|
Originally published In Press as doi:10.1074/jbc.M004913200 on July 11, 2000
J. Biol. Chem., Vol. 275, Issue 42, 32807-32815, October 20, 2000
A Targeted Apolipoprotein B-38.9-producing Mutation Causes
Fatty Livers in Mice Due to the Reduced Ability of Apolipoprotein
B-38.9 to Transport Triglycerides*
Zhouji
Chen ,
Robin L.
Fitzgerald,
Maurizio R.
Averna, and
Gustav
Schonfeld
From the Division of Atherosclerosis, Nutrition and Lipid Research,
Department of Medicine, Washington University School of Medicine,
St. Louis, Missouri 63110
Received for publication, June 6, 2000, and in revised form, July 10, 2000
 |
ABSTRACT |
Nonphysiological truncations of apolipoprotein
(apo) B-100 cause familial hypobetalipoproteinemia (FHBL) in humans and
mice. An elucidation of the mechanisms underlying the FHBL phenotypes may provide valuable information on the metabolism of apo B-containing lipoproteins and the structure-function relationship of apo B. To
generate a faithful mouse model of human FHBL, a subtle mutation was
introduced into the mouse apo B gene by targeting embryonic stem
cells using homologous recombination followed by removal of the
selection marker gene by Cre-loxP-mediated site-specific recombination.
The engineered mice bear a premature stop codon at residue 1767 and a
42-base pair loxP inserted into intron 24 of the apo B gene, thus
closely resembling the apo B-38.9-producing mutation in humans. Apo
B-38.9 was the sole apo B protein in homozygote (apob38.9/38.9) plasma. In heterozygotes
(apob+/38.9), apo B-100 and apo B-48
were reduced by 75 and 40%, respectively, and apo B-38.9 represented
20% of total circulating apo B. Hepatic apo B-38.9 mRNA levels
were reduced by 40%. In cultured
apob+/38.9 hepatocytes, apo B-100 was
produced in trace quantities, and the synthesis rate of apo B-38.9
relative to apo B-48 was reduced by 40%. However, almost
equimolar amounts of apo B-38.9 and apo B-48 were secreted into the
media. Pulse-chase studies revealed that apo B-38.9 was secreted
at a faster rate and more efficiently than apoB-48. Nevertheless, both
apob+/38.9 and
apob38.9/38.9 mice had reduced hepatic triglyceride
secretion rates and fatty livers. Thus, low mRNA levels or
defective secretion of apo B-38.9 may not be responsible for the FHBL
phenotypes caused by the apo B-38.9 mutation. Rather, a reduced
capacity of apo B-38.9 for triglyceride transport may account for the
fatty livers in these mice.
| |
INTRODUCTION |
Apolipoprotein B (apo
B)1 is the major structural
protein component of the triglyceride-rich very low density
lipoproteins (VLDLs) secreted from the liver and the chylomicrons
secreted from the intestine. The full-length apo B (apo B-100) is
composed of 4536 amino acid residues (1). Due to a posttranscriptional modification of the apo B gene (apob) mRNA that converts
codon 2153, CAA, to a stop codon, UAA, the apo B protein also naturally exists in a truncated form corresponding to the
NH2-terminal 48% of apo B-100, designated apo B-48 (2, 3).
In humans, apo B-48 is produced only by the intestine, whereas
both the intestine and the liver in rodents secrete apo B-48
(4, 5). High levels of plasma apo B-containing lipoproteins are a
major risk for the development of atherosclerotic diseases. Therefore,
mechanisms controlling apo B synthesis and secretion are under active investigation.
Nonsense and frameshift mutations in apob that
produce nonphysiological COOH-terminal truncations of apo B-100 cause
familial hypobetalipoproteinemia (FHBL) in humans, an autosomal
codominant disorder characterized by low levels (<5th percentile) of
plasma apo B and low density lipoprotein (LDL) cholesterol (6, 7). Numerous forms of truncated apo B, with sizes ranging from apo B-2 to
apo B-89, have been identified in FHBL subjects (6, 8), and they are
usually present in plasma at much lower concentrations than apo B-100
due to low production and high fractional catabolic rates (6, 9-11).
The molecular mechanism(s) underlying these metabolic alterations is
yet to be elucidated.
In addition to the abnormalities in plasma cholesterol and apo B
levels, fatty livers also occur in heterozygous FHBL subjects (12-15).
Truncation-producing mutations in apob might impair the ability of the liver to secrete triglycerides due to low synthesis rates of the truncated apo B, its enhanced intracellular degradation, or its impaired assembly with lipids, resulting in a reduced capacity of the mutant apo B for transporting triglycerides. Gaining insights into apo B and lipid metabolism in FHBL at the molecular and cellular level may provide information critical not only to the elucidation of
mechanisms for this disorder but also to defining further the structure-function relationship of apo B-100.
However, FHBL patients are usually asymptomatic (6, 7), posing ethical
barriers to obtaining liver and intestinal biopsies for in-depth
studies. In recent years, several lines of mice bearing various forms
of apo B truncations have been generated by targeting apob
in embryonic stem (ES) cells (16-19). In these mice, the targeted apob alleles were expressed at low levels, leading to very
low concentrations of truncated apo Bs in the plasma (16-19). It is not known whether these low observed levels of mRNA expression faithfully represent the mechanisms of FHBL existing in humans because
in the mice, multiple copies of the gene-targeting construct were
inserted into apob along with the desired mutations
(16-19). The presence of these foreign DNA sequences in the genome
might exert unnatural effects on the expression of the targeted gene or
on genes nearby (20). In the meantime, due to the low intracellular concentrations of the mutant apo Bs in the mouse livers, it has been
difficult to study their posttranslational fates. Consequently, little
is known about the cellular secretion and the function of the mutant
apo Bs. Furthermore, the effect of modifying mouse apob on
hepatic triglyceride secretion is yet to be explored.
To generate a faithful mouse model of FHBL, we have used ES cell
homologous recombination technology to introduce an apo
B-38.9-specifying single-nucleotide deletion into the exon 26 of mouse
apob. The selection marker gene sequence that was introduced
into intron 24 of apob during the homologous recombination
step was subsequently deleted using the Cre-loxP system (20, 21). Thus,
the targeted apob allele of the resultant mice contains only
a subtle mutation in the coding region plus a 34-base pair (bp) loxP
sequence inserted into the middle of intron 24. This model more closely
resembles the naturally occurring human apob mutations than
the previously reported apob gene-targeted FHBL mouse
models, in which large segments of extraneous DNA were retained. The
apo B-38.9-bearing mice display fatty livers as well as FHBL
phenotypes. We provide evidence suggesting that low mRNA levels or
defective secretion of apo B-38.9 may not be responsible for the
manifestation of these phenotypes. The apo B-38.9 truncation is
secreted by hepatocytes more efficiently than apo B-48, but it has a
reduced capacity for cellular triglyceride transport.
