Prenylcysteine lyase deficiency in mice results in the accumulation of farnesylcysteine and geranylgeranylcysteine in brain and liver.

In in vitro experiments, prenylcysteine lyase (Pcly) cleaves the thioether bond of prenylcysteines to yield free cysteine and the aldehyde of the isoprenoid lipid. However, the importance of this enzyme has not yet been fully defined at the biochemical or physiologic level. In this study, we show that Pcly is expressed at high levels in mouse liver, kidney, heart, and brain. To test whether Pcly deficiency would cause prenylcysteines to accumulate in tissues and result in pathologic consequences, we produced Pcly-deficient cell lines and Pcly-deficient mice (Pcly-/-). Pcly activity levels were markedly reduced in Pcly-/- cells and tissues. Pcly-/- fibroblasts were more sensitive than wild-type fibroblasts to growth inhibition when prenylcysteines were added to the cell culture medium. To determine if the reduced Pcly enzyme activity levels led to an accumulation of prenylcysteines within cells, mass spectrometry was used to measure farnesylcysteine and geranylgeranylcysteine levels in the tissues of Pcly-/- mice and wild-type controls. These studies revealed a striking accumulation of both farnesylcysteine and geranylgeranylcysteine in the brain and liver of Pcly-/- mice. This accumulation did not appear to be accompanied by significant pathologic consequences. Pcly-/- mice were healthy and fertile, and surveys of more than 30 tissues did not uncover any abnormalities. We conclude that prenylcysteine lyase does play a physiologic role in cleaving prenylcysteines in mammals, but the absence of this activity does not lead to major pathologic consequences.

A wide variety of cellular proteins are posttranslationally modified by cholesterol biosynthetic intermediates, a process generally termed protein prenylation (1)(2)(3). Protein prenylation involves the covalent attachment a 15-carbon farnesyl or a 20-carbon geranylgeranyl lipid to cysteine residue(s) at or near the carboxyl terminus of a protein via a thioether bond. There are two categories of prenylated proteins in mammalian cells, the Rab proteins and the CAAX 1 proteins. The Rab proteins are geranylgeranylated at a pair of carboxyl-terminal cysteines (4,5). CAAX proteins are either farnesylated or geranylgeranylated at the carboxyl-terminal cysteine (the C of the CAAX motif) (1)(2)(3). After prenylation, CAAX proteins generally undergo two additional modifications. First, the last three residues of the protein (i.e. the AAX of the CAAX motif) are endoproteolytically released. Second, the newly exposed isoprenylcysteine residue is methylated, converting the carboxylate anion of the isoprenylcysteine residue to an ␣-carboxyl methyl ester (6).
S-Prenylation of cysteines within proteins is a highly stable posttranslational modification. The stability of the modification and the abundance of isoprenylated proteins in cells suggested the existence of an enzymatic mechanism for degrading and disposing of prenylcysteine residues. Such an enzyme, termed prenylcysteine lyase (Pcly), 2 was recently identified (7)(8)(9). Pcly is a 505-amino acid flavin adenine dinucleotidedependent thioether oxidase and is located within lysosomes (8,9). Pcly recognizes both farnesylcysteine and geranylgeranylcysteine and their methyl esters with high affinity and cleaves the thioether bond to yield free cysteine (or cysteine methyl ester) and the aldehyde of the isoprenoid lipid (7). Interestingly, Pcly is only active against free prenylcysteines and not prenylcysteine residues within prenylated proteins or peptides (7). In humans, PCLY is expressed ubiquitously, with particularly high levels in brain, liver, kidney, and heart (7).
