Conserved amino acid residues in the COOH-terminal tail are indispensable for the correct folding and localization and enzyme activity of neutral ceramidase.

Several lines of evidence suggest that neutral ceramidase is involved in the regulation of ceramide-mediated signaling. Recently, the enzymes from mouse and rat were found to be localized at plasma membranes as a type II integral membrane protein, occasionally being detached from the cells after proteolytic processing of the NH(2)-terminal anchoring region (Tani, M., Iida, H., and Ito, M. (2003) J. Biol. Chem. 278, 10523-10530). We report here that conserved hydrophobic amino acid residues in the COOH-terminal tail are indispensable for the correct folding and localization, and enzyme activity of neutral ceramidase. Truncation of four, but not three, amino acid residues from the COOH terminus of rat neutral ceramidase resulted in a complete loss of enzyme activity as well as cell surface expression in HEK293 cells. Point mutation analysis revealed that Ile(758), the 4(th) amino acid residue from the COOH terminus, and Phe(756) are essential for the enzyme to function. The truncated and mutated enzymes were found to be retained in the endoplasmic reticulum (ER) and rapidly degraded without transportation to the Golgi apparatus. Treatment of the cells expressing the aberrant COOH-terminal enzyme with MG-132, a specific inhibitor for the proteasome, increased the accumulation of the enzyme in the ER, indicating that the misfolded enzyme was degraded by the proteasome. It was also found that the COOH-terminal tail was indispensable for the enzyme activity and correct folding of the prokaryote ceramidase from Pseudomonas aeruginosa, indicating that the importance of the COOH-terminal tail of the enzyme has been preserved through evolution.

lipid biomodulators for various cell functions (1)(2)(3). Both Cer and Sph have been shown to induce growth arrest and apoptosis (1,2), whereas S1P appears to promote cell growth and proliferation and suppress the apoptosis induced by Cer (3). Consequently, the balance of the cellular contents of Cer/Sph/ S1P is thought to regulate the diversity of cellular responses.
Ceramidase (CDase, EC 3.5.1.23) is an enzyme that cleaves the N-acyl linkage of Cer to produce Sph and free fatty acid (4). Three categories of CDases: acid, neutral, and alkaline, are clearly distinguished not only by their catalytic pH optima, but also their primary structures (5). Neutral CDase, which shows an optimum pH of 6.5-8.5, has been cloned from bacteria (6), Drosophila (7), mouse (8), rat (9), and human (10). Accumulating evidence suggests that neutral CDase regulates the intracellular content of Cer and thereby Cer-mediated signaling. For example, an increase of Cer caused by the IL-1␤-stimulated hydrolysis of SM was gradually normalized by up-regulation of neutral CDase as detected in mRNA and de novo synthesis levels after long term treatment with IL-1␤, showing the function of neutral CDase as a cytoprotective enzyme in mesangial cells exposed to inflammatory stress (11). In contrast, nitric oxide (NO) down-regulates the protein expression of neutral CDase via a protein kinase C-dependent mechanism in renal mesangial cells, resulting in an increase of Cer after treatment of the cells with NO (12). Very recently, the targeted expression of neutral CDase was found to reverse the retinal degradation by normalizing the Cer level in Drosophila arrestin and phospholipase C mutants (13). These results suggest that neutral CDases may be suitable targets in the therapeutic management of cytokine-induced inflammation and retinal degeneration.
Although the primary structure of neutral CDase is highly conserved from bacteria to mammals (8), the subcellular localization of the enzyme is quite different depending on its origin. Mammalian neutral CDases are present at the plasma membrane as a type II integral membrane protein, occasionally being detached from cells after proteolytic processing of the NH 2 -terminal anchoring region (14), whereas the bacterial and invertebrate enzymes are solely secretory proteins (7,15). This discrepancy may stem from whether a Ser/Thr-rich domain (mucin box) is present downstream of the NH 2 -terminal signal/ anchor sequence of the enzyme, because all mammalian CDases possess a mucin box, whereas the bacterial and invertebrate enzymes do not, and a mutant mammalian CDase lacking the mucin box or possible O-glycosylation sites in the mucin box was secreted into the medium instead of localizing at the cell surface (14).
