© 1995 American Heart Association, Inc.
Articles |
Inhibition of Protein Tyrosine Kinase Alters the Effect of Serum Basic Protein I on Triacylglycerols and Cholesterol Differently in Normal and HyperapoB Fibroblasts
From the Lipid Research Atherosclerosis Unit, Departments of Pediatrics (M.M.) and Medicine (P.O.K., M.M.), Johns Hopkins University School of Medicine, Baltimore, Md.
Correspondence to Peter Kwiterovich, Johns Hopkins Hospital, CMSC 604, 600 N Wolfe St, Baltimore, MD 21287-3654.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Abstract We studied whether the stimulatory effect of human serum basic protein I (BP I) on the formation of cell triacylglycerols and cholesterol may be mediated through protein tyrosine kinase in normal fibroblasts, and whether there was a deficiency in such a process in cells from subjects with hyperapobetalipoproteinemia (hyperapoB). Genistein, a highly specific inhibitor of tyrosine kinase phosphorylation, was used as a probe. When BP I (428.0 nmol/L) alone was added to F-12 medium without genistein, the mean mass of cell triacylglycerols doubled in six normal cell lines from healthy subjects, an effect that was decreased by 50% in six cell lines from subjects with hyperapoB (P=.007). The addition of genistein with BP I to normal cells decreased the stimulation of triacylglycerol formation by BP I by about 50% (P=.008), whereas genistein had little effect in the BP I–treated hyperapoB cells. The effect of genistein on the stimulation of triglyceride and cholesterol production by BP I was shown to be both time and concentration (92.5 nmol/mL medium nadir) dependent. In normal fibroblasts, BP I stimulated the rate of incorporation of both [14C]acetate (P=.0001) and [3H]mevalonolactone (P=.002) into unesterified cholesterol, an effect that was markedly deficient in the hyperapoB cells (P=.0001 for [14C]acetate and P=.0002 for [3H]mevalonolactone). In normal but not hyperapoB cells, genistein inhibited the significant stimulation by BP I of the rates of both [14C]acetate (P=.0001) and [3H]mevalonolactone (P=.04) incorporation into unesterified cholesterol. There was also a significantly greater stimulation by BP I of the rate of [14C]acetate incorporation into cell esterified cholesterol in normal cells than in hyperapoB cells (P=.003), an effect that was inhibited by genistein in both normal (P=.0009) and hyperapoB (P=.01) cells. BP I also stimulated to a greater extent the mass of total cholesterol (P=.0009) and unesterified cholesterol (P=.015), but to a lesser degree that of esterified cholesterol (P=.44), in normal cells than in hyperapoB cells. Herbimycin A and tyrphostin A47, two other inhibitors of protein tyrosine kinase, also significantly inhibited the effects of BP I on triacylglycerol and cholesterol mass in normal cells but not in hyperapoB cells. The effect of BP I on triacylglycerols and cholesterol formation in normal cells appeared to be mediated through a tyrosine kinase–dependent process that was deficient in hyperapoB cells.
Key Words: coronary artery disease • apolipoprotein B • second messengers • familial combined hyperlipidemia • acylation stimulatory protein
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
HyperapoB is a lipoprotein disorder prevalent in patients with premature coronary artery disease and is characterized by an increased number of small, dense, LDL particles.1 2 HyperapoB shares characteristics with familial combined hyperlipidemia, LDL subclass pattern B, familial dyslipidemic hypertension, and syndrome X.3 Two metabolic defects have been described in patients with hyperapoB. First, there is overproduction of apo B and VLDL particles in the liver, leading to an increased synthesis of small, dense LDL particles.3 4 Second, there is an abnormal clearance of postprandial triglyceride-rich particles and free fatty acids.5 6
We7 8 and others9 10 have previously studied the role of a low–molecular weight basic protein isolated from normal human serum in the pathogenesis of hyperapoB. We termed this protein BP I. It has an apparent molecular weight of 14 000 and an isoelectric point of 9.10.7 On Western immunoblot assays, BP I reacted with an antibody to ASP but not to antisera to two other basic proteins, sterol carrier protein–2 and protein 422, a fatty acid binding protein from adipose tissue.7
BP I normally stimulates the production of cellular triacylglycerols in fibroblasts and adipocytes, but such stimulatory activity is deficient in cells from patients with hyperapoB.7 8 9 10 11 12 BP I also stimulates the formation of esterified cholesterol in normal cells, an effect that was likewise significantly deficient in hyperapoB cells.8 The changes in the mass of cholesteryl esters were paralleled by similar changes in the mass of total but not unesterified cholesterol, suggesting that the pool of cellular free cholesterol was not depleted and raising the possibility that BP I may also affect the synthesis of cellular cholesterol.
Cianflone et al10 found that the deficiency in acylation stimulatory activity with ASP is accompanied by a decrease in the high-affinity binding of 125I-labeled ASP to fibroblasts from subjects with hyperapoB. We reported that inhibition of protein kinase C activity significantly decreased the effect of BP I in normal and hyperapoB cells,8 whereas the activation of protein kinase C stimulated triacylglycerol formation.8 BP I further increased triacylglycerol production under these conditions in the diacylglycerol–protein kinase C pathway, but at an earlier step, perhaps at the cell surface.
