Human Molecular Genetics Advance Access originally published online on October 26, 2005
Human Molecular Genetics 2005 14(23):3759-3773; doi:10.1093/hmg/ddi406
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Defective lysosomal arginine transport in juvenile Batten disease
1Center for Aging and Developmental Biology, Aab Institute of Biomedical Sciences, 2Department of Biochemistry and Biophysics and 3Department of Neurology, University of Rochester School of Medicine and Dentistry, Rochester, NY 14642, USA
* To whom correspondence should be addressed at: Center for Aging and Developmental Biology, Department of Biochemistry and Biophysics, PO Box 645, University of Rochester, School of Medicine and Dentistry, Rochester, NY 14642, USA. Tel: +1 5852731514; Fax: +1 5852761972; Email: david_pearce{at}urmc.rochester.edu
Received August 12, 2005; Accepted October 20, 2005
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Mutations in the CLN3 gene, which encodes a lysosomal membrane protein, are responsible for the neurodegenerative disorder juvenile Batten disease. A previous study on the yeast homolog to CLN3, designated Btn1p, revealed a potential role for CLN3 in the transport of arginine into the yeast vacuole, the equivalent organelle to the mammalian lysosome. Lysosomes isolated from lymphoblast cell lines, established from individuals with juvenile Batten disease-bearing mutations in CLN3, but not age-matched controls, demonstrate defective transport of arginine. Furthermore, we show that there is a depletion of arginine in cells derived from individuals with juvenile Batten disease. We have, therefore, characterized lysosomal arginine transport in normal lysosomes and show that it is ATP-, v-ATPase- and cationic-dependent. This and previous studies have shown that both arginine and lysine are transported by the same transport system, designated system c. However, we report that lysosomes isolated from juvenile Batten disease lymphoblasts are only defective for arginine transport. These results suggest that the CLN3 defect in juvenile Batten disease may affect how intracellular levels of arginine are regulated or distributed throughout the cell. This assertion is supported by two other experimental approaches. First, an antibody to CLN3 can block lysosomal arginine transport and second, expression of CLN3 in JNCL cells using a lentiviral vector can restore lysosomal arginine transport. CLN3 may have a role in regulating intracellular levels of arginine possibly through control of the transport of this amino acid into lysosomes.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Neuronal ceroid-lipofuscinoses (NCLs) are the most common group of progressive neurodegenerative diseases in children, with an incidence as high as 1 in 12 500 live births and with about 440 000 carriers in the USA (1
,2). These disorders are autosomal recessive, and have similar early symptoms and disease progression. Progression is characterized by vision loss, behavioral changes, decline in mental abilities, increased severity of seizures, loss of motor skills, parkinsonism and pre-mature death. The NCLs are characterized pathologically by the accumulation of autofluorescent hydrophobic material in the lysosomes of neurons and other cell types; however, the mechanism driving these cellular alterations and the manner in which they relate to the neurodegeneration in NCLs are unknown.
The most common NCL, the juvenile form (JNCL), typically referred to as Batten disease, results from mutations in CLN3 (3,4). CLN3 encodes a 438 amino acid protein with an estimated molecular weight of 55 kDa after modifications (for review, see 5). Approximately 85% of all JNCL patients harbor a 1.02 kb deletion in CLN3, which results in a frameshift and pre-mature stop in the transcript (1–4), which if translated would result in a truncated CLN3 polypeptide of 181 amino acids, of which the last 28 would be novel. Moreover, 31 other mutations in CLN3 have been identified in patients with Batten disease (http://www.ucl.ac.uk/ncl/CLN3.html).
The CLN3 protein is conserved across all eukaryotes (6). Proposed functions for CLN3 include lysosomal acidification, sequestration of lysosomal enzymes, degradation of proteins, small molecule transport, organelle fusion and apoptosis (5). Much of what we know about the possible function of CLN3 has come from studies on the yeast homolog, Btn1p. The yeast protein Btn1p is 39% identical and 59% similar to CLN3 (7). Btn1p localizes to the vacuole, which is orthologous to the lysosome (8). Furthermore, yeast strains lacking Btn1p have a lower than normal vacuolar pH during early growth that becomes normalized by later growth stages (9). Studies on human CLN3 have also revealed that mutations in CLN3 result in a disruption of lysosomal pH, indicating that pH is affected when CLN3 function is compromised (10,11). Interestingly, a yeast strain lacking Btn1p was shown to have a defect in vacuolar transport of arginine, and this defect was shown to be complemented by expression of CLN3, suggesting that altered vacuolar/lysosomal pH may well be a consequence of altered transport of small molecules into the vacuole (12). However, Btn1p and CLN3 do not share sequence homology to any known transporter, which may indicate that CLN3/Btn1p does not actually transport arginine but rather may play a role in the regulation of transport at the lysosomal/vacuolar membrane.