Very recently, a change in the COOH-terminal region of acid CDase was identified to be a novel mutation in Farber disease in which Cer is accumulated in lysosomes due to the inactivation of acid CDase (16). In the present study, we found that two conserved amino acid residues, Phe and Ile/Val, in the COOHterminal domain were completely conserved in neutral CDases from bacteria to humans. Thus, we examine the role of the conserved COOH-terminal tail of neutral CDases using mammalian and bacterial enzymes. In conclusion, we demonstrated that two conserved amino acids in the COOH-terminal tail are indispensable for the correct folding and localization, and enzyme activity of neutral CDase.

EXPERIMENTAL PROCEDURES
Materials-Anti-Rab6 antibody was kindly provided by Dr. S. Tanaka (Shizuoka University, Japan). HRP-labeled anti-mouse IgG antibody was purchased from Nacalai Tesque (Japan). Cy3-labeled anti-mouse IgG antibody, anti-FLAG M2 antibody, and benzyl-GalNAc were from Sigma. ECL plus, and HRP-and Cy3-labeled anti-rabbit IgG antibodies were obtained from Amersham Biosciences. MG-132 (benzylcarbonyl-Leu-Leu-Leu-aldehyde) and anti-Myc antibody were purchased from Calbiochem and Invitrogen, respectively. Anti-neutral CDase antibody was raised in a rabbit using the recombinant rat CDase as the antigen (9). C12-NBD-Cer was prepared as described in Ref. 17. HEK293 cells (JCRB9068, established by F. L. Graham) were obtained from the Human Science Research Resource Bank. All other reagents were of the highest purity available.
CDase Assay-The hydrolysis and reverse hydrolysis activities of neutral CDase were measured using C12-NBD-Cer (for the hydrolysis reaction) and NBD-dodecanoic acid and Sph (for the reverse hydrolysis reaction) as substrates (18).
Cell Culture and cDNA Transfection-HEK293 cells, human embryonic kidney cells, were grown in Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% fetal bovine serum (FBS), and 60 g/ml of kanamycin in a humidified incubator containing 5% CO 2 . cDNA transfection was carried out using LipofectAMINE TM Plus (Invitrogen) according to the instructions of the manufacturer.
Preparation of Culture Supernatants and Cell Lysates-At 18 h after transfection with the CDase cDNA, the medium was replaced with serum-free Opti-MEM (Invitrogen, 500 l/well on 24-well plates) and cultured for an additional 24 h. The cell culture medium was collected and subjected to centrifugation at 13,000 ϫ g for 5 min. The supernatant, supplemented with a 1:10 volume of 200 mM Tris-HCl, pH 7.5 containing 1% Triton X-100 and 33 g/ml of proteinase inhibitors (leupeptin, pepstatin, and chymostatin), was used as cell supernatant. Cells attached to the culture plate were rinsed with PBS and then lysed by adding 200 l of 10 mM Tris-HCl, pH 7.5, containing 0.5% Triton X-100 and 3.3 g/ml of the proteinase inhibitors described above. Lysates were collected by pipette and used as cell lysates.
Protein Assay, SDS-PAGE, and Western blotting-Protein content was determined by the bicinchoninic acid protein assay (Pierce) with bovine serum albumin as standard. SDS-PAGE was carried out according to the method of Laemmli (19). Protein transfer onto a polyvinyldifluoride membrane was performed using TRANS-BLOT S.D. (Bio-Rad) according to the method described in Ref. 20. After treatment with 3% skim milk in Tris-buffered saline (TBS) containing 0.1% Tween 20 (T-TBS) for 1 h, the membrane was incubated with primary antibody for 1 day at 4°C. After a wash with T-TBS, the membrane was incubated with HRP-conjugated secondary antibody for 2 h. After another wash with T-TBS, the ECL reaction was performed for 5 min as recommended by the manufacturer, and chemiluminescent signals were visualized with an ECL TM mini-camera (Amersham Biosciences).
