Dissociation between the Insulin-Sensitizing Effect of
Rosiglitazone and Its Effect on Hepatic and Intestinal
Lipoprotein Production
He´le`ne Duez, Benoît Lamarche, Kristine D. Uffelman, Rene´ Vale´ro, Linda Szeto, Simone Lemieux,
Jeffrey S. Cohn, and Gary F. Lewis
Departments of Medicine and Physiology (H.D., K.D.U., R.V., L.S., G.F.L.), Division of Endocrinology and Metabolism, University of
Toronto, Toronto, Ontario, Canada M5G 2C4; Institut des Nutraceutiques et Aliments Fonctionnels (B.L., S.L.), Universite´ Laval, Que´bec,
Canada G1K 7P4; and Clinical Research Institute of Montre´al (J.S.C.), Montre´al, Canada H2W 1R7
Context: Despite its potent, well-documented insulin-sensitizing effects, rosiglitazone (RSG) does
not effectively ameliorate the hypertriglyceridemia of insulin-resistant or diabetic individuals and
has even been shown to slightly but significantly increase triglyceride-rich lipoproteins (TRL) in
some studies. The mechanism of this effect is currently not known.
Objective: We investigated the effect of RSG treatment on TRL metabolism.
Design: This was a 12-wk, single-sequence, cross-over study of rosiglitazone vs. placebo for 6 wk.
Participants: Participants included 17 nondiabetic men with a broad range of insulin sensitivity.
Intervention: Intervention included rosiglitazone 8 mg/d vs. placebo for 6 wk.
Main Outcome Measure: TRL metabolism (concentration, production and catabolic rates) was
assessed in a constant fed state with a 12-h primed constant infusion of [D3]
L
-leucine and multi-
compartmental modeling.
Results: RSG treatment resulted in significant insulin sensitization with no change in body weight.
Fasting plasma triglyceride (TG) concentration, however, was higher with RSG vs. placebo (P
⫽
0.0006), as were fasting and fed TRL-TG, TRL-apoB-48, and TRL-apoB-100 (fed TRL-apoB-48: 0.93
⫾
0.08 vs. 0.76
⫾ 0.07 mg/dl, P ⫽0.017, and fed TRL-apoB-100: 15.57 ⫾ 0.90 vs. 13.71 ⫾ 1.27 mg/dl, P ⫽
0.029). This small but significant increase in plasma TRL concentration was explained by a tendency
for RSG to increase TRL production and reduce particle clearance, as indicated by the significantly
increased production to clearance ratios for both apoB-48-containing (0.43
⫾ 0.03 vs. 0.34 ⫾ 0.03,
P
⫽ 0.048) and apoB-100-containing (7.0 ⫾ 0.4 vs. 6.2 ⫾ 0.6, P ⫽ 0.029) TRL.
Conclusion: These data indicate dissociation between the insulin-sensitizing effects of RSG and
absence of anticipated reductions in production rates of apoB-100- and apoB-48-containing-TRL
particles, which may explain the absence of TG lowering seen in humans treated with this agent.
(J Clin Endocrinol Metab 93: 1722–1729, 2008)
I
ndividuals with insulin resistance and type 2 diabetes have an
increased risk of atherosclerotic cardiovascular disease (1).
Dyslipidemia is a prominent feature of these conditions and may
contribute to the increased risk of cardiovascular events (1, 2).
Typical diabetic dyslipidemia is characterized by elevated
plasma triglyceride-rich lipoproteins (TRL), low plasma high-
density lipoprotein (HDL)-cholesterol concentration and in-
creased numbers of small, dense low-density lipoprotein (LDL)
0021-972X/08/$15.00/0
Printed in U.S.A.
Copyright © 2008 by The Endocrine Society
doi: 10.1210/jc.2007-2110 Received September 20, 2007. Accepted February 12, 2008.
First Published Online February 19, 2008
Abbreviations: Apo, Apolipoprotein; FCR, fractional catabolic rate; FFA, free fatty acid;
HDL, high-density lipoprotein; HOMA-IR, homeostasis model assessment-insulin resis-
tance; LDL, low-density lipoprotein; PPAR, peroxisome proliferator-activated receptor; PR,
production rate; TG, triglyceride; TRL, triglyceride-rich lipoprotein; TZD, thiazolidinedione;
VLDL, very low-density lipoprotein.
O R I G I N A L
A R T I C L E
E n d o c r i n e
C a r e
1722
jcem.endojournals.org
J Clin Endocrinol Metab. May 2008, 93(5):1722–1729
particles (1). The elevation of TRL particles in insulin-resistant
states is contributed to by both hepatic [apolipoprotein (apo)-
B-100-containing] and intestinal (apoB-48-containing) lipopro-
teins in fasted and postprandial states (3, 4).
