Cover Page
The handle http://hdl.handle.net/1887/49552 holds various files of this Leiden University dissertation
Author: Mirzaian, Mina
Title: Analytical chemistry and biochemistry of glycosphingolipids : new developments and insights
Issue Date: 2017-06-14
Chapter 10
Adapting to a deficient
glycosphingolipid-degrading lysosomal glycosidase -
hypothesis review
Manuscript pending submission
Chapter 10
Adapting to a deficient glycosphingolipid-degrading lysosomal glycosidase – hypothesis review
Johannes M. Aerts1,2,*, André R. A. Marques2, Maria J. Ferraz1,2, Mina Mirzaian1, Paulo Gaspar2, Daniela Herrera Moro2, Saskia V. Oussoren1, Tanit L. Gabriel2, Kassiani Kytidou1, Patrick Wisse3, Herman S.
Overkleeft3, Marco van Eijk1, Rolf G. Boot1
Manuscript pending submission
1 Department of Medical Biochemistry, Leiden Institute of Chemistry, Leiden University, 2333 CC Leiden, The Netherlands
2 Department of Medical Biochemistry, Academic Medical Center, University of Amsterdam, 1105 AZ Amsterdam, The Netherlands
3 Department of Bio-organic Synthesis, Leiden Institute of Chemistry, Leiden University, 2333 CC Leiden, The Netherlands
* Corresponding author
Johannes M. Aerts
Department of Medical Biochemistry, room DE.1.19 Leiden Institute of Chemistry, Leiden University Einsteinweg 55, 2333 CC Leiden, The Netherlands.
Mail: j.m.f.g.aerts@lic.leidenuniv.nl
Abstract
Inherited deficiencies in glycosidases degrading glycosphingolipids (GSLs) in lysosomes are the cause of diseases named glycosphingolipidoses. These lysosomal storage disorders are relatively common with an estimated combined prevalence of at least 1 in 20 000 live births. Because lysosomal turnover of endogenous and endocytosed exogenous GSLs is considerable in most cell types, it may be expected that accumulation of GSLs is prominent during pronounced deficiency of metabolizing glycosidases.
Moreover, GSL accumulation should theoretically be progressive since there is no evidence for prominent compensatory down-regulation of their synthesis during failed lysosomal degradation.
However, cellular and tissue GSL accumulation seems to level in glycosphingolipidoses. An explanation for this biochemical paradox could be the existence of yet unknown or poorly appreciated biochemical adaptations to the primary defect in lysosomes. These would clarify the noted leveling of GSL accumulation in time.
Here we hypothesize on, and provide evidence for, such adaptive metabolic pathways in glycosphingolipidoses, inside and beyond lysosomes. We describe how the deficiency of lysosomal α-galactosidase A in Fabry disease (FD) is compensated by ongoing intralysosomal de-acylation of accumulating globotriaosylceramide substrate. Likewise, deficiency of glucocerebrosidase (GBA) in Gaucher disease (GD) patients is compensated by de-acylation of accumulating glucosylceramide.
The lysosomal acid ceramidase catalyzes these de-acylations. An additional, fundamentally distinct, adaptive metabolism takes place in the cytosol during GBA deficiency. The enzyme GBA2 mediates increased degradation of glucosylceramide in the cytosolic leaflet of membranes and generates glucosylated cholesterol in the process. The possible beneficial as well as harmful consequences of these adaptive metabolic pathways for patients with the glycosphingolipidoses GD and FD are discussed.
Chapter 10 Glycosphingolipid’s complex life cycle through various subcellular compartments
Glycosphingolipids (GSLs) are membrane constituents that are particularly abundant in the outer leaflet of the plasma membrane bilayer. They are composed of ceramide (Cer), N-acylated sphingosine, with attached to the C1-hydroxyl a sugar (glucose or galactose) that may be extended by addition of further monosaccharides (1, 2). By van der Waals interactions with cholesterol molecules, GSLs give rise to transient semi-ordered domains in membranes. In these so-called ‘lipid rafts’ specific proteins preferentially reside and related signaling events take place (3–5). The cellular life cycle of GSLs may involve several chemical modifications en route through various subcellular compartments (Figure 1). GSLs are continuously synthesized by cells, starting with the generation of a 3-ketosphinganine structure through condensation of serine and fatty acyl-CoA catalyzed the enzyme serine palmitoyltransferase (SPT) at the endoplasmic reticulum (ER). The predominance of 18 carbon sphingoid bases in mammalian sphingolipids stems from the preference of mammalian SPT for palmitoyl-CoA (6). Next, sphinganine is formed by a specific reductase and subsequent N-acylation by ceramide synthases (CERS) produces dihydroceramides that are rapidly converted to ceramides by dihydroceramide desaturase (DES) at the ER (6, 7). Subsequently, newly formed Cer may acquire inside the ER a galactose residue to yield galactosylceramide (GalCer) or be alternatively transported by the protein CERT to the cytosolic leaflet of membranes of the cis-Golgi apparatus (8). There it can be glucosylated to glucosylceramide (GlcCer) by the enzyme glucosylceramide synthase (GCS) (9). Newly formed GlcCer is in part translocated to the luminal leaflet of the Golgi membrane via an unknown mechanism. Inside the Golgi apparatus, GlcCer can be extended stepwise with further sugars through the sequential action of glycosyltransferases, yielding complex GSLs such as gangliosides and globosides (2, 10). Sulfation of specific GSLs may also take place, catalyzed by sulfotransferases (6). The impressive structural heterogeneity of GSLs, and its generation by biosynthetic enzymatic reactions, has been extensively reviewed (1, 6, 11). After completion of their biosynthesis in the Golgi apparatus, GSLs reach the outer leaflet of the plasma membrane. From there they may leave cells by transfer to nascent HDL particles, however most GSLs molecules stay in the plasma membrane and are ultimately internalized via endocytosis, ending up in multi-vesicular bodies inside late endosomes that are ultimately degraded within lysosomes.
Figure 1 | GSL life cycle in the cell. GSLs (black font) are synthesized in the ER (red) and Golgi (yellow) compartments starting with the condensation of serine and palmitoyl-coA building blocks. Catabolism of GSLs occurs in lysosomes (grey) through the sequential action of various hydrolases (green font)
Chapter 10 The intralysosomal degradation of complex GSLs is a sequential process in which terminal
sugar moieties, as a reverse of the biosynthetic pathway, are removed by glycosidases in a stepwise manner, often assisted by specific accessory proteins (GM2 activator protein and saposins A-D) (12).
During the lysosomal degradation ceramide is formed either from GalCer by galactocerebrosidase (GALC) or from GlcCer by glucocerebrosidase (GBA). Acid ceramidase (AC) finally cleaves Cer to fatty acid and sphingosine ((2S,3R,4E)-2-aminooctadec-4-ene-1,3-diol) inside lysosomes.
Sphingosine is subsequently exported to be immediately re-used by CERS enzymes in the salvage pathway to generate new Cer (13). Alternatively, sphingosine can be modified to sphingosine-1- phosphate (S1P) via sphingosine kinases (SK1 and SK2), thereafter S1P lysase (SPL) degrades it to phosphatidylethanolamine and 2-trans-hexadecenal (14, 15).
Defective lysosomal degradation of GSLs as cause for disease
The physiological relevance of intralysosomal GSL turnover is illustrated by a number of inherited glycosphingolipidoses in man, each caused by a specific deficiency in lysosomal GSL degradation.