| |
EXPERIMENTAL PROCEDURES |
Preparation of Targeting Construct--
To construct a sequence
replacement gene-targeting vector, an 11.5-kb
BglII-BglII genomic fragment spanning mouse
apob intron 23 through the 5'-end of exon 27 was isolated
from a BAC clone that contains a mouse apob insert derived
from W-4 (129/SvJ) ES cell genomic DNA (clone 12339) (Genome System,
St. Louis, MO). This genomic apob fragment was subcloned
into a modified, HindIII-negative Sp72 plasmid (Promega,
Madison, WI) at the BglII site (designated mB11.5K). Based
on sequence analysis, deletion of nucleotide 5449 on mouse apo B
cDNA is predicted to produce a premature stop codon at the same
position (i.e. residue 1767) as that occurring in the apo
B-38.9 mutation of human FHBL subjects (22) (Fig. 1A). To
generate this mutation, an 8-kb HindIII-HindIII
fragment spanning the entire exon 26 (Fig. 1B) was excised
from mB11.5K and subcloned into a pAlter-1 vector (Promega) at the
HindIII site for site-directed mutagenesis. A 38-base
oligomer was used as a mutagenic primer (5'-CTCAAAAATGGACAAGTTAACAGTGGAGACAAGTTC-3') to delete nucleotide 5449 and to create a novel HpaI site. After mutagenesis, a 7-kb HindIII-XhoI fragment was excised from the mutant
insert and subcloned into pSP-72 at HindIII and
XhoI sites. In the meantime, the 5'-region of the 11.5-kb
mouse apob BglII fragment was cut from mB11.5K using
XhoI (one XhoI site is located in exon 25, whereas the other is in the polylinker of Sp-72 encompassing the
5'-BglII site of the insert) and subcloned into a
HindIII-negative Sp 72 plasmid at XhoI. A
loxP-PGK-Neo-loxP cassette (20, 21) was inserted into the resultant
plasmid at HindIII site using a filling in/ligation strategy
that abolished the HindIII site. Thereafter, the entire insert containing a loxP-PGK-Neo-loxP cassette was excised from the
vector using XhoI and inserted into the Sp72 vector
containing the mutated exon 26 fragment at XhoI site. The
apo B-38.9-producing gene-targeting vector was obtained by digesting
the resultant plasmid with HindIII and BglII.
This construct contained the fragment of mouse apob spanning
intron 23 through the 3'-end of exon 26 with a loxP-Neo gene cassette
(1.7 kb) inserted in the middle of intron 24 (Fig. 1B).
Generation of Mutant Mice Producing Apo B-38.9
Truncations--
The 129/SvJ blastocyst-derived ES cells, RW4 cells
(Dr. T. J. Ley, Washington University School of Medicine) were
transfected with the gel-purified apo B-38.9 targeting construct, and
G418-resistant clones were isolated as described (20). Clones that had
undergone homologous recombination were identified by Southern analysis of HpaI-digested genomic DNA using a 0.3-kb
HindIII-HpaI genomic fragment (Fig.
1B) as an external probe. The fidelity of the targeted mutation was confirmed by sequence analysis.
To delete the PGK-neo cassette from the site of homologous
recombination, homologous recombinant ES cells were stably transfected with a plasmid encoding the Cre recombinase and PGK-hygro as described (20, 21); hygromycin-resistant clones were screened for deletion of the
PGK-neo marker by Southern analysis using a 0.5-kb
HindIII-XhoI genomic fragment as a probe
(internal probe; Fig. 1C).
C57BL/6 blastocysts were microinjected with ES cells from two PGK-neo
marker-deleted, apob-targeted ES cell clones and implanted into pseudopregnant Swiss Webster foster females as described previously (20) to produce chimeric progeny. Male chimeras were bred
with C57BL/6 females to generate heterozygous apo B-38.9-mice (apob+/38.9). The
apob+/38.9 mice were intercrossed to
create homozygotes (apob38.9/38.9). They were also
bred with apo B-48-only (apob48/48) mice (23) (The
Jackson Laboratory, Bar Harbor, ME) to generate apob48/38.9 mice.
All mice were weaned at 3 weeks of age, housed in a specific
pathogen-free barrier facility with a 12-h light/dark cycle, and fed a
regular mouse chow diet (Ralston Purina, St. Louis, MO). All mice used
in this study were of a mixed genetic background with 50% C57BL/6 and
50% 129/SvJ.
Anti-mouse Apo B Polyclonal Antibodies--
Rabbit anti-mouse
apo B antisera were developed in our laboratory previously (24). In
addition, to produce polyclonal antibodies specific for the
NH2-terminal region of mouse apo B, a glutathione S-transferase (GST) fusion protein containing amino acids
26-289 of mouse apo B was generated using a pGEX-2T vector (Amersham Pharmacia Biotech) and used as an antigen to immunize two
rabbits using a standard protocol.
Western Blot Detection of Apo B and Lipoprotein
Fractionation--
To perform Western blot analysis, mouse plasma (3 µl/well) or hepatocyte lysates (250 µg protein/well) were
electrophoresed on 3-12% gradient SDS-PAGE gels under reducing
conditions and electrotransferred onto Immobilon (Millipore Corp.,
Bedford, MA). Apo B protein bands were visualized using rabbit
anti-mouse apo B antisera or the anti-GST-apo B fusion protein
antiserum as primary antibodies and an ECL Western blot detection kit
(Amersham Pharmacia Biotech). The ECL signals were quantified both by
analyzing the density of the protein bands on Kodak XAR-5 film using
Sigma gel computer software (SPPS Science Corp., Chicago, IL) and by
direct measurement of the signal intensity using a Kodak image-analyzer system (Eastman Kodak Co.).
A fast performance liquid chromatography (FPLC) Superose column (24)
was used to assess the distribution of lipids and apo B within the
lipoprotein fractions of mouse plasma. The distribution of apo B in
each lipoprotein fraction was determined by Western blot analysis.
RNA Analysis--
Total RNA was isolated from mouse livers and
small intestines by a single step isolation method using
RNAZolTM B (Tel-Test, Inc., Friendswood, TX). A primer
extension assay was used to determine the relative amounts of
apob transcripts arising from different apob
alleles. To produce cDNA templates for this assay, a one-step
reverse transcription-polymerase chain reaction (RT-PCR) kit (Roche
Molecular Biochemicals) was used to generate a 1.35-kb fragment
of mouse apo B cDNA spanning the 3' region of exon 25 and the 5'
region of exon 26 encompassing the apo B-38.9 mutation from
liver or intestinal RNA. We used the upstream PCR primer
5'-ACAACTGGTCAGCCTCCTACACTGG-3' and the downstream PCR primer
5'-GTTGGTCAAATCTAGAGCACC-3'. The 1.35-kb RT-PCR products were
gel-purified and used as templates for primer extension assay. A
32P-labeled 29-base-oligo primer
(5'-TCTCTCACTGGACTTCTTCTCAAAAATGG-3') corresponding to sequence
immediately upstream of the apo B-38.9 mutation site was annealed to
the cDNA templates at 50 °C for 30 min and extended with
AMV-reverse transcriptase (Roche Molecular Biochemicals) in the
presence of dideoxy-GTP at 37 °C for 90 min. Because of a single-bp
deletion and an A  G exchange mutation introduced into the
apob38.9 allele, extension of
apob38.9 cDNA yielded a 34-bp product,
whereas the apob+ and
apob48 cDNA extension gave rise to a 42-bp
product. The extension products were resolved in an 8% polyacrylamide
gel containing 7.5 M urea. The gel was dried and subjected
to autoradiography and PhosphorImager quantification (Image
Quant version 3.2; Molecular Dynamics, Sunnyvale, CA). RNA samples from
three mice for each genotype were analyzed, and the data are presented
as mean ± S.D.