The thiazolidinedione (TZD) class of insulin-sensitizing
agents, agonists of the nuclear transcription factor peroxisome
proliferator-activated receptor (PPAR)-
␥, are widely used for the
treatment of type 2 diabetes (5). They improve insulin sensitivity
and glycemic control in individuals with type 2 diabetes (6, 7).
Although variable effects on plasma LDL-cholesterol concentra-
tion and LDL particle numbers have been reported, TZD treat-
ment of humans with type 2 diabetes more consistently raises
plasma HDL-cholesterol and increases LDL particle size (8).
However, despite the potent insulin-sensitizing properties of
TZDs and the link between insulin resistance and very low-den-
sity lipoprotein (VLDL) overproduction, the ability of these
agents to lower plasma triglycerides is variable. There are well-
documented differences between pioglitazone and rosiglitazone
with respect to their effects on TRL, with pioglitazone having a
modest triglyceride-lowering effect (8 –10), and rosiglitazone
having either no reduction or even a slight increase in plasma
triglycerides (8, 11, 12). To examine the mechanism of effect of
TZDs of TRL, it is necessary to go beyond the measurement of
plasma lipoprotein and apolipoprotein concentrations and to
determine the kinetics of TRL metabolism in vivo.
We have recently shown that diet-induced whole-body insu-
lin resistance in the Syrian golden hamster was associated with
mild hypertriglyceridemia and overproduction of both intestinal
and hepatic TRL (13, 14). Using this model of insulin resistance-
associated dyslipidemia, we demonstrated that rosiglitazone im-
proved whole-body insulin sensitivity and insulin signaling, and
partially reversed both hepatic and intestinal lipoprotein over-
production (15–17). This effect, however, may not be truly re-
flective of the effect of rosiglitazone treatment of humans with
insulin resistance.
In the present study, we investigated the effect of rosiglitazone
on steady-state fed TRL-apoB-48 and TRL-apoB-100 produc-
tion and clearance rates in men with a broad range of homeosta-
sis model assessment (HOMA) scores and plasma insulin con-
centrations. We specifically chose to examine nondiabetic
individuals in this study to eliminate the potential confounding
effect of rosiglitazone-induced improvements in glycemic con-
trol that occur in people with diabetes.
Subjects and Methods
Subjects and study design
Seventeen healthy, normoglycemic men, aged 30 – 60 yr with a
broad range of body weights (from 64.3 to 134.3 kg) and body mass
index (from 20.0 to 41.6 kg/m
2
) participated in the study for deter-
mination of apoB-containing lipoprotein kinetics (see baseline char-
acteristics in Table 1). Subjects were enrolled if their total plasma
cholesterol was 5.5 mmol/liter or less, HDL-cholesterol 0.8 mmol/
liter or greater, LDL-cholesterol 4.0 mmol/liter or less, and triglyc-
erides 4.0 mmol/liter or less to exclude those with extreme dyslipi-
demia. No subject was taking medications and all had a normal 75-g
oral glucose tolerance test performed immediately before enrollment
in the study. Fasting insulin concentrations ranged from 21.0 to 151.0
pmol/liter. HOMA-IR as an index of insulin resistance was calculated
as previously described (42) and ranged from 0.70 to 5.93 in the study
subjects. The study was designed and conducted as a 12-wk, single-
sequence, cross-over group design, with an initial 6-wk placebo treat-
ment period followed by 6-wk rosiglitazone (8 mg/d) treatment pe-
riod.
Rosiglitazone
and
placebo
tablets
were
supplied
by
GlaxoSmithKline Pharmaceuticals (Mississauga, Ontario, Canada).
Participants were monitored weekly for weight, vital signs, and fast-
ing blood glucose. Compliance with medications was assessed at the
weekly visits by tablet counting and was determined to be greater than
98% for all subjects. All participants were nonsmokers and none had
a previous history of cardiovascular disease or systemic illness. None
had any surgical intervention within 6 months before the studies.
TABLE 1.
