Human Molecular Genetics, 2001, Vol. 10, No. 13 1431-1439
© 2001 Oxford University Press
Biochemical analysis of mutations in palmitoyl-protein thioesterase causing infantile and late-onset forms of neuronal ceroid lipofuscinosis
Department of Internal Medicine and the Hamon Center for Therapeutic Oncology Research, University of Texas Southwestern Medical Center, 5323 Harry Hines Boulevard, Dallas, TX 75390-8593, USA
Received March 12, 2001; Revised and Accepted April 26, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Deficiency in a recently characterized lysosomal enzyme, palmitoyl-protein thioesterase (PPT), leads to a severe neurodegenerative disorder of children, infantile neuronal ceroid lipofuscinosis (NCL). Over 36 different mutations in the PPT gene have been described, and missense mutations have been interpreted in the light of the recently solved X-ray crystallographic structure of PPT. In the current study, we assessed the biochemical impact of mutations through the study of cells derived from patients and from the expression of recombinant PPT enzymes in COS and Sf9 cells. All missense mutations associated with infantile NCL showed no residual enzyme activity, whereas mutations associated with late-onset phenotypes showed up to 2.15% residual activity. Two mutations increased the Km of the enzyme for palmitoylated substrates and were located in positions that would distort the palmitate-binding pocket. An initiator methionine mutation (ATG
ATA) in two late-onset patients was expressed at a significant level in COS cells, suggesting that the ATA codon may be utilized to a clinically important extent in vivo. The most common PPT nonsense mutation, R151X, was associated with an absence of PPT mRNA. Mannose 6-phosphate modification of wild-type and mutant PPT enzymes was grossly normal at the level of the phosphotransferase reaction. However, mutant PPT enzymes did not bind to mannose 6-phosphate receptors in a blotting assay. This observation was related to the failure of the mutant expressed enzymes to gain access to ‘uncovering enzyme’ (N-acetylglucosamine-1-phosphodiester 
-N-acetyl glucosaminidase), presumably due to a block in transit out of the endoplasmic reticulum, where mutant enzymes are degraded. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
The neuronal ceroid lipofuscinoses (NCLs) are a group of inherited neurodegenerative disorders of children characterized by cognitive and motor deterioration, visual failure and seizures, accompanied by the accumulation of autofluorescent storage material in the brain and other tissues (1). Three major forms of NCL (infantile, late infantile and juvenile onset) are caused by mutations in the CLN1, CLN2 and CLN3 genes, respectively. The infantile form of NCL is caused by mutations in the CLN1/PPT gene, which encodes a lysosomal palmitoyl-protein thioesterase (PPT) (2). PPT removes the fatty acid palmitate from cysteine residues in post-translationally lipid-modified proteins (3,4). Over 36 different mutations in the PPT gene causing NCL have been described (2,5–9), and these include 14 missense mutations, several of which are associated with late-onset disease.
Infantile NCL is an autosomal recessive disorder, implying that the disease results from a simple deficiency of PPT enzymatic function, owing to defects in both maternal and paternal alleles. In this model, each allele contributes to the phenotype in an independent fashion, and the inheritance of one completely defective allele with one wild-type allele leads to an asymptomatic carrier state. The genotype/phenotype data in relation to PPT deficiency are in excellent agreement with this model, in that all probands inheriting any combination of nonsense or frameshift alleles have a classically severe disease, with onset <2 years of age and a rapidly progressive course (6,8–10). The recently solved X-ray crystal structure provides a clear explanation as to why nonsense and frameshift alleles lead to complete loss of PPT function (11). PPT is a classical globular monomeric enzyme which possesses an /ß hydrolase fold reminiscent of most lipases and a catalytic triad consisting of serine 115, aspartic acid 233 and histidine 289 (11). Of note, His289 lies very close to the C-terminus of the protein; consequently, frameshift and the majority of nonsense mutations cause this important residue to be deleted. In addition, mutation of an extreme C-terminal leucine (Leu305Gly) also leads to complete loss of enzyme activity (J.-Y. Lu, unpublished data). This is explained by stabilizing contacts of the leucine residue with other amino acids within the three-dimensional structure of the protein (11). Therefore, nonsense and frameshift mutations may be reliably predicted to lead to total loss of enzymatic activity. The clinical and enzymological data are consistent with this prediction (6).
The most common mutation in PPT worldwide is the R122W substitution prevalent in the Finnish population, which has a high carrier frequency for this allele of 1 in 70 (2). Outside of Finland, a nonsense mutation (R151X) accounts for 40% of disease alleles. A common variant of NCL, termed juvenile onset NCL with granular osmiophilic deposits, is commonly associated with a single missense mutation, T75P (5,10). The location of the known missense mutations within the three-dimensional structure of PPT has been used to predict effects on enzyme activity (11), but direct experimental data have been lacking.