Immunocytochemistry and Fluorescence Microscopy-Transfected cells were cultured on cover glass and then fixed with 3% paraformaldehyde in PBS for 15 min. After rinsing with PBS and 50 mM NH 4 Cl in PBS, cells were permeabilized, if necessary, by 0.1% Triton X-100 in

FIG. 1. Alignment of the COOH-terminal tails of two prototype neutral CDases (A) and wild-type and COOH-terminal-truncated rat enzymes (B).
A, sequences of neutral CDases were aligned using the CLASTAL algorithm (30). Gaps inserted into the sequences are indicated by dots. B, wild-type and truncated enzymes were tagged with FLAG at the NH 2 terminus.
PBS. After treatment with blocking buffer (5% skim milk in PBS) for 15 min, the samples were incubated with primary antibody (diluted 1:1000 with blocking buffer) at 4°C for 1 day followed by Cy3-labeled secondary antibody at room temperature for 2 h. Immunostained samples were examined with a confocal laser-scanning microscope (Digital Eclipse C1, Nikon, Japan).
Generation of Polyclonal Antibodies against Pseudomonas Neutral CDase-The rabbit antiserum against the Pseudomonas neutral CDase was raised against a full-length CDase (residues 1-646, Ref. 6), which had been expressed using a pET23a plasmid vector and Escherichia coli BL21(DE3) cells. To obtain the recombinant CDase, 1 ml of overnight culture was inoculated into 200 ml of LB in the presence of 100 g/ml of ampicillin and incubated at 25°C. After a 20-h incubation, isopropylthio-␤-D-galactopyranoside was added to a final concentration of 0.1 mM, and incubated at 25°C for 24 h. Cells were harvested by centrifugation at 5,000 ϫ g for 10 min at 4°C, and the pellet was sonicated in 10 mM Tris-HCl buffer, pH 7.5, containing 0.5% Triton X-100 and 0.5 mM ABSF. The insoluble fraction (inclusion bodies) was precipitated by centrifugation (15,000 ϫ g for 10 min), lysed, and subjected to SDS-PAGE. After electrophoresis, the gel was stained with Gelcode Blue Stain Reagent (Pierce). The CDase band was cut out and subjected to electroelution. The protein obtained was used for immunization.

Expression of Pseudomonas Neutral CDase and the Mutants-E. coli BL21 (DE3)pLysS cells transformed with a plasmid vector containing
Pseudomonas neutral CDase and the mutant DNA were grown at 25°C for 16 h in 10 ml of LB medium containing 100 g/ml of ampicillin and 35 g/ml of chloramphenicol with shaking. Then, isopropylthio-␤-Dgalactopyranoside was added to the culture to a final concentration of 0.1 mM to trigger transcription. After an additional 1 h culture at 25°C, cells were harvested by centrifugation, suspended in 1 ml of 10 mM Tris-HCl, pH 7.5, containing 0.1% Triton X-100 and 3.3 g/ml of the proteinase inhibitors (leupeptin, pepstatin, and chymostatin), and sonicated. After sonication, the solution was centrifuged at 13,000 rpm for 5 min, and the supernatant obtained was used as the crude enzyme solution.

