| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ARTICLES |
Departments of Cell Biology and Pharmacology, Georgetown University Medical Center, Washington, DC 20007
Address all correspondence and requests for reprints to: Dr. V. Papadopoulos, Department of Cell Biology, Georgetown University Medical Center, 3900 Reservoir Road, Washington, D.C. 20007. E-mail: papadopv{at}gunet.georgetown.edu
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
The peripheral-type benzodiazepine receptor (PBR) was originally
discovered because it binds the benzodiazepine diazepam
with relatively high affinity (3). Benzodiazepines are among the most
highly prescribed drugs due to their pharmacological actions in
relieving anxiety mediated through modulating the activity of
-aminobutyric acid receptors in the central nervous system (4). PBR
is another class of binding sites for benzodiazepines distinct from the
aforementioned neurotransmitter receptors. Further studies demonstrated
that in addition to benzodiazepines, PBR binds other classes of organic
compounds with high affinity (3). PBR, although present in all tissues
examined, was found to be particularly high in steroid producing
tissues, where it was primarily localized in the outer mitochondrial
membrane (OMM) (5). An 18 kDa isoquinoline-binding protein was
identified as PBR, cloned and expressed (6). It was then demonstrated
that PBR is a functional component of the steroidogenic machinery (6, 7) mediating cholesterol delivery from the outer to the inner
mitochondrial membrane (8). Further studies demonstrated that
pharmacologically induced reduction of adrenal PBR levels in
vivo resulted in decreased circulating glucocorticoid levels (6).
In addition, targeted disruption of the PBR gene in Leydig cells
resulted in the arrest of cholesterol transport into mitochondria and
steroid formation; transfection of the mutant cells with a PBR
complementary DNA (cDNA) rescued steroidogenesis (9).
Based on the known amino acid sequence of the human and mouse 18 kDa
PBR protein a three dimensional model of this receptor protein was
developed using molecular dynamics simulations (10, 11). The receptor
model developed, where the five transmembrane domains of PBR were
modeled as five
helices, was tested as a cholesterol carrier, and
it was shown that both the human and mouse PBR can accommodate
cholesterol and function as a channel. However, this is a hypothetical
model that needs to be tested.
In the present paper, we identify a cholesterol recognition amino acid sequence on PBR, common to all proteins shown to interact with cholesterol, and demonstrate that PBR functions as a cholesterol channel-like structure, mediating cholesterol transport across membranes.
| Materials and Methods |
|---|
|
|
|---|
Construction of the mouse PBR expression vector (pET15bPBR) and
expression of recombinant PBR in bacteria
The pET system (Novagen, Madison, WI) was used to express the
MA-10 mouse PBR (mPBR) recombinant protein. The insert containing full
length coding sequence as well as the NdeI and
XhoI site extensions at the 5' and 3' ends were generated by
PCR using the following primers: ATATATACATATGCCTGAATCCT-GGGTG and
ATACTCGAGTGGGTGCCTTCACTCTG, respectively. The MA-10 full length PBR
cDNA (14) was used as template. This mPBR fragment was inserted into
pET15b vector and linearized with NdeI and XhoI
downstream of the T7lac promoter. Recombinant mPBR expression vector
was used to transform the BL21(DE3) Escherichia coli strain
where the expression of recombinant mPBR protein was induced by 1
mM isopropyl-1-thiol-ß-D-galactopyranoside (IPTG). PBR
protein expression was monitored by SDS-PAGE followed by Coomassie Blue
staining or immunoblot analysis using anti-PBR antiserum (13). Binding
specificity of the IPTG-induced PBR in Escherichia coli was
determined in binding studies where specific binding of
3H-PK 11195 (1.0 nM) was measured in the
presence of the indicated concentrations of the indicated ligands.
