The cellular pathology of lysosomal diseases†
No conflicts of interest were declared.
Abstract
With a constitutive recycling function and the capacity to digest exogenous material as well as endogenous organelles in the process of autophagy, lysosomes are at the heart of the living cell. Dynamic interactions with other cellular components ensure that the lysosomal compartment is a central point of convergence in countless diverse diseases. Inborn lysosomal (storage) diseases represent about 70 genetically distinct conditions, with a combined birth frequency of about 1 in 7500. Many are associated with macromolecular storage, causing physical disruption of the organelle and cognate structures; in neurons, ectopic dendritogenesis and axonal swelling due to distension with membraneous tubules and autophagic vacuoles are observed. Disordered autophagy is almost universal in lysosomal diseases but biochemical injury due to toxic metabolites such as lysosphingolipid molecules, abnormal calcium homeostasis and endoplasmic reticulum stress responses and immune-inflammatory processes occur. However, in no case have the mechanistic links between individual clinico-pathological manifestations and the underlying molecular defect been precisely defined. With access to the external fluid-phase and intracellular trafficking pathways, the lysosome and its diseases are a focus of pioneering investment in biotechnology; this has generated innovative orphan drugs and, in the case of Gaucher's disease, effective treatment for the haematological and visceral manifestations. Given that two-thirds of lysosomal diseases have potentially devastating consequences in the nervous system, future therapeutic research will require an integrative understanding of the unitary steps in their neuro pathogenesis. Informative genetic variants illustrated by patients with primary defects in this organelle offer unique insights into the central role of lysosomes in human health and disease. We provide a conspectus of inborn lysosomal diseases and their pathobiology; the cryptic evolution of events leading to irreversible changes may be dissociated from the cellular storage phenotype, as revealed by the outcome of therapeutic gene transfer undertaken at different stages of disease. Copyright © 2011 Pathological Society of Great Britain and Ireland. Published by John Wiley & Sons, Ltd.
History of the lysosome
The discovery of the lysosome by Christian de Duve (who, for this and the discovery of the peroxisome, was awarded the Nobel Prize in Medicine or Physiology in 1974) marked the birth of cell biology as a radical new science 1. A fellow Belgian scientist, Henri-Gery Hers, identified the first lysosomal disorder, Pompe disease: gross accumulation of glycogen in the heart and skeletal muscles disorder was found to be due to a deficiency of a hitherto unknown acid maltase (acid α-1,4-glucosidase) 2. Acid maltase activity could be sedimented by centrifugation but was easily released into the supernatant by detergent treatment; this release by digitonin was accompanied by acid phosphatase previously localized to the lysosome 3. Hers's seminal work enabled him to predict correctly that: ‘other deposition diseases might be explained on the basis of the absence of other lysosomal enzymes’. The prescient term Hers later proposed for these conditions was ‘inborn lysosomal diseases’ 4, which expresses a defining and useful specificity and is reminiscent of Garrod. Table 1 depicts lysosomal diseases classified on the basis of this premise.
| Mucopoly-saccharidoses | Sphingolipidoses | Glycoproteinoses | Miscellaneous |
|---|---|---|---|
| Hurler, Scheie (I) | Tay–Sachs | α-Fucosidosis | Cathepsin C |
| Hunter (II) | Gaucher | Aspartylglucosaminuria | Cathepsin K |
| Sanfillipo A–D (III) | Krabbe | α-Mannosidosis | Pompe (GSD type II) |
| Morquio A, B (IV) | Anderson–Fabry | β-Mannosidosis | Niemann–Pick C types I, II |
| Maroteaux–Lamy (VI) | Farber | Sialidosis | Neurónal ceroid lipofuscinoses (CLN 1–10) |
| Sly (VII) | Metachromatic leukodystrophy | Galactosialidosis | Wolman (cholesterol ester) |
| Hyaluronidase deficiency (X) | Niemann–Pick A and B | Kanzaki | Hermansky–Pudlak (1–9) |
| GM1 gangliosidosis | Chédiak–Higashi | ||
| Sandhoff | Cystinosis | ||
| GM2 activator deficiency | Salla | ||
| Methylmalonic aciduria (CblF) | |||
| Danon | |||
| I-cell disease (mucolipidoses II and III) | |||
| Mucolipidosis (IV) | |||
| Multiple sulphatase deficiency | |||
| Sphingolipid activator protein deficiencies (A–D) | |||
| T2-family acidic endoribonuclease (RNASET2) | |||
| SCARB2/LIMP-2 deficiency |
To ‘explore the cell with a centrifuge’, de Duve adopted a biochemical stratagem, thus guaranteeing from the outset that the organelles identified were described in functional terms. The discovery of lysosomes (and peroxisomes), each, like mitochondria, with a unique complement of enzymes, allowed the interactions of these metabolic entities to be studied well in advance of their descriptive anatomy revealed by microscopy. Cell biology is an interdisciplinary science; without inhibition, its modern exponents freely utilize protein chemistry, molecular genetics to manipulate expression of nuclear genes, and the biophysics of fluorescence laser microscopy—including time-lapse and confocal techniques—solely to understand the living cell as an integrated universe of fluid but dynamically interacting structures.
Lysosomal diseases
Individually rare, it is now known that about 1 in 7500 live-born infants in many populations will have a lysosomal disease 5, 6. While these conditions shorten life, typically there is an interval after birth before the individual disorder becomes clinically manifest. Once established, the protean manifestations of lysosomal diseases do not relent; they impair the quality of life for the individual, their immediate family and carers. The personal and societal burden is immense 7.