Although the ability of recombinant PCLY to cleave prenylcysteines in vitro has been amply established (7)(8)(9), neither the biochemical importance nor the physiologic relevance of this enzyme has been fully defined. We do not know, for example, whether the absence of this enzyme would cause prenylcysteines to accumulate in mammalian cells since an alternate "disposal method," transport out of cells by cell-surface P-glycoprotein, has been suggested by the finding that prenylcys-teines are substrates for that transporter (10). If prenylcysteines were to accumulate, would they cause cell toxicity or tissue pathology? Also located within lysosomes is a thioesterase, palmitoyl-protein thioesterase 1 (PPT1), that cleaves fatty acids from acylated cysteines within proteins. Mutations in PPT1 cause a lysosomal storage disease in humans, infantile neuronal ceroid lipofuscinosis (Batten disease) (11). By analogy, Tschantz et al. (8,9) have speculated that the absence of prenylcysteine lyase might also cause a lysosomal storage disease. To address each of these issues and to better define the in vivo importance of this enzyme, we produced and analyzed Pcly-deficient mice and Pcly-deficient cell lines.

EXPERIMENTAL PROCEDURES
Analysis of PCLY Mutants-Site-directed mutagenesis (12) was used to introduce a variety of missense mutations into PCLY cDNA. The mutant PCLY cDNAs were used to produce recombinant baculoviruses, which were then used to infect Sf9 cells (8). A total of 2.5 ϫ 10 6 infected Sf9 cells were harvested and resuspended in 10 mM Tris buffer (pH 7.7) containing 0.2% Triton X-100 and protease inhibitors (8). The cells were incubated on ice for 15 min and then disrupted by drawing the solution up and down ten times through a 27-gauge needle. Samples were centrifuged at 100,000 ϫ g for 90 min at 4°C. The supernatant fluid (detergent extract) was collected, and the protein concentration was determined by a Lowry assay (13). All mutant proteins were expressed at similar levels as judged by immunoblot analysis with a PCLY-specific antibody (8).
PCLY Activity Measurements-PCLY activities in cell or tissue extracts were assessed by thin-layer chromatography (8). Briefly, 10 g of protein was incubated for 30 min at 37°C in 50 mM Tris (pH 7.7) containing 10 M [ 35 S]farnesylcysteine (ϳ40,000 dpm/reaction) in a final volume of 20 l. The reaction was stopped by adding 10 l of the thin-layer chromatography solvent (n-propanol:NH 4 OH:H 2 O (6:3:1, v/v)). Samples were processed by silica-gel thin-layer chromatography with a method that separates the reaction product, [ 35 S]cysteine, from the substrate, [ 35 S]farnesylcysteine (8). To assess the amount of [ 35 S]cysteine in each reaction, the plates were exposed to x-ray film.
A Pcly Gene-targeting Vector-A human PCLY cDNA (GenBank TM accession number AF181490) was used to identify a bacterial artificial chromosome (BAC) clone that spanned the mouse Pcly gene. A 9.8-kb XbaI fragment containing Pcly exons 4 -6 was subcloned into pCR-XL-TOPO (Invitrogen, Carlsbad, CA). The XbaI fragment was used to construct a gene-targeting vector designed to eliminate exon 6 of Pcly and replace it with a neomycin-resistance marker. The vector was generated in pKSloxPNT, which contains polylinker restriction sites, a thymidine kinase gene (tk), and a neomycin-resistance marker flanked by loxP sites. The 5Ј arm of the vector contained 2.5 kb of sequences extending from intron 4 to the end of intron 5; the 5Ј arm was amplified with oligonucleotides 5Ј-TGGCGGATCCGAGAGGGCATGGGAAAAC-ATGCAGTTCTTAGTGCTT-3Ј and 5Ј-CCTCGGATCCGAGGGTCATC-TGACAGCGTCTGGAGGCATTTGGGGC-3Ј and cloned into the BamHI site of pKSloxPNT. The 3Ј arm of the vector began with the 3Ј-untranslated sequences and extended 5.2 kb downstream of the Pcly gene. The 3Ј arm was amplified with oligonucleotides 5Ј-TGAGGCGCGCCCCTC-CCCCGAGCGTCCTGCTCTCCAAGGACCGAGT-3Ј and 5Ј-CGAGGCG-CGCCCGAGCTGCAGCAGAGGGCGGTGGGGGAGGGGGTGC-3Ј and then was cloned into the AscI site of pKSloxPNT. The orientation of both arms was verified by DNA sequencing and restriction endonuclease mapping.