Alignment and Truncation of COOH-terminal Region of Neu-
tral CDases-Although the primary structure of neutral CDase is highly conserved from bacteria to humans, the enzyme is classified into two prototypes; bacteria/invertebrate and mam-malian types. The latter possesses a Ser/Thr-rich mucin-like domain (mucin box) downstream of the NH 2 -terminal signal/ anchor sequence while the former does not (Fig. 1A). The mucin box was shown to retain the enzyme on the plasma membranes as a type II integral membrane protein. Thus mucin box-deleted mutant enzymes and bacteria/invertebrate enzymes were found to be secreted into the culture medium (7,14,15). In this study, we found that two amino acid residues, Phe and Ile/Val, in the COOH-terminal tail were completely conserved in both two types of the enzyme (Fig. 1A). To address the role of the conserved COOH-terminal tail, COOH-terminal-truncated mutant enzymes lacking 11, 6, 5, 4, and 3 amino acid residues from the COOH terminus, respectively, were constructed and expressed in HEK293 cells (Fig. 1B, ⌬11, ⌬6, ⌬5, ⌬4, ⌬3). Wildtype and truncated enzymes were tagged with FLAG at the NH 2 terminus. As shown in Fig. 2A, all truncated CDases except ⌬3 lost the enzyme activity as measured in terms of the hydrolysis (a) and synthesis (b) of Cer. The enzyme activity of wild-type and ⌬3 CDases was found in cell lysates as well as the medium (Fig. 2B). Western blotting analysis of the cell lysate using anti-FLAG antibody revealed that the wild-type and ⌬3 CDases showed two protein bands having a molecular mass of 133 kDa and 113 kDa (Fig. 2B). The 133-kDa protein seems to be a mature form in the Golgi and the 113-kDa protein, developing form in the ER (9). A 130-kDa mature form was also detected in the culture medium of wild-type and ⌬3 CDases when stained with anti-neutral CDase antibody (Fig.  2B). This secretory CDase is likely to be generated by processing of the NH 2 -terminal signal/anchor sequence from the mature form (14). In contrast, no mature forms were detected in the cell lysate or medium of ⌬11, ⌬6, ⌬5, and ⌬4 enzymes. These results suggest that the truncated CDases except ⌬3 were not transported from the ER to the Golgi and lost enzyme activity.
Next, we analyzed the subcellular localization of COOH-

FIG. 2. Expression of wild-type and COOH-terminal-truncated rat neutral CDases in HEK293 cells.
A, time course for hydrolysis (a) and reverse hydrolysis (b) activities. HEK293 cells were transfected with a plasmid vector containing wild-type or truncated enzyme cDNA. The hydrolysis and reverse hydrolysis activities of the cell lysates were determined using C12-NBD-Cer (for the hydrolysis reaction) and Sph and NBDdodecanoic acid (for the reverse reaction) as substrates. B, enzyme activity and protein expression of wild-type and truncated enzymes. The neutral CDase activity of the cell lysate and medium was measured using C12-NBD-Cer as a substrate, and the protein expression was determined by Western blotting using anti-FLAG (for cell lysates) and antineutral CDase antibodies (for culture supernatants).
terminal-truncated enzymes. As shown in Fig. 3A, wild-type and ⌬3 CDases were located at plasma membranes while ⌬11, ⌬6, ⌬5, and ⌬4 enzymes were likely to be retained in the ER and not be transported to the plasma membrane. Since human neutral CDase was reported to localize in mitochondria (10), we examined whether wild-type and truncated enzymes from rat were expressed in mitochondria. In this experiment, GFP with a mitochondria-targeting signal (mitochondria-GFP) was used as a marker for mitochondria. As shown in Fig. 3B, wild-type, ⌬4, and ⌬3 CDases were not co-localized with GFP-mitochondria, indicating that the rat enzymes are not transported to mitochondria under the conditions used in this study. In conclusion, the removal of more than three amino acid residues from the COOH terminus of rat neutral CDase resulted in a complete loss of enzyme activity and correct subcellular localization. The truncated enzymes lacking 232, 89, 56, and 23 amino acid residues from the COOH terminus showed no enzyme activity either (data not shown).
