- Email: [email protected]
The expression of late infantile neuronal ceroid lipofuscinosis (CLN2) gene product in human brains
Neuroscience Letters 257 (1998) 113–115
The expression of late infantile neuronal ceroid lipofuscinosis (CLN2) gene product in human brains Akira Oka a , b ,*, Yukiko Kurachi a , c, Masashi Mizuguchi d, Masaharu Hayashi e, Sachio Takashima a a
Department of Mental Retardation and Birth Defect Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry, Kodaira, Tokyo 187-8502, Japan b Division of Child Neurology, Institute of Neurological Sciences, Faculty of Medicine, Tottori University, Yonago, Tottori 683-8504, Japan c Department of Pediatrics, Faculty of Medicine, University of Tokyo, Bunkyo-ku, Tokyo 113-8655, Japan d Department of Pediatrics, Jichi Medical School, Minamikawachi, Tochigi 329-0498, Japan e Department of Clinical Neuropathology, Tokyo Metropolitan Institute of Neuroscience, Fuchu, Tokyo 183-8526, Japan Received 8 September 1998; accepted 24 September 1998
Abstract We raised polyclonal antibodies against a gene product responsible for late infantile neuronal ceroid lipofuscinosis (CLN2). By Western blotting, all three antisera recognized the CLN2 protein at approximately 49 kDa in human brain homogenates. Immunohistochemistry using the antisera demonstrated the granular labelling in the cytoplasm of cerebral neurons and glial cells. The immunoreactivity on Western blots was absent from the brain of a patient with CLN2. Our results suggest the usefulness of these antibodies for the diagnosis of CLN2, which currently requires demonstration of characteristic ultrastructure by electron microscopy. 1998 Elsevier Science Ireland Ltd. All rights reserved
Keywords: Late infantile ceroid lipofuscinosis; CLN2; Gene; Immunohistochemistry; Western blotting; Human
Neuronal ceroid lipofuscinoses (NCLs) comprise a group of hereditary progressive neurodegenerative diseases during childhood, which share clinicopathological features such as visual impairment, seizures, psychomotor deterioration, and the presence of autofluorescent lipopigments in neurons and other cells. Based on the age at onset and ultrastructures of the lipopigments, NCLs have been classified into subtypes [4]. To date, three responsible genes have been detected [1,8,9]. Mutations in the palmitoyl protein thioesterase gene on chromosome 1p32 cause infantile NCL (CLN1; Santavuori–Haltia disease) characterized by granular osmiophilic deposits (GROD) on electron microscopy as well as a variant of juvenile NCL with GROD [5,9]. In classic juvenile NCL (CLN3; Spielmeyer-Vogt disease; Batten disease) and a protracted variant, both of which typically show fingerprint profiles, defects of a gene of unknown * Corresponding author. Tel.: +81 859 348038; fax:+81 859 348135.
0304-3940/98/$ - see front matter PII S0304- 3940(98) 00787- 3
function on chromosome 16p12 (CLN3 gene) have been demonstrated [1,10]. Classical late-infantile NCL (CLN2; JanskyBielschowsky disease) is characterized by the presence of curvilinear bodies and a clinical onset between 2 and 4 years of age. Patients with CLN2 exhibit severe seizures, dementia, and ataxia, leading to death between 10 and 15 years of age. Recently, defects of a gene on 11p15 have been identified in CLN2 patients [8]. The CLN2 gene encodes a lysosomal protein of 46–48 kDa, which may function as a pepstatin-insensitive peptidase [8]. To investigate the expression of the CLN2 protein in human brains, we raised antibodies and performed immunoblotting and immunohistochemical examination of human brains. Specimens from the frontal lobes and liver were obtained post mortem from a patient with CLN2 as well as three subjects without pathological changes in either organ. In the CLN 2 patient, clinical onset occurred at age 2 years,
1998 Elsevier Science Ireland Ltd. All rights reserved
114
A. Oka et al. / Neuroscience Letters 257 (1998) 113–115