In the current study, we have chosen the most common nonsense mutation and a number of missense mutations in PPT for further analysis. We find that the common R151X mutation is associated with an absence of mRNA, and that late-onset mutations are associated with alleles that possess either a decreased Vmax (T75P, D79G) or altered Km for palmitate (Q177E, G250V). Severely reduced levels of immunoreactive protein, consistent with misfolding and endoplasmic reticulum (ER)-mediated protein degradation, was consistently observed and constitutes the major defect in PPT missense alleles.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Kinetic properties of mutant PPT enzymes
Table 1 is a summary of the properties of wild-type and mutant expressed PPT enzymes which were analyzed both in transient transfections of COS cells and after purification of their histidine-tagged forms expressed in Sf9 insect cells. In each instance, the data were normalized for expression of the PPT protein by immunoblotting and densitometric scanning, and the kinetic values obtained agreed well with those obtained for mammalian PPT purified from bovine brain (3,4). Five of the missense mutations [H39Q, G42E, R122W (the Finnish mutation), V181M and E184K] have no reliably detectable intrinsic enzymatic activity (<2% in this assay). These five mutations are associated with the most severe (infantile) form of NCL when present in the homozygous state or when co-inherited with the common R151X nonsense or R122W missense alleles (6). Several mutations have been associated repeatedly with late-onset phenotypes, and these were the only mutations associated with detectable expressed enzyme activity. The Q117E allele, with 7.3% of normal activity in this assay, is associated with disease onset at 3 years, whereas the T75P allele, with 6.8% of wild-type activity, is associated with onset at 5–9 years. Two other mutations (D79G and G250V) have been reported in isolated individuals and are also associated with a late-onset and protracted course.
|
In the case of two additional patients with late-onset phenotypes, the underlying genotype could only be interpreted in the context of our expression studies. These patients each inherited an initiation site mutation, ATG
ATA, together with private mutations (Y109D and Y247H). Interestingly, both private mutations showed no intrinsic enzymatic activity (Table 1), whereas, unexpectedly, a plasmid bearing the M1I mutation in pCMV5 expressed fully active enzyme, albeit at an expression level of only 10.8 ± 1.5% (Fig. 1). These findings suggest that the ATA codon, in the context of the normal PPT gene, may permit a low but significant level of translation initiation and provide a clinically important amount of enzyme.
|
A more detailed kinetic analysis was performed on highly purified mutant PPT enzymes derived through baculovirus expression in Sf9 cells (Table 1). The mutants Q177E and G250V each revealed substrate binding defects, as reflected in a 4- to 5-fold higher Km for both palmitoyl CoA and S-palmitoyl-thioglucoside. These observations are in excellent agreement with the insights gained from the previously described X-ray crystallographic structure of PPT (11). For example, the glutamine at position 177 was shown to make important hydrogen bond contacts with alanines at positions 171 and 183, which would be lost upon replacement of the glutamine by glutamic acid. Since Ala171 and Ala183 both make hydrophobic contact with the palmitate, loss of these contacts would be expected to lessen the affinity of the enzyme for its substrate. In a similar vein, glycine 250 is located in a ß-turn between ßa and ßb and has a positive
angle, so mutations to any other amino acid were expected to disrupt the antiparallel ß-motif. This disruption would lead to distortion of the palmitate-binding pocket and similarly result in a reduced substrate affinity.
Two additional late-onset mutations, T75P and D79G, showed no change in Km and only a slightly reduced Vmax to 
50% of normal. Both of these mutations are located on the helix 1. In the case of T75P, the substitution of a proline for threonine at the start of the helix was expected to disturb the geometry of this region, whereas in the case of D79G, the Asp79 side chain is hydrogen-bonded to Cys45 and Ile72. Loss of these contacts on mutation to Gly would lead to increased flexibility in this region. In light of the relatively minor changes in kinetic parameters associated with these two late-onset mutations, it appears likely that decreased enzyme stability, rather than decreased catalytic efficiency, plays a more important role in determining the phenotype (see below).
PPT activity and protein in NCL lymphoblasts
Expression studies were useful in assessing the impact of individual mutations on intrinsic catalytic activity in vitro, but provided no information on their impact in vivo. Ideally, analysis of PPT in affected brain tissue would be most informative, but it is not routinely available. Instead, we measured PPT activity in patient lymphoblasts. These results are summarized in Table 2. As anticipated, mutant PPT alleles that had no functional activity in expression studies (H39Q, G42E, R122W, V181M and E184K) also showed no activity in vivo, when inherited in combination with each other or with nonsense alleles. However, the alleles associated with juvenile onset NCL showed clearly detectable activity, which was at a maximum of 2.15% for the D79G allele. The mutation associated with three late infantile families (Q177E) showed an intermediate level of enzyme activity that was just detectable in the assay (1.08%). Therefore, there was a good correlation between the level of measurable enzymatic activity and the phenotype, with one exception. In two cell lines bearing an initiation methionine mutation, no activity could be detected in lymphoblasts, and yet the corresponding patients had late-onset phenotypes (one late infantile with onset at 3.5 years and one juvenile with onset at 5 years). Our expression studies indicated that the private mutations inherited in these cases (Y109D and Y247H) were inactive, and that the shared mutation M1I is functional, assuming a low level of expression occurs from the mutated ATA initiation codon, as seen in the COS cells. We speculate that initiation at the mutant ATA does not occur as efficiently in lymphoblasts as in brain or COS cells, accounting for the discrepancy.
|
A reduction in enzyme activity could result from quantitative as well as qualitative defects in PPT. Interestingly, severely reduced levels of immunoreactivity were seen in all PPT-deficient cell lines that we examined (Fig. 2). Unfortunately, the antibody was not sufficiently sensitive to provide precise quantitation, as was possible with the enzyme activity assay. However, in the current study and from our earlier analysis of cell lines from 37 families (6,8), we found that PPT deficiency is consistently associated with low levels of immunoreactive protein. Remarkably, we have observed no single example of PPT deficiency with normal levels of immunoreactive PPT protein in lymphoblasts to date.
