Skip to main content
NCBI home page
iScience logoLink to iScience
. 2020 Dec 31;24(2):102020. doi: 10.1016/j.isci.2020.102020

Urine proteomics analysis of patients with neuronal ceroid lipofuscinoses

Katharina Iwan 1, Robert Clayton 1, Philippa Mills 1,2, Barbara Csanyi 3, Paul Gissen 1,2,3, Sara E Mole 1,4, David N Palmer 5, Kevin Mills 1,2, Wendy E Heywood 1,2,6,∗
PMCID: PMC7822952  PMID: 33532713

Summary

The neuronal ceroid lipofuscinoses (NCL) are a group of 13 rare neurodegenerative disorders characterized by accumulation of cellular storage bodies. There are few therapeutic options, and existing tests do not monitor disease progression and treatment response. However, urine biomarkers could address this need. Proteomic analysis of CLN2 patient urine revealed activation of immune response pathways and pathways associated with the unfolded protein response. Analysis of CLN5 and CLN6 sheep model urine showed subtle changes. To confirm and investigate the relevance of candidate biomarkers a targeted LC-MS/MS proteomic assay was created. We applied this assay to additional CLN2 samples as well as other patients with NCL (CLN1, CLN3, CLN5, CLN6, and CLN7) and demonstrated that hexosaminidase-A, aspartate aminotransferase-1, and LAMP1 are increased in NCL samples and betaine-homocysteine S-methyltransferase-1 was specifically increased in patients with CLN2. These proteins could be used to monitor the effectiveness of future therapies aimed at treating systemic NCL disease.

Subject areas: Disease, Systems Biology, Proteomics

Graphical Abstract

graphic file with name fx1.jpg

Open in a new tab

Highlights

  • •

    The urine proteome is altered in humans and animals with NCL

  • •

    Hexosaminidase A and LAMP1 are increased in patients with NCL

  • •

    Betaine-homocysteine S-methyltransferase 1 is elevated in CLN2 patients

  • •

    Proteins altered in CLN5 and CLN6 sheep models are not affected in humans

Disease; Systems Biology; Proteomics

Introduction

The neuronal ceroid lipofuscinoses (NCLs, Batten disease) are a heterogeneous group of rare neurodegenerative diseases that present mainly in childhood and as a combined group represent the most common cause of childhood neurodegeneration and dementia. The group share phenotypic characteristics including seizures, visual loss, loss of motor skills, and cognitive function resulting in premature death, but each subtype is a distinct genetic entity with unique pathophysiological characteristics (Kohlschütter et al., 2019). Brain pathology reveals a massive loss of neurons, and most cells throughout the body contain lysosome-derived fluorescent storage bodies, the main component of which is the c subunit of mitochondrial ATP synthase or Saposins A and D (Palmer, 2015). So far, at least 13 genes have been identified that lead to NCL (CLN1–8, 10–14) (Butz et al., 2020; Williams and Mole, 2012). Of all the NCLs, CLN2 disease is one of the most commonly reported with an incidence in Europe varying between 0.15 and 0.78 per 100,000 live births. Autosomal recessive mutations in the TPP1 gene lead to loss or deficiency of the lysosomal enzyme tripeptidyl peptidase 1 (TPP1). TPP1 is a lysosomal serine protease that cleaves tripeptides from the N termini of small polypeptides during protein degradation inside the lysosomes (Golabek et al., 2003; Wlodawer et al., 2003). Five other NCLs occur due to a lack or defect of lysosomal-associated proteins, namely, palmitoyl protein thioesterase 1 (PPT1) in CLN1, CLN5 in CLN5, cathepsin D (CTSD) in CLN10, progranulin in CLN11, and cathepsin F (CTSF) in CLN13 diseases. Other NCL forms resulting from missing or defective lysosomal membrane proteins are CLN3, CLN7/MFSD8, and CLN12/ATP13A2 diseases. CLN6 and CLN8 diseases result from deficiencies of endoplasmic reticulum (ER) membrane proteins CLN6 and CLN8 (di Ronza et al., 2018; Mole et al., 2004). The very rare CLN4 disease results from defects in a chaperone protein DNAJC5 (Henderson et al., 2016), and CLN14 disease results from defects in the plasma membrane potassium channel KCTD7 gene (Staropoli et al., 2012).

To date, the only clinically approved treatment that has slowed disease progression is Brineura (cerliponase alfa) for CLN2 disease. This TPP1 enzyme replacement therapy (ERT) is delivered intra-cerebroventricularly into the brain ventricles of patients every fortnight (Schulz et al., 2018). Other treatment approaches in development for NCL include gene therapy, various small molecules, and the administration of immunosuppressive drugs to prevent neuroinflammation (Kohlschütter et al., 2019; Mole et al., 2019). Initial animal studies have shown promising results for an ERT with the recombinant proenzyme PPT1 for patients with CLN1 disease (Lu et al., 2015; Hu et al., 2012). ERT administered directly into the brain still cannot prevent loss of vision, hence additional intraocular/intravitreal treatment is necessary (Katz et al., 2014; Tracy et al., 2016; Whiting et al., 2014; Griffey et al., 2005; Mole et al., 2019). Clinical trials for gene therapy for patients with CLN2, CLN3, and CLN6 disease have begun (Liu et al., 2020; Kohlschütter et al., 2019), and promising data for an early clinical trial for CLN2 disease suggest an attenuation of neurological impairment. So far, preclinical and clinical studies indicate that a single treatment strategy will not suffice, as therapeutic strategies have to target the brain, the retina, and peripheral organs (Kohlschütter et al., 2019; Mole et al., 2019).

Clinical diagnoses of specific NCL forms are still challenging with investigations mainly only being conducted after NCL disease is suspected after the first symptoms are detected. Early or new-born diagnosis is crucial if a small therapeutic window exists where successful interventions can more effectively prevent or slow the neurological disease progression. Diagnosis of NCL used to require microscopic analysis of storage material using a skin biopsy. Nowadays, genetic testing and enzyme activity assays are the methods of choice. Although genetic tests can be used to diagnose the NCL form, they cannot monitor disease progression or response to treatments. Typically disease progression is determined using baseline clinical scores like the Unified Batten Disease Rating Scale (UBDRS) (de Blieck et al., 2013; Adams and Mink, 2013), the Weill Cornell CNS scale, and the Hamburg CLN2 scale to quantify seizures, loss of language, motor skills, and visual function. To accompany the clinical presentation, assessing what is happening at the molecular level will be crucial in monitoring early stages of disease and for preclinical studies or ongoing clinical trials. Robust molecular biomarkers are therefore needed to monitor even subtle changes in physiological or pathological disease progression.