3H-Cholesterol uptake by E. coli cells was examined using the indicated concentrations of control or IPTG-treated transformed bacteria incubated in the presence of 6.7 nM 3H-cholesterol (50.0 Ci/mmol) for 60 min at 37 C. Specific cholesterol uptake is defined as IPTG-induced minus basal values. Escherichia coli protoplasts were prepared from cells grown in LB at 37 C to the logarithmic phase of growth and centrifuged at 10,000 x g for 5 min. The cells were washed twice in 10 mM Tris-HCl buffer, pH 8.0, and the pellet was resuspended in a solution containing 20% (wt/wt) sucrose and 0.1 M Tris-HCl, pH 8.0. The cells were then suspended in 1 ml of buffer per 10 OD450 and mixed. Within 1 min, lysozyme was added from a 2 mg/ml stock solution in distilled water to a final concentration of 100 µg/ml at 37 C, and stirring was continued for the next 12 min. The cell suspension was diluted 1:10 with 0.1 M Na2EDTA prewarmed to 37 C. Continuous stirring and slow dilution over 2.5 min prevented cell lysis. More than 99% of the cells became spherical within 10 min (15). Protoplasts (100 µg in 0.3 ml final volume) were then incubated in the presence of 6.7 nM 3H-cholesterol (50.0 Ci/mmol) for 60 min at 37 C for cholesterol uptake assays. 3H-Cholesterol-labeled membranes of IPTG-induced transformed bacteria, incubated with 30 nM 3H-cholesterol, were treated with PK 11195 or clonazepam, and the 3H-cholesterol released was quantified by liquid scintillation spectrometry.
Site-directed mutagenesis
Mutations were performed using the QuikChange Site-Directed
Mutagenesis kit from Stratagene (La Jolla, CA). Briefly,
miniprep pET-PBR plasmid dsDNA was used as template. Synthetic
oligonucleotide primer pairs containing A147T, Y153S, R155L point
mutations and PBR deletions,
520,
4151,
108119,
120133,
141152,
153169, each complementary to the
opposing strand of the vector, were extended during temperature cycling
by pfu DNA polymerase. Upon incorporation of the
oligonucleotide primers, mutated plasmids containing staggered nicks
were generated. After temperature cycling, the products were treated
with DpnI is used to digest the parental DNA template and
select for the generated mutation. The nicked vector DNAs containing
the desired mutations were then transformed into Escherichia
coli. The mutated plasmids were prepared by ABI Prism Miniprep
Kit. The generated mutations and deletions were confirmed by sequencing
using the ABI Prism Dye Terminator Cycle Sequencing ready reaction kit
(Perkin Elmer Corp., PE Applied Biosystems,
Foster City, CA). DNA sequencing was performed at the Lombardi Cancer
Center Sequencing Core Facility (Georgetown University).
Radioligand binding assays
3H-1-(2-chlorophenyl)-N-methyl-N-(1-methyl-propyl)-3-isoquinolinecarboxamide
(PK 11195) binding studies were performed as we previously
described (7, 14). The dissociation constant (Kd) and the
number of binding sites (Bmax) were determined by Scatchard
plot analysis of the data using the LIGAND program (Biosoft,
Ferguson, MO) (16).
Protein measurement
Microgram amounts of protein were quantified using the
dye-binding assay of Bradford (17) using BSA as the standard.
Statistics
The results shown represent the means ± SD or
SEM from two to six independent experiments. Statistical
analysis was performed by ANOVA followed by the Student-Newman-Keuls
test or the Dunnett multiple comparisons test using the Instat (version
2.04) package from GraphPad, Inc. (San Diego, CA).
