The Journal of Neuroscience, June 19, 2013 • 33(25):10195–10208 • 10195

Disease Focus

Editor’s Note: Disease Focus articles provide brief overviews of a neural disease or syndrome, emphasizing potential links to basic neural mechanisms. They are presented in the hope of helping researchers identify clinical implications of their research. For more information, see http://www.jneurosci.org/misc/ifa_minireviews.dtl.

Gangliosides and Gangliosidoses: Principles of Molecular and Metabolic Pathogenesis

Konrad Sandhoff1 and Klaus Harzer2 1LIMES, Kekule´-Institut, 53121 Bonn, Germany and 2Neurometabolic Laboratory, Klinik fu¨r Kinder- und Jugendmedizin, University of Tu¨bingen, 72076 Tu¨bingen, Germany

Gangliosides are the main of neuronal plasma membranes. Their surface patterns are generated by coordinated processes, involving biosynthetic pathways of the secretory compartments, catabolic steps of the endolysosomal system, and intracellular traffick- ing. Inherited defects in biosynthesis causing fatal neurodegenerative diseases have been described so far almost exclusively in mouse models, whereas inherited defects in ganglioside catabolism causing various clinical forms of GM1- and GM2-gangliosidoses have long been known. For digestion, gangliosides are endocytosed and reach intra-endosomal vesicles. At the level of late endosomes, they are depleted of membrane-stabilizing like cholesterol and enriched with bis(monoacylglycero)phosphate (BMP). Lysosomal catabolism is catalyzed at acidic pH values by cationic activator proteins (SAPs), presenting lipids to their respective , electrostatically attracted to the negatively charged surface of the luminal BMP-rich vesicles. Various inherited defects of ganglioside hydrolases, e.g., of ␤-galactosidase and ␤-, and of GM2-activator protein, cause infantile (with tetraparesis, , blindness) and different protracted clinical forms of GM1- and GM2-gangliosidoses. yielding proteins with small residual catabolic activities in the give rise to juvenile and adult clinical forms with a wide range of clinical symptomatology. Apartfrompatients’differencesintheirgeneticbackground,clinicalheterogeneitymaybecausedbyratherdiversesubstratespecificities and functions of lysosomal hydrolases, multifunctional properties of SAPs, and the strong regulation of ganglioside catabolism by membrane lipids. Currently, there is no treatment available for neuronal ganglioside storage diseases. Therapeutic approaches in mouse models and patients with juvenile forms of gangliosidoses are discussed.

Introduction bers of -laden cells in the nervous gand (1963). Six major and many minor 1Gangliosides had been discovered by system and the visceral organs, mental re- ganglioside species have been identified in Ernst Klenk in the 1930s (Klenk, 1937, tardation, and impaired vision or blind- mammalian brains (Yu et al., 2011). Inher- 1939) when he analyzed postmortem ness. With the help of biochemical, ited defects in ganglioside can brain tissues of patients with Tay–Sachs enzymatic, molecular genetic, morpho- cause fatal neurodegenerative diseases. Best disease, a fatal infantile form of amaurotic logical, and animal studies, the clinical known so far are inherited defects in lyso- idiocy. At that time the clinical diagnosis picture of amaurotic idiocy has been re- somal ganglioside catabolism, the GM1- “amaurotic idiocy” comprised an inher- solved in the last five decades into two ma- and GM2-gangliosidoses. ited disease characterized by high num- jor groups of diseases at the molecular Up to now, no defects in ganglioside level: lysosomal ganglioside storage dis- biosynthesis have been described in hu- Received Feb. 19, 2013; revised April 30, 2013; accepted May 9, 2013. eases, comprising GM1-gangliosidoses mans with the exception of two reports on This work was supported by the Deutsche Forschungsgemeinschaft. and 4 forms of GM2-gangliosidoses, and three patients with an infantile onset re- Dedicated on the occasion of her 60th birthday to Ingeborg Kra¨geloh- 10 forms of neuronal ceroid lipofuscino- fractory epilepsy (Simpson et al., 2004; Mann, professor of Neuropediatrics, Tu¨bingen. Correspondence should be addressed to either of the following: Konrad ses poorly characterized so far (Jalanko Fragaki et al., 2013), caused by a defi- Sandhoff, LIMES, Kekule´-Institut, Gerhard-Domagk-Str. 1, 53121 Bonn, and Braulke, 2009). ciency of ganglioside GM3-synthase. It Germany,E-mail:[email protected];orKlausHarzer,Neurometabolic Gangliosides are typical components of is associated with psychomotor delay, Laboratory, Klinik fu¨r Kinder- und Jugendmedizin, University of Tu¨bingen, neuronal plasma membranes. The first gan- blindness, deafness, and respiratory chain 72076 Tu¨bingen, Germany, E-mail: [email protected]. DOI:10.1523/JNEUROSCI.0822-13.2013 glioside structure, that of ganglioside GM1 dysfunction. In addition, one patient with Copyright © 2013 the authors 0270-6474/13/3310195-14$15.00/0 (Fig. 1), was elucidated by Kuhn and Wie- a genetically undisclosed basis of infantile 10196 • J. Neurosci., June 19, 2013 • 33(25):10195–10208 Sandhoff and Harzer • Gangliosides and Gangliosidoses

Figure1. Pathwayoflysosomalsphingolipiddegradation.TheeponymsofknownmetabolicdiseasesandthoseofSAPs(red)necessaryforinvivodegradationareindicated.Hydrolasesaregiven in green. Heterogeneity in the lipid part of the is not indicated. Variant AB, Variant AB of GM2- ͓deficiency of GM2-activator protein (GM2-AP)͔; Sap, sphingolipid activator protein (modified from Sandhoff and Kolter, 1995). Sandhoff and Harzer • Gangliosides and Gangliosidoses J. Neurosci., June 19, 2013 • 33(25):10195–10208 • 10197

GM2-synthase deficiency, a ganglioside all a-series and b-series gangliosides triggers systems are also involved, but with minor GM3 storage and loss of higher ganglio- the formation of 0-series gangliosides, clinical relevance, though lysosomal stor- sides has been described, passed away at which are not observed in the wild-type age bodies are demonstrable at places. age 3.5 months (Fishman et al., 1975). brain. Their additional synthesis, however, In addition to the neuronal involve- However, blocks in ganglioside bio- cannot compensate the loss of all the major ment, an extraneural involvement (Table synthesis generated in -manipulated neuronal ganglioside structures. Surprising 1) is striking in infantile and partly present mice cause a broad spectrum of patholog- was that the total ganglioside content in the in juvenile patients with GM1-gangli- ical phenotypes (Proia, 2003). brains of the various mutant mice stayed al- osidosis; even an intrauterine manifesta- most unchanged at the wild-type level tion with hydrops fetalis is possible, and a Ganglioside biosynthesis and its defects (Sandhoff, 2012). dysmorphic (gargoyle-like) face is regular in genetically manipulated mice The ubiquitous block of GlcCer syn- in early infantile patients. Juvenile and Glycolipids are expressed in a cell-type- thesis, the precursor of the main glyco- adult GM1-gangliosidosis patients have a and stage-specific manner, gangliosides sphingolipids (GSLs), is lethal in mice at predominant or almost pure neurologic being the main glycolipids of neuronal embryonic day 6 (Yamashita et al., 1999). picture (Suzuki et al., 2001). Extraneural surfaces. Their surface patterns are gener- However, its cell-type-specific knock out manifestations are absent in GM2- ated by coordinated processes, involving in neurons (Jennemann et al., 2005), gangliosidosis with the exception of vari- biosynthetic and catabolic steps and intra- hepatocytes (Jennemann et al., 2010), or ant 0, also called , in cellular trafficking. keratinocytes (Jennemann et al., 2007) which slight visceromegaly is frequent Their hydrophobic membrane an- still allowed complete embryogenesis but (Gravel et al., 2001) and some involve- chors, the residues, are formed with the birth of fatally sick animals. ment of bones and even the heart can oc- in the endoplasmic reticulum (ER), glyco- Apparently, some complex glycolipids de- cur (Venugopalan and Joshi, 2002). sylation of ceramide takes place at the rived from GlcCer such as the stage- The clinical phenotypes in the gangli- Golgi and trans-Golgi network (TGN), specific embryonic antigens (Kannagi et osidoses can, though with some caveats, and their constitutive degradation mainly al., 1983; Eggens et al., 1989) involved in be correlated to the different types and in the endolysosomal compartment (Kolter cellular adhesion processes are needed at numbers of substrates, which are accumu- and Sandhoff, 1999). After formation of early embryonic stages for development lated in these diseases. Examples of the glucosylceramide (GlcCer) from cer- of a multicellular organism (Kolter et al., substrates of the acid hydrolases whose amide (Cer) at the cytosolic surface of 2000). deficiencies define the diseases are ganglio- Golgi membranes and its flip to the lumi- Gangliosides have been discovered as sides, asialogangliosides, other glycolipids, nal surface of Golgi membranes, GlcCer GSLs of neuronal membranes by lipid specific oligosaccharides and, in case of acid gets converted to lactosylceramide (Lac- analysis of Tay–Sachs disease (Klenk, ␤-galactosidase, mucopolysaccharides. In Cer). At the luminal surfaces of Golgi and 1937, 1939), but their complex chemical the deficiencies, the patterns of accumulated TGN membranes, LacCer is sialylated to structure was elucidated only later in the substrates can be explained in a simplified form small precursor gangliosides (GM3, 1960s (Kuhn and Wiegandt, 1963). Their format as follows. The main substrates GD3, and GT3). They are then converted storage in neuronal paved the of the acid ␤-galactosidase, which are by membrane-bound promiscuous glyco- way for the analysis of their cellular deficient in GM1-gangliosidosis, are GM1- syltransferases to complex ganglioside metabolism and lysosomal catabolism ganglioside (Fig. 1), specific oligosaccha- structures, which reach cellular surfaces (Sandhoff et al., 1969; Sandhoff, 1977) rides, and keratan sulfate, so that this disease by vesicular transport of the secretory is marked not only by ganglioside storage pathway. This concept of “combinatorial Clinical phenotypes, correlations to (whose main pathologic correlate is neuro- ganglioside biosynthesis” (Kolter et al., pathological and biochemical nal degeneration) but also by oligosacchari- 2002) allows the generation of cell- phenotypes dosis and mucopolysaccharidosis, with the specific patterns on neuronal surfaces de- All types of gangliosidoses are caused by extraneuronal clinical involvement due pending on the availability of precursor inherited defects in ganglioside catabo- to the latter. The main substrate of the molecules, expression of biosynthetic en- lism (Fig. 1) and can manifest with severe, ␤- A deficient in GM2- zymes and their regulation. Generation extremely variable clinical symptoms (Ta- gangliosidosis variant B (also called and analysis of mutant mice with defects ble 1) in almost any age group, although Tay–Sachs disease) or suppressed in the in the pathways of ganglioside biosynthe- most patients have an early infantile onset GM2-activator-deficient variant AB, is sis supported this model (Kolter et al., and a late infantile or juvenile, fatal end of GM2-ganglioside, so the disease is “only” a 2002; Sandhoff and Kolter, 2003). Mutant their disease, which is dominated by the ganglioside storing disorder. The main sub- mice defective in b-series gangliosides central nervous involvement (Fig. 2A) strates of the ␤-hexosaminidases A and B were viable and without apparent neuro- (Lyon et al., 1996a). The cerebral degen- deficient in GM2-gangliosidosis variant 0 logical abnormalities (Kawai et al., 2001; eration is marked by the lipid storage are GM2-ganglioside and oligosaccharides, Okada et al., 2002) and reach an almost present in neurons with the ultrastruc- so the disease includes gangliosidosis and normal life span. Mice defective in turally characteristic “membranous cy- oligosaccharidosis, with some visceral in- GalNAc-transferase, missing the major toplasmic bodies,” which are secondary, volvement due to the latter. brain gangliosides, were viable but suf- “storage” lysosomes (Fig. 2B). These stor- Another correlation can, though with fered from instability of brain structures age bodies distend the neurons (Fig. 2C), some reasonable doubts, be assumed be- with detachment of myelin sheets from which eventually are lost, but the exact tween the time (as patients’ age) of onset axonal membranes (Schnaar, 2010) and reason is unclear (for example, cell inac- or manifestation, or the duration (chro- male infertility. A combined defect of tivity by ganglioside-induced impairment nicity) of the disease, and the rudimentary GM3-synthase and GM2-synthase causes of synaptic transmission is discussed). catabolic functions of the defective (if not an early fatal disease. Here, the absence of The peripheral and autonomous nerve completely absent) acid hydrolases in the 10198 • J. Neurosci., June 19, 2013 • 33(25):10195–10208 Sandhoff and Harzer • Gangliosides and Gangliosidoses