Northern blot analysis was used to determine the size and the levels of
apob transcripts in the livers of apo B-38.9-producing mice
and their wild type littermates. Thirty micrograms of total RNA were
electrophoresed on a 0.8% agarose-formaldehyde gel and transferred and
immobilized onto a GeneScreen nylon membrane (NEN Life Science
Products). A SacI-HpaI-digested fragment
of mouse apob corresponding to the 3'-region of exon 26 was
radiolabeled with [ -32P]dCTP using a random primer
labeling kit (Roche Molecular Biochemicals) and used to probe
apob transcripts. Northern hybridization was carried out
using Rapid-hyb buffer according to the manufacturer's instructions
(Amersham Pharmacia Biotech). The blots were also stripped and
probed with a rat glyceraldehyde-3-phosphate dehydrogenase cDNA
(Sigma). The hybridization signals were quantified by PhosphorImager (Molecular Dynamics, Sunnyvale, CA). The relative apo B mRNA levels are expressed as ratios of apoB mRNA/glyceraldehyde-3-phosphate dehydrogenase mRNA. Data are presented as mean ± S.D
(n = 3 animals).
Preparation of Mouse Hepatocyte Cultures--
To isolate mouse
hepatocytes, the portal vein of each mouse (males, 10-12 weeks old)
was cannulated under anesthesia with Metophane and pentobarbitol, and
the liver was perfused with oxygenated Ca2+- and
Mg2+-free Hanks'-buffered saline solution for 5 min
followed by perfusion of oxygenated Hanks'-buffered saline solution
containing 0.05% collagenase (Type IV; Sigma) for 5-10 min.
The perfusate was drained through an incision in the hepatic vein.
After perfusion, the liver was removed and submerged into ice-cold
DMEM, and the cells were released gently. The resultant cell suspension
was then filtered through a nylon mesh and centrifuged at 50 × g for 1 min. The cell pellets were resuspended in ice-cold
DMEM and repelleted. After three wash cycles, cells were resuspended in
DMEM containing 10% fetal bovine serum (FBS). Viability of the cells
(~80%) was determined by Trypan exclusion. Cells were plated onto
six-well plates (0.7 × 106 cells/well) or 100-mm
dishes (8 × 106 cells/dish) coated with
poly-D-lysine (Sigma) and incubated at 37 °C under 5%
CO2 in 10% FBS/DMEM. After 1 h of attachment, the cell monolayers were washed twice and incubated in 10% FBS/DMEM until
use. All experiments involving cultured hepatocytes were commenced 7-8
h after the cells were cultured.
Apo B Secretion from Cultured Hepatocytes--
To assess the apo
B secretion by the cultured hepatocytes, the cells cultured on 100-mm
dishes were washed three times with phosphate-buffered saline and then
once with DMEM. They were then cultured in 10 ml of serum-free DMEM for
3 h. The conditioned media were collected and concentrated using
Centricon-30 (Amicon Inc., Beverly, CA), and the cells were washed
twice with phosphate-buffered saline and harvested. Aliquots of the
concentrated media and cell lysates were subjected to SDS-PAGE and
Western blot analysis to determine the relative amounts of cellular and
secreted apo B. The relative apo B contents were normalized by cellular
protein contents. Three independent experiments were performed. The
data are presented as mean ± S.D.
Metabolic Labeling and Pulse-Chase Studies to Determine Apo B
Synthesis and Secretion--
Continuous labeling experiments were
performed to determine the synthetic rates and secretion rates of wild
type apo B and apo B-38.9 in hepatocytes from heterozygous mice. After
the initial 7-h incubation period, cells on six-well plates were washed
three times with phosphate-buffered saline and incubated in methionine (Met)- and cysteine (Cys)-free DMEM for 30 min to diminish the cellular
pool of Met and Cys. After this incubation, the media were replaced
with 1 ml of Met- and Cys-free DMEM containing 150 µCi of
35S-Promix (530 MBq/ml; Amersham Pharmacia Biotech), an
L-[35S]methionine and
L-[35S]cysteine metabolic labeling solution,
for the specified time periods. At the end of each incubation period,
media were collected and spun to remove trace amounts of the cells, and
the cell monolayers were lysed in an immunoprecipitation buffer (150 mM NaCl, 5 mM EDTA, 50 mM Tris, pH
7.4, 0.0625 M sucrose, 0.5% Triton X-100, and 0.5% sodium
deoxycholate). Appropriate amounts of a mixture of protease inhibitors
(Roche Molecular Biochemicals) were added to the media and cell lysates.
To determine the secretion efficiency of the newly synthesized apo B,
cells were washed and labeled as described above for 30 min. The media
were replaced with 1 ml of DMEM containing 10 mM Met and 3 mM Cys, and cells were cultured for the specified time
periods. Cells and media were then processed as described above for
continuous labeling experiments.
Immunoprecipitations were carried out to quantify the labeled apo B in
the cell or the medium. For this purpose, 100 µl of 5×
immunoprecipitation buffer were added to each tube containing the
conditioned media. Media and cell lysates were incubated at 4 °C
with gentle rotation for 12 h and then centrifuged at 10,000 × g for 5 min to remove any unsolubilized proteins or cell
debris. The resultant media and cell lysates were incubated with rabbit anti-mouse apo B antisera for 6-h at 4 °C, and the immunoreaction complex was precipitated with protein A-agarose (Life Technologies, Inc.). After being washed five times with 0.5× immunoprecipitation buffer, the immunoprecipitates were dissolved in SDS-PAGE gel loading
buffer and resolved on a 3-12% gradient gel under reducing conditions. The gels were dried, and the radioactivity associated with
apo B bands was quantified using a PhosphorImager. Three independent
experiments were carried out for these studies.
Determination of Triglyceride Secretion from Cultured
Hepatocytes--
After the initial culture period, cells on six-well
plates were washed three times with phosphate-buffered saline and then incubated in 1 ml of DMEM containing 200 µCi of
[2-3H]glycerol (20 Ci/mmol; American Radiolabeled
Chemicals, Inc., St. Louis, MO) for the specified time periods. At the
end of each incubation, media were collected and centrifuged to remove
any detached cells. Cell monolayers were scraped and harvested. Lipid in the media and cells were extracted (25) and separated by TLC. The
triglyceride spots were scraped off and counted for 3H
radioactivity. Triplicate incubations were performed for each time point.
Determination of in Vivo Secretion Rates--
Hepatic production
of VLDL-triglyceride was measured in littermates of
apob+/+,
apob+/38.9, and
apob38.9/38.9 mice (male, 14 weeks old) after
intravenous injection of Triton WR 1339 (Sigma) as described (26). Mice
were fed a fat-free, high carbohydrate diet for 8 h prior to the
experiments, and Triton WR 1339 in 100 µl of saline (500 mg/kg
of body weight) was injected into the mice under light anesthesia with
Metophane. Tail vein blood samples were taken at the specified times
after injection for triglyceride measurement.