|
PPT expression levels
Low levels of immunoreactive protein may result from decreased expression at the mRNA level or increased protein turnover due to protein misfolding and degradation. In order to assess the first possibility, we performed RNA blotting on a panel of cell lines from PPT-deficient patients. (Fig. 3). PPT mRNA was readily detectable in all samples, with the exception of homozygotes for the common nonsense mutations R151X and L10X. Consistent with this observation, both R151X and L10X alleles would be predicted to undergo nonsense-mediated RNA decay (see Discussion).
|
PPT glycosylation and mannose 6-phosphate receptor binding
Several of the late-onset mutations (D79G and the common T75P allele) were catalytically active and were associated with normal mRNA levels but showed reduced protein levels in the lymphoblast samples, suggesting that increased protein turnover plays a role in producing the phenotype. Increased protein turnover could result from improper folding, leading to ER-mediated protein degradation; alternatively, in rare instances, mistargeting of mutant lysosomal enzymes may lead to increased turnover. To assess protein stability of all of the PPT mutant proteins, we expressed mutant PPT enzymes in COS cells and followed the synthesis and degradation in [35S]methionine pulse-chase experiments. We found that both mutant and wild-type enzymes turned over with a half-life of between 3 and 5 h (data not shown). However, these studies must be interpreted with caution, since under conditions of overexpression, the measured half-life may not reflect the turnover under physiological conditions. For example, in the case of the R122W mutant, the mutant protein expressed in COS cells accumulates in the ER, whereas under physiological conditions of normal expression, the degradation rate keeps up with the synthetic rate and no immunoreactive PPT can be demonstrated in the brain (2). Unfortunately, the best currently available anti-human PPT antibody is not sufficiently sensitive for determining PPT half-life under conditions of endogenous expression in lymphoblasts or fibroblasts, the only accessible patient tissues. Endogenous levels of PPT in these tissues are extremely low compared with those in the brain, and pulse-chase analysis was not technically feasible. Immunofluorescence assays using overexpression in COS cells also did not yield interpretable information regarding the mutants, as the results were very dependent on expression level. For this reason, a biochemical approach was used to assess lysosomal trafficking of PPT mutants.
Lysosomal enzymes including PPT are targeted to lysosomes through specific recognition of a mannose 6-phosphate determinant present on the oligosaccharide chains of the protein (12,13). In order to assess targeting of wild-type and mutant PPT enzymes to lysosomes through the mannose 6-phosphate receptor pathway, we first tested the ability of the mutants to recognize biotinylated mannose 6-phosphate receptors in a recently described receptor blotting assay (14,15). In this assay, whole-cell extracts of COS cells expressing wild-type and mutant PPT enzymes were subjected to SDS–PAGE, blotted onto nitrocellulose filters and probed with a biotinylated mannose 6-phosphate receptor. First, we examined wild-type PPT in this assay, and tested the effect of glycosylation site mutations that might disrupt the interaction (Fig. 4). PPT has sites of asparagine-linked glycosylation at amino acid positions 197, 212 and 232 (11,12). Strongest mannose 6-phosphate receptor binding was seen when all three sites were intact (Fig. 4, lane 2). Mutation of these sites, singly or in combination, reduced the level of binding (Fig. 4, lanes 4–9 and 11). The glycosylation site at amino acid position 232 showed the greatest influence on binding because mutation of this site alone markedly reduced binding (Fig. 4, compare lanes 2 and 6) whereas mutation at either of the other two sites had only a minor effect. In addition, the site at position 232 was the only site that could support a low level of mannose 6-phosphate receptor binding alone (Fig. 4, lanes 7–9).
|
Perhaps somewhat surprisingly, none of the disease-associated PPT mutants bound to the mannose 6-phosphate receptor regardless of the phenotype (Fig. 5). The failure of receptor binding was not a function of enzyme activity, because a structurally subtle but catalytically inactive PPT mutant (Ser115Ala) showed nearly normal binding in this assay (Fig. 5, lane 12).
