| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J. Biol. Chem., Vol. 275, Issue 26, 19560-19566, June 30, 2000
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
From Cell Biochemistry, Hannah Research Institute,
Ayr KA6 5HL, Scotland
Received for publication, March 15, 2000, and in revised form, April 11, 2000
The two isoforms of carnitine
palmitoyltransferase I (CPT I; muscle (M)- and liver (L)-type) of the
mitochondrial outer membrane have distinct kinetic characteristics with
respect to their affinity for one of the substrates
(L-carnitine) and the inhibitor malonyl-CoA. Moreover, they differ markedly in their hysteretic behavior with respect to malonyl-CoA and in their response to changes in the in
vivo metabolic state. However, the two proteins are 62%
identical and have the same overall structure. Using liver
mitochondria, we have previously shown that the protein is polytopic
within the outer membrane, comprising a 46-residue cytosolic N-terminal sequence, two transmembrane segments (TM1 and TM2) separated by a
27-residue loop, and a large catalytic domain (also cytosolic) (Fraser,
F., Corstorphine, C. G., and Zammit, V. A. (1997)
Biochem. J. 323, 711-718). We have now conducted a
systematic study on six chimeric proteins constructed from combinations
of three linear segments of rat L- and M-CPT I and on the two parental
proteins to elucidate the effects of altered intramolecular
interactions on the kinetics of CPT activity. The three segments were
(i) the cytosolic N-terminal domain plus TM1, (ii) the loop plus TM2, and (iii) the cytosolic catalytic C-terminal domain. The kinetic properties of the chimeric proteins expressed in Pichia
pastoris were studied. We found that alterations in the
combinations of the N-terminal plus TM1 and C-terminal domains as well
as in the N terminus plus TM1/TM2 pairings resulted in changes in the
Km values for carnitine and palmitoyl-CoA and the
sensitivity to malonyl-CoA of the L-type catalytic domain. The changes
in affinity for malonyl-CoA and palmitoyl-CoA occurred independently of
changes in the affinity for carnitine. The kinetic characteristics of the M-type catalytic domain and, in particular, its malonyl-CoA sensitivity were much less susceptible to influence by exchange of the
other two segments of the protein. The marked difference in the
response of the two catalytic domains to changes in the N-terminal
domain and TM combinations explains the previously observed differences
in the response of L- and M-CPT I to altered physiological state in
intact mitochondria and to modulation of altered lipid molecular order
of the mitochondrial outer membrane in vivo and in
vitro.
The enzyme carnitine palmitoyltransferase I (CPT
I)1 plays a major role in the
regulation of mitochondrial In mammals, CPT I exists as two distinct isoforms that are the products
of different genes: the liver (L) type or CPT I is an integral membrane protein that has two transmembrane
segments (TM1 and TM2) within the N-terminal 15% of its primary sequence (see Figs. 1 and 2) (12). In the mitochondrial outer membrane,
it adopts a polytopic orientation such that both the N- and C-terminal
(catalytic) domains, composed of ~46 and 651 residues, respectively,
are exposed on the outer (cytosolic) surface of the membrane (12-14).
A short loop of 27 amino acid residues connects the two transmembrane
segments and protrudes into the intermembrane space between the outer
and inner membranes of the mitochondrion (12). Previous results
demonstrated that N/C-terminal domain interactions are important in
determining the kinetic characteristics of L-CPT I (12, 14).
Proteolytic cleavage of the extreme N terminus causes total loss of the
malonyl-CoA sensitivity of the enzyme and partial loss of its catalytic
activity (12). In addition, interactions of regions of the molecule
with the membrane have been suggested to explain the marked dependence
of the kinetic properties of the enzyme (particularly its malonyl-CoA
sensitivity) on the molecular order of the membrane lipids both
in vivo (15) and in vitro (16). Functional
mutagenesis studies have confirmed that the N-terminal 18 amino acid
residues (and particularly Glu-3) of the L-CPT I protein are required
for malonyl-CoA sensitivity (17, 18), a parameter that is also affected
by mutation of His-5 and, on the C-terminal domain, of His-140 (19).
Deletion of the N-terminal 30 amino acid residues of M-CPT I also
largely abolishes its malonyl-CoA sensitivity (20).
This study was undertaken systematically to investigate the roles of
the different domains of the molecule in substrate and inhibitor
interactions and to exploit the availability of isoforms (L- and
M-type) that have distinct kinetic parameters. These isoforms are
closely related in amino acid sequence, having an overall 62% sequence
identity, with regions, including the cytosolic N-terminal sequence, of
much higher similarity. They are therefore presumed to have the same
overall structure (see Fig. 1). By determining the kinetic parameters
of chimeras constructed from different domains of L- and M-CPT I
isoforms, it was reasoned that it would be possible to elucidate the
roles of different regions of the molecule in determining the
distinctive properties of the two isoforms. Studies on other
transmembrane proteins have shown that it is highly energetically
favorable to have physical interaction between transmembrane helices
(see Ref. 21). Thus, our chimeras were designed to perturb the native
interactions between TM1 and TM2 and/or the putative interactions
between the N- and C-terminal domains.