Genistein, an isoflavone antibiotic, is a broad but highly specific
competitive inhibitor of ATP in the tyrosine kinase
reaction.13 14 15 The ATP binding domain is highly conserved
among tyrosine kinases. Because of its broad spectrum, we first used
genistein as a probe to determine whether the effects of BP I on
triacylglycerol and cholesterol metabolism in normal cells may be
mediated through tyrosine phosphorylation and whether there is a
deficiency in such a process in cells from patients with hyperapoB. Two
other inhibitors of protein tyrosine kinase were also assessed.
Herbimycin A, a natural inhibitor, irreversibly blocks and enhances the
degradation of the epidermal growth factor receptor and, for example,
the second messenger tyrosine kinases Src and phospholipase C
-1.13 16 Tyrphostin A47 (3,4-dihydroxybenzylidine
thiocyanoacetamide), a synthetic inhibitor, blocks the substrate domain
of protein tyrosine kinases and exhibits a range of
selectivity.13 17 18
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Human Materials
Six subjects with hyperapoB, from six unrelated kindreds with familial hyperapoB, and seven unrelated normal subjects were studied. These subjects have been characterized previously.7 8 After informed consent was obtained from the subjects, fibroblasts were grown from skin biopsies taken from the forearm as described.7 8 Cells were used between passages 5 to 10. BP I was isolated and purified from healthy volunteers as previously described.7 8
Protocol for Cell Experiments
Fibroblasts (105) were seeded and grown in minimal
essential medium containing 10% (vol/vol) fetal calf serum, 1% amino
acids, 100 U penicillin/mL, and 100 mg streptomycin/mL for 6 days. The
medium was then changed to a supplemented serum-free medium for 24
hours, as described previously.7 At zero time,
oleate:albumin (4.6:1, 10 nmol/L oleate) was added to the
medium without BP I (control cells) or to medium to which 6 µg/mL BP
I (428.6 nmol/L) had been added, either in the presence or absence of
the indicated concentration of genistein (0 to 100 µg/mL medium; 0 to
370 nmol/mL) (Upstate Biotechnology Inc, catalog No. 19110). This
concentration of BP I was that previously found to exert its maximal
effect on lipid synthesis in fibroblasts.7 The cells were
incubated for 6 hours, then the medium was removed, the cells were
washed, and the lipids were extracted as described.7 8 The
mass of individual cell lipids or the incorporation of radioactivity
into unesterified or esterified cholesterol was measured
(see below). Sodium hydroxide was added to the cell residue and dried,
and the cell protein content was determined by the method of Lowry et
al.19 Duplicate dishes of cells were used for each
experimental condition. In all cell experiments with BP I, the values
from the control cells grown in the absence of BP I (with and without
genistein) were substracted from values from the cells grown in the
presence of BP I (with and without genistein).
In one experiment, the effect of genistein (92.5 nmol/mL) was compared to the effects of two other inhibitors of protein tyrosine kinase, herbimycin A (200 and 400 nmol/mL)16 and tyrphostin A47 (25 and 50 nmol/mL)19 (Calbiochem). The concentrations used were those previously found to inhibit protein tyrosine kinase second messengers.16 18
Endogenous Synthesis of Cellular
Cholesterol
The incorporation of sodium [1-14C]acetate or
mevalonolactone-5-3H into cell unesterified
cholesterol and esterified cholesterol was
studied by use of the method of Goldstein and Brown,20
modified as follows. Supplemented serum-free medium was added to
confluent fibroblasts. After 6 hours, 10 µL
14C[acetate]/mL medium (100 nmol/L, 0.1 µCi/µL) and
10 µL [3H]mevalonolactone/mL medium (33 nmol/L, 0.1
µCi/µL) were added to the medium and the incubation continued for
an additional 18 hours. At that time, oleate:albumin (4.6:1, 10
nmol/L oleate) was added to the medium without BP I (with or without
92.5 nmol/mL genistein [25 µg/mL]) (control cells) or to a medium
to which 6 µg of BP I was added, with or without 92.5 nmol/mL
genistein, and the incubation continued for an additional 6 hours. The
medium was removed, the cells were washed, and the lipids were
extracted as described.7 8 The extract was dried and then
taken up in hexane. Lipid carriers were added to the extract, which was
then subjected to thin-layer
chromatography16 ;
[3H]cholesteryl esters were used to correct for recovery.
The spots corresponding to unesterified cholesterol and
esterified cholesterol were scraped and the radioactivity
was determined by scintillation counting. The data are expressed as
nanomoles of [14C]acetate or
[3H]mevalonolactone incorporated per milligram of cell
protein.