|
Failure of the naturally occurring mutant PPTs to bind to mannose 6-phosphate receptors might be explained by a lack of mannose 6-phosphate determinants on the asparagine-linked oligosaccharides of mutant PPT proteins. The mannose 6-phosphate comes to modify lysosomal enzymes in a two-step reaction (reviewed in ref. 16). In the first step, N-acetylglucosamine 1-phosphate is transferred by a phosphotransferase to hydroxyl groups on the C-6 of mannose residues in N-linked high mannose oligosaccharides to generate phosphodiesters. In the second step, a phosphodiesterase or ‘uncovering enzyme’ removes the N-acetylglucosamine to reveal the mannose 6-phosphate. To assess these two steps during the processing of the wild-type and mutant PPT enzymes, phosphate labeling of PPT enzymes expressed in COS cells was compared (Fig. 6 and Table 1). Phosphate incorporation into mutant PPT enzymes was indistinguishable from wild-type (Fig. 6, compare lane 3 with lanes 5, 7 and 9). The phosphate was bound to asparagine-linked high-mannose oligosaccharides on the expressed proteins, because the phosphate is sensitive to endoglycosaminidase H (Endo H) (Fig. 6). Therefore, the defect in mannose 6-phosphate receptor binding did not seem to reside grossly at the level of the phosphotransferase (see Discussion). To assess the status of the ‘cover’ on the mannose 6-phosphate residue, labeled extracts as shown in Figure 5 were subjected to digestion with alkaline phosphatase, an enzyme which dephosphorylates mannose 6-phosphate but not the ‘covered’ N-acetylglucosamine phosphate (17). We found that the phosphate bound to the mutant PPT enzymes was not sensitive to alkaline phosphatase, but that the phosphate bound to the wild-type enzyme was partially sensitive (data not shown). Therefore, the failure of the mutants to bind to the mannose 6-phosphate receptor appears to be related to the continued presence of the ‘cover’, blocking their interaction. Copious amounts of wild-type PPT were secreted into the medium, whereas secretion of mutants was not detectable, or only barely detectable in some cases (data not shown). We relate all of these findings to failure of the mutants to exit from the ER, a hallmark of ER-mediated protein degradation, as discussed below.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
In the current study, we analyzed the effect of known mutations on PPT structure and function. The most common nonsense mutation (R151X, accounting for 40% of all non-Finnish alleles) is characterized by the absence of detectable mRNA. This observation explains the frequent failure to detect PPT mutations using methods involving PCR amplification of cDNA (i.e. RT–PCR) (6). The absence of mRNA associated with this allele is consistent with the location of the mutation (in the 5' portion of exon 5), because nonsense mutations located >50–55 bp upstream of 3' splice junctions result in message instability, through a poorly understood process known as nonsense-mediated RNA decay (reviewed in ref. 18). The antibiotic gentamicin has been used recently in cultured cells to suppress certain nonsense mutations, presumably through its effects on ribosome function. In one example, gentamicin was shown to promote read-through translation of a nonsense allele causing Duchenne muscular dystrophy (19,20). Unfortunately, gentamicin was ineffective in increasing expression of PPT in cell lines homozygous for the R151X mutation (J.-Y. Lu, unpublished data). It is interesting to speculate that gentamicin may not have been effective in this situation because the mRNA is subject to nonsense-mediated RNA decay, but many more observations of nonsense mutations in PPT and in other genes will be needed in order to confirm or dismiss this hypothesis.
Of the 12 missense mutations analyzed in this study, four were associated with low but measurable enzymatic activity and were found in patients with late-onset phenotypes. The residual PPT activity in lymphoblasts in these subjects ranged from 1.08 to 2.15%. As a rule, residual activity in cultured cells from patients with lysosomal storage disorders, when viewed over a spectrum of clinical phenotypes, tend to be clustered within a narrow range that is often well below 5% of the normal level. Conzelmann and Sandhoff (21) have discussed theoretical reasons why very small changes in residual lysosomal enzyme activity may lead to large differences in the rate of substrate storage, and how this translates to clinical severity. In several instances (such as in this paper) it has been possible to relate residual activity to clinical phenotype. For example, residual activities of ß-hexosaminidase were 0.1% of the normal level for infantile Tay–Sachs patients, 0.5% for late infantile cases, and 2–4% for adult patients (22). In cultured skin fibroblasts, activities of -L-iduronidase of 0.1–0.5% of the normal level were found for the severe Hurler patients and 0.3–1.4% for the milder Scheie patients. However, in other disorders, such correlations have been largely unsuccessful (23). For example, lysosomal neuraminidase activity in sialidosis fibroblasts ranges from 0.5 to 1.5%, and shows no correlation over a broad spectrum of clinical manifestations (24). Similarly, distinct mutations in lysosomal acid lipase lead to clinically different disorders (Wolman and cholesterol ester storage disease) and yet enzymatic activity in fibroblasts or leukocytes does not distinguish the phenotype (25). In addition, the residual activity of sphingomyelinase or glucocerebrosidase reported from cases of similar severity is said to vary greatly and provide no clear distinction between infantile and adult cases. This variation has been related to practical difficulties in sphingolipid enzymology (21). In the current study, the correlation between clinical severity and residual activity fell within a remarkably small range (2%). We attribute this to the exceptional sensitivity and low background provided by the thiopalmitoyl-ß-D-glucoside substrate (26).
PPT immunoreactivity was severely reduced in the lymphoblasts of all patients we examined, and although precise quantitation was not possible, it was evident that in no case were detectable amounts of catalytically inactive protein found. Therefore, although our expression studies indicated that some of the mutants (T75P and D79G) were close to catalytically normal, the reduced levels of immunoreactive protein in tissues from patients with the disorder suggests that these mutant proteins are unstable in vivo. Although two mutants with significantly reduced substrate affinity were identified, the severely reduced immunoreactive protein observed with these alleles is also consistent with decreased stability of the mutant enzymes. Unfortunately, for technical reasons, this hypothesis could not be tested directly under physiological conditions of endogenous levels of PPT expression.
In the course of this study, we identified a mutation in an initiation codon (ATGATA) which yields 10% of normal expression when expressed transiently in COS cells. Given that the two mutations co-inherited with this allele did not have any residual enzyme activity, this observation suggests that the ATA codon may be utilized and result in the translation of a low but clinically important amount of enzyme in the brain. Translation initiation of ATA (and other non-ATG triplets) has been observed in mammalian cells (27,28), particularly when the AUA triplet is in the context of a favorable sequence context, as defined by the rules of Kozak (29). In these cases, methionine is the initiating amino acid regardless of which triplet is utilized, because tRNAiMet is the only species to gain access to the initiator site. The consequence is the production of small amounts of an entirely normal polypeptide. We found evidence for translation at the ATA codon in PPT in COS cells and perhaps brain (through clinical data) but not lymphoblasts. It might be interesting in the future to compare translation at ATA codons in transfected neurons versus lymphoblasts, to see whether there is a tissue-specific difference in translation at the ATA codon in different tissues.