Generation of Pcly-deficient Mice-The sequence-replacement genetargeting vector was linearized with NotI and introduced by electroporation into mouse embryonic stem (ES) cells (strain 129/SvJae). Mouse ES cells were cultured on mitomycin C-treated STO feeder cells in medium containing G418 (250 g/ml) and 1-(2Ј-deoxy-2Јfluoro-␤-D-arabinofuranosyl)-5-iodouracil (0.2 M) (14). Drug-resistant ES cell colonies were picked on the 10th day after electroporation. Targeted colonies (Pclyϩ/-) were identified by Southern blot analysis of XbaIdigested genomic DNA with a 3Ј flanking probe amplified from BAC DNA with primers 5Ј-AATACATCGTCCTTAACAATTTGA-3Ј and 5Ј-TACCATCTGAGCCACACCACCAGC-3Ј. Four targeted clones, each with a single integration event, were injected into C57BL/6 blastocysts (15) to produce chimeric mice, which were bred to establish lines of Pcly knockout mice. All mice described here had a mixed genetic background (ϳ50% C57BL/6 and ϳ50% 129/SvJae). The mice were weaned at 21 days of age, housed in a barrier facility with a 12-h light/12-h dark cycle, and fed a chow diet containing 4.5% fat (Ralston Purina, St. Louis, MO).
Measurement of Prenylcysteines by Mass Spectrometry-Nine-monthold Pclyϩ/ϩ and Pcly-/-mice were anesthetized with avertin, and samples of blood, liver, and brain were obtained. Tissues or blood samples were pooled from 8 -10 mice of each genotype. Levels of farnesylcysteine (FC) and geranylgeranylcysteine (GGC) in tissues of Pclyϩ/ϩ and Pcly-/-mice were determined by mass spectrometry. Briefly, tissue samples were homogenized in 3 volumes (w/v) of 20 mM Tris-Cl (pH 7.7), and the resultant extracts were processed by the bioanalytical laboratory of PPD Discovery (Madison, WI); blood samples were provided directly to the laboratory without prior homogenization. Tissue extracts and blood samples were subjected to the following extraction technique: to each of three independent 25-l samples (tissue homogenate or blood) was added 75 l of acetonitrile containing 5 pmol of geranylcysteine (GC, used as an internal standard to monitor efficiency of extraction and recovery during liquid chromatography/mass spectrometry analysis). Samples were mixed for 5 min at room temperature and centrifuged for 5 min at 5000 ϫ g in a tabletop centrifuge, and the supernatant fluids were transferred to a 0.2-ml polypropylene injection vial and capped. Samples were then injected into a PerkinElmer Life Sciences Sciex API3000 Triple Quadrupole liquid chromatography/ tandem mass spectrometry mass spectrometer fitted with a C18 separation column. The column was developed with a gradient of water to acetonitrile; both solvents contained 0.1% formic acid for ion-suppression during the chromatography. In preliminary trial and calibration separations, the elution positions for GC, FC, and GGC were identified; unambiguous identification of all three prenylcysteines was afforded by direct mass spectrometric analysis of the eluted compounds. Standard curves for concentrations from 10 -5000 nM of each metabolite (FC, GGC) were generated by analysis of pure standards dissolved in acetonitrile. The levels of FC and GGC in each of the samples analyzed were determined from comparison of the peak areas for each compound with the standard curves, and the values were converted to units of nmol/ gram of starting material (i.e. tissue or blood).