Demographic and biochemical (fasting plasma metabolites, hormones, and lipids) characteristics of subjects during
the placebo (PL) and rosiglitazone (RSG) treatment periods
Subject
Age
(yr)
Weight
(kg)
Glucose
(mmol/liter)
Insulin
(pmol/liter)
HOMA-IR
FFA
(mmol/liter)
PL
RSG
PL
RSG
PL
RSG
PL
RSG
PL
RSG
1
40
124
122
5.5
4.5
105
92
4.28
3.07
0.503
0.355
2
54
120
118
4.6
4.2
70
58
2.39
1.80
0.446
0.402
3
39
134
133
4.0
4.4
59
49
1.75
1.60
0.366
0.362
4
56
86
86
6.1
5.3
114
81
5.15
3.18
0.518
0.351
5
60
83
84
4.7
4.6
120
99
4.18
3.37
0.271
0.348
6
47
95
94
4.6
4.1
87
37
2.96
1.12
0.486
0.315
7
55
107
106
5.7
5.3
123
114
5.19
4.48
0.232
0.243
8
59
98
100
5.3
5.2
151
107
5.93
4.12
0.326
0.177
9
30
85
86
4.8
4.7
79
91
2.79
3.18
0.448
0.304
10
50
66
65
5.1
5.3
60
45
2.27
1.77
0.381
0.351
11
41
64
64
5.1
4.5
41
35
1.56
1.18
0.392
0.442
12
42
77
79
4.7
4.7
43
18
1.50
0.63
0.491
0.475
13
40
70
72
5.0
5.2
67
64
2.49
2.47
0.221
0.154
14
50
79
81
4.5
4.7
21
25
0.70
0.86
0.541
0.79
15
50
76
76
5.6
5.0
51
33
2.12
1.22
0.343
0.448
16
58
83
84
4.8
4.4
41
34
1.47
1.10
0.396
0.356
17
51
67
69
4.5
4.8
46
53
1.54
1.90
0.166
0.47
Mean
⫾
SEM
48
⫾ 2 89 ⫾ 5 89 ⫾ 5 5.0 ⫾ 0.1 4.8 ⫾ 0.1 75.2 ⫾ 8.7 60.9 ⫾ 7.4 2.84 ⫾ 0.37 2.18 ⫾ 0.29 0.38 ⫾ 0.03 0.37 ⫾ 0.03
P value (PL vs. RSG)
0.584
0.044
0.0028
0.0018
0.750
P values are by paired t test.
J Clin Endocrinol Metab, May 2008, 93(5):1722–1729
jcem.endojournals.org
1723
TRL apoB-containing lipoprotein kinetics studies
Kinetics studies for assessment of TRL-apoB metabolism were per-
formed at the end of the 6-wk placebo treatment period and again at the
end of the 6-wk rosiglitazone treatment period. Subjects were admitted
to the Endocrine/Metabolic Diagnostic Center of the Toronto General
Hospital at 1500 h after fasting from midnight the previous night. An iv
line was placed in a superficial vein in each forearm, one for infusion and
one for blood sampling, and a baseline fasting sample was drawn. Sub-
jects were provided with a mixed meal (American Heart Association
phase 1 diet) at approximately 1700 h, and at approximately 0700 h the
following morning, subjects were instructed to begin ingesting 15 iden-
tical hourly volumes of a liquid food supplement called Boost (Mead
Johnson Nutritionals, Ottawa, Ontario, Canada), each equivalent to one
15th of their total daily caloric needs, using the Harris Benedict equation
to determine the total energy requirements (based on height, weight, age,
and activity factors). Boost contains 20% of total calories from protein,
62% carbohydrate, and 18% fat (of the total energy derived from fat,
25% was polyunsaturated fatty acids, 65% monounsaturated fatty
acids, and 13% saturated fatty acids). Three hours after starting to ingest
Boost, subjects received a primed constant infusion (10
mol/kg bolus
followed by 10
mol/kg䡠h for 12 h) of deuterium-labeled leucine (18)
(
L
-[5,5,5-
2
H
3
]leucine; 98%, Cambridge Isotope Laboratories, Andover,
MA) to enrich apoB-100 and apoB-48 in hepatic and intestinally derived
lipoprotein particles, respectively, to calculate the production and clear-
ance rates of the particles as previously described (19). Blood samples
were collected at 1, 2, 3, 5, 7, 9, 10, 11, and 12 h.
The Research Ethics Board of the University Health Network, Uni-
versity of Toronto, approved the study, and all subjects gave written
informed consent before their participation.
Sample processing
TRLs were isolated at each time point by centrifugation, and delipi-
dated proteins were separated by preparative 3.3% SDS-PAGE and
stained with Coomassie R-250. ApoB-100 and apoB-48 gel slices were
excised, hydrolyzed and derivatized to allow for the determination of
plasma leucine isotopic enrichment as previously described (18 –21). En-
richment of samples with deuterium-labeled leucine was measured by gas
chromatography/mass spectrometry (Agilent 5973 GC/MS; Agilent
Technologies Canada Inc., Mississauga, Ontario, Canada), and tracer to
tracee ratios (corrected for background leucine in gel slices) were calcu-
lated from isotopic ratios for each sample as previously described (22).
Laboratory measurements
Triglycerides, cholesterol, and free fatty acids (FFAs) were deter-
mined using commercially available kits (Roche Diagnostics, Mann-
heim, Germany; Wako Industrials, Osaka, Japan). Plasma insulin con-
centrations were assayed by RIA using a human specific insulin kit (Linco
Research, St. Louis, MO). Glucose was measured enzymatically using a
Beckman Glucose Analyzer II (Beckman Instruments Corp., Fullerton,
CA). Total apoB in plasma and TRL was measured by electroimmuno-
assay as previously described (23). ApoB-48 and apoB-100 mass in the
TRL fraction was quantified using analytical SDS-PAGE as previously
described (24). LDL size was determined on polyacrylamide gradient gels
(4 –16%) using molecular mass calibration markers (Pharmacia Biotech
Inc., Uppsala, Sweden).
