Molecular Genetics and Metabolism 102 (2011) 170–175

Contents lists available at ScienceDirect

Molecular Genetics and Metabolism

j ourna l homepage: www.e lsev ie r.com/ locate /ymgme

Pathology of the first autopsy case diagnosed as mucolipidosis type III α/β suggestingautophagic dysfunction

Hiroshi Kobayashi a,b,c,⁎, Junko Takahashi-Fujigasaki d, Takahiro Fukuda d, Ken Sakurai a,b, Yohta Shimada a,Koichi Nomura e, Masamichi Ariga a,b, Toya Ohashi a,b,c, Yoshikatsu Eto c, Takanobu Otomo f,Norio Sakai f, Hiroyuki Ida a,b,c

a Department of Gene Therapy, Institute of DNA Medicine, The Jikei University School of Medicine, Japanb Department of Pediatrics, The Jikei University School of Medicine, Japanc Department of Genetic Disease and Genome Science, The Jikei University School of Medicine, Japand Division of Neuropathology, The Jikei University School of Medicine, Japane Division of Pathology, The Jikei University School of Medicine, Japanf Department of Pediatrics, Osaka University Graduate School of Medicine, Japan

⁎ Corresponding author. Department of Gene TherapThe Jikei University School of Medicine, Japan. Fax: +8

E-mail address: hrkb@jikei.ac.jp (H. Kobayashi).

1096-7192/$ – see front matter © 2010 Elsevier Inc. Aldoi:10.1016/j.ymgme.2010.09.014

a b s t r a c t

a r t i c l e i n f o

Article history:Received 29 September 2010Accepted 29 September 2010Available online 3 November 2010

Keywords:Mucolipidosis type IIII-cell deseaseDorsal root ganglions (DRG)AutopsyAutophagosomeFoamy cell proliferationOdocytes

Mucolipidosis type III (MLIII) is amild form ofMucolipidosis type II (MLII, I-cell disease) of late onset, of whichalmost no pathological study has been reported, as it is a very rare disease. We encountered the case of a 23-year-old man of Japanese and Caucasian mixed parentage diagnosed with MLIII by enzyme assay andgenotyping. He died suddenly due to severe dilated cardiomyopathy. On the day after his death, autopsy wasperformed, and accumulation of Luxol Fast Blue (LFB) positive material was found to be most severe in theneuronal cells of dorsal root ganglions (DRG). Electromicroscopic DRG revealed the neuronal cytoplasm wasfilled with a zebra-body-like membranous matrix. We tried immunohistochemistry to investigate themechanism of such accumulation in the DRG that resulted in double positive anti-ubiquitin antibody (FK-2)and anti-LC3 antibody (as specific marker for autophagy) staining, and speculated activating ofautophagosome pathway, and ‘zebra-body’ should be suspected as dysfunctional autophagosome. We alsodetected foamy cell proliferation in the dura mater, Auerbach's plexus (peripheral nervous system),podocytes of almost all glomeruli, cartilage tissue in lumbar discs, and in cardiac muscle. We tried FK-2 andanti-LC3 antibody staining also for the podocytes, the area with the most marked proliferation of foamy cells,but the result was negative. This led us to speculate that these pathological findings, namely, accumulation ofLFB-positive material and foamy fibroblast proliferation, might be the forms of dysfunctional autophagosomeat various stages of development. This pathological study of MLIII supports the theory that MLIII is a mild typeof MLII because of the close similarity of their pathological findings.

y, Institute of DNA Medicine,1 3 3433 1230.

l rights reserved.

© 2010 Elsevier Inc. All rights reserved.