For the purpose of these studies, the molecule was divided into three
discrete regions: the N-terminal domain plus TM1, the loop plus TM2,
and the C-terminal domain (see Fig. 1). Thus, six chimeras were
constructed and expressed in Pichia pastoris, which has
previously been shown to be a good expression system for CPT I (22,
23). The use of an expression background free from endogenous CPT
activity allowed detailed kinetic analyses to be performed. The kinetic
parameters for the substrates carnitine and palmitoyl-CoA and the
inhibitor malonyl-CoA were examined for the chimeric CPT I molecules
and the two expressed parental enzymes. Our results reveal a more
complex set of intramolecular interactions than has hitherto been
reported (19). The kinetic characteristics of chimeric proteins
containing the C-terminal domain of L-CPT I are dependent on the
precise pairings between the respective N- and C-terminal domains and
between TM1 and TM2, but the C-terminal domain of M-CPT I is much less
sensitive to alterations in these interactions. The results are
discussed in relation to the differences in responses of the two
isoforms to altered conditions both in vivo and in
vitro.
Materials--
These were obtained as described previously (23),
except where stated.
Construction of Expression Plasmids--
Full-length cDNAs
for the rat L-CPT I (LLL) and M-CPT I (MMM) isoforms were obtained by
screening a rat skeletal muscle cDNA library (Stratagene no.
937510). Probes corresponding to nucleotides 82-534
(GenBankTM/EBI accession number L07736) of L-CPT I and
nucleotides 21-512 of M-CPT I (accession number D43623) were generated
by PCR from rat heart cDNA using oligonucleotides
AGCAATAGGTCCCCACTCAA and GCTCATTTTGCCGTGTTCTG and oligonucleotides
CCCAGGATGGCGGAAGCACACC and CCGGCTGGACAGGAGACGAACAC, respectively.
Library screening was performed as described previously (23). The
cDNAs were excised from the libraries into pBluescript SK(
To generate the chimeras LLM and MML (see Fig. 2A), a silent
AflII restriction site (CTTAAG corresponding to amino acids
Leu-Lys; residues 126 and 127 and residues 128 and 129 of the L- and
M-type isoforms, respectively) was introduced into the cDNAs for
both isoforms (shown in italics in the oligonucleotides below). This single nucleotide change was introduced by overlap extension PCR. For
L-CPT I, two independent PCRs (20 cycles each) were performed with
primers A (TTGCGGCCGCAATAGGTCCCCACT) and B
(GCAGCACCTTAAGCGAGTAGCG) and primers C
(GCTACTCGCTTAAGGTGCTGC) and D (TCCTCGAGGCCTTACAGATTCCAG) using the rat L-CPT I cDNA as template. The two products were annealed and reamplified for a further five cycles of PCR using primers
A and D. The same method was used for rat M-CPT I using TTGCGGCCGCAGGATGGCGGAAGCACACCAGGCAGTA (primer E),
GCAGCAGCTTAAGGGTTTGTCGGACCGACAAACCCTTAAGCTGCTGC, and primer F (GGTCTAGACAAGGGCCGCACAGAATCCA). In each case, a novel NotI site (shown in italics) was also introduced upstream of
each of the respective start codons. All PCRs were performed using Pfu TURBO DNA polymerase (Stratagene).
The LLL cDNA was subcloned into pGAPZ B using the NotI
site from pBluescript and the naturally occurring EcoRI site
in the 3'-untranslated region. The PCR product (containing the
AflII site) was used to replace the 5'-ends of the cDNAs
upstream of the unique KpnI site. The MMM cDNA was
subcloned into pcDNA3.1 at the NotI-XhoI
sites, using sites from the pBluescript vector. The 5'-end
(NotI-HindIII) was then replaced with the PCR
product, and the whole cDNA was transferred to pGAPZ B using
NotI and the XbaI site from the pcDNA vector.
The sequence ATTTTTTTATTC (encoding amino acids 121-124, IFLF) in the
M-CPT I cDNA, which resembles a known yeast polyadenylation signal,
was subsequently changed to
ATCTTCCTCTTC in an attempt to
increase expression levels without changing the amino acid sequence. A
cleaved PCR product (primer E, above, and
AGCTTAAGAGTTTGTCGGAAGAGGAAGATGCCTGTCGCCCAG) was used to
replace the NotI-AflII fragment.
The chimeras LLM and MML (see Fig. 2A) were generated from
LLL and MMM by swapping the NotI-AflII
restriction fragments. The chimeras LMM and MLL were generated
individually. Briefly, LMM was made by replacing the
SphI-AflII fragment of LLM (corresponding to the
loop and TM2) with a PCR product generated from the corresponding region of M-CPT I (with the polyadenylation site removed), except that
a 5'-SphI site was added and a single amino acid change (Val to Ile) was introduced into the loop (PCR primers
ATGCATGCTAAAATCGATATCAGTATGGGGCTGGTCC and F above) (see Fig.
2A). MLL was made by replacing the
NcoI-AflII region in MML with a PCR product
generated from the corresponding region of L-CPT I with an added
5'-NcoI site (PCR primers
TTCCTCCATGGGCATGATCGCAAAGAT and D above). LML and MLM were
subsequently generated from LMM and MLL by swapping
NotI-AflII restriction fragments. All the mutations and junctions were verified by sequencing.
Transformation and Expression in P. pastoris--
P.
pastoris strain X-33 was used throughout these studies. All
procedures were performed as recommended by the expression system
distributor, Invitrogen (San Diego, CA). Expression was driven by the
constitutive glyceraldehyde-3-phosphate dehydrogenase promoter (24)
using the vector pGAPZ. High copy number clones were selected where
necessary using 0.5 or 1 mg/ml Zeocin. Cultures for activity
measurements were grown for 24 h in YPD medium (1% yeast extract,
2% peptone, 2% glucose), and cell-free extracts were prepared as
described previously (23).