Analysis of lipoprotein production and clearance rates
Stable isotope enrichment curves for apoB-48 and apoB-100 were
fitted to a three-compartment model using SAAM II computer software
(SAAM II Institute, Seattle, WA) as previously described (22). Each sub-
ject was in steady state with respect to apoB-48 concentrations so frac-
tional catabolic rate (FCR) was equivalent to fractional synthetic rate.
Kinetic parameters were derived by analyzing individual enrichment
curves and only those in which the coefficient of variation for apoB-48
or apoB-100 modeling was less than 25% were included. For this reason
the apoB-48 kinetic results are reported in only 11 of the 17 study par-
ticipants because only 11 subjects had acceptable TRL apoB-48 enrich-
ment vs. time curves for both placebo and rosiglitazone-treated studies.
Production rates were derived using the FCR of the TRL apoB-100
and apoB-48 and multiplied by their pool sizes measured over the 12 h
of the study per kilogram body weight, where pool size
⫽ plasma con-
centration (milligrams per deciliter) between time 3 h and time 12 h of
the kinetic
⫻ plasma volume (0.045 liter/kg).
Details of the laboratory measurements, including the determination
of TRL apoB-containing lipoprotein kinetics are included in the online
data supplement, published as supplemental data on The Endocrine So-
ciety’s Journals Online Web site at http://jcem.endojournals.org.
Statistics
Results are presented as mean
⫾
SEM
. Paired t tests were used to
compare patients during the placebo vs. rosiglitazone treatment period.
ANOVA was used to analyze TRL-triglycerides (TG), TRL-apoB-100,
and TRL-apoB-48 increase over the time. All analyses were performed
with SPSS (version13; SPSS Inc., Chicago, IL). For all of the analyses, a
P value
⬍ 0.05 was considered significant.
Results
Demographic characteristics and the effect of
rosiglitazone treatment on anthropometric indices and
insulin sensitivity (Table 1)
The clinical characteristics of subjects after the placebo and
rosiglitazone treatment periods are given in Table 1. Rosiglita-
zone administration did not induce weight gain over the 6-wk
treatment period. Waist circumference and body mass index
were also not affected by rosiglitazone treatment (data not
shown). As anticipated, rosiglitazone significantly decreased
plasma insulin (P
⫽ 0.003) and glucose (P ⫽ 0.044). Accord-
ingly, the HOMA-IR index was significantly reduced after ros-
iglitazone treatment (P
⫽ 0.002). There was no difference in FFA
levels after rosiglitazone treatment. These data indicate that ros-
iglitazone administration improved insulin sensitivity, without a
significant change in body fat.
Effect of rosiglitazone treatment on fasting plasma
(Table 2) and TRL (Table 3) lipids and apolipoprotein B
concentrations
As shown in Table 2, rosiglitazone treatment induced a sig-
nificant increase in total plasma cholesterol (P
⫽ 0.003), apoB
(P
⫽ 0.004), and triglycerides (P ⫽ 0.0006). HDL-cholesterol
increased with rosiglitazone treatment (P
⫽ 0.035). No change
was observed in plasma LDL-cholesterol concentration or LDL
particle size after rosiglitazone treatment (P
⫽ 0.942).
TRL composition was examined in more detail (Table 3).
Fasting TRL-TGs were significantly elevated after rosiglitazone
treatment (P
⫽ 0.007), and a parallel increase in TRL-apoB-48
(P
⬍ 0.001) and TRL apoB-100 (P ⫽ 0.005) was observed after
rosiglitazone administration.
Effect of rosiglitazone on TRL-TG and TRL-apoB-48 and
apoB-100 concentrations during liquid formula ingestion
(Fig. 1)
Mean TRL-TG, TRL-apoB-100, and TRL-apoB-48 concen-
trations at fasting (0 h) and during the 15-h fat feeding study are
1724
Duez et al.
Rosiglitazone and Lipoprotein Metabolism
J Clin Endocrinol Metab, May 2008, 93(5):1722–1729
illustrated in Fig. 1, A–C, respectively. TRL-TG in the fed state
tended to be higher in rosiglitazone-treated patients. Similarly,
both TRL-apoB-48 (mean fed state rosiglitazone, 0.93
⫾ 0.08 vs.
placebo, 0.76
⫾ 0.07, P ⫽ 0.017) and TRL-apoB-100 (mean fed
state rosiglitazone, 15.57
⫾ 0.90 vs. placebo, 13.71 ⫾ 1.27, P ⫽
0.029) were higher after rosiglitazone treatment.