1. Introduction

Mucolipidosis type III (MLIII) is an autosomal recessive diseasecaused by deficiency of the lysosomal enzyme N-acetylglucosamine-1-phosphotransferase (GlcNAc-phosphotransferase), which is clinicallycharacterized by developmental delay and dysostosis multiplex,partially overlapping with mucopolysaccharidosis (MPS) [1]. GlcNAc-phosphotransferase acts in thefirst step of the synthesis of themannose6 phosphate (m6p) recognition marker on the lysosomal enzymes,which is recognized by m6p receptors for uptake and transport tolysosomes. In many eukaryotic cells, proper targeting of newlysynthesized acid hydrolases to the lysosome is mediated by the m6p

recognition system [2]. Newly synthesized lysosomal proenzymes thatlack m6p recognition markers are released into the extracellularmedium instead of being incorporated into lysosomes, as a result,lysosomes are enzyme deficient and cells accumulate undigestedstorage material, while elevated lysosomal enzyme levels are presentin the serum. In humans, deficient GlcNAc-phosphotransferase activitygives rise to the lysosomal storage diseases known as mucolipidosis IIand III. GlcNAc-phosphotransferase is a hexamic enzyme consisting ofthree subunits, α2, β2, and γ2. The α and β subunits are encoded byGNPTAB and the g subunit byGNPTG. A mutation inGNPTAB causes bothsevere type ML (MLIIα/β) and attenuated type ML (MLIIIα/β), while amutation in GNPTG causes only attenuated type ML (MLIIIγ) [3,4].

There have been some human pathological studies on MLII andanimal (mice) pathological studies on MLIII, but no human patholog-ical study on MLIII [5–7]. Recent reports have shown that abnormallysosomal storage interferes the autophagy pathway and the ubiquitin

171H. Kobayashi et al. / Molecular Genetics and Metabolism 102 (2011) 170–175

pathway in lysosomal storage diseases, while in MLII and IIIfibroblasts, inhibition of autophagosome formation restores mito-chondrial function [8]. We have experienced an autopsy case of MLIII,and in this study, tried to investigate the mechanism of pathologicalaccumulation, including the autophagy system.

2. Case presentation

This index case was that of a 23-year-old man of mixed race[Japanese and Caucasian (English)], who at 9 months old wasdetected delayed growth and development and coarse face, and hadsubsequently been diagnosed with mucolipidosis type III (MLIII) by

Fig. 1. X-ray of the patient. a. Narrow airway due to the accumulation of a soft tissuemass. b. Ccardiac failure. • LvDd: Left ventricular diastolic dimension (mm), • EF: ejection fraction (%)

enzyme assay and DNA diagnosis (Exon1: c.10ANC,p.K4Q, Exon13:c.2189delT, p.L730fs). His familial history was unremarkable.

When he was 7 years old, he had mastoiditis, his airway had beenprogressively narrowing due to accumulation of soft tissue mass(Fig. 1-a), and at 12, he developed symptoms of cardiopulmonarydysfunction. Since the age of 16 he had required O2 supply with NIV(non-invasive ventilation), and had suffered from frequent episodesof pneumonia, exudative otitis media, and auditory / visual impair-ment. He had scoliosis and severe bone dysplasia (Fig. 1-b,c). Hisgrowth retardation was very severe, with 88 cm of body length and19 kg of body weight, both were more than 3SD below the averagevalues for a 22-year-old man. At 20 he had auditory and visual

hest XP: scoliosis and cardiac hypertrophy. c. Dysostosis multiplex. d. Progression of his, • Bisphosphonate (pamidronate) therapy; 21y1m–21y3m.

172 H. Kobayashi et al. / Molecular Genetics and Metabolism 102 (2011) 170–175

impairment, but his intelligence was normal, he studied by corre-spondence every day and could have an intellectual conversationunder NIV. Mobility was aided by his electric wheel chair.

When he was 21, therapy with bisphosphonates for his severebone pain was stopped as it had not been effective. Furthermore,cardiac failure worsened as his ejection fraction (EF) decreased to 51%[9]. (Fig. 1-d) For the next 2 years his condition was relatively stable,but he died suddenly when he was 23 because of further deteriorationof cardiopulmonary failure due to dilated-hypokinetic hypertrophiccardiomyopathy (HCM)[10,11] , diagnosed by ultra sound cardiogra-phy (UCG) and pathological study. (Figs. 1-d and 2-a–e.)