and typical neurological symptoms such as epilepsy, myoclonus, and ataxia were observed. Following his death at the age of 13 years, autopsy revealed loss of neurons as well as the accumulation of autofluorescent lipopigments in the remaining neurons and glial cells. The presence of curvilinear bodies was observed by electron microscopy, confirming the diagnosis. For Western blotting analysis, frozen samples from the frontal lobes and liver were used. For immunohistochemistry, brain tissues of control patients were fixed in 10% formalin for 2–3 weeks, then embedded in paraffin, and cut into 4 mm-thick sections. Antisera were raised in Japanese White rabbits against three peptides synthesized by solid-phase techniques according to the sequence [8]. These were amino acid residues 196 to 209 plus the carboxy-terminal cysteine (peptide N; LHLGVTPSVIRKRYC), 320 to 333 plus the aminoterminal cysteine (peptide I; CVHTVSYGDDEDSLSS), and 549 to 563 plus the amino-terminal cysteine (peptide C; CGTPNFPALLKTLLNP). Immunization, collection, titration, and absorption of the antisera were performed as described previously [6]. Western blot analyses were performed by a standard method [7]. For the extraction of proteins, samples were thawed and homogenized in a 2% SDS loading buffer. After centrifugation, supernatants were collected, and protein assay was performed using the method of Bradford. Samples (60 mg/lane) were separated on a 10% SDS polyacrylamide gel, and electrophoretically transferred to an Immobilon polyvinylidene difluoride membrane (Millipore; Bedford, MA, USA). After the membrane was blocked with 8% skim milk overnight at 4°C, it was incubated with each antiserum overnight at 4°C, and then with biotinylated rabbit anti-goat IgG (1:500) for 1 h. Three antisera against different peptides were used for Western blotting; antiCLN2-N2 against peptide N (1:2000), anti-CLN2-I1 against peptide I (1:500), and anti-CLN2-C1 against peptide C (1:1500). Bands were visualized by avidin-biotin procedures using biotinylated goat anti-rabbit IgG (1:1000), alka-
line phosphatase-conjugated avidin-biotin complex, and Vector Black (Vector, Burlingame, CA, USA). Immunostaining was performed with the anti-CLN2 antibodies. After deparaffinization, the sections were immersed in 0.3% of hydrogen peroxide in methanol for 20 min to block the endogenous peroxide activity, and then, for antiCLN2-N2 antibody, heated in a microwave oven for 9 min at 90°C for antigen retrieval. Non-specific binding was blocked with a 10% goat serum. The sections were incubated with anti-CLN2-N2 antibody (1:800) or anti-CLN2-I1 antibody (1:1000) overnight at 4°C, and then processed according to the avidin-biotin procedures by using diaminobenzidine as a chromogen. Between the steps, the sections were thoroughly washed with phosphate buffered saline (145 mM NaCl, 2.1 mM NaH2PO4, 8.0 mM Na2HPO4) three times. On the Western blots of human brain homogenates of adult controls, all antisera (Fig. 1A, lanes 1, 4, and 7) recognized a band at approximately 49 kDa, apparently corresponding to the CLN2 protein [8]. The band was not detected by the respective preimmune antisera (Fig. 1A, lanes 2, 5, and 8) or antisera after absorption (Fig. 1A, lanes 3, 6, and 9). Three additional bands were found at approximately 60, 35, and 30 kDa (Fig. 1A, lanes 1, 4, and 7), the identity of which were unknown. It is possible that they were derived from posttranslational modification. Then, we applied the anti-CLN2-N2 antibody, which most clearly detected the 49 kDa protein, to Western blotting in samples from CLN2 patient. The CLN2 protein, which appeared in homogenates of the control cerebrum and liver (Fig. 1B, lanes 1 and 2), was not detected in those of the CLN2 patient (Fig. 1B, lanes 3 and 4). Immunohistochemistry using anti-CLN2-N2 antibody and anti-CLN2-I1 antibody was performed in the human cerebral cortex without neuropathological changes. Similar immunolabelling was observed by both antibodies, and the reactivity was observed in the cytoplasm of neurons and glial cells (Fig. 2A,B). There were intensely labelled gran-
Fig. 1. (A) Western blots of adult human cerebral homogenates, immunolabelled by anti-CxLN2-N2 (lane 1), anti-CLN-I1 (lane 4), and anti-CLN2C1 (lane 7) as well as the respective preimmune sera (lanes 2, 5, and 8) and antisera after absorption (lanes 3, 6, and 9). (B) Western blots with anti-CLN2-N2 of a control human cerebrum (lane 1) and liver (lane 2) as well as those of a cerebrum (lane 3) and liver (lane 4) of a CLN2 patient.