We report that lysosomes isolated from lymphoblast cell lines derived from patients with JNCL, but not age-matched controls are defective for transport of arginine. Previous reports have shown the presence of a cation amino acid transport system in the membrane of lysosomes termed system c (13–15). System c is the transport route for cationic amino acids across the lysosomal membrane, and was initially characterized in fibroblasts (13,14). The finding that amino acid transport in the lysosome was mediated by a transporter gave further support to the concept that small molecules could not simply diffuse through the lysosomal membrane. Although system c was characterized as means to better understand the mechanism of action of cysteamine on treating patients with cystinosis, the findings strongly suggested that arginine and lysine were recognized as substrates and that the transport of these molecules was likely to occur through one transport route (13–15). Despite this initial characterization, little is known about cationic amino acid transport into lysosomes, so we have further characterized the process of arginine transport into the lysosome. We confirm that lysosomes have an arginine transport system that is ATP- and cation-dependent, and requires the activity of the vacuolar H+-ATPase. JNCL is a rare disease and patient samples are equally rare. Thus to extrapolate our findings on defective lysosomal arginine transport in JNCL, we have explored the role of CLN3 in this process. We demonstrate that CLN3 is unlikely to be the actual arginine transporter, but it is required for the transport of arginine into the lysosome, by specifically inhibiting lysosomal arginine transport with an antibody to CLN3, and by restoring transport through expression of CLN3 in JNCL cells. Although other processes may be disrupted by mutation of CLN3 in JNCL, we propose that an underlying cause of altered lysosomal function in JNCL could result from altered transport of arginine across the lysosomal membrane.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Isolation and assay of lysosomal fractions purity
Studies on lysosomal transporter systems have typically used crude lysosomal fractions, also referred as crude granular fractions, which are obtained by differential centrifugation (16
–20). We have utilized isopyknic fractionation that has been optimized for lysosomal isolation from EBV-lymphoblasts (16,20–23). In order to assay the purity of our fractions, we performed western blots of the different fractions obtained throughout the isolation process, using antibodies targeted to different organelles (Fig. 1). Identical results were obtained for each disease cell line and the corresponding matching control. Typically, most studies measure the degree of enrichment of lysosomes by determining the activity or presence of lysosomal enzymes. However, it has been reported that JNCL lysosomes have an increased activity of certain lysosomal enzymes (24), which in our case will make these measurements inaccurate for adequate normalization between control and disease lysosomes. We have, therefore, used total protein concentration as the normalization method.
|
The precise protocol used for lysosomal isolation is described in Materials and Methods. Figure 1A shows low concentration of all organelles in the first supernatant fraction. Figure 1B shows an increased enrichment of almost all organelles, and although this is the fraction that has been widely used as an enriched lysosomal fraction to assay transport systems in lysosomes, it is clear that this fraction has a high degree of contamination of peroxisomes (PEX1), endoplasmic reticulum (GRP-78), golgi (GGA-2) and mitochondria (VDAC). We, therefore, performed further purification of this re-suspended pellet with percoll gradient centrifugation. Lysosomes were retained within the bottom-third layer of this percoll gradient due to a different buoyancy to other organelles (Fig. 1C). Finally, after four to five rounds of high-speed centrifugation to wash off the percoll from the samples, the lysosomal fractions are highly enriched in lysosomes, with only trace levels of ER (GRP-78) and peroxisomes (PEX-1) (Fig. 1D). Note that the studies further described characterize the transport of arginine into these lysosomes (Fig. 1D). Examination of arginine transport into organelles present in the fraction depicted in Figure 1C, and notably the absence of lysosomes, does not show any inherent arginine transport activity. Thus, any minor organellar contamination of the lysosomal fraction is not contributing to arginine transport. Furthermore, the properties of lysosomes in regard to this isolation procedure and buoyancy for JNCL and control lysosomes are identical. Note that as this study centers on CLN3, we have confirmed the presence of CLN3 by incubating the lysosome-enriched fractions from controls or, JNCL patients homozygous for the 1.02 kb deletion with an antibody to CLN3 (Ab Q-438). We detect a specific band at approximately 55 kDa that is only present on controls but not on JNCL samples (Fig. 1E).
Defective lysosomal arginine transport in JNCL
Altered amino acid transport at the lysosome resulting in lysosomal dysfunction, and potentially resulting in disease and in particular lysosomal storage disorders have been previously reported (25). JNCL results from mutations in CLN3 (1–4). Previous studies have shown that CLN3 restores defective arginine transport in yeast strains lacking the CLN3-homolog, Btn1p. Therefore, we have compared amino acid transport of arginine into lysosomes isolated from lymphoblasts of patients with JNCL with age–gender-matched controls. Figure 2A demonstrates that arginine uptake into lysosomes isolated from lymphoblasts is linear over a period of 3 min. There is a clear deficiency in lysosomal arginine transport into JNCL lysosomes. In Figure 2Aa, we show clear differences in the rate of lysosomal arginine transport for lysosomes isolated from lymphoblasts of four different individuals confirmed to be homozygous for the 1.02 kb deletion in CLN3 or compound heterozygous for the 1.02 kb deletion plus a point mutation, when compared with their age–gender-matching lymphoblast controls (Table 1). The decrease in arginine transport is significant and is present whether the patient is a male (P
0.01) or a female (P0.001).
|
|
As lysosomal storage disorders likely share similar alterations of lysosomal function, we have examined arginine transport in isolated lysosomes from patients confirmed to have infantile-NCL (INCL), late infantile-NCL (LINCL) or Niemann pick disease type C (NPC). As evidenced in Figure 2B, transport of arginine into lysosomes isolated from INCL, LNCL and NPC lymphoblasts does not appear to be compromised when compared with normal age–gender-matching controls, suggesting that the decrease in lysosomal arginine transport is specific to JNCL.
Lysosomes from JNCL patients have been reported to have an increased intra-lysosomal pH (11), which could affect the pH gradient and/or the membrane potential across the lysosomal membrane. So, this change in 
pH could affect the net transport of other molecules into the lysosome. We have, therefore, examined the transport of radiolabeled glutamine, histidine, glycine and lysine into JNCL and matched control lysosomes. Although little is known about the lysosomal transport of histidine and glutamine, it is highly possible that due to its charge at physiologic pH, histidine is transported through the same lysosomal cationic transport system that transports lysine and arginine, known as system c (13–15). In contrast, we expect glutamine to be transported through lysosomal system d (26) and glycine to be transported through any of the small amino acid transporters reported previously (23,27). In Figure 3A, we show that JNCL and control lysosomal fractions show no difference in the transport of glycine, histidine or glutamine. Curiously, we see no difference in the transport of lysine which is also a substrate for the cationic amino acid lysosomal transporter, system c, suggesting that the function of this system may not be compromised in JNCL lysosomes.
