme in cLINCL fibroblasts (lanes 1-3).

mutations either in heterozygous or homozygous form. This finding indicates that loss of TPP-I activity disclosed in these subjects is caused by absence of TPP-I or its low levels, barely detectable by such Western blotting of whole lysates or subcellular fractions. These observations suggest that TPP-I protein bearing missense point muta- tions analysed at present is unstable and undergoes rapid degradation. Although a subset of reactive astroglial cells showed some TPP-I immunoreac- tivity in four cLINCL cases by immunocytochem- istry, the pattern of staining was different from that typical for lysosomes. Thus, either TPP-I was mistargeted in these cases or this particular staining represents an artifact associated with tissue processing. Neither cultured fibroblasts nor transformed lymphoblasts from these subjects demonstrated TPP-I immunostaining by immuno- fluorescence and laser-scanning confocal micro= scopy analysis.

Of interest, although the activity of TPP-I also was low in cLINCL subject with atypical, pro- tracted clinical course, subcellular fractionation technique allowed detection of small amounts of TPP-I in lysosomal fraction from cultured fibro- blasts from this subject. This observation suggests that the presence of even small amounts of TPP-I in lysosomes is able to slow down significantly CLN2 disease process, representing a good prognostic for future enzymatic replacement therapy.

It was reported in the past that TPP-I activity is increased in brain tissues from CLN1 and CLN3 subjects. TM We show at present that increase in TPP-I immunoreactivity in CLN1 and CLN3

Original article: TPP-I in lyososomal disorders 79

individuals is more p rominen t in glial cells and macrophages than neurons. This f inding suggests that TPP-I is up regu la ted in cells wi th phagocyt ic proper t ies to part ic ipate in degrada t ion of pro- t e ins /po lypep t ides and resolution of d a m a g e d tissues. This hypothes is is suppor t ed by our data showing that TPP-I immunoreac t iv i ty is weake r in MPS and GM cases, which demons t ra te less severe brain tissue d a m a g e than CLN1 and CLN3 sub- jects. Fur thermore , in cul tured p r i m a ry fibroblasts and t rans formed lymphoblas t s f rom CLN3 sub- jects, both the activity and protein levels of TPP-I were comparab le wi th control values. Because fibroblasts and t rans formed lymphoblas t s f rom subjects wi th juvenile fo rm of NCL do not reveal CLN3 p h e n o t y p e in vitro, 2° it can be a s sumed that lysosomal s torage per se also could induce TPP-I expression. Thus, it seems that complex biological processes associated wi th cell injury and d a m a g e d tissue resolut ion underl ie TPP-I mobil izat ion in lysosomal s torage disorders other than CLN2 examined at present .

Acknowledgements Suppor ted by N I H grant NS38988, the Batten 's Disease Suppor t and Research Association, and the N e w York State Office for Mental Retardat ion and Deve lopmenta l Disabilities.

References 1 Sleat DE, Donnelly RJ, Lackland H et al. Association

of mutations in a lysosomal protein with classical late-infantile neuronal ceroid lipofuscinosis. Science 1997; 277: 1802-1805.

2 Sleat DE, Gin RM, Sohar Iet al. Mutational analysis of the defective protease in classic late-infantile neuronal ceroid lipofuscinosis, a neurodegenerative lysosomal storage disorder. Am J Hum Genet 1999; 64: 1511-1523.

3 Rawlings ND, Barrett AJ. Tripeptidyl-peptidase I is apparently the CLN2 protein absent in classical late- infantile neuronal ceroid lipofuscinosis. Biochim Bio- phys Acta 1999; 1429: 496-500.

4 Wisniewski KE, Kida E, Patxot OF, Connell F. Variability in the clinical and pathological findings in the neuronal ceroid lipofuscinosis: review of data and observations. Am J Med Genet 1992; 42: 525-532.

5 Wisniewski KE, Kaczmarski A, Kida E et al. Re- evaluation of neuronal ceroid lipofuscinoses: Atypical juvenile onset may be the result of CLN2 mutations. Mol Genet Metab 1999; 66: 248-252.

6 Williams RE, Gottlob I, Lake BD. CLN2.Classic late infantile NCL. In: Goebel HH et al. (eds). The Neuronal Ceroid Lipofuscinoses (Batten disease). Amsterdam: IOS Press, 1999: 37-54.

7 Vines D, Warburton MJ. Purification and character- isation of tripeptidyl aminopeptidase I from rat spleen. Biochim Biophys Acta 1998; 1384: 233-242.

8 Ezaki J, Tanida I, Kanehagi N, Kominami E. A lyso- somal proteinase, the late infantile neuronal ceroid lipofuscinosis gene (CLN2) product, is essential for degradation of a hydrophobic protein, the subunit c of ATP synthase. J Neurochem 1999; 72: 2573-2582.

9 Liu C-G, Sleat DE, Donnelly RJ, Lobel P. Structural organization and sequence of CLN2, the defective gene in classical late infantile neuronal ceroid lipofuscinosis. Genomics 1998; 50: 206-212.

10 Junaid MA, Wu G, Pullarkat RK. Purification and characterization of bovine brain lysosomal pepstatin- insensitive proteinase, the gene product deficient in the human late-infantile neuronal ceroid lipofusci- nosis. J Neurochem 2000; 74: 287-294.

11 McDonald JK, Hoisington AR, Eisenhauer DA. Partial purification and characterization of an ovarian tripeptidyl peptidase: a lysosomal exo- peptidase that sequentially releases collagen- related (Gly-Pro-X) triplets. Biochem Biophys Res Commun 1985; 126: 63-71.

12 Palmer DN, Jolly RD, van Mil HC et al. Different pattern of hydrophobic protein storage in different forms of neuronal ceroid lipofuscinosis (NCL, Batten disease). Neuropediatrics 1997; 28: 45-48.

13 Zhong N, Wisniewski KE, Hartikainen Jet al. Two common mutations in the CLN2 gene underlie late infantile neuronal ceroid lipofuscinosis. Clin Genet 1998; 54: 234-238.

14 Vines DJ, Warburton MJ. Classical late infantile neuronal ceroid lipofuscinosis fibroblasts are defi- cient in lysosomal tripeptidyl peptidase I. FEBS Lett 1999; 443: 131-135.

15 Warburton MJ, BemardiniF. Tripeptidyl-peptidase I deficiency in classical late-infantile neuronal ceroid lipofuscinosis brain tissue. Evidence for defective peptidase rather than proteinase activity. J Inherit Metab Dis 2000; 23: 145-154.

16 Sohar I, Sleat DE, Jadot M, Lobel P. Biochemical characterization of a lysosomal protease deficient in classical late infantile neuronal ceroid lipofuscinosis (LINCL) and development of an enzyme-based assay for diagnosis and exclusion of LINCL in human specimens and animal models. J Neurochem 1999; 73: 700-711.

17 Kurachi Y, Oka A, Mizuguchi M et al. Rapid immunologic diagnosis of classic late infantile neuronal ceroid lipofuscinosis. Neurology 2000; 54: 1676-1680.

18 Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applica- tions. Proc Natl Acad Sci USA 1979; 76: 4350-4359.

19 Golabek AA, Kida E, Walus Met al. CLN3 protein regulates lysosomal pH and alters intracellular processing of Alzheimer's amyloid-fl protein pre- cursor and cathepsin D in human cells. Mol Genet Metab 2000; 70: 203-213.

20 Lake BD, Rowan SA. Light and electron microscopic studies on subunit c in cultured fibroblasts in late infantile and juvenile Batten disease. Neuropediatrics 1997; 28- 56-59.