For every step in lysosomal degradation of more common glycosphingolipids, except for that of lactosylceramide, a corresponding inherited disease in man is now known (for reviews see references (16–19)). Examples of these glycosphingolipidoses are GM1 gangliosidosis, GM2 gangliosidosis, Fabry disease (FD, globotriaosylceramidosis), Gaucher disease (GD, glucosylceramidosis) and Krabbe disease (galactosylceramidosis) being due to genetic deficiencies in β-hexosaminidase, β-galactosidase, α-galactosidase A, glucocerebrosidase (acid β-glucosidase) and galactocerebrosidase, respectively (Table 1). The successful therapeutic intervention for type 1 Gaucher disease (GD1) by means of supplementing patients with lacking enzyme through chronic intravenous infusions (enzyme replacement therapy, ERT) has boosted similar therapy approaches for other glycosphingolipidoses.
This development has worldwide stimulated screening for individuals with abnormalities in lysosomal glycosidases in targeted risk groups as well as among newborns (20, 21). The estimated prevalence of the combined glycosphingolipidoses has long been thought to be around 1 in 20 000. Increasingly late onset and atypical variants of glycosphingolipidoses are recognized, particularly for the X-linked FD, raising dramatically the total prevalence (20).
Table 1 | The paradox of inherited (glyco)sphingolipidoses.
Disease Gene Protein Main storage material
GM1 gangliosidosis GLB1 Acid β-galactosidase GM1 ganglioside GM2 gangliosidosis (Tay-Sachs) HEXA β-hexosaminidase α subunit GM2 ganglioside GM2 gangliosidosis (Sandhoff) HEXB β-hexosaminidase β subunit GM2 ganglioside
Fabry GLA α-galactosidase A Globotriaosylceramide (Gb3)
Gaucher GBA Glucocerebrosidase Glucosylceramide (GlcCer)
Metachromatic leukodystrophy ARSA Arylsulfatase A Sulfatide
Krabbe GALC Galactosylceramidase Galactosylceramide (GalCer)
Niemann-Pick type A and B SMPD1 Acid sphingomyelinase Sphingomyelin
Niemann-Pick type C NPC1/NPC2 NPC1/NPC2 Cholesterol, GSLs & sphingomyelin
Farber ASAH1 Acid ceramidase Ceramide
In most cell types de novo synthesis of GSLs is impressive since their turnover is fast with estimated half-lives in the order of hours (6, 11). Furthermore, cells endocytose significant amounts of GSL- rich lipoproteins. A (near) complete block in lysosomal degradation of GSLs is therefore predicted to result soon in an overt, and ongoing, accumulation of these lipids in cells. Lysosomal GSL storage in glycosidase-deficient cells indeed generally develops quickly, being observed already in utero with some glycosidase knockout mouse models (18). However, GSL accumulation in glycosphingolipidoses patients and mouse models in some cases follows not the prediction: after a fast initial accumulation, subsequent storage of GSLs increases often only marginal. This conflicts with the “critical threshold”
hypothesis, proposed by Conzelmann and Sandhoff, predicting a linear increase in storage accumulation (22). Equally puzzling, complete absence of a GSL-degrading glycosidase may lead in both man and mice to overt disease relatively late. A striking example for this is presented by the complete deficiency of GBA activity in man and mouse. The substrate for GBA, GlcCer, is formed during lysosomal degradation of all gangliosides, globosides and lactosylceramide and consequently a major degradative flux has to be catalyzed by GBA. Already in utero there is considerable lysosomal catabolism of GlcCer, nevertheless mice and humans lacking active GBA can develop more or less normally as fetus (23, 24). Only at birth a fatal impairment of skin barrier function becomes apparent.
The skin defect is not thought to be due to lysosomal lipid storage, but rather to a crucial role for GBA in establishing an appropriate ratio in ceramide and GlcCer molecules in extruded lamellar bodies forming the lipid barrier in the stratum corneum (25). Although upon autopsy marked GlcCer deposition is detected in several organs of the so-called ‘collodion baby’ variant of GD, organ function and development do not seem overtly affected, at least until birth (26, 27). Another example in the same line is presented by males suffering from classic FD by lack of any residual α-galactosidase A (GLA) protein. GLA is a 49 kDa lysosomal enzyme that catalyzes the degradation of the globoside Gb3, a relatively abundant GSL in for example blood cells, endothelial cells, cardiomyocytes, fibroblasts and podocytes (24). Despite usually complete deficiency of GLA, classic FD males express the first overt symptoms in skin, nociceptive neurons and eye only at juvenile age (28). Gb3 lipid deposits, referred to as zebra-bodies, are already observed with electron microscopy in fetal endothelial cells, but major tissues involved in the pathology of FD such as heart, kidney and brain only manifest symptoms in adult life (28). Studies with male FD mice lacking GLA have revealed more clearly that accumulation of Gb3 in tissues already starts to level after the first months of life (29, 30). Similar leveling of storage lipid deposition with age has been noted for LIMP2-deficient mice (Gaspar et al. to be published, chapter 13). LIMP2 is a lysosomal membrane protein encoded by the SCARB2 gene that mediates
Chapter 10 the transport of newly synthesized GBA to lysosomes (31). LIMP2 binds via its coiled coil domain in
the endoplasmic reticulum fully folded GBA and moves it in complex to late endosomes/lysosomes where the low local pH causes dissociation (31, 32). Most cell types and tissues of LIMP2-deficient mice show extremely low levels of active GBA and associated with this very marginal increase in GlcCer. Again, with increasing age GlcCer storage in tissues of LIMP2-deficient mice does not become more prominent (Gaspar et al. to be published, chapter 13). These findings for GLA- and LIMP2- deficient mice nicely illustrate the non-linear increase in GSL storage with age. This finding cannot be explained by a simple feedback regulation since there are no reports on any prominent reduction of GSL biosynthesis in response to an intralysosomal defect in their degradation. This is in sharp contrast to the subtle regulation of cellular cholesterol levels by sensing of sterol concentrations in membranes followed by transcriptional and post-translational regulation of biosynthetic and other modifying enzymes (33). During the last decade Ballabio and co-workers elucidated how lysosomal dysfunction is sensed and responded to. Following lysosomal stress by indigestible macromolecules, the transcription factor TFEB translocates to the nucleus where it increases transcription of genes encoding protein constituents of lysosomes (34). Subsequent increased de novo synthesis of entire lysosomes may contribute to slowing down of lysosomal lipid storage buildup, but the beneficial effect of this up-regulation is short term and intrinsically limited by the total cellular volume that can be occupied by lysosomes.
Glycosphingolipidoses thus still present the paradox of an initial phase of rapid accumulation of GSL storage followed by a far less progressive phase. An appealing explanation for this riddle could be the induction of cellular adaptations in metabolism of GSLs following their storage in lysosomes (Figure 2). Adaptive responses would fit with the noted leveling of GSL storage with age. Such postulated biochemical adaptations are at present either yet unknown or poorly appreciated pathways.
Figure 2 | The paradox of glycosphingolipidoses. Theoretical argument for existence of adaptive metabolism for GSL clearance.
Intralysosomal conversion of glycosphingolipids to glycosphingoid bases as adaptive rescue.