Mass Measurements of Lipids in Fibroblasts
The masses of total cholesterol, unesterified
cholesterol, and esterified cholesterol were
determined by gas-liquid chromatography as described
previously.8 Unesterified cholesterol was
measured directly and total cholesterol was measured after
saponification. The mass of cholesterol that was esterified
was calculated by subtraction of the mass of unesterified
cholesterol from the mass of total cholesterol.
The mass of triacylglycerols was determined
enzymatically by use of a commercially available kit (Seradyn
Triacylglycerols Procedure).8 The data
are expressed as nanomoles of lipid per milligram of cell
protein.
Statistical Analysis
Tests of significance within a group were performed with
Student's paired t test, and the tests between the normal
and hyperapoB cells were performed with the two-sample t
test.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Effect of Genistein Alone on the Mass of Cellular Lipids in Normal and HyperapoB Fibroblasts
There were no significant differences between the masses of triacylglycerols, total cholesterol, unesterified cholesterol, and esterified cholesterol in six normal and six hyperapoB cell lines after the cells were grown in F-12 medium that contained oleate:albumin but not genistein or BP I (control condition) (Table 1
). When genistein (92.5 nmol/mL medium) was
added to the control medium in the absence of BP I, the masses of the
total cholesterol and unesterified cholesterol
decreased significantly in both the normal and hyperapoB cells; the
mass of triacylglycerols also decreased
significantly in the hyperapoB cells but not in the normal cells (Table 1). When the masses of these lipids in the normal and hyperapoB cells
were compared after genistein treatment, the decrease in both total
cholesterol and unesterified cholesterol was
significantly greater in the hyperapoB cells than in the normal cells
(Table 1). No differences were found in cells for the content of
esterified cholesterol (Table 1).
|
Time Course of the Effect of Genistein on the Formation of Cellular
Triacylglycerols by BP I
We determined the effect of genistein (92.5 nmol/mL) over time (0
to 12 hours) on the stimulation by BP I of the mass of
triacylglycerols in normal and hyperapoB
fibroblasts. The effect of BP I on the mass of cell
triacylglycerols without genistein was time
dependent, and plateaued at 6 hours in normal cells (Fig 1, top). When genistein was added to BP I in normal
cells, the time curve had a pattern similar to that without genistein,
but the mass of triacylglycerols was about 50%
lower at each time point. BP I with or without genistein had little
effect on the mass of triacylglycerols with time in
the hyperapoB cells (Fig 1, top). The mass of
triacylglycerols in the hyperapoB cells with BP I
alone was lower than that observed in the dose-response curve (Fig 1,
bottom) or in a subsequent experiment (Fig 2), perhaps
because of experimental variability.
|
|
Dose-Response of the Effect of Genistein on the Formation of
Cellular Triacylglycerols by BP I
Increasing amounts of genistein (0 to 370 nmol/mL medium) were
added to normal and hyperapoB cells with or without BP I and incubated
for 6 hours (see "Methods"). Without BP I, increases in the
amount of genistein in the medium had little effect on the mass of
triacylglycerols in cells incubated in F-12 medium
alone (data not shown). After incubation with BP I alone, the mass of
triacylglycerols in the normal cells was about
twice that in the hyperapoB cells (Fig 1
, bottom). When genistein was
added with BP I, the mass of cell triacylglycerols
in the normal cells decreased in a concentration-dependent fashion, up
to a level of 92.5 nmol genistein/mL medium, and leveled off thereafter
(Fig 1, bottom). The mass of triacylglycerols in
the hyperapoB cells increased somewhat at low levels of genistein, but
then fell at concentrations of 92.5 nmol genistein/mL medium and
plateaued thereafter (Fig 1, bottom).
The mass of triglyceride in the hyperapoB cells after
genistein treatment (92.5 nmol/mL) was somewhat lower (7 to 10 nmol
triglyceride/mg cell protein) than that in the normal cells
(Fig 1, top and bottom).
Dose-Response of the Effect of Genistein on the Formation of
Cellular Total Cholesterol, Unesterified
Cholesterol, and Esterified Cholesterol by
BP I
With no genistein in the medium, BP I stimulated the masses of
total cholesterol, unesterified cholesterol,
and esterified cholesterol in normal cells, an effect that
was inhibited by genistein in a dose-dependent fashion and that
plateaued at a level of 92.5 nmol genistein/mL medium (Fig 3). In hyperapoB cells, little stimulation of cell
cholesterol with BP I alone was observed, and genistein had
no inhibitory effect. The effects of BP I on the masses of
total, unesterified, and esterified cholesterol in normal
cells and the inhibition of genistein on such stimulation were also
time dependent, and plateaued at 6 hours (data not shown).
|
After the concentration and time dependence of the effect of genistein were determined, the significance of genistein's effect on the stimulation of lipid mass by BP I was assessed further (see below) by use of a larger number of normal and hyperapoB cell lines, a concentration of genistein of 92.5 nmol/mL medium, and a 6-hour period of incubation with BP I.