Christian de Duve and succeeding investigators have placed the molecular physiology of the lysosome at the heart of cellular activity but the lexicon of its constituents is burgeoning 8-10. The field has been informed, and in countless instances defined, by scientific exploration of lysosomal diseases 11, 12. That deficiency of lysosomal acid maltase activity impairs glycogen degradation was unexpected, but the finding identified Pompe disease as a paradigm for other inborn errors of the lysosome affecting the ebb and flow of complex cellular macromolecules. 13. The discovery of acid maltase deficiency had far-reaching significance because it emphasized the specific role of lysosomes in the constitutive turnover and dynamic remodelling of intracellular components through autophagy 1, 14.
The taxonomy of lysosomal diseases is unsatisfactory: this reflects the evolving scientific context of discovery and understanding, spanning a biochemical categorization (according to the principal storage molecules identified; Table 1) or assignment according to the defective molecular cell physiology (functional classification). The right-hand column includes several miscellaneous disorders, reflecting the protean functions of the lysosome. While an impartial taxonomy based on a genetic classification might be preferred, understanding lysosomal disorders due to defects in non-lysosomal proteins requires an integrated and convincing biological approach.
Examples of trafficking and biogenesis defects include Niemann–Pick disease type C, the Hermansky–Pudlak syndrome (1–9) and Chédiak–Higashi disease. In Hermansky–Pudlak syndrome, there is defective biogenesis of organelles due to mutations in adaptin molecules that control trafficking of proteins to lysosomes, melanosomes and platelet δ-granules 15. Chédiak–Higashi disease is a microtubule polymerization defect due to mutations in the gene encoding a lysosomal trafficking regulator; the disease impairs phagocytosis and viral defences and partial albinism occurs. Both disorders affect the lytic granule, a secretory lysosome which carries out the granule functions of cytotoxic T lymphocytes and natural killer cells—as well as neutrophilic polymorphonuclear leukocytes and other haematogenous cells 16.
Functional complementation of the lysosome
Biochemical characterization of cells cultured from the tissues of patients with different lysosomal diseases, by Elizabeth Neufeld and colleagues, led to the discovery of functional complementation 17. Mutual and reciprocal secretion of factors that correct the biochemical abnormality in individual lysosomal diseases was shown in fibroblasts obtained from genetically distinct mucopolysaccharidoses (now known as MPS 1 and MPS 2). These studies ultimately showed how nascent proteins destined for the lysosome were sorted to the organelle on the basis of a molecular signature—the mannose 6-phosphate moiety serving as a lysosomal recognition signal. Later identification of the defect in patients with an inclusion body disease—I-cell disease (mucolipidoses II and III)—led to the discovery that specific enzymes responsible for the decorating lysosomal glycoproteins with the mannose 6-phosphate residue, are defective in this disorder 18. High-mannose oligosaccharides promote folding of the nascent lysosomal glycoproteins in the endoplasmic reticulum by interacting with lectin-like chaperone molecules; after phosphorylation in the Golgi, the proteins bind to Man-6-P receptors, a key step in their transport to lysosomes. In I-cell disease, nascent glycoproteins destined for the lysosome are synthesized but are incorrectly sorted to the organelle appear in the plasma 19, 20. It turns out that the mannose 6-phosphate receptor system (cation-independent mannose-phosphate/insulin-like growth factor II receptor system) is leaky, so that a quota of the lysosomal proteins egress normally from cells—only to be taken up via such receptors on the surface of other cells at a distance. This secretion–recapture system has recently been exploited for specific enzyme augmentation therapies, now licensed for the treatment of several lysosomal diseases 21, 22 (Table 2). The notable exception to this dogma is the first (and most successful) therapy for a lysosomal disease: uptake of therapeutic imiglucerase (Cerezyme™, partially deglycoslyated human β-glucocerebrosidase) by the diseased macrophages in patients with Gaucher's disease is targeted to mannose-binding lectins expressed on the macrophage surface 23-25.
| Gaucher's disease types 1Ψ * and 3* †‡ |
| Anderson–Fabry disease* |
| Mucopolysaccharidosis type I* |
| Mucopolysaccharidosis type II* |
| Mucopolysaccharidosis type IIIA |
| Mucopolysaccharidosis type IVA |
| Mucopolysaccharidosis VI* |
| Glycogenosis type ll A* (Pompe/acid maltase deficiency) |
| Niemann–Pick disease type B |
| Niemann–Pick disease type C* † |
| α-Mannosidosis |
| Metachromatic leukodystrophy |
- Ψ Imiglucerase (Cerezyme™), velaglucerase (VPRIV™) authorised; taliglucerase is in late-stage clinical trials for type I Gaucher disease.
- * Licensed in Europe under orphan drug legislation.
- † Miglustat (Zavesca™) licensed orally active agent to inhibit formation of glycosphingolipids.
- ‡ Eliglustat tartrate; orally active substrate inhibitor in Phase 3 clinical trials.
Further research into those diseases where the recognition signal directing nascent lysosomal enzymes to the organelle is independent of mannose 6-phosphate, has led to the identification of internal residues within certain lysosomal proteins, such as acid β-glucosylceramidase (deficient in Gaucher's disease), which are recognized by the lysosomal membrane protein, LIMP-2. This protein proves to be the mediating chaperone for β-glucosylceramidase in certain tissues, such as the brain and kidney glomerulus 26—a discovery requiring sophisticated biochemistry but also supported by the study of another informative human disease: action myoclonus–renal failure syndrome, in which the macrophages appear normal but the patient has other consequences of enzyme deficiency at other sites, such as the brain, as well as the kidney. Mice lacking SCARB2/Limp2—an ancestral protein conserved in Caenorhabditis elegans– also have cochear injury, but the kidney appears principally affected by pelvi–ureteric obstruction rather than glomerular collapse. The condition has very few, if any, features in common with Gaucher's disease but myoclonic epilepsy is a feature of the chronic neuronopathic variant 27, 28.
