(Received for publication, April 17, 1997, and in revised form, May 15, 1997)
From the Department of Biology, San Diego State University, San Diego, California 92182
To date, isopentenyl diphosphate:dimethylallyl diphosphate isomerase (IPP isomerase; EC 5.3.3.2) is presumed to have a cytosolic localization. However, we have recently shown that in permeabilized cells lacking cytosolic components, mevalonate can be converted to cholesterol, implying that all of the enzymes required for the conversion of mevalonate to farnesyl diphosphate are found in the peroxisome. To provide unequivocal evidence for the subcellular localization of IPP isomerase, in this study, we have cloned the rat and hamster homologues of IPP isomerase and identified the signal that targets this enzyme to peroxisomes. In addition, we also demonstrate that IPP isomerase is regulated at the mRNA level.
The isoprenoid biosynthetic pathway is ubiquitous to all living organisms. A few of the important end products of this complex pathway include: dolichols; vitamins A, D, E, and K; steroid hormones; carotenoids; bile acids; and cholesterol (1). The enzyme isopentenyl diphosphate:dimethylallyl diphosphate isomerase (IPP isomerase1; EC 5.3.3.2) plays a crucial role in this pathway by catalyzing the interconversion of isopentenyl diphosphate (IPP) to its highly electrophilic isomer, dimethylallyl diphosphate (2). These two isomers are the building blocks for the successive head-to-tail condensation reactions that result in the synthesis of farnesyl diphosphate (FPP), and ultimately, cholesterol (3).
Recently, it has been shown by our group and others that peroxisomes contain a number of enzymes involved in cholesterol biosynthesis that were previously thought to be cytosolic. Specifically, peroxisomes have been shown to contain acetoacetyl-CoA thiolase (4, 5), 3-hydroxy-3-methylglutaryl coenzyme A synthase (6), 3-hydroxy-3-methylglutaryl coenzyme A reductase (7-9), mevalonate kinase (10, 11), phosphomevalonate kinase (12), mevalonate diphosphate decarboxylase (12), and FPP synthase (13). Both mevalonate kinase and FPP synthase seem to be localized predominantly, if not exclusively, to peroxisomes (11, 13). To date, IPP isomerase is presumed to have a cytosolic localization (1); however, the following three observations have led us to believe that the enzyme is localized to peroxisomes: (i) in permeabilized cells lacking cytosolic components, mevalonate can be converted to cholesterol in equal amounts to that observed in nonpermeabilized cells, therefore suggesting that the cytosol does not contain enzymes necessary for the conversion of mevalonate to FPP (12); (ii) IPP isomerase activity in tissues from patients with peroxisome-deficient diseases (Zellweger and neonatal adrenoleukodystrophy) is 50% of that found in tissues from control patients (13); and (iii) the deduced amino acid sequence from the human isomerase cDNA, which has been recently cloned (14) and characterized (15), contains two putative peroxisomal targeting sequences.
At the C-terminal end of human isomerase is a putative peroxisomal targeting sequence 1 (PTS1) consisting of YRM (single-letter amino acid notation), and at the N-terminal end is a putative PTS2 sequence consisting of HLX5QL (where X designates any amino acid). The consensus sequence for the PTS1 motif is (S/A/C)(K/H/R)(L/M) (16); however, many tripeptide combinations that do not adhere to this consensus sequence were able to target Saccharomyces cerivisiae malate dehydrogenase to yeast peroxisomes (17). The PTS2 consensus sequence is usually near the N-terminal end of the protein and is the nonapeptide (R/K)(L/V/I)X5(H/Q)(L/A) (18).
In this study, we have cloned the rat and hamster homologues of IPP isomerase and identified the signal that targets this enzyme to peroxisomes. In addition, we demonstrate that IPP isomerase is regulated at the mRNA level in liver from rats treated with compounds known to modulate the levels of cholesterol biosynthesis.
