1. Introduction
The term neuronal ceroid lipofuscinoses (NCL) refers to a group of autosomal recessive neurodegenerative disorders, presenting with myoclonic epilepsy, psychomotor delay, progressive loss of vision, and early death
[1][2][3][4]. It is worth noting that NCL are the most common cause of childhood dementia
[3][5][6]. Cardinal to NCL pathology is the toxic levels of protein aggregates in the central nervous system (CNS)
[4][5][7], which, more specifically, are aggregates of lipopigments (lipofuscin) within lysosomes
[4][8][9]. They are the most prevalent neurodegenerative disorders of childhood, with an incidence in the USA estimated at 1.6–2.4/100,000
[2][4][10][11][12], while, in Scandinavian countries, the incidence varies between: 2–2.5/100,000 in Denmark
[2][10][11], 2.2/100,000 in Sweden
[2][10][11], 3.9/100,000 in Norway
[2][10][11], 4.8/100,000 in Finland
[2][10][12], and 7/100,000 in Iceland
[2][10]. The rarity of incidence and scarcity of the disease models have limited comprehensive understanding of the pathological factors that lead to disease progression. To date, 13 autosomal recessive, as well as one autosomal dominant gene, variants of NCL have been described, each involving functional defects in lysosomal protein
[3][4][13][14][15][16]. Aggregated lysosomal lipofuscin affects the neuronal cytoskeleton and cellular trafficking, resulting in neuronal loss and pathological glial proliferation and activation
[4][7]. More specifically, defective proteins (corresponding genes designated
CLN for “ceroid lipofuscinosis, neuronal”) include secreted lysosomal proteins (CLN1/PPT1, CLN2/TPP1, CLN5, CLN10/CTSD, CLN13/CTSF, and CLN11/GRN) or membrane proteins (CLN3, CLN7/MFSD8, and CLN12/ATP13A2)
[4][17][18][19][20]. Two of them are the membrane proteins of the endoplasmic reticulum (CLN6 and CLN8). Other NCL proteins are cytoplasmic (CLN4/DNAJC5 and CLN14/KCTD7), which peripherally associate with cell membranes
[4][17][18][19][20].
NCLs are grouped on pathological grounds, due to the common presence of the neuronal and extra-neuronal accumulations of autofluorescent pigments, despite the different underlying biochemical etiologies
[2][3][21]. In all forms of NCL, lipopigment storage material accumulates in macrophages, neurons, and some somatic tissues, including vascular endothelial cells and smooth muscle cells
[3][22]. Neuronal loss is profound and extensive in most patients with NCL, leading to cortical gray matter atrophy, cerebellar atrophy, and secondary ventricular enlargement, all of which progress throughout the course of the disease
[3][22][23]. The degree of atrophy and ventricular enlargement varies with the form of NCL
[3][22][23]. The prominent activation of microglia and astrocytes precedes (and perhaps causes) neuronal loss, accurately predicting its distribution
[3][24][25][26].
NCL is classified into five primary types, as shown in Table 1.
The most common NCL types are the CLN1 (also known as “Infantile NCL”), CLN2 (“Late Infantile”), CLN3 (“Juvenile NCL”), and CLN6 (“variant LINCL”) diseases
[4][17][18][19].
Table 2 shows all known types of neuronal ceroid lipofuscinoses, together with the genes whose mutations are responsible for their occurrence.
Table 2 also contains the age of onset of the disease and the proteins whose functions are impaired.
Table 2. NCL diseases and encoded genes
[19][21][31][32].
In 1998, a database was established that includes published mutations and sequence changes in genes causing NCL, together with unpublished data, included with permission. All mutation details can be found in the publicly available NCL mutation database at the following Internet address:
www.ucl.ac.uk/ncl-disease (accessed on 16 May 2022).
2. Genetics and Pathophysiology
2.1. Animal Models in NCL Pathology
The development of various animal models has provided scientists with a range of tools for studying the effects of the mutations responsible for NCL. In NCL, as in many other neurodegenerative diseases, there is usually a subset of tissues or cell populations that are selectively susceptible to pathogenic agents
[5][33][34][35].
Animal models (especially the widely used mouse models of NCL) and their characteristics allow researchers to develop disease phenotypes at the organismal level, as they largely reproduce certain disease-relevant phenotypes, with obvious species limitations
[5][34][35][36][37][38][39][40][41][42][43][44]. With the advent of more accurate and precise technologies to generate mouse models with specific mutations, attempts have been made to replicate the most common human disease-causing mutations in mice, which would allow for a better representation and understanding of the pathological changes observed in human NCL
[5][45][46][47][48]. Naturally, in addition to mouse models, larger animal models of NCL have been developed in dogs
[5][49][50][51][52] and sheep
[5][53]. These models have proved very useful for understanding the anatomical and pathophysiological spread of disease, as they more closely mimic the human disease and provide a means to test the delivery and dosing of therapeutic agents in a way that is not possible in mice
[5].