CPT Assay--
CPT I activity was measured at different
carnitine, palmitoyl-CoA, and malonyl-CoA concentrations, as indicated,
as described previously (23). Assays were performed on cell-free yeast
extracts rather than mitochondrion-enriched fractions for reasons
discussed previously (23). It was shown that for both L- and M-CPT I, identical kinetic behavior was observed in the two fractions (23). When
concentrations of carnitine were varied, that of palmitoyl-CoA was
fixed at 135 µM. When concentrations of palmitoyl-CoA
were varied, that of carnitine was fixed at 500 µM. For
malonyl-CoA inhibition studies, palmitoyl-CoA and carnitine
concentrations were fixed at 35 and 500 µM, respectively.
Presentation of Data and Statistical Analysis--
Data were
fitted to the Michaelis-Menten equation for palmitoyl-CoA and carnitine
or to the equation for simple competitive inhibition in the case of
malonyl-CoA. Curve fitting was performed using Sigma-Plot software with
nonlinear regression analysis as described previously (23).
For each of the three parameters (malonyl-CoA, palmitoyl-CoA, and
carnitine) and for each catalytic domain (L- or M-type), separate
general linear models were used to examine the effects of the identity
(L- or M-type) of the remaining two regions of the chimeric molecules.
Where the two-way statistical interaction between the two regions was
significant, Tukey's post-hoc multiple comparisons test was used to
examine the differences between the four mean values. In general, it
was found that residuals in the various models when applied to the raw
data increased with the mean, and so the values were log-transformed
prior to analysis. Since the data were unbalanced, to ensure that
p values were genuine, sequential sums of squares in the
general linear model were also investigated with the order of the
factors reversed, and in addition, the results were checked using the
residual maximum likelihood method (REML). Minitab software (Version
12) was used to apply the general linear models, and Genstat 5 was used
for REML.
Expression of Wild-type Rat L- and M-CPT I
This study describes the first heterologous expression of rat
M-CPT I in P. pastoris. In preliminary experiments, only
very low levels of M-CPT I activity could be detected, even when
selection for high copy number genomic integration events with high
Zeocin concentrations was used. This was presumed to be due to
premature termination of transcripts in AT-rich regions and has been
observed in Pichia and other eukaryotic systems. In
particular, the sequence TTTTTATA is known to promote premature
polyadenylation in Saccharomyces cerevisiae, and a similar
sequence has been shown to cause termination when the human
immunodeficiency virus type 1 gp120 protein is expressed in P. pastoris (25). When the sequence TTTTTATTC present in rat M-CPT I
(encoding amino acids 122-124) was mutated (without changing the amino
acid sequence; see "Experimental Procedures"), higher levels of
expressed activity were achieved, although it was still necessary to
select for high copy number integrants. It is noteworthy that in human
M-CPT I, the corresponding sequence is TTCTTCTTC, and the protein was
readily expressed from the glyceraldehyde-3-phosphate dehydrogenase
promoter from single copy clones (22). For rat L-CPT I, good levels of
CPT activity were achieved when present at single copy within the
Pichia genome (after selecting for transformants with 0.1 mg/ml Zeocin).
Expression and Kinetic Characterization of Chimeric CPT I
Constructs
For the purpose of these studies, the molecule was divided into
three discrete regions (Fig. 1) based on
the primary amino acid sequence (Fig. 2):
the N-terminal domain plus TM1, the loop plus TM2, and the C-terminal
domain (residues 1-82, 83-126, and 127-772, respectively, in L-CPT
I). Due to the low degree of sequence conservation observed in the loop
of CPT I isoforms from different mammalian species and the low degree
of functional importance inferred from the topology of the protein,
this region was not considered separately, but in conjunction with TM2.
The N-terminal domain shows a very high degree of sequence conservation
between the L- and M-type isoforms (Fig. 2A). In particular,
the first 18 residues in L- and M-CPT I are identical in all known
mammalian CPT I sequences. Therefore, the N-terminal domain was not
addressed separately, but in combination with TM1. Using all
combinations of the three regions allowed generation of six chimeras,
plus the two parental molecules. No amino acids were omitted at the joins in the chimeras. The single conservative Val-77 For chimeras containing the L-type catalytic domain, high levels of CPT
activity were expressed from single copy clones. For chimeras
containing the M-type C-terminal domain, high copy number integrants
had to be used to achieve readily measurable activity levels, as was
the case for the parental M-CPT I (MMM). In the constructs MML, LML,
and LMM, the T-rich nucleotide sequence was altered as described above
for MMM. All the kinetic characteristics of the expressed chimeras are
given in Table I, but in the interests of
clarity, they are considered separately, below, for those containing the L- or M-type catalytic domain.