Various biomarker studies in patients with different NCLs have previously been conducted to identify possible biomarkers of disease progression. Using proteomics, Sleat et al. (2017) investigated cerebrospinal fluid (CSF) and postmortem brain samples of patients with CLN1, CLN2, and CLN3 disease. Initially they investigated lysosomal proteins, i.e., TPP1, beta-hexosaminidase subunit alpha (HEXA), and cathepsins, as they are known to be altered in many lysosomal storage disorders (Sleat et al., 2017). Other studies analyzed CSF from patients with CLN2 disease and identified metabolites that correlated to disease severity scores (Sindelar et al., 2018). Urine is a highly desirable body fluid for biomarker analysis because collection is non-invasive, feasible in relatively large volumes, and at frequent intervals; because it is inexpensive to preserve; and, due to its high urea content, urine proteins are considered more stable than serum or plasma proteins. The urinary proteins present derive from glomerular filtrations of plasma, renal tubule excretion, and urogenital tract secretion, with fluctuations representing normal physiological changes as well as pathological changes due to renal or systemic disease (Harpole et al., 2016; Kalantari et al., 2015).

One of the greatest challenges in biomarker research in rare diseases is accessing enough samples for a viable study. Often researchers turn to animal or cell models to address this. One such model is the slime mold Dictyostelium, which has revealed how secretion is impaired in NCL (Huber et al., 2014; Huber, 2020). Proteins observed to be altered in Dictyostelium were also observed to be altered in human patient tissues (Sleat et al., 2017).

Experimental design

We have used both human and animal models in a two-phase proteomic strategy to identify proteins altered in patients with NCL.

Phase 1 the “discovery” step is an unbiased eyes-open approach that uses label-free proteomic profiling to study urine from patients with CLN2 disease and CLN5 and CLN6 disease sheep models. The technology is semi-quantitative and lengthy, therefore small sample cohorts are analyzed using this approach.

Phase 2 involves a targeted analysis using multiple reaction monitoring mass spectrometry coupled with liquid chromatography (LC-MS). A candidate biomarker peptide(s) is accurately quantitated using this technique, and multiple peptides can be measured in one analysis, therefore creating a multiplex approach. Proteins of interest identified from the phase 1 profiling analysis are considered as potential biomarker candidates for corresponding CLN2, CLN5, and CLN6 diseases in human patients. As additional biomarkers can be included we also selected peptides from other biomarker candidates described in previous NCL omics studies (Sleat et al., 2017, 2019) to confirm these in our analyses. Using the targeted multiplex assay we applied this to a larger separate cohort of CLN2 disease patient samples and patient samples with CLN5 and CLN6 disease to confirm phase 1 findings. To see if the biomarkers may be specific for the NCL type we also included analyses of the NCL diseases CLN1, CLN3, and CLN7.

Results

Phase 1: Proteomic screening of urine from human and animal models of NCL disease identify candidate biomarkers and affected pathways

Untargeted proteomic analyses of two patients with CLN2 disease and comparison with four age-matched healthy control samples demonstrated an increase in 61 proteins (p value < 0.05 with a normalized fold change >2) and 11 proteins with reduced levels (p value < 0.05 and normalized fold change < -2, Table S1). Additionally, proteomic analyses of urine from 3 CLN5-affected sheep and 13 CLN6-affected sheep were compared with seven normal control sheep (Tables S2 and S3). The volcano plot in Figure 1A summarizes the alterations found in the urinary proteome of patients with CLN2 disease and the CLN5 and CLN6 sheep models. The CLN5 profile shows the least number of significant proteins altered among the three NCL urine profiles.

Figure 1.

Figure 1

Open in a new tab

Label-free proteomic analysis of urine from CLN2 patients and CLN5 and CLN 6 sheep models

(A) Volcano plots summarizing the proteomic data. Proteins highlighted in red are significanlty upregulated, proteins highlighted in blue are downregulated, whereas the proteins in gray are not significantly changed (p (ANOVA) < 0.05).

(B) Summary of the affected pathways in CLN2 patient urine as determined by Ingenuity Pathway Analysis (Qiagen). This shows which cellular pathways are altered significantly in urine of patients with CLN2 disease compared with healthy controls. Only proteins that were significantly (p (ANOVA) < 0.05) changed with a fold change >2 or < −2 were used for this analysis. The y axis depicts how significantly changed a pathway is in a negative logarithmic scale of the p values. The cutoff is 1.3 as this is the equivalent to a p value of 0.05. The activity pattern is color coded: bars in gray indicate it is unknown if the pathway is activated or inactivated, bars in blue label negative z values and therefore inhibited pathways, whereas positive z values and therefore activated pathways are highlighted in orange.

To gain some insight into the effects of CLN2 disease on biological pathways an Ingenuity Pathway Analysis (IPA, QIAGEN Inc., https://www.qiagenbioinformatics.com/products/ingenuitypathway-analysis) was conducted on the differentially expressed proteins (Figure 1B). Pathways showing no activity pattern (z = 0) were omitted. The most significantly activated pathway is the BAG2 signaling pathway, with altered levels of proteins Annexin A2, Heat shock 70 kDa protein 1A/B, and Heat shock 70 kDa proteins 2 and 8. Interestingly, the neuroinflammation signaling pathway and the synaptogenesis signaling pathway are both significantly activated, whereas the natural killer cell signaling pathway is inhibited. A functional enrichment gene ontology analysis for cellular components (CC) performed using the DAVID bioinformatics website (Huang et al., 2008, 2009) confirmed that the differentially expressed proteins detected in the urine proteome were derived from the lysosomal lumen and the lysosome (Table S4). The proteins affected primarily in patients with CLN2 disease were GOT1 aspartate aminotransferase and FMO3 dimethylaniline monooxygenase, which were absent in controls. IPA analysis was also performed on the CLN5 and CLN6 disease sheep urine proteomic data (Figure S1).

The CLN5 and CLN6 sheep datasets had one significant pathway in common, the LRX/RXR activation pathway. Interestingly, the ER stress response pathway was affected in the CLN6 sheep, as was an unfolded protein response pathway and pathways associated with the immune response. In CLN5-affected sheep, however, there were greater effects on pathways associated with carbohydrate metabolism. Significant proteins affected in CLN5 sheep urine showed small fold changes. Only one protein, beta enolase, a muscle-specific form of enolase, was increased more than 2-fold. The number of significantly affected proteins in CLN6 sheep urine was greater than that seen in CLN5 urine, whereas the fold changes were similarly small. The most affected protein, ERICH6B, glutamate-rich protein 6B of unknown function, showed a >2-fold reduction in amount.

Phase 2: confirmation of human and animal model discovery targets in an extended NCL patient cohort

A multiplexed, targeted proteomic assay was developed to confirm if the changes in pathways and protein levels observed in the proteomic profiling analyses could be detected in a separate and larger number of 9 CLN2 patient samples and 13 controls. A selection of proteins from the sheep proteomic analyses was also included to see if they were relevant in human CLN5 (n = 2) and CLN6 (n = 2) urine. Candidate selection criteria are given in methods, and the proteins included in the multiplex assay are shown in Table 1. Urine from patients with other NCL disorders, CLN1 (n = 2), CLN3 (n = 1), and CLN7 (n = 1), was also analyzed to determine whether our initial discovery findings were specific to CLN2, CLN5, and CLN6 or if similar findings could be attributed to all NCL disorders. Further sample details are given in Table S7 in methods.