Table 1. The main clinical hallmarks and symptoms in gangliosidoses Disease Manifestation Regular symptoms/signs Facultative symptoms/signs Remarks GM1-gangliosidosis Infantilea,b Dysmorphic face, severe dysostosis, hepato- Cherry-red macular spotc, nystagmus, Rapid or 1–2 year downhill course with megaly, unable to sit, muscle hypotonia, , , joint contrac- psychomotor deterioration, general later on spasticity, visual failure, foam cells tures, hematologic signs, vacuolated dystrophy, feeding problems in bone marrow, oligosacchariduria, blood lymphocytesd mucopolysacchariduria Juvenilea By 4 years or later: dysarthria, extrapyramidal Slight intellectual impairment, mild bone Some patients have also parts or attenu- signs including changes, foam cells in bone marrow ated forms of the symptoms listed for the infantile disease, other patients belong to the chronic group Adulta As in juvenile disease, but with more insidious Psychosis Disease is chronic rather than with only onset adult manifestation Chronic Combines late infantile, juvenile, and sometimes adult manifestations in which the above listed symptoms are partly or gradually developed GM2-gangliosidosis, variants B, Infantilee Loss of head/trunk control, smiling, by about 6 Megalencephaly, visceral involvement and Downhill course mostly less rapid than in B1 and 0 months, cherry-red macular spot c, high oligosacchariduria only in variant 0 early GM1-gangliosidosis startle response to noise, muscle hypotonia by 1 year, epilepsy, tetraparesis with some spasticity, dementia, blindness Juvenile/subacute Onset by 3 to 6 years, dysarthria, loss of Seizures, irritability, psychiatric signs, Clinical variability even in siblingsg, above speech, spastic paraparesis, later on denervation muscle atrophy (EMG), remark to juvenile GM1-gangliosidosis paraplegyf, pyramidal signsf, cerebellar areflexia, cherry red macular spotc here also apt , mental deterioration Adult Clumsiness, lower motor neuron disease, Psychosis High clinical variability, oldest patient was denervation muscle atrophy, cerebellar 76 years ataxia Chronic See remark to chronic GM1-gangliosidosis GM2-gangliosidosis, variant AB Most as in variants B, B1, 0 Cherry red macular spotc No adult patients known aGrouping of GM1-gangliosidosis patients as type I, II, and III of the disease according to the age at clinical onset is questionable by the often ambiguous onset or insidious or chronic manifestation through all, except the early infantile, age groups (Lyon et al, 1996b). bEarlier named “generalized gangliosidosis.” cSee Fig. 2D. dSee Fig. 2E. eEarlier named “infantile amaurotic idiocy”; relatively high prevalence of variant B (Tay–Sachs disease) in Ashkenazim. fSee Fig. 2A. gMany patients are Ashkenazim, but a genetic isolate of variant B1 is in Spain. patients. This “residual activity hypothe- substrate (e.g., the ganglioside/activator ing the presence of normal sialidase activity, sis” postulates higher residual functions protein complex) binding sites, protein- or by molecular gene analysis. for late starting/manifesting or chronic folding-relevant sites, or subdomains in the forms of gangliosidoses than for early (in- , which are required for the catabo- Summary outline of fantile) disease forms (see Protracted clin- lism of different substrates, and suggests clinical symptomatology ical forms of gangliosidoses and the that these sites are differentially or alterna- The main symptomatology of the gangli- threshold theory, below). Although the tively defective in different patients who osidoses is given as a partially simplified low residual, probably pathophysiologi- have then very variable diseases caused by summary in Table 1. Clinical details cally relevant activities of the defective hy- their ␤-galactosidase defects. On the other should be found in the special literature drolases cannot always reliably be assessed hand, a special GM1-ganglioside-storing (Lyon et al., 1996a, b; Gravel et al., 2001; under laboratory conditions in vitro, that disease (associated with oligosaccharide Suzuki et al., 2001). Pathologists have hypothesis seemed to be acceptable for storage) results from an indirect deficiency found that in gangliosidosis the brain is many patients who were studied enzymat- sometimes slightly infiltrated with inflam- of acid ␤-galactosidase (and sialidase, and a ically. As to those residual functions, it is matory cells considered to be a response specific sulfatase) in galactosialidosis, which interesting that in late manifesting GM1- to the dysfunction of the lipid-laden neu- is caused by the genetic deficiency of the so- gangliosidosis the almost only substrate rons and to the abundance of antigeni- called protective protein (identical to car- relevant for the biochemical phenotype is cally acting gangliosides. Here follow the ganglioside accumulation, particu- boxypeptidase A) (d’Azzo et al., 2001). some supplementary clinical remarks. larly in neurons, while for the breakdown When this protein is defective, the func- Neuroimaging has no essential role in the of other substrates the residual functions tional complex of the normal protein with characterization of gangliosidoses, but in ␤ seem to be nearly sufficient. acid -galactosidase and sialidase (and the late disease stages cerebral and cerebellar However, in a certain type of genetic sulfatase) cannot be formed, and these en- atrophy can often be documented and acid ␤-galactosidase deficiency, also a suf- zymes, though having a normal structure, slight T2-weighted hyperintensities may ficient catabolism of GM1-ganglioside is are subject to rapid decay. The resulting appear earlier in the brainstem and white possible, whereas the breakdown of other clinical phenotype is very similar to that in matter (Wilken et al., 2008). Magnetic substrates is severely impaired, and the re- classic GM1-gangliosidosis (and to that in resonance spectroscopy (N-acetyl aspar- sulting disease is not gangliosidosis but mu- the genetic sialidase defect, sialidosis). So an tate, myoinositol, and other signals) may copolysaccharidosis IV (Morquio) type B enzymatic diagnosis of GM1-gangliosidosis be helpful in documenting neuronal loss (Suzuki et al., 2001). This indicates different has always to be confirmed by demonstrat- and inflammation (Assadi et al., 2008). Sandhoff and Harzer • Gangliosides and Gangliosidoses J. Neurosci., June 19, 2013 • 33(25):10195–10208 • 10199

Figure2. Clinicalandmorphologicalobservationsingangliosidoses.A,NearendstageinaSpanish13-year-oldpatientwithGM2-gangliosidosisvariantB1(photographcourtesyofH.H.Goebel, Mainz). B, GM2-gangliosidosis variant 0 in a 23-week-old fetus; and part of a cortical neuron; early stage of membranous cytoplasmic bodies (two clusters marked by crosses), i.e., lipid storing secondary lysosomes, ϫ36,000 (courtesy W. Schlote, Tu¨bingen). C, The large cells are neurons distended by lysosomal storage in the medulla oblongata of a 1.5-year-old patient with GM1- gangliosidosis. The small dark bodies are nuclei from increased numbers of glial cells (“reactive gliosis”). Cresyl violet stainϫ300 (courtesy of H.U. Benz, Tu¨bingen). D, Cherry red macular spot in an infantilepatientwithGM2-gangliosidosisvariantB(Tay–Sachsdisease).Thedarkredspotinthemiddleofthelightcentralareaissecondarytolipidstorageinneuronalcellsinthisarea;thestoring cells have lost their processes that normally cover the fovea centralis. The fovea normally appears as yellow, but is changed to the red macular spot showing the color of the choroidea behind the retina. E, Lymphocytes with clusters of vacuoles (storage lysosomes) in blood smear of an early infantile patient with GM1-gangliosidosis, indicating the “generalized storage disorder.” May- Gru¨nwald stain; ϫ1500 (courtesy of H.U. Benz, Tu¨bingen).