In plasma of patients suffering from glycosphingolipidoses, glycosphingoid bases corresponding with the accumulating GSLs in cells and tissues are markedly increased (Figure 3A). Examples are the 300-fold increase in glucosylsphingosine (GlcSph) in plasma of GD patients and the more modest elevation of GlcSph noted in plasma of patients suffering from Action Myoclonus Renal failure Syndrome (AMRF) caused by mutations in the SCARB2 gene encoding LIMP2 (35–37). Likewise, in plasma of classic FD patients the glycosphingoid base globotriaosylsphingosine (lysoGb3) is 200-fold increased above normal (38). Elevated levels of galactosylsphingosine have earlier been reported for
Krabbe disease patients developing GalCer storage as the result of galactocerebrosidase deficiency (39, 40). These findings with patients are recapitulated by the investigation of glycosphingoid bases and corresponding glycosphingolipids in mouse models with deficiencies in glucocerebrosidase, α-galactosidase A and galactocerebrosidase. In plasma and tissues of these animal models storage of the primary GSL substrate of the deficient glycosidase is accompanied by marked increases in corresponding glycosphingoid bases (41). The subsequent question to be addressed is “how do these excessive amounts of glycosphingoid bases arise?”. A first clue for this was provided by the early work of Yamaguchi et al. demonstrating that pharmacological inhibition of GBA in fibroblasts led to formation of GlcSph, but not in cells deficient in the lysosomal enzyme AC (42). We recently studied in more detail the possible role of lysosomal AC in formation of glycosphingoid bases during lysosomal glycosidase deficiency (43). Formation of GlcSph by cells, following inhibition of lysosomal GBA, was found to be abolished by genetic loss of AC as well as its pharmacological inhibition. Incubation of cells with
13C5-isotope encoded GlcCer with the label in the lipid’s sphingosine moiety resulted in prominent formation of 13C5-isotope encoded GlcSph only when GBA in cells was inhibited beforehand. A similar finding was made for classic FD fibroblasts. A specific inhibitor of AC could also abolish the production lysoGb3 by these cells (43). Thus, there is experimental proof for a direct role of AC in conversion of accumulating GlcCer and Gb3 in lysosomes to corresponding glycosphingoid bases GlcSph and lysoGb3. The amphiphilic nature of these glycosphingoid bases allows them to egress from lysosomes and even from cells as we experimentally observed. GlcSph and lysoGb3 are water soluble compounds, in sharp contrast to their corresponding GSLs. They are not associated with lipoproteins in plasma like GlcCer and Gb3. In urine samples, they are not enriched in the proximal tube cell sediment, but recovered in the cell-free supernatant. Plasma and urine concentrations of the glycosphingoid bases correlate with each other in FD as well as GD patients. Interestingly, in Gaucher disease and Fabry disease patients most of the urinary GlcSph and lysoGb3 molecules differ in their C18-sphingoid base structure from the major species in plasma, being more hydroxylated and possibly methylated (Figure 3B) (36, 44). These modifications appear to be largely introduced in the bases themselves in the kidney since they do not occur to the same extent in urinary GlcCer and Gb3. Likely, local monooxygenase activity of CYPs, the superfamily of detoxifying enzymes containing a heme cofactor, is contributing to the noted modifications. Another structural heterogeneity observed in plasma and urinary GSLs and bases alike is the presence of an additional double bond in the sphingosine moiety. These 4E,14Z- dienes usually constitute around 10% of the total GlcCer, GlcSph, Gb3 and lysoGb3.
Chapter 10 Figure 3 | AC-mediated deacylation of GlcCer and Gb3. A) Deacylation by AC of GlcCer and Gb3 to
GlcSph and lysoGb3, respectively. B) Isoforms of lysoGb3 found in the urine of 2 classical Fabry patients.
Glycosphingoid bases, a blessing or a curse?
Conversion of accumulating GSLs during deficiency of GBA or GLA into corresponding gly- cosphingoid bases provides, at least in part, an explanation for the non-linear increase of stored GSL in time (Figure 2). Intralysosomal de-acylation by AC offers the organism a new pathway to eliminate indigestible GSL macromolecules as water-soluble glycosphingoid bases from cells and even from the body by excretion via bile and urine (Figure 3A). This solution mimics the body’s handling of excessive non-digestible cholesterol, being secreted as such in bile or as water-soluble metabolite like bile acid.
LIMP2-deficient mice illustrate nicely how de-acylation of non-digestible GlcCer offers most cells and tissues the possibility to prevent overt lysosomal storage. This is exemplified by the finding that liver of LIMP2-deficient mice with near complete absence of GBA shows with microscopic analysis almost no storage deposits and with biochemical analysis a very modest increase in GlcCer in combination with a 6.7 times increase in GlcSph as well as 4.5-fold elevated biliary GlcSph (Gaspar et al. unpublished, chapter 13). Likely, the ability to avoid accumulation of lipid-laden lysosomes in hepatocytes contribu- tes to the noticed normal liver function of LIMP2-deficient mice.
The choice to de-acylate non-digestible GSL in glycosphingolipidoses by AC activity has obviously its prize: it exposes the body to abnormally high concentrations of glycosphingoid bases (Figure 4). This may not be without longer term consequences since it has been claimed that gly- cosphingoid bases are biologically active and may exert negative effects (for detailed reviews on this topic see references (24, 45–48)). Briefly, the production of galactosylsphingosine in brain of Krabbe disease patients is thought to contribute to the devastating neuropathology (48). Glucosylsphingosine, abnormally produced in GD, and to lesser extent AMRF patients, is proposed to exert toxic effects at high concentrations. For instance, GlcSph has been proposed to promote lysis of red blood cells, impair cell fission during cytokinesis, damage specific neurons, hamper growth, impair bone forma- tion by osteoblasts, and promote chronic inflammation via activation of phospholipase A2 (45, 46).
It is appealing to speculate that GlcSph may contribute to signs and symptoms in GD patients such as occurrence of hemolysis, multinucleated macrophages, neuropathology, growth retardation, bone deterioration and chronic low grade inflammation. In the same line, gammopathies and lymphoma occur with increased incidence in GD patients (49). Investigations by Cox and co-workers firstly de- monstrated a correlation between the occurrence of lymphoma and plasma GlcSph levels in Gbatm1Karl/
tm1KarlTg(Mx1-cre)1Cgn/0 mice with inducible GBA knock-down in the white blood cell lineage (50, 51). Recently, Nair and colleagues reported that excessive GlcSph acts in GD patients as auto-antigen driving B-cell proliferation and thus directly promotes the development of multiple myeloma, a blood cell cancer occurring with increased incidence in GD patients (52).
LysoGb3, excessively generated in classic FD patients, has also been considered to be patho- logical (53, 54). It has recently been proposed to play a direct role in pain experienced by the patients through sensitization of nociceptive neurons (55). LysoGb3 has also been proposed as culprit in renal disease by causing podocyte damage and fibrosis in the kidney of FD patients (56). Plasma lysoGb3 levels in FD patients have indeed been found to correlate with pain (54). A correlation of plasma (or urinary) lysoGb3 with renal complications has however not been observed. Finally, there is experi- mental evidence that lysoGb3 promotes proliferation of smooth muscle cells in vitro and such effect might explain the characteristic increase in intima media thickness in vasculature of FD patients (38, 53). Theoretically, sphingoid bases like GlcSph and lysoGb3 might also be indirectly harmful as struc- tural mimic of sphingosine-1-phosphate (S1P), interfering with processes governed by this sphingoid base and its receptors (57). Of note, the putative negative effects of the glycosphingoid bases in GD1 and classic FD all occur relatively late in patients’ lives. This implies that associated pathologies would require prolonged exposure to high concentrations of these compounds. In conclusion, the potential toxicity of glycosphingoid bases and putative direct pathological role in glycosphingolipidoses urgent- ly warrants further examination.
Figure 4 | Pro and con of GSL base formation. Pro: avoidance cellular dysfunction by accumulation of dysfunctional lipid laden lysosomes; con: possible toxic side effects.
Chapter 10 Exploiting glycosphingoid base abnormalities for diagnostic purposes and disease monitoring
Whether elevated glycosphingoid bases exert positive or negative effects, or both, their abnormal high concentrations in blood and urine of patients with a glycosphingolipidoses can be used to bio- chemically confirm diagnosis and to demonstrate onset of pathological GSL accumulation (24, 33, 38, 39, 58, 59). A major step forward has been the recent development of suitable methods for very sensitive quantification of glycosphingoid bases in complex biological samples. Glycosphingoid bases like GlcSph and lysoGb3 can be in tandem accurately quantified by LC-MS/MS and the use of 13C5 isotope-encoded natural sphingoid bases as internal standards (Figure 5) (36, 60–62).With advan- ced LC-MS/MS methods, we detected average 300-fold increases in plasma GlcSph in symptomatic GD patients and 200-fold increased lysoGb3 in plasma of males with classic FD (Figure 5D)(36, 60).