Table 1.

Multiplex panel of proteins listed with their corresponding gene names and cellular location

Gene CLN2 patients (label-free proteomics) Location
GOT1 Aspartate aminotransferase, cytoplasmic Cytoplasm
CTSH Pro-cathepsin H Lysosome
PEBP1 Phosphatidylethanolamine-binding protein 1 Cytoplasm
CTSZ Cathepsin Z Lysosome
AKT1 RAC-alpha serine/threonine-protein kinase Nucleosome/cytoplasm
GALNS N-acetylgalactosamine-6-sulfatase Lysosome
BHMT1 Betaine-homocysteine S-methyltransferase 1 Cytoplasm
LAMP1 Lysosome-associated membrane glycoprotein 1 Lysosome/endosome
HEXA Beta-hexosaminidase subunit alpha Lysosome
Literature research NCL biomarker
CTSL1 Cathepsin L1∗ Lysosome
TPP1 Tripeptidyl-peptidase 1∗ Lysosome
PAM Peptidyl-glycine alpha-amidating monooxygenase# Secreted
GM2A Ganglioside GM2 activator° Secreted
NAGA Alpha-N-acetylgalactosaminidase∗ Lysosome
CLN5 and CLN6 sheep (label free proteomics)
LEAP2 Liver-expressed antimicrobial peptide 2 Secreted
DSC2 Desmocollin-2 Cell membrane
SCGN Secretagogin Secreted
LAMP2 Lysosome-associated membrane glycoprotein 2 Lysosome/endosome
CLN6 sheep (label-free proteomics)
CTSB Cathepsin B Lysosome
Open in a new tab

The list is subdivided by the origin of the proteins, i.e., if they were selected from the label-free proteomics study, found in literature (∗ Sleat et al., 2017; #Doll et al., 2017; ° Sjödin et al., 2019) or from the label-free proteomics study of the urine of CLN5 and CLN6 sheep. The gene names labeled in bold were reliably detected in human urine.

Multivariate-principal component analysis (PCA) was performed to determine if the proteins in the biomarker panel are able to collectively distinguish between the control and patient samples. Figure 2 shows a PCA plot of all samples apart from one omitted outlier (patient CLN2 (4)-post). The first two components (t [1] and t[2]) explain 27% and 19.2% of the variance in urine. The variation within the dataset roughly separates the patient samples from the control samples. Some patient samples were present within the cluster of control samples (particularly CLN6 and CLN7) indicating they have a normal urine profile. The other NCL urines show a dispersed profile indicating a large degree of variation, which is away from the cluster of control samples. This indicates that these NCL urine profiles are abnormal compared to controls.

Figure 2.

Figure 2

Open in a new tab

Multivariate analysis of targeted proteomics data of NCL patient urine

Principle-component analysis of all patient (apart from one outlier, namely, CLN2(4)-post) and control samples using SIMCA presents two almost separate clusters to distinguish between NCL and control samples.

(A and B) (A) Score plot of the samples with the control samples cluster highlighted. Samples outside the control cluster are also highlighted. (B) Loading plot of the measured proteins indicating which proteins are driving the variation/change in the score plot.

The loading plot of the multivariate analysis indicated that lysosome-associated membrane glycoprotein 1 (LAMP1), peptidyl-glycine alpha-amidating monooxygenase (PAM), and aspartate aminotransferase, cytoplasmic (GOT1) are the biggest drivers of the difference observed in the two clusters, i.e., controls and patients.

Proteins affected in CLN2 disease: Figure 3A shows univariate analysis of proteins in CLN2 patient urine compared with age- and sex-matched controls. Peptide ratios representative of betaine-homocysteine S-methyltransferase 1 (BHMT1), LAMP1, and TPP1 were significantly altered in CLN2 disease samples. As patients with CLN2 disease have mutations in the TPP1 gene this result was expected, with the level of TPP1 approaching zero. Conversely, levels of urinary BHMT1 and LAMP1 were increased, on average 17.8- and 3.4-fold, respectively, compared with the control samples. Two of these three proteins, TPP1 and BHMT1, are specific for CLN2 (Figure 3B): BHMT1 is increased 16-fold and TPP1 is decreased up to 9-fold relative to patients with other NCL types (CLN1, CLN3, CLN5, CLN6, and CLN7). Comparison of the levels of BHMT1, LAMP1, and TPP1 in pre-treatment and ongoing treatment (post) samples did not show any visible trend (only two paired samples available so statistical testing is not possible), although for one patient, TPP1 did increase to a measurable amount (Figure 3D), suggesting that the ERT delivered into the brain ventricles may not stay confined to the CNS. However, the other proteins in this patient did not decrease suggesting that the levels of TPP1 coming from the CNS are not at a level sufficient to correct the rest of the diseased organ systems.

Figure 3.

Figure 3

Open in a new tab

Targeted analysis of CLN2 patient samples

The y axis shows the amount of each protein normalized to the amount of creatinine in each sample. As the f-test showed that the variance is unequally distributed a two-sample t test assuming unequal variance was performed for each protein. For box and whisker plot: error bars are the 95% confidence interval, the bottom and top of the box are the 25th and 75th percentiles, the line inside the box is the 50th percentile (median), and any outliers are shown as open circles.

(A) Comparison of patients with CLN2 disease (color coded) with age- and sex-matched controls. The amount of LAMP1 is significantly increased in the CLN2 disease patient samples (p (two tail) = 0.010). The amount of BHMT1 shows a borderline increase in the CLN2 disease patient samples (p (two tail) = 0.05). The amount of TPP1 shows a significant decrease in the CLN2 disease patient sample (p (one-tail = 0.040).

(B) Comparison of patients with CLN2 disease with all other patients with NCL (CLN1, CLN3, CLN5, CLN6, and CLN7). The amount of BHMT1 (p (two tail) = 0.059) shows a borderline significant increase, whereas there is a significant decrease in TPP1 (p (one tail) = 0.015) in patients with CLN2 disease.

(C) Comparison of two patients with CLN2 disease pre- and post-treatment for the three significant proteins identified in CLN2 samples compared with controls. Statistical analysis is not possible as n = 2.

ERT administered into the brain might not alter the proteins that come from the periphery, i.e., kidney lysosomes. Levels of BHMT1 have increased in Patient ID-1 with ongoing treatment, but levels of LAMP1 and TPP1 have remained constant, whereas patient ID-3 reveals a constant BHMT1 level but increasing levels for LAMP1 and TPP1 with ongoing treatment.

Proteins assessed in CLN5 and CLN6 sheep models: liver-enriched antimicrobial peptide-2 (LEAP2), desmocolin 2 (DSC2), cathepsin B (CTSB), secretogogin (SCGN), and lysosomal-associated membrane protein 2 (LAMP2) were identified as significantly altered in the sheep urine proteomic analysis and included in the targeted LC-MS/MS assay (Tables S2 and S3). CTSB and SCGN were not detectable in human urine. Figure 4A shows label-free proteomics data for increased levels of DSC2 in CLN5 and CLN6 (p< 0.017), increased LAMP2 in CLN5 (p < 0.028), and reduced LEAP2 in the CLN5 (p < 0.01) and CLN6 (p < 0.003) sheep urine. Targeted analysis (Figure 4B) of LAMP2, LEAP2, and DSC2 for CLN5 and CLN6 patients did not show a difference relative to controls.