Quantification of Liver Lipid Contents and Oil red O
Staining of Liver Sections--
Lipids were extracted from liver
tissues as described (25). The dried lipid extracts were dissolved in
1% Triton X-100 in chloroform, dried under a stream of N2,
and redissolved in H2O as described (27) for determination
of triglycerides, total cholesterol, and phospholipids using enzymatic
kits (WAKO Chemicals USA, Inc., Richmond, VA). The hepatic lipid
concentrations were expressed as µg of lipid/mg of protein.
Oil-red O staining of livers was performed on frozen liver sections (10 µm in thickness). After being fixed in 40% formaldehyde, the
sections were stained in 0.3% Oil red O in 60% isopropanol for 10 min, washed in tap water, and then counterstained in hematoxylin.
Miscellaneous Procedures--
Plasma levels of cholesterol,
triglycerides, and phospholipids were determined enzymatically (WAKO
Chemicals USA, Inc.). Cellular protein contents were determined using a
modified Lowry method (28). Student's t test and analysis
of variance were performed to determine the levels of significance of differences.
| |
RESULTS |
Apo B-38.9-producing Mice--
The apo B-38.9-producing mutation
in humans is caused by a single base pair deletion at nucleotide 5449, leading to a frameshift and subsequently converting codon 1767 (Val)
into a stop codon (22). Deleting the nucleotide at the same position in
mouse apob also converts codon 1767 into a stop codon (Fig.
1A). A replacement type
gene-targeting vector was used to introduce this mutation into mouse ES
cell apob via homologous recombination, followed by removal
of the marker gene from the targeted allele via Cre-loxP-mediated site-specific recombination (Fig. 1B). After electroporation
of the gene-targeting construct into ES cells, 120 colonies were picked
and screened for homologous recombination events by Southern blot
analysis with a 3'-flanking probe that is external to the targeting
construct (Fig. 1B). Two targeted clones were identified, expanded, and transfected with a PGK-hygro-Cre vector to express the
Cre recombinase. Clones were screened for deletion of the marker gene
using a Neo gene probe and a mouse apo B probe corresponding to exon 25 (internal probe, Fig. 1B). Five PGK-neo-negative clones were
identified, as indicated by the shift in the size of the HpaI-HpaI fragment hybridized to the internal
probe on Southern analysis (4.7 kb with the Neo gene versus
3.0 kb without the Neo gene) and the lack of hybridization to the
labeled Neo gene cDNA probe (not shown). High percentage male
chimeras were bred with C57BL/6 females. The
apob+/38.9 offspring developed normally
and were fertile. To avoid any potential confounding effect of
expression of the Cre-transgene on lipoprotein metabolism, we used only
Cre gene-negative offspring for further breeding. Typical Southern blot
analysis and PCR analysis used to screen for mutant animals are shown
in Fig. 1, panels C, D, and E. Western
analysis showed that the mouse and human apo B-38.9 are of very similar
apparent molecular weights (Fig. 1, F and G).
View larger version (41K):
[in this window]
[in a new window]
|
Fig. 1.
Generation of apo B-38.9-bearing mice.
A, alignment of partial nucleotide sequences of human and
mouse apo B cDNA demonstrating the position of premature stop codon
caused by the apo B-38.9-specifying mutation. The nucleotide deleted in
apo B-38.9-carrying FHBL humans (22) is shown in boldface
and indicated by an arrow, whereas the resultant premature
stop codons (TAA) are boldface and underlined.
Partial sequence of the mutagenic (mut.) primer used for
site-directed mutagenesis to create this mutation in the mouse
apob is also shown, and the mismatched nucleotides meant to
create a novel HpaI site are underlined.
B, gene targeting strategy. A sequence replacement-type gene
targeting construct was prepared using a 11.5-kb
BglII-BglII mouse apo B genomic fragment (intron
23 to exon 27) isolated from a BAC clone containing an RW-4
(129/SvJ) ES cell-derived apob insert as described under
"Experimental Procedures." The targeting construct contained the
fragment of mouse apob spanning exon 23 through the 3'-end
of exon 26 with a loxP-PGK neo-loxP cassette (1.7 kb) inserted to
intron 24 at HindIII by filling in-ligation, thereby
abolishing the internal HindIII restriction site. A 0.3-kb
mouse genomic fragment (HindIII-HpaI)
corresponding to the 3'-end of apob exon 26 was used as an
external probe for Southern analysis to screen for homologous
recombinant ES cell clones. This probe hybridized to 9- and 6-kb
HpaI-digested genomic fragments of the wild type and
targeted apob alleles, respectively. Overexpression of Cre
recombinase in the targeted ES cells induced deletion of the PGK
neo-loxP sequence, leaving only subtle mutations in the targeted
allele. An internal probe spanning the 3'-end of intron 24 and exon 25 (HindIII-XhoI) was used for Southern analysis to
screen for PGK neo-loxP-deleted clones. C, a typical
Southern blot showing hybridization of 32P-labeled external
probes to HpaI-digested genomic DNA from
apob+/+,
apob+/38.9, and
apob38.9/38.9 mice. D, Southern analysis
confirming the absence of PGK-neo sequence in the apo B-38.9 carrying
mice. DNA was digested with HpaI and 32P-labeled
internal probes were used for Southern analysis. The 9-kb bands
represent the wild type alleles, and the 4.7-kb band represents the
targeted allele before the PGK-neo sequence was deleted (ES cell clone
89). The reduction in the size of the targeted allele-band from 4.7 to
3 kb indicates the absence PGK-neo marker gene in the
apob+/38.9 mice. E, PCR and
restriction-digestion analysis confirming apo B-38.9 mutation. A 296-kb
fragment encompassing the apo B-38.9 mutation site was amplified and
restriction-digested with HpaI. Lanes 1, 3, and
5, digestion with HpaI; lanes 2, 4, and 6, uncut controls. F and G,
detection of apo B-38.9 by Western analysis. Two microliters of mouse
(F) or human (G) plasma were separated on a
3-12% SDS-PAGE gel. Polyclonal rabbit anti-mouse apo B antisera were
used for detection of mouse apo B, whereas a monoclonal anti-human apo
B antibody (C1.4) was used for the human samples. Note that mouse apo
B-38.9 and human apo B-38.9 are of nearly identical apparent molecular
weights.
|
|
Intercrosses of apob+/38.9 mice produced
viable offspring apob38.9/38.9. The
apob38.9/38.9 mice were apparently healthy and
fertile. However, the numbers of apob38.9/38.9 mice
born were significantly smaller than predicted (p < 0.05 by  2). Of 158 offspring, there were 38 apob+/+, 106 apob+/38.9, and 14 apob38.9/38.9 mice. The
apob+/38.9 mice were also mated with
apob48/48 mice (23) to generate
apob48/38.9 mice.