Analysis of Pcly-/-Cells and Tissues-Full necropsies were performed on four different 4-and 10-month-old Pcly-/-mice and littermate Pclyϩ/ϩ controls, and more than 30 tissues were examined by routine histology. To determine whether the brains of 10-month-old Pcly-/-mice contained increased amounts of lipofuscin, we examined brain tissue for autofluorescence as described recently for mice lacking palmitoyl protein thioesterase I (19). For these studies, 10-month-old Pclyϩ/ϩ, Pclyϩ/-, and Pcly-/-mice were anesthetized with avertin and perfusion-fixed with 4% paraformaldehyde. Brains were removed and immersed in formalin for 24 h. Samples were processed for paraffin embedding, and sagittal sections (5-10 m) were mounted on polylysinecoated slides. Deparaffinized sections from each sample were examined with a UV-equipped Eclipse E600 microscope (Nikon) at 470 nm (excitation)/525 nm (emission). Finally, to determine if Pcly-/-ES cells had any lysosomal abnormalities, and in particular a lysosomal storage disease phenotype, we examined Pcly-/-and Pclyϩ/ϩ ES cells by transmission electron microscopy. Pcly-/-and Pclyϩ/ϩ ES cells were also examined 10 days after they had been mitotically inactivated with mitomycin C.

RESULTS
Pcly Expression Pattern in Mice-Pcly expression in wildtype mouse tissues was determined by multiple-tissue Northern blot analysis. A major Pcly transcript, ϳ4 kb in length, was detected in all tissues. The highest levels of Pcly expression were in the liver, kidney, heart, and brain (Fig. 1A). In addition to the 4-kb transcript, a 5-kb transcript was detected in liver, kidney, and brain; a smaller 1.5-2-kb band was present in liver and kidney.
Production of Pcly-deficient Mice-We identified a strain 129/Sv BAC spanning the entire Pcly gene and also identified within GenBank TM an expressed sequence tag containing the complete mouse Pcly coding sequences (accession no. AK004799). The protein-coding sequences in the BAC and the expressed sequence tag were identical (Fig. 1B). To inactivate Pcly, we adopted a strategy to eliminate the carboxyl-terminal portion of the protein (sequences encoded by exon 6, see Fig.  1B). We were confident that removal of this large segment of the protein would cause a null mutation because site-directed mutagenesis experiments revealed that several different missense mutations in that segment of the gene abolished Pcly enzymatic activity (Fig. 1C). After electroporation of the genetargeting vector (Fig. 1D) into ES cells, we identified seven targeted clones (from a total of 294). Four ES clones were used to produce Pclyϩ/-and Pcly-/-mice. No Pcly transcripts were detectable in Pcly-/-tissues when the Northern blot was hybridized with exon 6 sequences. Small amounts of a truncated transcript were observed when the Northern blot was hybridized with a cDNA probe containing sequences from exons 4 -5 (Fig. 1E).
Increased Susceptibility of Pcly-deficient Fibroblasts to Prenylcysteine Compounds-Pcly-deficient cells-both ES cells and fibroblasts-appeared normal morphologically. The growth rates of Pcly-/-and Pclyϩ/ϩ fibroblasts were indistinguishable under standard culture conditions (Fig. 3A), as were those of Pcly-/-and Pclyϩ/ϩ ES cells (not shown). Because of the known pharmacological effects and toxicity associated with treatment of cells with prenylcysteines (20, 21), we hypothesized that the deficiency in Pcly activity might render Pcly-/fibroblasts more susceptible to toxicity or growth inhibition by exogenous prenylcysteine compounds. To test this possibility, Pcly-/-and Pclyϩ/ϩ fibroblasts were grown for 4 days in the presence of increasing concentrations (up to 50 M) of either FCME (Fig. 3B) or GGCME (Fig. 3C). While treatment with both these prenylcysteines retarded cell growth in fibroblasts regardless of the genotype, the Pcly-/-cells were significantly more sensitive to the growth inhibition than Pclyϩ/ϩ cells (Fig.  3, B and C). We considered the possibility that the observed differences could have been due to the release of methanol from FCME and GGCME within the cell culture medium. However, we doubt that this is the case. Adding methanol (at concentrations up to 250 M) to the cell culture medium did not retard the growth of either Pclyϩ/ϩ or Pcly-/-fibroblasts (data not shown).