The meticulous investigation of patients with lysosomal defects, unusual diagnostic biochemistry and rare clinical phenotypes has identified several activators, modifiers and other cofactors which are required for the physiological action of some lysosomal enzymes. Among this group are the sphingolipid activator proteins (saposins) 29. Rare patients with Gaucher's disease have been reported with normal acid β-glucosidase activity when their leukocytes or fibroblasts are assayed with (artificial) water-soluble substrates used in routine diagnosis: deficiency of saposin C in these patients however impairs the ability of β-glucosylceramidase to hydrolyse the natural sphingolipid substrate, β-glucosylceramide 30. Intensive studies of patients with multiple sulphatase deficiency (Austin's disease), using either a genetic or biochemical stratagems, led to the identification of sulphatase modifying factor (SUMF1) 31, 32. SUMF1 ensures the formation of a formylglycine residue essential for the activity of all 12 known human sulphatases by catalysing the oxidation of a conserved cysteine at the active site 31. Multiple sulphatase deficiency is characterized by ichthyosis (steroid sulphatase), metachromatic leukodystrophy (arylsulphatase A) and protean features of mucopolysaccharidoses II, IIIA, IIID, IV and VI (iduronate sulphatase, heparin sulphatase, N-glucosamine 6-sulphatase, N-galactosamine-6-sulphatase and arylsulphatase B) but arises from mutations in the single SUMF1 gene.
In the half-century since identification, we have a useful dossier of lysosomal diseases that is at least partially understood and a few disorders that can be effectively treated. The debt of science to medicine in this field is immense, and the catholic role of the lysosome in recycling membrane and other organelles, as well as the turnover of molecules and particulate matter delivered through endocytosis and phagocytosis, ensures its status as a workhorse of the cell.
Pathological features of lysosomal diseases
Methods familiar to medical scientists, such as microscopy and basic histochemical analysis, attracted attention to the most obvious end-stage abnormalities, and the inaccurate term ‘lysosomal storage diseases’ (LSDs) is persistent. Not all lysosomal diseases are storage disorders, and while the tinctorial properties of intralysosomal material in particular tissues in a given disease may be striking, the cause of neurological deficits, with or without cell death, of visceral fibrosis, of hypertrophy or hyperplasia and related inflammatory changes, sometimes accompanied by vascular obstruction, is often enigmatic. Investment in therapeutics has led to a rebirth of clinical and pathological descriptions of lysosomal diseases but it now obvious that former classifications fail to provide useful mechanistic insight into each newly emerging disease complex.
Lysosomes are near-ubiquitous organelles and lysosomal diseases can affect any tissue or organ system. Imbued with a wide complement of acid hydrolases, each with specificity towards a particular chemical linkage, lysosomal enzymes act in concert to degrade a diverse constellation of macromolecular complexes, which may be predominantly lipid, polysaccharide, protein or nuclear acid in nature. These substrates are broken down by the stepwise action of individual enzymes in the coordinate hydrolysis of successive specific bonds; if one member of the enzymatic machinery is deficient, then the digestive sequence is arrested.
In Pompe disease, failure to remodel glycogen polymers leads to the accumulation of massive glycogen molecules within the vacuolar spaces of an expanded lysosomal compartment and also the cytoplasm. In a typical lysosomal disease, these ‘storage’ deposits are sequestrated principally in those organs where pathological injury is identified, and tissues in which lysosomal turnover of the parent macromolecule is most rapid.
In Tay–Sachs disease (GM2 gangliosidosis), one of the first clinically recognized lysosomal storage diseases associated with prominent neurological features, the microscopic appearance of the neurones thoughout the nervous system in the end-stage of the disease is almost universally abnormal 33 (Figure 1). Accumulation of the glycosphingolipid, GM2 ganglioside, due to deficiency of β-hexosaminidase A generates whorled bodies (membraneous cytoplasmic bodies) representing abnormal lysosomes which may displace the nucleus. Ultimately, the disease is characterized by a gain in brain weight due to gliosis, accompanied by neuronal atrophy and axonal defects of abnormalities in myelin and the formation of spheroid bodies, as demonstrated by studies in knock-out mice that replicate the enzymatic defect. While it is tempting to invoke a direct ‘cause-and-effect’ between the accumulation of membraneous cytoplasmic bodies and neuronal death, the complex effects on individual neurons and massive recruitment of microglia and astroglia defy current understanding.
Timed therapeutic intervention reveals irretrievable disease in GM2 gangliosidosis. (A) Kaplan–Meier survival curves of untreated and rAAV-treated Sandhoff disease (SD) mice. Survival of SD mice infused bilaterally into the striatum and cerebellum with rAAV at 4 (4WPBI), 8 (8WPBI) and 12 (12WPBI) weeks of age. Controls were untreated wild-type (WT) and SD mice. Infusion of rAAV 2/2 at the presymptomatic age of 12 weeks did not improve survival. (B–E) Histochemical detection and (F) chromatographic analysis of glycoconjugates from SD mouse brain. (B–E) PAS staining of brain sections; the mice were killed at 12 weeks of age: (B) WT and (C) untreated SD. SD mouse killed at the humane end point of: (D) 145 days of age after rAAV infusion at 12 weeks of age; and (E) 120 days old untreated SD. (F) Glycosphingolipids were extracted from each cerebral hemisphere and processed individually: 1, 21 week-old WT; 2, untreated SD at 16 weeks; SD mice infused bilaterally into the striatum and cerebellum with rAAV 2/2 at: 3, 4 weeks of age and killed at 31 weeks; 4, 8 weeks of age and killed at 22 weeks; and 5, 12 weeks of age and killed at 16 weeks. SD mice were killed when they reached their humane end point. (G, H) No direct correlation was found between glycoconjugate accumulation and microglia activation with survival when therapy was initiated just before the development of clinical manifestations
Disturbed autophagy
Autophagy was codified by de Duve in 1963; he identified membrane-bound vesicles containing endogenous organelles undergoing digestion in a manner now subclassified as macroautophagy; various stimuli, including starvation and glucagon administration, stimulate this process. In Pompe disease, the glycogen substrate accumulates in membrane-bound vacuoles as well as the cytoplasm—confirming the prescient insights of de Duve and his colleagues 14.