Rabbit antisera raised against a GST fusion protein containing amino
acids 26-289 of mouse apo B were used to determine the precise
relative molar concentrations of plasma apo B-100, apo B-48, and
apo-B38.9 in the mice. Because apo B-100, apo B-48, and apo B-38.9 all
contained the entire apo B protein sequence included in this fusion
protein, these antisera are expected to recognize epitopes common to
these three different forms of apo B, thus providing accurate molar
ratios for these apo Bs. On Western blotting, the anti-mouse apo B and
the anti-fusion protein antibodies provided similar apo B protein
ratios (data not shown), indicating that the anti-mouse apo B antiserum
(R272) also recognized epitopes that are common to mouse apo B-100, apo
B-48, and apo B-38.9. The relative concentrations of plasma apo B-100,
apo B-48, and apo B-38.9 in various mice are shown in Fig.
2. Apo B-38.9 accounted for 20 and 30%
of the total plasma apo B molecules in
apob+/38.9 and
apob48/38.9 mice, respectively (Fig. 2A).
Plasma levels of apo B proteins were reduced by 45 and 85% in
apob+/38.9 and
apob38.9/38.9 mice, respectively (Fig.
2A). As previously reported in other FHBL mouse models
(16-19), a much greater decrease was seen in plasma levels of apo
B-100 than those of apo B-48 (70% versus 40%) (Fig.
2A).
View larger version (47K):
[in this window]
[in a new window]
|
Fig. 2.
Effects of the apo B-38.9-producing mutation
on plasma apo B levels. Plasma samples were obtained from mice
after a 4-h fast and subjected to SDS-PAGE (3-12% gel) (2 µl/well).
Rabbit antisera raised against a GST-mouse apo B (amino acids 26-289)
fusion protein were used as primary antibodies for Western analysis
using an ECL kit. The ECL signals were quantified as described under
"Experimental Procedures." A, each data point represents
the mean ± S.D. (n = 4 animals). B, a
representative Western blot.
|
|
There were no significant differences in plasma concentrations of
triglycerides, total cholesterol, and phospholipids between apob+/+ mice and
apob+/38.9,
apob48/48, and apob48/38.9 mice
(Table I). However, in the
apob38.9/38.9 mice, there were significant decreases
in plasma levels of triglycerides, cholesterol, and phospholipids
(Table I). On FPLC analysis, cholesterol peaks corresponding to VLDL
and LDL diminished in the plasmas of
apob+/38.9 and
apob38.9/38.9 mice (Fig.
3A), whereas significant
decreases in HDL cholesterol were observed only in the
apob38.9/38.9 plasma. Because VLDL and LDL
cholesterol accounts for only a very small fraction of plasma
cholesterol in mice, the reduction in VLDL and LDL cholesterol did not
lead to a significant decrease in plasma cholesterol in the
apob+/38.9 mice. On Western blot
analysis of the FPLC fractions of the
apob+/38.9 plasma, apo B-38.9 eluted
with HDL-sized particles (fractions 25-32) (Fig. 3B).
View larger version (39K):
[in this window]
[in a new window]
|
Fig. 3.
Distribution of cholesterol
(A) and the apo B moieties (B) in
FPLC-separated lipoprotein fractions of plasma from apo B-38.9
heterozygotes (Het.), homozygotes
(Homo.), and their wild type littermates
(WT). Mouse plasmas were fractionated by FPLC,
and aliquots of the FPLC fractions were used for cholesterol
determination and Western blot analysis. Distributions of apo B-100 and
apo B-48 in FPLC fractions were similar in plasmas of the heterozygous
and wild type mice, and distributions of apo B-38.9 in FPLC fractions
were also similar in plasmas of heterozygous and homozygous mice. Only
the Western blot of heterozygous plasma FPLC fractions is shown
(B).
|
|
Relative Levels of Apo B mRNA Expression by the Wild Type and
Mutant Alleles--
We examined the ratios of
apob+ to apob38.9 mRNA
transcripts by a primer extension assay. Fragments of mouse apo B
cDNA spanning exon 25 and the 5'-region of exon 26 that
included the apo B-38.9 mutation site (1.3 kb) were amplified
from mouse liver and intestinal RNA by RT-PCR (Fig.
4A) and used as templates for
primer extension reactions. Extension of templates from the tissues of
apob+/38.9 mice yielded two bands,
whereas templates from apob+/+ or
apob38.9/38.9 mice produced only single bands
(Fig. 4B). No extension of the labeled primer occurred when
the RT step was omitted in the RT-PCR (Fig. 4B),
demonstrating the specificity of the assay. Levels of
apob38.9 transcripts were 62 ± 3% of those of
the apob+ allele in the livers and intestines of
apo+/38.9 (n = 3) and
apob48/38.9 mice (n = 3).
View larger version (86K):
[in this window]
[in a new window]
|
Fig. 4.
Allele-specific analysis of relative levels
apo B mRNA expression using a primer extension assay.
A, generation of cDNA templates by RT-PCR. Mouse apo B
cDNA fragments corresponding to a region including exon 25 and the
5'-end of exon 26 were amplified from liver (Liv.) and
intestinal (Intest.) RNA. The specific PCR products
(indicated by an arrow) were gel-purified and used for
primer extension assay. B, primer extension of apo B
cDNA templates. Extension of apob38.9 cDNA
templates yielded a 34-bp product, whereas the
apob+ and apob48 cDNA
extension gave rise to a 42-bp product. The bands were quantified using
a PhosphorImager.
|
|
On Northern blot analysis, there were no differences in the sizes of
apob+ and apob38.9
transcripts (Fig. 5B).
Compared with apo B wild type littermates, apo B mRNA levels in the
liver of apob+/38.9 and
apob48/38.9 mice were reduced by approximately 22%
(Fig. 5A), whereas those of the
apob38.9/38.9 mice were reduced by 47%, confirming
the results obtained from the primer extension assay (Fig. 4). The
decrease in apob38.9 mRNA levels fully accounted
for the difference in total apo B mRNA levels between the
apob+/+ and
apob+/38.9 mice, indicating that apo
B-38.9 mutation does not affect the mRNA expression of the wild
type allele.
View larger version (61K):
[in this window]
[in a new window]
|
Fig. 5.
Northern blot analysis. Thirty
micrograms of total RNA were separated in 0.8% agarose-formaldehyde
gels. A mouse apo B cDNA fragment (~ 1.2 kb) corresponding to the
3'-end of exon 26 was labeled with 32P and used for
Northern hybridization. The hybridization signals were quantified using
a PhosphorImager. The ratios of apo B/glyceraldehyde-3-phosphate
dehydrogenase (GAPDH) mRNA are summarized in
A (mean ± S.D.; n = 4), and a typical
blot is shown in B. * and ** denote the significance of
differences (p < 0.05 and p < 0.01 (n = 4), respectively) compared with the values
of apob+/+.
|
|
Synthesis and Secretion of Apo B-48 and Apo B-38.9 by Cultured
Hepatocytes--
Next, we determined in cultured hepatocytes the
intracellular contents of apo B and the amounts of apo B secreted into
the cultured media using Western blot analysis. In the
apob+/+ hepatocytes, apo B-100 and apo B-48
accounted for 5 and 95% of the total cellular apo B, respectively
(Fig. 6A). In the
apob+/38.9 cells, apo B-48 levels were
reduced by 50% compared with the apob+/+ cells
(Fig. 6A). Furthermore, the intensity of the apo B-38.9 band
was approximately 53% of that of the apo B-48 in the
apob+/38.9 cells, whereas apo B-100 was
barely detectable (Fig. 6A). In apob38.9/38.9 cells, apo B-38.9 was present at
levels approximately 50% of those of apo B-48 in
apob+/+ cells (Fig. 6A). After a 3 h-incubation period in DMEM, conditioned media from the
apob+/38.9 hepatocytes contained 50%
less apo B-48 than the media from the apob+/+
cells (Fig. 6B). Apo B-100 bands in media were visible only
after a prolonged exposure time (not shown). Unexpectedly, equal
amounts of apo B-48 and apo B-38.9 accumulated in media of
apob+/38.9 hepatocytes (Fig.