Restriction enzymes were
purchased from New England Biolabs. Chemically competent
Escherichia coli InvF
, electrocompetent E. coli TOP10F
, molecular weight markers, TA cloning kit, cDNA synthesis copy kit, and all cloning and expression vectors were obtained from Invitrogen (San Diego, CA). Microbiological reagents were
from Difco. Cholestyramine (Questran) was obtained from Bristol Laboratories. Automated DNA sequencing was performed, and DNA synthetic
oligonucleotides were synthesized by the SDSU Microchemical Core
Facility (San Diego, CA). The 5
RACE kit was purchased from Life
Technologies, Inc. and used according to the provided instructions.
DNA probes were labeled with [32P]dCTP using the Nick Translation kit (Boehringer Mannheim). Zeta Probe GT membrane (used for Northern and Southern analysis), C/P Lift membrane (used for plasmid library screening), and Trans-Blot transfer medium (used for Western analysis) were purchased from Bio-Rad. Western transfer was performed as described (11). Northern and Southern transfers were performed using the Stratagene Posiblot pressure blotter and pressure control station (Stratagene) as described by the manufacturer's protocol. Hybridization conditions for Northern analysis, Southern analysis, or plasmid library screening were as follows. Membranes were incubated overnight at a temperature of 50 °C with radiolabeled probe in a solution of 50% formamide, 6% SDS, 100 mM Na2PO4, and 200 mM NaCl; membranes were washed for 15 min, three times, at 55 °C using 0.1% SDS and 0.5 × SSC, and exposed on a PhosphorImager screen (Molecular Dynamics).
Standard molecular biology techniques (19) were routinely used for cloning and restriction endonuclease digestions. Plasmid mini- and large scale preps were performed using QIAprep Spin Miniprep kit and Qiagen Plasmid Midi kit, respectively (Qiagen). DNA or RNA purifications from agarose gels were performed using the Gene Clean kit (Bio 101). Secondary antibodies used in immunofluorescence were from Molecular Probes. Anti-HA monoclonal antibody (12CA5) was from Boehringer Mannheim. All other biochemicals were purchased from Sigma.
DNA AmplificationWhere stated, DNA amplification was
carried out by the polymerase chain reaction (PCR). In these
experiments, the reaction mixture contained 5 µL of 10 × Taq polymerase PCR buffer (Boehringer Mannheim);
deoxynucleotide triphosphates, 200 µM; DNA
oligonucleotide primers, 0.2 mM; template DNA, 30 ng
(except for 5 RACE, where templates were added as per kit
instructions); and 1.25 units of Taq polymerase in a final
reaction volume of 50 µl. Thermocycling conditions were as follows:
initial denature, 94 °C for 3 min; 40 rounds of cycling, 94 °C
for 15 s, 55 °C for 45 s, and 72 °C for 50 s; final
extension, 72 °C for 10 min. The reactions were performed in a
Perkin-Elmer 2400 thermocycler.
Male Sprague-Dawley rats (150-240 g) and a New Zealand White rabbit were maintained on a 12-h light/dark cycle. Three rats each were fed the standard laboratory diet; or a diet supplemented with a combination of 5.0% cholestyramine for 2 days followed by 5.0% cholestyramine plus 0.1% mevinolin for 3 days; or the standard diet supplemented with 2% cholesterol for 15 days. Rats were not fasted prior to decapitation (4-6 h into light cycle), and 6.0 g of liver were extracted from each rat; 2.0 g used for total RNA and subsequent poly(A)+ RNA isolation, and 4.0 g were used for preparation of liver homogenates.