The prenatal and postnatal laboratory di- Macromolecules and membrane compo- Our view of the topology of lysosomal agnosis of gangliosidoses is usually made nents reach the lysosomal compartment for ganglioside digestion enzyme biochemically (e.g., in blood sam- digestion by autophagy (Florey and Over- For their digestion, gangliosides of cellular ples), and/or molecularly. Many aspects holtzer, 2012), phagocytosis (Delves et al., surfaces reach luminal intra-endolysosomal of genotype–phenotype correlation in 2006; Florey and Overholtzer, 2012), and by vesicles or intra-endosomal membranes these diseases were described in the special endocytotic pathways (Kolter and Sandhoff, (IMs) (Fig. 3; Burkhardt et al., 1997; Mo¨bius literature. Therapeutic strategies such as 2005). Whereas water-soluble lysosomal et al., 1999), which are generated during en- bone marrow transplantation (BMT), hydrolases can attack water-soluble macro- docytosis. Luminal vesicles are formed by use of drugs with substrate reducing, molecules directly, the digestion of ganglio- successive steps of vesicle budding and fis- chaperon-like and anti-inflammatory ef- sides, GSLs, and membranes needs a more sion controlled by the endosomal-sorting fects, and intracerebral or intrathecal en- complex cooperation between soluble hydro- complex proteins (Wollert and Hurley, zyme replacement are under investigation lases, lipid binding and transfer proteins, and 2010). The vesicles are prepared for lyso- (see Therapeutic approaches, below). luminal intra-endolysosomal vesicles (Kolter somal digestion by a lipid-sorting process and Sandhoff, 2005; Kolter and Sandhoff, beginning at the level of endosomes (Kolter Pathway of ganglioside degradation Topology of endocytosis and lysosomal 2010). Due to the lipid phase problem, water- and Sandhoff, 2005; Gallala et al., 2011). digestion of gangliosides soluble glycosidases hardly attack amphiphilic Cholesterol is sorted out by two sterol bind- Lysosomes are intracellular stomachs. They GSLs as components of vesicular membranes, ing proteins, NPC-2 and NPC-1 (Abdul- degrade macromolecules and release their but need the help of membrane perturbing Hammed et al., 2010), and components as nutrients into the cytosol for and lipid binding proteins, the sphingolipid is degraded by acid sphingomyelinase salvage pathways and energy metabolism. activator proteins (SAPs). (ASM; V. Oninla, B. Breiden, K. Sandhoff, 10200 • J. Neurosci., June 19, 2013 • 33(25):10195–10208 Sandhoff and Harzer • Gangliosides and Gangliosidoses

Figure 3. Proposed topology of endocytosis and lysosomal lipid and membrane digestion (Gallala et al., 2011). “A section of the plasma membrane is internalized by way of coated pits or caveolae. These membrane patches include GSLs (red) and receptors such as epidermal growth factor receptor (EGFR; blue). These vesicles fuse with the early endosomes, which mature to late endosomes.Endosomalperimetermembranesforminvaginations,controlledbyESCRTproteins(WollertandHurley,2010),whichbudoff,formingintra-endosomalvesicles.Lipidsortingoccursat thisstage.ThepHofthelumenisϳ5.AtthispH,ASMisactiveanddegradessphingomyelinoftheintra-endosomalvesiclestoceramide,whereastheperimetermembraneisprotectedagainstthe action of ASM by the glycocalyx facing the lumen. This decrease in sphingomyelin, coupled with the increase in ceramide, facilitates the binding of cholesterol to NPC2 and its transport to the perimetermembraneofthelateendosomewhereitistransferredtoNPC1(Kwonetal.,2009).Thisproteinmediatestheexportofcholesterolthroughtheglycocalyx,eventuallyreachingcholesterol bindingproteinsinthecytosol.Ultimately,lateendosomesfusewithlysosomes.TheGSLsareharboredinintralysosomalvesiclesthatfacethelumenofthelysosome,andaredegradedbyhydrolases with the assistance of SAPs. The products of this degradation are exported to the cytosol or loaded on CD1b immunoreceptors and exported to the plasma membrane for antigen presentation. Gradients of pH in the lysosol, and intra-endolysosomal vesicle content of cholesterol (Chol), BMP, sphingomyelin (SM, hypothetical), and ceramide (Cer, hypothetical) are shown” (modified from Kolter and Sandhoff, 2010). unpublished observations). Anionic bis- tralysosomal vesicle and membrane struc- The nonspecific and versatile GM1 de- (monoacylglycero)phosphate (BMP) that tures. These are enriched with anionic BMP, grading ␤-galactosidase occurs as part of a stimulates ganglioside catabolic steps sub- which attracts polycationic proteins, glycosi- lysosomal multi-enzyme complex, contain- stantially is formed from phosphatidyl glyc- dases, and SAPs, to the ganglioside-containing ing sialidase, cathepsin A (the so-called pro- erol in the IMs (Gallala and Sandhoff, 2011; membrane surfaces in the acidic intralyso- tective protein) (d’Azzo et al., 2001), and Gallala et al., 2011). Whereas the lysosomal somal environment. Whereas GM1 hydrolyz- N-acetylaminogalacto-6-sulfate sulfa- perimeter membrane seems to be quite re- ing ␤-galactosidase and GM2 splitting HexA tase (Pshezhetsky and Ashmarina, 2001). sistant against lysosomal degradation, IMs bind to these anionic surfaces, they do not at- Due to changes of the substrate specificity are digested with the help of lipid binding tack the membrane bound ganglioside sub- of variant patient , an inherited proteins (SAPs) and hydrolytic enzymes. ␤ strates in the absence of membrane perturbing defect of GM1- -galactosidase may also Perimeter membranes are protected by lead to a major accumulation of galac- lipid binding and transfer proteins. The first a thick glycocalyx enriched in polylac- tose containing keratan sulfate and oligo- case of GM1 storage was described in 1963 as a tosamine structures, and by a high choles- saccharides, a phenotype called Morquio special form of infantile amaurotic idiocy terol level (Hay, 1989; Appelqvist et al., syndrome, type B (mucopolysaccharido- 2012), the chaperon HSP70 (Kirkegaard et (Jatzkewitz and Sandhoff, 1963). of sis IV B) (Neufeld and Muenzer, 2001; al., 2010), and presumably also by a high vesicle-bound GM1 by lysosomal ganglioside Okumiya et al., 2003). ␤ lateral pressure attenuating the interaction -galactosidase needs the presence of Hydrolysis of GM2 is facilitated at an- with GM2AP, a lipid binding protein essen- GM2AP or saposin B (Sap-B; Wilkening et ionic vesicular surfaces by the cooperation tial for ganglioside catabolism (Giehl et al., al., 2000). Anionic lipids like BMP in the of two proteins, HexA (␣,␤) and the 1999). GM1 carrying vesicles stimulate GM1 hy- membrane perturbing and lipid binding of ganglioside catabolism drolysis efficiently. Inherited defects of the protein GM2AP (Fig. 4)(Kolter and and its diseases GM1 degrading ␤-galactosidase cause Sandhoff, 2006). At acidic pH values, the Ganglioside catabolism proceeds in a stepwise GM1-gangliosidoses (Okada and O’Brien, latter one is a protonated cationic amphi- procedure (Fig. 1) at the surface of luminal in- 1968). philic protein and can bind to the anionic, Sandhoff and Harzer • Gangliosides and Gangliosidoses J. Neurosci., June 19, 2013 • 33(25):10195–10208 • 10201

Figure 4. Mechanism of GM2-AP-Liftase. A, Model for the interaction of GM2-activator protein with luminal lysosomal membranes (IM) in the degradation of ganglioside GM2 (modified from Wendeleretal.,2004a).“GM2-APinteractswiththemembranebydippingthetwoexposedhydrophobicloops,V90–W94andV153–L163,intotheapolarpartofthemembrane.GangliosideGM2 is recognized by specific sites at the rim of the cavity. In the open protein conformation, the large hydrophobic area reaching from the apolar phase of the membrane to the activator’s cavity lowers the energy barrier for lipids leaving the membrane in an upward direction. After the ceramide tail has arrived inside the activator’s cavity, the inward-bending of the hydrophobic loop V153–L163 is favored, and the conformation changes to the closed form (Wright et al., 2003). This folding in of the hydrophobic loop leaves a more polar patch close to the membrane. The activator is then anchored only by the loop V90–W94. It may now rotate slightly upward to expose all polar patches more fully to the solvent, and it may also leave the membrane and interact with the degrading enzyme. The photoaffinity label analogs C(n ϭ 5,1)-TPD-GM2 were photo-incorporated specifically into the region V153–L163” (Sandhoff, 2012). B, BMP and GM2-AP stimulate the hydrolysis of LUV-boundgangliosideGM2.ThedegradationofgangliosideGM2insertedintoLUVsdopedwith0mol%BMP(E),5mol%BMP(Ⅺ),10mol%BMP(·),20mol%BMP(Œ),or25mol%BMP(f) was measured in the presence of increasing concentrations of GM2-AP (0–2.2 ␮M)(Werth et al., 2001).