|
Previous studies on lysosomes isolated from fibroblasts indicated that cationic amino acid transport maximizes as the pH gradient increases (14
). Maximum cationic amino acid transport was observed at cytosolic pH
7, suggesting that an adequate pH is required to maximize the lysosomal transport of cationic amino acids (14). Furthermore, it was shown that under physiologic conditions (cytosolic pH 7.0 and 37°C), the net flow of lysine in lysosomes is outward (13), whereas arginine net flow is inward (14). As JNCL lysosomes were able to efficiently transport lysine, we therefore decided to test lysine efflux in JNCL lysosomes. Figure 3B shows no difference in lysine efflux between lysosomal fractions isolated from JNCL lymphoblasts and their normal controls. Thus, only the net transport of arginine is affected in JNCL lysosomes. Altered lysosomal pH in JNCL may alter the lysosomal pH gradient, and this alteration in the pH may have an effect on the transport of arginine into JNCL lysosomes. However, it is unlikely that this decrease in arginine transport could be explained solely as a consequence of a generalized transport impairment, as amino acids that are substrates for the same transport system such as lysine, or amino acids transported through different lysosomal transport systems requiring pH gradients such as system d, or systems e/f do not show the same impairment. We have assayed the amino acid content of JNCL and control cells used to isolate lysosomes for these studies and demonstrate that intracellular arginine levels in lymphoblasts are decreased when compared with their normal controls (Fig. 3C). No other intracellular amino acid levels were found to be significantly affected (data not shown). This provides an important correlate that indicates that the disruption in lysosomal transport of arginine that we see could potentially translate into an overall alteration in the cellular availability of the semi-essential amino acid arginine. Despite the differences in arginine levels, when cytosolic pH was measured for JNCL cells compared with their corresponding controls, no significant differences were found (JNCL pH 7.84±0.14, controls pH 7.77±0.18).
Characterization of arginine transport in lymphoblast lysosomes
It has been previously reported that specific properties in lysosomal transport systems may differ depending on the characteristics of the cell type (18,28). Lysosomal transport of cationic amino acids was originally characterized in lysosomes isolated from fibroblasts, and it was shown to be actively mediated by a transporter designated system c (13–15). Overall, little is known about the transport of arginine and other amino acids into lysosomes. To further understand the apparent defect in arginine transport in lysosomes isolated from JNCL lymphoblasts, we thought it prudent to further characterize lysosomal transport of arginine in our normal cell lines. As a first step in this characterization, we measured the rate of arginine transport in lysosomes isolated from cells obtained from normal subjects. We have evaluated the kinetics of arginine transport and show in Figure 4A, an Eadie–Hofstee plot indicating that lysosomal arginine transport in lymphoblasts follows a first-order reaction. Moreover, the Km for arginine transport in our fractions is 0.34 mM, which is similar to the Km previously reported for fibroblast lysosomes (13,14). In Figure 4C, we show that arginine transport into the lysosome is ATP-dependent, as it is clear that upon exclusion of ATP we see no arginine transport. Furthermore, upon addition of only Mg2+ lysosomes are unable to transport arginine (data not shown). In addition, we show in Figure 4B that lysosomal arginine transport requires the activity of the vacuolar-type H+-ATPase (v-ATPase), as addition of the v-ATPase blocker bafilomycin A which completely blocked v-ATPase activity (not shown) clearly decreases arginine transport (P<0.001). However, this mode of inhibition does not completely ablate arginine transport, but interestingly decreases the level of arginine transport to a level similar to that observed in JNCL lysosomes.
|
To evaluate whether a decrease in v-ATPase activity could account for the decrease in arginine transport in JNCL lysosomes, we assayed v-ATPase activities in our JNCL lysosomal fractions and compared them with their matching controls. In Table 2, we demonstrate that JNCL lysosomes have a 69.8% decrease in the activity of the v-ATPase compared with their matching controls (P<0.001). A similar decrease (63.9%) was found when we tested v-ATPase activity in INCL which do not have defective arginine transport, when compared with their matched controls (P=0.008). Interestingly, v-ATPase activity in LINCL lysosomes which we also show to have normal arginine transport is significantly increased (P<0.001) when compared with controls. However, v-ATPase activity in NPC and their matching controls is similar. Taken together, we conclude that although NCL lysosomes have a decrease in v-ATPase activity, this decrease cannot specifically account for decreased arginine transport in JNCL lysosomes, and as noted earlier, inhibition of v-ATPase activity does not completely block arginine transport into lysosomes. Nevertheless, we aware that a decrease in the activity of the v-ATPase could exert some contribution to the decreased lysosomal arginine transport, although as transport of other amino acids is unaffected (see Fig. 3A) and this effect would likely be minor.
|
We have further investigated the effect of other cationic and neutral amino acids on arginine transport. In Figure 5A, we have pre-incubated lysosomes isolated from normal cells with cold glutamine, cold histidine or cold lysine, and assayed arginine transport. Except for lysine, which is a substrate for the cationic transporter (system c), these pre-incubations do not result in a significant decrease or increase in lysosomal arginine transport. Lysine, therefore, competitively inhibits arginine transport (P<0.001). Glutamine is not a substrate for system c, and is likely to be transported into the lysosome by another transport system such as system d. Histidine is considered a cationic amino acid similar to arginine and lysine. The fact that we do not see a significant decrease in lysosomal arginine transport upon pre-incubation with histidine, although we demonstrate in Figure 3A that histidine is transported into the lysosome, suggests that histidine may be transported through system c with a lower affinity than arginine and lysine; or that it may be transported into the lysosome through a different lysosomal transport system.
|
To further characterize the directionality of arginine transport in lysosomes isolated from normal lymphoblasts, we have examined arginine efflux in control samples. Figure 5B shows that arginine efflux in lysosomes isolated from normal lymphoblasts is linear over time, beginning after approximately 5 min and continuing for about 6–7 min. To test the effect of cationic, neutral and hydrophobic amino acids on
pH and consequently on the lysosomal efflux rate of arginine, normal lysosomes pre-loaded with 14C-arginine were incubated with lysine, histidine, glutamine and glycine. In Figure 5B, we show that when lysosomes are incubated with any of these amino acids, changes toward a decrease in lysosomal arginine efflux are detected. Therefore, changes in membrane potential or changes in the proton gradient resulting from the transport of these other amino acids into the lysosome could lead to changes in net transport of other molecules such as arginine. Importantly, our studies on JNCL lysosomes did not reveal any differences in the transport of these molecules, so although arginine transport has the potential to be modulated in this way, we think it is unlikely to be the cause of altered arginine transport in JNCL. However, future studies on the lysosomal membrane potential would be more definitive. Note that efflux studies could not be performed on JNCL lysosomes, as their decreased transport of arginine does not allow sufficient pre-loading of the lysosome with this amino acid.