Figure 4.

Figure 4

Open in a new tab

Evaluation in patient urine of protein biomarker candidates identified from CLN5 and CLN6 sheep models

Three proteins, DSC1, LEAP2, and LAMP2, identified in the proteomic analysis were detectable in humans. For the box and whisker plot: error bars are the 95% confidence interval, the bottom and top of the box are the 25th and 75th percentiles, the line inside the box is the 50th percentile (median), and any outliers are shown as open circles.

(A) Proteomic profiling data from the CLN5 and CLN6 sheep model analysis for these three proteins. DSCN2 (p < 0.017) and LAMP2 (p < 0.02 for CLN5) were increased and LEAP2 was decreased (p < 0.01).

(B) Targeted analysis of these proteins in CLN5 and CLN6 patient urine. The proteins are not significantly affected in humans and DSCN2 is decreased in CLN6, which is not observed in the sheep model.

Proteins affected in NCL: Figure 5A shows the comparison of all NCL patient samples as a disease group with controls. HEXA, LAMP1, TPP1, and GOT1 are significantly altered (p < 0.05). LAMP1 and HEXA are increased 3-fold on average, and GOT1 is elevated 6-fold. TPP1 is reduced when compared with controls for all NCL samples. Figure 5B shows a breakdown of the NCL groups compared with each other and controls for all proteins that have shown significant change. BHMT1 is specific to CLN2 as observed previously. HEXA is increased in CLN5, CLN3, and CLN1 patients and in some patients with CLN2 and CLN6. GOT1 and LAMP1 are raised in CLN5, CLN3, and some CLN2 patients. TPP1, although the affected protein in CLN2 disease, is also reduced in all other NCL forms relative to controls. The CLN7 patient sample, as indicated in the multivariate analysis, did not show many protein changes apart from reduced TPP1.

Figure 5.

Figure 5

Open in a new tab

Proteins affected across multiple NCL forms

For the box and whisker plot: error bars are the 95% confidence interval, the bottom and top of the box are the 25th and 75th percentiles, the line inside the box is the 50th percentile (median), and any outliers are shown as open circles.

(A) Comparison of patients with CLN1, CLN2, CLN3, CLN5, CLN6, and CLN7 disease (color coded) with age- and sex-matched controls. The difference in the amounts of GOT1 (p (two tail) = 0.023), HEXA (p (two tail) = 0.025), and LAMP1 (p (two tail) = 0.0004) are statistically significant. The amount of TPP1 shows a significant decrease (p (one tail) = 0.024).

(B) A breakdown of the NCL forms compared with controls for each of the affected protein.

Discussion

Using an animal model and human patient samples, we have employed a two-phase proteomic discovery and validation strategy to identify and confirm urinary proteins altered in patients with NCL. Urine is an easily accessible tissue for monitoring possible biomarkers for disease progression, and treatment and is a valuable tissue resource as the kidneys present a high proportion of mitochondria and lysosomes, both known to be affected in NCL disease.

The proteomic profiling analyses were performed on samples from two patients with CLN2 disease. Proteins observed to be altered included many proteins associated with the lysosome and is consistent with CLN2 being caused by a deficiency in a lysosomal enzyme involved in protein degradation. Other pathways also potentially affected in CLN2 urine include signaling pathways, e.g., those involved in inactivation of BAG2 signaling and inflammatory response-associated pathways. The BAG2 signaling pathway, which includes heat shock proteins involved in the control of folding or degradation of misfolded proteins, are typically either increased or decreased in neurodegenerative disorders and implicated in the pathogenesis of various neurodegenerative diseases (Soto and Pritzkow, 2018). Many neurodegenerative diseases are primarily disorders of protein misfolding and accumulation of storage material. Protein misfolding would be expected to be affected in CLN2 disease due to the lysosomal accumulation of mitochondrial ATP-synthase subunit-c, which has been shown to be stored in lysosome-derived organelles in many but not all NCL, including CLN 2, 5, 6 diseases (Palmer, 2015; Chen et al., 2004; Palmer et al., 1989, 1992). Protein misfolding is clearly a pathological pathway in the brain and other tissues that we can observe in patient urine, and is consistent with the systemic pathology of CLN2 disease that affects nearly all cell types in the body. Other potentially affected pathways also include integrin-like kinase (ILK) signaling, which promotes cell growth. Altered cell growth has previously been described in NCL disease models (Guo et al., 1999; Huber et al., 2014; Mahmood et al., 2013; Savchenko et al., 2017; Sleat et al., 2017). Other pathways that may be affected involve inflammation reflective of neuroinflammation, which is common to the ovine diseases and several other NCLs (Palmer et al., 2013; Kay et al., 2006; Oswald et al., 2005; Nelvagal et al., 2020), that is known to occur due to stress of the ER from the unfolded protein response (Marotta et al., 2017). Proteins affected in patients with CLN2 disease included FMO3, which is likely to be present due to its role in metabolism of administered drugs. Phosphatidylethanolamine-binding protein 1 (PEBP1) is a protease inhibitor with multiple functions; why it is increased in CLN2 is unknown. GALNS N-acetylgalactosamine-6-sulfatase and beta-hexosaminidase alpha are part of the glycosaminoglycan degradation pathway, which has been previously implicated in NCL disease (Sleat et al., 2019).

Subsequent targeted analysis was performed on a larger cohort of samples from patients with CLN2 disease undergoing ERT treatment (n = 7, for two patients pre- and post-treatment paired samples were available) and a small number of other patients with NCL including two patients with CLN1, one with CLN3, two with CLN5, two with CLN6, and one with CLN7 disease. We compared data of the NCL disease group, involving all samples, and that of the CLN2 disease group with controls. Proteins that were found specific for CLN2 disease included BHMT1 as shown in Figures 3A and 5B. BHMT1 is a zinc-dependent cytosolic enzyme that catalyzes the transfer of a methyl group from betaine to homocysteine. This reaction results in increased dimethylglycine and methionine (Teng et al., 2011). High concentrations of BHMT may indicate higher production of the substrate homocysteine. High concentrations of homocysteine in plasma are associated with cardiovascular diseases, renal insufficiency, and cognitive impairment and usually indicate vitamin B12 deficiency. Why higher BHMT1 and possibly homocysteine occur specifically in patients with CLN2 disease is unknown, but they could have potential as markers for monitoring future treatments for patients with CLN2 disease that also target the systemic disease as well as the brain.