BMP-rich membrane surfaces, and lift a the additional storage of neutral glycolip- tomatology. At the biochemical level, the ganglioside molecule (Fig. 4). A HexA ids, especially of in the visceral clinical heterogeneity is paralleled by a binding helix of the GM2AP facilitates organs, and of oligosaccharides. The rare variation of the extent and the pattern of formation of the Michaelis–Menten com- defect of GM2AP results in AB-variant of accumulation and by different plex for the degradation of GM2, the Tay– GM2-gangliosidoses (Sandhoff et al., levels of residual catabolic activities de- Sachs ganglioside, and glycolipid GA2. 1989; Sandhoff and Kolter, 2003). Genet- tected in cultures of patients’ fibroblasts Inherited defects of either of the proteins ically manipulated knock-out mice are (Fig. 5). Catabolic activities were assayed result in fatal neurodegenerative diseases available as models for these diseases to in cell cultures using the physiological, ra- (Kolter and Sandhoff, 2006). Tay–Sachs study their course and pathogenesis, and diolabeled lipid substrates instead of the disease is caused by mutations in the to investigate therapeutic approaches. water-soluble synthetic substrates, which HexA gene with loss of HexA (␣,␤) and S serve widely for diagnostic purposes. The (␣,␣) activity. HexA mutations that still Protracted clinical forms of water-soluble substrates may well lead to allow the formation of the heterodimeric gangliosidoses and the threshold theory erroneous data because of the lipid phase HexA (␣,␤) and affect only its activity Complete or nearly complete loss of GM1 problem and the absent control by against anionic substrates result in B1 or GM2 catabolic activities results in fatal GM2AP (and other activators) of the ca- variant of GM2-gangliosidosis (Kytzia et infantile forms of storage diseases. How- tabolism of physiological lipid substrates al., 1983; Kytzia and Sandhoff, 1985). On ever, allelic mutations, which produce such as GM2 ganglioside. According to the other hand, inherited defects in HexB proteins with some residual catabolic ac- Figure 5, GM2 cleavage by HexA was al- gene cause the loss of HexA (␣,␤) and tivities, give rise to protracted clinical most absent in fibroblasts of infantile pa- HexB (␤,␤) activities in Sandhoff disease. forms often described as late infantile, ju- tients, and significantly higher in those of Here, the combined deficiency of both venile, or chronic diseases (Sandhoff et al., juvenile and adult patients (Kytzia et al., major hexosaminidase isoenzymes causes 1989) with a wide range of clinical symp- 1984; Leinekugel et al., 1992). This obser- 10202 • J. Neurosci., June 19, 2013 • 33(25):10195–10208 Sandhoff and Harzer • Gangliosides and Gangliosidoses

Figure 5. Residual catabolic activity correlates with clinical forms of GM2 gangliosidoses (modified from Sandhoff, 2012). A, Steady-state substrate concentration as a function of enzyme concentration and activity (Conzelmann and Sandhoff, 1983). The model underlying this theoretical calculation assumes influx of the substrate into the lysosomal compartment at a constant rate ϭ ………… ϭ ϭ (vi)anditssubsequentutilizationbythecatabolicenzyme.Blueline [S]eq steady-statesubstrateconcentration; theoreticalthresholdofenzymeactivity;——- criticalthreshold value, taking limited solubility of substrate into account; red line ϭ turnover rate of substrate (flux rate). B, To experimentally verify the basic assumptions for this model, studies were performed in cell culture (Leinekugel et al., 1992). The radiolabeled substrate ganglioside GM2 was added to cultures of skin fibroblasts with different activities of ␤-hexosaminidase A and its uptake and turnovermeasured.Thecorrelationbetweenresidualenzymeactivityandtheturnoverrateofthesubstratewasessentiallyaspredicted:degradationrateofgangliosideGM2increasedsteeplywith residualactivity,toreachthecontrollevelataresidualactivityofϳ10–15%ofnormal.Allcellswithanactivityabovethiscriticalthresholdhadanormalturnover.Comparisonoftheresultsofthese feeding studies with the clinical status of the donor of each cell line basically confirmed our notions but also revealed the limitations of the cell culture approach. vation can well be modeled on the basis of the disease will also contribute to the clin- mulation of nondegradable substrates but a greatly simplified kinetic model (Con- ical course and disease pathogenesis, e.g., also affect the transcription rate of many zelmann and Sandhoff, 1983, 1991; misregulation of proteins and intracellu- other . Mutations in the HexB gene Leinekugel et al., 1992) presented in Fig- lar trafficking, formation of lytic and toxic and the subsequent loss of HexA and ure 5. A nice correlation was observed be- compounds (Nilsson et al., 1982; Igisu HexB activity in genetically manipulated tween the predicted and the measured and Suzuki, 1984), depletion of precursor mice trigger a significant downregulation lipid substrate turnover, and also between pools, alterations of composition and of ϳ300 transcripts and a significant up- the decrease of the ganglioside GM2 turn- function of membranes, mislocalized regulation of another ϳ250 transcripts in- over and the clinical course of the disease. storage compounds like GM1, which acti- cluding genes of the immune system Ten to 20% of normal GM2 cleaving ac- vate ER-chaperons and trigger neuronal (Myerowitz et al., 2002). tivity appear already compatible with apoptosis (Tessitore et al., 2004), and al- Genetically manipulated mice are not normal life. Similar observations were tered regulation of endolysosomal pro- always faithful copies of the human dis- reported for metachromatic leukodystro- teins and enzymes by an increased load of ease. In contrast to Tay–Sachs patients, phy (MLD), Gaucher, Sandhoff, and lipids (see below). Also the formation of HexA knock-out mice reach an almost ASM-deficient Niemann–Pick disease meganeurites and an increase in synaptic normal life span and hardly get sick. The (Kudoh et al., 1983; Kytzia et al., 1984; spines may well disturb the connectivity reason seems to be a small difference in Leinekugel et al., 1992; Graber et al., in the nervous system (Purpura and Su- the substrate specificity of murine and hu- 1994). zuki, 1976). man sialidase. Whereas human sialidase is Of course, additional factors induced A genetic and loss of a cata- almost inactive, murine sialidase has a mi- by the mutation and the storage process in bolic activity may not only result in accu- nor but significant cleaving activity on the Sandhoff and Harzer • Gangliosides and Gangliosidoses J. Neurosci., June 19, 2013 • 33(25):10195–10208 • 10203 storage compound ganglioside GM2 and mieux et al., 2006). Therefore, mutations values, enzymes, and SAPs for significant thereby opens a catabolic bypass (Sand- affecting the dimer formation produce inac- degradation rates, but also a sufficient hoff and Jatzkewitz, 1967; Sango et al., tive monomeric subunits causing infantile concentration of anionic lipids such as 1995). Mutant mice develop only a minor diseases. The broad and overlapping speci- BMP in the liposomal membranes. storage process in some neurons, e.g., in ficity of murine hexosaminidases A, B, and S Both, GM2AP and Sap-B, allow that the pyramidal cells of the cortex (Taniike is highlighted by mice lacking both subunits ␤-galactosidase hydrolyzes GM1. There- et al., 1995). Apparently, the catabolic ac- of ␤-hexosaminidase. They store ganglio- fore, neither a defect of GM2AP nor of tivity of the bypass (sequential action of sides and glycosaminoglycans and display Sap-B causes a GM1 accumulation, since murine sialidase to convert GM2 to GA2, phenotypic, pathologic, and biochemical the other one left efficiently facilitates the and hex B to convert GA2 to LacCer) is features of mucopolysaccharidoses, not reaction. However, the reaction is rather sufficient in almost all neurons of the ce- seen in Tay–Sachs and Sandhoff disease slow in neutral liposomes and is stimu- rebral cortex except in the murine pyra- (Sango et al., 1996). lated 2- to 9-fold in the presence of an midal cells and in a few others. These cells The first activator protein described, anionic phospholipid such as BMP, phos- probably have a rather high biosynthetic the activator protein, is now phatidylinositol, dolicholphosphate, and rate for gangliosides, thereby slightly ex- called Sap-B. It was discovered as an es- phosphatidylserine (Wilkening et al., ceeding their maximal catabolic capacity sential for the lysosomal catabo- 2000). Even at acidic, lysosomal pH val- (Vmax of the catabolic system; Fig. 5)in lism of (Mehl and Jatzkewitz, ues, these phospholipids convey negative case of hex A knocked out. Further patho- 1964). Its deficiency causes a late infantile charges to the vesicular surfaces, which at- genetic cascades activated in these dis- form of MLD with a pronounced sulfatide tract and bind the protonated and posi- eases, altered lipid trafficking, altered storage (Mehl and Jatzkewitz, 1964; Ste- tively charged hydrolases and SAPs to autophagy, and induction of inflamma- vens et al., 1981). Sap-B combines with catalyze the hydrolytic reaction. This view tion, have been reviewed (Vitner et al., sulfatides, and the formed stoichiometric also applies to the degradation of lipo- 2010; Xu et al., 2010). soluble complex then forms a Michaelis– somal ganglioside GM2 by HexA in the Menten complex with presence of GM2AP. GM2AP is essential Many proteins involved in ganglioside (Mehl and Jatzkewitz, 1964; Mraz et al., for the enzymatic hydrolysis of ganglio- metabolism are nonspecific 1976). Though this suggested a high lipid side GM2, even in the presence of high The inherited defect of an endolysosomal binding specificity of Sap-B, we now BMP concentrations. Its absence causes protein usually affects only one step in know that the human Sap-B is not specific the fatal neurodegenerative AB variant of ganglioside and sphingolipid catabolism. for sulfatide binding, but rather a promis- GM2-gangliosidosis in vivo. Still, the reac- However, a more careful investigation of cuous lipid binding protein (Ga¨rtner et tion is stimulated by anionic BMP Ͼ100- the proteins revealed that many of them al., 1983; Sun et al., 2008). It even stimu- fold (Werth et al., 2001)(Fig. 4B). Also have rather promiscuous functions and, lates the hydrolysis of bacterial lipids by downstream reactions (Fig. 1) are stimu- therefore, have a broader effect on metab- bacterial hydrolases (Li et al., 1988). Also lated by anionic membrane lipids. At con- olism than originally expected. For glyco- the other saposins have broad functions centrations of anionic phospholipids Ͼ10 sidases this has been known for a long (Doering et al., 1999) and are rather un- mol% in the liposomal membranes, the time, and it was no surprise to find out specific lipid binding proteins, like the cleavage of GlcCer by ␤-glucosidase ␤ that GM1- -galactosidase hydrolyzes GM2AP (Wendeler et al., 2004b). Even the (Wilkening et al., 1998) and that of cer- ␤ many other -galactosides in addition to involved in the biosyn- amide by acid (Linke et al., ganglioside GM1 and GA1 (Kolter and thesis of complex gangliosides at Golgi 2001) proceeds even in the absence of a Sandhoff, 2006). Mutations in the membranes are rather versatile. There are SAP. ␤-galactosidase gene can even change the only one GalNAc-, one Gal-, and two sialyl- enzyme’s substrate specificity and affect transferases available in Golgi membranes the catabolism of GSLs and glycosamino- to convert the precursors LacCer, GM3, Therapeutic approaches glycans differently. GD3, and GT3 into the different complex Currently, there is no treatment available It was also no surprise to realize that gangliosides of the 0-, a-, b-, and c-series for patients with ganglioside storage dis- human ␤-hexosaminidases have broad (Kolter et al., 2002). eases (van Gelder et al., 2012). A main ob- and partially overlapping substrate speci- stacle is the blood–brain barrier, which ficities. Again some mutations in the Membrane lipids are regulators of prevents the passage of therapeutic en- HexA gene may also change substrate ganglioside catabolism zymes and proteins into the brain. For the specificity of the patients’ HexA. Those Promiscuous functions of endolysosomal prevention of substrate storage, substrate that disrupt enzyme activity only against proteins and enzymes make it difficult to influx into the lysosomal compartment anionic substrates (like GM2) and not understand the precise molecular pathol- must not exceed this compartment’s cat- against neutral substrates (like globoside) ogy of the diseases caused by a defect of abolic capacity (see Protracted clinical cause B1 variant of GM2-gangliosidosis any of these proteins. The complexity is forms of gangliosidoses and the threshold (Kytzia et al., 1983), whereas those that further increased by the strong lipid de- theory, above; Conzelmann and Sand- affect HexA activity against all substrates pendency of many endolysosomal pro- hoff, 1983, 1991; Leinekugel et al., 1992). including neutral substrates cause Tay– teins, given that the lipid composition of This means that therapeutic methods Sachs disease (variant B of GM2- many organelle membranes is mostly un- should either increase the catabolic rate gangliosidosis). Active ␤-hexosaminidases known, especially in patients’ tissues. In above the threshold or decrease the influx are the homodimers HexB (␤,␤) and HexS vitro experiments have shown that cata- rate of substrates below the maximum (␣,␣), and the heterodimer HexA (␣,␤). bolic steps at surfaces of substrate carrying catabolic capacity of the affected cells, e.g., Each of them carries two active sites at the liposomes—simulating luminal endoly- by reducing the cellular substrate biosyn- domain interfaces (Maier et al., 2003; Le- sosomal vesicles—not only need low pH thetic rate (Fig. 6). Therapeutic strategies 10204 • J. Neurosci., June 19, 2013 • 33(25):10195–10208 Sandhoff and Harzer • Gangliosides and Gangliosidoses