6B), indicating that apob+/+ and
apob+/38.9 cells may have secreted
nearly equimolar amounts of apo-48 and apoB-48 plus apoB-38.9,
respectively. Indeed, despite the 2-fold difference in the cellular apo
B contents of apob+/+ and
apob38.9/38.9 cells, equal molar amounts of apo B
were secreted into the media (Fig. 6B).
View larger version (50K):
[in this window]
[in a new window]
|
Fig. 6.
Mass measurements of cellular apo B contents
and apo B secretion by apob+/+,
apob+/38.9, and
apob38.9/38.9 hepatocytes. Hepatocytes
were isolated and cultured for 7 h in 10% FBS/DMEM. Thereafter,
the cell monolayers were washed and cultured for 3 h in serum-free
DMEM. Relative apo B levels in the cells and in the conditioned media
were determined by Western analysis using rabbit anti-GST-apo B (amino
acids 26-289) antisera. Each column represents the
mean ± S.D. of the numbers of independent experiments
(n = 4 for apob+/+ and
apob+/38.9; n = 2 for
apob38.9/38.9). Representative Western blots are
shown in the bottom panels. Lanes 1-3, 250 µg
of protein of apob+/+,
apob+/38.9, and
apob38.9/38.9 hepatocyte lysates; lanes
4-6, media from apob+/+,
apob+/38.9, and
apob38.9/38.9 hepatocytes (corresponding to 600 µg
of cell protein).
|
|
During a continuous labeling period of 3 h, there were 1.6-2
times as much labeled apo B-48 as apo B-38.9 in the
apob+/38.9 cells (Fig.
7A), but similar amounts of
35S-labeled apo B-48 and apo B-38.9 were secreted into the
media (Fig. 7B), confirming the mass measurements (Fig.
6B). The total amounts of labeled apo B-48 and apo B-38.9
(i.e. secreted plus cellular) increased linearly within the
3 h-incubation period (Fig. 7C). Signals for the apo B-100
bands were barely detectable (data not shown).
View larger version (31K):
[in this window]
[in a new window]
|
Fig. 7.
Continuous metabolic labeling of apo B in
cultured apob+/38.9
hepatocytes. After 7 h in 10% FBS/DMEM following isolation,
hepatocytes (0.7 × 106) were labeled with
35S-Promix for 0, 45, 90, and 180 min in Met- and Cys-free
DMEM. Apo B proteins in the cell lysate or the medium were
immunoprecipitated and resolved on SDS-PAGE (3-12% gel) and
quantified using a PhosphorImager, and the intensities of the signal
were expressed as arbitrary units. Similar results were obtained from
two additional experiments. Signals for apo B-100 were barely
detectable and are not shown in the figure.
|
|
Pulse-chase experiments were then performed to determine the secretion
efficiency of the newly synthesized apo B-48 and apo B-38.9 from the
apob+/38.9 hepatocytes. Again, equal
amounts of the labeled apo B-48 and apo B-38.9 were secreted into the
media at the indicated chase time points despite the fact that cells
synthesized less apo B-38.9 than apo B-48. (Fig.
8, A and B). At the
end of the chase period, approximately 27 and 40% of the initially
labeled apo B-48 and apo B-38.9, respectively, were secreted into the
media (Fig. 8B). Apo B-38.9 also appeared to be secreted at
a faster rate than apo B-48 (Fig. 8B). Thus, both metabolic
labeling experiments confirmed the results of mass measurements that
the apo B-38.9 mutation does not reduce the secretion rates of the
resultant truncated apo B relative to apoB-48 from the same cells.
View larger version (33K):
[in this window]
[in a new window]
|
Fig. 8.
Pulse-chase analysis of apo B secretion by
cultured apob+/38.9
hepatocytes. After a 7-h culture period in 10% FBS/DMEM following
isolation, hepatocytes (0.7 × 106) were labeled with
35S-Promix for 30 min and chased for 0, 30, 60, and 120 min. Apo B proteins in the cell lysate or the medium were
immunoprecipitated and resolved by SDS-PAGE (3-12% gel) and
quantified using a PhosphorImager. Data are expressed as a percentage
of the intensities of cellular apo B signal obtained at 0 min of chase.
Similar results were obtained from two additional experiments. Signals
for apo B-100 were barely detectable and are not shown in the
figure.
|
|
All of the above experiments were also performed with
apob48/38.9 hepatocytes. The results were very
similar to those of the apob+/38.9 cells
(data not shown).
Effects of Apo B-38.9 Mutation on Hepatic Triglyceride
Secretion--
The secretion rates of newly synthesized triglycerides
decreased by 40% in the hepatocytes of
apob+/38.9 mice and by 70% in
hepatocytes of apob38.9/38.9 mice relative to the
apob+/+ cells, whereas more newly synthesized
triglycerides were accumulated by the
apob+/38.9 and
apob38.9/38.9 cells than by the
apob+/+ cells (Fig.
9). Likewise, in vivo studies
using Triton WR-1339 to block plasma triglyceride clearance showed that
triglyceride secretion rates were decreased by 30 and 70% in the
heterozygotes and homozygotes, respectively (Fig.
10).
View larger version (23K):
[in this window]
[in a new window]
|
Fig. 9.
Triglyceride synthesis and secretion by
cultured hepatocytes from apob+/+
(WT), apob+/38.9
(Het.), and apob38.9/38.9
(Homo.) mice. After 7 h in 10%
FBS/DMEM following isolation, hepatocytes (0.8 × 106)
were labeled with [2-3H]glycerol for 0, 30, 60, 90, and
120 min in 2 ml of DMEM. Lipids were extracted from the media, and
triglycerides were isolated by TLC. Each data point
represents mean ± S.D. of triplicate incubations.
|
|
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 10.
Plasma triglyceride accumulations due to
injection of Triton WR-1339. Littermates of
apob+/+ (WT),
apob+/38.9 (Het.), and
apob38.9/38.9 (Homo.) mice were fed a
fat-free, high carbohydrate diet for 8 h and injected with Triton
WR-1339. Blood samples were taken at the indicated time points. Plasma
triglyceride concentrations were determined. The values obtained within
30 s after injection were treated as baseline values and
subtracted from values obtained at later time points. Each data
point represents mean ± S.D. (n = 4 animals).
|
|
Fatty Livers Resulting from Apo B-38.9 Mutation--
Hepatic
triglyceride contents were increased 2.3- and 5-fold in the
heterozygotes and homozygotes, respectively (Fig.