| Results and Discussion |
|---|
|
|
|---|
40 pmol/mg protein; 7), and as
all eukaryotic cells contains endogenous cholesterol; and 2) the
Escherichia coli DE3 cells which do not express PBR (this
study), do not have endogenous cholesterol (18) and do not form
steroids. Thus, using these cell models we attempted to correlate PBR
expression with the cholesterol transport function. In addition, we
used a method with radiolabeled cholesterol (12) to quantify
cholesterol movement. This method allows for the distinction between
the exogenously supplied cholesterol and the endogenous cholesterol,
permitting an easy quantification, and direct measurement of the
pregnenolone formed, in the case of the steroidogenic cells where
cholesterol transported to IMM is cleaved by the P450scc to generate
pregnenolone. Data shown in Fig. 1
|
|
|
PBR is an 18 kDa hydrophobic protein with five putative transmembrane
domains located in the OMM. As a first step in defining the regions of
the receptor involved in the interaction with the drug ligand and
cholesterol, we constructed mutant PBRs with the deletions indicated in
the left panel of Fig. 4
. The
location of the five transmembrane regions of the receptor (I to V) is
also shown in Fig. 4
. It should be noted that the amino-terminus of
mitochondrial membrane proteins is directed toward the inside of the
organelle, whereas the carboxy-terminus is in the cytoplasmic side. The
right panel of Fig. 4
shows the effect of deletion of
specific amino acid sequences on PBR ligand binding and cholesterol
uptake examined in the bacterial system described above. Deletion of
amino acid sequences 520 and 4151 in the amino-terminus of the
receptor decreased by 3045% the ability of PBR to bind the ligand PK
11195. Our results are in agreement with previous studies on human PBR
expressed in yeast (22), although in those studies deletion of the
amino-terminus of human PBR completely abolished the ability of the
receptor to bind PK 11195. Deletion of amino acids 120133 in the
fourth transmembrane domain also decreased PBR ligand binding by 45%.
Smaller decrease (25%) of PK 11195 binding was also seen when the
regions 141152 and 153169 were deleted. These results suggest that,
although the amino- terminus of the receptor may confer the ability
to bind drug ligands, such as the isoquinoline PK 11195, amino acid
sequences in the fourth transmembrane domain may participate in the
formation of the ligand binding site. It should be noted that deletions
affecting PK 11195 ligand binding did not had a major effect on the
ability of the recombinant receptor expressed in bacterial membranes to
take up radiolabeled cholesterol.
|
Recent studies by Pikuleva et al. (23) with another protein
that interacts with cholesterol, the enzyme P450scc, indicated that a
Tyrosine in the active site of the P450scc interacts with the side
chain of cholesterol. Aligning the P450scc active site amino acid
sequence with the carboxy-terminus of PBR indicated that there might be
a common amino acid consensus pattern in these two molecules
recognizing cholesterol (Table 1
). This
consensus pattern is composed of a neutral and hydrophobic amino acid,
such as Leucine or Valine, a neutral and polar amino acid, such as
Tyrosine, and a basic amino acid, such as Arginine or Lysine. One to
five different amino acids may be placed between these three coding
amino acids. Thus, the proposed consensus pattern is
L/V-(X)15-Y-(X)15-R/K-.
Leucine or Valine will interact with the hydrophobic side chain of
cholesterol and Tyrosine will interact with the polar 3'OH-group of
cholesterol, whereas the Arginine or Lysine may help create a pocket.
This hypothesis was tested (Fig. 5
).
Replacement of Y153 by Serine or R156 by Leucine completely abolished
the ability of PBR to take up radiolabeled cholesterol. Mutation and
replacement of A147 with Threonine, did not affect cholesterol uptake
by bacteria expressing the mutated receptor. Figure 5
also shows that
the wild-type and mutated recombinant receptor proteins were expressed
at equal levels upon IPTG induction.
|
|
-monooxygenase (31), cholesterol
oxidase (32), cholesterol dehydrogenase (33), bile-salt-activated
lipase precursor (cholesterol esterase) (34), and acyl-CoA cholesterol
acyltransferase (35) we looked for the presence of the cholesterol
recognition/interaction amino acid consensus pattern
L/V-(X)15-Y-(X)15-R/K-
in these proteins. Table 1
-actin,
nonmuscle and smooth muscle myosin light chain did not contain this
cholesterol recognition/interaction consensus pattern. However, given
any tyrosine there is a reasonably high probability that this consensus
amino acid sequence will be found in many proteins. Indeed, a motif
search through the various gene data banks indicated that this amino
acid consensus pattern is present in various proteins. This is not
surprising because it is known that the cholesterol/protein interaction
plays a role not only in cholesterol transport and/or storage but also
in protein stability, folding, and/or localization. Thus, it is
possible that only in some proteins this consensus sequence will be
functional. The strength and specificity of the interaction of a
protein containing this consensus amino acid sequence with cholesterol
may be due either to the presence of a certain microenvironment, or the
location of the consensus sequence within the protein, or a specific
conformation of the protein that allows the use of this amino acid
sequence. In the latter case, it is also possible that the consensus
sequence identified represents only a portion, may the core, of a
larger motif to be identified.