A. Oka et al. / Neuroscience Letters 257 (1998) 113–115
115
Fig. 2. Photomicrographs of the sections of a human control cerebral cortex, immunolabelled by anti-CLN2-N2 (A) and anti-CLN-I1 (B). Sections were counterstained lightly with hematoxylin. Scale bar, 15 mm.
ules in the cytoplasm. The staining completely disappeared in immunohistochemistry using the respective preimmune antisera and the antisera after absorption (data not shown). Thus, we raised antisera that recognized the CLN2 protein (Fig. 1A), and the immunohistochemical study indicated the presence of reactivity in human cerebral neurons and glial cells (Fig. 2). The dense granular labelling appeared to correspond to cytoplasmic organelles, such as lysosomes. The distribution of the CLN2 protein shown by this study appeared to be compatible with the neuropathological alterations of CLN2 [2], such as the accumulation of lipopigments within the cytoplasm of these cells. We also demonstrated, by Western blotting, the lack of this protein in the brain and hepatic tissues of a CLN2 patient (Fig. 1B). Currently, the diagnosis of NCLs including CLN2 depends on the demonstration of characteristic ultrastructures, which requires electron microscopic examination [3]. Therefore, clinical application of the antisera for the diagnosis of CLN2 may be beneficial. Since we only examined a single case of CLN2 in this study, further investigations are necessary. This study was supported by grants from the Ministry of Health and Welfare, and the Ministry of Education, Science, Sports, and Culture of Japan. [1] Consortium, The International Batten Disease, Isolation of a novel gene underlying Batten disease, CLN3, Cell, 82 (1995) 949–957. [2] Friede, R.L., Ceroid-lipofuscinosis and miscellaneous lipidosis. In Developmental Neuropathology, 2nd edn., Springer, Berlin, 1989, pp. 448–460.
[3] Goebel, H.H., Morphologic diagnosis in neuronal ceroid lipofuscinosis, Neuropediatrics, 28 (1997) 67–69. [4] Goebel, H.H. and Sharp, J.D., The neuronal ceroid-lipofuscinosis: recent advances, Brain Pathol., 8 (1998) 151–162. [5] Mitchison, H.M., Hofmann, S.L., Becerra, C.H.R., Munroe, P.B., Lake, B.D., Crow, Y.J., Stephenson, J.B.P., Williams, R.E., Hofman, I.L., Taschner, P.E.M., Martin, J.-J., Philippart, M., Andermann, E., Andermann, F., Mole, S.E., Gardiner, R.M. and O’Rawe, A.M., Mutations in the palmitoyl-protein thioesterase gene (PPT; CLN1) causing juvenile neuronal ceroid lipofuscinosis with granular osmiophilic deposits, Hum. Mol. Genet., 7 (1998) 291–297. [6] Mizuguchi, M., Takashima, S., Kakita, A., Yamada, M. and Ikeda, K., Lissencephaly gene product localization in the central nervous system and loss of immunoreactivity in Miller–Dieker syndrome, Am. J. Pathol., 147 (1995) 1142–1151. [7] Oka, A. and Takashima, S., Induction of cyclo-oxygenase 2 in brains with Down’s syndrome and dementia of Alzheimer type: specific localization in affected neurones and axons, NeuroReport, 8 (1997) 1161–1164. [8] Sleat, D.E., Donnelly, R.J., Lackland, H., Liu, C.-G., Sohar, I., Pullarkat, R.K. and Lobel, P., Association of mutations in a lysosomal protein with classical late-infantile neuronal ceroid lipofuscinosis, Science, 277 (1997) 1802–1805. [9] Vesa, J., Hellsten, E., Verkruyse, L.A., Camp, L.A., Rapola, J., Santavuori, P., Hofmann, S.L. and Peltonen, L., Mutations in the palmitoyl protein thioesterase gene causing infantile neuronal ceroid lipofuscinosis, Nature, 376 (1995) 584–587. [10] Wisniewski, K.E., Zhong, N., Kaczmarski, W., Kaczmarski, A., Kida, E., Brown, W.T., Schwarz, K.O., Lazzarini, A.M., Rubin, A.J., Stenroos, E.S., Johnson, W.G. and Wisniewski, T.M., Compound heterozygous genotype is associated with protracted juvenile neuronal ceroid lipofuscinosis, Ann. Neurol., 43 (1998) 106–110.