Use of Six Chimeric Proteins to Investigate the Role of
Intramolecular Interactions in Determining the Kinetics of Carnitine
Palmitoyltransferase I Isoforms*
, and
ABSTRACT
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
-oxidation in all mammalian tissues. CPT
I catalyzes the first step in the transport of long-chain acyl groups
from the cytosol into the mitochondrial matrix, where they undergo
-oxidation. Inhibition by malonyl-CoA is a major regulatory property
of CPT I and confers to the enzyme the ability to signal to the cell
the relative availability of carbohydrate and lipid fuels (see Ref. 1
for recent review). Therefore, the enzyme plays a central role in the
control of hepatic, cardiac, and skeletal muscle and pancreatic
-cell function and is a potential pharmacological target for the
treatment of metabolic disorders ranging from diabetes to insulin
resistance and coronary heart disease (2-5)
form and the muscle (M)
type or form (6, 7). When assayed in isolated mitochondria, these
two CPT I variants differ markedly in key kinetic characteristics in
that M-CPT I has a much lower IC50 for malonyl-CoA and a
higher Km for L-carnitine than L-CPT I
(8, 9). However, their affinity for long-chain acyl-CoA is not markedly
different (9). They also differ in their responses to altered
physiological states (the malonyl-CoA sensitivity and Km for palmitoyl-CoA are altered only for L-CPT I
(10)) and incubation conditions of isolated mitochondria (L-CPT I shows pronounced hysteretic behavior with respect to malonyl-CoA, whereas M-CPT I does not (11)).
EXPERIMENTAL PROCEDURES
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
), and
the sequences of the coding regions were verified to be identical to
those already published.
RESULTS
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
Ile change in
LML and LMM (Fig. 2A) is in the loop region and is also a
substitution that occurs between rat or mouse and human in L-CPT I. The
chimeras are referred to using a three-letter nomenclature as shown in Fig. 2A. Thus, LLL and MMM refer to the two parental
proteins.
View larger version (16K):
[in a new window]
Fig. 1.
Diagrammatic representation of the different
regions of L-CPT I structure in relation to its membrane topology.
M-CPT I is thought to have the same overall structure (see the
Introduction). The positions where joins in the chimeras were made are
indicated by asterisks. N-domain, N-terminal
domain (residues 1-46); C-domain, C-terminal domain
(residues 126-773).
View larger version (79K):
[in a new window]
Fig. 2.
Primary sequences of the N-terminal regions
of the CPT I constructs with the positions of the N-terminal domain,
TM1, loop, TM2, and the start of the C-terminal domain.
A shows the sequences of the six chimeras generated. Regions
corresponding to L-CPT I are shown in white on black, with
those from M-CPT I in black on white. The two isoleucine
residues shown in black on gray (residue 77 in LML and LMM)
represent a conservatively changed amino acid present in neither of the
parent molecules. B shows the aligned amino acid sequences
of rat L- and M-CPT I with conservative (black on gray) and
nonconserved (white on black) differences highlighted.
Kinetic parameters of CPT I chimeras
Kinetic Properties of Chimeras with the L-type Catalytic Domain (MLL, LML, and MML)
Velocity versus substrate/inhibitor concentration
curves for palmitoyl-CoA, carnitine, and malonyl-CoA are given in Fig.
3. For both carnitine and palmitoyl-CoA,
the data obeyed standard Michaelis-Menten kinetics, thus enabling the
accurate calculation of Km values by graphical
methods. The assay conditions were designed (26) to give high values
for the IC50 for malonyl-CoA (concentration required to
inhibit CPT I activity by 50%). This enabled the more accurate
determination of IC50 values, thus allowing the detection
of subtle changes in this parameter.
|
Malonyl-CoA Inhibition-- The IC50 for malonyl-CoA was the same for LLL and LML (38.4 ± 2.24 and 35.1 ± 4.1 µM, respectively), suggesting that TM1/TM2 interactions do not affect malonyl-CoA sensitivity as long as the N/C-terminal domain interactions (LxL) present in the parental form are maintained. By contrast, the chimeric enzymes MML and MLL both had IC50 values for malonyl-CoA that were >1.5-fold higher than those of the parental control (Table I), indicating that disruption of the N/C-terminal domain interactions negatively affects malonyl-CoA action.
Km for Palmitoyl-CoA-- The same pattern of changes was observed for the affinity for palmitoyl-CoA as that observed above for malonyl-CoA (Table I). Thus, LLL and LML had very similar Km values (46.0 ± 5.1 and 36.9 ± 7.2 µM, respectively; not significantly different), whereas those for MML and MLL (63.9 ± 6.6 and 64.7 ± 6.5 µM, respectively) were both 1.5-fold higher (p < 0.01) than the value for the parental protein. Therefore, in chimeric CPT I proteins with the L-type catalytic domain, the effects of malonyl-CoA and palmitoyl-CoA on activity appear to be modulated by alterations in N/C-terminal domain interactions, whereas changes in TM1/TM2 interactions do not affect either parameter.
Km for Carnitine-- The chimeric protein MML had the same Km for carnitine (173.6 ± 16.4 µM) as the parental protein, LLL (153.3 ± 11.1 µM), indicating that contrary to the case with malonyl-CoA and palmitoyl-CoA, alterations of N/C-terminal domain interactions did not affect this kinetic parameter (Table I). However, mismatching of the TM regions (in MLL or LML) resulted in a >2-fold increase in the Km for carnitine (to 369.5 ± 39.8 and 376.3 ± 49.1 µM, respectively), indicating that the matched TM1/TM2 interactions that are preserved when they both originate from either the L- or M-type isoform (as in MML or LLL) are crucial for retention of the low Km for carnitine displayed by L-CPT I (compared with that of M-CPT I; see below).