Autophagy, critical for growth, development and survival in eukaryotic organisms, is a highly conserved process regulated by more than 30 autophagy-related proteins (Atg), many of which have been discovered in yeast 34. Lysosomes digest intracellular components and other organelles in autophagosomes—a process greatly enhanced under conditions of nutrient starvation. These structures have a double-membrane and are formed by the merging of lysosomes with elongated phagophores which arise by homotypic fusion from precursor structures that originate from plasma membrane and other components such as mitochondria.
A definitive indication of the importance of autophagy in health is provided by another inborn error of lysosomal metabolism, Danon disease—a vacuolar skeletal myopathy and cardiomyopathy with neurodegenerative features in affected hemizygous males. The X-chromosome-linked LAMP-2 lysosomal membrane glycoprotein is important for the maturation and integrity of lysosomes. LAMP-2 is mutated in Danon disease, in which the fusion of lysosomes with autophagosomes mediated by dynein is impaired. These discoveries provides strong evidence that defective macroautophagy is fundamental to the development of this fatal disease and, moreover, that the contribution of lysosomes to this process is essential for life 35.
Neurones do not replicate in adult life and, with abundant oxidative metabolism and extensive membrane structures, have a highly developed lysosomal network for membrane recycling: lysosomes are trafficked up and down axonal fibres using filamentous molecular motors to ensure that, as the cytoplasm extends to the far reaches and ramifications of the axon, distribution of the organelle is maintained 36. In lysosomal diseases, electron microscopy of neurones and other cells shows formation of vacuoles and lysosomes distended by recognizable cellular cargo—residual organelles, glycogen and membrane fragments, which are direct evidence of disturbed autophagy 37.
It is now apparent that, even with activation of compensatory pathways, a more generalized defect of autophagic clearance occurs. Not only can the cell no longer compensate for failing macromolecular turnover by exocytosis of undigested material, but in several lysosomal diseases affecting the brain [multiple sulphatase deficiency (MSD), Sanfilippo disease type A (MPS IIIA), GM2 gangliosidosis] activation of molecular signatures of enhanced (and functionally arrested) expansion of the autophagic apparatus have been reported. In embryonic cells from MSD and MPS IIIA mice, there is greater processing of the cytoplasmic LC3-I molecule to its cleaved and lipidated product, LC3-II—a universal hallmark of autophagic activation. Decreased intracellular co-localization of autophagosome punctae containing LC3-II molecules with the lysosomal membrane protein marker, LAMP-1, shows that the normal fusion of the activated complement of autophagosomes with lysosomes is arrested 38. The consequential effect of defective autophagy is the build-up of effete cellular matter, including ubiquitinated proteins normally destined for recycling in the lysosomal compartment.
Coordinate genetic control of lysosomal biogenesis and autophagy
In many lysosomal diseases, expansion of the lysosomal network is accompanied by increased activity (and release) of many lysosomal proteins 39. A coordinated control mechanism with enhanced transcriptional expression was long suspected; recently Ballabio and colleagues have shown, by pattern discovery analysis of the promoter region of genes encoding known lysosomal proteins, that in most a common palindromic motif (coordinated lysosomal expression and regulation, CLEAR), which mediates transcriptional activation, is present 40. The CLEAR element binds a single nuclear transcription factor, TFEB, of the MiT/TFE subfamily; the factor was also found to be a target of silencing by a microRNA, miR-128. Further studies in cultured cells showed that TFEB activates transcription of many target lysosomal genes; over-expression of TFEB increased formation of new lysosomes with expansion of the compartment in each cell. Expression studies in stable transfectants showed that TFEB enhanced the degradation of glycosaminoglycans, as well as the breakdown of mutant polyQ-expanded huntingtin polymers in an inducible striatal cell model of Huntington's disease.
During starvation, autophagy and lysosomal protein synthesis are enhanced, with expansion of both cellular compartments. It has now become apparent that TFEB, the master regulator of lysosomal biogenesis, also coordinates the formation of autophagosomes and their fusion with lysosomes (Figure 2). Andrea Ballabio and colleagues have shown that the transcriptional activation of proteins destined for the autophagosome as well as the lysosome was up-regulated by TFEB 41. Under resting conditions, phosphorylation of a critical serine residue (ser 142) in the factor prevents its translocation to the nucleus; this phosphorylation is catalysed by the extracellular signal-regulated kinase 2 (ERK2), which is regulated by signalling pathways sensitive to activation by nutrients in the mitogen-activated protein kinase pathway 42. Hitherto mTOR (mammalian target of rapamycin) has been considered to coordinate cellular responses to nutrients. In nutrient-rich conditions, the kinase is activated and stimulates growth while inhibiting autophagy; reciprocal effects pertain when the supply of nutrients is poor, leading to induction of autophagy. The new findings indicate a control mechanism orchestrated by ERK2–MAPCK, which is distinct from the mammalian target of rapamycin (mTORC1) pathway of transcriptional regulation 43.