Biochemical Analysis of Pcly-/-Mice-Direct measurements of Pcly activity in extracts of tissues from wild-type mice showed that the highest activity levels were in the liver, followed by kidney, brain, and heart (Fig. 4A). Consistent with the data obtained from the Pcly-/-cells (see Fig. 2), enzyme activities were markedly reduced in tissues derived from Pcly-/mice (reductions of 90%, 97%, 92%, and 74% in liver, kidney, brain, and heart, respectively) (Fig. 4A).
To determine if the reduction in Pcly activity levels in Pcly-/mice would lead to an accumulation of the substrates of the enzyme (i.e. endogenous prenylcysteines derived from prenylprotein turnover) instead of being simply transported out of the tissue, we used mass spectrometry to measure levels of prenylcysteines in blood, brain, and liver of 9-month-old Pcly-/-and Pclyϩ/ϩ mice. Indeed, while blood levels of prenylcysteines were virtually unaffected in the setting of Pcly-deficiency, this analysis revealed an impressive accumulation of FC in liver and brain of Pcly-/-mice (50-and 40-fold increase, respectively) (Fig. 4B). Geranylgeranylcysteine also accumulated in liver (10-fold) and brain (30-fold) of Pcly-/-mice (Fig. 4C). We did not observe an accumulation of prenylcysteines in the blood of Pcly-/-mice, perhaps because few prenylcysteines are released into the blood or because those substances are cleared rapidly from that compartment.
We also attempted to quantify levels of prenylcysteine methyl esters in tissues by mass spectrometry, but were unable to detect these compounds due to the lability of these methyl esters in tissue extracts. When FCME and GGCME were added exogenously to tissue extracts, they were rapidly and completely converted to the corresponding demethylated compounds (i.e. FC, GGC) (not shown). Hence, we believe that the analysis of FC and GGC in the tissue extracts accurately reflected the total level of farnesylated and geranylgeranylated cysteines in these tissues.
Pathologic Analysis of Pcly-/-Mice-The accumulation of prenylcysteine residues did not appear to result in significant consequences. Blood chemistries including calcium, phosphate, glucose, cholesterol, triglycerides, alanine aminotransferase, aspartate aminotransferase, and creatinine were normal in Pcly-/-mice; serum lipids in Pcly-/-and Pclyϩ/ϩ mice did not differ on a high-fat diet (not shown). In addition, we analyzed more than 30 different tissues from 4-and 10-month-old Pcly-/-mice with routine histologic stains, and observed no abnormalities.
In the case of palmitoyl protein thioesterase I-deficient mice, the presence of lipofuscin was particularly apparent by fluorescence microscopy (19). To determine if increased levels of prenylcysteines might lead to increased autofluorescence within tissues, we examined brain sections from 10-month-old Pclyϩ/ϩ and Pcly-/-mice with a UV-equipped microscope. Regardless of the genotype, autofluorescence levels were highest in the cerebellum, but lower levels of autofluorescence could also be detected in pons, hippocampus, and frontal cortex. However, we did not observe any differences in the levels of autofluorescence in brains of Pclyϩ/ϩ and Pcly-/-mice (not shown).
To look for abnormalities in intracellular organelles, we examined Pclyϩ/ϩ and Pcly-/-ES cells by transmission electron microscopy, after staining the cells with imidazole-osmium tetroxide (which stains unsaturated lipids). No differences were observed in Pcly-/-and Pclyϩ/ϩ ES cells. The lysosomes appeared normal, and there was no increase lipid-staining material anywhere within the cell (not shown).