Of special interest to steroidogenesis is the observation that the
cholesterol recognition/interaction amino acid consensus pattern was
found in the polypeptide DBI (3, 6) and the precursor StAR protein
(26). In search of a cytosolic steroidogenesis-stimulating factor, a
protein was purified shown to be identical to DBI (3). DBI was
originally purified from brain by monitoring its ability to displace
diazepam from its recognition sites in synaptosomes. DBI
is also identical to the acyl-CoA-binding protein (36). Purified DBI
was shown to stimulate intramitochondrial cholesterol transport and
increase pregnenolone formation by isolated mitochondria (6). Later on,
it was demonstrated that this action of DBI was mediated by PBR (3, 6).
In addition, DBI was shown to increase cholesterol loading onto
isolated P450scc (37). Thus, the identification of the cholesterol
recognition/interaction amino acid consensus pattern in DBI may help
understand its role in steroidogenesis and its direct effect on
P450scc. Interestingly, we showed in the past that the naturally
occurring processing product of DBI, the triakontatetrapeptide TTN
(DBI1750), but not the octadecaneuropeptide ODN
(DBI3350), was able to mimic the effect of DBI on
mitochondrial steroidogenesis (38). The finding that in DBI the
cholesterol recognition/interaction amino acid consensus pattern is
located in the amino acid sequence 25 to 32 (Table 1
) may now explain
this result. It should be also noted that the cholesterol
recognition/interaction amino acid consensus pattern is found in the
middle of the acyl-CoA-binding protein signature domain (amino acids 19
to 37) important in forming the acyl-CoA-binding site (36).
StAR has been found in gonadal and adrenal cells, where it is newly synthesized in response to trophic hormones, as a cytoplasmic precursor protein of 37 kDa targeted to mitochondria (26). StAR synthesis in Leydig cells begins 60 min after addition of the hormone and then parallels the capacity of the cells to produce steroids in response to tropic hormones (26, 39). The 37 kDa StAR precursor further undergoes cleavage to produce the 30 kDa mitochondrial mature StAR protein and its phosphorylated counterpart (26). This protein processing is believed to occur at the level of the outer/inner mitochondrial membrane contact sites and it has been proposed to be responsible for cholesterol transport from outer to inner mitochondrial membrane (26). Considering that the cholesterol recognition/interaction amino acid consensus pattern is found in the amino-terminus of the StAR precursor protein, which is removed from the mature protein, it is possible that the function of the precursor StAR protein is to shuttle cholesterol from intracellular stores to the outer mitochondrial membrane.
In conclusion, the results presented herein demonstrate that PBR may have a channel-like function for cholesterol in the OMM. The steroidogenic pool of cholesterol, coming from various intracellular sources, is recognized by the cholesterol recognition/interaction amino acid consensus pattern L/V-(X)15-Y-(X)15-R/K- present in the carboxy-terminus of PBR in the OMM. This pool of cholesterol enters in the OMM at the PBR sites where it remains without mixing with other membrane components. Ligand binding to the receptor induces the release of this cholesterol. Considering that PBR has been shown to be associated with the voltage-dependent anion channel (6), found in the outer/inner mitochondrial membrane contact sites, the released cholesterol could now directly access the P450scc in the IMM where it will be cleaved to pregnenolone, precursor of all steroids.