TFEB coordinates the functions of two organelle systems in cellular recycling and autophagy. The master regulator of lysosomal biogenesis, TFEB (a nuclear transcription factor), stimulates formation of autophagosomes and their fusion with lysosomes. Transcription of nuclear genes whose products are destined for the autophagosome, as well as the lysosome, is up-regulated by TFEB. Phosphorylation of a critical residue (ser 142) mediated by the extracellular signal-regulated kinase 2 (ERK2) prevents nuclear translocation; this phosphorylation is sensitive to nutrient activation in the mitogen-activated protein kinase pathway. Mammalian target of rapamycin (mTOR) also regulates cellular responses to nutrients. In nutrient-rich conditions, ERK2 is activated and stimulates growth while inhibiting autophagy; reciprocal effects pertain when the supply of nutrients is poor and these induce autophagy
These impressive discoveries, principally based on long-known clinical phenomena, show that lysosomal biogenesis and organelle-specific protein synthesis are genetically programmed and coordinated with induction of key autophagy functions by a common master regulator. Transcriptional activation by the TFEB mechanism provides a network for control of the functions of two distinct organelle systems in recycling and autophagy (Figure 2).
Storage: inconstant relationships and frank poisoning
While biochemical classifications of lysosomal diseases often reveal clinico-pathological features that occur in biochemically related clusters, it is increasingly apparent that each disorder is best considered as a singularity. In this context, with ∼70 known and genetically distinct lysosomal disorders, we have identified specific examples to illustrate what is known of numerous divergent pathways of pathogenesis.
Gaucher's disease and Krabbe's disease
Superficially, these glycosphingolipidoses have similar biochemical effects and several cellular features in common; they are nonetheless informative, since they illustrate how discrepant clinico-pathological manifestations arise from distinct mechanisms of disease. Gaucher's disease is an autosomal recessive condition, due to deficient activity of lysosomal β-glucosylceramidase. The natural substrates for this acid β-glucosidase (D-glucosyl-N-acylsphingosine glucohydrolase; EC 3.2.1.45) are mixtures of N-acyl-sphingosyl-1-O-β-D glucosides with varying acyl (fatty acid) and sphingosine moieties, including those such as β-glucosylsphingosine which are devoid of fatty acids. Krabbe's disease, also an autosomal recessive disorder, is caused by mutations affecting the activitiy of lysosomal β-galactosylceramidase (D-galactosyl-N-acylsphingosine galactohydrolase; EC 3.2.1.46); like its enzymatically related but structurally distinct, member of the β-glycanase family, this acid β-galactosidase has a range of natural substrates, including analogues devoid of fatty acids, such as β-galactosylsphingosine 44.
The massive visceromegaly of Gaucher's disease gives rise to the most familiar and usually most prominent signs of the condition as it appears most frequently in childhood and adolescence—marked enlargement of spleen prematurely destroys the formed elements of the blood with anaemia, susceptibility to infection and a marked bleeding tendency. Infiltration by engorged macrophages (Gaucher's cells) is associated with a prominent fibrotic reaction in the affected tissues. The skeleton is also affected: sheets of abnormal macrophages in the marrow cavity impair the microcirculation and are associated with painful episodes of osteonecrosis; there is also bone demineralization, with the risk of fragility fractures. Gaucher's disease is characterized by a sustained inflammatory reaction, plasma-protein abnormalities, weight loss, fatigue and an increased metabolic rate; polyclonal and monoclonal lymphoproliferation of B cells may lead to myeloma and related cancers. In the nervous system, Parkinsonism is a late complication, but in severely affected younger patients, isolated neuronophagia in isolated nuclei and the cerebellum is accompanied by disordered eye movements, ataxia and other brain-stem neurological signs; cortical disease with myoclonic epilepsy and hearing loss occur in severe cases 45.
The pathologically enlarged and often multinucleate ‘storage’ cell is a striking feature of Gaucher's disease but, while the condition may be associated with a 100-fold increase in splenic mass, the excess β-glucosylceramide in this organ amounts only to < 2% of the additional weight—a stark illustration of the inflammatory consequences of the biochemical defect and the pathological complexity of lysosomal diseases 46. The Gaucher's macrophage has a lysosomal compartment distended by polymerized aggregates of glycolipid (stored) molecules aligned in a head-to-tail conformation (Figure 3A). These cells evoke local tissue reactions and have greatly increased expression of numerous lysosomal acid hydrolases with enhanced plasma activities, eg tartrate-resistant acid phosphatase, β-hexosaminidases, cathepsins and β-galactocerebrosidase. Expression of chitotriosidase increases up to 10 000-fold and has turned out to be a useful clinical biomarker and, as with several other secreted proteins, activity in the plasma is restored to almost normal by macrophage-targeted enzyme therapy (Figure 3B) 47. The Gaucher's cell's is also a direct source of numerous cytokines and chemokines, such as CCL18, which is chemotactic for activated T cells, dendritic cells and possibly homing of other lymphocytes 48, 49.
Targeted therapy to pathological macrophage in Gaucher's disease. (A) Pathological macrophages result from storage of N-acyl-sphingosyl-1-O-β-D glucosides due to deficiency of glucosylceramidase, a lysosomal acid β-glucosidase. (B) Enzymatic targeting to the lysosomal compartment utilizes the mannose lectin/receptor on the macrophage surface
In contrast, the principal clinico-pathological manifestations of Krabbe's disease, due to deficiency of β-galactocerebrosidase, are restricted to the nervous system and characterized by widespread degeneration of cerebral white matter and loss of myelin in peripheral nerves. Typically, onset is in infancy, causing irritability and spasticity; blindness, seizures and dementia supervene rapidly. The alternative common term for this disease is ‘globoid-cell leukodystrophy’, which describes the neuropathology. Extensive demyelination with loss of oligodendroglia and proliferation of astroglia is widespread in the brain and spinal cord; there is widespread axonal degeneration and peripheral nerves are also affected. Like Gaucher's cells, the globoid cells are large multinucleate cells of mononuclear macrophage origin: globoid cells are prominent around capillaries but are found throughout the brain and spinal cord, particularly where demyelination is active 50. Oligodendrocytes, specialized glia responsible for the formation of myelin, express β-galactosylceramide as an abundant cell-surface marker; these cells disappear early from neural tissue and myelination ceases. Although electron microscopy shows that globoid cells contain abnormal inclusions scattered freely in the cytoplasm, with the appearance of galactosylceramide, no storage occurs in single-membrane-bound organelles corresponding to lysosomes.