These characteristic abnormalities are of great value to verify diagnosis of patients identified in scree- ning programs based on detection of gene abnormalities or reduced enzyme activity in dried blood spots. Diagnosis based on these tests is sometimes ambiguous, for example in the case of mutations with unknown consequence or marked residual enzymatic activity. Demonstration of elevated GlcSph allows sensitive confirmation of diagnosis for GD (36). Likewise, demonstration of markedly eleva- ted plasma lysoGb3 confirms diagnosis of FD in males (60). Adding to the sensitivity of diagnostic use, glycosphingoid base abnormalities already occur in GD patients and male Fabry patients at very young age, prior to overt symptomatology. Marginal increases in plasma glycosphingoid bases should however be treated with caution since these might not be related to the presumed primary defect. For example, we noted that plasma samples of GD patients show besides a more than 100 times increased GlcSph also a modest elevation in lysoGb3 level (43, 63). This is highly relevant since in recent years individuals with abnormalities of unknown significance in the GLA gene are increasingly regarded to be at risk for an atypical manifestation of FD (64, 65). Contrary to classic FD patients, atypical patients express no characteristic acroparesthesia’s and corneal clouding early in life, but only develop one of the isolated late onset symptoms such as unexplained stroke, cardiomyopathy or renal disease. The relative high frequency of such symptoms in the general population makes it likely that simply by mere chance an individual with a GLA polymorphism develops such a common symptom. It should obviously be avoided that in these cases a faulty diagnosis of FD is made. Particularly for an X-linked disorder as FD an incorrect diagnosis in a male has serious consequences since all daughters are la- belled as obligate carriers potentially developing disease and requiring preventive, extremely costly, therapeutic intervention by enzyme replacement therapy. To avoid faulty diagnoses a threshold value of 1.3 pmol/mL plasma lysoGb3 was recently proposed to distinguish true atypical Fabry patients from individuals with alpha-galactosidase A abnormalities without significance (65). This threshold was based on data from 10 non-matched controls and unfortunately plasma from individuals with unex- plained stroke, cardiomyopathy or renal disease in the presence of normal GLA was not analysed. We cannot recommend the strict use of the proposed threshold since in our experience plasma lysoGb3 levels in nearly every GD1 patient examined exceed 1.3 pmol/mL. It is actually conceivable that diverse causes for lysosomal stress, including chronic exposure to lysomotropic drugs, may cause non-specific modest plasma elevations (up to a few pmol/mL).
Figure 5 | LC-MS/MS quantification of glycosphingoid bases. A) 13C5-encoded isotope standards of GlcSph and lysoGb3. B) M/z ratio for analyte and internal standard for GlcSph. C) M/z ratio for analyte and internal standard for lysoGb3. D) GlcSph levels in GD1 patients (n=69) and lysoGb3 in classical FD patients (n=20).
A less debated application of glycosphingoid base measurements is their use to monitor disease manifestation and progression in GD1 patients and classic FD patients (35, 66–70). An intrinsic limitation of circulating glycosphingoid bases is that their origin is unknown and consequently they do not reflect a particular symptom. Increases in plasma levels do however generally point to on- going disease progression and should alert treating clinicians. Regular measurements of plasma glycosphingoid bases in GD and FD patients receiving costly enzyme replacement therapy should be advocated since it gives objective general information on efficacy of the intervention. A lack of response to ERT in plasma GlcSph of GD patients, or plasma lysoGb3 of FD patients, strongly suggests that the treatment is ineffective. Striking in this respect is the noted fast relapse in plasma GlcSph in GD patients following ERT interruption or major dose reductions as well as relapses seen in classic FD males following the formation of neutralizing antibodies against the therapeutic enzyme (66, 69). The measurement of glycosphingoid bases is also of great value to examine the efficacy of experimental therapeutic intervention in animal models. An illustration for this is provided by the work of Dahl and colleagues on gene therapy of GD1 in mice with a reduced GBA activity in white blood cells (71).
Their study and an earlier one by Mistry and co-workers provides information on the relationship of plasma GlcSph and GlcCer-laden macrophages (Gaucher cells) (72). The induction of Gaucher cells in mouse models was found to be associated with a marked increase of plasma GlcSph, indicating that these cells are a major source of the circulating glycosphingoid base. Another strong indication for this is provided by the earlier noted proportional changes in plasma GlcSph and plasma chitotriosidase, a validated biomarker for Gaucher cells (35).
Out of the lysosome box: metabolism of GlcCer in the cytosol mediated by the enzyme GBA2 The present axiom states that GSL degradation is exclusively restricted to lysosomes. An exception to
Chapter 10 this is formed by GlcCer, the only GSL synthesized and present in the cytosolic leaflet of membranes.
Two decades ago we discovered the existence of a non-lysosomal glucosylceramidase, presently named GBA2. The enzyme was found to differ from GBA in inhibitor specificity and ability to degrade artificial β-xyloside substrates (73). The strong membrane association of GBA2 and its intrinsic lability upon dissociation from membranes with detergents hampered purification. Cloning of the cDNA encoding GBA2, independently by Yildiz and colleagues and Boot and co-workers (74, 75), shed the first light on the protein’s structural features. GBA2 is a non-glycosylated 927-amino acid protein (Figure 6A) that is synthesized in the cytosol and subsequently associates strongly with membranes (75).
Literature reports on the subcellular localization of GBA2 are conflicting, ranging from endosomes to the endoplasmic reticulum (75, 76). GBA2 lacks a true transmembrane domain and most likely the catalytic pocket is pressed tightly against the cytosolic leaflet of membranes, fitting with the early observation that the enzyme preferentially uses substrate while embedded in the membrane (73).
At present no 3D-structure of GBA2 is available. Very recently a crystal structure was published for the slightly homologous TxGH116 β-glucosidase from Thermoanaerobacterium xylanolyticum, revealing a N-terminal domain, primarily formed by a two-sheet β-sandwich, and a C-terminal (α/α)6 solenoid domain (77). The C-terminal domain contains the residues previously proposed as the catalytic nucleophile and general acid/base in the archaeal β-glucosidase from Sulfolobus solfataricus and human GBA2. The residues binding the glucose in the −1 subsite are highly conserved between TxGH116 β-glucosidase and human GBA2 (77). GBA2 is a retaining β-glucosidase that uses double displacement in its catalytic mechanism governed by glutamate 527 and aspartate 677 as nucleophile and acid/base residues respectively (78). Hydrophobic iminosugars like AMP-deoxynojirimycin (AMP-DNM, IC50 of 1 nM) ((Figure 6B) and N-butyl-deoxynojirimycin (Zavesca; IC50 of 250 nM) are potent inhibitors of GBA2 enzymatic activity (79, 80). A survey of genomes shows that GBA2 is an ancient and evolutionarily conserved protein, with homologues being present in the three domains of life (Archaea, Bacteria, and Eukarya) (81). GD patients treated with Zavesca at concentrations that inhibit GBA2 activity generally tolerate the drug well. On the other hand, there are now several recent reports of patients with defects in the GBA2 gene developing spastic paraplegia and cerebellar ataxia (82–87). The disease has an early onset and involves muscle weakness and spasticity in upper and lower limbs, ataxia, axonal neuropathy, cognitive impairment, thin corpus callosum, and cerebellar and cerebral atrophy. Likewise in zebrafish knock-down of GBA2 causes neuropathology: antisense morpholino oligonucleotides targeting the zebrafish GBA2 orthologous gene led to abnormal motor behavior and axonal shortening/branching of motor neurons (88). However, GBA2 knock-down in mice does not result in any overt neurological phenotype (Figure 6C)(74). GBA2-deficiency in male mice causes reduced spermatocyte fertility due to a defect in acrosome formation in early post-meiotic germ cells, and malformation of the sperm head (89). These defects are caused by extra-lysosomal accumulation of GlcCer affecting the organization of cytoskeletal structures at the interface between Sertoli and germ cells (90). GBA2 mice also show impaired liver regeneration due to the accumulation of GlcCer which affects cytokine- and growth factor-mediated signaling pathways (91).