Effect of rosiglitazone on hepatic and intestinal
lipoprotein production and clearance rates (Fig. 2)
No significant changes in TRL-apoB-48 and TRL-apoB-100
production and clearance rates were observed after rosiglitazone
treatment (Fig. 2). However, TRL-apoB-48 and TRL-apoB-100
production rates tended to slightly increase after rosiglitazone
treatment [TRL-apoB-48 production rate (PR): rosiglitazone,
1.52
⫾ 0.21 mg/kg䡠d vs. placebo, 1.25 ⫾ 0.26 mg/kg䡠d, P ⫽ 0.288
and TRL-apoB-100 PR: rosiglitazone, 28.93
⫾ 3.57 mg/kg䡠d vs.
placebo, 26.68
⫾ 3.42 mg/kg䡠d, P ⫽ 0.410], whereas clearance
rates tended to decrease (TRL-apoB-48 FCR: rosiglitazone,
3.80
⫾ 0.56 pool/d vs. placebo, 4.41 ⫾ 0.92 pool/d, P ⫽ 0.665,
and TRL-apoB-100 FCR: rosiglitazone, 4.21
⫾ 0.51 pool/d vs.
placebo, 4.42
⫾ 0.42 pool/d, P ⫽ 0.645). However, the PR to
FCR ratio was significantly increased after rosiglitazone treat-
ment for both TRL-apoB-48 and TRL-apoB-100 (TRL-apoB-
48: rosiglitazone, 0.43
⫾ 0.03 vs. placebo 0.34 ⫾ 0.03, P ⫽
0.048, and TRL-apoB-100: rosiglitazone, 7.0
⫾ 0.4 vs. placebo,
6.2
⫾ 0.6, P ⫽ 0.029).
TABLE 3.
Fasting TRL TG and apoB levels during the placebo (PL) and rosiglitazone (RSG) treatment periods
Subject
TRL
TRL
TRL
TG (mmol/liter)
ApoB-100 (mg/dl)
ApoB-48 (mg/dl)
PL
RSG
PL
RSG
PL
RSG
1
1.11
0.65
11.0
12.1
0.33
0.27
2
0.79
0.72
10.7
10.3
0.26
0.28
3
0.81
1.29
13.8
15.5
0.45
0.62
4
3.32
4.22
23.1
30.8
0.83
1.14
5
0.44
1.34
9.2
20.2
0.47
0.64
6
0.64
0.96
10.2
11.0
0.47
0.53
7
0.70
1.54
6.1
12.8
0.33
0.45
8
1.49
1.72
9.4
10.7
0.61
0.79
9
0.27
0.53
6.2
6.4
0.20
0.36
10
0.78
1.53
5.3
10.9
0.31
0.63
11
0.74
0.78
6.5
8.5
0.57
0.70
12
0.58
1.30
7.1
12.3
0.24
0.68
13
0.71
0.51
15.2
10.4
0.30
0.27
14
0.55
0.67
8.2
10.1
0.55
0.67
15
0.70
0.91
6.5
11.5
0.26
0.47
16
0.52
0.32
8.3
7.9
0.45
0.48
17
0.69
1.14
12.1
19.6
0.30
0.32
Mean
⫾
SEM
0.87
⫾ 0.17
1.18
⫾ 0.21
9.94
⫾ 1.07
13.00
⫾ 1.42
0.41
⫾ 0.04
0.55
⫾ 0.05
P value (PL vs. RSG)
0.0071
0.0051
0.0005
P values are by paired t test.
TABLE 2.
Plasma lipids and apoB concentrations in the fasted state during the placebo (PL) and rosiglitazone (RSG) treatment
periods
Subject
Total cholesterol
(mmol/liter)
HDL-C
(mmol/liter)
LDL-C
(mmol/liter)
TGs
(mmol/liter)
Apo B
(mg/dl)
LDL size
(nm)
PL
RSG
PL
RSG
PL
RSG
PL
RSG
PL
RSG
PL
RSG
1
4.45
4.45
1.06
1.17
2.96
2.67
1.49
1.39
83
92
23.71
23.86
2
4.83
5.07
0.83
1.25
3.56
3.22
1.43
1.37
97
107
24.78
24.47
3
4.79
4.91
1.25
1.02
2.70
2.38
1.48
1.89
81
84
25.42
24.70
4
5.21
6.90
0.99
1.05
2.07
1.93
4.00
5.10
154
171
24.78
24.47
5
4.31
5.01
0.97
0.81
2.92
3.14
0.88
1.85
68
78
23.46
23.77
6
3.71
4.56
0.93
0.96
2.35
2.91
0.64
0.96
60
76
23.77
24.55
7
4.09
5.06
0.95
1.19
3.05
2.37
1.31
2.20
98
115
24.71
25.87
8
5.42
6.08
0.97
1.28
3.26
3.36
1.96
2.61
106
117
25.70
25.70
9
4.39
4.16
0.90
1.08
3.25
2.72
0.61
0.72
100
99
25.85
25.61
10
4.21
4.96
1.02
1.06
2.52
2.89
1.38
2.06
81
83
25.93
25.37
11
3.52
3.50
0.89
1.08
2.07
1.71
0.98
1.18
96
105
24.02
24.32
12
3.85
4.04
1.05
1.20
2.32
1.92
0.96
1.88
85
96
25.16
24.32
13
5.04
4.94
0.93
0.94
3.12
3.43
1.29
1.25
98
82
24.17
24.17
14
5.59
6.13
1.16
1.38
4.04
3.73
1.13
1.28
85
97
23.95
23.66
15
4.81
4.53
1.09
1.04
3.35
2.83
0.97
1.36
78
87
23.39
22.95
16
4.76
5.28
1.57
1.70
2.94
3.27
0.74
0.61
85
77
22.95
23.62
17
4.25
4.97
0.94
0.87
2.50
3.25
1.09
1.69
96
109
22.66
23.17
Mean
⫾
SEM
4.54
⫾ 0.14 4.97 ⫾ 0.20 1.03 ⫾ 0.04 1.12 ⫾ 0.05 2.88 ⫾ 0.13 2.81 ⫾ 0.14 1.31 ⫾ 0.19 1.73 ⫾ 0.25 92.2 ⫾ 4.9 98.5 ⫾ 5.5 24.38 ⫾ 0.25 24.39 ⫾ 0.21
P value (PL vs. RSG)
0.0032
0.035
0.487
0.0006
0.0038
0.942
P values are by paired t test.