Unlike Gaucher's disease, the striking feature of Krabbe's disease is the lack of increased β-galactosylceramide content in the brain and the absence of ‘storage’ within single-membrane-bound organelles when the cognate lysosomal hydrolase is lacking; this is particularly noteworthy, given that the glycosphingolipid constitutes > 15% of the dry weight of myelin. The simple glycosphingolipid metabolite, the cationic water-soluble lysolipid, β-D-galactosylsphingosine, is implicated in pathogenesis 51. In Krabbe's disease brain, β-galactosylsphingosine (‘psychosine’) accumulates up to 100-fold greater than normal at micromolar concentrations in affected tissues. The molecule has wide-ranging cellular effects and preferentially accumulates in lipid rafts, in which it has been demonstrated in the brain of patients with Krabbe's disease.
Psychosine is highly cytotoxic and inhibits protein kinase C (PKC), a key signalling molecule 52; in oligodendrocytes and Schwann cells, psychosine reduces PKC-dependent phosphorylation of myelin basic protein. The toxic action of psychosine on oligodendrocytes may be mediated by the generation of the MAP kinase agonists, lysophosphatidylcholine and arachidonic acid by activation of secretory phospholipase A2 53; inhibitors of this phospholipase prevent psychosine-induced apoptosis of cultured oligodendroglia. Additional actions, mediated by cytokines, have been demonstrated in cell cultures obtained from patients with Krabbe's disease and the genetically authentic model in the Twitcher mouse, in which activation of transcriptional factor C/EPB and astrocytes induces the production of several cytokines, including interleukin-β1, TNFα and interleukin 6.
Quasi-pathophysiological concentrations of the psychosines (β-galactosylsphingosine and β-glucosylsphingosine) also inhibit cytokinesis, and thus can induce formation of multinucleated cells 54. Exposure of human U937 monocytes to psychosines in culture induces the formation of multinucleated macrophage-like cells which closely resemble Gaucher's or globoid cells (Figure 3A). This effect, associated with increased intracellular calcium concentrations, appears to be mediated by a member of the G-protein-coupled receptor 1 family (GPR), encoded by the proton-sensing T cell death-associated receptor gene (TDAG8; GPR 65) 55: binding of cognate lysosphingolipids, such as sphingophosphorylcholine, with greater affinity for this orphan GPCR decreases the concentration of cyclic AMP and increases intracellular [Ca2+] but does not inhibit cytokinesis in all cells (Figure 4). Psychosines induce several phenomena, including arrest of cell (but not nuclear) division, as well as intracellular clumping of F-actin polymers with formation of vacuoles reminiscent of endocytic membranes. It thus remains possible that the formation of multinucleated Gaucher's and globoid cells are due to other downstream effects of psychosine–TDAG8 on cell-membrane dynamics. Nonetheless, it is remarkable that a single class of glycosphingolipid molecules induces pathological effects, mediated at least in part by interacting with a specific cell receptor, and which mimic those observed in Gaucher's and Krabbe's diseases.
Lysolipids in the pathogenesis of Gaucher's and Krabbe's diseases. The lysosphingolipids β-glucosyl- and β-galactosylsphingosine (‘psychosine’) exert their effect by increasing cAMP after binding to a member of the G-protein-coupled receptor 1 family (GPR) encoded by the proton-sensing T cell death-associated receptor gene (TDAG8; GPR 65). Psychosines induce several phenomena, including arrest of cell division and intracellular actin clumping
The mechanism by which formation of the psychosines and other bioactive lysoslipids occurs is unclear, although two main biochemical pathways are possible: deacylation of glucosyl- and galactosylceramides by an as-yet unknown enzyme system or by the direct glycoslation of sphingosine (utilizing active sugar–nucleotide intermediates in the Cleland–Kennedy pathway). Recent experiments using cultured human fibroblasts exposed to various inhibitors of glycosphingolipid synthesis and β-glucosylceramidase suggest that both pathways are utilized—deacylation of β-glucosylceramide appears to be in part mediated by lysosomal acid ceramidase, since its metabolism to β-glucosylsphingosine is greatly reduced in cells from patients with Farber disease, in whom acid ceramidase is deficient. Whether this or another enzyme contributes to the formation of β-galactosphingosine, is unknown.
In Fabry disease, due to deficiency of α-galactosidase A, globotriaosylceramide (Gb3) accumulates within lysosomes, principally in epithelial and endothelial cells. The disease principally affects the kidney tubule and glomerulus, the heart, peripheral nerves, skin and cerebral blood vessels. Recent studies have shown markedly increased plasma concentrations of the deacylated congener of globotriaosylceramide, lyso-Gb3 56, in Fabry disease; this is is a poor substrate for α-galactosidase A, but stimulates cell proliferation and fibrotic responses in vitro.
Other mechanisms of disease
Pathogenesis of neurodenerative diseases
The glycosphingolipidoses and related lysosomal diseases with neurodegenerative features have been modelled in either naturally occurring or transgenic animals. Such models are invaluable for experimental studies and especially therapeutic research.
Faulty gene function in the living brain of mice with GM2 gangliosidosis can be restored using recombinant adeno-associated viral vectors (rAAV) to supply the missing function of the murine HexA and HexB genes encoding β-hexosaminidases A and B: gene therapy given at 4 weeks of age rescues mice that invariably die with paralysis before 18 weeks. Animals treated only once with several injections at 4 weeks, before disease becomes manifest, have a normal or near-normal lifespan (2 years), with preservation of mental function and mobility; normal gene activity is restored in the brain and spinal cord and the associated inflammation of the brain is corrected 57.