Effect of Genistein on the Stimulation of the Mass of
Triacylglycerols by BP I in Normal and
HyperapoB Fibroblasts
When BP I (428.0 nmol/L) was added to F-12 medium without
genistein, the average mass of triacylglycerols
doubled in six normal cell lines, but the acylation stimulatory
activity of BP I in six hyperapoB cells was significantly decreased
(P=.007) (Fig 2). The addition of genistein with BP I to the
normal cells decreased the stimulation of
triacylglycerol formation by BP I by about 50%
(P=.008) (Fig 2), whereas genistein had little effect on the
stimulated triacylglycerol mass in the BP
I–treated hyperapoB cells (P=.62) (Fig 2). The mean mass of
triacylglycerols in normal cells incubated with
genistein and BP I was approximately equal to that in hyperapoB
fibroblasts treated with BP I alone (Fig 2). Thus, genistein
essentially obliterated the significant differences between the
responses of the normal and hyperapoB fibroblasts to BP I.
Effect of BP I With and Without Genistein on
Cholesterol Synthesis and Mass in Normal and
HyperapoB Fibroblasts
The effects of BP I, with and without genistein, on the rates of
[14C]acetate and [3H]mevalonolactone
incorporation into cell unesterified cholesterol and
esterified cholesterol were studied in four normal and four
hyperapoB cell lines. The masses of cell total cholesterol,
unesterified cholesterol, and esterified
cholesterol were measured concomitantly.
[14C]Acetate Incorporation Into Cell Unesterified
Cholesterol in F-12 Medium With and Without BP I
After 24 hours in F-12 lipid-deficient medium, there was no
significant difference in the rate of incorporation of
[14C]acetate into unesterified cholesterol
between the normal cells (9.7±1.9 nmol · mg cell
protein-1 · h-1 [mean±SEM]) and the
hyperapoB cells (7.8±1.4 nmol · mg cell
protein-1 · h-1,
P=.45). Addition of BP I to the F-12 medium produced a
2.34-fold increase in [14C]acetate incorporation into
unesterified cholesterol in the normal fibroblasts
(22.5±1.3 nmol · mg cell
protein-1 · h-1,
P=.0001), but no stimulation was observed in the hyperapoB
cells (7.4±1.1 nmol · mg cell
protein-1 · h-1,
P=.458). The effect of BP I on the stimulation of
[14C]acetate incorporation into unesterified
cholesterol in the hyperapoB cells was therefore markedly
deficient (P=.0001, Fig 4). The addition of
genistein to BP I–containing medium significantly reduced the
stimulation of [14C]acetate incorporation in the normal
cells (P=.0001) but not in the hyperapoB cells
(P=.06) (Fig 4).
|
[14C]Acetate Incorporation Into Cell Esterified
Cholesterol in F-12 Medium With and Without BP I
After 24 hours in F-12 lipid-deficient medium, there was no
significant difference in the rate of incorporation of
[14C]acetate into esterified cholesterol
between the normal cells (0.30±0.05 nmol · mg cell
protein-1 · h-1) and the hyperapoB cells
(0.44±0.12 nmol · mg cell
protein-1 · h-1,
P=.32). Addition of BP I to F-12 medium produced a 10.8-fold
increase in [14C]acetate incorporation into esterified
cholesterol in the normal cells (3.25±0.51 nmol · mg
cell protein-1 · h-1,
P=.0004), and a 3.31-fold increase (1.46±0.36 nmol · mg
cell protein-1 · h-1,
P=.005) in the hyperapoB cells. The effect of BP I on the
stimulation of [14C]acetate incorporation into cell
esterified cholesterol in the hyperapoB fibroblasts was
therefore significantly deficient (P=.003) (Fig 4). The
addition of genistein to BP I–containing medium significantly
inhibited the stimulatory effect of BP I in both the normal
(P=.0009) and the hyperapoB (P=.01) cells (Fig 4).
[3H]Mevalonolactone Incorporation Into Cell
Unesterified Cholesterol in F-12 Medium With and
Without BP I
After 24 hours in F-12 lipid-free medium, the rate of
incorporation of [3H]mevalonolactone into unesterified
cholesterol was higher in the hyperapoB cells (2.8±0.20
nmol · mg cell protein-1 · h-1) than
in the normal cells (0.3±0.09 nmol · mg cell
protein-1 · h-1,
P=.0001). Upon addition of BP I, the incorporation of
[3H]mevalonolactone into unesterified
cholesterol in normal cells increased 17-fold (to
4.58±0.37 nmol · mg cell
protein-1 · h-1,
P=.002), whereas this effect of BP I in hyperapoB cells was
markedly less (1.1-fold, P=.28). Thus, the response in the
hyperapoB fibroblasts to the stimulation of
[3H]mevalonolactone incorporation into unesterified
cholesterol was significantly deficient
(P=.0002) (Fig 4). Genistein significantly inhibited the
stimulation of [3H]mevalonolactone incorporation into
unesterified cholesterol by BP I in the normal cells
(P=.04) (Fig 4), but actually stimulated such incorporation
in the hyperapoB cells (to 2.82±0.40 nmol · mg cell
protein-1 · h-1,
P=.016).