With the permission of his family, we performed an autopsy of thiscase on the day after his death. To investigate the mechanism ofautophagic reaction in the accumulation, we tried immunohistochem-istry for the dorsal root ganglion (DRG), as peripheral nervous system;severe lipofustin granule distribution was detected by Luxol-Fast Blue(LFB) staining.

3. Methods

3.1. Genotyping (amplification and DNA sequencing)

These were done as previously reported in Ref. [12].

Fig. 2. General pathology, (all scale bars: 10 μm). a,b: End-stage hypertrophic cardiomyopa. Macroscopic pathology of the tricuspid atresia; severe calcification and hypertrophy. b. Mfibrosis in the cardiac interstitial substance (Masson stain). d. Cardiac muscle in the left ventrsevere proliferation of foamy fibroblasts. f–h. Foamy fibroblasts degeneration in the cartiladegenerated or ruined autophagosomes. *d–h: Foamy or vacuolar degeneration (arrow).

3.2. Immunohistochemistry

Sections of formalin-fixed, paraffin-embedded dorsal root ganglia(DRG) and kidneys were used for immunohistochemical staining. Thesections were autoclaved in 10 mM citrate buffer (pH6.0) for 10 minfor pre-treatment. Then, the sections were incubated with anti-LC3antibody (rabbit polyclonal, 1:500, Medical & Biological LaboratoriesCo., Ltd., Nagoya, Japan, code no. PM036), a specific marker forautophagosome, or anti-poly-ubiquitin antibody (mouse monoclonal1:2000, FK2, Medical & Biological Laboratories Co., Ltd.), and stainedwith 2,3-diaminobenzidine (DAB) hydrochloride as the chromogen.To clarify the spatial relationship of LC3 and poly-ubiquitin,immunofluorescence double-labelling was performed. For the doublelabelling, secondary antibodies conjugated with Alexa 488 (Invitro-gen, Carlsbad, CA) and Cy3 (Jackson ImmunoResearch Laboratories,Inc., West Grove, PA) were used.

4. Results

4.1. General pathology

The patient suffered from severe HCM in the end stage. The left andright ventricles were severely dilated and the trabeculae carneae was

athy (dilated type) resulted in cardiac failure and was the direct cause of his death:acroscopic pathology of the mitral valve; severe calcification and hypertrophy. c. Severeicle; destruction and vacuolar degeneration of myocytes. e. Cardiac valve (mitral valve);ge (f), Auerbach's plexus (g), and podocytes (h, most severe), was assumed to be the

173H. Kobayashi et al. / Molecular Genetics and Metabolism 102 (2011) 170–175

hypertrophic. The mitral, aortic and tricuspid valves were thick due tosevere calcification (Fig. 2-a,b). Microscopically, vacuolar degenera-tion of myocytes and severe interstitial fibrosis were observed in theventricles (Fig. 2-c,d). In the thickened heart valves, large amounts ofvacuolated fibroblasts were observed (Fig. 2-e). Similar proliferationof vacuolated fibroblasts was also observed in the modestly thickeneddura mater.

Vacuolated changes were detected in chondrocytes of cartilages(Fig. 2-f), and in Schwann cells in Auerbach's plexus of the intestines(Fig. 2-g). In the kidneys, most of the glomeruli showed strikingchanges with marked foamy transformation of podocytes (Fig. 2-h).No remarkable alteration was observed in the renal tubular cells andendothelial cells. His renal function was normal except for the