Effects of Point Mutation of the Conserved Amino Acids in the COOH-terminal Tail on Enzyme Activity and Intracellular Lo-calization-The truncated enzyme ⌬3 retained the activity whereas ⌬4 did not. Therefore, Ile 758 of the rat enzyme, the 4th amino acid residue from the COOH terminus, is likely to be integral to the enzyme activity and correct subcellular localization (Fig. 1A). In addition to Ile 758 , Phe 756 is also conserved in all enzymes (Fig. 1A). Thus, we examined the effects of point mutations of Ile 758 and Phe 756 on enzyme activity and subcellular localization. All mutant and wild-type enzymes were tagged with Myc at the COOH terminus and expressed in HEK293 cells. As shown in Fig. 4A, replacement of Ile 758 with Asp or Arg almost completely abolished the enzyme activity. On the other hand, no decrease in enzyme activity was observed when Ile 758 was replaced with Val, which is conserved in all neutral CDases except the rat enzyme. Replacement of Ile 758 with Phe decreased the enzyme activity to 20% in comparison with wild-type CDase. Western blotting analysis using anti-Myc antibody detected the 113-kDa ER form, but not the 133-kDa mature form, in the cell lysates when I758D and I758R mutants were overexpressed in HEK293 cells. No enzyme activity or protein band of I758D and I758R was found in the medium. Similarly, F756D and F756R, but not F756I, showed no enzyme activity in cell lysates and culture medium. Correspondingly, the 133-kDa and 130-kDa mature forms were not found in the cell lysate and culture medium, respectively. The replacement of Phe 756 with Ile had little effect. Next, the cell surface distribution of the enzyme was examined with (permeable condition) or without (unpermeable condition) Triton X-100 treatment. Mutants I758D, I758R, F756D, and F756R, all of which lost enzyme activity completely, were not stained with anti-Myc antibody under unpermeable conditions, indicating they were not expressed on the surface of cells (Fig.  4B, lower panels). However, under permeable conditions these mutant CDases were detected in the ER at almost the same level as the wild-type enzyme (Fig. 4B, upper panels). The cell surface expression of the mutants I758F and F756I is somewhat lower than that of the wild-type and I758V enzymes. This result is well consistent with the enzyme activity levels. No reduction in enzyme activity or cell-surface expression was observed when Glu 757 , a non-conserved amino acid, was replaced with Arg (control experiment). In summary, these results clearly indicate that the conserved amino acid residues in the COOH-terminal tail, Ile 758 and Phe 756 , are integral to the activity and correct subcellular localization of the enzyme. It is noteworthy that Ile 758 or Phe 756 can be changed with another hydrophobic, but not a charged, amino acid without a marked loss of enzyme activity.
Glycosylation of Truncated and Mutated CDases-We compared the glycosylation of wild-type and mutant CDases using endoglycosidase H (endo H) and glycopeptidase F (PNGF). The former specifically hydrolyzes the N,NЈ-chitobiose linkage in high-mannose type N-glycans whereas the latter hydrolyzes not only high mannose-type, but also complex and hybrid-type N-glycans. As previously reported (9), the 113-kDa CDase was sensitive to endo H, whereas the 133-kDa CDase was somewhat resistant to endo H (Fig. 5A). The 133-kDa band was converted to a 105-kDa band after PNGF treatment. In contrast, both the I758D and ⌬4 CDases showed only an endo H-sensitive 113-kDa band (Fig. 5A), indicating that the mutant enzymes possess high mannose, but not complex/hybrid, type N-glycans. Treatment of the cells overexpressing wild-type CDase with benzyl-GalNAc, an inhibitor of O-glycosylation, resulted in a reduction in the molecular mass of the 133-kDa mature form, but not the 113-kDa ER form (Fig. 5B). In contrast, the I758D and ⌬4 CDases were not affected by the inhibitor, indicating that these mutant enzymes are not glycosylated with O-glycans (Fig. 5B). ⌬3 CDase seems to enter the ER/Golgi FIG. 3. Subcellular localization of wild-type and COOH-terminal-truncated rat neutral CDases. A, HEK293 cells overexpressing the CDases were fixed, permeabilized with Triton X-100, and then stained with anti-FLAG antibody followed by anti-mouse IgG-Cy3. The arrows indicate the expression of CDase at the plasma membrane. B, HEK293 cells were co-transfected with a plasmid vector containing CDase and that containing pCMV/Myc/mitochondria (Invitrogen) as a mitochondrial marker. Cells were fixed, permeabilized with Triton X-100, and stained with anti-FLAG antibody followed by anti-mouse IgG-Cy3. Left and center panels show the expression of CDase and mitochondria-GFP. Merged images are shown in the right panels. pathway and so be glycosylated properly, although the processing is somewhat slower than that of the wild-type enzyme (Fig.  5, A and B). Next, we examined the intracellular localization of the wild-type and I758D CDases. The Myc signal of wild-type CDase was partly co-localized with the signal for Rab6, a marker protein for the Golgi apparatus, but that of I758D mutant CDase was not (Fig. 5C). Similarly, ⌬4 and F756D CDases were not co-localized with Rab6 (data not shown). These results suggest that the removal of more than 3 amino acids from the COOH terminus or point mutations of two conserved amino acids in the COOH-terminal tail stop the transport of the enzyme from the ER to the Golgi apparatus, resulting in an absence of the protein on plasma membranes (Figs. 3A and 4B). It was concluded that the aberrant COOHterminal mutant CDases lacking enzyme activity are not transported to the Golgi apparatus leading to the immature/incorrect glycosylation of the protein.