J Clin Endocrinol Metab, May 2008, 93(5):1722–1729
jcem.endojournals.org
1725
Discussion
In the present study, we investigated the effect of 6-wk treatment
with rosiglitazone on TRL metabolism in the fed state by modeling
the stable isotope enrichment of protein (apoB-48 and apoB-100)
moieties of TRL particles. We were thus able to examine both in-
testinally and hepatically derived TRL metabolism. We specifically
chose to study nondiabetic individuals to exclude a potential con-
founding effect of improvements in glycemic control that would be
anticipated to occur with rosiglitazone treatment of patients with
type 2 diabetes. Because insulin resistance is associated with over-
production of TRL particles by both the intestine (22) and the liver
(25), we hypothesized that the well-documented improvement in
whole-body and hepatic insulin sensitivity with rosiglitazone would
lower plasma TRL concentrations by reducing their production
rates. Instead we found that, despite significant insulin sensitization
in our subjects (reduction in plasma glucose, insulin, and conse-
quently HOMA-IR), there was no reduction in either apoB48-con-
taining or apoB-100-containing TRL production. In fact, we ob-
served a small but significant increase in plasma TRL
concentrations within the normal range, which can be explained by
a tendency for rosiglitazone to increase TRL production and reduce
particle clearance, as indicated by the significantly increased PR to
FCR ratios for both apoB48-containing and apoB-100-containing
TRL. Subjects with a wide range of insulin sensitivity were recruited
in our study. It is interesting that rosiglitazone treatment does not
improve TRL metabolism, even in subjects with the lowest pre-
treatment insulin sensitivity (i.e. those who were expected to benefit
the most from insulin sensitization). Thus, our findings demonstrate
that rosiglitazone does not reduce either hepatic or intestinal TRL
production, despite improvement in whole-body insulin sensitivity,
and rather suggests dissociation between its insulin-sensitizing
properties and its effects on TRL metabolism.
The effect of rosiglitazone on fasting plasma TGs is variable,
with most studies reporting no change (26 –29) or an increase in
plasma TGs (8, 11, 12). Several investigators have examined the
effect of TZDs on postprandial TG metabolism. In patients with
type 2 diabetes, rosiglitazone was reported to decrease TG levels
after an oral fat load (30, 31), the latter also showing a reduction
in TRL-TG. In contrast, James et al. (32) and Chappuis et al. (12)
recently demonstrated an increase in fasting and postprandial
TG levels and reported an increase of plasma apoB-48 concen-
tration after a mixed meal in rosiglitazone-treated patients, com-
TRL apoB-48 concentratio
n
(mg/dl
)
C
post-placebo
post-RSG
A
TRL TG
concentration
(mM)
Time (hour)
0
0.50
1.00
1.50
2.00
2.50
0 1 2 3 4 5 6 7 8 9 1011 12 13 14 15
TRL apoB-100 concentration
(mg/dl
)
B
Time (hour)
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Time (hour)
0
0.20
0.40
0.60
0.80
1.00
1.20
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
0
2
4
6
8
10
12
14
16
18
20
post-placebo
post-RSG
post-placebo
post-RSG
FIG. 1. TRL-TG, TRL-apoB-100, and TRL-apoB-48 concentrations over the time course of the kinetic study. TRL-TG (A), TRL-apoB-100 (B), and TR-apoB-48
(C) were measured throughout the 15-h study in subjects after the placebo period (empty squares) or after the rosiglitazone (RSG) treatment period
(filled squares). Subjects began to ingest a liquid formula hourly at 0700 h (0 time), after an overnight fast. A primed, continuous infusion of deuterium-
labeled leucine was started after 3 h of formula ingestion, and TRL-apoB kinetics parameters were calculated during the subsequent 12-h study period
(time 3 to 15 h). Values are mean
⫾
SEM
for each group. The overall significance over the time as analyzed by ANOVA was P
⫽ 0.027 for TRL-apoB-100
and P
⬍ 0.001 for TRL-apoB-48.