The expression of late infantile neuronal ceroid lipofuscinosis (CLN2) gene product in human brains Akira Oka a , b ,*, Yukiko Kurachi a , c, Masashi Mizuguchi d, Masaharu Hayashi e, Sachio Takashima a a
Department of Mental Retardation and Birth Defect Research, National Institute of Neuroscience, National Center of Neurology and Psychiatry, Kodaira, Tokyo 187-8502, Japan b Division of Child Neurology, Institute of Neurological Sciences, Faculty of Medicine, Tottori University, Yonago, Tottori 683-8504, Japan c Department of Pediatrics, Faculty of Medicine, University of Tokyo, Bunkyo-ku, Tokyo 113-8655, Japan d Department of Pediatrics, Jichi Medical School, Minamikawachi, Tochigi 329-0498, Japan e Department of Clinical Neuropathology, Tokyo Metropolitan Institute of Neuroscience, Fuchu, Tokyo 183-8526, Japan Received 8 September 1998; accepted 24 September 1998
Abstract We raised polyclonal antibodies against a gene product responsible for late infantile neuronal ceroid lipofuscinosis (CLN2). By Western blotting, all three antisera recognized the CLN2 protein at approximately 49 kDa in human brain homogenates. Immunohistochemistry using the antisera demonstrated the granular labelling in the cytoplasm of cerebral neurons and glial cells. The immunoreactivity on Western blots was absent from the brain of a patient with CLN2. Our results suggest the usefulness of these antibodies for the diagnosis of CLN2, which currently requires demonstration of characteristic ultrastructure by electron microscopy. 1998 Elsevier Science Ireland Ltd. All rights reserved
Keywords: Late infantile ceroid lipofuscinosis; CLN2; Gene; Immunohistochemistry; Western blotting; Human
Neuronal ceroid lipofuscinoses (NCLs) comprise a group of hereditary progressive neurodegenerative diseases during childhood, which share clinicopathological features such as visual impairment, seizures, psychomotor deterioration, and the presence of autofluorescent lipopigments in neurons and other cells. Based on the age at onset and ultrastructures of the lipopigments, NCLs have been classified into subtypes [4]. To date, three responsible genes have been detected [1,8,9]. Mutations in the palmitoyl protein thioesterase gene on chromosome 1p32 cause infantile NCL (CLN1; Santavuori–Haltia disease) characterized by granular osmiophilic deposits (GROD) on electron microscopy as well as a variant of juvenile NCL with GROD [5,9]. In classic juvenile NCL (CLN3; Spielmeyer-Vogt disease; Batten disease) and a protracted variant, both of which typically show fingerprint profiles, defects of a gene of unknown * Corresponding author. Tel.: +81 859 348038; fax:+81 859 348135.
0304-3940/98/$ - see front matter PII S0304- 3940(98) 00787- 3
function on chromosome 16p12 (CLN3 gene) have been demonstrated [1,10]. Classical late-infantile NCL (CLN2; JanskyBielschowsky disease) is characterized by the presence of curvilinear bodies and a clinical onset between 2 and 4 years of age. Patients with CLN2 exhibit severe seizures, dementia, and ataxia, leading to death between 10 and 15 years of age. Recently, defects of a gene on 11p15 have been identified in CLN2 patients [8]. The CLN2 gene encodes a lysosomal protein of 46–48 kDa, which may function as a pepstatin-insensitive peptidase [8]. To investigate the expression of the CLN2 protein in human brains, we raised antibodies and performed immunoblotting and immunohistochemical examination of human brains. Specimens from the frontal lobes and liver were obtained post mortem from a patient with CLN2 as well as three subjects without pathological changes in either organ. In the CLN 2 patient, clinical onset occurred at age 2 years,
1998 Elsevier Science Ireland Ltd. All rights reserved
114
A. Oka et al. / Neuroscience Letters 257 (1998) 113–115