Fig. 3. Neuropathology 1. a. In hematoxillin-eosin stained tissue, we detected swelling cytopthe dorsal root ganglion (DRG) and autophagic function. b. In Kluver-Barrera's (KB) stained tc. In neuronal cells of the DRG, we detected striking accumulation of zebra bodies by electronLC3 antibody and 2,3-diaminobenzidine (DAB) hydrochloride as the chromogen: LC3 immpresent case (MLIII) stained with anti-LC3 antibody and DAB hydrochloride: in the cytoplasmin the control. f. DRG of an age-matched control stained with FK2 and DAB hydrochloride: d(MLIII) stained with FK2 and DAB hydrochloride: in the ganglion cells, dot-like accumulatio

secondary renal failure due to cardiac failure with end-stage HCM. Thesections of kidney were immunostained with the LC3 and FK2antibodies, however accumulation of FK2 positive materials wasinconspicuous (data not shown). There was also a spermatogenesisdisorder, as well as extra-medullary hematopoiesis (liver and spleen),and hemophagocytic hystiocyte proliferation in his lymph nodes.

4.2. Neuropathology

The brain weighed 1450 g, and had a normal appearance onmacroscopic examination. The dura mater was modestly thickened.

Microscopically, the cerebral cortex, white matter, basal ganglia,thalamus, hypothalamus, brain stem and cerebellum were essentially

lasm of the thalamus (ventral lateral nucleus), (scale bar: 20 μm). b–g. Accumulation inissue, too many lipofustin granules were observed in neuronal cells, (scale bar: 20 μm).microscopy (EM), (scale bar: 1 μm). d. DRG of an agematched control stained with anti-unopositive fine dots were detected in the cytoplasm of ganglion cells. e. DRG of theof ganglion cells, the amount of LC3 immunopositive fine dots was almost the same as

ot-like accumulation was less apparent in the ganglion cells. g. DRG of the present casen of poly-ubiquitin was revealed by the FK2. *d–e: scale bar: 20 μm.

174 H. Kobayashi et al. / Molecular Genetics and Metabolism 102 (2011) 170–175

normal. However, neuronal accumulation of lipofuscin granules wasconspicuous for his age, (22 years old). Moderate accumulation wasobserved in neurons of the thalamus (Fig. 3-a) and hypothalamus.Occasionally, Luxsol Fast Blue positive materials (LFB-PSMs) wererevealed in the neurons by Kluver–Barrera stain. Mild lipofuscindeposition was detected in the lateral and internal geniculate nuclei,dentate nucleus, locus coeruleus, and spinal anterior horn. The DRGshowed most the prominent storage of lipofuscin and LFB-PSMs(Fig. 3-b). Tissue blocks were sampled from the DRG and embedded inEpon. Ultra-thin sections were cut and examined by electronmicroscopy. Ultrastructurally, the ganglion cells contained amountsof zebra-like inclusion bodies (Fig. 3-c). Then, we performedimmunohistochemical analysis of the DRG using anti-LC3 and anti-ubiquitin antibodies (FK2). In both age matched control and this caseof ML III patient, LC3 immunopositive fine dots were detected in thecytoplasm of ganglion cells (Fig. 3-d,e). In ganglion cells, dot-likeaccumulation of poly-ubiquitin was revealed by FK2 (Fig. 3-g),although accumulation was less apparent in DRG of control patients(Fig. 3-f). The FK2 positive dot-like structures were merged with LC3positive structures (Fig. 4-c,f). From these results, autophagy build upmight be active in the DRG, or in age-matched controls, it might beactivated by severe stress just before death (Fig. 3-d), but in this caseof MLIII, the built-up autophagosome must have been dysfunctionalbecause it included the accumulation of poly-ubiquitin in its body.

5. Discussion

Mucolipidosis type III has been understood as late onset and mildtype of mucolipidosis type II (MLII, I-cell disease), although nopathological study has been reported [1]. We encountered the case ofa 23-year-old man of Japanese and Caucasian mixed parentagediagnosed asMLIII by enzymeassay and genotyping,whodied suddenly

Fig. 4. Neuropathology 2. a. and d. Immunohistochemistry (IHC) with FK2, indicating the ppositive dot staining indicating the presence of the active form of LC3 (LC3-II) distributelocalization for ubiquitin and autophagosome (scale bar: 10 μm).