In addition to being a precursor for steroid hormone synthesis, cholesterol is an essential structural element of cellular membranes and a precursor for the synthesis of bile acids and lipoproteins. Mammalian cells obtain cholesterol by internalization of low density lipoproteins or by de novo synthesis in the endoplasmic reticulum. The subcellular distribution of cholesterol suggests that cholesterol is trafficked and incorporated quickly from the sites of acquisition to the target membrane (40). Thus, a tissue and cell specific cholesterol homeostasis is achieved. Considering the widespread occurrence of PBR and its tissue and cell specific subcellular localization (3, 6), these results suggest a more general role for PBR in intracellular cholesterol transport and compartmentalization.
| Acknowledgments |
|---|
| Footnotes |
|---|
2 Supported by a Research Career Development Award (HD-01031) from the
National Institute of Child Health and Human Development, NIH. ![]()
Received May 19, 1998.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Das, M. A. Davis, and L. L. Rudel Identification of putative active site residues of ACAT enzymes J. Lipid Res., August 1, 2008; 49(8): 1770 - 1781. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Kirkham, S. J. Nixon, M. T. Howes, L. Abi-Rached, D. E. Wakeham, M. Hanzal-Bayer, C. Ferguson, M. M. Hill, M. Fernandez-Rojo, D. A. Brown, et al. Evolutionary analysis and molecular dissection of caveola biogenesis J. Cell Sci., June 15, 2008; 121(12): 2075 - 2086. [Abstract] [Full Text] [PDF] |
||||
![]() |
X.-H. Sun, D. C. Flynn, V. Castranova, L. L. Millecchia, A. R. Beardsley, and J. Liu Identification of a Novel Domain at the N Terminus of Caveolin-1 That Controls Rear Polarization of the Protein and Caveolae Formation J. Biol. Chem., March 9, 2007; 282(10): 7232 - 7241. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Hager-Braun, H. Katinger, and K. B. Tomer The HIV-Neutralizing Monoclonal Antibody 4E10 Recognizes N-Terminal Sequences on the Native Antigen. J. Immunol., June 15, 2006; 176(12): 7471 - 7481. [Abstract] [Full Text] [PDF] |
||||
![]() |
R. G. Parton, M. Hanzal-Bayer, and J. F. Hancock Biogenesis of caveolae: a structural model for caveolin-induced domain formation. J. Cell Sci., March 1, 2006; 119(Pt 5): 787 - 796. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Liu, Z.-X. Yao, and V. Papadopoulos Cytochrome P450 17{alpha} Hydroxylase/17,20 Lyase (CYP17) Function in Cholesterol Biosynthesis: Identification of Squalene Monooxygenase (Epoxidase) Activity Associated with CYP17 in Leydig Cells Mol. Endocrinol., July 1, 2005; 19(7): 1918 - 1931. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Jamin, J.-M. Neumann, M. A. Ostuni, T. K. N. Vu, Z.-X. Yao, S. Murail, J.-C. Robert, C. Giatzakis, V. Papadopoulos, and J.-J. Lacapere Characterization of the Cholesterol Recognition Amino Acid Consensus Sequence of the Peripheral-Type Benzodiazepine Receptor Mol. Endocrinol., March 1, 2005; 19(3): 588 - 594. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. Hauet, Z.-X. Yao, H. S. Bose, C. T. Wall, Z. Han, W. Li, D. B. Hales, W. L. Miller, M. Culty, and V. Papadopoulos Peripheral-Type Benzodiazepine Receptor-Mediated Action of Steroidogenic Acute Regulatory Protein on Cholesterol Entry into Leydig Cell Mitochondria Mol. Endocrinol., February 1, 2005; 19(2): 540 - 554. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. NAVARRO, B. ANAND-APTE, and M.-O. PARAT A role for caveolae in cell migration FASEB J, December 1, 2004; 18(15): 1801 - 1811. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. A. Spencer, P. Wang, D. Li, J. S. Russel, D. H. Blank, J. Huuskonen, P. E. Fielding, and C. J. Fielding Benzophenone-containing cholesterol surrogates: synthesis and biological evaluation J. Lipid Res., August 1, 2004; 45(8): 1510 - 1518. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Lindemann, A. Koch, B. Degenhardt, G. Hause, B. Grimm, and V. Papadopoulos A Novel Arabidopsis thaliana Protein is a Functional Peripheral-Type Benzodiazepine Receptor Plant Cell Physiol., June 15, 2004; 45(6): 723 - 733. [Abstract] [Full Text] [PDF] |