11). No significant changes occurred in
hepatic phospholipids and total cholesterol contents (Fig. 11). Oil red
O staining of frozen liver sections confirmed the quantitative analysis
(Fig. 12). Differences in the sizes of
intracellular lipid droplets were also readily detected in freshly
isolated hepatocytes under the light contrast microscope (Fig. 12,
insets).
View larger version (36K):
[in this window]
[in a new window]
|
Fig. 11.
Hepatic lipid contents of
apob+/+ (WT),
apob+/38.9 (Het.), and
apob38.9/38.9 (Homo.)
mice. Lipids were extracted from livers and assayed for contents
of triglycerides, phospholipids, and total cholesterol. Lipid levels
are presented as mg/g of liver protein. Each column
represents mean ± S.D. (n = 6). ** denotes the
significance of differences (p < 0.01) compared with the
values of abob+/+.
|
|
View larger version (104K):
[in this window]
[in a new window]
|
Fig. 12.
Oil red O staining of frozen liver sections
from apob+/+,
apob+/38.9, and
apob38.9/38.9 mice. Oil red O staining was
performed as described under "Experimental Procedures."
Insets show the presence of large lipid droplets in the
hepatocytes of apob38.9/38.9 mice under a light
contrast microscope.
|
|
| |
DISCUSSION |
The FHBL associated with apo B-truncation-producing mutations is
the best characterized subset of FHBL (6, 7). However, although the
molecular bases of the genetic defects have been established, the
corresponding cellular and molecular mechanisms that produce the FHBL
phenotypes are relatively poorly understood. Previous studies using
apob-modified mice suggested that reduced mRNA levels of
the mutant apob allele and the presumed resulting low
production rates of apo B-containing lipoproteins might be primarily
responsible for the FHBL phenotypes (16-19). In the current study, we
have utilized a novel gene-targeting strategy to generate a mouse model
that closely resembles the natural apo B-38.9 mutation in humans. In
this FHBL model, levels of the apob38.9 mRNA
expression and intrahepatocytic synthetic rates of apo B-38.9 were
reduced by approximately 40%. However, we did not note corresponding decreases in the rates of apo B-38.9 secretion. In fact, apo B-38.9 was
secreted more efficiently than apo B-48. Thus, we demonstrated a
dissociation between low apo B-38.9 mRNA levels and low apo B-38.9
synthesis rates on the one hand and the high secretion efficiency of
apo B-38.9 on the other. Despite efficient apo B-38.9 secretion, rates
of hepatic triglyceride secretion were reduced. This indicates that apo
B-38.9 is defective for triglyceride transport, resulting in
development of fatty livers in the mice. These are new insights into
the structure-function relationship of apo B, as well as the FHBL syndrome.
The apo B-38.9 mouse model reported here is unique in that only a
single base pair deletion (nucleotide 5449) was introduced into the
mouse apob coding region and one copy of Lox-P sequence (43 bp) was inserted into the middle of intron 24. Therefore, unlike the
complex alterations introduced into apob in the previously reported FHBL mice (16-19), the apob modifications
occurring in our mouse model are relatively minor and are very close to
replicating a natural apo B-truncation-producing mutation in humans.
The apo B-38.9-bearing mice displayed characteristics typical of FHBL in humans, including low plasma levels of apo B-100 and apo B-38.9 and
LDL cholesterol. Plasma levels of HDL cholesterol were dramatically reduced in the homozygous mice. Similar observations have been made in
prenatal mice homozygous for apob null-knockout (29), adult
mice homozygous for the apo B-70-truncation (16), and mice bearing a
liver-specific knockout of the microsomal triglyceride transport
protein large subunit gene (30, 31). The decreased HDL levels in these
mice appeared to be mainly due to enhanced HDL catabolism because apo
AI synthesis was not affected (16, 29). We also observed no significant
changes in the liver contents of apo AI protein and apo AI secretion by
the hepatocytes due to the apo B-38.9 mutation (data not shown). Viable
apob38.9/38.9 mice were born and reached adulthood
in significant numbers, albeit fewer were born than expected. Apo B
plays an essential role in yolk sac nutrient transport for normal mouse
embryonic development (29, 32). The ability of
apob38.9/38.9 mice to survive indicates that apo
B-38.9 may be capable of performing this transporting role, at least in part.
One of the major issues we wanted to clarify was whether and to what
extent the apo B-38.9-specifying mutation affects the steady-state
level of apoB mRNA, given the previously reported low
apoB mRNA levels (~ 10-30% of normal) in the apoB
truncation-bearing mice (16, 18, 19). On allele-specific mRNA
analysis and Northern blotting, the apo B-38.9 mutation did not affect
the levels of mRNA expression of the wild type allele in
heterozygotes, but it decreased the levels of mRNA expression of
the apob38.9 allele by 40-50%. The magnitude of
this decrease is much smaller than those reported in other genetically
engineered FHBL mice (16, 18, 19), including the apo B-39-producing
mice, which bear an apob premature stop codon only 18 residues downstream from the apo B-38.9-COOH terminus (19). The mutant
apob81 allele was expressed at an even lower level
than the apob39 allele (19) although the apo B-81-
and apo B-39-bearing mice were produced using an identical
gene-targeting strategy (18, 19). In contrast with premature stop
codons in the nonphysiological apo B-truncation sites, placing a
nonsense mutation by in/out gene targeting at the apo B-editing site
(apo B-48) had no effect on apo B mRNA (23). In fact, premature
stop codons can be associated with either low (33-35) or unaltered
(36-38) mRNA expression for different genes. Thus, it is not clear
whether the differences in mutant apob mRNA levels
between the various mice are related to the different gene-targeting
strategies used, to the specific location of the premature stop codon,
or to the length of the open reading frame on the apob transcript.
Apo B-48 synthesis and secretion were reduced by 50%, as expected, in
apob+/38.9 hepatocytes compared with
apob+/+ hepatocytes, indicating that the apo
B-38.9 mutation does not interfere with the production rates of apo
B-48 from the normal allele. A surprising finding is that although the
apo B-38.9 mutation decreased apob38.9 mRNA
levels and the rates of intrahepatocytic apo B-38.9 synthesis by
40-50%, it did not cause a corresponding decrease in the cellular secretion of apo B-38.9 because apo B-38.9 was secreted more
efficiently than apoB-48 by hepatocytes. In fact, nearly equal molar
amounts of total apo B were secreted by the
apob+/+,
apob+/38.9, and
apob38.9/38.9 hepatocytes. Previous studies (39, 40)
using rat hepatoma cells overexpressing truncated apo B proteins have
suggested that COOH-terminal truncation of apo B-100 does not impair
the secretion of the resultant truncated apo B. The present study, for
the first time, provides data verifying these observations under
physiological settings. Unlike the situation in human livers, in which
the apo B-100 is the sole apob product, apo B-48 accounted
for more than 95% of the wild type apoB allele product from
hepatocytes of the apob+/+ and
apob+/38.9 mice, consistent with
previous observations (5, 30, 41). The relative rates of secretion of
apoB-100 and apoB 38.9 remain to be determined in
apob100/38.9 mice.