Kinetic Properties of Chimeras Containing the M-type Catalytic Domain (LMM, MLM, and LLM)
IC50 for Malonyl-CoA-- The IC50 for malonyl-CoA of the parental enzyme MMM was 10-fold lower (3.3 ± 0.2 µM) than that of the parental L-CPT I (LLL). All the chimeras containing the M-type catalytic domain had very similar IC50 values (Table I), suggesting that the degree of malonyl-CoA sensitivity of the C-terminal domain of M-CPT I is determined solely by the structural characteristics within that domain and is not dependent on the identity of the rest of the molecule.
Km for Palmitoyl-CoA-- As expected from previous studies on liver and muscle mitochondria, the Km for palmitoyl-CoA of M-CPT I (MMM; 57.4 ± 5.5 µM) was similar to that of L-CPT I (LLL; 46.0 ± 5.1 µM). The affinity for palmitoyl-CoA was not affected (Table I) so long as either the N/C-terminal domain interaction was preserved (in MLM) or the pairing of TM regions from the same isoform was present (in LLM). However, simultaneous disruption of both sets of interactions, as achieved in LMM, resulted in a 2-fold increase (p < 0.001) in Km for palmitoyl-CoA to 103.7 ± 9.0 µM. This illustrates that the C-terminal domain of M-CPT I, unlike that of L-CPT I, can accommodate changes in N/C-terminal domain interactions without loss of affinity for palmitoyl-CoA.
Km for Carnitine--
The Km for
carnitine of M-CPT I (MMM; 780 ± 80 µM) was 4-fold
higher than that of LLL, as expected from the relative affinities for
this substrate shown by the respective enzymes in isolated rat muscle
and liver mitochondria (9). Substitution of the N-terminal domain and
TM1 (in LMM) significantly (p = 0.02) lowered the
Km by 20% (to 612 ± 10 µM).
This effect was reversed in LLM, i.e. when TM2 was also
substituted. Substitution of TM2 only (by liver-type TM2, in MLM)
resulted in an even bigger (2-fold) increase in
Km for carnitine to 1.3 ± 0.2 mM. These observations indicate that the higher Km of
M-CPT I is not solely determined by the characteristics of the
C-terminal domain. Indeed, the M-type catalytic domain showed a wide
range of values for Km for carnitine across the
chimeric constructs (Table I and Fig. 4),
with the lowest value resulting from substituting the N-terminal domain
and TM1 region and the highest value resulting from substituting
TM2.
|
| |
DISCUSSION |
|---|
|
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
|
|---|
Using the Pichia expression system, we were able to determine accurate kinetic parameters for L- and M-CPT I and the six chimeric derivatives. In particular, the lack of sigmoidicity for the CPT activity versus substrate concentration curves allowed fitting of the data to the standard Michaelis-Menten equation. The kinetics of M- and L-CPT I differ most strikingly with respect to the IC50 for malonyl-CoA and the Km for carnitine when measured in muscle and liver mitochondria (8, 9). These differences were fully preserved in the proteins expressed from the respective parental constructs in P. pastoris.
The primary amino acid sequence of the 46-residue cytosolic N-terminal domain of CPT I is identical or highly conserved in the L- and M-type isoforms, with three semiconservative differences (Fig. 2B). However, there is a notable difference between sequences of their transmembrane segments. M-CPT I has a basic residue (Arg-52) within the hydrophobic stretch of TM1 (Fig. 2B). This L/M-CPT I difference is conserved in all known mammalian CPT I sequences. This may make this TM region shorter than that in L-CPT I and extend the cytosolic N-terminal domain accordingly. Alternatively, this charged residue might restrict the mobility of the transmembrane (presumed) helix within the membrane environment, contributing toward an overall more rigid structure for M-CPT I. These possibilities are currently being investigated.
We have previously suggested that interaction between the N-terminal region and the catalytic C-terminal domain of L-CPT I is important in determining malonyl-CoA sensitivity (14). Similarly, we have emphasized that it is highly likely that TM1 and TM2 interact strongly with each other as well as with annular membrane lipids of the protein (21). These predictions have now been tested. From the data, we conclude that the precise nature of these interactions is important in determining the kinetics of the L-CPT I catalytic domain. This is contrary to previous conclusions by Swanson et al. (19) derived from analysis of two rat CPT I chimeras expressed in a mammalian cell line. The reason for this discrepancy is not immediately apparent, as the chimera involving the L-type C-terminal domain studied by those authors was similar to our MLL, which shows a >2-fold increase in Km for carnitine (Table I). Moreover, in the aforementioned study, no measurements of the Km for palmitoyl-CoA were performed, presumably because of the high background activity of endogenous CPT I and CPT II in COS cells (19). It was also concluded that the IC50 for malonyl-CoA was not affected by N/C-terminal domain interactions for either M- or L-CPT I. Although this is true for the M-type catalytic domain, it is evidently not so for the L-type catalytic domain (Table I). A chimera similar to our MLL (using rat L- and human M-CPT I (20)) was also found to have a higher IC50 for malonyl-CoA. However, the simultaneous disruption of N/C-terminal domain and TM1/TM2 interactions in that chimera, coupled with the absence of other relevant chimeras, allowed no definite conclusions to be drawn by those authors about the relative importance of the different types of intramolecular interactions for the L-type catalytic domain. The effect of similar disruptions on the M-type catalytic domain was not addressed in their study. Our study thus benefits from the ability both to determine accurate kinetics in the absence of endogenous CPT activity and to resolve the effects of different intramolecular interactions through the use of several chimeras. Any potential for intermolecular interaction between the endogenous CPT I molecules and those expressed experimentally was also eliminated in our studies.