and typical neurological symptoms such as epilepsy, myoclonus, and ataxia were observed. Following his death at the age of 13 years, autopsy revealed loss of neurons as well as the accumulation of autofluorescent lipopigments in the remaining neurons and glial cells. The presence of curvilinear bodies was observed by electron microscopy, confirming the diagnosis. For Western blotting analysis, frozen samples from the frontal lobes and liver were used. For immunohistochemistry, brain tissues of control patients were fixed in 10% formalin for 2–3 weeks, then embedded in paraffin, and cut into 4 mm-thick sections. Antisera were raised in Japanese White rabbits against three peptides synthesized by solid-phase techniques according to the sequence [8]. These were amino acid residues 196 to 209 plus the carboxy-terminal cysteine (peptide N; LHLGVTPSVIRKRYC), 320 to 333 plus the aminoterminal cysteine (peptide I; CVHTVSYGDDEDSLSS), and 549 to 563 plus the amino-terminal cysteine (peptide C; CGTPNFPALLKTLLNP). Immunization, collection, titration, and absorption of the antisera were performed as described previously [6]. Western blot analyses were performed by a standard method [7]. For the extraction of proteins, samples were thawed and homogenized in a 2% SDS loading buffer. After centrifugation, supernatants were collected, and protein assay was performed using the method of Bradford. Samples (60 mg/lane) were separated on a 10% SDS polyacrylamide gel, and electrophoretically transferred to an Immobilon polyvinylidene difluoride membrane (Millipore; Bedford, MA, USA). After the membrane was blocked with 8% skim milk overnight at 4°C, it was incubated with each antiserum overnight at 4°C, and then with biotinylated rabbit anti-goat IgG (1:500) for 1 h. Three antisera against different peptides were used for Western blotting; antiCLN2-N2 against peptide N (1:2000), anti-CLN2-I1 against peptide I (1:500), and anti-CLN2-C1 against peptide C (1:1500). Bands were visualized by avidin-biotin procedures using biotinylated goat anti-rabbit IgG (1:1000), alka-
line phosphatase-conjugated avidin-biotin complex, and Vector Black (Vector, Burlingame, CA, USA). Immunostaining was performed with the anti-CLN2 antibodies. After deparaffinization, the sections were immersed in 0.3% of hydrogen peroxide in methanol for 20 min to block the endogenous peroxide activity, and then, for antiCLN2-N2 antibody, heated in a microwave oven for 9 min at 90°C for antigen retrieval. Non-specific binding was blocked with a 10% goat serum. The sections were incubated with anti-CLN2-N2 antibody (1:800) or anti-CLN2-I1 antibody (1:1000) overnight at 4°C, and then processed according to the avidin-biotin procedures by using diaminobenzidine as a chromogen. Between the steps, the sections were thoroughly washed with phosphate buffered saline (145 mM NaCl, 2.1 mM NaH2PO4, 8.0 mM Na2HPO4) three times. On the Western blots of human brain homogenates of adult controls, all antisera (Fig. 1A, lanes 1, 4, and 7) recognized a band at approximately 49 kDa, apparently corresponding to the CLN2 protein [8]. The band was not detected by the respective preimmune antisera (Fig. 1A, lanes 2, 5, and 8) or antisera after absorption (Fig. 1A, lanes 3, 6, and 9). Three additional bands were found at approximately 60, 35, and 30 kDa (Fig. 1A, lanes 1, 4, and 7), the identity of which were unknown. It is possible that they were derived from posttranslational modification. Then, we applied the anti-CLN2-N2 antibody, which most clearly detected the 49 kDa protein, to Western blotting in samples from CLN2 patient. The CLN2 protein, which appeared in homogenates of the control cerebrum and liver (Fig. 1B, lanes 1 and 2), was not detected in those of the CLN2 patient (Fig. 1B, lanes 3 and 4). Immunohistochemistry using anti-CLN2-N2 antibody and anti-CLN2-I1 antibody was performed in the human cerebral cortex without neuropathological changes. Similar immunolabelling was observed by both antibodies, and the reactivity was observed in the cytoplasm of neurons and glial cells (Fig. 2A,B). There were intensely labelled gran-
Fig. 1. (A) Western blots of adult human cerebral homogenates, immunolabelled by anti-CxLN2-N2 (lane 1), anti-CLN-I1 (lane 4), and anti-CLN2C1 (lane 7) as well as the respective preimmune sera (lanes 2, 5, and 8) and antisera after absorption (lanes 3, 6, and 9). (B) Western blots with anti-CLN2-N2 of a control human cerebrum (lane 1) and liver (lane 2) as well as those of a cerebrum (lane 3) and liver (lane 4) of a CLN2 patient.