due to severe dilated period hypertrophic cardiac myopathy. Thispatient, as already mentioned, was of mixed race and his genotype wasnot found in the Japanese population [12], but was reported by Kudo etal. and Cathey et al. [13,14] As for the enzyme assay of GlcNAc-phosphotransferase, we could not perform it because of some clinicalreasons, but the deficiency had been detected in France at around thetime he was one year old (data not shown). Concerning generalpathology, we detected foamy cell proliferation in various organs suchas the dura mater, Auerbach's plexus (peripheral nervous system),podocytes in almost all glomeruli, cartilage tissue of his lumbar disc, andcardiacmuscle.We identified threemain causes of death at the autopsy,(i) severe bone dysplasia as the remote cause of cardio-pulmonaryfailure, (ii) cardiomyopathy (end-stage, dilated-hypokinetic HCM), and(iii) pulmonary failure. The influence of intravenous pamidronatetreatment was unclear, but we thought it was not the direct trigger forhis death. Dilated-hypokinetic HCM and bone dysplasia should havebeen influenced by the accumulation of proliferative foamy fibroblasts,although we detected no accumulation in the lungs or bronchi, andspeculated pulmonary failure was a secondary symptom due tohemorrhage and scoliosis. Besides, despite the severe accumulation inhis podocytes, his renal function was normal until his death. Wespeculated it was due to the normal structure of renal tubules and bloodvessel endothelium. These foamy fibroblasts have been reported to bethe result of the accumulation of various lysosomal substrates. [1,15]Martin et al. [15] reported that the podocyteswere essentially normal inthe longest surviving patient (10 years), while patients who died atearlier ages showed heavy vacuolation, but comparingwith our study, itappears to represent the only histologic difference as the author of thispaper reported [15]. To investigate the progression of autophagy, wetried immunohistochemistry for the podocytes with the most severefoamyfibroblast proliferation, using anti-LC3 antibody and FK-2, but didnot detect cells showing positive staining. It led us to speculate that

resence of ubiquitin. b. and e. IHC with anti-LC3 antibody conjugated with Alexa 488:d in the membrane of autophagosomes. c. and f. Merge phase suggesting the same

175H. Kobayashi et al. / Molecular Genetics and Metabolism 102 (2011) 170–175

these foamy cells might be the final form or the slough of autophago-somes. In the neuropathology examination, we detected accumulationof LFB positive material in neuronal cells, thalamus (ventral lateralnucleus Fig. 3-a), hypothalamus, anterior horn of the spine, and DRG(Fig. 3-b), but not in the cortex. These findings were consistent with hisnormal intelligence, although it was unclear whether he had sensorydisturbance or not. We tried immunohistochemistry to investigate themechanism of the severe accumulation in the DRG, and detectedpositive staining for both FK-2 and anti-LC3 antibody stain, suggestingactivation of both the ubiquitin-proteosome and autophagosomepathways. This finding led us to speculate that the presence of ‘zebra-bodies’ or inclusion bodies should be suspected as dysfunctionalautophagosome. Hara et al. [16] reported that loss of autophagy causesneurodegeneration even in the absence of any disease-associatedmutant protein. In that report, mice deficient for Atg5 (autophagy-related 5), specifically in neural cells, developed progressive deficits inmotor function that was accompanied by the accumulation ofcytoplasmic inclusion bodies in neurons, suggesting that continuousclearance of diffuse cytosolic proteins is important for preventing theaccumulation of abnormal proteins that lead to neurodegeneration.