Phagemid cDNA Library Construction and Isolation of Full-length cDNA ClonesRat liver poly(A)+ RNA
from cholestyramine plus mevinolin (C+M) fed rats was used to construct
an oligo(dT)-primed, bidirectional phagemid cDNA library as per the
Copy Kit cDNA synthesis kit (Invitrogen). A 667-bp PCR product
amplified from C+M rat liver cDNA using the primers Av71 (5-TAA
CCA CCT CGA GAA GCA ACA GGT TC-3
) and Kp711 (5
-TCA CAT TCG GTA CCT
TTT CTC ATG GTC-3
) was used as a probe to screen approximately 500,000 colonies from a rat liver cDNA library. Four clones were isolated
and sequenced (both strands in triplicate), all of which were identical
with the exception of the length of the 5
untranslated region (UTR).
To determine the longest 5
UTR, 5
RACE was performed using primer
Kp711 as GSP1 and degenerate primer E207 (5
-ATR TAR TCD ATY TCR TGY
TC-3
) as GSP2.
The full-length hamster CHO K1 clone (termed pHamIPPI) was obtained by
assembling the 5 and 3
RACE products amplified from CHO K1
poly(A)+ RNA. 5
RACE on the CHO K1 template was performed
using primer H993R (5
-TGC CTA TTC ATC ACA CTC ATC CT-3
) as GSP1 and
primer Kp711 as GSP2. The 3
RACE was performed identically to the 5
RACE except 1 µg of CHO K1 poly(A)+ RNA was primed with
oligonucleotide TUP (5
-GGC CAC GCG TCG ACT AGT
AC(T)16-3
), reverse transcribed, then subsequently
amplified by PCR using primer set UAP20 (5
-GGC CAC GCG TCG ACT AGT
AC-3
) and Av71. Both strands of all subcloned fragments were sequenced in triplicate.
Total RNA was prepared from nine individual rat livers by the method of Chomczynski and Sacchi (21). Extraction of poly(A)+ RNA was done using oligo(dT)-cellulose type 3 (Collaborative Biomedical Products) according to the supplier's protocol. For each lane, 10 µg of poly(A)+ RNA were electrophoresed through a 1.0% agarose gel. The gel was transferred to nylon membrane and cross-linked in a Stratagene Crosslinker according to the manufacturer's instructions. The membrane was probed with a [32P]dCTP-labeled 667-bp isomerase PCR product.
Southern Blot AnalysisGenomic DNA was purified from rat liver using the QIAamp tissue kit (QIAGEN), and 15 µg of DNA were digested with each indicated restriction enzyme. The samples were electrophoresed on a 0.8% agarose gel. The gel was transferred to nylon, and the membrane was cross-linked using a Stratagene Crosslinker. The blot was probed with a radiolabeled 180-bp NcoI/BamHI restriction fragment of rat IPP isomerase.
Fusion Protein Construction and PurificationA 280-bp
NcoI digestion product of rat isomerase was subcloned into
the prokaryotic expression vector pThioHisB. The vector was transformed
into TOP10 E. coli, and a single colony was grown overnight
at 37 °C in 10 ml LB + ampicillin media. The overnight culture was
then added to 1 liter of LB + ampicillin media and grown for 3 h.
Ispropyl-1-thio--D-galactopyranoside was added to a
final concentration of 1 mM, and the culture was grown for an additional 4.5 h. The bacterial cells were pelleted by
centrifugation, resuspended at a 10 × concentration in PBS, and
treated with 4 successive freeze/thaw sonications. DNase I was added to
a final concentration of 0.5 unit/ml, and the sample was centrifuged. The lysate supernatant was discarded, and the pellet was washed once
with RIPA buffer (10 mM NaH2PO4, pH
7.5, 5 mM EDTA, 5 mM EGTA, 100 mM
NaCl, 1% (v/v) Triton X-100, 0.1% (w/v) SDS, 0.5% (w/v) sodium
deoxycholate, and 0.01% NaN3). After centrifugation, the
pellet was then washed three times with PBS + 0.1% SDS. The remaining
pellet was solubilized as described (11) and electrophoresed on a 12%
SDS-polyacrylamide gel. The proteins in the gel were stained with
Coomassie Brilliant Blue R. The band corresponding to the
thioredoxin-isomerase fusion protein was excised and electroeluted using a Centricon-10 and Centrilutor apparatus (Amicon), followed by
concentration and equilibration in PBS in the Centricon-10.