[3H]Mevalonolactone Incorporation Into Cell
Esterified Cholesterol in F-12 Medium With and Without BP
I
BP I did not stimulate [3H]mevalonolactone into cell
esterified cholesterol in the normal cells (Fig 4). Only a
small stimulation occurred in the hyperapoB cells, and this effect of
BP I was increased to a small extent upon the addition of genistein
(P=.27) (Fig 4).
Effect of BP I With and Without Genistein on the Mass of
Cholesterol in Normal and HyperapoB Cells
BP I stimulated the masses of total cell cholesterol
(P=.0009) and unesterified cholesterol
(P=.015) to a significantly greater extent in normal cells
than in hyperapoB cells (Fig 5). The mass of esterified
cholesterol in BP I–treated cells was 1.7-fold higher in
the normal cells than in the hyperapoB cells (P=.438). This
stimulation of cell cholesterol in the normal cells was
inhibited by genistein for total cholesterol
(P=.0002), unesterified cholesterol
(P=.003), and esterified cholesterol
(P=.077) (Fig 5). Genistein also inhibited the (albeit
decreased) stimulation by BP I of the mass of total
cholesterol (P=.002), unesterified
cholesterol (P=.029), and esterified
cholesterol (P=.005) in the hyperapoB cells (Fig 5).
|
Comparison of the Effects of Genistein, Herbimycin A, and
Tyrphostin A47 on the Stimulation of Cellular Masses of
Triacylglycerols and Cholesterol in
Normal and HyperapoB Cells Treated With BP I
There were no significant differences for the mean mass of
triacylglycerols between normal and hyperapoB cells
for any of the control conditions (F-12 alone or F-12 with genistein,
herbimycin A, or tyrphostin A47) (Table 2). After
treatment with BP I alone (uncorrected cell data), there was a
threefold increase in the mean mass of
triacylglycerols in the normal cells, whereas there
was a highly significant deficiency in the response of the hyperapoB
cells to BP I alone (Table 2). This significant difference between
normal and hyperapoB cells in terms of the stimulatory activity of BP I
was obliterated when genistein, herbimycin A, or tyrphostin A47 was
added to F-12 medium containing BP I (uncorrected cell data) (Table 2).
Similar results were obtained when the mean triglyceride
data were corrected for the control condition (Table 2).
|
In the normal fibroblasts, when genistein (P=.001),
herbimycin A (P=.001), or tyrphostin A47 (P=.005)
was added to BP I, there was a highly significant fall in the
stimulatory activity of BP I. In contrast, when any one of the three
inhibitors was added to BP I in the hyperapoB cells, there
was no significant decrease in the stimulatory activity of BP I for
genistein (P=.62), herbimycin A (P=.97), or
tyrphostin A47 (P=.40) (uncorrected cell data) (Table 2).
Genistein was found in this experiment to be the most potent
inhibitor of the effect of BP I on cell
triglyceride mass, and there was no significant difference
between the control condition with genistein and the condition with BP
I plus genistein for normal cells (P=.91) or for hyperapoB
cells (P=.31). Although herbimycin A significantly depressed
the effect of BP I on the mean mass of triglycerides in
normal cells (P=.001), there was still some residual
acylation stimulatory activity compared with the control condition
alone in both normal cells (P=.05) and hyperapoB cells
(P=.03) (Table 2). Also, tyrphostin A47 significantly
depressed the effect of BP I in normal fibroblasts (P=.005),
but there was still some stimulatory activity of BP I above that found
in the control condition in both the normal cells (P=.004)
and, to a lesser extent, the hyperapoB cells (P=.03) (Table 2).
Genistein again significantly decreased the mass of total
cholesterol when added to F-12 control medium in both the
normal (P=.007) and hyperapoB (P=.002)
fibroblasts (Table 2). In a similar fashion, herbimycin A
(P=.007) and tyrphostin A47 (P=.006) suppressed
the mass of total cholesterol in F-12 control medium alone
in normal fibroblasts. Again, an even greater decrease in the total
mass of cholesterol in hyperapoB fibroblasts in F-12 medium
alone was observed with both herbimycin A (P=.0005) and
tyrphostin A47 (P=.002). The decrease in the mass of total
cholesterol for each of the three inhibitors of
protein tyrosine kinase was significantly more pronounced in the
hyperapoB fibroblasts than in normal fibroblasts (Table 2).
When BP I was added to F-12 medium alone, in the normal cells there was a highly significant 1.7-fold increase in the mass of total cholesterol (P=.005). This stimulatory effect on the mass of cholesterol with BP I in normal cells was significantly inhibited with genistein (P=.05), herbimycin A (P=.01), and tyrphostin A47 (P=.009). In the hyperapoB cells, in the presence of F-12 medium with BP I, there was no significant stimulation of mass of total cholesterol (P=.30).
When the normal and hyperapoB fibroblasts were compared, there was
significantly higher mass of total cholesterol in the
normal fibroblasts for all four experimental conditions according to
either the uncorrected or corrected cell data (Table 2).