Triglyceride secretion was decreased in the hepatocytes of
apob+/38.9 mice and severely impaired in
apob38.9/38.9 mice, indicating that apo B-38.9,
although secreted in expected molar quantities, may not be as capable
as apo B-48, the physiological form of truncated apo B, for
transporting triglycerides. This finding is consistent with a recent
study of apo B-39-bearing mice that showed that the particle volume of
plasma apo B-39-VLDL was only about half that of the apo B-48-VLDL (19)
and with studies of rat hepatoma cells transfected with truncated human apo B cDNAs that demonstrated that the length of the apo B
truncation is positively correlated with the size but inversely
correlated with the buoyant density of the truncated apo B-containing
lipoprotein particle (42). Compatible observations have been made on
apo B-containing lipoproteins in human FHBL subjects (6, 22, 43);
e.g. human kinetic studies have shown that the majority of
apo B-43.7-containing lipoproteins enter the plasma as HDLs (43).
Several cases of fatty livers have been reported in heterozygous FHBL
human subjects (12-15). A previous study has shown that truncating apo
B at residue 1768 (i.e. apo B-38.95), only one residue
downstream from the apo B-38.9-COOH terminus, caused fatty liver in a
heterozygous FHBL patient (12). Furthermore, we have recently
demonstrated that human subjects heterozygous for apo B-4 or apo B-9
truncations have reduced hepatic VLDL-triglyceride secretion rates
(44). The results of the present study confirm and greatly extend the
findings in humans. Thus, selected mutations in the apo B gene could
cause a subset of fatty livers.
Although apo B-48 and apo B-38.9 were secreted at similar rates in
apob+/38.9 mice, the plasma
concentration of apo B-38.9 was only about half of that of apo B-48.
Likewise, the apo B-38.9 levels in plasmas of heterozygous FHBL human
subjects were much lower than those of apo B100 (22). Most of the human
(22) and mouse apo B-38.9 particles are of HDL particle sizes, whereas
the apo B-48 particles are much larger. Due to their smaller sizes, the
Lp B-38.9 particles may be distributed in the interstitial fluid, as
well as in plasma, thereby diluting the plasma pool of apo B-38.9.
Furthermore, apo B-38.9 may be cleared from plasma at a faster rate
than apo B-48. We have previously shown that human lipoproteins
containing short apo B truncations, including apo B-38.9, were cleared
from the plasma much more rapidly than the apo B-100-LDL particles
(45). We subsequently identified a renal proximal tubular endocytic receptor, megalin, as the key receptor involved in catabolism of apo
B-70.5-containing lipoproteins (46). The megalin binding site of apo B
appears to be in the NH2-terminal region within apo B-31
(47). Thus, megalin may also play an important role in rapid clearance
of apo B-38.9 particles from the plasma.
In conclusion, the results of this study strongly suggest that low
levels of mRNA expression or a defective secretion of apo B-38.9
may not be responsible for manifestation of the FHBL phenotypes and
fatty livers in our apo B-38.9-bearing mice. Rather, our study implies
that positioning a premature stop codon 386 residues upstream from the
apo B-48-COOH terminus in a mouse results in an apo B-38.9 with an
enhanced hepatic secretion efficiency but decreased triglyceride transporting capacity and fatty livers.
| |
ACKNOWLEDGEMENT |
We thank Dr. Timothy Ley for helpful
discussions and the ES cell core at Washington University School of
Medicine for the excellent assistance they provided for ES cell culture
and mouse blastocyst microinjection.
| |
FOOTNOTES |
*
This work was supported by National Institutes of Health
Grants R37 HL-424460 and RO1 HL-59515 and a grant from the Alan and Edith Wolf charitable fund.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: Division of
Atherosclerosis, Nutrition and Lipid Research, Dept. of Medicine, Washington University School of Medicine, Box 8046, 660 S. Euclid Ave.,
St. Louis, MO 63110. Tel.: 314-747-4352; Fax: 314-362-3513; E-mail:
zchen@im.wustl.edu.
Published, JBC Papers in Press, July 11, 2000, DOI 10.1074/jbc.M004913200
| |
ABBREVIATIONS |
The abbreviations used are:
apo, apolipoprotein;
VLDL, very low density lipoproteins;
LDL, low density
lipoproteins;
HDL, high density lipoproteins;
FHBL, familial
hypobetalipoproteinemia;
ES, embryonic stem;
bp, base pair(s);
kb, kilobase(s);
FPLC, fast performance liquid chromatography;
RT, reverse
transcription;
PCR, polymerase chain reaction;
PAGE, polyacrylamide gel
electrophoresis;
FBS, fetal bovine serum;
DMEM, Dulbecco's modified
Eagle's medium;
GST, glutathione S-transferase.
| |
REFERENCES |
| 1.
|
Chan, L.
(1992)
J. Biol. Chem.
267,
25621-25624
|
| 2.
|
Powell, L. M.,
Wallis, S. C.,
Pease, R. J.,
Edwards, Y. H.,
Knott, T. J.,
and Scott, J.
(1987)
Cell
50,
831-840
|
| 3.
|
Chen, S.-H.,
Habib, G.,
Yang, C. Y.,
Gu, Z. W.,
Lee, B. R.,
Weng, S. A.,
Silberman, S. R.,
Cai, S. J.,
Deslypere, J. P.,
Rossened, M.,
Gotto, A. M., Jr.,
Li, W. H.,
and Chan, L.
(1987)
Science
238,
363-366
|
| 4.
|
Greeve, J.,
Altkemper, I.,
Diesterich, J. H.,
Greten, H.,
and Windler, E.
(1993)
J. Lipid Res.
34,
1367-1383
|
| 5.
|
Higuchi, K.,
Kitagawa, K.,
Kogishi, K.,
and Takeda, T.
(1992)
J. Lipid Res.
33,
1753-1764
|
| 6.
|
Schonfeld, G.
(1995)
Annu. Rev. Nutr.
15,
23-34
|
| 7.
|
Linton, M. F.,
Farese, R. V., Jr.,
and Young, S. G.
(1993)
J. Lipid Res.
34,
521-541
|
| 8.
|
Wu, J.,
Kim, J.,
Li, Q.,
Kwok, P.-Y.,
Cole, T. G.,
Cefalu, B.,
Averna, M.,
and Schonfeld, G.
(1999)
J. Lipid Res.
40,
955-959
|
| 9.
|
Parhofer, K. G.,
Barrett, P. H.,
Bier, D. M.,
and Schonfeld, G.
(1992)
J. Clin. Invest.
89,
1931-1937
|
| 10.
|
Krul, E. S.,
Parhofer, K. G.,
Barrett, P. H.,
Wagner, R. D.,
and Schonfeld, G.
(1992)
J. Lipid Res.
33,
1037-1050
|
| 11.
|
Parhofer, K. G.,
Barrett, P. H.,
Aguilar-Salinas, C. A.,
and Schonfeld, G.
(1996)
J. Lipid Res.
37,
844-852
|
| 12.
|
Tarugi, P.,
Lona | |
TOP
ABSTRACT
INTRODUCTION