A. Oka et al. / Neuroscience Letters 257 (1998) 113–115
115
Fig. 2. Photomicrographs of the sections of a human control cerebral cortex, immunolabelled by anti-CLN2-N2 (A) and anti-CLN-I1 (B). Sections were counterstained lightly with hematoxylin. Scale bar, 15 mm.
ules in the cytoplasm. The staining completely disappeared in immunohistochemistry using the respective preimmune antisera and the antisera after absorption (data not shown). Thus, we raised antisera that recognized the CLN2 protein (Fig. 1A), and the immunohistochemical study indicated the presence of reactivity in human cerebral neurons and glial cells (Fig. 2). The dense granular labelling appeared to correspond to cytoplasmic organelles, such as lysosomes. The distribution of the CLN2 protein shown by this study appeared to be compatible with the neuropathological alterations of CLN2 [2], such as the accumulation of lipopigments within the cytoplasm of these cells. We also demonstrated, by Western blotting, the lack of this protein in the brain and hepatic tissues of a CLN2 patient (Fig. 1B). Currently, the diagnosis of NCLs including CLN2 depends on the demonstration of characteristic ultrastructures, which requires electron microscopic examination [3]. Therefore, clinical application of the antisera for the diagnosis of CLN2 may be beneficial. Since we only examined a single case of CLN2 in this study, further investigations are necessary. This study was supported by grants from the Ministry of Health and Welfare, and the Ministry of Education, Science, Sports, and Culture of Japan. [1] Consortium, The International Batten Disease, Isolation of a novel gene underlying Batten disease, CLN3, Cell, 82 (1995) 949–957. [2] Friede, R.L., Ceroid-lipofuscinosis and miscellaneous lipidosis. In Developmental Neuropathology, 2nd edn., Springer, Berlin, 1989, pp. 448–460.
[3] Goebel, H.H., Morphologic diagnosis in neuronal ceroid lipofuscinosis, Neuropediatrics, 28 (1997) 67–69. [4] Goebel, H.H. and Sharp, J.D., The neuronal ceroid-lipofuscinosis: recent advances, Brain Pathol., 8 (1998) 151–162. [5] Mitchison, H.M., Hofmann, S.L., Becerra, C.H.R., Munroe, P.B., Lake, B.D., Crow, Y.J., Stephenson, J.B.P., Williams, R.E., Hofman, I.L., Taschner, P.E.M., Martin, J.-J., Philippart, M., Andermann, E., Andermann, F., Mole, S.E., Gardiner, R.M. and O’Rawe, A.M., Mutations in the palmitoyl-protein thioesterase gene (PPT; CLN1) causing juvenile neuronal ceroid lipofuscinosis with granular osmiophilic deposits, Hum. Mol. Genet., 7 (1998) 291–297. [6] Mizuguchi, M., Takashima, S., Kakita, A., Yamada, M. and Ikeda, K., Lissencephaly gene product localization in the central nervous system and loss of immunoreactivity in Miller–Dieker syndrome, Am. J. Pathol., 147 (1995) 1142–1151. [7] Oka, A. and Takashima, S., Induction of cyclo-oxygenase 2 in brains with Down’s syndrome and dementia of Alzheimer type: specific localization in affected neurones and axons, NeuroReport, 8 (1997) 1161–1164. [8] Sleat, D.E., Donnelly, R.J., Lackland, H., Liu, C.-G., Sohar, I., Pullarkat, R.K. and Lobel, P., Association of mutations in a lysosomal protein with classical late-infantile neuronal ceroid lipofuscinosis, Science, 277 (1997) 1802–1805. [9] Vesa, J., Hellsten, E., Verkruyse, L.A., Camp, L.A., Rapola, J., Santavuori, P., Hofmann, S.L. and Peltonen, L., Mutations in the palmitoyl protein thioesterase gene causing infantile neuronal ceroid lipofuscinosis, Nature, 376 (1995) 584–587. [10] Wisniewski, K.E., Zhong, N., Kaczmarski, W., Kaczmarski, A., Kida, E., Brown, W.T., Schwarz, K.O., Lazzarini, A.M., Rubin, A.J., Stenroos, E.S., Johnson, W.G. and Wisniewski, T.M., Compound heterozygous genotype is associated with protracted juvenile neuronal ceroid lipofuscinosis, Ann. Neurol., 43 (1998) 106–110.