ϭ Figure 6. Intracellular metabolic flux of sphingolipids. vi the influx rate of the substrate into the lysosome; for further information see the text. Modified from Kolter and Sandhoff, 1999). in use such as enzyme replacement therapy thetic pathway, such as iminosugars of the A combination of different methods for (ERT) for adult Gaucher disease and those nojirimycin type (Butters et al., 2005;Fantur patients with significant residual enzymatic under study in animal models such as BMT et al., 2012; Suzuki et al., 2012) and reduces activity, e.g., substrate reduction together yielded so far no substantial and long- the influx of substrates into the lysosomes with enzyme enhancement therapy, non- lasting improvements in mouse models (Kolter and Sandhoff, 1999). However, steroidal anti-inflammatory drugs, and an- with GM1- and GM2-gangliosidosis. How- treatment of two Tay–Sachs patients with tioxidants is expected to yield better results. ever, intracerebroventricular administra- N-butyldeoxynojirimycin (Miglustat) did tion of a modified recombinant HexB not arrest their clinical deterioration (Bembi Gene therapy (Matsuoka et al., 2011) and a highly phos- et al., 2006), whereas a stable neurological To circumvent the blood–brain barrier, phomannosylated recombinant human picture was reported for two patients with functional copies of the mutant gene were HexA reduced the level of storage com- Sandhoff disease after Miglustat treatment directly inserted into murine brain cells pounds (Tsuji et al., 2011) in the HexA- and (Tallaksen and Berg, 2009; Wortmann et al., with the help of retroviral and lentiviral HexB-deficient mouse brain efficiently. 2009). vectors (Biffi and Naldini, 2005). The Additional approaches using low However, reducing the ganglioside bio- functional enzyme should be stably over- molecular compounds penetrating the synthesis with inhibitors of the ceramide expressed in the transfected mutant cells blood–brain barrier have been evaluated glucosyltransferase, N-butyldeoxynojirimy- and secreted, thereby cross-correcting ad- in murine models of gangliosidoses, but cin or N-butyldeoxygalactonojirimycin, led jacent cells (Arfi et al., 2005; Martino et al., so far without substantial success. They to an increased survival of Sandhoff mice up 2005). include the use of chemical chaperons, to 2 months (Andersson et al., 2004; Jeyaku- Intracerebral injection of an adenovi- substrate analogs, or active site inhibitors mar et al., 2004). ral vector encoding the ␤-subunit of (Matsuda et al., 2003; Tropak et al., 2004) In a clinical Phase I/II trial, treatment of HexA increased HexA activity to nearly to stabilize variant patient hydrolases dur- late onset GM2-gangliosidosis patients with normal levels in HexB-deficient mice ing their biosynthesis and folding, and a presumed chaperon of ␤-hexosamini- (Bourgoin et al., 2003). Also, the intracra- thereby enhance their hydrolytic activity dases, pyrimethamine, for a 4 month period nial inoculation of recombinant adeno- as introduced by Fan et al. (1999). enhanced their leukocyte HexA levels, but associated viral (AAV) vectors encoding Another approach is the substrate reduc- clinical efficacy could not be assessed both, human ␣- and ␤ Hex-subunit genes, tion therapy. It uses inhibitors of the biosyn- (Clarke et al., 2011). resulted in an almost generalized correc- Sandhoff and Harzer • Gangliosides and Gangliosidoses J. Neurosci., June 19, 2013 • 33(25):10195–10208 • 10205 tion of the catabolic lysosomal defect in quantitative understanding of the degra- Appelqvist H, Sandin L, Bjo¨rnstro¨m K, Saftig P, the same mice (Cacho´n-Gonza´lez et al., dative sphingolipid pathways. Garner B, Ollinger K, Kågedal K (2012) Sen- 2006). Animals survived more than a year We also do not know to what extent sitivity to lysosome-dependent cell death is di- rectly regulated by lysosomal cholesterol with delayed onset of the disease, reduced the lipid load of the affected lysosomes content. PLoS One 7:e50262. CrossRef ganglioside storage and inflammation, disturbs their primary functions as cellu- Medline preserved motor function, and rescue of lar stomachs (Tamboli et al., 2011)tode- Arfi A, Bourgoin C, Basso L, Emiliani C, Tancini acute neurodegenerative illness (Sargeant grade macromolecules and release the B, Chigorno V, Li YT, Orlacchio A, Poenaru L, et al., 2012). products into the cytosol as nutrients Sonnino S, Caillaud C (2005) Bicistronic AAV-mediated neonatal gene delivery for biosynthetic pathways and energy me- lentiviral vector corrects beta-hexosaminidase by intracerebroventricular injection seems tabolism. One should also keep in mind deficiency in transduced and cross-corrected human Sandhoff fibroblasts. Neurobiol Dis to be an effective approach to treat GM1- that the cellular uptake of essential factors, 20:583–593. CrossRef Medline gangliosidoses. It resulted in a widespread for example, the uptake of vitamin B12 Assadi M, Baseman S, Janson C, Wang DJ, Bila- appearance of the missing ␤-galactosidase involving several proteins, and that of niuk L, Leone P (2008) Serial 1H-MRS in ϩ activity in the murine brain and a normal- Fe 2 via transferrin and the transferrin GM2 gangliosidoses. Eur J Pediatr 167: ization of all GSL levels (Broekman et al., receptor, may well be impaired by the 347–352. CrossRef Medline 2007; Baek et al., 2010). lipid load causing a traffic jam in the en- Baek RC, Broekman ML, Leroy SG, Tierney LA, Though these results are promising, it dolysosomal system (Jeyakumar et al., Sandberg MA, d’Azzo A, Seyfried TN, Sena- Esteves M (2010) AAV-mediated gene deliv- will take many more investigations and 2009). The storage of GSLs in gangli- ery in adult GM1-gangliosidosis mice corrects progress to develop a real gene therapy for osidoses also triggers side effects such as lysosomal storage in CNS and improves sur- GSL and sphingolipid storage diseases. the accumulation of hydrophobic mate- vival. PLoS One 5:e13468. CrossRef Medline rial including phospholipids and hydro- Bembi B, Marchetti F, Guerci VI, Ciana G, Ad- Conclusions and outlook phobic protein fragments in the storage dobbati R, Grasso D, Barone R, Cariati R, To identify the molecular basis of amau- granules (“membranous cytoplasmic Fernandez-Guillen L, Butters T, Pittis MG rotic idiocy of the lipid accumulating type, bodies”) (Samuels et al., 1963), but also (2006) Substrate reduction therapy in the in- fantile form of Tay-Sachs disease. it was important to isolate the storage transmembrane spans of mitochondrial ␤ 66:278–280. CrossRef Medline compounds and clarify their chemical ATP synthase and amyloid- -peptide Biffi A, Naldini L (2005) Gene therapy of storage structures. This allowed a search for the (Keilani et al., 2012). As in multiple sclero- disorders by retroviral and lentiviral vectors. defective catabolic steps to identify the en- sis, functionally impaired neurons give Hum Gene Ther 16:1133–1142. CrossRef zymes and lipid binding proteins in- rise to an immune response, and are Medline volved, their genomic structures, and the prone to digestion by microglia and in- Bourgoin C, Emiliani C, Kremer EJ, Gelot A, Tan- mutations causing the different gangli- vaded macrophages (Wada et al., 2000; cini B, Gravel RA, Drugan C, Orlacchio A, Poenaru L, Caillaud C (2003) Widespread osidoses. The fact that the main storage Wekerle, 2005). distribution of beta-hexosaminidase activity compounds (gangliosides GM1 and Although we have learned a lot about in the brain of a Sandhoff mouse model after GM2) are preferentially synthesized in the function of gangliosides, their bio- coinjection of adenoviral vector and manni- neuronal cells is crucial for the start of chemistry, enzymology, and cell biology, tol. Gene Ther 10:1841–1849. CrossRef neurodegenerative diseases and compli- we are still far away from a real under- Medline cates all therapeutic approaches. So far, standing of the molecular and cellular Broekman ML, Baek RC, Comer LA, Fernandez JL, efficient ERT is not feasible, since the pathogenesis of ganglioside storage dis- Seyfried TN, Sena-Esteves M (2007) Complete correction of enzymatic deficiency and neuro- blood–brain barrier prevents cerebral up- eases and the development of therapeutic chemistry in the GM1-gangliosidosis mouse take of therapeutic proteins. Also genetic approaches. However, the successful dis- brain by neonatal adeno-associated virus- approaches that appeared to be promising section of the clinical picture of amaurotic mediated gene delivery. Mol Ther 15:30–37. in mouse models are still not reliable idiocy into 5 different ganglioside storage CrossRef Medline (Cacho´n-Gonza´lez et al., 2006). diseases involving 4 different genes, and a Burkhardt JK, Hu¨ttler S, Klein A, Mo¨bius W, There is no simple model to under- group of 10 so far not extensively analyzed Habermann A, Griffiths G, Sandhoff K stand the complex pathogenesis of mono- neuronal ceroidlipofuscinoses, may en- (1997) Accumulation of sphingolipids in SAP-precursor (prosaposin)-deficient fibro- genetic gangliosidoses. The primary lipid courage the search for similar clinical en- blasts occurs as intralysosomal membrane storage affects the endolysosomal system tities of monogenetic diseases but with structures and can be completely reversed by of neurons and obviously does not trigger causative diversity, among the high num- treatment with human SAP-precursor. Eur an effective feedback inhibition of the stor- ber of endemic inherited diseases, the mo- J Cell Biol 73:10–18. Medline age compounds’ biosynthesis. However, the lecular basis of which still has to be Butters TD, Dwek RA, Platt FM (2005) Imino storage may well impair the turnover of elucidated. sugar inhibitors for treating the lysosomal glyco- membranes and macromolecules, affecting . Glycobiology 15:43R–52R. CrossRef Medline neuronal metabolism and function. References Abdul-Hammed M, Breiden B, Adebayo MA, Ba- Cacho´n-Gonza´lez MB, Wang SZ, Lynch A, Degradation of gangliosides requires balola JO, Schwarzmann G, Sandhoff K Ziegler R, Cheng SH, Cox TM (2006) Effec- the interaction of rather unspecific hydro- (2010) Role of endosomal membrane lipids tive gene therapy in an authentic model of lases and lipid binding proteins at surfaces and NPC2 in cholesterol transfer and mem- Tay-Sachs-related diseases. Proc Natl Acad Sci of intraluminal vesicles and membrane brane fusion. J Lipid Res 51:1747–1760. U S A 103:10373–10378. CrossRef Medline structures. These complex interactions, CrossRef Medline Clarke JT, Mahuran DJ, Sathe S, Kolodny EH, including the substantial dependency of Andersson U, Smith D, Jeyakumar M, Butters TD, Rigat BA, Raiman JA, Tropak MB (2011) An open-label Phase I/II clinical trial of pyrim- many catabolic steps on lipid binding co- Borja MC, Dwek RA, Platt FM (2004) Im- proved outcome of N-butyldeoxygalactonojiri- ethamine for the treatment of patients factors (SAPs) and the lipid composition mycin-mediated substrate reduction therapy in affected with chronic GM2 gangliosidosis of the intraluminal, lysosomal mem- a mouse model of Sandhoff disease. Neurobiol (Tay-Sachs or Sandhoff variants). Mol Genet branes, seem to prevent an exact and Dis 16:506–515. CrossRef Medline Metab 102:6–12. CrossRef Medline 10206 • J. Neurosci., June 19, 2013 • 33(25):10195–10208 Sandhoff and Harzer • Gangliosides and Gangliosidoses