Figure 6 | GBA2 in silico, in vitro and in vivo. A) Homology model of GBA2 secondary structure. B) Chemical structure of nanomolar GBA2 inhibitor AMP-DNM and activity-based probe (ABP 1) targeted against GBA and GBA2. C) Labeling of GBA2 and GBA by ABP 1 and immunoblotting of GBA2 and tubulin in brain homogenates of mice heterozygous, wt and knock-out for Gba2. Scale bar = 20 μm. D) In situ visualization of GBA2 labeled in vivo following i.c.v. injection of ABP 1. E) Immunostaining of GBA and GBA2 in the cerebellum of wt mouse. Scale bar = 100 μm.
The discovery of GBA2 in the nineties soon led us to speculate that the enzyme might degrade more GlcCer when GBA activity is reduced as is the case in GD and that such excessive degradation of GlcCer by GBA2 may exert toxic effects and contribute to the symptomatology (92). Mistry and colleagues tested this hypothesis many years later by crossing in GBA2 deficiency in mice with inducible GBA deficiency. Indeed it was observed that lack of GBA2 significantly rescued the clinical phenotype of GD type I mice, particularly the bone-marrow formation defect (93). We and others observed that cells from GD patients show higher levels of GBA2 protein and increased in vitro enzymatic activity when compared to corresponding control cells (94, 95). A similar phenomenon is occurs in brain of NPC mice (96). In NPC cells and tissues, the levels of GBA are known to be significantly reduced and increased GlcCer is observed (97, 98). To test our hypothesis that increased GBA2 activity during GBA deficiency is harmful, we examined the effect of genetic loss of Gba2 in Npc1nih mice that develop a phenocopy of NPC disease in man with major motor neuron loss at end stage disease. This investigation revealed that NPC animals with concomitant GBA2 deficiency live longer and develop
Chapter 10 loss of motor coordination at later age (Figure 7C)(95). The finding fits well with the noted prolonged
survival of Purkinje cells, motor neurons with a relatively high GBA2 content (Figure 6D-E)(95). The beneficial effect of reducing GBA2 in NPC mice was recapitulated by daily administration of 1 mg AMP-DNM per kilo to each animal, a dose sufficient to inhibit GBA2 in the brain of mice (Figure 7B). Our investigation and that by Mistry and colleagues (93), provide evidence that excessive activity of GBA2 has harmful effects. The molecular mechanism for this remains currently elusive. Excessive generation of Cer from GlcCer by GBA2 in the cytosolic leaflet might be detrimental (Figure 7A) given the proposed role of cytosolic Cer in apoptosis and its stimulation of inflammation through activation of PLA2 resulting in increased prostaglandin E2 (92). The question should also be posed whether GBA2 is capable of catalyzing more reactions than just hydrolysis of GlcCer.
Figure 7 | Detrimental role of excessive GBA2 activity during GBA deficiency. A) Scheme of postulated excessive compensatory activity of GBA2 during GBA deficiency. B) Mean survival of Npc1nih mice treated with one mg per kilogram per day of AMP-DNM in the diet and untreated control. C) Mean survival of Npc1-deficient mice wt or knock-out for Gba2.
GBA2, a versatile β-glucosidase generating GlcChol by transglucosylation
A new dimension was recently added to the adaptive responses in glycosphingolipidoses by the realization that the β-glucosidase GBA2 not only removes the glucose moiety from GlcCer (Figure 8A) but also may subsequently transfer it to a cholesterol molecule (99). In this so-called transglucosylation glucosyl-β-D-cholesterol or 1-O-cholesteryl-β-D-glucopyranoside (GlcChol) is produced from GlcCer (Figure 8B).
144 Chapter 10
Figure 8 | Transglucosylation. A) Hydrolysis of GlcCer by a β-glucosidase yielding free glucose and ceramide.
B) Transglycosylation of cholesterol catalyzed by a β-glucosidase using GlcCer as donor of the glucose moiety and leading to the formation of GlcChol. Shown in both schemes are the catalytic residues of the β-glucosidase: acid-base (top) and nucleophile (bottom).
First the catalytic mechanism of retaining glycosidases as GBA and GBA2 (100), coined
‘double-displacement’ by its discoverer Koshland Jr., deserves closer inspection (Figure 8A). The catalytic pocket of the retaining glycosidase employs two adjacent carboxylic acid residues, spaced ~5.5 Å apart, with one acting as catalytic nucleophile and the other as acid/base residue. The deprotonated carboxylate of the nucleophile attacks the substrate’s anomeric C1 carbon, while the carboxylic acid side-chain of the acid/base donates a proton to the inter-glycosidic oxygen. The aglycone is next expelled while concurrently a glycosyl-enzyme intermediate is formed, with an inversed configuration at the anomeric center. To deglycosylate the enzyme, the now deprotonated side-chain of the acid/base abstracts a proton from an incoming water molecule, forming a nucleophilic hydroxyl that attacks the anomeric center (C1) of the glycosyl-enzyme adduct and causes release of the glycone with overall retention of configuration (Figure 8A).
Of interest, several retaining glycosidases have been found to also efficiently transglycosylate, i.e. transferring the released sugar from the substrate to an acceptor other than a free hydroxyl. A thorough historical account of the realization that glycosidases may transglycosylate is provided by the review of Hehre (101). Documented acceptors in transglycosylation reactions in vitro are sugars as noted for chitinases (102), but also hydrophobic alcohols like retinol and sterol. Glew and co-workers firstly demonstrated that purified GBA can catalyze the transfer of the glucose moiety from 4-methylumbelliferyl-β-glucoside to retinol and other alcohols (103). Akiyama and colleagues more recently reported that in vitro GBA generates through transglucosylation 25-NBD-cholesterol- glucoside from GlcCer and artificial 25-NBD-cholesterol (104). We recapitulated their finding with natural cholesterol as acceptor (99). Artificial β-glucosides like 4-methylumbelliferyl-β-glucoside as well as natural GlcCer were found to act as donor in the in vitro reaction (Figure 8B). Next, we discovered that the enzyme GBA2 can also generate GlcChol in vitro through transglucosylation, again using GlcCer as donor. Not surprisingly, GlcChol also proved to be an excellent substrate for in vitro hydrolysis by GBA and GBA2. To establish physiological relevance of GlcChol, we decided to first investigate its natural occurrence in mammalian cells and tissues. For plant and fungal species
HOHO O HO
HO
O C13H28 HN
OH O
C15H32 H
O O
O O
ceramide
GlcCer
HO O HO HO
HOO O O O
H OH
HOHO O HO
HO OH OH O
O O glucose
HO O HO HO
HO OCer H O O
O O GlcCer
HO O HO HO
HOO O O O
HO O HO HO
HO O OH O
O O glucose ceramide
O H H H
H H
H H H
H
GlcChol
cholesterol
A B
Figure 8 | Transglucosylation. A) Hydrolysis of GlcCer by a β-glucosidase yielding free glucose and ceramide.
B) Transglycosylation of cholesterol catalyzed by a β-glucosidase using GlcCer as donor of the glucose moiety and leading to the formation of GlcChol. Shown in both schemes are the catalytic residues of the β-glucosidase:
acid-base (top) and nucleophile (bottom).