A potentially significant finding in this study was the inability of all mutant PPT enzymes expressed in COS cells to bind to the mannose 6-phosphate receptor in blotting assays. To understand this finding, it is important to appreciate that the mannose 6-phosphate receptor recognition involves only the sugar phosphate and makes no contact with the polypeptide backbone of lysosomal enzymes. In addition, contact with two mannose 6-phosphate residues (normally located on adjacent branches of oligosaccharide chains) is required for high affinity binding. Therefore, we interpret the failure of recognition as the lack of a step needed to generate sufficient mannose 6-phosphate on the mutant enzymes. As discussed above, mannose 6-phosphate addition is a two-step process involving a phosphotransferase and an uncovering enzyme that removes a GlcNac residue and exposes the phosphate. The phosphotransferase reaction itself occurs in a sequential manner at two (or more) sites located within the secretory route, probably both proximal and distal to the KDEL retrieval site in the early Golgi. The first phosphorylation event occurs in the ER or intermediate compartment and the second in the early or even mid-Golgi (30,31). We have documented at least one phosphorylation event on all of the mutant PPT enzymes we examined and therefore none of the mutations appears to grossly affect phosphotransferase recognition. In contrast, the mutants were resistant to alkaline phosphatase treatment, suggesting that they are in ‘covered’ form. The uncovering enzyme was recently shown to reside primarily in the trans-Golgi network (TGN), and to cycle between the TGN and the plasma membrane (32). Our data are most consistent with the interpretation that mutant lysosomal enzymes (under conditions of overexpression) are still ‘covered’ and have been retained in a compartment that lies within or between the first phosphotransferase step and before the TGN, the site of uncovering enzyme. These data are also consistent with earlier data indicating that brefeldin A, a fungal metabolite which induces Golgi stack collapse into the ER, prevents uncovering of lysosomal enzymes but not phosphate transfer (33). Therefore, the primary consequence of the disease-associated missense mutations analyzed in this study is failure to progress through the ER. This conclusion is consistent with the increasingly recognized role of the ER in the degradation of improperly folded polypeptides and our repeated finding of low levels of immunoreactive PPT in mutant cell lines.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Analysis of transformed lymphoblast cell lines
Epstein–Barr virus-transformed lymphoblastoid cell lines derived from peripheral blood mononuclear cells were grown in RPMI 1640 (Gibco) supplemented with 10% heat- inactivated fetal bovine serum, 10 U/ml penicillin, 100 µg/ml streptomycin and 0.25 µg/ml amphotericin B. For analysis by PPT assay and immunoblotting, cells were harvested and washed with phosphate-buffered saline (PBS) pH 7.0, and resuspended in buffer containing 50 mM Tris–HCl pH 7.4, 150 mM NaCl, 2 mM EDTA, 1 µg/ml pepstatin and 1 µg/ml leupeptin. Cells were sonicated for two pulses of 6 s each at a setting of 3 using a Branson probe sonifier on ice, and homogenates were centrifuged at 100 000x g for 1 h at 4°C. The clear supernatants were removed and subjected to assays for PPT activity and immunoblotting. Protein concentrations were determined using the Micro DC protein assay (BioRad). Immunoblotting was performed as described previously (6) using 50 µg of total protein per lane and developed using a rabbit anti-human PPT polyclonal antibody and an ECL reagent kit from Amersham Pharmacia. For RNA blotting, total RNA was extracted directly from cells using Trizol reagent (Gibco). Blots were probed for 3 h at 65°C in Rapid Hyb buffer (Amersham) with a random hexamer-primed [32P]-labeled cDNA insert corresponding to the entire human PPT coding region, and washed in 2x SSC, 0.1% SDS at ambient temperature for 30 min and in 0.1x SSC, 0.1% SDS at 65°C for 40 min.
Thioesterase assays
Assays utilizing the fluorescent substrate 4-methylumbelliferyl-6-thiopalmitoyl-ß-D-glucoside were performed using the one-stage procedure exactly as described by van Diggelen et al. (26). Fluorescence was measured at 360 nm (excitation) and 465 nm (emission). Fluorescence measurements were quantified against a standard curve of increasing concentrations of 4-methylumbelliferone (Sigma). Palmitoyl CoA hydrolase assays using [3H]palmitoyl CoA were performed as described previously (3). For determination of Vmax and apparent Km, substrate concentrations were varied between 20 and 640 µM. Hydrolysis times were adjusted such that the extent of substrate hydrolysis never exceeded 10% in the assays.
Site-directed mutagenesis and expression of PPT in COS cells
The plasmid pCMV5-hPPT was used as a template for the introduction of all PPT missense mutations (both naturally occurring and designed) by PCR, using the Chameleon system (Stratagene) and appropriate mutagenic oligonucleotides essentially as described by Bellizzi et al. (11). Sequences of mutagenic oligonucleotides are available upon request. The PPT coding region in mutagenized expression constructs was fully sequenced to rule out the presence of secondary mutations. COS cell transfections, preparation of cell lysates and normalization of the data for PPT protein expression by scanning densitometry were performed exactly as described by Bellizzi et al. (11). Briefly, cells were harvested in PBS pH 7.0, sonicated, and centrifuged for 15 min at 15 800 g in an Eppendorf microcentrifuge at 4°C. Clear supernatants were subjected to PPT activity assays and immunoblotting as described above.