In this study we pathologically analyzed a case of humanMLIII, whichhas been thought to be a mild type of MLII because the pathologicalfindings are almost the same [15,17]. Recent reports have shown thatabnormal lysosomal storage interferes with the autophagy pathway andtheubiquitin pathway in lysosomal-storagediseases. [18–20]Otomoet al.demonstrated accumulation of autolysosomes and increased levels of p62and ubiquitin proteins in cultured fibroblasts, and that these autophagicelevationsweremilder inMLIII comparedwithMLII [8]. They also showedinhibitionof autophagic formation restoresmitochondrial function inMLIIand III skin fibroblasts, suggesting that mitochondria were directlyimpaired by increased autophagic formation. In another report, MLIIandMLIII skin fibroblastswere found to havemany inclusion bodies filledwith substrates [21]. It is possible that these storagematerials are involvedin complex downstream pathways of autophagy. In this autopsy case, wedetected autophagic dysfunction especially in the DRG, but found no signof autophagy in other organs such as podocytes, suggesting that thesepathologicalfindings, i.e., accumulationof LFBpositivematerial and foamycell proliferation, might be forms of dysfunctional autophagosomes atvarious stages of development.

In conclusion, we presented the first report of human autopsy case ofMLIII whose general pathology and neuropathology demonstrated itcorresponded to amild type ofMLII, and these pathological findingswerethought to be explained as dysfunctional autophagic mechanismthroughout. These findings raise the possibility of a new approach forthe analysis and therapy of MLIII patients.

Acknowledgements

We thank Ms K. Ishibashi and Ms N. Takabayashi in Division ofNeuropathology, Jikei University school of Medicine, for technicalassistance. This study is supported by the grants from the Research onMeasures for Intractable Diseases, the Ministry of Health, Labour andWelfare in Japan.

References

[1] S. Kornfeld, W.S. Sly, I-cell diseases and pseudo-Hurler polydystrophy: disordersof lysosomal enzyme phosphorylation, in: C.R. Scriber, A.L. Beaudet, W.S. Sly, D.Valle (Eds.), The Online Metabolic and Molecular Bases of Inherited Disease(OMMBID), McGraw-Hill, New York, 2010, http://www.ommbid.com/.

[2] S. Kornfeld, Trafficking of lysosomal enzymes in normal and disease states, J. Clin.Invest. 77 (1986) 1–6.

[3] M. Kudo, M. Bao, A. D'Souza, F. Ying, H. Pan, B.A. Roe,W.M. Canfield, The alpha-andbeta- subunits of the human UDP-N-acetylglucosamine:lysosomal enzyme N-acetylglucosamine-1-phosphotransferase are encoded by a single cDNA, J. Biol.Chem. 280 (2005) 36141–36149.

[4] A. Raas-Rothschild, V. Cormier-Daire, M. Bao, E. Genin, R. Salomon, K. Brewer, M.Zeigler, H.Mandel, S. Toth, B. Roe, A.Munnich,W.M.Canfield,Molecular basis of variantpseudo-Hurler polydystrophy (mucolipidosis IIIC), J. Clin. Invest. 105 (2000) 673–681.

[5] P. Vogel, B.J. Payne, R. Read, W.S. Lee, C.M. Gelfman, S. Kornfeld, Comparativepathology of murine mucolipidosis types II and IIIC, Vet. Pathol. 46 (2009) 313–324.

[6] C.M. Gelfman, P. Vogel, T.M. Issa, C.A. Turner, W.S. Lee, S. Kornfeld, D.S. Rice, Micelacking a/b subunits of GlcNAc-1-phosphotransferase exhibit growth retardation,retinal degeneration, and secretory cell lesions, Invest. Ophthalmol. Vis. Sci. 48(2007) 5221–5228.

[7] K. Nagashima, K. Sakakibara, H. Enodo, Y. Konishi, Y. Suzuki, T. Abe, I-cell disease(mucolipidosis II). Pathological and biochemical studies of an autopsy case, ActaPathol. Jpn 27 (1977) 251–264.

[8] T. Otomo, K. Higaki, E. Namba, K. Ozono, N. Sakai, Inhibition of autophagosomeformation restores mitochondrial function mucolipidosis II III skin fibroblasts,Mol. Genet. Metab. 98 (4) (2009) 393–399.