Threshold effects revealed by therapeutic intervention
To explore pathological progression, gene therapy was administered bilaterally in the striatum and cerebellum of Sandhoff disease mice at 4, 8 and 12 weeks (see Figure 1). Untreated mutant mice survive to a predefined humane end-point at 121 ± 6 days. Mice injected at 4 weeks reached this end-point at 286 days. In those injected at 8 and 12 weeks of age, survival was only 170 and 117 days, respectively. Unexpectedly, brain storage and neuroinflammation were almost completely cleared when gene transfer was delayed (unpublished data). Our findings mirror the results of similar experiments carried out with delayed gene therapy in a murine neuronal ceroid lipofuscinosis model (CNL2) and are significant for mechanistic understanding of the evolution of GM2 gangliosidoses 58. Dissociation of GM2 storage and neuroinflammation from advanced stages of disease provides a new insight on the disease process: mice continue to die from the effects of GM2 gangliosidosis, even though GM2 storage has been ameliorated. Thus, there are irretrievable aspects to the pathogenesis and defined windows for intervention. Elucidation of the molecular pathways implicated in the irreversible pathology will be highly informative about pathogenesis.
Cytopathological abnormalities in the brain
Taking the example of GM2 gangliosidosis, there is strong evidence of injury to vulnerable neurons: neuroaxonal dystrophy, with focal expansion of arborized regions of the axon to form spheroids, is prominent. Axonal spheroids harbour numerous membranous vesicles and tubular structures with multivesicular bodies and distended autophagosomes; the structural integrity of the cell is disrupted and it seems likely that such gross abnormalities will reflect profound disturbances in the propagation of action potentials, neuronal activity and connectivity 37. However, at the time of writing, the molecular pathways leading to these catastrophic cellular changes are unclear. It seems probable that the master regulator of autophagy and lysosomal biogenesis, TFEB 40 (Figure 2), will be implicated in the process, which in effect represents a failure of cellular compensation.
That ganglioside accumulation cannot account for all the injury to the nervous system is shown by the improved survival of GM2 gangliosidosis mice transplanted with bone marrow transplantation from wild-type animals; this has very little effect on the deficient enzyme activity or ganglioside storage 59. One group has suggested that an autoimmune reaction against the ganglioside may contribute to the neurological disease, and others have described autoantibodies (to glutamic acid carboxylase) in juvenile neuronal ceroid lipofuscinosis (CLN 3) mice 60-62. Abnormalities of the thymus have been described in several models of GM2 gangliosidosis and impaired macrophage degradation of normally apoptotic thymic lymphocytes leads to enhanced secretion of the CXCL13 thymic chemokine with dysregulation of autoimmunity 63-65.
Gangliosides and other active sphingolipids, such as sphingosphine-1-phosphate and ceramide derivatives, are highly bioactive and strongly implicated in many diseases 66. Gangliosides localized to lipid rafts decorate many key cellular antigens and serve as receptors 67. Although it is not difficult to impute changes in key receptor functions and glycolipid signalling molecules in the ultimate pathogenesis of lysosomal diseases, particularly those in which membrane glycosphingolipid turnover is impaired as a result of hydrolytic defects, quantification of the lipid intermediates in the pathological state at various stages in the evolution of the disease has yet to be done. Comprehensive quantitative analysis of cellular bioactive lipids and sphingolipids, including gangliosides, is a highly specialized analytical field 68: so far, the systematic application of contemporary mass spectrometry methods to the study of lysosomal diseases, in which sphingolipid signalling pathways are strongly implicated, is in its infancy.
Other pathways of cell injury and inflammation
Disturbed calcium signalling
Derangements in intracellular calcium signalling have been discovered in several lysosomal diseases, but distinct abnormalities occur in individual diseases and identification of a common pathological mechanism is elusive 69. In GM2 gangliosidosis, increased cytosolic Ca2+ concentrations are accompanied by decreased Ca2+ uptake by the sarco-endoplasmic reticulum–Ca2+ ATPase (SERCA)—an effect appears attributed to the excess sialyl moieties of the ganglioside 70. Similar changes in SERCA activity have been reported in Niemann–Pick A disease but the sialyl moieties are not implicated 71. In murine models of neuronopathic Gaucher's disease, enhanced Ca2+ release from endoplasmic reticulum has been shown; this has been attributed to be due to dysregulation of the ryanodine receptor by a direct action of the excess glucosylceramide in this disorder 72, 73. The lysosphinglipid, glucosylsphingosine, appears not to affect the receptor at low concentration but stimulates egress of Ca2+ by other mechanisms and can act as a ryanodine agonist 74. In all, the contributions of homeostatic changes in Ca2+ to cellular pathogenesis in these diseases do not seem to have a unifying mechanism.
High internal concentrations of calcium in the lysosome are regulated by the action of nicotinic acid dinucleotide phosphate (NAADP) on membrane two-pore channels 75. In the neurodegenerative lysosomal disorder Niemann–Pick disease type C (NPC), mutations in the cognate lysosomal membrane protein cause severe cerebellar disease and dementia; there is impaired endosomal-lysosomal trafficking of lipid cargoes accompanied by accumulation of cholesterol, sphingosine and. glycospingolipids 76. Defective lysosomal uptake of calcium and impaired Ca2+ egress in response to NAADP was shown in NPC 1 cells—changes attributable to lysosomal accumulation of sphingosine base; pharmacological correction of the calcium abnormality with curcumin, a polyphenol anti-inflammatory and calcium-modulating agent, restored the endosomal trafficking defect and pathological storage—with improved survival in NPC mice 77. Since curcumin is a food additive, these findings may have clinical significance.