Histidine-tagged constructs and Sf9 expression, purification and analysis
EcoRI/HindIII inserts of pCMV5-hPPT (containing the entire coding region of human PPT with or without desired missense mutations) (4) were subcloned into the plasmid pGEM3, opened at a unique PstI site and ligated to annealed synthetic oligonucleotides (5'-GCATCACCATCATCACCACCTGCA-3' and 5'-GGTGGTGATGATGGTGATGCTGCA-3') to yield his-tagged constructs. The resulting translation products possess a hexahistidine tag immediately downstream of the signal peptide, resulting in a mature processed protein with the predicted N-terminal sequence A-L-Q-(H)6-L-Q-H. Tagged inserts were subsequently subcloned into an EcoRI/HindIII-digested pFastBac baculovirus expression vector (Bac-to-Bac expression kit, Gibco) for expression in Sf9 cells. Sf9 cells (1.2 x 106 cells/ml) were grown in 10% fetal bovine serum containing IPL 41 supplemented with 50 µg/ml gentamicin, 2.5 µg/ml fungizone, 0.1% (v/v) Pluronic F 68, 1x tryptose phosphate broth and 4.0 gm/l TC yeastoleate. Cells were infected with recombinant virus at 0.1 MOI and harvested 60 h post-transfection. Harvested cells were washed with PBS pH 7.0 and resuspended in 25 ml of a buffer containing 20 mM Tris–HCl pH 7.0, 100 mM NaCl and protease inhibitor cocktail (Sigma P8849) (Buffer A). Cells were lysed by sonication using a Branson sonifier at a setting of 3 for five pulses of 5 s and centrifuged at 100 000 g for 1 h. The supernatant was loaded onto a 2 ml DE52 column previously equilibrated with Buffer A to remove contaminating proteins. The unbound (flow-through) fraction was adjusted to 10 mM imidazole by the addition of 500 mM imidazole pH 7.5, and loaded onto a 1 ml Ni2+-NTA agarose column (Qiagen). The column was washed with 20 ml of a buffer containing 20 mM Tris–HCl pH 7.5, 500 mM NaCl (Buffer B) and with 20 ml of Buffer B containing 25 mM imidazole pH 7.5. The recombinant PPT was stored on the column in buffer containing 20 mM Tris–HCl, 100 mM NaCl, 0.02% NaN3 pH 7.0 and 0.2 mM PMSF and used directly in enzyme assays. Kinetic experiments using histidine-tagged PPT prepared in this manner yielded virtually identical results to unmodified PPT.
Mannose 6-phosphate receptor and immunoblotting
Simian COS-1 cells growing in 60 mm plates were transiently transfected with 1 µg of pCMV5-hPPT (wild-type or mutant) and 6 µl of FuGENE 6 reagent (Roche). Cells were harvested at 65–72 h post-transfection into buffer containing 10 mm HEPES pH 7.9, 1.5 mM MgCl2, 10 mM KCl, 0.5 mM DTT, 0.2 mM PMSF, 50 mM NaF, 1 mM Na3VO4 and 0.5 mM ß-glycerophosphate and were homogenized by sonication. The homogenates were centrifuged at 12 000x g for 10 min, and supernatants were subjected to electrophoresis in 12% SDS-polyacrylamide slab gels, and transferred to nitrocellulose. To determine the level of PPT protein expression, the filter was blotted against PPT antibody as described previously (11). For mannose 6-phosphate receptor blotting, filters were blocked for 1 h in PBS-T [0.025% (v/v) Tween-20 in PBS] containing 5% (w/v) non-fat dry milk, washed in PBS-T, and incubated for 4 h with biotinylated mannose 6-phosphate receptor (0.125–0.8 µg/ml) in PBS-T containing 5 mM ß-glycerophosphate essentially as described by Sleat et al. (14). The filters were washed and incubated for 1 h with avidin DH: biotinylated horseradish peroxidase H complex according to the directions supplied by the manufacturer (Vectastain ABC kits, Vector Laboratories) and developed using ECL chemiluminescence reagents (Amersham).
Phosphate labeling and immunoprecipitation
COS1 cells were maintained and transfected as described above. For phosphate labeling, cells that were 60 h post-transfection were incubated in phosphate-deficient medium for 1 h and then labeled for 3 h with 400 µCi/ml [32P]orthophosphate (ICN) and in fresh culture medium for an additional 3 h. Cells were harvested and washed three times with ice-cold PBS, pH 7.0. Cell lysates were prepared by the addition of 0.25 ml of lysis buffer consisting of Dulbecco’s phosphate-buffered saline (without calcium and magnesium) containing 1% (v/v) Triton X-100, 1 µg/ml leupeptin, 1 µg/ml pepstatin and 0.2 mM PMSF for 1 h on ice, followed by centrifugation at 15 800 g in a microfuge for 30 min at 4°C. To immunoprecipitate the [32P]phosphate-labeled PPT, cell lysates were pre-cleared with preimmune IgG (5 µg of IgG per 50 µg of total protein) at 4°C for 1 h. The preimmune complex was absorbed on Protein A agarose beads by incubating for 1 h at 4°C and centrifuged at 720 g for 5 min. Affinity-purified anti-human PPT antibody (5 µg) was added to the precleared cell lysate and incubated for 14–16 h at 4°C. The immune complexes were collected on Protein A agarose beads (Sigma) for 2 h at 4°C. The pellets were washed three times with ice-cold PBS pH 7.0. Endoglycosaminidase H treatment of pellets (as indicated in the legend to Fig. 5) was performed according to the manufacturer’s directions (Glyko). Pellets were treated with SDS–PAGE buffer, boiled for 5 min, and subjected to 12% SDS–PAGE and autoradiography.