First the catalytic mechanism of retaining glycosidases as GBA and GBA2 (100), coined ‘double- displacement’ by its discoverer Koshland Jr., deserves closer inspection (Figure 8A). The catalytic pocket of the retaining glycosidase employs two adjacent carboxylic acid residues, spaced ~5.5 Å apart, with one acting as catalytic nucleophile and the other as acid/base residue. The deprotonated carboxylate of the nucleophile attacks the substrate’s anomeric C1 carbon, while the carboxylic acid side-chain of the acid/base donates a proton to the inter-glycosidic oxygen. The aglycone is next expelled while concurrently a glycosyl-enzyme intermediate is formed, with an inversed configuration at the anomeric center. To deglycosylate the enzyme, the now deprotonated side-chain of the acid/base abstracts a proton from an incoming water molecule, forming a nucleophilic hydroxyl that attacks the anomeric center (C1) of the glycosyl-enzyme adduct and causes release of the glycone with overall retention of configuration (Figure 8A).
Of interest, several retaining glycosidases have been found to also efficiently transglycosylate, i.e.
transferring the released sugar from the substrate to an acceptor other than a free hydroxyl. A thorough
historical account of the realization that glycosidases may transglycosylate is provided by the review of
Hehre (101). Documented acceptors in transglycosylation reactions in vitro are sugars as noted for
chitinases(102), but also hydrophobic alcohols like retinol and sterol. Glew and co-workers firstly
demonstrated that purified GBA can catalyze the transfer of the glucose moiety from 4-
methylumbelliferyl-
β-glucoside to retinol and other alcohols (103). Akiyama and colleagues more
recently reported that in vitro GBA generates through transglucosylation 25-NBD-cholesterol-
glucoside from GlcCer and artificial 25-NBD-cholesterol (104). We recapitulated their finding with
natural cholesterol as acceptor (99). Artificial
β-glucosides like 4-methylumbelliferyl-
β-glucoside as
well as natural GlcCer were found to act as donor in the in vitro reaction (Figure 8B). Next, we
Chapter 10 the existence of sterol-glucosides is well documented (105), but mammals have been little studied
in this respect. Murofushi and co-workers proposed the presence of GlcChol in cultured human fibroblasts and gastric mucosa, but solid analytical proof for this was not provided (106, 107). We therefore developed a sensitive quantitative detection of GlcChol by LC-MS/MS using 13C5-isotope labeled GlcChol as internal standard. With this method GlcChol was detected in human plasma and cultured cells. Analysis of dissected mouse organs revealed that GlcChol is present in almost all tissues examined (99). The relative high amounts in the thymus, several nanomoles per gram of wet weight, are of interest in view of noted abnormalities in NKT and B-cells in GBA-deficient GD patients (108–
110). It has been proposed by Mistry and colleagues that elevated GlcCer or GlcSph via binding to CD1 may be causing this phenomenon (108), but based on our finding GlcChol may also be a serious candidate in this respect.
To obtain insight in the metabolism of GlcChol, both its biosynthesis and degradation were more closely investigated. The related GlcCer is formed by the enzyme glucosylceramide synthase (GCS, EC2.4.1.80) by transfer of the glucose-moiety from UDP-glucose to Cer (111). However, Akiyama and colleagues demonstrated that GCS does not synthesize GlcChol (112), a finding independently confirmed by our group. To determine whether GBA or GBA2 are responsible for formation of GlcChol through transglucosylation reactions with GlcCer as donor, we studied GD1 mice, GBA-deficient LIMP-2 KO mice and GBA2-deficient mice (99). The mice deficient in GBA showed modestly elevated GlcChol in several tissues whereas the GBA2-deficient animals presented a very marked reduction in the sterolglucoside. These findings suggest that in vivo GBA2 is largely responsible for biosynthesis of GlcChol, while GBA normally degrades it. Consistent with this interpretation is the noted increase in plasma GlcChol in symptomatic GD1 patients. GBA2 is well positioned to de facto generate GlcChol when associated to the cytosolic leaflet of endosomes containing both GlcCer and high concentrations of cholesterol. There the transglucosylation equilibrium of the enzyme will favor formation of GlcChol.
Intrinsically, the local concentrations of donors (GlcCer and GlcChol) and acceptors (ceramide and cholesterol) determine the transglucosylation equilibrium of a retaining β-glucosidase. This is nicely illustrated by our finding that high intralysosomal cholesterol concentrations favor formation of GlcChol by GBA rather than its degradation normally catalyzed by the enzyme. In liver of NPC mice, GlcChol was found to be 25-fold increased and induction of lysosomal cholesterol accumulation in cells with U18666A causes a rapid increase in GlcChol, which is abolished by selective inactivation of GBA (99). Of further interest is our finding that pharmacological inhibition of GCS leads to reduction of GlcChol in cultured cells, plasma of mice and plasma of GD patients (99). This strongly suggests that the availability of GlcCer is an important driver in formation of GlcChol through transglucosylation.
Apparently, the exquisite balance of various GlcCer metabolizing enzymes as well as cholesterol concentrations will determine GlcChol formation in subcellular compartments.
The physiological implication of GlcChol is at present subject of speculation. GlcChol is far more water soluble than cholesterol and therefore intrinsically more suited for non-vesicular transport between compartments. Tentatively, water soluble GlcChol formed by transglucosylation at one cellular site could be transported and via the reverse reaction reconverted back at the destination site to more inert cholesterol without any need for ATP. In the context of GD GlcChol deserves reflection. It seems likely that secondary abnormalities in GlcChol occur in cells of GD patients with an imbalance in GBA and GBA2 activities. Future research will need to address whether such abnormalities in GlcChol or in other glucosylated metabolites contribute to particular symptoms associated with GD.
Conclusion and Outlook
As reviewed, adaptive metabolism occurs in glycosphingolipidoses like GD and FD. There is solid proof for formation of glycosphingoid bases in response to intralysosomal accumulation of GSLs.
There is strong evidence pointing to cytosolic metabolism of GlcCer by GBA2 that may generate GlcChol as side product and this pathway appears increased during GBA deficiency.
It may be questioned if these two processes complete the metabolic adaptations occurring in res- ponse to lysosomal GSL accumulation or that others are still missing. In the case of more complex GSLs with extended oligosaccharide moieties a potential alternative compensatory reaction comes to mind, i.e. direct enzymatic removal of the entire glycan of GSLs. Such enzymes actually do exist in nature and are named endoglycoceramidases (EGC) (113, 114). At present several EGCs, with little mutual sequence homology, have been identified in mollusk, leech, earthworm and several pathogenic cestode parasites (115), however not yet in vertebrates. A search for the existence of such enzymes in mammals seems warranted, particularly since a compatible enzymatic activity has been reported by Basu et al. for rats (116–119). Unfortunately, their finding has never been followed up by purification of the responsible enzyme protein. The recent availability of activity-based probes recognizing a variety of retaining β-glucosidases might assist in the discovery of an elusive EGC in mammals. Briefly, cyclophellitol-epoxides and cyclophellitol-aziridines with a particular sugar con- figuration react with high specificity with corresponding retaining glycosidases through irreversible linkage to the catalytic nucleophilic residue. Cyclophellitol-epoxides tagged at C6 with a fluorophore react specifically with GBA while cyclophellitol-aziridines tagged at C1 with a fluorophore show broad in class reactivity with several β-glucosidases (120, 121). We already observed that bacterial EGCase reacts well with a β-glucopyranosyl-configured cyclophellitol-aziridine tagged with either a fluorophore or biotin (122). The biotin-tagged ABP might be conveniently employed in a future search for the elusive mammalian endoglycoceramidase and facilitate its purification and subsequent identification by proteomics.