A striking observation of this study is that the kinetics with respect to palmitoyl-CoA and to carnitine respond to changes in different characteristics of the chimeras. Affinity for palmitoyl-CoA was mostly responsive to altered N/C-terminal domain interactions, especially for the L-type catalytic domain. By contrast, the Km for carnitine was mostly responsive to TM1/TM2 interactions and especially for the M-type catalytic domain. As shown in Fig. 4, these interactions mean that a continuous range of Km values for carnitine can be obtained for proteins expressed from different combinations of the three regions of L- or M-CPT I from which the six chimeras were constructed.
In rat liver mitochondria, alterations in membrane lipid order affect the kinetic parameters of L-CPT I for malonyl-CoA and palmitoyl-CoA (15, 16). This might arise through altered N/C-terminal domain interactions or, directly or indirectly, through altered inter-TM interactions. The present data suggest that the latter is not the case. Indeed, it was the Km for carnitine that was altered by perturbation of TM interactions, and not the kinetic parameters for palmitoyl-CoA or malonyl-CoA (see above). Importantly, this effect on the Km for carnitine was observed for both the L- and M-type series of constructs (Fig. 4 and Table I), suggesting that the kinetics of carnitine binding to either isoform is dependent on TM1/TM2 interactions. To our knowledge, effects of membrane lipid order on carnitine kinetics have not been examined; however, our results predict that this parameter would not be greatly influenced by changes in membrane lipid composition that occur in vivo (15). Our data show that the kinetics for carnitine, on the one hand, and for malonyl-CoA and palmitoyl-CoA, on the other, appear to be affected independently. This suggests that the inverse relationship between the values of these two parameters for CPT I assayed in mitochondria from different tissues (9) does not result from an obligatory link between the respective binding sites. Indeed, whereas a continuous range of affinities for carnitine can be produced by altering the combinations of L- and M-CPT I domains used in the constructs (Fig. 4), IC50 values for malonyl-CoA fall into two discrete groups, depending on the identity of the catalytic domain (Table I; cf. Ref. 19), although the value for the L-type catalytic domain is capable of modulation by the structure of the rest of the molecule (Table I).
The difference between the C-terminal domains of the two CPT I isoforms in their dependence on N/C-terminal domain and TM1/TM2 interactions may not only reflect the major differences in kinetics between them (see Introduction), but may also account for the differences in their respective responses to conditions that alter membrane lipid molecular order in vivo (15) and in isolated mitochondria in vitro (11, 16). It may also explain why the activity and malonyl-CoA sensitivity of the M-type catalytic domain are more resistant to N-terminal truncations (17, 20). Importantly, the changes in kinetic parameters observed for L-CPT I upon mismatching of N- and C-terminal domains or TM1/TM2 pairings were of the same magnitude as those observed under the same assay conditions for L-CPT I in liver mitochondria isolated from rats in different metabolic states (26-28). Thus, in liver mitochondria isolated from fasted rats, the IC50 for malonyl-CoA and Km for palmitoyl-CoA of L-CPT I are both 2-4-fold higher than those in mitochondria from fed animals (28). This suggests that conformational changes of the same order are established within L-CPT I when induced either by altered membrane composition in vivo or through changes in intramolecular interactions produced in our chimeras.
In chimeras containing the L-type C-terminal domain, the effects of palmitoyl-CoA and malonyl-CoA were altered by substitution of the N terminus and TM1 (in MxL). As discussed above, the most obvious differences between L- and M-type amino acid sequences for this region are in TM1 rather than in the cytosolic N terminus. It is suggested that conformational changes that affect the positioning (laterally or transversely) of TM1 relative to the membrane (but independent of the identity of TM2) are important in determining the way that the cytosolic N-terminal domain emerges from the membrane and interacts with the C-terminal domain. It is possible that it is solely the arginine residue in muscle-type TM1 that accounts for the differences between LxL and MxL, namely the raised IC50 and Km for palmitoyl-CoA for the latter. Studies are being conducted to explore this possibility and to investigate the role of the other conservative/semiconservative differences between the amino acid sequences of the respective N-terminal domains.
The fact that the malonyl-CoA sensitivity of the M-CPT I catalytic domain is much less influenced by changes in either N/C-terminal domain or TM1/TM2 interactions and, by implication, by any changes in these interactions brought about by altered membrane fluidity in vivo explains the previous observations that the kinetics of this isoform in heart and muscle do not change under different physiological conditions (10). Similarly, it is interesting that, unlike L-CPT I, heart mitochondrial CPT I does not show any hysteretic behavior with respect to malonyl-CoA in vitro (11). Coupled with the fact that the IC50 for malonyl-CoA of M-CPT I is already >10-fold lower than that of L-CPT I, these observations suggest that the catalytic domain of M-CPT I adopts a much more rigid conformation that maintains its structure in a highly malonyl-CoA-sensitive state independently of changes in N/C-terminal domain and TM1/TM2 interactions mediated by changes in membrane lipid molecular order. However, as is the case for L-CPT I (1, 12, 14, 17), the presence of an intact cytosolic N terminus is still required for malonyl-CoA sensitivity of M-CPT I (20).