Conzelmann E, Sandhoff K (1983) Partial en- membranes and with monolayers at the air- ER, Khanna R, Valenzano KJ, Checler F, Bux- zyme deficiencies: residual activities and the water interface. Eur J Biochem 261:650–658. baum JD, Yanagisawa K, Lockhart DJ, Wustman development of neurological disorders. Dev Medline BA, Gandy S (2012) Lysosomal dysfunction in Neurosci 6:58–71. CrossRef Medline Graber D, Salvayre R, Levade T (1994) Accurate a mouse model of Sandhoff disease leads to ac- Conzelmann E, Sandhoff K (1991) Biochemical differentiation of neuronopathic and non- cumulation of ganglioside-bound amyloid-beta basis of late-onset neurolipidoses. Dev Neuro- neuronopathic forms of Niemann-Pick dis- peptide. J Neurosci 32:5223–5236. CrossRef sci 13:197–204. CrossRef Medline ease by evaluation of the effective residual Medline d’Azzo A, Andria G, Strisciulio P, Galjaard H lysosomal sphingomyelinase activity in intact Kirkegaard T, Roth AG, Petersen NH, Mahalka (2001) Galactosialidosis. In: The metabolic cells. J Neurochem 63:1060–1068. Medline AK, Olsen OD, Moilanen I, Zylicz A, Knudsen and molecular bases of inherited diseases, Ed 8 Gravel R, Kaback MM, Proia RL, Sandhoff K, Su- J, Sandhoff K, Arenz C, Kinnunen PK, Ny- (Scriver CR, Beaudet AL, Sly WS, Valle D, zuki K, Suzuki K (2001) The GM2 gangli- landsted J, Ja¨a¨ttela¨ M (2010) Hsp70 stabi- eds), pp 3811–3826. New York: McGraw-Hill. osidoses. In: The metabolic and molecular lizes lysosomes and reverts Niemann-Pick Delves P, Martin S, Burton D, Roit I (2006) bases of inherited diseases, Ed 8 (Scriver CR, disease-associated lysosomal pathology. Na- Roit’s essential immunology, Ed 11. Malden, Beaudet AL, Sly WS, Valle D, eds), pp 3827– ture 463:549–553. CrossRef Medline MA: Blackwell Publishing. 3876. New York: McGraw-Hill. Klenk E (1937) Die Fettstoffe des Gehirns Doering T, Holleran WM, Potratz A, Vielhaber G, Hay ED (1989) Extracellular matrix, cell skele- bei Amaurotischer Idiotie und Niemann- Elias PM, Suzuki K, Sandhoff K (1999) Sph- tons, and embryonic development. Am J Med Pick’scher Krankheit. Ber Ges Physiol ingolipid activator proteins are required for Genet 34:14–29. CrossRef Medline 96:659–660. epidermal permeability barrier formation. Igisu H, Suzuki K (1984) Progressive accumula- Klenk E (1939) Niemann-Pick’sche Krankheit J Biol Chem 274:11038–11045. CrossRef tion of toxic metabolite in a genetic leukodys- und Amaurotische Idiotie. Hoppe-Seyler’s Z Medline trophy. Science 224:753–755. CrossRef Physiol Chem 262:128–143. CrossRef Eggens I, Fenderson B, Toyokuni T, Dean B, Medline Kolter T, Sandhoff K (1999) Sphingolipids— Stroud M, Hakomori S (1989) Specific in- Jalanko A, Braulke T (2009) Neuronal ceroid li- their metabolic pathways and the pathobio- teraction between Lex and Lex determinants. pofuscinoses. Biochim Biophys Acta 1793: chemistry of neurodegenerative diseases. A possible basis for cell recognition in preim- 697–709. CrossRef Medline Angew Chem Int Ed Engl 38:1532–1568. plantation embryos and in embryonal carci- Jatzkewitz H, Sandhoff K (1963) On a biochem- CrossRef noma cells. J Biol Chem 264:9476–9484. ically special form of infantile amaurotic idi- Kolter T, Sandhoff K (2005) Principles of lyso- Medline ocy. Biochim Biophys Acta 70:354–356. somal membrane digestion: stimulation of Fan JQ, Ishii S, Asano N, Suzuki Y (1999) Ac- CrossRef Medline sphingolipid degradation by sphingolipid ac- celerated transport and maturation of lyso- Jennemann R, Sandhoff R, Wang S, Kiss E, tivator proteins and anionic lysosomal lipids. somal alpha-galactosidase A in Fabry Gretz N, Zuliani C, Martin-Villalba A, Ja¨ger Annu Rev Cell Dev Biol 21:81–103. CrossRef lymphoblasts by an enzyme inhibitor. Nat R, Schorle H, Kenzelmann M, Bonrouhi Medline Med 5:112–115. CrossRef Medline M, Wiegandt H, Gro¨ne HJ (2005) Cell- Kolter T, Sandhoff K (2006) Sphingolipid me- Fantur KM, Wrodnigg TM, Stu¨tz AE, Pabst BM, specific deletion of glucosylceramide syn- tabolism diseases. Biochim Biophys Acta Paschke E (2012) Fluorous iminoalditols act thase in brain leads to severe neural defects 1758:2057–2079. CrossRef Medline as effective pharmacological chaperones after birth. Proc Natl Acad Sci U S A 102: Kolter T, Sandhoff K (2010) Lysosomal degra- against gene products from GLB(1) alleles 12459–12464. CrossRef Medline dation of membrane lipids. FEBS Lett 584: causing GM1-gangliosidosis and Morquio B Jennemann R, Sandhoff R, Langbein L, Kaden S, Ro- 1700–1712. CrossRef Medline disease. J Inherit Metab Dis 35:495–503. thermel U, Gallala H, Sandhoff K, Wiegandt H, Kolter T, Magin TM, Sandhoff K (2000) CrossRef Medline Gro¨ne HJ (2007) Integrity and barrier func- Biomolecule function: no reliable prediction Fishman PH, Max SR, Tallman JF, Brady RO, Ma- tion of the epidermis critically depend on from cell culture. Traffic 1:803–804. CrossRef claren NK, Cornblath M (1975) Deficient glucosylceramide synthesis. J Biol Chem 282: Medline Ganglioside Biosynthesis: a novel human sph- 3083–3094. Medline Kolter T, Proia RL, Sandhoff K (2002) Combi- ingolipidosis. Science 187:68–70. CrossRef Jennemann R, Rothermel U, Wang S, Sandhoff R, natorial ganglioside biosynthesis. J Biol Chem Medline Kaden S, Out R, van Berkel TJ, Aerts JM, 277:25859–25862. CrossRef Medline Florey O, Overholtzer M (2012) Autophagy pro- Ghauharali K, Sticht C, Gro¨ne HJ (2010) Kudoh T, Velkoff MA, Wenger DA (1983) Up- teins in macroendocytic engulfment. Trends Hepatic deficiency and take and metabolism of radioactively labeled Cell Biol 22:374–380. CrossRef Medline liver function in mice. Hepatology 51:1799– sphingomyelin in cultured skin fibroblasts Fragaki K, Ait-El-Mkadem S, Chaussenot A, Gire 1809. CrossRef Medline from controls and patients with Niemann- C, Mengual R, Bonesso L, Be´ne´teau M, Ricci Jeyakumar M, Smith DA, Williams IM, Borja MC, Pick disease and other lysosomal storage dis- JE, Desquiret-Dumas V, Procaccio V, Ro¨tig A, Neville DC, Butters TD, Dwek RA, Platt FM eases. Biochim Biophys Acta 754:82–92. Paquis-Flucklinger V (2013) Refractory epi- (2004) NSAIDs increase survival in the CrossRef Medline lepsy and mitochondrial dysfunction due to Sandhoff disease mouse: synergy with Kuhn R, Wiegandt H (1963) Die Konstitution GM3 synthase deficiency. Eur J Hum Genet N-butyldeoxynojirimycin. Ann Neurol 56: der Ganglio-N-tetraose und des Gangliosids 21:528–534. CrossRef Medline 642–649. CrossRef Medline GI. Chem Ber 96:866–880. CrossRef Gallala HD, Sandhoff K (2011) Biological func- Jeyakumar M, Williams I, Smith D, Cox TM, Platt Kwon HJ, Abi-Mosleh L, Wang ML, Deisenhofer tion of the cellular lipid BMP—BMP as a key FM (2009) Critical role of iron in the patho- J, Goldstein JL, Brown MS, Infante RE (2009) activator for cholesterol sorting and mem- genesis of the murine gangliosidoses. Neuro- Structure of N-terminal domain of NPC1 re- brane digestion. Neurochem Res 36:1594– biol Dis 34:406–416. CrossRef Medline veals distinct subdomains for binding and 1600. CrossRef Medline Kannagi R, Cochran NA, Ishigami F, Hakomori S, transfer of cholesterol. Cell 137:1213–1224. Gallala HD, Breiden B, Sandhoff K (2011) Andrews PW, Knowles BB, Solter D (1983) CrossRef Medline Regulation of the NPC2 protein-mediated Stage-specific embryonic antigens (SSEA-3 Kytzia HJ, Sandhoff K (1985) Evidence for cholesterol trafficking by membrane lipids. and -4) are epitopes of a unique globo-series two different active sites on human beta- J Neurochem 116:702–707. CrossRef ganglioside isolated from human teratocarci- hexosaminidase A. Interaction of GM2 activator Medline noma cells. EMBO J 2:2355–2361. Medline protein with beta-hexosaminidase A. J Biol Ga¨rtner S, Conzelmann E, Sandhoff K (1983) Ac- Kawai H, Allende ML, Wada R, Kono M, Sango K, Chem 260:7568–7572. Medline tivator protein for the degradation of globotri- Deng C, Miyakawa T, Crawley JN, Werth N, Kytzia HJ, Hinrichs U, Maire I, Suzuki K, Sandhoff aosylceramide by human alpha-galactosidase. Bierfreund U, Sandhoff K, Proia RL (2001) K (1983) Variant of GM2-gangliosidosis with J Biol Chem 258:12378–12385. Medline Mice expressing only monosialoganglioside hexosaminidase A having a severely changed Giehl A, Lemm T, Bartelsen O, Sandhoff K, Blume GM3 exhibit lethal audiogenic seizures. J Biol substrate specificity. EMBO J 2:1201–1205. A (1999) Interaction of the GM2-activator Chem 276:6885–6888. CrossRef Medline Medline protein with phospholipid-ganglioside bilayer Keilani S, Lun Y, Stevens AC, Williams HN, Sjoberg Kytzia HJ, Hinrichs U, Sandhoff K (1984) Diag- Sandhoff and Harzer • Gangliosides and Gangliosidoses J. Neurosci., June 19, 2013 • 33(25):10195–10208 • 10207