|
ACKNOWLEDGEMENTS |
|---|
We thank David Sleat and Peter Lobel for helpful advice and for the biotinylated mannose 6-phosphate receptor used in this study, and Paula Bezerra for expert technical assistance. This study was supported by a grant from the National Institutes of Health (NS 36867), the Robert A. Welch Foundation, and the Batten Disease Support and Research Association.
|
FOOTNOTES |
|---|
+ To whom correspondence should be addressed. Tel: +1 214 648 4911; Fax: +1 214 648 4940; Email: sandra.hofmann@utsouthwestern.eduThe authors wish it to be known that, in their opinion, the first two authors should be regarded as joint First Authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
1 Goebel, H.H. (1995) The neuronal ceroid-lipofuscinoses. J. Child Neurol., 10, 424–437.
2 Vesa, J., Hellsten, E., Verkruyse, L.A., Camp, L.A., Rapola, J., Santavuori, P., Hofmann, S.L. and Peltonen, L. (1995) Mutations in the palmitoyl protein thioesterase gene causing infantile neuronal ceroid lipofuscinosis. Nature, 376, 584–587.[Medline]
3 Camp, L.A. and Hofmann, S.L. (1993) Purification and properties of a palmitoyl-protein thioesterase that cleaves palmitate from H-Ras. J. Biol. Chem., 268, 22566–22574.
4 Camp, L.A., Verkruyse, L.A., Afendis, S.J., Slaughter, C.A. and Hofmann, S.L. (1994) Molecular cloning and expression of palmitoyl-protein thioesterase. J. Biol. Chem., 269, 23212–23219.
5 Mitchison, H.M., Hofmann, S.L., Becerra, C.H., Munroe, P.B., Lake, B.D., Crow, Y.J., Stephenson, J.B., Williams, R.E., Hofman, I.L., Taschner, P.E.M. et al. (1998) Mutations in the palmitoyl-protein thioesterase gene (PPT; CLN1) causing juvenile neuronal ceroid lipofuscinosis with granular osmiophilic deposits. Hum. Mol. Genet., 7, 291–297.
6 Das, A.K., Becerra, C.H.R., Yi, W., Lu, J.-Y., Siakotos, A.N., Wisniewski, K.E. and Hofmann, S.L. (1998) Molecular genetics of palmitoyl-protein thioesterase deficiency in the U.S. J. Clin. Invest., 102, 361–370.[Web of Science][Medline]
7 Mole, S.E., Mitchison, H.M. and Munroe, P.B. (1999) Molecular basis of the neuronal ceroid lipofuscinoses: mutations in CLN1, CLN2, CLN3, and CLN5. Hum. Mutat., 14, 199–215.[Web of Science][Medline]
8 Waliany, S., Das, A.K., Gaben, A., Wisniewski, K.E. and Hofmann, S.L. (2000) Identification of three novel mutations of the palmitoyl-protein thioesterase-1 (PPT1) gene in children with neuronal ceroid-lipofuscinosis. Hum. Mutat., 15, 206–207.
9 Salonen, T., Jarvela, I., Peltonen, L. and Jalanko, A. (2000) Detection of eight novel palmitoyl protein thioesterase (PPT) mutations underlying infantile neuronal ceroid lipofuscinosis (INCL;CLN1). Hum. Mutat., 15, 273–279.[Web of Science][Medline]
10 Munroe, P.B., Greene, N.D., Leung, K.Y., Mole, S.E., Gardiner, R.M., Mitchison, H.M., Stephenson, J.B. and Crow, Y.J. (1998) Sharing of PPT mutations between distinct clinical forms of neuronal ceroid lipofuscinoses in patients from Scotland. J. Med. Genet., 35, 790.
11 Bellizzi, J.J.,III, Widom, J., Kemp, C., Lu, J.Y., Das, A.K., Hofmann, S.L. and Clardy, J. (2000) The crystal structure of palmitoyl protein thioesterase 1 and the molecular basis of infantile neuronal ceroid lipofuscinosis. Proc. Natl Acad. Sci. USA, 97, 4573–4578.
12 Verkruyse, L.A. and Hofmann, S.L. (1996) Lysosomal targeting of palmitoyl-protein thioesterase. J. Biol. Chem., 271, 15831–15836.
13 Hellsten, E., Vesa, J., Olkkonen, V.M., Jalanko, A. and Peltonen, L. (1996) Human palmitoyl protein thioesterase: evidence for lysosomal targeting of the enzyme and disturbed cellular routing in infantile neuronal ceroid lipofuscinosis. EMBO J., 15, 5240–5245.[Web of Science][Medline]
14 Sleat, D.E., Chen, T.L., Raska, K.,Jr and Lobel, P. (1995) Increased levels of glycoproteins containing mannose 6-phosphate in human breast carcinomas. Cancer Res., 55, 3424–3430.
15 Sleat, D.E., Sohar, I., Lackland, H., Majercak, J. and Lobel, P. (1996) Rat brain contains high levels of mannose-6-phosphorylated glycoproteins including lysosomal enzymes and palmitoyl-protein thioesterase, an enzyme implicated in infantile neuronal lipofuscinosis. J. Biol. Chem., 271, 19191–19198.