1726
Duez et al.
Rosiglitazone and Lipoprotein Metabolism
J Clin Endocrinol Metab, May 2008, 93(5):1722–1729
pared with controls, and one of these studies (32) reported an
increase of plasma apoB-48 concentration after a mixed meal in
rosiglitazone-treated patients, compared with controls, a finding
that is consistent with ours. Our results showing no reduction in
either hepatic or intestinal TRL production and no increase in the
clearance of the particles, despite ameliorated insulin sensitivity
after rosiglitazone treatment, may explain the absence of robust
TG-lowering in earlier studies.
The effect of pioglitazone, another member of the TZD class
of insulin-sensitizing agents, on TRL metabolism was recently
examined by Nagashima et al. (9) in patients with type 2 diabe-
tes. Consistent with our findings, they also showed no reduction
in hepatic VLDL production with pioglitazone treatment. In con-
trast to our results, in which there was a trend toward a reduction
in TRL clearance with rosiglitazone treatment, pioglitazone low-
ered plasma TG by approximately 30% by stimulating
VLDL-TG but not VLDL particle (i.e. apoB) clearance via a
mechanism postulated to be due to increased LPL mass and de-
creased apoC-III (9). This effect may be attributed to partial
PPAR
␣ activation by pioglitazone (33), although an indirect ef-
fect on TRL clearance secondary to improved glycemic control
cannot be excluded. More recently, pioglitazone treatment was
reported to lower fasting and postprandial TG levels in patients
with type 2 diabetes, independent of any change in LPL activity,
and was associated with a significant decrease in hepatic lipase
activity (10). The difference between the effect of rosiglitazone
and pioglitazone on plasma TRL concentrations was recently
confirmed in the first large-scale, prospective, double-blinded,
head-to-head comparison study (8). It appears, therefore, that
pioglitazone and rosiglitazone have common (insulin sensitizing)
and distinct (lipid related effects) properties. The difference in
effects on TRL metabolism appears to be confined to differences
in the clearance of TRL triglycerides, whereas their effects on
TRL production rates are similar.
Insulin sensitization by a variety of methods, including weight
reduction and exercise, is associated with reductions in VLDL
secretion and plasma TRL concentrations (34 –36), which is why
we were surprised that rosiglitazone did not effectively reduce
TRL secretion. VLDL secretion is, to a large extent, driven by the
increased FFA flux to liver that characterizes insulin resistance
and type 2 diabetes (37). FFA lowering has been demonstrated in
some human studies secondary to an improvement of insulin
sensitivity in muscle and/or adipose tissue (28, 30, 38, 39). In the
present study, however, rosiglitazone treatment failed to de-
crease fasting plasma FFA concentration, although we cannot
exclude the possibility that it may have resulted in lower post-
prandial FFA flux to the liver. The fact that our subjects were not
diabetic, or the short period of treatment might explain the dif-
ference from the above-mentioned studies. A reduction of FFA
flux to the liver would be anticipated to decrease TRL produc-
tion, which was not the case in the present study. The significant
increase in plasma total cholesterol levels, and possibly in plasma
VLDL-cholesterol, indicates that failure to decrease hepatic cho-
lesterol with rosiglitazone may contribute to the VLDL overse-
cretion as well. We did not measure liver fat in the present study,
but the anticipated reduction in liver fat content would be an-
other factor expected to reduce VLDL secretion. It may be pos-
tulated that a direct effect of rosiglitazone in stimulating hepatic
lipogenesis might have counteracted the other insulin-sensitizing
effects of rosiglitazone. To the best of our knowledge, no pre-
vious study has addressed whether TZD treatment results in
direct stimulation of hepatic de novo lipogenesis. This could
occur through a PPAR
␥-LXR-sterol regulatory element-binding
protein-1c-mediated mechanism (40) or a PPAR
␥2-dependent
but LXR/sterol regulatory element-binding protein-1c-indepen-
dent mechanism as recently demonstrated in a mouse model of
insulin resistance- and steatosis-associated dyslipidemia (41) or
by other yet unknown direct effects of rosiglitazone on TRL
particle biogenesis and TRL-lipid clearance. The mechanism ac-
counting for the puzzling dissociation between the hepatic and
whole-body insulin-sensitizing effects of rosiglitazone and TRL
metabolism requires further investigation.