Structural data for CPT I are not yet available in view of its nature
as an integral membrane protein. In their absence, studies such as
those presented here are invaluable for furthering our understanding of
structure/function relationships within the molecule. Our studies
extend the understanding of the complexity of the intramolecular
interactions of CPT I as it relates to the distinct kinetics of the two
isoforms. In particular, our studies provide insights into the
molecular mechanisms underlying their different hysteretic behavior
with respect to the inhibitor malonyl-CoA and the response to altered
physiological state previously observed for the liver- and muscle-type
CPT I isoforms. Indeed, especially in the case of the liver-type
isoform, any future understanding derived from the determination of its
static structure will need to be supplemented by studies
addressing conformational changes within the molecule induced by
intramolecular interactions.
| |
ACKNOWLEDGEMENTS |
|---|
We thank Dr. S. Brocklehurst (Biomathematics and Statistics Scotland) for performing the statistical analyses and C. Narain for excellent assistance.
| |
FOOTNOTES |
|---|
* The work was supported by the British Diabetic Association and the Scottish Executive.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
To whom correspondence should be addressed. Tel.:
44-0-1292-674058; Fax: 44-0-1292674059; E-mail:
zammitv@hri.sari.ac.uk.
Published, JBC Papers in Press, April 13, 2000, DOI 10.1074/jbc.M002177200
| |
ABBREVIATIONS |
|---|
The abbreviations used are: CPT I, mitochondrial overt carnitine palmitoyltransferase; L, liver; M, muscle; TM, transmembrane segment; PCR, polymerase chain reaction.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
REFERENCES
|
|---|
| 1. | Zammit, V. A. (1999) Biochem. J. 343, 505-515 |
| 2. | Foley, J. E. (1992) Diabetes Care 15, 773-784 |
| 3. | McGarry, J. D. (1998) Am. J. Clin. Nutr. 67, 500S-504S |
| 4. | Ruderman, N. B., Saha, A. K., Vavvas, D., and Witters, L. A. (1999) Am. J. Physiol. 276, E1-E18 |
| 5. | Zammit, V. A. (1998) Diabetes Nutr. Metab. 11, 200-211 |
| 6. | Esser, V., Britton, C. H., Weis, B. C., Foster, D. W., and McGarry, J. D. (1993) J. Biol. Chem. 268, 5817-5822 |
| 7. | Yamazaki, N., Shinohara, Y., Shima, A., and Terada, H. (1995) FEBS Lett. 363, 41-45 |
| 8. | Saggerson, E. D. (1982) Biochem. J. 202, 397-405 |
| 9. | McGarry, J. D., Mills, S. E., Long, C. S., and Foster, D. W. (1983) Biochem. J. 214, 21-28 |
| 10. | Mynatt, R. L., Lappi, M. D., and Cook, G. A. (1992) Biochim. Biophys. Acta 1128, 105-112 |
| 11. | Lloyd, A. C., Carpenter, C. A., and Saggerson, E. D. (1986) Biochem. J. 237, 289-291 |
| 12. | Fraser, F., Corstorphine, C. G., and Zammit, V. A. (1997) Biochem. J. 323, 711-718 |
| 13. | Fraser, F., Corstorphine, C. G., and Zammit, V. A. (1996) Biochem. Soc. Trans. 24, 184S |
| 14. | Zammit, V. A., Fraser, F., and Corstorphine, C. G. (1997) Adv. Enzyme Regul. 37, 295-317 |
| 15. | Zammit, V., Corstorphine, C., Kolodziej, M., and Fraser, F. (1998) Lipids 33, 371-376 |
| 16. | Kolodziej, M. P., and Zammit, V. A. (1990) Biochem. J. 272, 421-425 |
| 17. | Shi, J., Zhu, H., Arvidson, D. N., Cregg, J. M., and Woldegiorgis, G. (1998) Biochemistry 37, 11033-11038 |
| 18. | Shi, J., Zhu, H., Arvidson, D. N., and Woldegiorgis, G. (1999) J. Biol. Chem. 274, 9421-9426 |
| 19. | Swanson, S. T., Foster, D. W., McGarry, J. D., and Brown, N. F. (1998) Biochem. J. 335, 513-519 |
| 20. | Shi, J., Zhu, H., Arvidson, D. N., and Woldegiorgis, G. (2000) Biochemistry 39, 712-717 |
| 21. | Zammit, V. (1999) Prog. Lipid Res. 38, 199-224 |
| 22. | Zhu, H., Shi, J., de Vries, Y., Arvidson, D. N., Cregg, J. M., and Woldegiorgis, G. (1997) Arch. Biochem. Biophys. 347, 53-61 |
| 23. | Jackson, V. N., Cameron, J. M., Zammit, V. A., and Price, N. T. (1999) Biochem. J. 341, 483-489 |
| 24. | Waterham, H. R., Digan, M. E., Koutz, P. J., Lair, S. V., and Cregg, J. M. (1997) Gene (Amst.) 186, 37-44 |
| 25. | Scorer, C. A., Buckholz, R. G., Clare, J. J., and Romanos, M. A. (1993) Gene (Amst.) 136, 111-119 |
| 26. | Drynan, L., Quant, P. A., and Zammit, V. A. (1996) Biochem. J. 317, 791-795 |
| 27. | Grantham, B. D., and Zammit, V. A. (1988) Biochem. J. 249, 409-414 |
| 28. | Grantham, B. D., and Zammit, V. A. (1986) Biochem. J. 239, 485-488 |