To further characterize lysosomal arginine transport, we have tested the effect that anions and cations could exert on arginine transport. Lysosomes isolated from normal cells were pre-incubated with different monovalent, divalent cations or chloride, and lysosomal arginine transport measured. In Figure 5C, we show that pre-incubation with NaCl, KCl and CaCl2 results in an increase in lysosomal transport of arginine (P0.001). As the presence of chloride could also affect the pH and explain the increase in lysosomal arginine transport, we tested arginine transport after pre-incubating with Na gluconate or LiCl, to test whether chloride is the molecule modulating lysosomal arginine transport. Figure 5C shows that it is cations rather than anions exerting an increase in lysosomal arginine transport, as a similar increase to the one exerted by NaCl is seen when we use Na gluconate (P<0.001), whereas we did not see any changes in arginine transport when the lysosomal fractions were pre-incubated with LiCl. This observation rules out the possibility that increases in lysosomal arginine transport could be mediated by chloride. Therefore, lysosomal arginine transport is likely mediated in a cation-dependent manner, similar to the reported divalent cation effect that Mg2+, Mn2+, KCl and Ca2+ exert on the lysosomal transport of cystine in lysosome granular fractions isolated from leucocytes and liver cells (29–31).
Blocking CLN3 function inhibits lysosomal arginine transport
CLN3 is a transmembrane protein localized to the late endosome/lysosome (32,33). To assess the role that CLN3 may play in lysosomal arginine transport, we have pre-incubated isolated lysosomes from normal lymphoblasts with proteinase K, an endolytic protease that cleaves peptide bonds at the carboxylic sides of aliphatic, aromatic or hydrophobic amino acids. Upon exposure to a low concentration of proteinase K, degradation of only the exposed portion of lysosomal transmembrane proteins will occur, altering not only the function of these proteins, but also damaging the membrane gradient required for amino acid transport. Lysosomes treated in this way show a decrease in the ability to transport arginine (Fig. 6A).
|
A remaining question is whether CLN3 is involved in the transport of arginine into the lysosome. To address this, we have pre-incubated lysosomes from normal lymphoblasts with an antibody targeted to the predicted cytosolic region, amino acids 250–264, QPLITRTEAPESKPGS (Ab Q-438) of human CLN3. Note that this antibody has been reported to react with the CLN3 protein (34
–36). Upon incubation of the lysosome-enriched fractions from controls or from JNCL patients homozygous for the 1.02 kb deletion with Ab-438Q, we detect a specific band at approximately 55 kDa that is only present on controls but not on JNCL samples, strongly supporting the specificity of this antibody to CLN3 (Fig. 1E). The region of CLN3 that Ab Q-438 recognizes has been experimentally shown to be on the cytoplasmic side of the lysosomal membrane (reviewed in 5). The cytoplasmic-facing portion of CLN3 should, therefore, be accessible to being bound by an antibody directed to this region, and upon binding it is expected to disrupt the normal function of CLN3. In Figure 6B, we show that arginine transport is decreased upon pre-incubation with Ab Q-438. Furthermore, the use of anti-rabbit antibody as a negative control shows no effect on arginine transport, and heat denaturation of Ab Q-438 results in a loss of the blocking effect that Ab Q-438 has on arginine transport activity in lysosomes (Fig. 6B). Negative controls for this experiment included pre-incubation of lysosome-enriched fractions with anti-rabbit IgG (Fig. 6B) which had no effect on transport. It is noteworthy that lysine transport was unaffected by pre-incubation with Ab-Q438 (data not shown).
Restoration of lysosomal arginine transport by expression of CLN3
JNCL cells were permanently transfected with a lentiviral vector harboring either hCLN3 followed by an IRES-EGFP to act as a reporter (JNCL+hCLN3), or a lentiviral vector harboring only the EGFP (JNCL+GFP) and used as a negative control. Upon antibiotic selection and corroboration of the expression of the reporter, more than 90% of the transfected cells permanently express the GFP reporter, suggesting that the rate of transfection is extremely efficient and stable throughout time. Lysosomes were then isolated and 14C-arginine transport was assayed as described previously.
Results clearly show that upon expression of CLN3 in JNCL cells, arginine transport into the lysosomes is restored back to the levels previously reported for controls (Fig. 6C). Furthermore, when 14C-arginine transport in lysosomes isolated from JNCL+GFP is assayed, transport levels remain similar to those previously reported for JNCL cells, suggesting that the increase in arginine transport is due to the re-expression of a functional CLN3 instead of an effect of the vector.
We show in Table 2 that upon measuring v-ATPase activity in JNCL+hCLN3 samples, the activity is similar to control levels, suggesting that upon re-expression of CLN3, JNCL cells also regain normal v-ATPase activity. As expected, when the v-ATPase activity of the JNCL+GFP lysosome samples was measured, rates were similar to the ones reported for JNCL lysosomes, suggesting again that restoration of v-ATPase activity in JNCL+hCLN3 lysosomes is due to the expression of a functional CLN3.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Characteristics of transporter proteins at the lysosomal membrane have been described using lysosomal fractions from human cells isolated by isopyknic fractionation. We have, therefore, utilized isopyknic fractionation to isolate lysosomes and show that only lysosomes, not other organelles, account for the transport of the substrates.