[9] C. Robinson, N. Baker, J. Noble, A. King, G. David, D. Sillence, P. Hofman, T. Cundy,The osteodystrophy of mucolipidosis type III and the effects of intravenouspamidronate treatment, J. Inherit. Metab. Dis. 25 (2002) 681–693.

[10] H. Ashrafian, M.J. Mason, A.G. Mitchell, Regression of dilated-hypokinetichypertrophic cardiomyopathy by biventricular cardiac pacing, Europace 9(2007) 50–54.

[11] K.M. Harris, P. Spirito, M.S. Maron, A.G. Zenovich, F. Formisano, J.R. Lesser, S.Mackey-Bojack, W.J. Manning, J.E. Udelson, B.J. Maron, Prevalence, clinical profile,and significance of left ventricular remodeling in the end-stage phase ofhypertrophic cardiopathy, Circulation 114 (2006) 216–225.

[12] T. Otomo, T. Muramatsu, T. Yorifuji, T. Okuyama, H. Nakabayashi, T. Fukao, T.Ohura, M. Yoshino, A. Tanaka, N. Okamoto, K. Inui, K. Ozono, N. Sakai,Mucolipidosis II and III alpha/beta: mutation analysis of 40 Japanese patientsshowed genotype-phenotype correlation, J. Hum. Genet. 54 (2009) 145–151.

[13] M. Kudo, M.S. Brem, W.M. Canfield, Mucolipidosis II (I-cell disease) andmucolipidosis IIIA (classical pseudo-hurler polydystrophy) are caused bymutations in the GlcNAc-phosphotransferase alpha/beta-subunits precursorgene, Am. J. Hum. Genet. 78 (2006) 451–463.

[14] S.S. Cathey, J.G. Leroy, T. Wood, K. Eaves, R.J. Simensen, M. Kudo, R.E. Stevenson, M.J.Friez, Phenotype and genotype in mucolipidoses II and III alpha/beta: a study of 61probands, J. Med. Genet. 47 (2010) 38–48.

[15] J.J. Martin, J.G. Leroy, M. van Eygen, C. Ceuterick, I-cell disease. Further reportpathology, Acta Neuropathol. 64 (1984) 234–242.

[16] T. Hara, K. Nakamura, M. Matsui, A. Yamamoto, Y. Nakahara, R. Migishima, M.Yokoyama, K. Mishima, I. Saito, H. Okano, N. Mizushima, Suppression of basalautophagy in neural cells causes neurodegenerative disease in mice, Nature 441(2006) 885–889.

[17] J.J. Martin, J.P. Leroy, P. Farriaux, G. Fontaine, J. Desnick, A. Cabello, I-cell disease(mucolipidosis II): a report on its pathology, Acta Neuropathol. 33 (1975) 285–305.

[18] C. Settembre, A. Fraldi, L. Jahreiss, C. Spampanato, C. Venturi, D. Medina, R. dePablo, C. Tacchetti, D.C. Rubinsztein, A. Ballabio, A block of autophagy in lysosomalstorage disorders, Hum. Mol. Genet. 17 (2008) 119–129.

[19] P. Bifsha, K. Landry, L. Ashmarina, S. Durand, V. Seyrantepe, S. Trudel, C. Quiniou, S.Chemtob, Y. Xu, R.A. Gravel, R. Sladek, A.V. Pshezhetsky, Altered gene expressionin cells from patients with lysosomal storage disorders suggests impairment ofthe ubiquitin pathway, Cell Death Differ. 14 (2007) 511–523.

[20] A. Ballabio, V. Gieselmann, Lysosomal disorders: from storage to cellular damage,Biochim. Biophys. Acta 1793 (2009) 684–696.

[21] I. Kawashima, M. Ohsawa, T. Fukushige, Y. Nagayama, Y. Niida, M. Kotani, Y.Tajima, T. Kanekura, T. Kanzaki, H. Sakuraba, Cytochemical analysis of storagematerials in cultured skin fibroblasts from patients with I-cell disease, Clin. Chim.Acta 378 (2007) 142–146.