The lack of TPP1 in the brain is known to cause neurological disease, but the deficiency of TPP1 outside the CNS could also indicate a wider functional impairment. This may be associated with the progression of cardiac pathology suggested to occur in some NCL forms, including CLN2 disease (Fukumura et al., 2012; Katz et al., 2017; Ostergaard et al., 2011). It is assumed that upon administration of ERT directly into the CSF the neurological burden is relieved, and a longer lifespan becomes possible, which may result in pathology developing in the untreated organs becoming apparent (Katz et al., 2017). Treatments that target not just the CNS will likely also be needed, and the ability to monitor urinary TPP1 could be of benefit in this scenario. Another protein significantly increased in urine of patients with CLN2 disease is LAMP1, but this is not specific to CLN2. This lysosomal-associated membrane protein is situated in a dynamic equilibrium between lysosomes, endosomes, and the plasma membrane and is partially responsible for maintaining lysosomal integrity, pH, and catabolism (Xicoy et al., 2019; Eskelinen, 2006). LAMP1 is a widely used biomarker for lysosomes and lysosomal storage disorders but is usually measured in plasma or cells (Meikle et al., 1997). The elevated levels of LAMP1 in urine of patients with NCL may indicate the higher burden on the lysosome. Other urinary proteins increased in NCL disorders include GOT1 and HEXA. HEXA is increased in most patients with NCL; however, it is not increased sufficiently in patients with CLN2 disease to be significant when compared with the corresponding controls. HEXA is part of the glycosphingolipid degradation pathway (Breiden and Sandhoff, 2020; Hepbildikler et al., 2002). Specific deficiency of HEXA results in the lysosomal storage disorder called Tay Sachs disease (Okada and O'Brien, 1969). This observation of HEXA in urine here confirms a previous study that showed that the enzymatic activity of HEXA in CSF from patients with CLN2 and CLN3 disease is significantly increased (Sleat et al., 2017). HEXA was also found to be altered in the Dictyostelium NCL model of CLN3 disease, indicating this enzyme is affected across different species model systems and may be a suitable biomarker candidate. Like HEXA, CLN5 has been described to also be a glycolytic hydrolase (Huber and Mathavarajah, 2018), indicating this particular molecular function is relevant in NCL disease pathology.

The fact that HEXA is increased but most other lysosomal enzymes are not indicates that it may not derive from lysosomes of disrupted cells. The HEXA could come from mast cells as it is a known marker of mast cell degranulation (Moon et al., 2014). Mast cell activity has not been specifically assessed in NCL, but mast cells have been observed to surround the optic nerve in a mouse model for CLN3 disease indicating compromise of the blood-brain barrier (Sappington et al., 2003). Proteomic analysis of CLN1 and CLN2 mouse model brain tissue has shown increased levels of CD63 (Sleat et al., 2019), which is thought to play a role in activation of mast cells and basophils (Kabashima et al., 2018). Therefore it is possible that the presence of HEXA is indicative of a mast cell-associated immune response in NCL.

Glutamic-oxaloacetic transaminase 1 is a cytosolic enzyme that plays a role in amino acid metabolism and the urea and tricarboxylic acid cycles. In principle GOT1 is a bidirectional enzyme, which turns it into an important regulator of glutamate concentration, the major excitatory neurotransmitter of the CNS, and also of aspartic acid levels. The form seen in urine is likely derived from kidney and plasma, which itself contains GOT1 from liver and muscle. Clinically GOT1 in plasma is routinely measured as aspartate aminotransferase 1 (ALT) and is a known marker of tissue damage. Increased plasma ALT has been observed in a canine model of CLN2 disease undergoing intracerebroventricular gene therapy (Katz et al., 2017), which also noted that treated dogs went on to develop cardiac disease, which is the likely source of the raised ALT. The elevated GOT1/ALT seen in NCL is probably due to tissue damage. It is plausible that this could be from the heart as progressive cardiac disease has been reported in children and adults with NCL (Ostergaard et al., 2011; Katz et al., 2017; Fukumura et al., 2012).

We considered the specificity of these proteins as biomarkers for NCL or Lysosomal Storage Disorders (LSDs). LAMP1 is already well documented as a general marker of lysosome dysfunction and lysosomal disease (Meikle et al., 1997). We have profiled urine from patients with mucopolysaccharidosis and Niemann Pick C disease (unpublished data, Table S9). GOT1 was not affected, but interestingly, levels of BHMT are significantly reduced in NPC patient urine, the opposite of that observed in CLN2 disease. Why this occurs is unknown, but it does indicate that methionine metabolism is perturbed in LSDs.

ALT (GOT1) and HEXA are routinely measured in the clinical setting to monitor tissue damage and mast cell degranulation, so it would be possible to include these proteins when monitoring the effect of future novel systemic treatments. In summary, we have identified BHMT1 as a new biomarker in CLN2 disease and GOT1 (ALT) for systemic NCL disease as well as the known marker LAMP1. Furthermore, although we were unable to demonstrate that biomarkers in the sheep CLN5 and CLN6 models can translate to humans, we have confirmed the involvement of HEXA as observed in previous studies on the eukaryotic model system Dictyostelium (Huber et al., 2014).

Limitations of the study

The scarcity of available patient samples due to these disorders being rare means it will always be difficult to ascertain statistical significance in such studies. The targeted analysis unfortunately did not confirm the proteins detected in the sheep as altered in human CLN5 and CLN6 disease samples. CTSB, alpha-N-acetylgalactosaminidase (NAGA), and RAC-alpha serine/threonine-protein kinase (AKT1) could not be detected in the human urine. Animal models can be very useful for understanding disease pathology and designing treatments, but their application to identify biomarkers in humans as in this scenario is limited so far.

Resource availability

Lead contact

Wendy E Heywood wendy.heywood@ucl.ac.uk.

Materials availability

This study did not generate new unique reagents.

Data and code availability

Original data have been deposited to PeptideAtlas at http://www.peptideatlas.org/PASS/PASS01636.

Methods

All methods can be found in the accompanying Transparent methods supplemental file.

Acknowledgments

The authors would like to thank the Batten Disease Support and Research Association (BDSRA), USA, Noah's Hope/Hope for Bridget and Drew's Hope for funding this project. Medical Research Council (Grant No MR/R025134/1) supported research (S.E.M.). All research at Great Ormond Street Hospital NHS Foundation Trust and UCL Great Ormond Street Institute of Child Health is made possible by the NIHR Great Ormond Street Hospital Biomedical Research Center. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR, or the Department of Health. We would also like to thank Martin Wellby and Dr. Katharina Russell, Lincoln University, for collecting the ovine samples.

Author contributions

K.I. created the targeted assay, performed the targeted analysis, and wrote the manuscript. R.C. performed the sheep urine analysis. P.M. enabled provision of patient samples, study design, and concept. B.C. provided patient samples. P.G. enabled provision of patient samples, study design, and concept. S.E.M., study design and concept and expertise. D.N.P. provided sheep model samples. K.M., study design and concept and resources. W.E. H., funding, concept, study design, and wrote manuscript. All authors contributed to manuscript writing and revision.

Declaration of Interests

P.G. is an investigator on BioMarin-sponsored studies and has received research grants and speaker honoraria. S.E.M. receives financial support from BioMarin Pharmaceutical Inc. to maintain the NCL Mutation Database. W.E.H. has received a travel award from BioMarin Pharmaceutical Inc. The remaining authors declare no competing interests.