Susceptibility to Degradation by Trypsin-The loss of enzyme activity and incorrect subcellular localization of the aberrant COOH-terminal enzymes could be attributed to misfolding of the proteins. To assess this possibility, the susceptibility of mutant CDases to trypsin was examined since trypsin-sensitivity may reflect the misfolding of proteins (21). As shown in Fig. 6, a notable difference in susceptibility to trypsin was observed between the wild-type and mutant CDases. Both the 133-and 113-kDa protein bands of wild-type CDase were quite resistant to proteolysis by trypsin up to a concentration of 100 ng/l, the 133-kDa mature form still remaining after digestion for 30 min with 500 ng/l of trypsin. In contrast, I758D and ⌬4 CDases were completely degraded by trypsin at a concentration of 50 ng/l, being 10-times as sensitive to trypsin as the wildtype enzyme. It is noted that ⌬3 CDase was relatively resistant to trypsin compared with I758D and ⌬4 CDases. These results may indicate that a mutation or truncation of the COOHterminal tail leads to incorrect folding of the neutral CDase.
Degradation of COOH-terminal-truncated and -mutated CDases by Proteasome-We examined the effects of cycloheximide (CHX), an inhibitor for de novo protein synthesis, on the stability of CDases overexpressed in HEK293 cells. As shown in Fig. 7A, the 113-kDa form disappeared quickly when de novo synthesis of the protein was stopped by CHX while the 133-kDa mature forms of the wild-type and ⌬3 CDases were detected after treatment of the cells with CHX for 4 h. The quick disappearance is probably caused by the conversion of the 113-kDa form into the 133-kDa form. The I758D and ⌬4 CDases disappeared very quickly after treatment of the cells with CHX, suggesting that they were degraded much faster than the wildtype enzyme in cells. To examine whether the proteasome is involved in the quick disappearance of the I758D and ⌬4 CDases, cells were incubated with MG-132, a specific inhibitor for the proteasome (22). Treatment with MG-132 resulted in a marked accumulation of I758D and ⌬4 CDases in the cell lysates (Fig. 7B). In contrast, MG-132 had no notable effects on the content of wild-type CDase regardless of the attachment of a FLAG tag at the NH 2 -terminal (Fig. 7B). These results suggest that the I758D and ⌬4 CDases, which are likely to be folded incorrectly in the ER, are degraded quickly by the proteasome.