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  E. Lopez-Vinas, A. Bentebibel, C. Gurunathan, M. Morillas, D. de Arriaga, D. Serra, G. Asins, F. G. Hegardt, and P. Gomez-Puertas Definition by Functional and Structural Analysis of Two Malonyl-CoA Sites in Carnitine Palmitoyltransferase 1A J. Biol. Chem., June 22, 2007; 282(25): 18212 - 18224. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Borthwick, V. N. Jackson, N. T. Price, and V. A. Zammit The Mitochondrial Intermembrane Loop Region of Rat Carnitine Palmitoyltransferase 1A Is a Major Determinant of Its Malonyl-CoA Sensitivity J. Biol. Chem., November 3, 2006; 281(44): 32946 - 32952. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kerner, A. M. Distler, P. Minkler, W. Parland, S. M. Peterman, and C. L. Hoppel Phosphorylation of Rat Liver Mitochondrial Carnitine Palmitoyltransferase-I: EFFECT ON THE KINETIC PROPERTIES OF THE ENZYME J. Biol. Chem., September 24, 2004; 279(39): 41104 - 41113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Gobin, L. Thuillier, G. Jogl, A. Faye, L. Tong, M. Chi, J.-P. Bonnefont, J. Girard, and C. Prip-Buus Functional and Structural Basis of Carnitine Palmitoyltransferase 1A Deficiency J. Biol. Chem., December 12, 2003; 278(50): 50428 - 50434. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Pan, I. Cohen, F. Guillerault, B. Feve, J. Girard, and C. Prip-Buus The Extreme C Terminus of Rat Liver Carnitine Palmitoyltransferase I Is Not Involved in Malonyl-CoA Sensitivity but in Initial Protein Folding J. Biol. Chem., November 27, 2002; 277(49): 47184 - 47189. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. R. van der Leij, K. B. Cox, V. N. Jackson, N. C. A. Huijkman, B. Bartelds, J. R. G. Kuipers, T. Dijkhuizen, P. Terpstra, P. A. Wood, V. A. Zammit, et al. Structural and Functional Genomics of the CPT1B Gene for Muscle-type Carnitine Palmitoyltransferase I in Mammals J. Biol. Chem., July 19, 2002; 277(30): 26994 - 27005. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Morillas, P. Gomez-Puertas, B. Rubi, J. Clotet, J. Arino, A. Valencia, F. G. Hegardt, D. Serra, and G. Asins Structural Model of a Malonyl-CoA-binding Site of Carnitine Octanoyltransferase and Carnitine Palmitoyltransferase I. MUTATIONAL ANALYSIS OF A MALONYL-CoA AFFINITY DOMAIN J. Biol. Chem., March 22, 2002; 277(13): 11473 - 11480. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Nicot, J. Relat, G. Woldegiorgis, D. Haro, and P. F. Marrero Pig Liver Carnitine Palmitoyltransferase. CHIMERA STUDIES SHOW THAT BOTH THE N- AND C-TERMINAL REGIONS OF THE ENZYME ARE IMPORTANT FOR THE UNUSUAL HIGH MALONYL-CoA SENSITIVITY J. Biol. Chem., March 15, 2002; 277(12): 10044 - 10049. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. N. Jackson, V. A. Zammit, and N. T. Price Identification of Positive and Negative Determinants of Malonyl-CoA Sensitivity and Carnitine Affinity within the Amino Termini of Rat Liver- and Muscle-type Carnitine Palmitoyltransferase I J. Biol. Chem., December 1, 2000; 275(49): 38410 - 38416. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Fraser, R. Padovese, and V. A. Zammit Distinct Kinetics of Carnitine Palmitoyltransferase I in Contact Sites and Outer Membranes of Rat Liver Mitochondria J. Biol. Chem., June 1, 2001; 276(23): 20182 - 20185. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Morillas, P. Gomez-Puertas, R. Roca, D. Serra, G. Asins, A. Valencia, and F. G. Hegardt Structural Model of the Catalytic Core of Carnitine Palmitoyltransferase I and Carnitine Octanoyltransferase (COT). MUTATION OF CPT I HISTIDINE 473 AND ALANINE 381 AND COT ALANINE 238 IMPAIRS THE CATALYTIC ACTIVITY J. Biol. Chem., November 21, 2001; 276(48): 45001 - 45008. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
)
and rat muscle (MMM) (
) CPT I isoforms. The mean data from four
to five velocity versus inhibitor or substrate concentration
curves obtained for separate yeast expressions are shown. In
A, values (mean ± S.E.) from four to five separate
preparations are expressed as a percentage of control activity obtained
in the absence of malonyl-CoA. In B and C, the
velocity versus substrate concentration curves are given for
palmitoyl-CoA and carnitine, respectively. Individual values
(means ± S.E., n = 3-5) are expressed as a
percentage of Vmax obtained for the same
preparation. The Vmax values (mean ± S.E.)
for L-CPT I (LLL) and M-CPT I (MMM) were 8.26 ± 0.83 and
1.85 ± 0.27 nmol of palmitoylcarnitine formed per min/mg of
yeast protein, respectively, when measured at a constant carnitine
concentration (500 µM) and 8.37 ± 1.08 and
3.55 ± 0.27 nmol of palmitoylcarnitine formed per min/mg,
respectively, when measured at a constant palmitoyl-CoA concentration
(135 µM).