Many biochemical and physiological studies have been initiated to determine the characteristic cellular defect underlying JNCL, although only a few traits have been truly associated with the disease. In summary, JNCL, as other NCLs, is characterized by the accumulation of autofluorescent hydrophobic material in the cytoplasm of neurons, and to a lesser extent many other cell types. More specifically, these deposits have been localized to lysosomes (37,38). The structure of this electron-dense storage material varies by the type of NCL, but is described as being curvilinear or fingerprint-like in both late infantile (CLN2) and juvenile (CLN3) forms of NCL (39). Protein sequencing and immunological studies have revealed that subunit c of the mitochondrial ATP synthase is the major component of the storage material in late infantile and juvenile NCLs (40,41). Biochemical and molecular approaches have demonstrated that accumulation of mitochondrial ATP synthase subunit c is not a result of increased expression of the P1 and P2 nuclear genes that encode the protein, nor does the stored protein have a different encoded sequence from that for normal (42,43). In JNCL fibroblasts, although initially located in the mitochondria, mitochondrial ATP synthase subunit c accumulated in lysosomes, whereas the degradation of another inner mitochondrial membrane protein, cytochrome oxidase subunit IV, was unaffected, with no lysosomal accumulation (44,45). It is, therefore, presumed that a defect in, or the absence of CLN3, perturbs lysosomal function, such that degradation of subunit c is compromised.
We report that lysosomal arginine transport is ATP-, vATPase- and cation-dependent. Moreover, we demonstrate that JNCL lysosomes have a defect in this transport and that JNCL cells have a depletion in arginine levels. It is important to recognize that despite this defect, residual transport of arginine into the lysosome remains, and that while arginine levels are depleted, arginine albeit at decreased levels, is still present. Thus, a complete metabolic defect in arginine utilization or synthesis in JNCL is unlikely. Arginine is considered an essential amino acid in newborns, and is semi-essential in adults due to limited de novo synthesis. Therefore, it is possible that irrespective of the effect of CLN3 mutation on arginine in cells, there is an over-riding requirement to maintain basal levels of arginine for adequate cell function (46–49).
An intriguing question is what, if any, is the effect of altered arginine levels in JNCL? We show that arginine levels are decreased, suggesting that there may be a compromise in intracellular levels of arginine. Arginine depletion may impact certain metabolic pathways for which it is essential. Arginine is the main substrate for the production of nitric oxide (NO) (50–52), a small radical that has been implicated in numerous mechanisms that regulate cell function and immune response. Altered regulation in the production of NO has been linked to severe alterations in cell viability (48,49). Nitric oxide is also important for vasodilation, and is considered a potent neuromodulator and neurotransmitter (49,51). It is well documented that JNCL selectively affects the central nervous system (CNS). Arginine levels in the CNS depend entirely on the transport of this amino acid across the blood–brain barrier through the well-characterized cationic-transport-denominated system Y+ (51). Therefore, the amount of NO production in the CNS depends on the amount of extracellular arginine that is transported actively across the blood–brain barrier (51). Nitric oxide production depends also on the availability and effectiveness of astrocytes to store and transport the cationic amino acid into neurons (51,53). Interestingly, transgenic mice lacking the major plasma membrane cationic transporter Y+ (CAT2) have decreased arginine transport into astrocytes, and decreased production of NO (54). In addition, studies have shown that animals exposed to high diets of arginine or with increased levels of brain amino acids such as arginine have increased susceptibility to seizure induction (55–59). Other reports have described the correlation between decreases in NO production and seizure susceptibility (59–62). These contradictory findings have been the focus of debate, and it has been suggested that imbalances in NO production and availability, whether there is a decrease or increase, could cause severe consequences in the nervous system. Thus, it is tempting to suggest that as JNCL cells have a decrease in arginine availability, that patients could also have an imbalance in the arginine/NO pathway. Intriguingly, a cln3-knockout mouse model for juvenile NCL has been shown to have age-dependent variations in blood serum arginine levels (63). It is important to further explore what effect—if any—CLN3 would have on arginine transport in the whole cell at the plasma membrane, as regulatory mechanisms governing transport at this membrane could also be affected in the absence of a functional CLN3. In summary, any imbalance of arginine levels in the CNS would have a direct effect on the production and availability of NO, thus increasing the susceptibility to seizures. This is particularly important if we consider that epilepsy is one of the clinical hallmarks of JNCL.
The question remains, what does altered lysosomal arginine transport tell us about JNCL? Arginine, being an amino acid, is an organic cation. Thus, decreased arginine levels in JNCL cells and altered lysosomal transport of arginine have the potential to alter cationic balance. We present data indicating that transport of other amino acids, which is likely facilitated by different transport proteins, is unaffected in lysosomes isolated from JNCL cells. This suggests that altered arginine levels may not be sufficient to alter either the pH or the electrochemical potential that drives transport of these other amino acids. A possible role for CLN3 in homeostatic or osmotic balance of the lysosome was implied by a previous study that indicated that sucrose-induced vacuolization of human fibroblasts results in increased expression of several lysosomal gene products, including CLN3 (64). Moreover, studies on the yeast lacking the homolog to CLN3, Btn1p, have suggested that alterations in intracellular arginine concentration may underlie a mechanism to maintain cationic balance (65). An imbalance of arginine resulting from a defect in CLN3 accompanied by a decreased v-ATPase activity could alter conditions in the lysosome that result in the characteristic accumulation of autofluorescent storage material in the lysosomes of JNCL cells. Moreover, altered lysosomal arginine transport may have the potential to affect the production and net flow of other molecules into and out of the lysosome that we have not investigated in this study. Therefore, further experimentation is required to evaluate whether or not the transport of other molecules into or out of the lysosome is compromised by mutation of CLN3. However, altered lysosomal arginine transport in JNCL cells represents a bona fide biochemical abnormality that can be associated to JNCL.