Effects of COOH-terminal Truncation and Mutation on Pseudomonas CDase-To investigate the role of the COOHterminal tail of bacterial CDases, DNA constructs encoding truncated and point-mutated CDases of Pseudomonas aeruginosa were designed (Fig. 8A) and expressed in E. coli BL21 (DE3)pLysS cells. Val 665 in the COOH-terminal tail of Pseudomonas CDase corresponds to Ile 758 of the neutral CDase, which was indispensable for the activity and correct folding of rat enzyme as described above. Fig. 8B shows the CDase activity in the lysate of transformed cells. As expected, the mutation of Val 665 to Asp or removal of Val 665 completely abrogated the enzyme activity (Fig. 8B). However, ⌬5 CDase possessing Val 665 retained the enzyme activity though the reaction was slower than for the wild-type CDase (Fig. 8B). The protein expression levels of these three mutant enzymes were almost the same as wild-type levels when determined by Western blotting using anti-Pseudomonas CDase antibody (Fig. 8C). To investigate the susceptibility to proteolysis, wild-type and mutated enzymes were treated with trypsin for 30 min at different concentrations. It was found that the V665D and ⌬6 mutants were very sensitive to trypsin compared with the wild-type enzyme (Fig. 8D). Taken together, it is concluded that the FIG. 6. Susceptibility to trypsin of wild-type and aberrant COOH-terminal rat neutral CDases. HEK293 cells were transfected with a plasmid vector containing wild-type, truncated, or mutated CDase cDNA. The cell lysates (protein concentration, 0.3 g/l) were incubated at 37°C for 30 min with trypsin at concentrations of 10, 20, 50, 100, and 500 ng/l. The reaction was terminated by the addition of SDS (final concentration, 1%) and boiling for 3 min and subjected to SDS-PAGE, followed by Western blotting using anti-neutral CDase antibody.

FIG. 7. Degradation of wild-type and aberrant COOH-terminal rat neutral CDases by the proteasome.
A, effects of CHX on the processing of wild-type, I758D, ⌬4, and ⌬3 CDases. CDase-overexpressing HEK293 cells were treated with CHX (50 g/ml) in DMEM supplemented with 10% FBS for the period indicated. B, effects of MG-132 on the degradation of wild-type, I758D, and ⌬4 CDases. CDase-overexpressing HEK293 cells were treated with 5 mM MG-132 in DMEM supplemented with 10% FBS at 37°C for 12 h. Cells were harvested and the CDase proteins were detected by Western blotting using anti-Myc antibody (wild type and I758D) and anti-FLAG antibody (FLAG-tagged wild type, and ⌬4), respectively.

FIG. 5. Glycosylation and processing of wild-type and aberrant COOH-terminal rat neutral CDases.
A, analysis of the N-glycosylation of rat neutral CDase. HEK293 cells were transfected with a plasmid vector containing wild-type, truncated, or mutated CDase cDNA. The cell lysates were treated with endoglycosidase H (H) and glycopeptidase F (F) as recommended by the manufacturer and subjected to SDS-PAGE, followed by Western blotting using anti-Myc antibody (wild type and I758D) and anti-FLAG antibody (⌬4 and ⌬3). B, inhibition of O-glycosylation by benzyl-GalNAc. After cDNA transfection, the cells were incubated with or without 5 mM benzyl-GalNAc in DMEM supplemented with 10% FBS, and cultured for an additional 20 h. Cells were lysed and subjected to the SDS-PAGE, followed by Western blotting using anti-Myc antibody (wild type and I758D) and anti-FLAG antibody (⌬4 and ⌬3). C, localization of neutral CDase in the Golgi apparatus. Wild-type or I758D CDase-expressing cells were stained with anti-Myc antibody followed by anti-mouse IgG-Cy3 (left) and anti-Rab6 antibodies followed by anti-rabbit IgG-Alexa Fluor 488 (center). Images are merged (right). An arrow indicates the colocalization of neutral CDase and Rab6, a marker protein for the Golgi apparatus.