Antibodies were raised in a rabbit. Following collection of preimmune serum, the rabbit received 0.1 mg of electroeluted thioredoxin-isomerase fusion protein that was emulsified 2:1 with Hunter's Titer Max (CytRx Corp.) and injected intramuscularly at two sites. After 1 month, the rabbit was given a second set of injections. One month after the second set of injections, blood was taken, and serum was recovered.
Liver Homogenate Preparation and Western Blot AnalysisLivers from the rats fed a cholestyramine plus mevinolin diet, normal diet, and cholesterol diet (each in triplicate) were homogenized in homogenization buffer (0.25 M sucrose, 5 mM Tris-HCl, 1 mM EDTA, 0.1% ethanol, containing 1 µg/ml aprotinin, 10 µg/ml cycloheximide, 0.1 M dithiothreitol, 20 mM leupeptin, 1 µg/ml antipain, 1 µg/ml chymostatin, 1 µg/ml pepstatin A, and 0.1 M phenylmethylsulfonyl fluoride, pH 7.5, at 4 °C). The protein concentration of the homogenate was determined by the BCA method (Pierce Chemical Co.) using bovine serum albumin as a standard, and 200 µg of protein from each sample were electrophoresed on a 12% SDS-polyacrylamide gel as described (20). The gel was transferred to nitrocellulose, and immunoblot analysis was performed as described previously (11). The anti-thioredoxin-isomerase antiserum was added at a 1:1000 dilution, and the secondary antibody (horseradish peroxidase protein A conjugate) was used at a 1:5000 dilution. The blot was then processed using the ECL kit (Amersham).
Internal Epitope Tagged Expression Vector ConstructionTwo
DNA oligonucleotides were used to introduce the HA epitope into the
hamster isomerase. The first primer, termed COFRev (5-ATC AGG
GACGTC ATA AGG GTA AAT TCG TGT TAG ATA ATG C-3
, AatII underlined) was used in conjunction with primer T7 (5
-AAT ACG ACT CAC
TAT AGG-3
) to PCR-amplify a 522-bp region using the full-length clone,
pHamIPPI, as template for the reaction. The product was TA cloned
(pCR2.1), and the resulting vector was termed pCOFRT7. The second
primer, COFFor (5
-TAC TAT GACGTC CCT GAC TAT GCT TGG GGT
GAA CAT GAA ATT GAT T-3
, AatII underlined) was used in
conjunction with primer M13Rev (5
-CAG GAA ACA GCT ATG ACC-3
) to PCR
amplify a 2014-bp region using pHamIPPI as template. The product was
double digested with KpnI and AatII, gel
purified, and ligated into pCOFRT7, which had been previously cut with
AatII and KpnI. The resulting construct contained
a DNA sequence such that, when translated, would code for the amino
acids P136, Y137, D138,
V139, P140, D141, Y142,
and A143 instead of the native isomerase amino acids
Y136, K137, A138, Q139,
S140, D141, G142, and
I143. Because hamster isomerase has a Tyr at position 138, the final result was a DNA sequence encoding the HA epitope from amino
acids 135 to 143 (YPYDVPDYA). The internally HA-tagged isomerase
cDNA construct was then digested with NotI and
HindIII, and the 1173-bp digestion product was gel purified
and ligated into NotI and HindIII sites of the
mammalian expression vector pcDNA3.1. The subsequent clone was
termed pIsoHA, and the sequence of the insert was confirmed by
sequencing.
A deletion of the putative C-terminal PTS1 targeting signal was made by
PCR-amplifying the template, pIsoHA, with primer T7 and primer
IsoPTS1 (5
-TCA AAGCTT ATT CTT TAT ATT TTC TCG TG-3
; HindIII site underlined, stop codon in boldface).