The remaining question is what is the role of CLN3 in lysosomal arginine transport? Considerably more evidence is required to demonstrate that CLN3 directly mediates lysosomal arginine transport, as its sequence and topology provide limited support for it being a transport protein (5). This study and others have characterized the transport of arginine and lysine into the lysosome by the same system (system c). We show that only arginine and not lysine transport is affected in lysosomes isolated from JNCL lymphoblasts, indicating that system c function may not be compromised. A simple explanation would be that as endogenous levels of arginine appear to be decreased in JNCL cells, this results in the downregulation of the transport system. However, the ability to transport lysine by the same transport system is unaffected. Therefore, we conclude that there is something specific about arginine transport as opposed to arginine and lysine transport in lysosomes that remains to be uncovered, and that CLN3 may play a central role in this difference. CLN3 could specifically regulate substrate specificity of system c, or proteins involved in regulating solute transport into the lysosome. Further studies on other solutes that would be transported into the lysosome are underway. An alternative explanation would be that the CLN3-defect specifically results in a decrease in arginine turnover, which consequently manifests as decreased arginine levels, and thus decreased lysosomal arginine transport. However, there have been no previous reports of altered arginine metabolism in juvenile Batten disease to support this notion.
As an antibody to CLN3 was able to block arginine transport, but not lysine transport, CLN3 does appear to mediate a direct effect on arginine transport, perhaps by interacting with a protein or proteins involved directly in arginine transport, the regulation of arginine transport or delivery of the transporter to the lysosomal membrane. This is further supported by the restoration of arginine transport into JNCL lysosomes upon expression of CLN3.
A final question is whether a connection exists between CLN3-dependent lysosomal arginine transport and v-ATPase activity? We show a decrease in v-ATPase activity in JNCL lysosomes that can be corrected upon permanent expression of CLN3. v-ATPases are versatile and important proton pumps (66–68). These pumps are thought to regulate and maintain intra-luminal acidification of lysosomes, which are required for among other things, optimal conditions for proteolytic activity for lysosomal protein turnover. JNCL lysosomes have been shown to have increased intra-lysosomal pH, that could be caused in part, by the decrease in the v-ATPase activity that we report. Further experiments will be required to demonstrate whether CLN3 exerts a direct influence on the v-ATPase, or indeed the electrochemical gradient at the lysosomal membrane. In the CNS, the v-ATPase is necessary for maintaining the electrochemical gradient required for proper function of catecholamine transporters in synaptic vesicles, and uncoupling of this gradient has been shown to induce parkinsonism (66–70). Thus, if the altered activity in v-ATPase activity we see in JNCL lymphoblasts occurs in the CNS, this could represent a contributing factor to the parkinsonism associated to mid- to late-stages of this disease (71). Again, more detailed studies on the effect of altered CLN3 function in the CNS, and in particular on all neurotransmitter systems are required.
A key to pinpointing CLN3 function may be to determine the role that altered arginine levels have on the lysosome and indeed the cell. As CLN3 seems to affect lysosomal arginine transport and arginine levels, rather than those for lysine, understanding any imbalance of arginine in NO production, or perhaps how arginine acts as an organic cation and whether the v-ATPase has a functional link with CLN3 needs to be further explored in JNCL.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Cell cultures
Blood samples from a child confirmed to have LINCL, as well as blood samples from JNCL patients homozygous for the 1.02 kb deletion or heterozygous for a point mutation and the 1.02 kb deletion were collected, lymphocytes isolated using Ficoll-Paque and transformed following a protocol established previously (72
). B95-8 lymphoblasts, PPT1 lymphoblasts, NPC lymphoblasts and control lymphoblasts cell lines were obtained from Coriell Cell Repositories. Control lymphoblasts were selected to match gender, age and split ratio for each disease cell line (Table 1). Cells were grown on 75 cm2 flasks (Corning 430641) using 1640 RPMI medium (Gibco 61870-036) plus 15% FBS (Gibco 26140-079) and antibiotic–antimycotic (Gibco 15240-062) at 37°C and 5% CO2. Cells were split every 3–4 days once log phase was reached, and re-seeded at a density of 300 000 cells/ml. A total of 120 flasks were required to reach the suggested cell density of 5x109 (16).
Generation of lentiviral vector
A p4-ENTR vector (Invitrogen 11818-010) containing either CLN3 and IRES2-EGFP sequences (JNCL+hCLN3) or only the EGFP sequence as negative control (JNCL+GFP) were constructed. LR re-combination reactions using the pLenti6/UbC//V5-DEST vector (Invitrogen V499-10), viral packaging using 293FT cells and titering of the full lentiviral vector were performed using Invitrogen Gateway System and ViraPower Lentiviral Expression System manual instructions. Presence of GFP and CLN3 was confirmed by PCR, and correct insertion of the clone was further confirmed by sequencing analysis. Results obtained from sequencing were blasted using PubMed-Entrez showing 100% alignment with CLN3. Lentiviral transduction and expression efficiency as well as titering were performed following Invitrogen specifications and using HT-1080 cells (ATCC CCL-121). Packed virus was concentrated by ultracentrifugation (20 000g for 2 h at 4°C) using centricon filters (Millipore YM-50000) as previously described (73).
Transfection of JNCL cells
MOIs between 2 and 4 were used to transfect JNCL cells, and were suitable enough to efficiently transfect EBV-transformed B lymphocytes. Cells were transfected in 48-well plates following Virapower Lentiviral Expression System's instructions. Clones were selected by antibiotic resistance plus expression of the GFP reporter. Two to three weeks after transfection, expression of the reporter gene showed that >90% of the cells were permanently transfected. Lysosomes were isolated following a protocol previously described (see below).