nosis of infantile and juvenile forms of GM2 microscopy after endocytosis in fibroblasts. Naturwissenschaften 82:403–413. CrossRef gangliosidosis variant 0. Residual activities to- J Histochem Cytochem 47:1005–1014. CrossRef Medline ward natural and different synthetic sub- Medline Sandhoff K, Kolter T (2003) Biosynthesis and strates. Hum Genet 67:414–418. CrossRef Mraz W, Fischer G, Jatzkewitz H (1976) Low degradation of mammalian glycosphingolip- Medline molecular weight proteins in secondary lyso- ids. Philos Trans R Soc Lond B Biol Sci 358: Leinekugel P, Michel S, Conzelmann E, Sandhoff somes as activators of different sphingolipid 847–861. CrossRef Medline K (1992) Quantitative correlation between hydrolases. FEBS Lett 67:104–109. CrossRef Sandhoff K, Jatzkewitz H, Peters G (1969) Die the residual activity of beta-hexosaminidase A Medline infantile amaurotische Idiotie und verwandte and arylsulfatase A and the severity of the re- Myerowitz R, Lawson D, Mizukami H, Mi Y, Tifft Formen als Gangliosid-Speicherkrankheiten. sulting lysosomal storage disease. Hum Genet CJ, Proia RL (2002) Molecular pathophysi- Naturwissenschaften 56:356–362. CrossRef 88:513–523. Medline ology in Tay-Sachs and Sandhoff diseases as Medline Lemieux MJ, Mark BL, Cherney MM, Withers SG, revealed by gene expression profiling. Hum Sandhoff K, Conzelmann E, Neufeld EF, Kaback

Mahuran DJ, James MN (2006) Crystallo- Mol Genet 11:1343–1350. CrossRef Medline MM, Suzuki K (1989) The GM2 gangli- graphic structure of human beta-hexosamini- Neufeld EF, Muenzer J (2001) The Mucopoly- osidosis. In: The metabolic basis of inherited dase A: interpretation of Tay-Sachs mutations saccharidoses. In: The metabolic and molecu- disease. (Scriver CR, Beaudet AL, Sly WS, and loss of GM2 ganglioside hydrolysis. J Mol lar bases of inherited disease, Ed 8 (Scriver CR, Valle D, eds), pp 1807–1839. New York: Biol 359:913–929. CrossRef Medline Beaudet AL, Sly WS, Valle D, eds), pp 3421– McGraw-Hill. Li SC, Sonnino S, Tettamanti G, Li YT (1988) 3452. New York: McGraw-Hill. Sango K, Yamanaka S, Hoffmann A, Okuda Y, Characterization of a nonspecific activator Nilsson O, Månsson JE, Håkansson G, Svenner- Grinberg A, Westphal H, McDonald MP, protein for the enzymatic hydrolysis of glyco- holm L (1982) The occurrence of psychosine Crawley JN, Sandhoff K, Suzuki K, Proia RL lipids. J Biol Chem 263:6588–6591. Medline and other glycolipids in spleen and liver from (1995) Mouse models of Tay-Sachs and Linke T, Wilkening G, Sadeghlar F, Mozcall H, Ber- the three major types of Gaucher’s disease. Sandhoff diseases differ in neurologic pheno- nardo K, Schuchman E, Sandhoff K (2001) In- Biochim Biophys Acta 712:453–463. CrossRef type and ganglioside metabolism. Nat Genet terfacial regulation of acid ceramidase activity. Medline 11:170–176. CrossRef Medline Stimulation of ceramide degradation by lyso- Okada M, Itoh Mi M, Haraguchi M, Okajima T, Sango K, McDonald MP, Crawley JN, Mack somal lipids and sphingolipid activator proteins. Inoue M, Oishi H, Matsuda Y, Iwamoto T, ML, Tifft CJ, Skop E, Starr CM, Hoffmann J Biol Chem 276:5760–5768. CrossRef Medline Kawano T, Fukumoto S, Miyazaki H, Furu- A, Sandhoff K, Suzuki K, Proia RL (1996) Lyon G, Adams RD, Kolodny EH (1996a) Early in- kawa K, Aizawa S, Furukawa K (2002) Mice lacking both subunits of lysosomal fantile progressive metabolic encephalopathies: b-series Ganglioside deficiency exhibits no ␤-hexosaminidase display gangliosidosis clinical problems and diagnostic considerations. definite changes in the neurogenesis and the and mucopolysaccharidosis. Nat Genet 14: In: Neurology of hereditary metabolic diseases sensitivity to Fas-mediated apoptosis but im- 348–352. CrossRef Medline of children, Ed 2 (Lyon G, Adams RD, Kolodny pairs regeneration of the lesioned hypoglossal Sargeant TJ, Drage DJ, Wang S, Apostolakis AA, EH, eds), pp 45–123. New York: McGraw-Hill. nerve. J Biol Chem 277:1633–1636. CrossRef Cox TM, Cacho´n-Gonza´lez MB (2012) Lyon G, Adams RD, Kolodny EH (1996b) Medline Characterization of inducible models of Tay- Childhood and adolescent hereditary meta- Okada S, O’Brien JS (1968) Generalized gangli- Sachs and related disease. PLoS Genet bolic disorders. In: Neurology of hereditary osidosis: beta-galactosidase deficiency. Sci- 8:e1002943. CrossRef Medline metabolic diseases of children Ed 2 (Lyon G, ence 160:1002–1004. CrossRef Medline Schnaar RL (2010) Brain gangliosides in axon- Adams RD, Kolodny EH, eds), pp 177–281. Okumiya T, Sakuraba H, Kase R, Sugiura T myelin stability and axon regeneration. FEBS New York: McGraw-Hill. (2003) Imbalanced substrate specificity of Lett 584:1741–1747. CrossRef Medline Maier T, Strater N, Schuette CG, Klingenstein R, mutant beta-galactosidase in patients with Simpson MA, Cross H, Proukakis C, Priestman DA, Sandhoff K, Saenger W (2003) The X-ray crys- Morquio B disease. Mol Genet Metab 78:51– Neville DC, Reinkensmeier G, Wang H, Wiz- tal structure of human beta-hexosaminidase B 58. CrossRef Medline nitzer M, Gurtz K, Verganelaki A, Pryde A, Pat- provides new insights into Sandhoff disease. Proia RL (2003) Glycosphingolipid functions: ton MA, Dwek RA, Butters TD, Platt FM, J Mol Biol 328:669–681. CrossRef Medline insights from engineered mouse models. Phi- Crosby AH (2004) Infantile-onset symptom- Martino S, Marconi P, Tancini B, Dolcetta D, De los Trans R Soc Lond B Biol Sci 358:879–883. atic epilepsy syndrome caused by a homozygous Angelis MG, Montanucci P, Bregola G, Sand- CrossRef Medline loss-of-function mutation of GM3 synthase. Nat hoff K, Bordignon C, Emiliani C, Manservigi Pshezhetsky AV, Ashmarina M (2001) Lyso- Genet 36:1225–1229. CrossRef Medline R, Orlacchio A (2005) A direct gene transfer somal multienzyme complex: biochemistry, Stevens RL, Fluharty AL, Kihara H, Kaback MM, strategy via brain internal capsule reverses the genetics, and molecular pathophysiology. Shapiro LJ, Marsh B, Sandhoff K, Fischer G biochemical defect in Tay-Sachs disease. Hum Prog Nucleic Acid Res Mol Biol 69:81–114. (1981) Cerebroside sulfatase activator defi- Mol Genet 14:2113–2123. CrossRef Medline CrossRef Medline ciency induced metachromatic leukodystro- Matsuda J, Suzuki O, Oshima A, Yamamoto Y, Purpura DP, Suzuki K (1976) Distortion of neu- phy. Am J Hum Genet 33:900–906. Medline Noguchi A, Takimoto K, Itoh M, Matsuzaki Y, ronal geometry and formation of aberrant Sun Y, Witte DP, Ran H, Zamzow M, Barnes S, Yasuda Y, Ogawa S, Sakata Y, Nanba E, Higaki synapses in neuronal storage disease. Brain Cheng H, Han X, Williams MT, Skelton MR, K, Ogawa Y, Tominaga L, Ohno K, Iwasaki H, Res 116:1–21. CrossRef Medline Vorhees CV, Grabowski GA (2008) Neuro- Watanabe H, Brady RO, Suzuki Y (2003) Samuels S, Korey SR, Gonatas J, Terry RD, Weiss logical deficits and glycosphingolipid accu- Chemical chaperone therapy for brain pathol- M (1963) Studies in Tay-Sachs disease. IV. mulation in saposin B deficient mice. Hum ogy in G(M1)-gangliosidosis. Proc Natl Acad Membranous cytoplasmic bodies. J Neuro- Mol Genet 17:2345–2356. CrossRef Medline Sci U S A 100:15912–15917. CrossRef path Exp Neurol 22:81–97. CrossRef Medline Suzuki Y, Oshima A, Nanba E (2001) Medline Sandhoff K (1977) The biochemistry of sphin- ␤-Galactosidase deficiency (␤-galactosidosis): Matsuoka K, Tamura T, Tsuji D, Dohzono Y, Kita- golipid storage diseases. Angew Chem Int Ed GM1 gangliosidosis and Morquio B disease. In: kaze K, Ohno K, Saito S, Sakuraba H, Itoh K Engl 16:273–285. CrossRef Medline The metabolic and molecular bases of inherited (2011) Therapeutic potential of intracerebro- Sandhoff K (2012) My journey into the world of diseases, Ed 8 (Scriver CR, Beaudet AL, Sly WS, ventricular replacement of modified human sphingolipids and sphingolipidoses. Proc Jpn Valle D, eds), pp 3755–3809. New York: beta-hexosaminidase B for GM2 gangliosidosis. Acad Ser B Phys Biol Sci 88:554–582. McGraw-Hill. Mol Ther 19:1017–1024. CrossRef Medline CrossRef Medline Suzuki Y, Ichinomiya S, Kurosawa M, Matsuda J, Mehl E, Jatzkewitz H (1964) Eine Cerebrosidsulfa- Sandhoff K, Jatkewitz H (1967) A particle- Ogawa S, Iida M, Kubo T, Tabe M, Itoh M, Hi- tase aus Schweineniere. Hoppe-Seylers Z bound sialyl lactosidoceramide splitting gaki K, Nanba E, Ohno K (2012) Therapeutic Physiol Chem 339:260–276. CrossRef Medline mammalian sialidase. Biochim Biophys Acta chaperone effect of N-octyl 4-epi-beta- Mo¨bius W, Herzog V, Sandhoff K, Schwarzmann G 141:442–444. CrossRef Medline valienamine on murine G(M1)-gangliosidosis. (1999) Intracellular distribution of a biotin- Sandhoff K, Kolter T (1995) Glykolipide der Mol Genet Metab 106:92–98. CrossRef Medline labeled ganglioside, GM1, by immunoelectron Zelloberfla¨che-Biochemie ihres Abbaus. Tallaksen CM, Berg JE (2009) Miglustat therapy 10208 • J. Neurosci., June 19, 2013 • 33(25):10195–10208 Sandhoff and Harzer • Gangliosides and Gangliosidoses