The results for the cellular mass of
triacylglycerols and cholesterol were
similar to those reported in Table 2 when the concentrations of
herbimycin A and tyrphostin A47 used were halved (data not shown).
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Genistein, an isoflavone antibiotic, is a competitive inhibitor of ATP binding to the catalytic domain of tyrosine kinase, a highly conserved structural feature among all tyrosine kinases.21 In the present study, we first used genistein as a broad probe to determine whether the inhibition of tyrosine kinase phosphorylation in cultured normal human and hyperapoB fibroblasts affected the biochemical effects of BP I. We found significant differences between the effects of genistein in normal and hyperapoB fibroblasts, strongly suggesting that tyrosine kinase phosphorylation has a role in the pathogenesis of hyperapoB. This tenet was supported by the observation that two other inhibitors of protein tyrosine kinase, herbimycin A and tyrphostin A47, that had different mechanisms of action had effects similar to those of genistein in both the normal and hyperapoB cells. Our results, however, do not permit us to determine whether the observed effect(s) of these inhibitors involve a transmembrane tyrosine kinase receptor or a membrane-associated tyrosine kinase molecule,22 23 or whether the effect occurs at the postreceptor level24 (see also below). We also did not determine whether the physiological effects of BP I take place at a point before, directly at, or after genistein inhibition. The fact that the effect of genistein on the biochemical actions of BP I was both time and concentration dependent, however, suggests that a high-affinity mechanism of genistein on the effects of BP I may be involved.24 Finally, any generalized inhibition of a number of tyrosine kinases by genistein was taken into account, in part, by the control condition, in which the effect of genistein alone on cellular lipids was studied. The effect of genistein alone was corrected for, and it was therefore against this background that the effect of genistein on the stimulatory biochemical actions of BP I in normal and hyperapoB cells was studied.
In agreement with the results of our previous work,8 there
were no differences between normal and hyperapoB fibroblasts in the
masses of triacylglycerols, total
cholesterol, unesterified cholesterol, and
esterified cholesterol in F-12 medium containing
oleate:albumin without BP I (control condition). The addition
of genistein without BP I produced a small decrease (about 10%) in the
mass of triacylglycerols in normal cells and a
somewhat larger decrease (about 25%) in the hyperapoB cells (Table 1).
It is therefore unlikely that the major effect of genistein on
triacylglycerol metabolism was mediated
by perturbation of the uptake of free fatty acids from albumin
and the subsequent incorporation of free fatty acids into
triacylglycerols. In both the normal and hyperapoB
cells, the magnitude of the inhibitory effect of genistein
on the mass of total cholesterol and unesterified
cholesterol in the absence of basic proteins was greater
than that observed for triacylglycerols. This
effect was significantly greater in the hyperapoB cells, suggesting
that these cells may have a defect that renders them more susceptible
to the inhibitory effect of genistein.
We confirmed our previous observations that BP I alone essentially doubled the mass of cell triacylglycerols by 6 hours in normal cells but that there was a 50% deficiency in such acylation stimulatory activity in hyperapoB cells.8 Genistein, at a maximum inhibitory concentration of 25 µg/mL medium (92.5 nmol/mL), significantly decreased this effect of BP I in normal cells, suggesting that a significant part of the biochemical effect of BP I is mediated through a pathway requiring tyrosine kinase phosphorylation. It was also noteworthy that the addition of genistein to BP I in hyperapoB cells had little effect on the mass of triacylglycerols. Because the mass of triacylglycerols produced by normal cells by means of stimulation with BP I in the presence of genistein (as well as herbimycin A and tyrphostin A47) was very similar to that found in the hyperapoB cells in the absence of genistein, the acylation stimulatory deficiency with BP I in hyperapoB cells appears to involve a pathway that is blocked by these protein tyrosine kinase inhibitors. The basis for the residual effect of BP I stimulation of triacylglycerols in genistein-treated normal and hyperapoB fibroblasts is not known.
We had previously found that BP I stimulated the formation of cell esterified cholesterol in normal fibroblasts, and that this effect was decreased in hyperapoB fibroblasts.8 However, the pool of cell unesterified cholesterol was not depleted, suggesting that these may be significant differences in cholesterol biosynthesis in normal and hyperapoB cells treated with BP I. These observations prompted us to examine cholesterol production. Data from the present study indicate, for the first time, that BP I stimulates the production of cholesterol in normal fibroblasts. The inhibition of this effect by genistein was probably not mediated through competitive inhibition of HMG CoA reductase, the rate-limiting enzyme of cholesterol biosynthesis, because the incorporation of [3H]mevalonolactone into unesterified cholesterol was also decreased by genistein. The decrease in cholesterol production by genistein is also unlikely to be due to inhibition of phosphorylation of HMG CoA reductase. Sato and coworkers25 have shown that phosphorylation of HMG CoA reductase lowers its catalytic activity. Thus, inhibition of phosphorylation of HMG CoA reductase by genistein might be expected to increase, not decrease, cholesterol biosynthesis. BP I may conceivably increase the transcription of the HMG CoA reductase gene, stabilize its steady state mRNA level, or enhance its translation. These possible effects of BP I may be mediated through tyrosine kinase phosphorylation, because genistein inhibits the normal stimulation of cholesterol production by BP I. Finally, the data further indicate that the hyperapoB cells have a defect in the stimulatory effect of BP I on cholesterol production.