Published: February 19, 2021

Footnotes

Supplemental information can be found online at https://doi.org/10.1016/j.isci.2020.102020.

Supplemental information

Document S1. Transparent methods, Figures S1 and S2, and Tables S1–S

10

mmc1.pdf (1.4MB, pdf)
Data S1 Data file of exported proteomics datasets related to Figures 1–5
mmc2.xlsx (695.1KB, xlsx)

References

  1. Adams H.R., Mink J.W. Neurobehavioral features and natural history of juvenile neuronal ceroid lipofuscinosis (batten disease) J. Child Neurol. 2013;28:1128–1136. doi: 10.1177/0883073813494813. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Breiden B., Sandhoff K. Mechanism of secondary ganglioside and lipid accumulation in lysosomal disease. Int. J. Mol. Sci. 2020;21:2566. doi: 10.3390/ijms21072566. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Butz E.S., Chandrachud U., Mole S.E., Cotman S.L. Moving towards a new era of genomics in the neuronal ceroid lipofuscinoses. Biochim. Biophys. Acta Mol. Basis Dis. 2020;1866:165571. doi: 10.1016/j.bbadis.2019.165571. [DOI] [PubMed] [Google Scholar]
  4. Chen R., Fearnley I.M., Palmer D.N., Walker J.E. Lysine 43 is trimethylated in subunit C from bovine mitochondrial ATP synthase and in storage bodies associated with batten disease. J. Biol. Chem. 2004;279:21883–21887. doi: 10.1074/jbc.M402074200. [DOI] [PubMed] [Google Scholar]
  5. de Blieck E.A., Augustine E.F., Marshall F.J., Adams H., Cialone J., Dure L., Kwon J.M., Newhouse N., Rose K., Rothberg P.G. Methodology of clinical research in rare diseases: development of a research program in juvenile neuronal ceroid lipofuscinosis (JNCL) via creation of a patient registry and collaboration with patient advocates. Contemp. Clin. Trials. 2013;35:48–54. doi: 10.1016/j.cct.2013.04.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. di Ronza A., Bajaj L., Sharma J., Sanagasetti D., Lotfi P., Adamski C.J., Collette J., Palmieri M., Amawi A., Popp L. CLN8 is an endoplasmic reticulum cargo receptor that regulates lysosome biogenesis. Nat. Cell Biol. 2018;20:1370–1377. doi: 10.1038/s41556-018-0228-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Doll S., Dreßen M., Geyer P.E., Itzhak D.N., Braun C., Doppler S.A., Meier F., Deutsch M.-A., Lahm H., Lange R. Region and cell-type resolved quantitative proteomic map of the human heart. Nat. Commun. 2017;8:1469. doi: 10.1038/s41467-017-01747-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Eskelinen E.-L. Roles of LAMP-1 and LAMP-2 in lysosome biogenesis and autophagy. Mol. Aspects Med. 2006;27:495–502. doi: 10.1016/j.mam.2006.08.005. [DOI] [PubMed] [Google Scholar]
  9. Fukumura S., Saito Y., Saito T., Komaki H., Nakagawa E., Sugai K., Sasaki M., Oka A., Takamisawa I. Progressive conduction defects and cardiac death in late infantile neuronal ceroid lipofuscinosis. Dev. Med. Child Neurol. 2012;54:663–666. doi: 10.1111/j.1469-8749.2011.04170.x. [DOI] [PubMed] [Google Scholar]
  10. Golabek A.A., Kida E., Walus M., Wujek P., Mehta P., Wisniewski K.E. Biosynthesis, glycosylation, and enzymatic processingin vivo of human tripeptidyl-peptidase I. J. Biol. Chem. 2003;278:7135–7145. doi: 10.1074/jbc.M211872200. [DOI] [PubMed] [Google Scholar]
  11. Griffey M., Macauley S.L., Ogilvie J.M., Sands M.S. AAV2-mediated ocular gene therapy for infantile neuronal ceroid lipofuscinosis. Mol. Ther. 2005;12:413–421. doi: 10.1016/j.ymthe.2005.04.018. [DOI] [PubMed] [Google Scholar]
  12. Guo W.X., Mao C., Obeid L.M., Boustany R.M. A disrupted homologue of the human CLN3 or juvenile neuronal ceroid lipofuscinosis gene in Saccharomyces cerevisiae: a model to study Batten disease. Cell Mol. Neurobiol. 1999;19:671–680. doi: 10.1023/A:1006992704108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Harpole M., Davis J., Espina V. Current state of the art for enhancing urine biomarker discovery. Expert Rev. Proteomics. 2016;13:609–626. doi: 10.1080/14789450.2016.1190651. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Henderson M.X., Wirak G.S., Zhang Y.Q., Dai F., Ginsberg S.D., Dolzhanskaya N., Staropoli J.F., Nijssen P.C., Lam T.T., Roth A.F. Neuronal ceroid lipofuscinosis with DNAJC5/CSPalpha mutation has PPT1 pathology and exhibit aberrant protein palmitoylation. Acta Neuropathol. 2016;131:621–637. doi: 10.1007/s00401-015-1512-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Hepbildikler S.T., Sandhoff R., Kölzer M., Proia R.L., Sandhoff K. Physiological substrates for human lysosomal β-hexosaminidase S. J. Biol. Chem. 2002;277:2562–2572. doi: 10.1074/jbc.M105457200. [DOI] [PubMed] [Google Scholar]
  16. Hu J., Lu J.-Y., Wong A.M.S., Hynan L.S., Birnbaum S.G., Yilmaz D.S., Streit B.M., Lenartowicz E.M., Thompson T.C.M., Cooper J.D., Hofmann S.L. Intravenous high-dose enzyme replacement therapy with recombinant palmitoyl-protein thioesterase reduces visceral lysosomal storage and modestly prolongs survival in a preclinical mouse model of infantile neuronal ceroid lipofuscinosis. Mol. Genet. Metab. 2012;107:213–221. doi: 10.1016/j.ymgme.2012.05.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Huang D.W., Sherman B.T., Lempicki R.A. Bioinformatics enrichment tools: paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res. 2008;37:1–13. doi: 10.1093/nar/gkn923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Huang D.W., Sherman B.T., Lempicki R.A. Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat. Protoc. 2009;4:44–57. doi: 10.1038/nprot.2008.211. [DOI] [PubMed] [Google Scholar]
  19. Huber R.J. Molecular networking in the neuronal ceroid lipofuscinoses: insights from mammalian models and the social amoeba Dictyostelium discoideum. J. Biomed. Sci. 2020;27:64. doi: 10.1186/s12929-020-00653-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Huber R.J., Mathavarajah S. Cln5 is secreted and functions as a glycoside hydrolase in Dictyostelium. Cell Signal. 2018;42:236–248. doi: 10.1016/j.cellsig.2017.11.001. [DOI] [PubMed] [Google Scholar]