The recognition of adaptive metabolism prompts further studies on the excessive lysoGb3 in FD and the excessive GlcSph and GlcChol in GD as pathological factors contributing to symp- tomatology. If so, could novel drugs be envisioned to ameliorate these pathological effects? Unfor- tunately, inhibition of AC, the enzyme responsible for generation of glycosphingoid bases, does not seem wise since it will cause impaired lysosomal degradation of ceramide mimicking Farber disease, a severe neurological disorder. However, further investigations on the therapeutic value of available brain-permeable combined inhibitors of GBA2 and GCS are appealing for the juvenile neuronopa- thic variant of GD (type 3) are appealing, particularly given the positive outcome observed for the well tolerated inhibitor treatment of NPC mice (99, 123) and similar positive findings made earlier by other investigators with Sandhoff mice (124), another glycosphingolipidoses.
Chapter 10 References
1. Thudichum, J. 1884. A treatise on the chemical constitution of the brain. Bailliere Tindall and Cox, London.
2. Wennekes, T., R. J. B. H. N. van den Berg, R. G. Boot, G. a van der Marel, H. S. Overkleeft, and J. M. F. G.
Aerts. 2009. Glycosphingolipids-nature, function, and pharmacological modulation. Angew. Chem. Int. Ed.
Engl. 48: 8848–69.
3. Hancock, J. F. 2006. Lipid rafts: contentious only from simplistic standpoints. Nat. Rev. Mol. Cell Biol. 7: 456–
4. 62. Sonnino, S., and A. Prinetti. 2013. Membrane domains and the “lipid raft” concept. Curr. Med.
Chem. 20: 4–21.
5. Lingwood, D., and K. Simons. 2010. Lipid rafts as a membrane-organizing principle. Science (80-. ). 327:
46–50.
6. Merrill, A. H. 2011. Sphingolipid and glycosphingolipid metabolic pathways in the era of sphingolipidomics.
Chem. Rev. 111: 6387–422.
7. Schulze, H., and K. Sandhoff. 2014. Sphingolipids and lysosomal pathologies. Biochim. Biophys. Acta - Mol.
Cell Biol. Lipids. 1841: 799–810.
8. Hanada, K., K. Kumagai, S. Yasuda, Y. Miura, M.
Kawano, M. Fukasawa, and M. Nishijima. 2003.
Molecular machinery for non-vesicular trafficking of ceramide. Nature. 426: 803–9.
9. Ichikawa, S., H. Sakiyama, G. Suzuki, K. I. Hidari, and Y. Hirabayashi. 1996. Expression cloning of a cDNA for human ceramide glucosyltransferase that catalyzes the first glycosylation step of glycosphingolipid synthesis. Proc. Natl. Acad. Sci. U. S. A. 93:
12654.
10. D’Angelo, G., S. Capasso, L. Sticco, and D. Russo.
2013. Glycosphingolipids: synthesis and functions.
FEBS J. 280: 6338–53.
11. Merrill, A. H., M. D. Wang, M. Park, and M. C.
Sullards. 2007. (Glyco)sphingolipidology: an amazing challenge and opportunity for systems biology. Trends Biochem. Sci. 32: 457–68.
12. Kolter, T., and K. Sandhoff. 2010. Lysosomal degradation of membrane lipids. FEBS Lett. 584:
1700–1712.
13. Trinchera, M., R. Ghidoni, S. Sonnino, and G.
Tettamanti. 1990. Recycling of glucosylceramide and sphingosine for the biosynthesis of gangliosides and sphingomyelin in rat liver. Biochem. J. 270: 815–
14. Pyne, S., D. R. Adams, and N. J. Pyne. 2016. 20.
Sphingosine 1-phosphate and sphingosine kinases in health and disease: Recent advances. Prog. Lipid Res.
62: 93–106.
15. Serra, M., and J. D. Saba. 2010. Sphingosine 1-phosphate lyase, a key regulator of sphingosine 1-phosphate signaling and function. Adv. Enzyme Regul. 50: 349–62.
16. Platt, F. M. 2014. Sphingolipid lysosomal storage disorders. Nature. 510: 68–75.
17. Schulze, H., and K. Sandhoff. 2011. Lysosomal lipid storage diseases. Cold Spring Harb. Perspect. Biol. 3:
18. Xu, Y.-H., S. Barnes, Y. Sun, and G. A. Grabowski.
2010. Multi-system disorders of glycosphingolipid and ganglioside metabolism. J. Lipid Res. 51: 1643–
19. Futerman, A. H., and G. van Meer. 2004. The cell 75.
biology of lysosomal storage disorders. Nat. Rev. Mol.
Cell Biol. 5: 554–65.
20. Wang, R. Y., O. A. Bodamer, M. S. Watson, and W. R.
Wilcox. 2011. Lysosomal storage diseases: diagnostic confirmation and management of presymptomatic individuals. Genet. Med. 13: 457–84.
21. Matern, D., D. Oglesbee, and S. Tortorelli. 2013.
Newborn screening for lysosomal storage disorders and other neuronopathic conditions. Dev. Disabil.
Res. Rev. 17: 247–53.
22. Conzelmann, E., and K. Sandhoff. Partial enzyme deficiencies: residual activities and the development of neurological disorders. Dev. Neurosci. 6: 58–
23. Farfel-Becker, T., E. B. Vitner, and A. H. Futerman. 71.
2011. Animal models for Gaucher disease research.
Dis. Model. Mech. 4: 746–752.
24. Ferraz, M. J., W. W. Kallemeijn, M. Mirzaian, D.
Herrera Moro, A. Marques, P. Wisse, R. G. Boot, L.
I. Willems, H. S. Overkleeft, and J. M. Aerts. 2014.
Gaucher disease and Fabry disease: new markers and insights in pathophysiology for two distinct glycosphingolipidoses. Biochim. Biophys. Acta. 1841:
811–25.
25. Holleran, W. M., Y. Takagi, and Y. Uchida. 2006.
Epidermal sphingolipids: metabolism, function, and roles in skin disorders. FEBS Lett. 580: 5456–
26. Staretz-Chacham, O., T. C. Lang, M. E. LaMarca, 66.
D. Krasnewich, and E. Sidransky. 2009. Lysosomal storage disorders in the newborn. Pediatrics. 123:
1191–207.
27. Sidransky, E., D. M. Sherer, and E. I. Ginns. 1992.
Gaucher disease in the neonate: a distinct Gaucher phenotype is analogous to a mouse model created by targeted disruption of the glucocerebrosidase gene.
Pediatr. Res. 32: 494–8.
28. Germain, D. P. 2010. Fabry disease. Orphanet J. Rare Dis. 5: 30.
29. Durant, B., S. Forni, L. Sweetman, N. Brignol, X.-L. Meng, E. R. Benjamin, R. Schiffmann, and J.-S. Shen. 2011. Sex differences of urinary and kidney globotriaosylceramide and lyso- globotriaosylceramide in Fabry mice. J. Lipid Res. 52:
1742–6.
30. Taguchi, A., H. Maruyama, M. Nameta, T. Yamamoto, J. Matsuda, A. B. Kulkarni, H. Yoshioka, and S.
Ishii. 2013. A symptomatic Fabry disease mouse model generated by inducing globotriaosylceramide synthesis. Biochem. J. 456: 373–383.
31. Reczek, D., M. Schwake, J. Schröder, H. Hughes, J.
Blanz, X. Jin, W. Brondyk, S. Van Patten, T. Edmunds, and P. Saftig. 2007. LIMP-2 is a receptor for lysosomal mannose-6-phosphate-independent targeting of beta- glucocerebrosidase. Cell. 131: 770–83.