Intracellular pH measurement
Cytosolic pH in lymphoblasts was measured following the protocol previously described by Ng et al. (74–76). Briefly 1x107 cells at log phase were harvested by centrifugation at 200g and 4°C for 10 min. Cells were washed twice with HBSS buffer, centrifuged and re-suspended in 1 ml of RPMI media containing 5 µM of the pH fluorescent dye BCECF-AM (Sigma-Aldrich 14562) and incubated at 37°C 5% CO2 for 30 min. Cells were then harvested and re-suspended in 1 ml of a buffer containing 140 mM NaCl, 5 mM KCl, 0.8 mM MgSO4, 1.8 mM CaCl2, 5 mM glucose and 15 mM HEPES, pH 7.4, without BCECF-AM for 10 min at 37°C. The resting intracellular pH (pHi) was determined by aliquoting 200 µl of the sample into a 96-well plate and measuring the fluorescence of the intracellular dye using a SpectraMax M5 multi-detection microplate reader at 37°C (Molecular Devices Corp., Sunnyvale, CA, USA), with dual excitation wavelengths at 439 and 500 nm and an emission wavelength of 530 nm. Fluorescence ratios were calculated as previously described (74) and calibration was performed by re-suspending cells in KCl-nigericin buffer at pH 6–8 (76). Experiments were performed in triplicates for each JNCL cell line and the corresponding controls were described in Table 1.
Lysosome isolation
Lysosome-enriched fractions from disease and control cells were purified following the protocol previously described by Harms et al. (16) with slight modifications. The final pellet was re-suspended in 2 ml of 1x isotonic isolation buffer pH 7.0. This fraction was considered highly enriched in lysosomes and different from the established protocol; it was not further purified by free flow electrophoresis. Enriched fractions of isolated lysosomes were either used immediately for uptake experiments or stored at –80°C overnight. To assay purity of the different lysosomal fractions, western blots using antibodies raised against peroxisomes (PEX-1), lysosomes (LAMP-1), endoplasmic reticulum (GRP-78), golgi (GGA-2) and mitochondria (VDAC) were performed. To wash out the percoll from the top layer, which is expected to retain most of the non-lysosomal organelles, we did sequential centrifugations and washes with isotonic buffer identical to those performed with lysosomal fractions, obtaining similarly after four to five washes, a pellet was re-suspended in 1x isotonic isolation buffer of pH 7.0. Protein concentration for lysosomal and non-lysosomal fractions was determined by Bradford Assay using a BioRad spectrophotometer and BSA as standards.
Uptake experiments
14C-radiolabeled amino acids were used throughout the different experiments, and were obtained from Amersham Pharmacia. For each experiment, 32.5 µg of total protein from isolated lysosomes was mixed with isotonic uptake buffer (0.25 M sucrose, 5 mM MgCl2, 20 mM HEPES, pH 7.0) using a 1:2 v/v ratio, and kept at 4°C in an eppendorf tube. Upon the addition of 2 mM ATP, a blank sample was taken and washed twice with 5 ml of ice-cold 1x PBS. The mix was then incubated for 1 min at 37°C to allow the reaction to occur. After this period, 12.5 µCi/ml of the radiolabeled amino acid was added and aliquots were taken at time 0 and 30 s increments for 3 min, and collected using GF-C Whatman filters in a vacuum manifold (Millipore, Bedford, MA, USA). The reaction was stopped by two 5 ml washes of ice-cold 1x PBS. The radioactivity retained on the filter was quantified using a Beckmann Scintillation counter.
Background or non-specific radioactivity was determined by performing the same experiment at 4°C, and results obtained at 4°C were subtracted from those obtained at 37°C. Longer uptake periods were used initially, but we found that at this concentration, the linear uptake occurs in the first 3 min of the reaction. To calculate kinetics, different concentrations of 14C-arginine were used to measure arginine lysosomal transport, and the data were plotted in an Eadie–Hofstee plot to calculate the ordinate intercept (Vmax) and the slope (–Km).
Efflux experiments
Efflux experiments were performed following the protocol previously described by Steinherz et al. (17) with slight modifications. Briefly, 32.5 µg of total protein from isolated lysosomes was mixed with isotonic uptake buffer, followed by the addition of 12.5 µCi/ml of 14C-radiolabeled amino acid. The mixture was incubated initially for different periods of time at 37 C to determine the appropriate pre-loading time. Once incubation time was due, isotonic efflux buffer (0.25 M sucrose, 5 mM MgCl2, 20 mM HEPES, 2 mM ATP, pH 7.0) was added to the uptake mixture using a 50:1 v/v ratio. Aliquots were taken at efflux time 0 and 30 s intervals for 10 min; however, no differences in the efflux of arginine or lysine were detected after 6–7 min. For efflux experiments using cold amino acids, 5 mM of the cold amino acid re-suspended in isotonic uptake buffer was added to the uptake mixture and the efflux was then measured as described earlier. The addition of cold amino acid into the isotonic efflux buffer (5 mM) did not cause a significant change in the pH of the buffer.
Competitive inhibition experiments
Cold amino acids of 0.5 mM, 5 mM and 50 mM (histidine, glutamine or lysine) or different concentrations of a specific cation (NaCl, KCl, CaCl2, LiCl or Na gluconate) were re-suspended in isotonic uptake buffer, pH 7.0, and added concomitantly with the addition of ATP. After addition, uptake was performed as described earlier. Uptake rates are presented in Results using specific concentrations of amino acids or cations.
v-ATPase activity
Lysosomal v-ATPase activity was assayed by following the release of Pi. The reaction was performed as follows. Lysosome-enriched fractions (32.5 µg total protein re-suspended n isotonic buffer) were mixed with 2 mM ATP, 3 vol of Buffer A (100 mM MES–Tris buffer, 80 mM KCl, 6 mM MgCl2, 150 mM NaCl, pH 7.0) and incubated at 30 or 37°C for 5 min. After incubation, the reaction was started upon the addition of 2 mM of ATP and the mixture incubated for 20–25 min at 30 or 37°C, respectively. The ATPase reaction was stopped after 25 min by the addition of 2 ml of molybdate solution (2% v/v H2SO4, 0.5% w/v ammonium molybdate and 0.5% w/v SDS) and 0.2 ml of ascorbic acid (2% w/v). The color was allowed to develop over 5 min at room temperature. The absorbance was measured at 750 nm and the value related to a Pi concentration curve using standard solutions of KH2PO4.