Abnormalities of iron metabolism
The lysosome plays a role in the retrieval of storage iron from ferritin and haemosiderin, a proteolytic and polymerized ferritin iron–protein complex. When loaded with iron, the organelle has long been known as the ‘siderosome’. Mice with GM1 and GM2 gangliosidosis are characterized by progressive depletion of iron in brain tissue. Key regulators of systemic iron homeostasis, hepatic peptide hormone hepcidin and the inflammatory cytokine IL-6, were increased in the serum when compared with age-matched control animals. Transferrin concentrations were reduced, reflecting a progressive inability of the brain to acquire iron from the circulation, whereas reciprocal expression of the transferrin receptor was up-regulated. The iron exporter ferroportin was up-regulated in the brain. Administration of iron prolonged survival in the mutant mice by about one-third and also delayed the onset of disease 78. Given the therapeutic effect of iron observed in coherent genetic models of the gangliosidoses in animals, there is a strong case for exploring these findings as far as possible in human patients.
Integrated perspectives: lysosomes in common disorders
The central position of the lysosome in cell biology and in all aspects of growth, development and survival of eukaryotic organisms and Metazoa, confers upon it the need for comprehensive understanding. While the inborn errors of lysosomal function have proved to be highly informative, we should not forget that the exploitation of receptor-mediated pathways for delivery of therapeutic proteins has generated immense revenues and lifted the profits of orphan Biotech companies to sustained blockbuster status. Finally, the legion of other diseases in which lysosomal function is critical to pathogenesis should not be forgotten.
The unfolded protein (UPR) and the heat-shock responses serve to protect cells from the toxic effects of misfolded protein aggregates; activation of these systems may prematurely deplete cells of protein or simply fail to compensate for misfolding that overwhelms quality control mechanisms in the endoplasmic reticulum. The UPR has been reported to be activated in GM1 gangliosidosis 79 and in palmitoyl–protein thioesterase-1 deficiency and in fibroblasts from patients with neuronopathic Gaucher's disease. However, it is not surprising that activation of intracellular chaperones and the CHOP ER stress-response transcription factor was not observed in a murine model of neuronopathic Gaucher's disease, in which glucosylceramidase was effectively absent. This suggests that the UPR is not activated by the disturbance of glycosphingolipid metabolism 80.
There is a direct role for the lysosome in the breakdown of intracytoplasmic proteins, whether or not they form aggregates, and the lysosome is implicated in several major neurological diseases. These include: Parkinson's disease (α-synuclein in Lewy bodies); Huntington's disease and related triplet-repeat diseases (polyyglutamime residues in aggregates of huntingtin); aggregated tau in the tauopathies; the abnormal aggregated isoforms of prion proteins that accumulate in the synaptic structures in patients with Creutzfeldt–Jakob disease; Bunina bodies and ubiquitinated skein and/or spherical inclusions in amyotrophic lateral sclerosis—and in the abundant formation of axonal spheroids that characterize the autophagic arrest 81. Recently it has been found that activation of the unfolded response enhances autophagy with increased molecular display of LC3 in the cerebral neurons of patients with Alzheimer's disease 82.
Parkinsonism occurs with a higher than expected frequency in patients with Gaucher's disease and in carriers (heterozygotes) otherwise unaffected; moreover, patients with Parkinson's disease in many populations have an increased frequency of mutations in the glucosylceramidase gene (GBA1) and Lewy bodies harbouring the enzyme protein are found in their brains 83.The finding that heterozygous mutations in the Gaucher's disease-related GBA1 gene, which encodes lysosomal β-glucosylceramidase, are the most frequent genetic determinants of Parkinson's disease, presents a tantalizing challenge for pathogenesis 83. Recent studies have revealed a reciprocal relationship between deficiency of glucosylceramidase and accumulation of α-synuclein in lysosomes which further inhibits the enzyme; a positive feedback mechanism has been proposed. 84. A role for lysosomal proteins in autoimmunity is apparent from the recent report of genome-wide association studies in a very large international cohort of patients with multiple sclerosis and controls: a region of the human genome to which only three genes, including Galc, localize, shows the greatest odds ratio for association with multiple sclerosis 85. Since Galc is essential for the maintenance of normal myelin integrity, the molecular nature of the predisposition and strong association with one Galc allele (odds ratio ∼1.22) is of critical importance in evaluating this common neurodegenerative disorder. Deficiency of the lysosomal enzyme, tartrate-resistant acid phosphatase Acp5, has recently been found to cause diverse autoimmune phenomena with lupus and and spodyloendochondroplasia 86. The phosphatase is expressed in osteolasts, macrophages and dendritic cells and dephosphorylates the skeletal protein osteopontin, also known as Eta-1-a 87. This Acp5 has, as postulated, a common role in osteoclestic function as well as in the regulation of innate immunity 88.
Conclusion
Cell biology of the lysosome has a large debt to medicine on account of the rich discoveries that emerge from the study of rare single-gene disorders. With almost universal applicability to disturbances of the organelle in numerous acquired human diseases, sustained exploration of this field holds the promise of therapeutic discovery—and scientific repayment on an unprecedented scale.
Author contributions
Both authors contributed to the concept, content, writing and revision of this article; MBCG conducted the experiment depicted in Figure 1.
Acknowledgements
The authors' research is currently supported by the US National Institutes of Health (UDI award); Cambridge in America; the National Tay–Sachs and Allied Diseases Association, and Sparks, the children's medical charity. We gratefully acknowledge support from the European Union, 7th Framework Programme ‘Euclyd—a European Consortium for Lysosomal Storage Diseases’ Health F2/2008 Grant Agreement No. 201678 (EP); and the Metabolic theme of the Cambridge Biomedical Research Centre of the National Institute of Health Research.