  21. Huber R.J., Myre M.A., Cotman S.L. Loss of Cln3 function in the social amoeba Dictyostelium discoideum causes pleiotropic effects that are rescued by human CLN3. PloS One. 2014;9:e110544. doi: 10.1371/journal.pone.0110544. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kabashima K., Nakashima C., Nonomura Y., Otsuka A., Cardamone C., Parente R., De Feo G., Triggiani M. Biomarkers for evaluation of mast cell and basophil activation. Immunol. Rev. 2018;282:114–120. doi: 10.1111/imr.12639. [DOI] [PubMed] [Google Scholar]
  23. Kalantari S., Jafari A., Moradpoor R., Ghasemi E., Khalkhal E. Human urine proteomics: analytical techniques and clinical applications in renal diseases. Int. J. Proteomics. 2015;2015:782798. doi: 10.1155/2015/782798. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Katz M.L., Coates J.R., Sibigtroth C.M., Taylor J.D., Carpentier M., Young W.M., Wininger F.A., Kennedy D., Vuillemenot B.R., O'Neill C.A. Enzyme replacement therapy attenuates disease progression in a canine model of late-infantile neuronal ceroid lipofuscinosis (CLN2 disease) J. Neurosci. Res. 2014;92:1591–1598. doi: 10.1002/jnr.23423. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Katz M.L., Johnson G.C., Leach S.B., Williamson B.G., Coates J.R., Whiting R.E.H., Vansteenkiste D.P., Whitney M.S. Extraneuronal pathology in a canine model of CLN2 neuronal ceroid lipofuscinosis after intracerebroventricular gene therapy that delays neurological disease progression. Gene Ther. 2017;24:215–223. doi: 10.1038/gt.2017.4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kay G.W., Palmer D.N., Rezaie P., Cooper J.D. Activation of non-neuronal cells within the prenatal developing brain of sheep with neuronal ceroid lipofuscinosis. Brain Pathol. 2006;16:110–116. doi: 10.1111/j.1750-3639.2006.00002.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kohlschütter A., Schulz A., Bartsch U., Storch S. Current and emerging treatment strategies for neuronal ceroid lipofuscinoses. CNS Drugs. 2019;33:315–325. doi: 10.1007/s40263-019-00620-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Liu W., Kleine-Holthaus S.M., Herranz-Martin S., Aristorena M., Mole S.E., Smith A.J., Ali R.R., Rahim A.A. Experimental gene therapies for the NCLs. Biochim. Biophys. Acta Mol. Basis Dis. 2020;1866:165772. doi: 10.1016/j.bbadis.2020.165772. [DOI] [PubMed] [Google Scholar]
  29. Lu J.-Y., Nelvagal H.R., Wang L., Birnbaum S.G., Cooper J.D., Hofmann S.L. Intrathecal enzyme replacement therapy improves motor function and survival in a preclinical mouse model of infantile neuronal ceroid lipofuscinosis. Mol. Genet. Metab. 2015;116:98–105. doi: 10.1016/j.ymgme.2015.05.005. [DOI] [PubMed] [Google Scholar]
  30. Mahmood F., Fu S., Cooke J., Wilson S.W., Cooper J.D., Russell C. A zebrafish model of CLN2 disease is deficient in tripeptidyl peptidase 1 and displays progressive neurodegeneration accompanied by a reduction in proliferation. Brain. 2013;136:1488–1507. doi: 10.1093/brain/awt043. [DOI] [PubMed] [Google Scholar]
  31. Marotta D., Tinelli E., Mole S.E. NCLs and ER: a stressful relationship. Biochim. Biophys. Acta Mol. Basis Dis. 2017;1863:1273–1281. doi: 10.1016/j.bbadis.2017.04.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Meikle P.J., Brooks D.A., Ravenscroft E.M., Yan M., Williams R.E., Jaunzems A.E., Chataway T.K., Karageorgos L.E., Davey R.C., Boulter C.D. Diagnosis of lysosomal storage disorders: evaluation of lysosome-associated membrane protein LAMP-1 as a diagnostic marker. Clin. Chem. 1997;43:1325–1335. [PubMed] [Google Scholar]
  33. Mole S.E., Anderson G., Band H.A., Berkovic S.F., Cooper J.D., Kleine Holthaus S.M., McKay T.R., Medina D.L., Rahim A.A., Schulz A., Smith A.J. Clinical challenges and future therapeutic approaches for neuronal ceroid lipofuscinosis. Lancet Neurol. 2019;18:107–116. doi: 10.1016/S1474-4422(18)30368-5. [DOI] [PubMed] [Google Scholar]
  34. Mole S.E., Michaux G., Codlin S., Wheeler R.B., Sharp J.D., Cutler D.F. CLN6, which is associated with a lysosomal storage disease, is an endoplasmic reticulum protein. Exp. Cell Res. 2004;298:399–406. doi: 10.1016/j.yexcr.2004.04.042. [DOI] [PubMed] [Google Scholar]
  35. Moon T.C., Befus A.D., Kulka M. Mast cell mediators: their differential release and the secretory pathways involved. Front. Immunol. 2014;5:569. doi: 10.3389/fimmu.2014.00569. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Nelvagal H.R., Lange J., Takahashi K., Tarczyluk-Wells M.A., Cooper J.D. Pathomechanisms in the neuronal ceroid lipofuscinoses. Biochim. Biophys. Acta Mol. Basis Dis. 2020;1866:165570. doi: 10.1016/j.bbadis.2019.165570. [DOI] [PubMed] [Google Scholar]
  37. Okada S., O'Brien J.S. Tay-Sachs disease: generalized absence of a beta-D-N-acetylhexosaminidase component. Science. 1969;165:698–700. doi: 10.1126/science.165.3894.698. [DOI] [PubMed] [Google Scholar]
  38. Ostergaard J.R., Rasmussen T.B., Molgaard H. Cardiac involvement in juvenile neuronal ceroid lipofuscinosis (Batten disease) Neurology. 2011;76:1245–1251. doi: 10.1212/WNL.0b013e31821435bd. [DOI] [PubMed] [Google Scholar]
  39. Oswald M.J., Palmer D.N., Kay G.W., Shemilt S.J., Rezaie P., Cooper J.D. Glial activation spreads from specific cerebral foci and precedes neurodegeneration in presymptomatic ovine neuronal ceroid lipofuscinosis (CLN6) Neurobiol. Dis. 2005;20:49–63. doi: 10.1016/j.nbd.2005.01.025. [DOI] [PubMed] [Google Scholar]
  40. Palmer D.N. The relevance of the storage of subunit c of ATP synthase in different forms and models of Batten disease (NCLs) Biochim. Biophys. Acta. 2015;1852:2287–2291. doi: 10.1016/j.bbadis.2015.06.014. [DOI] [PubMed] [Google Scholar]
  41. Palmer D.N., Barry L.A., Tyynela J., Cooper J.D. NCL disease mechanisms. Biochim. Biophys. Acta. 2013;1832:1882–1893. doi: 10.1016/j.bbadis.2013.05.014. [DOI] [PubMed] [Google Scholar]