COOH-terminal tail is indispensable for the correct folding and activity of the Pseudomonas enzyme. DISCUSSION Interestingly, the shortened rat neutral CDases differing in length by only one amino acid (⌬4 versus ⌬3 enzyme) have quite different fates when overexpressed in HEK293 cells, i.e. ⌬4 CDase is not transported from the ER to the Golgi apparatus, instead it is degraded possibly by the proteasome, while ⌬3 enzyme is transported normally to the plasma membranes via the Golgi and partly detached from the cells. The reason for the difference was revealed by point mutation analysis: the 4th amino acid residue from the COOH terminus, Ile 758 , is essential for the correct folding and localization of the enzyme. In addition, Phe 756 , another conserved amino acid residue at the COOH-terminal tail, was also shown to be very important for enzyme functions. Therefore, two conserved amino acid residues in the COOH-terminal tail are indispensable for the correct folding and localization, and enzyme activity of the neutral CDase. It is noteworthy that these two hydrophobic amino acids could be replaced with other hydrophobic amino acids without a marked loss of function while their replacement with non-hydrophobic amino acids caused critical damage to the enzyme. X-ray crystallography or NMR data is, however, re-quired to reveal how the COOH-terminal tail contributes to the correct folding of the enzyme.
In eukaryotic cells, the ER has the inherent function of ensuring that only correctly folded and assembled proteins are forwarded to their destinations. The term quality control has been adopted to describe the process of conformation-dependent molecular sorting of newly synthesized proteins in the ER (23). Most commonly, misfolded or incompletely assembled proteins are retained in the ER and eventually degraded by the ubiqutin-proteasome pathway (23). The fact that the truncated and mutated neutral CDases were retained in the ER and degraded possibly by the proteasome may indicate that the aberrant COOH-terminal sequence resulted in a misfolding of the protein leading to elimination via quality control mechanisms coupled to the ER.
There have been several reports that COOH-terminal residues are important for the correct folding and secretion of proteins (21, 24 -26). Occasionally, COOH-terminal mutations are known to cause genetic disorders (16,27). In human ␣-1proteinase inhibitor deficiency, some variants are not secreted due to frameshift mutations leading to products with aberrant COOH-terminal sequences (28). The low level of the protein results in a failure to inhibit elastase released from activated or disintegrating neutrophils, and causes pulmonary damage in FIG. 8. Effects of COOH-terminal truncation and point mutation on Pseudomonas neutral CDase. A, COOH-terminal amino acid sequences of the wild-type, point-mutated, and truncated Pseudomonas neutral CDases. B, CDase activity of the wild-type and aberrant COOH-terminal CDases. Cell lysate (0.13 g of protein each) was incubated with 550 pmol of C12-NBD-Cer in 20 l of 25 mM Tris-HCl buffer, pH 8.5, containing 0.25% Triton X-100 and 2.5 mM CaCl 2 for the periods indicated. C, expression of Pseudomonas wild-type and aberrant COOH-terminal CDases in E. coli BL21(DE3)pLysS cells. Cell lysates were subjected to 7.5% SDS-PAGE, followed by Western blotting using anti-Pseudomonas CDase antibody. D, susceptibility to trypsin of Pseudomonas wild-type and aberrant COOH-terminal CDases. The cell lysates (protein concentration, 0.27 g/l) were incubated at 37°C for 30 min with trypsin at concentrations of 0, 2.5, 5, 10, 50, and 100 ng/l. The reaction was terminated by the addition of SDS (final concentration, 1%) and boiling for 3 min and subjected to SDS-PAGE followed by Western blotting using anti-neutral CDase antibody. emphysema (28). Farber disease is an autosomal recessive disorder caused by a deficiency of acid CDase (29). In one patient with a severe Farber phenotype, there was an insertion of the base T at the stop codon, which caused a frameshift and change in the COOH-terminal structure (16). These reports may indicate that a frameshift mutation at the COOH terminus can lead to a complete loss of function of the neutral CDase although a genetic deficiency of the enzyme has yet to be identified. This is the first report demonstrating the importance of the COOH-terminal tail of neutral CDase, which would be involved in the regulation of Cer-mediated signaling (11)(12)(13), for folding and processing leading to the enzyme activity and distribution. It should be emphasized that the role of the COOH-terminal tail of the enzyme has not changed throughout evolution, because aberrant COOH-terminal sequences drastically affect the functions of not only mammalian but also bacterial enzymes.