We had previously found that inhibitors of protein kinase C
obliterated the effects of BP I in cultured human fibroblasts but that
there were no differences between the normal and hyperapoB
cells.8 Furthermore, C-8, a analogue of diacylglycerol,
stimulated the formation of triacylglycerol about
fivefold in both normal and hyperapoB cells, suggesting that there was
no defect at that point in the diacylglycerol/protein kinase C pathway
in hyperapoB cells. Nonreceptor tyrosine protein kinases, which contain
SH2 and SH3 domains, become physically
associated with, and phosphorylated by, an
activated transmembrane receptor protein tyrosine kinase.
Included in these SH2 and SH3 proteins is one
of several isoforms of phospholipase C, phospholipase C -1, which
cleaves the phospholipid phosphatidylinositol-4,5-diphosphate into the
second messengers diacylglycerol and inositol triphosphate, which
in turn stimulate protein kinase C and mobilize
Ca2+, respectively.22 These
SH2 and SH3 proteins also possess tyrosine
protein kinase activity, serving to amplify the tyrosine kinase signal.
Herbimycin A has been shown to directly inhibit phospholipase C
-1.16 Thus, the defect in hyperapoB cells may reside in
a transmembrane receptor protein tyrosine kinase or in a prototype
cytoplasmic signaling protein such as phospholipase C -1. Finally,
as with insulin (and other hormones), one cannot conclude that all
responses to BP I depend on receptor- or cell surface–associated
tyrosine kinase activity alone. Indeed, genistein itself has been shown
in rat adipocytes to inhibit differentially postreceptor effects of
insulin without inhibiting the insulin receptor tyrosine
kinase.23
Accelerated premature atherosclerosis is often found in patients with hyperapoB. An important component of atherosclerosis is uncontrolled cell growth in the vessel wall. Because protein tyrosine kinases are involved in the control of cell growth, it is possible that a defect in such a pathway in hyperapoB cells might contribute to increased cell proliferation in the arterial wall. The synthetic inhibitor used here, tyrphostin A-47 (also referred to as RG50864 and AG213) inhibits the epidermal growth factor receptor 800 times as potently as it inhibits the insulin receptor.13 17 AG213 also inhibits the effect of the proliferative signals of both basic fibroblast growth factor and platelet-derived growth factor in rabbit vascular smooth muscle cells and human bone marrow fibroblasts.13
The combined use of inhibitors of protein tyrosine kinase and BP I in cultured human fibroblasts appears to further distinguish patients with hyperapoB from healthy control subjects. Taken together, these data strongly indicate that a second messenger–, tyrosine kinase–dependent process is involved in the pathogenesis of hyperapoB. Future studies will be directed toward improving the understanding of the molecular basis of the disorder and its role in premature coronary artery disease.
|
Acknowledgments |
|---|
This work was supported by the following grants from the National Institutes of Health: 1 P50 HL-47212-03 (Specialized Center of Research in Arteriosclerosis), HL-31497, R01-HD32193, General Clinical Research Center Program, RR-52, RR-35, and CLINFLO. We thank Pauline Gugliotta and Sally Tharp for preparing this manuscript and Dr Paul Bachorik for helpful comments.
Received January 11, 1995; accepted May 23, 1995.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1. Kwiterovich PO Jr, Coresh J, Bachorik PS. Prevalence of hyperapoB and other lipoprotein phenotypes in premature coronary artery disease. Am J Cardiol. 1993;71:631-639. [Medline] [Order article via Infotrieve]
2. Coresh J, Kwiterovich PO Jr, Smith HH, Bachorik PS. Association of LDL particle size, density and composition with premature coronary artery disease in men and women. J Lipid Res. 1993;34:1687-1697. [Abstract]
3. Kwiterovich PO Jr. Genetics and molecular biology of familial combined hyperlipidemia. Curr Opin Lipidol. 1993;4:133-143.
4. Teng B, Sniderman AD, Soutar AK, Thompson GR. Metabolic basis of hyperapobetalipoproteinemia: turnover of apolipoprotein B in low density lipoprotein and its precursors and subfractions compared with normal and familial hypercholesterolemia. J Clin Invest. 1986;77:663-672.
5.
Genest J, Sniderman AD, Cianflone K, Teng B, Wacholder
S, Marcel Y, Kwiterovich PO Jr. Hyperapobetalipoproteinemia:
plasma lipoprotein responses to oral fat load.
Arteriosclerosis. 1986;6:297-304.
6. Sniderman A, Kwiterovich PO Jr. Hyperapobetalipoproteinemia and LDL and HDL2 Heterogeneity: Proceedings of the Workshop on Lipoprotein Heterogeneity. Washington, DC: 1978;293-304. US Dept of Health and Human Services publication NIH 87-2646.