FIG. 1. Pcly expression pattern in the mouse and a strategy to inactivate Pcly in mice.
A, mouse multiple-tissue poly(A) ϩ RNA blot (Clontech) showing the tissue pattern of Pcly expression in wild-type mice. The blot was hybridized with a 32 P-labeled Pcly cDNA (top panel) probe, and the blot was exposed to x-ray film for 8 h at -80°C. The same blot was also probed with a glyceraldehyde-3-phosphate dehydrogenase cDNA (bottom panel). B, alignment of the predicted amino acid sequences for human and mouse Pcly. An asterisk indicates conserved amino acid residues. Boxed area represents the region encoded by exon 6. Arrows designate the sites where point mutations have been shown to inactivate the protein.
C, PCLY mutants were tested for enzymatic activity as described under "Experimental Procedures." Results are expressed as the percentage of wild-type enzyme activity levels. D, sequence-replacement strategy to inactivate Pcly. Genotyping of cell lines and mice was performed by Southern blot with a 3Ј flanking probe. A representative Southern blot illustrating Pclyϩ/ϩ, Pclyϩ/-, and Pcly-/-ES cells is shown. E, Northern blots of total RNA from the liver, kidney, brain, and heart of Pclyϩ/ϩ and Pcly-/-mice. Three probes were used: a Pcly cDNA probe containing sequences from exon 4 -5 (top panel), a Pcly cDNA probe containing exon 6 sequences (middle panel), and a glyceraldehyde-3-phosphate dehydrogenase probe (bottom panel). In the top panel, arrows indicate the positions of 1-and 2-kb truncated Pcly transcripts in Pcly-/-mice. The 1.5-2-kb band in wild-type liver and kidney, observed with poly(A) ϩ RNA (see A), was never observed in Northern blots with total RNA samples.

DISCUSSION
Protein prenylation, unlike many other posttranslational modifications, is a stable modification of proteins. Prenylated proteins have a half-life of ϳ20 h and comprise ϳ2% of total cellular proteins (7,8). The relatively short half-life of prenylproteins, along with the stability of the modification, suggested that cells must possess a mechanism for disposal of prenylcysteines. Previous studies identified an enzyme, Pcly, capable of degrading prenylcysteines in vitro (7)(8)(9), but none of these studies addressed whether the enzyme was truly relevant or important for prenylcysteine metabolism in vivo. In the current study, we addressed the biochemical and physiologic importance of Pcly with a gene knockout experiment and were able to show, unequivocally, that Pcly is indeed relevant to prenylcysteine metabolism in mammalian tissues.
Our studies establish that Pcly is the main enzymatic activity for degrading prenylcysteines in mammalian cells. Prenylcysteine lyase activity was markedly reduced in Pcly-/-ES cells, in Pcly-/-fibroblasts, and in the tissues of Pcly-/-mice. In the majority of our experiments, enzymatic activities in Pcly-/-cells and tissues were at or just slightly above the background levels for our assay. Therefore, we cannot exclude the possibility that very low levels of a redundant bioactivity exist in mammalian cells. We have looked for a Pcly family member within the expressed sequence tag and genomic databases, for it is conceivable that another yet-to-be-identified gene encodes a protein that can metabolize prenylcysteines in the setting of Pcly deficiency. In the course of this data mining, we found a hypothetical human protein (GI:13278789) that exhibits 35% homology with a 330-amino acid segment of human PCLY. It would be of interest to determine whether that clone encodes an enzyme with some activity against prenylcysteines (i.e. a PCLY2). This possibility may not be farfetched. PPT1 is a lysosomal enzyme that removes palmitoyl groups from cysteines in proteins, and its absence causes infantile neuronal ceroid lipofuscinosis (11). Recently, cloning of a PPT2 gene has been reported (22). Despite the fact that PPT1 and PPT2 share only 18% identity at the amino acid level, PPT1 and PPT2 have comparable palmitoyl-CoA thioesterase activities. Interestingly, PPT2 appears to display a distinct substrate specificity in cells, and its expression does not correct the metabolic defect in PPT1-deficient cells (22).