The resulting 716-bp product was digested with EcoRI and
HindIII and ligated into the EcoRI and
HindIII sites of pcDNA3.1. This vector was termed
pIsoHA
PTS1.
Transformed fibroblast cell lines from a patient with Zellweger syndrome, FAIR-T, belonging to complementation group 2, and a patient with rhizomelic chondrodysplasia punctata, BRO-T, belonging to complementation group 11 were kindly provided by Dr. Suresh Subramani (University of California, San Diego, CA). A control skin fibroblast cell line was obtained from Dr. Peter Pentchev (National Institutes of Health, Bethesda, MD) and used during passage 16. All cells were grown in monolayer at 37 °C in an atmosphere of 5% CO2. Cell lines from the patients with Zellweger syndrome and rhizomelic chondrodysplasia punctata were maintained in Dulbecco's modified Eagle's minimal essential medium (Life Technologies, Inc.) containing 100 units/ml penicillin, 100 µg/ml streptomycin sulfate, and 10% (v/v) fetal calf serum (Life Technologies, Inc.). CHO K1 cells were maintained in the same medium, except this medium was supplemented with 5% fetal bovine serum instead of 10% fetal bovine serum. Cells were transfected using Perfect Transfection Pfx-7 lipid (Invitrogen) according to the manufacturer's protocol.
Immunofluorescence MicroscopyCells on coverslips were washed in PBS and fixed in 3.0% paraformaldehyde in PBS for 15 min. Cells were permeabilized with 1% Triton X-100 in PBS for 5 min and then washed with 0.1% Tween 20 in PBS (also used for subsequent washes). A mixture of rabbit anti-catalase and mouse anti-HA (both at a final dilution of 1:50) was applied to the coverslips and incubated for 60 min. The cells were washed, and a mixture of secondary reagents consisting of fluorescein conjugate of goat anti-rabbit IgG (H + L) antibody (at a final dilution of 1:100) and Texas Red-X conjugate of goat anti-mouse IgG (H + L) antibody (at a final dilution of 1:200) was applied to the coverslips for 60 min. The cells were washed extensively, and the coverslips were mounted on microscope slides for observation with a Nikon fluorescence microscope.
Based on the
cDNA sequence of human IPP isomerase (14), two oligonucleotide
primers were constructed and used in RT-PCR on rat liver
poly(A)+ RNA to amplify 667 bp of the isomerase coding
region. This PCR product was used as a probe to screen an oligo
dT-primed cDNA library made from liver of a rat fed a C+M diet.
These two drugs up-regulate mRNA levels of
3-hydroxy-3-methylglutaryl coenzyme A reductase and a number of other
enzymes involved in the cholesterol biosynthetic pathway (22). A total
of approximately 500,000 colonies were screened, and four individual
clones were isolated and sequenced. All four clones had 5 UTRs of
different lengths, varying from 72 to 385 nucleotides (data not shown).
To identify the size of the longest 5
UTR of rat isomerase, 5
RACE
was performed on poly(A)+ RNA isolated from the liver of a
C+M-fed rat. The PCR product indicated that the longest 5
UTR was 385 nucleotides, giving a size of 1182 bp for the complete cDNA of rat
IPP isomerase (data not shown).
The full-length cDNA for hamster IPP isomerase was isolated
utilizing 5 and 3
RACE. The experiments were performed using poly(A)+ RNA isolated from CHO K1 cells. Eight individual,
overlapping, partial clones were isolated and sequenced (data not
shown).
The deduced amino acid sequences from both the rat and hamster
cDNAs were compared with the deduced amino acid sequence of human
IPP isomerase (Fig. 1). The following results were
obtained: (i) hamster isomerase has an 89.4% identity and a 95.6%
similarity to human isomerase; (ii) rat isomerase has an 86.3%
identity and a 92.5% similarity to human isomerase; and (iii) hamster
and rat isomerases have a 91.6% identity, and a 96.0% similarity to
one another.