in juvenile Sandhoff disease. J Inherit Metab Venugopalan P, Joshi SN (2002) Cardiac in- somal degradation on vesicular membrane Dis 32 [Suppl 1]:S289–S293. CrossRef volvement in infantile Sandhoff disease. J Pae- surfaces. Enhanced glucosylceramide degra- Medline diatr Child Health 38:98–100. CrossRef dation by lysosomal anionic lipids and activa- Tamboli IY, Hampel H, Tien NT, Tolksdorf K, Medline tors. J Biol Chem 273:30271–30278. CrossRef Breiden B, Mathews PM, Saftig P, Sandhoff K, Vitner EB, Platt FM, Futerman AH (2010) Com- Medline Walter J (2011) Sphingolipid storage affects mon and uncommon pathogenic cascades in Wilkening G, Linke T, Uhlhorn-Dierks G, Sand- autophagic metabolism of the amyloid pre- lysosomal storage diseases. J Biol Chem 285: hoff K (2000) Degradation of membrane- cursor protein and promotes A␤ generation. 20423–20427. CrossRef Medline bound ganglioside GM1. Stimulation by J Neurosci 31:1837–1849. CrossRef Medline Wada R, Tifft CJ, Proia RL (2000) Microglial ac- bis(monoacylglycero)phosphate and the acti- Taniike M, Yamanaka S, Proia RL, Langaman C, tivation precedes acute in vator proteins SAP-B and GM2-AP. J Biol Bone-Turrentine T, Suzuki K (1995) Neuro- Sandhoff disease and is suppressed by bone Chem 275:35814–35819. CrossRef Medline pathology of mice with targeted disruption of marrow transplantation. Proc Natl Acad Sci Wollert T, Hurley JH (2010) Molecular mecha- Hexa gene, a model of Tay-Sachs disease. Acta U S A 97:10954–10959. CrossRef Medline nism of multivesicular body biogenesis by Neuropathol 89:296–304. CrossRef Medline Wekerle H (2005) Immune pathogenesis of ESCRT complexes. Nature 464:864–869. Tessitore A, del P Martin M, Sano R, Ma Y, Mann L, . Neurol Sci 26 [Suppl 1]: CrossRef Medline Ingrassia A, Laywell ED, Steindler DA, Hender- S1–S2. Medline Wortmann SB, Lefeber DJ, Dekomien G, Wil- shot LM, d’Azzo A (2004) GM1-ganglioside- Wendeler M, Hoernschemeyer J, Hoffmann D, lemsen MA, Wevers RA, Morava E (2009) mediated activation of the unfolded protein Kolter T, Schwarzmann G, Sandhoff K Substrate deprivation therapy in juvenile response causes neuronal death in a neurode- (2004a) Photoaffinity labelling of the human Sandhoff disease. J Inherit Metab Dis 32 generative gangliosidosis. Mol Cell 15:753–766. GM2-activator protein. Eur J Biochem 271: [Suppl 1]:S307–S311. CrossRef Medline CrossRef Medline 614–627. CrossRef Medline Wright CS, Zhao Q, Rastinejad F (2003) Struc- Tropak MB, Reid SP, Guiral M, Withers SG, Wendeler M, Hoernschemeyer J, John M, Werth tural Analysis of Lipid Complexes of GM2- Mahuran D (2004) Pharmacological en- N, Schoeniger M, Lemm T, Hartmann R, Kes- hancement of beta-hexosaminidase activity sler H, Sandhoff K (2004b) Expression of the Activator Protein. J Mol Biol 331:951–964. in fibroblasts from adult Tay-Sachs and GM2-activator protein in the methylotrophic CrossRef Medline Sandhoff Patients. J Biol Chem 279:13478– yeast Pichia pastoris, purification, isotopic la- Xu YH, Barnes S, Sun Y, Grabowski GA (2010) 13487. CrossRef Medline beling, and biophysical characterization. Pro- Multi-system disorders of glycosphingolipid Tsuji D, Akeboshi H, Matsuoka K, Yasuoka H, tein Expr Purif 34:147–157. CrossRef Medline and ganglioside metabolism. J Lipid Res 51: Miyasaki E, Kasahara Y, Kawashima I, Chiba Werth N, Schuette CG, Wilkening G, Lemm T, 1643–1675. CrossRef Medline Y, Jigami Y, Taki T, Sakuraba H, Itoh K Sandhoff K (2001) Degradation of membrane- Yamashita T, Wada R, Sasaki T, Deng C, Bier- (2011) Highly phosphomannosylated en- bound ganglioside GM2 by beta -hexosamini- freund U, Sandhoff K, Proia RL (1999) A vi- zyme replacement therapy for GM2 gangli- dase A. Stimulation by GM2 activator protein tal role for glycosphingolipid synthesis during osidosis. Ann Neurol 69:691–701. CrossRef and lysosomal lipids. J Biol Chem 276:12685– development and differentiation. Proc Natl Medline 12690. CrossRef Medline Acad Sci U S A 96:9142–9147. CrossRef van Gelder CM, Vollebregt AA, Plug I, van der Wilken B, Dechent P, Hanefeld F, Frahm J (2008) Medline Ploeg AT, Reuser AJ (2012) Treatment op- Proton MRS of a child with Sandhoff disease Yu RK, Tsai YT, Ariga T, Yanagisawa M (2011) tions for lysosomal storage disorders: devel- reveals elevated brain hexosamine. Eur J Pae- Structures, biosynthesis, and functions of oping insights. Expert Opin Pharmacother 13: diatr Neurol 12:56–60. CrossRef Medline gangliosides–an overview. J Oleo Sci 60: 2281–2299. CrossRef Medline Wilkening G, Linke T, Sandhoff K (1998) Lyso- 537–544. CrossRef Medline