||||
![]() |
M.-K. Chen, K. Baidoo, T. Verina, and T. R. Guilarte Peripheral benzodiazepine receptor imaging in CNS demyelination: functional implications of anatomical and cellular localization Brain, June 1, 2004; 127(6): 1379 - 1392. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Ostuni, K. Marazova, G. Peranzi, B. Vidic, V. Papadopoulos, R. Ducroc, and J.-J. Lacapere Functional characterization and expression of PBR in rat gastric mucosa: stimulation of chloride secretion by PBR ligands Am J Physiol Gastrointest Liver Physiol, June 1, 2004; 286(6): G1069 - G1080. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Nollmann, R. Gilbert, T. Mitchell, M. Sferrazza, and O. Byron The Role of Cholesterol in the Activity of Pneumolysin, a Bacterial Protein Toxin Biophys. J., May 1, 2004; 86(5): 3141 - 3151. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. Giatzakis and V. Papadopoulos Differential Utilization of the Promoter of Peripheral-Type Benzodiazepine Receptor by Steroidogenic Versus Nonsteroidogenic Cell Lines and the Role of Sp1 and Sp3 in the Regulation of Basal Activity Endocrinology, March 1, 2004; 145(3): 1113 - 1123. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Bribes, D. Carriere, C. Goubet, S. Galiegue, P. Casellas, and J. Simony-Lafontaine Immunohistochemical Assessment of the Peripheral Benzodiazepine Receptor in Human Tissues J. Histochem. Cytochem., January 1, 2004; 52(1): 19 - 28. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Gazouli, Z. Han, and V. Papadopoulos Identification of a Peptide Antagonist to the Peripheral-Type Benzodiazepine Receptor That Inhibits Hormone-Stimulated Leydig Cell Steroid Formation J. Pharmacol. Exp. Ther., November 1, 2002; 303(2): 627 - 632. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Culty, L. Luo, Z.-X. Yao, H. Chen, V. Papadopoulos, and B. R. Zirkin Cholesterol Transport, Peripheral Benzodiazepine Receptor, and Steroidogenesis in Aging Leydig Cells J Androl, May 1, 2002; 23(3): 439 - 447. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Li, B. Degenhardt, D. Tobin, Z.-x. Yao, K. Tasken, and V. Papadopoulos Identification, Localization, and Function in Steroidogenesis of PAP7: A Peripheral-Type Benzodiazepine Receptor- and PKA (RI{alpha})-Associated Protein Mol. Endocrinol., December 1, 2001; 15(12): 2211 - 2228. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. M. Stocco Tracking the Role of a StAR in the Sky of the New Millennium Mol. Endocrinol., August 1, 2001; 15(8): 1245 - 1254. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. Li, Z.-x. Yao, B. Degenhardt, G. Teper, and V. Papadopoulos Cholesterol binding at the cholesterol recognition/ interaction amino acid consensus (CRAC) of the peripheral-type benzodiazepine receptor and inhibition of steroidogenesis by an HIV TAT-CRAC peptide PNAS, January 23, 2001; (2001) 31461598. [Abstract] [Full Text] |
||||
![]() |
N. Boujrad, B. Vidic, M. Gazouli, M. Culty, and V. Papadopoulos The Peroxisome Proliferator Perfluorodecanoic Acid Inhibits the Peripheral-Type Benzodiazepine Receptor (PBR) Expression and Hormone-Stimulated Mitochondrial Cholesterol Transport and Steroid Formation in Leydig Cells Endocrinology, September 1, 2000; 141(9): 3137 - 3148. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. A. Yeliseev and S. Kaplan TspO of Rhodobacter sphaeroides. A STRUCTURAL AND FUNCTIONAL MODEL FOR THE MAMMALIAN PERIPHERAL BENZODIAZEPINE RECEPTOR J. Biol. Chem., February 25, 2000; 275(8): 5657 - 5667. [Abstract] [Full Text] |