Dietary Regulation of Rat Liver IPP Isomerase mRNA Levels
To investigate if IPP isomerase mRNA levels are
modulated by compounds known to affect cholesterol biosynthesis, we
performed Northern blot analysis on poly(A)+ RNA isolated
from rats fed a control diet, a diet containing cholesterol, or a diet
containing C+M. When probed with a radiolabeled isomerase fragment, a
2.4-kb transcript was observed in each lane (Fig.
2A). Relative to the control diet, the 2.4-kb
transcript was significantly increased in the C+M diet, whereas in the
cholesterol diet, the 2.4-kb transcript was significantly decreased.
Additionally, a second message of 1.1 kb was detected in the lanes from
the C+M diet. The presence of this message and the observation that the
2.4-kb message is significantly larger than the size predicted by
sequencing of 5 RACE products and library clones led us to further
analyze both of these transcripts. We determined by RT-PCR on gel
purified rat liver poly(A)+ RNA corresponding to 2.4 and
1.1 kb that both messages contained the coding region of rat IPP
isomerase (data not shown). A possible explanation for the size
discrepancy is alternative splicing of the 3
UTR to give a 3
UTR size
of approximately 1 kb. Including a poly(A)+ tail of 200 bp,
this would yield a final transcript size of approximately 2.4 kb. We
have performed 3
RACE on isolated rat liver poly(A)+ RNA
to investigate possible alternative splicing of the 3
UTR; however,
the results were inconclusive (data not shown).
To investigate if IPP isomerase protein levels are modulated by
compounds known to affect cholesterol biosynthesis, we performed immunoblot analysis on liver homogenates isolated from rats fed a
control diet, a diet containing cholesterol, or a diet containing C+M.
When blotted using antiserum against the thioredoxin-isomerase fusion
protein, there was a marked increase of a 29-kDa band in the lanes
corresponding to the C+M diet as compared with the normal diet and a
diet containing cholesterol (Fig. 3). The isomerase band
of 29 kDa is in excellent agreement with the molecular weight of human
IPP isomerase when overexpressed in E. coli, as judged by
SDS-PAGE (15). There was no visible difference in the 29-kDa band
between the lanes corresponding to the cholesterol and normal diets.
These data, and the data obtained from Northern analysis, indicate that
a diet containing C+M up-regulates rat liver IPP isomerase both at the
mRNA and protein levels.
IPP Isomerase Is Present as a Single Copy Gene in the Rat Genome
To identify whether IPP isomerase is present as a single
copy gene, a Southern blot was performed using rat genomic DNA. The blot was probed with a 180-bp rat isomerase fragment, chosen due to its
high degree of conservation among several species (15). The Southern
blot revealed a single band hybridizing in each lane (Fig.
4), providing evidence that rat IPP isomerase is present as a single copy gene. These data further suggest that the two transcripts (2.4 and 1.1 kb) visualized by Northern analysis (Fig. 2A) originated from a single gene.
IPP Isomerase Is Targeted to Peroxisomes by a PTS1 Motif
Because all of the mammalian isomerases have putative PTS1
and PTS2 motifs, we first constructed a eukaryotic expression vector containing the full coding sequence of hamster IPP isomerase with an
internal HA epitope tag. This vector was termed pIsoHA (Fig. 5A). An internal HA tag was chosen instead of
C- or N-terminal HA tags so as not to disrupt the potential PTS1 or
PTS2 function.
To determine in which subcellular compartment IPP isomerase is
localized, we transfected pIsoHA into control human fibroblast cells.
The cells were then simultaneously immunolabeled with anti-catalase antibody (Fig. 6A) and anti-HA antibody (Fig.
6B). The immunofluorescence pattern for catalase was
superimposable over the pattern obtained with the HA antibody. Similar
results were obtained in CHO cells when labeled with anti-catalase
antibody (Fig. 6C) and anti-HA antibody (Fig.
6D). These results show that hamster IPP isomerase is
colocalized with catalase to peroxisomes.