16 Kornfeld, S. (1990) Lysosomal enzyme targeting. Biochem. Soc. Trans, 18, 367–374.[Web of Science][Medline]
17 Isidoro, C., Radons, J., Baccino, F.M. and Hasilik, A. (1990) Suppression of the ‘uncovering’ of mannose-6-phosphate residues in lysosomal enzymes in the presence of NH4Cl. Eur. J. Biochem., 191, 591–597.[Web of Science][Medline]
18 Hentze, M.W. and Kulozik, A.E. (1999) A perfect message: RNA surveillance and nonsense-mediated decay. Cell, 96, 307–310.[Web of Science][Medline]
19 Howard, M.T., Shirts, B.H., Petros, L.M., Flanigan, K.M., Gesteland, R.F. and Atkins, J.F. (2000) Sequence specificity of aminoglycoside-induced stop condon readthrough: potential implications for treatment of Duchenne muscular dystrophy. Ann. Neurol., 48, 164–169.[Web of Science][Medline]
20 Barton-Davis, E.R., Cordier, L., Shoturma, D.I., Leland, S.E. and Sweeney, H.L. (1999) Aminoglycoside antibiotics restore dystrophin function to skeletal muscles of mdx mice. J. Clin. Invest., 104, 375–381.[Web of Science][Medline]
21 Conzelmann, E. and Sandhoff, K. (1983) Partial enzyme deficiencies: residual activities and the development of neurological disorders. Dev. Neurosci., 6, 58–71.[Web of Science][Medline]
22 Conzelmann, E., Kytzia, H.J., Navon, R. and Sandhoff, K. (1983) Ganglioside GM2 N-acetyl-ß-D-galactosaminidase activity in cultured fibroblasts of late-infantile and adult GM2 gangliosidosis patients and of healthy probands with low hexosaminidase level. Am. J. Hum. Genet., 35, 900–913.[Web of Science][Medline]
23 Litjens, T., Brooks, D.A., Peters, C., Gibson, G.J. and Hopwood, J.J. (1996) Identification, expression, and biochemical characterization of N-acetylgalactosamine-4-sulfatase mutations and relationship with clinical phenotype in MPS-VI patients. Am. J. Hum. Genet., 58, 1127–1134.[Web of Science][Medline]
24 Bonten, E.J., Arts, W.F., Beck, M., Covanis, A., Donati, M.A., Parini, R., Zammarchi, E. and d’Azzo, A. (2000) Novel mutations in lysosomal neuraminidase identify functional domains and determine clinical severity in sialidosis. Hum. Mol. Genet., 9, 2715–2725.
25 Anderson, R.A., Bryson, G.M. and Parks, J.S. (1999) Lysosomal acid lipase mutations that determine phenotype in Wolman and cholesterol ester storage disease. Mol. Genet. Metab., 68, 333–345.[Web of Science][Medline]
26 van Diggelen, O.P., Keulemans, J.L., Winchester, B., Hofman, I.L., Vanhanen, S.L., Santavuori, P. and Voznyi, Y.V. (1999) A rapid fluorogenic palmitoyl-protein thioesterase assay: pre- and postnatal diagnosis of INCL. Mol. Genet. Metab., 66, 240–244.[Web of Science][Medline]
27 Drabkin, H.J. and RajBhandary, U.L. (1998) Initiation of protein synthesis in mammalian cells with codons other than AUG and amino acids other than methionine. Mol. Cell. Biol., 18, 5140–5147.
28 Peabody, D.S. (1989) Translation initiation at non-AUG triplets in mammalian cells. J. Biol. Chem., 264, 5031–5035.
29 Kozak, M. (1999) Initiation of translation in prokaryotes and eukaryotes. Gene, 234, 187–208.[Web of Science][Medline]
30 Gabel, C.A. and Kornfeld, S. (1984) Targeting of ß-glucuronidase to lysosomes in mannose 6-phosphate receptor-deficient MOPC 315 cells. J. Cell Biol., 99, 296–305.
31 Dittmer, F. and von Figura, K. (1999) Phosphorylation of arylsulphatase A occurs through multiple interactions with the UDP-N-acetylglucosamine-1-phosphotransferase proximal and distal to its retrieval site by the KDEL receptor. Biochem. J., 340, 729–736.
32 Rohrer, J. and Kornfeld, R. (2001) Lysosomal hydrolase mannose 6-phosphate uncovering enzyme residues in the trans-Golgi network. Mol. Biol. Cell, in press.
33 Chardin, P. and McCormick, F. (1999) Brefeldin A: the advantage of being uncompetitive. Cell, 97, 153–155.[Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  B. Chang, M. N. A. Mandal, V. R.M. Chavali, N. L. Hawes, N. W. Khan, R. E. Hurd, R. S. Smith, M. L. Davisson, L. Kopplin, B. E.K. Klein, et al. Age-related retinal degeneration (arrd2) in a novel mouse model due to a nonsense mutation in the Mdm1 gene Hum. Mol. Genet., December 15, 2008; 17(24): 3929 - 3941. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Gupta, A. A. Soyombo, A. Atashband, K. E. Wisniewski, J. M. Shelton, J. A. Richardson, R. E. Hammer, and S. L. Hofmann Disruption of PPT1 or PPT2 causes neuronal ceroid lipofuscinosis in knockout mice PNAS, November 20, 2001; 98(24): 13566 - 13571. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||