The use of CRISPR-Cas9 has enabled the generation of a detailed and genetically comprehensive sheep model of CLN1 disease
[5][54], as well as a porcine model of CLN3 disease
[5][55]. In all likelihood, researchers expect similar models to be developed for other forms of NCL
[5].
2.2. Anatomical Regions Affected by NCL Pathology
A constant characteristic sign in NCL is cerebral and cerebellar atrophy co-occurring with enlargement of the lateral ventricles in the brain
[5]. It should be noted, however, that atrophy is not a uniform process; thus, some regions are affected much earlier than others
[5][23][56][57][58][59]. Studies in mouse models have shown that the thalamus and cerebellum are particularly vulnerable regions in various forms of NCL, and somatosensory regions of the cortex are affected earlier and more severely than motor regions
[5][42][60][61][62][63]. Indicative changes include marked glial activation, accumulation of autofluorescent storage material (AFSM), and neuronal loss, as well as marked loss of interneurons in the cortex and hippocampus. The stage and progression of pathology has also been determined in larger animal models of NCL, confirming that these regions may, indeed, be clinically relevant
[5][51][52][64][65][66]. Patients with NCL are often characterised by sensory and motor deficits
[5][13]. Spinal cord involvement, as demonstrated in CLN1, may be responsible for this
[67]. A study in animal models, as well as autopsy studies in humans, suggest that spinal cord pathology occurs with high frequency in many types of NCL
[5][22][51][54][55][68].
Most patients with NCLs suffer from progressive loss of vision
[3][5][13]. This has necessitated a detailed examination of the visual pathway, which has shown marked retinal degeneration in these patients
[13][69][70]. In many forms of NCL, based on animal models, it is also often accompanied by optic nerve degeneration
[5][71][72][73][74][75][76][77]. Furthermore, mouse models have been shown to exhibit significant pathology in central visual pathways within the dorsolateral geniculate nucleus of the thalamus and visual cortex in many forms of NCL
[42][60][61][62], and it is likely that other retinal receptive nuclei are also affected. Another issue with mouse models is that rodent species rely much more on sensory information from the eyeballs than on visual information
[78], which may explain why the somatosensory pathways through the posterior ventral thalamic nucleus to the barrel-field cortex are so severely impaired in the mouse models of NCL
[61][62][63][79][80], even more so than the central visual pathways
[42][60][61][62].
In NCLs, as in many other known storage diseases, findings provide evidence of significant cardiac pathology from both clinical observations and experimental studies
[5][81][82][83]. These are best understood in CLN3 disease and also include evidence of autonomic nervous system dysfunction
[84]. Furthermore, research reports also refer to the intestinal problems occurring in children with NCL
[5][13].
Since most of the proteins in the NCL are widely expressed in different tissues and cell types, in all likelihood, researchers suspect that the deficiency of these proteins also affects other organ systems in the body (including outside the CNS)
[19][85][86]. Therefore, treating only the regions in the brain occupied by the disease process may not be sufficient to achieve therapeutic success. Making the identification of other, unexpected sites where pathology develops seem important. While the involvement of the brain by the disease process appears to be the primary concern, other disease effects, associated with other locations of disease development (although not to the same extent as the brain), will potentially also require therapeutic intervention. Treating these previously overlooked pathologies may provide additional clinical benefits, improving quality of life
[5].
2.3. Cell Types in NCL Pathology
NCL affects different cell types
[87][88][89][90]. Dysfunction and loss of the number of properly functioning neurons are characteristic features, bearing in mind that neuronal population types differ in their sensitivity to being affected
[5][88]. One type of neuron that is more susceptible to damage in NCL is the interneuron population. The greatest loss of this population occurs in the thalamus, cerebral cortex, and hippocampus, where they account for the majority of neurons lost in the early stages of the disease
[60][61][79].
Their greater vulnerability to damage may be due to their properties, mainly electrophysiological and bioenergetic
[91]. What is clear in NCLs is that AFSMs are deposited in lysosomes, and their function is impaired; as it seems, the preserved normal lysosomal function is critical for the proper functioning of interneurons
[47][92]. Interestingly, interneuron populations have also been shown to play an important role in the neurodegenerative pathways of various diseases, such as epilepsy
[93], Alzheimer’s
[94], amyotrophic lateral sclerosis (ALS)
[95], and Parkinson’s
[96].