To determine which of the PTS targeting signals are used by the hamster isomerase, we used two distinct cell lines derived from patients with peroxisomal disorders. One cell line, BRO-T, shown to be deficient in the peroxisomal import of only PTS2 proteins, was from a patient diagnosed with rhizomelic chondrodysplasia punctata, belonging to complementation group 11 (23). The other cell line, FAIR-T, shown to be deficient in the peroxisomal import of both PTS1 and PTS2 proteins, was from a patient diagnosed with Zellweger syndrome, belonging to complementation group 2 (23). To first determine if the putative PTS2 is responsible for peroxisomal import, the construct pIsoHA was transfected into the PTS2-deficient cell line BRO-T. The cells were double-labeled with anti-catalase antibody (Fig. 6E) and anti-HA antibody (Fig. 6F). The data again demonstrated a superimposable punctate pattern when labeled with anti-catalase and anti-HA antibodies. These results suggest that IPP isomerase does not use the putative PTS2 for peroxisomal targeting.
To determine if the isomerase is targeted via the PTS1 machinery, we transfected pIsoHA into the FAIR-T cell line and again immunolabeled the cells with anti-catalase antibody (Fig. 6G) and anti-HA antibody (Fig. 6H). As expected, catalase labeling showed a cytosolic fluorescence pattern, i.e. catalase is imported by the PTS1 mechanism (24), hence it is localized to the cytosol in cells deficient in this import. A similar cytosolic pattern was revealed over the same field of cells labeled with anti-HA antibody. These results suggest that the transfected isomerase uses PTS1.
To provide direct evidence that the putative PTS1 of isomerase is
necessary for peroxisomal targeting, a second expression vector was
constructed, pIsoHAPTS1, in which the HRM tripeptide was deleted
(Fig. 5B). This deletion construct was transfected into the
PTS2-deficient cell line BRO-T. The immunofluorescence pattern was
consistent with peroxisomal labeling when anti-catalase antibody was
used (Fig. 7A), whereas a cytosolic labeling
pattern was obtained when the anti-HA antibody was used (Fig.
7B). Thus, these data show that the HRM tripeptide is
necessary for the targeting of IPP isomerase to peroxisomes.
The HRM tripeptide is present at the C terminus of both the rat and hamster homologues of IPP isomerase. However, the human homologue at its C terminus has a YRM tripeptide (Fig. 1).
The PTS1 consensus sequence of (S/A/C)(K/H/R)(L/M) was derived from extensive mutational analysis, where peroxisomal proteins from other organisms or nonperoxisomal proteins were used as reporters. Because of the formulation of this consensus sequence, more peroxisomal proteins have been identified, the PTS1-like tripeptide of which does not fit this exact sequence. However, it has been demonstrated recently that many amino acid substitutions can be made at the first position of a homologous protein without compromising the PTS1 function (17). Thus, this strongly suggests that human isomerase is also targeted to peroxisomes by a PTS1, because the RM dipeptide meets the consensus sequence criteria.
The peroxisomal enzymes required for conversion of mevalonate to FPP (i.e. mevalonate kinase, phosphomevalonate kinase, mevalonate diphosphate decarboxylase, isopentenyl diphosphate isomerase, and FPP synthase) have now been all cloned and sequenced. Four of the five enzymes, mevalonate kinase (25), phosphomevalonate kinase (26), mevalonate diphosphate decarboxylase (27), and isomerase (14), contain a conserved putative PTS1 or PTS2, supporting the concept of a targeted transport into peroxisomes. This is the first study to unequivocally demonstrate that IPP isomerase, one of the five enzymes involved in the conversion of mevalonate to FPP, uses a PTS1 for peroxisomal targeting. FPP synthase does not contain a currently identifiable PTS1 or PTS2, yet it is selectively localized to peroxisomes. It seems likely that targeting to peroxisomes must include alternative mechanisms, not yet defined.