There are two other reasons to think that Pcly is the principal route for disposing of prenylcysteines in cells, even if minute amounts of another prenylcysteine-degrading enzyme ex- ist. First, mass spectrometric analysis showed a significant accumulation of both farnesylcysteine and geranylgeranylcysteine in the brain and liver of Pcly-/-mice compared with tissues from aged-matched control mice. Second, Pcly-/-fibroblasts, when challenged with increasing doses of exogenous prenylcysteines, displayed increased sensitivity to the toxicity of these compounds.
We did not note an accumulation of prenylcysteines in the blood of Pcly-/-mice. We do not know the reason for the absence of prenylcysteine accumulation in blood, although we hypothesize that the production of prenylcysteines is probably minimal in erythrocytes. Alternatively, it is possible that prenylcysteines readily diffuse away from blood cells into the plasma and are rapidly cleared by the kidney.
One could speculate that the accumulation of farnesylcysteine and geranylgeranylcysteine in brain and liver might inhibit the methylation of prenylated proteins by isoprenylcysteine carboxyl methyltransferase (6). However, we do not consider this to be likely, because the concentrations of the prenylcysteines in the Pcly-/-cells were in the picomolar range, significantly below the micromolar concentrations required to inhibit isoprenylcysteine carboxyl methyltransferase (6,20,23,24). In keeping with this prediction, we found that extracts from Pcly-/-cells and Pclyϩ/ϩ cells were equally effective in methylating recombinant farnesyl-K-Ras. 3 We had speculated that the absence of prenylcysteine lyase in mammalian lysosomes might cause, either directly or indirectly, a lysosomal storage disease (8,9) akin to that occurring in the setting of PPT1 mutations (11,19,25). A deficiency in PPT1 leads to the accumulation of a finely granular autofluorescent sudanophilic storage material (lipofuscin) in the brain (19,25). However, no such pathologic findings were noted in Pcly-/-mice. We were unable to detect histologic abnormalities in a survey of more than 30 tissues from Pcly-/-mice, and there was no increase in autofluorescence within the brains of Pcly-/-mice. A key difference between PPT1 and Pcly is that PPT1 cleaves palmitate residues from proteins, whereas Pcly acts only on free prenylcysteines after the protein has been degraded. The accumulation of free prenylcysteines in the setting of Pcly deficiency (as opposed to palmitoylated proteins in the setting of PPT1 deficiency) apparently has little proclivity to form lipofuscin either directly by inducing aggregation of proteins or indirectly by interfering with the action of other lysosomal enzymes.
The absence of overt pathology in Pcly-deficient mice raises the possibility that the cell might have more than one strategy for dealing with prenylcysteines. In yeast, at least one isoprenylated peptide, a-factor, is transported out of cells by an ABC transporter, Ste6p (26). A related protein, P-glycoprotein, has been shown to be capable of transporting prenylcysteines out of human cells (10). In the setting of Pcly deficiency, it is conceivable that P-glycoprotein-mediated transport of prenylcysteines out of cells partially prevents the intracellular accumulation of prenylcysteines and thereby prevents the pathologic consequences of the enzyme deficiency. However, it is far from clear that P-glycoprotein functions in a meaningful way to dispose of prenylcysteine residues in normal cells. For example, prenylation of proteins is ubiquitous, but most mammalian tissues do not express appreciable levels of P-glycoprotein (27).
In conclusion, our data support the hypothesis that catabolism of prenylcysteines in mammals requires Pcly. In the absence of that enzyme, the cell's ability to cleave prenylcysteines is severely compromised, and prenylcysteine residues accumulate. Despite this accumulation, Pcly-deficient mice are born at the expected mendelian frequency, are fertile, and are free of obvious pathology.