32. Blanz, J., J. Groth, C. Zachos, C. Wehling, P. Saftig, and M. Schwake. 2010. Disease-causing mutations
2 (LIMP-2) reveal the nature of binding to its ligand beta-glucocerebrosidase. Hum. Mol. Genet. 19: 563–
33. Goldstein, J. L., R. A. DeBose-Boyd, and M. S. Brown. 72.
2006. Protein sensors for membrane sterols. Cell. 124:
35–46.
34. Settembre, C., A. Fraldi, D. L. Medina, and A. Ballabio.
2013. Signals from the lysosome: a control centre for cellular clearance and energy metabolism. Nat. Rev.
Mol. Cell Biol. 14: 283–96.
35. Dekker, N., L. van Dussen, C. E. M. Hollak, H.
Overkleeft, S. Scheij, K. Ghauharali, M. J. van Breemen, M. J. Ferraz, J. E. M. Groener, M. Maas, F. A.
Wijburg, D. Speijer, A. Tylki-Szymanska, P. K. Mistry, R. G. Boot, and J. M. Aerts. 2011. Elevated plasma glucosylsphingosine in Gaucher disease: relation to phenotype, storage cell markers, and therapeutic response. Blood. 118: e118-27.
36. Mirzaian, M., P. Wisse, M. J. Ferraz, H. Gold, W. E.
Donker-Koopman, M. Verhoek, H. S. Overkleeft, R. G. Boot, G. Kramer, N. Dekker, and J. M. F. G.
Aerts. 2015. Mass spectrometric quantification of glucosylsphingosine in plasma and urine of type 1 Gaucher patients using an isotope standard. Blood Cells, Mol. Dis. 54: 307–314.
37. Gaspar, P., W. W. Kallemeijn, A. Strijland, S. Scheij, M. Van Eijk, J. Aten, H. S. Overkleeft, A. Balreira, F. Zunke, M. Schwake, C. Sá Miranda, and J. M.
F. G. Aerts. 2014. Action myoclonus-renal failure syndrome: diagnostic applications of activity-based probes and lipid analysis. J. Lipid Res. 55: 138–
38. Aerts, J. M., J. E. Groener, S. Kuiper, W. E. Donker-45.
Koopman, A. Strijland, R. Ottenhoff, C. van Roomen, M. Mirzaian, F. a Wijburg, G. E. Linthorst, A. C. Vedder, S. M. Rombach, J. Cox-Brinkman, P. Somerharju, R.
G. Boot, C. E. Hollak, R. O. Brady, and B. J. Poorthuis.
2008. Elevated globotriaosylsphingosine is a hallmark of Fabry disease. Proc. Natl. Acad. Sci. U. S. A. 105:
2812–2817.
39. Miyatake, T., and K. Suzuki. 1973. Additional deficiency of psychosine galactosidase in globoid cell leukodystrophy: an implication to enzyme replacement therapy. Birth Defects Orig. Artic. Ser. 9:
136–40.
40. Svennerholm, L., M. T. Vanier, and J. E. Månsson. 1980.
Krabbe disease: a galactosylsphingosine (psychosine) lipidosis. J. Lipid Res. 21: 53–64.
41. Ferraz, M. J., A. R. A. Marques, P. Gaspar, M. Mirzaian, C. van Roomen, R. Ottenhoff, P. Alfonso, P. Irún, P.
Giraldo, P. Wisse, C. Sá Miranda, H. S. Overkleeft, and J. M. Aerts. 2015. Lyso-glycosphingolipid abnormalities in different murine models of lysosomal storage disorders. Mol. Genet. Metab. 117:
186–93.
42. Yamaguchi, Y., N. Sasagasako, I. Goto, and T.
Kobayashi. 1994. The synthetic pathway for glucosylsphingosine in cultured fibroblasts. J.
Biochem. 116: 704–10.
43. Ferraz, M. J., A. R. A. Marques, M. D. Appelman, M. Verhoek, A. Strijland, M. Mirzaian, S. Scheij, C. M. Ouairy, D. Lahav, P. Wisse, H. S. Overkleeft, R. G. Boot, and J. M. Aerts. 2016. Lysosomal
glycosphingolipid catabolism by acid ceramidase:
formation of glycosphingoid bases during deficiency of glycosidases. FEBS Lett. 590: 716–25.
44. Lavoie, P., M. Boutin, and C. Auray-Blais. 2013.
Multiplex analysis of novel urinary lyso-Gb3-related biomarkers for Fabry disease by tandem mass spectrometry. Anal. Chem. 85: 1743–52.
45. Ballabio, A., and V. Gieselmann. 2009. Lysosomal disorders: from storage to cellular damage. Biochim.
Biophys. Acta. 1793: 684–96.
46. Gieselmann, V. 1995. Lysosomal storage diseases.
Biochim. Biophys. Acta. 1270: 103–36.
47. Cox, T. M. 2003. Future perspectives for glycolipid research in medicine. Philos. Trans. R. Soc. Lond. B.
Biol. Sci. 358: 967–73.
48. Pastores, G. M. 2009. Krabbe disease: an overview.
Int. J. Clin. Pharmacol. Ther. 47 Suppl 1: S75- 49. de Fost, M., T. A. Out, F. A. de Wilde, E. P. M. Tjin, S. 81.
T. Pals, M. H. J. van Oers, R. G. Boot, J. F. M. G. Aerts, M. Maas, S. vom Dahl, and C. E. M. Hollak. 2008.
Immunoglobulin and free light chain abnormalities in Gaucher disease type I: data from an adult cohort of 63 patients and review of the literature. Ann. Hematol.
87: 439–449.
50. Pavlova, E., S. Wang, J. Archer, N. Dekker, J. Aerts, S. Karlsson, and T. Cox. 2013. B cell lymphoma and myeloma in murine Gaucher’s disease. J. Pathol. 231:
88–97.
51. Pavlova, E. V, J. Archer, S. Wang, N. Dekker, J. M.
Aerts, S. Karlsson, and T. M. Cox. 2015. Inhibition of UDP-glucosylceramide synthase in mice prevents Gaucher disease-associated B-cell malignancy. J.
Pathol. 235: 113–24.
52. Nair, S., A. R. Branagan, J. Liu, C. S. Boddupalli, P. K. Mistry, and M. V. Dhodapkar. 2016. Clonal Immunoglobulin against Lysolipids in the Origin of Myeloma. N. Engl. J. Med. 374: 555–561.
53. Rombach, S. M., B. van den Bogaard, E. de Groot, J. E. M. Groener, B. J. Poorthuis, G. E. Linthorst, B.-J. H. van den Born, C. E. M. Hollak, and J. M. F.
G. Aerts. 2012. Vascular aspects of Fabry disease in relation to clinical manifestations and elevations in plasma globotriaosylsphingosine. Hypertension. 60:
998–1005.
54. Biegstraaten, M., C. E. M. Hollak, M. Bakkers, C. G.
Faber, J. M. F. G. Aerts, and I. N. van Schaik. 2012.
Small fiber neuropathy in Fabry disease. Mol. Genet.
Metab. 106: 135–41.
55. Choi, L., J. Vernon, O. Kopach, M. S. Minett, K. Mills, P.
T. Clayton, T. Meert, and J. N. Wood. 2015. The Fabry disease-associated lipid Lyso-Gb3 enhances voltage- gated calcium currents in sensory neurons and causes pain. Neurosci. Lett. 594: 163–8.
56. Sanchez-Niño, M. D., A. B. Sanz, S. Carrasco, M. A. Saleem, P. W. Mathieson, J. M. Valdivielso, M. Ruiz-Ortega, J. Egido, and A. Ortiz. 2011.
Globotriaosylsphingosine actions on human glomerular podocytes: implications for Fabry nephropathy. Nephrol. Dial. Transplant. 26: 1797–
802.
57. Barbey, F., N. Brakch, A. Linhart, N. Rosenblatt-Velin, X. Jeanrenaud, S. Qanadli, B. Steinmann, M. Burnier,