  42. Palmer D.N., Fearnley I.M., Medd S.M., Walker J.E., Martinus R.D., Bayliss S.L., Hall N.A., Lake B.D., Wolfe L.S., Jolly R.D. Lysosomal storage of the DCCD reactive proteolipid subunit of mitochondrial ATP synthase in human and ovine ceroid lipofuscinoses. Adv. Exp. Med. Biol. 1989;266:211–222. doi: 10.1007/978-1-4899-5339-1_15. discussion 223. [DOI] [PubMed] [Google Scholar]
  43. Palmer D.N., Fearnley I.M., Walker J.E., Hall N.A., Lake B.D., Wolfe L.S., Haltia M., Martinus R.D., Jolly R.D. Mitochondrial ATP synthase subunit c storage in the ceroid-lipofuscinoses (Batten disease) Am. J. Med. Genet. 1992;42:561–567. doi: 10.1002/ajmg.1320420428. [DOI] [PubMed] [Google Scholar]
  44. Sappington R.M., Pearce D.A., Calkins D.J. Optic nerve degeneration in a murine model of juvenile ceroid lipofuscinosis. Invest. Ophthalmol. Vis. Sci. 2003;44:3725–3731. doi: 10.1167/iovs.03-0039. [DOI] [PubMed] [Google Scholar]
  45. Savchenko E., Singh Y., Konttinen H., Lejavova K., Mediavilla Santos L., Grubman A., Karkkainen V., Keksa-Goldsteine V., Naumenko N., Tavi P. Loss of Cln5 causes altered neurogenesis in a mouse model of a childhood neurodegenerative disorder. Dis. Model. Mech. 2017;10:1089–1100. doi: 10.1242/dmm.029165. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Schulz A., Ajayi T., Specchio N., de Los Reyes E., Gissen P., Ballon D., Dyke J.P., Cahan H., Slasor P., Jacoby D. Study of intraventricular cerliponase alfa for CLN2 disease. N. Engl. J. Med. 2018;378:1898–1907. doi: 10.1056/NEJMoa1712649. [DOI] [PubMed] [Google Scholar]
  47. Sindelar M., Dyke J.P., Deeb R.S., Sondhi D., Kaminsky S.M., Kosofsky B.E., Ballon D.J., Crystal R.G., Gross S.S. Untargeted metabolite profiling of cerebrospinal fluid uncovers biomarkers for severity of late infantile neuronal ceroid lipofuscinosis (CLN2, batten disease) Sci. Rep. 2018;8:15229. doi: 10.1038/s41598-018-33449-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Sjödin S., Brinkmalm G., Öhrfelt A., Parnetti L., Paciotti S., Hansson O., Hardy J., Blennow K., Zetterberg H., Brinkmalm A. Endo-lysosomal proteins and ubiquitin CSF concentrations in Alzheimer’s and Parkinson’s disease. Alzheimers Res. Ther. 2019;11:82. doi: 10.1186/s13195-019-0533-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Sleat D.E., Tannous A., Sohar I., Wiseman J.A., Zheng H., Qian M., Zhao C., Xin W., Barone R., Sims K.B. Proteomic analysis of brain and cerebrospinal fluid from the three major forms of neuronal ceroid lipofuscinosis reveals potential biomarkers. J. Proteome Res. 2017;16:3787–3804. doi: 10.1021/acs.jproteome.7b00460. [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Sleat D.E., Wiseman J.A., El-Banna M., Zheng H., Zhao C., Soherwardy A., Moore D.F., Lobel P. Analysis of brain and cerebrospinal fluid from mouse models of the three major forms of neuronal ceroid lipofuscinosis reveals changes in the lysosomal proteome. Mol. Cell Proteomics. 2019;18:2244–2261. doi: 10.1074/mcp.RA119.001587. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Soto C., Pritzkow S. Protein misfolding, aggregation, and conformational strains in neurodegenerative diseases. Nat. Neurosci. 2018;21:1332–1340. doi: 10.1038/s41593-018-0235-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Staropoli J.F., Karaa A., Lim E.T., Kirby A., Elbalalesy N., Romansky S.G., Leydiker K.B., Coppel S.H., Barone R., Xin W. A homozygous mutation in KCTD7 links neuronal ceroid lipofuscinosis to the ubiquitin-proteasome system. Am. J. Hum. Genet. 2012;91:202–208. doi: 10.1016/j.ajhg.2012.05.023. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Teng Y.-W., Mehedint M.G., Garrow T.A., Zeisel S.H. Deletion of betaine-homocysteine S-methyltransferase in mice perturbs choline and 1-carbon metabolism, resulting in fatty liver and hepatocellular carcinomas. J. Biol. Chem. 2011;286:36258–36267. doi: 10.1074/jbc.M111.265348. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Tracy C.J., Whiting R.E.H., Pearce J.W., Williamson B.G., Vansteenkiste D.P., Gillespie L.E., Castaner L.J., Bryan J.N., Coates J.R., Jensen C.A., Katz M.L. Intravitreal implantation of TPP1-transduced stem cells delays retinal degeneration in canine CLN2 neuronal ceroid lipofuscinosis. Exp. Eye Res. 2016;152:77–87. doi: 10.1016/j.exer.2016.09.003. [DOI] [PubMed] [Google Scholar]
  55. Whiting R.E.H., Narfström K., Yao G., Pearce J.W., Coates J.R., Castaner L.J., Jensen C.A., Dougherty B.N., Vuillemenot B.R., Kennedy D. Enzyme replacement therapy delays pupillary light reflex deficits in a canine model of late infantile neuronal ceroid lipofuscinosis. Exp. Eye Res. 2014;125:164–172. doi: 10.1016/j.exer.2014.06.008. [DOI] [PubMed] [Google Scholar]
  56. Williams R.E., Mole S.E. New nomenclature and classification scheme for the neuronal ceroid lipofuscinoses. Neurology. 2012;79:183–191. doi: 10.1212/WNL.0b013e31825f0547. [DOI] [PubMed] [Google Scholar]
  57. Wlodawer A., Durell S.R., Li M., Oyama H., Oda K., Dunn B.M. A model of tripeptidyl-peptidase I (CLN2), a ubiquitous and highly conserved member of the sedolisin family of serine-carboxyl peptidases. BMC Struct. Biol. 2003;3:8. doi: 10.1186/1472-6807-3-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Xicoy H., Peñuelas N., Vila M., Laguna A. Autophagic- and lysosomal-related biomarkers for Parkinson’s disease: lights and shadows. Cells. 2019;8:1317. doi: 10.3390/cells8111317. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Transparent methods, Figures S1 and S2, and Tables S1–S

10

mmc1.pdf (1.4MB, pdf)
Data S1 Data file of exported proteomics datasets related to Figures 1–5
mmc2.xlsx (695.1KB, xlsx)

Data Availability Statement

Original data have been deposited to PeptideAtlas at http://www.peptideatlas.org/PASS/PASS01636.

Articles from iScience are provided here courtesy of Elsevier

RESOURCES