Despite the fact that AFSMs accumulate in many cell types, NCLs have always been considered neuronal diseases, which is also reflected in their name, i.e., “neuronal ceroid lipofuscinoses”
[5][83]. Nowadays, an early and significant effect of NCLs on other cells, which are also part of the CNS, is increasingly pointed out. These include astrocytes and microglia, which are observed in a state of activation, both biochemically and histologically
[5][48][60][61][97]. Microglia activation generally occurs at a stage immediately preceding neuronal loss, and its location has a greater predictive value, in relation to the site of neuronal loss, than the site of storage material appearance
[3][98]. Certainly, the question of whether glial activation contributes to neuronal population loss (which is a suggested pathophysiological pathway in other lysosomal or neurodegenerative disorders) still needs to be clarified
[5][99][100][101]. Primary cell culture experiments investigating the properties of glial cells derived from mouse models of CLN1 and CLN3 have shown that astrocytes and microglia have significant functional and morphological defects, although they vary widely between these forms of NCL
[5][25][26]. In co-culture systems, astrocytes and microglia have been shown to damage or negatively affect the survival of both healthy and mutant neurons
[5][25]. These experiences highlight the role that glial cells appear to play in the pathogenesis of NCL. In addition to the innate immune changes evidenced by glial activation in the CNS, there is also evidence for a general humoral immune response in NCL
[102][103], with a possible autoimmune component, particularly in CLN3
[5][104]. Researchers should discuss the results and how they can be interpreted, from the perspective of previous studies and the working hypotheses. The findings and their implications should be discussed in the broadest context possible. Future research directions may also be highlighted.
2.4. Cellular Model
Mutations in the NCL involve the endo-lysosomal system. Lysosomes are responsible for recycling endogenous and internalized molecules, which are involved in maintaining tissue homeostasis
[105][106]. They are also known to be involved in many cellular processes, such as nutrient sensing, calcium and bio-metal homeostasis, cell growth axonal transport, and synaptic homeostasis
[105][106]. Because of this, autophagy function, which is essential for normal cellular homeostasis, is impaired in the NCL. Despite the disclosure of these disorders in various forms of NCL, the exact mechanism that leads to the consequent changes that cause cell death still remains unknown
[47][80][107][108][109]. Similar disturbances in the autophagy mechanism have been revealed in other lysosomal disorders
[110][111].
Animal models have also shown changes in synaptic vesicle density, exocytosis, and electrophysiological changes, which are indicative of synaptic dysfunction
[62][65][108][112][113][114]. Despite the evidence that lysosomes are transported into the synapse to facilitate synaptic remodelling, pre-synaptic autophagy, synaptic vesicle sorting, and overall synaptic homeostasis
[5][115][116][117], the exact mechanisms causing NCL to affect the lysosomal system, leading to synaptic dysfunction, are still not understood
[5][115][116][117]. However, the question of synaptic dysfunction remains essential for a comprehensive understanding of the entire pathomechanism, mainly due to the fact that it often precedes the stage of neuronal loss. It is highly likely that this is due to altered conductivity within the neurons, as well as the presence of other signalling defects within cells
[5][118]. Disease involvement of the (already mentioned) astrocytes and microglia is also one of the postulated mechanisms leading to loss of synaptic function
[5][118][119][120][121].
The discovery of the cellular mechanisms responsible for controlling the expression of many endo-lysosomal proteins is undoubtedly one of the major achievements of molecular biology. These include the discovery of the coordinated lysosomal expression and regulation (CLEAR) transcription initiation site, the EB transcription factor (TFEB), and its phosphorylation by mTORC1. Of particular relevance to the pathophysiology of NCLs was the demonstration of dysregulation of this pathway in many NCL types
[5][122][123][124][125][126]. Preclinical studies in a mouse model provided evidence that the aforementioned dysregulation may be a viable therapeutic target
[127][128], mainly because an upregulation of TFEB increases the overall cellular clearance, which may bypass the existing lysosomal dysfunction
[127][128].
2.5. Genotype–Phenotype Correlations
All of the groups of neuronal ceroid lipofuscinoses are monogenic disorders, so each represents a distinct disease entity. Genes whose mutations are responsible for the occurrence of NCL encode seemingly unrelated proteins, including the soluble lysosomal enzymes and membrane proteins located in various organelles, including the lysosome
[3][19][129][130]. All types of NCL are inherited autosomal recessively, except neuronal ceroid lipofuscinosis type 4 (CLN4)
[3][129]. For most NCLs, there is a recognizable classical disease phenotype associated with the complete loss of gene function, due to intracellular mislocalization or degradation of the mutant protein
[129]. In addition, there are forms of the disease that may have a more chronic course, and some of the predicted classical phenotypes may not be present, as a result of milder mutations that do not cause complete arrest of protein function
[3][24]. There are examples of mutations associated with a specific phenotype, such as a missense mutation in CLN8 or a 1 kb intragenic deletion that underlies the most common form of NCL, i.e., juvenile CLN3 disease
[3][24][129]. The most common mutations are a 1 kb deletion in CLN3 and two mutations in CLN2
[3][23][24].