7.
Kwiterovich PO Jr, Motevalli M, Miller M.
Acylation-stimulatory activity in hyperapoB fibroblasts:
enhanced cholesterol esterification with a new serum basic
protein. Proc Natl Acad Sci U S A. 1990;87:8980-8984.
8.
Kwiterovich PO Jr, Motevalli M, Miller M. The
effect of three serum basic proteins on the mass of lipids in normal
and hyperapoB fibroblasts. Arterioscler
Thromb. 1994;14:1-7.
9.
Cianflone KM, Sniderman AD, Walsh MJ, Vu HT, Gagnon J,
Rodriguez MA. Purification and characterization of
acylation-stimulating protein. J Biol
Chem. 1989;264:426-430.
10. Cianflone KM, Maslowska MH, Sniderman AD. Impaired response of fibroblasts from patients with hyperapobetalipoproteinemia to acylation-stimulating protein. J Clin Invest. 1990;85:722-730.
11. Baldo A, Sniderman AD, St-Luce S, Kohen R, Maslowska M, Hoang B, Monge JC, Bell A, Mulay S, Cianflone K. The adipsin-acylation stimulating protein system and regulation of intracellular triglyceride synthesis. J Clin Invest. 1993;92:1543-1547.
12.
Kwiterovich PO Jr, Motevalli M, Miller M, Bachorik PS,
Kafonek SD, Chatterjee S, Beaty T, Virgil D. Further insights
into the pathophysiology of hyperapobetalipoproteinemia.
Clin Chem. 1991;37:317-326.
13.
Levitzki A, Gazit A. Tyrosine kinase inhibition:
an approach to drug development. Science. 1995;267:1782-1788.
14.
Akiyama T, Ishida J, Nakagawa S, Ogawara H, Watanabe S,
Itoh N, Shibuya M, Fukami Y. Genistein, a specific
inhibitor of tyrosine specific protein kinases.
J Biol Chem. 1987;262:5592-5595.
15.
Hill TD, Dean NM, Mordan LJ, Lau AF, Kanemitsu MY,
Boynton AL. PDGF-induced activation of phospholipase C is not
required for induction. Science. 1990;248:1660-1663.
16.
Park DJ, Min HK, Rhee SG. IgE-induced tyrosine
phosphorylation of phospholipase C -1 in rat
basophilic leukemia cells. J Biol Chem. 1991;266:24237-24240.
17. Levitzki A. Tyrphostins: potential antiproliferative agents and novel molecular tools. Biochem Pharmacol. 1990;40:913-918. [Medline] [Order article via Infotrieve]
18. Satoh S, Rensland H, Pfitzer G. Ras proteins increase Ca2+ responsiveness on smooth muscle contractions. FEBS Lett. 1993;324:211-215. [Medline] [Order article via Infotrieve]
19.
Lowry OH, Rosebrough NJ, Farr AL, Randall RJ.
Protein measurement with the Folin phenol reagent.
J Biol Chem. 1951;193:265-275.
20.
Goldstein JL, Brown MS. Familial
hypercholesterolemia: identification of a
defect in the regulation of 3-hydroxy-3-methylglutaryl coenzyme A
reductase activity associated with overproduction of
cholesterol. Proc Natl Acad Sci U S A. 1973;70:2804-2808.
21. Yarden Y. Growth factor receptor tyrosine kinases. Annu Rev Biochem. 1988;57:443-477. [Medline] [Order article via Infotrieve]
22. Bolen JB. Non-receptor tyrosine protein kinases. Oncogene. 1993;8:2025-2031. [Medline] [Order article via Infotrieve]
23. Koch CA, Anderson D, Moran MF, Ellis C, Pawson T. SH2 and SH3 domains: elements that control interactions of cytoplasmic signaling protein. Science. 1992;252:668-674.
24.
Abler A, Smith JA, Randazzo PA, Rothenberg PL, Jarett
L. Genistein differentially inhibits postreceptor effects of
insulin in rats. J Biol Chem. 1992;267:3946-3951.
25.
Sato R, Goldstein JL, Brown MS. Replacement of
serine-871 of hamster 3-hydroxy-3-methylglutaryl CoA reductase prevents
phosphorylation by a AMP-activated kinase and
blocks inhibition of sterol synthesis induced by ATP depletion.
Proc Natl Acad Sci U S A. 1993;90:9261-9265.
This article has been cited by other articles:
 |
|
 |
|
  P. O. Kwiterovich Jr and M. Motevalli Differential Effect of Genistein on the Stimulation of Cholesterol Production by Basic Protein II in Normal and HyperapoB Fibroblasts Arterioscler Thromb Vasc Biol, January 1, 1998; 18(1): 57 - 64. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Motevalli, P. J. Goldschmidt-Clermont, D. Virgil, and P. O. Kwiterovich Jr. Abnormal Protein Tyrosine Phosphorylation in Fibroblasts from Hyperapobetalipoproteinemia Subjects J. Biol. Chem., September 26, 1997; 272(39): 24703 - 24709. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||