Intracellular amino acid concentration
3x106 lymphoblasts from control or JNCL patients were washed twice with PBS following heat denaturation at 100°C for 10 min. Lysed cells were then washed with 1x potassium phosphate buffer (PPB) pH 7.3, and centrifuged twice at 10 000 g for 5 min. Supernatant was collected and lyophilized using a SPD SpeedVac. Pellet was re-suspended in 20 µl of 0.1 M phosphoric acid and homoserine was added as an internal control. Amino acid concentrations were then measured using HPLC and following the manufacturer's protocol (ESA, Inc., Boston, MA, USA).
Arginine transport blocking experiments
Proteinase K
Proteinase K (Sigma P-2308) at a final concentration of 0.1 µg/ml was added concomitantly with ATP, and samples were incubated on ice for 30 min. The sample was then incubated for 1 min at 37°C and uptake experiments were performed as previously described.
Bafilomycin A1
To block the activity of v-ATPase, isotonic uptake buffer containing 5 nM of Bafilomycin-A1 (Sigma B-1793) was used. The samples were then incubated on ice for 90 s, warmed at 37°C for 1 min followed by uptake protocol.
CLN3 antibody blocking
An antibody recognizing the predicted cytoplasmic loop of CLN3 (Ab Q-438) kindly donated by Dr Michael J. Bennett was used. This antibody was raised against a specific peptide sequence (QPLITRTEAPESKPGS) that is predicted to be part of a cytosolic region of CLN3 (34–36). The blocking experiments were performed adding either Ab Q-438 or Anti-Rabbit IgG antibody used as negative control (Amersham Pharmacia, Piscataway, NJ, USA) at a 1:1000 dilution to the uptake buffer. Incubation was performed at room temperature for different time points, and after the incubation time, uptake experiments were performed as described. Further experiments were performed incubating for 10 min.
For denaturing experiments, Ab Q-438 antibody (1:100) was heat-denatured at 100°C for 30 min. Following denaturing, the sample was cooled down and added to uptake buffer at 4°C (1:1000). Uptake experiments were then performed as usual.
Western blot
To show specificity of the antibody to CLN3, 50 µg total protein from control or JNCL lysosome-enriched fractions were ran on a 10% SDS and western blot was performed. Overnight incubation of Ab Q-438 (1:1000) was followed by 1 h incubation with 1:2500 HRP anti-rabbit secondary antibody. Samples were normalized to ß-actin content. Image shown is representative of results obtained from four different controls and four different JNCL samples homozygous for the 1.02 kb deletion.
Statistical analysis
Results comparing three or more samples were analyzed using ANOVA and Bonferroni tests. Results comparing two samples were analyzed by t-test. Only P-values <0.05 were considered significant. Statistical analysis was performed using SigmaPlot and GraphPad, and corroborated using SPSS and JMP statistical programs.
|
ACKNOWLEDGEMENTS |
|---|
We wish to thank the BDSRA and the families that kindly donated blood for the generation of the cell lines used in this study. Thanks to Paul G. Rothberg for genotyping of cell lines. This work was supported by NIH grant R01 NS36610 and the Luke and Rachel Batten Disease Foundation.
Conflict of Interest statement. None declared.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
- Goebel, H.H. (1995) The neuronal ceroid lipofuscinoses. J. Chil. Neurol., 10, 424–437.
- International Batten Disease Consortium (1995) Isolation of a novel gene underlying Batten disease. Cell, 82, 949–957.[CrossRef][ISI][Medline]
- Mole, S.E. (2004) The genetic spectrum of human neuronal ceroid-lipofuscinoses. Brain Pathol.,14, 70–76.[ISI][Medline]
- Munroe, P.B., Mitchison, H.M., O'Rawe, A.M., Anderson, J.W., Boustany, R.M., Lerner, T.J., Taschner, P.E., de Vos, N., Breuning, M.H., Gardiner, R.M. and Mole, S.E. (1997) Spectrum of mutations in the Batten disease gene, CLN3. Am. J. Hum. Gen., 61, 310–316.[ISI][Medline]
- Phillips, S.N., Benedict, J.W., Wiemer, J.M. and Pearce, D.A. (2005) CLN3, the protein associated with Batten disease: structure, function and localization. J. Neurosci. Res., 79, 573–583.[CrossRef][ISI][Medline]
- Taschner, P.E., de Vos, N. and Breuning, M.H. (1997) Cross-species homology of the CLN3 gene. Neuropediatrics, 28, 18–20.[ISI][Medline]
- Pearce, D.A. and Sherman, F. (1997) BTN1, a yeast gene corresponding to the human gene responsible for Batten's disease, is not essential for viability, mitochondrial function, or degradation of mitochondrial ATP synthase. Yeast, 13, 691–697.[CrossRef][ISI][Medline]
- Croopnick, J.B., Choi, H.C. and Mueller, D.M. (1998) The subcellular location of the yeast Saccharomyces cerevisiae homologue of the protein defective in the juvenile form of Batten disease. Biochem. Biophys. Res. Commun., 250, 335–341.[CrossRef][ISI][Medline]
- Pearce, D.A., Ferea, T., Nosel, S.A., Das, B. and Sherman, F. (1999) Action of Btn1p, the yeast orthologue of the gene mutated in Batten disease. Nat. Genet., 22, 55–58.[CrossRef][ISI][Medline]
- Golabek, A.A., Kida, E., Walus, M., Kaczmarski, W., Michalewski, M. and Wisniewski, K.E. (2000) CLN3 protein regulates lysosomal pH and alters intracellular processing of Alzheimer's amyloid–beta protein precursor and cathepsin D in human cells. Mol. Gen. Metab., 70, 203–213.[CrossRef][ISI][Medline]