In conclusion, we have demonstrated that, despite a signifi-
cant improvement in insulin sensitivity, rosiglitazone does not
Fractional Catabolic Rate (pools/da
y
)
TRLapoB-100
TRLapoB-48
TRLapoB-100
TRLapoB-48
TRLapoB-100
Pr
oduction
Rate (mg/kg.
d
a
y
)
0
1.0
2.0
3.0
4.0
5.0
6.0
post-placebo
post-RSG
0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
0
5
10
15
20
25
30
35
post-placebo
post-RSG
B
A
TRLapoB-4
8
Pr
oduction
Rate (mg/kg.
d
a
y
)
FIG. 2. Effect of rosiglitazone (RSG) on TRL-apoB-48 and TRL-apoB-100 production and catabolic rates. TRL-apoB-100 (n
⫽ 17) and TRL-apoB-48 (n ⫽ 11)
PRs (A) and FCRs (B) after 6 wk treatment with placebo or RSG. No statistically significant differences existed between placebo and RSG treatment
periods.
J Clin Endocrinol Metab, May 2008, 93(5):1722–1729
jcem.endojournals.org
1727
improve fasting and postprandial lipemia in men with a broad
range of insulin sensitivity. Our observations highlight the fact
that not all clinical effects of PPAR
␥ agonists, such as the TZDs,
can be anticipated according to the insulin-sensitizing effects of
these drugs, which is but one of many downstream effects of
PPAR
␥ activation.
Acknowledgments
We are indebted to Patricia Harley, R.N., for her assistance with subject
recruitment and conducting the clinical protocol.
Address all correspondence and requests for reprints to: Dr. Gary F.
Lewis, Toronto General Hospital, 200 Elizabeth Street, EN12-218, To-
ronto, Ontario, Canada M5G 2C4. E-mail: gary.lewis@uhn.on.ca.
This work was supported by the Canadian Institutes for Health Re-
search (MOP-43839), Glaxosmithkline Canada, and the Heart and
Stroke Foundation of Canada. G.F.L. holds a Canada Research Chair in
Diabetes and is a Career Investigator of the Heart and Stroke Foundation
of Canada. B.L. holds a Canada Research Chair in Nutrition and Car-
diovascular Health. H.D. is the recipient of a Postdoctoral Fellowship
Award from the Heart and Stroke Foundation of Canada.
Present address for J.S.C.: Heart Research Institute, Nutrition and
Metabolism Group, Camperdown, Sydney, New South Wales 2050,
Australia.
Disclosure statement: G.F.L. received a grant from GlaxoSmithKline,
which provided funding in part for this study. All other authors have
nothing to disclose.
References
1. Ginsberg HN 2000 Insulin resistance and cardiovascular disease. J Clin Invest
106:453– 458
2. Lewis GF, Steiner G 1996 Hypertriglyceridemia and its metabolic conse-
quences as a risk factor for atherosclerotic cardiovascular disease in non-in-
sulin-dependent diabetes mellitus. Diabetes Metab Rev 12:37–56
3. Schaefer EJ, McNamara JR, Shah PK, Nakajima K, Cupples LA, Ordovas JM,
Wilson PW; Framingham Offspring Study 2002 Elevated remnant-like particle
cholesterol and triglyceride levels in diabetic men and women in the Framing-
ham Offspring Study. Diabetes Care 25:989 –994
4. Taskinen MR 2003 Diabetic dyslipidaemia: from basic research to clinical
practice. Diabetologia 46:733–749
5. Staels B, Fruchart JC 2005 Therapeutic roles of peroxisome proliferator-ac-
tivated receptor agonists. Diabetes 54:2460 –2470
6. Dormandy JA, Charbonnel B, Eckland DJ, Erdmann E, Massi-Benedetti M,
Moules IK, Skene AM, Tan MH, Lefe`bvre PJ, Murray GD, Standl E, Wilcox
RG, Wilhelmsen L, Betteridge J, Birkeland K, Golay A, Heine RJ, Kora´nyi L,
Laakso M, Moka´n M, Norkus A, Pirags V, Podar T, Scheen A, Scherbaum W,
Schernthaner G, Schmitz O, Skrha J, Smith U, Taton J; PROactive investigators
2005 Secondary prevention of macrovascular events in patients with type 2
diabetes in the PROactive Study (PROspective pioglitAzone Clinical Trial In
macroVascular Events): a randomised controlled trial. Lancet 366:1279 –1289
7. Sidhu JS, Kaposzta Z, Markus HS, Kaski JC 2004 Effect of rosiglitazone on
common carotid intima-media thickness progression in coronary artery disease
patients without diabetes mellitus. Arterioscler Thromb Vasc Biol 24:930 –934
8. Goldberg RB, Kendall DM, Deeg MA, Buse JB, Zagar AJ, Pinaire JA, Tan MH,
Khan MA, Perez AT, Jacober SJ; GLAI Study Investigators 2005 A comparison
of lipid and glycemic effects of pioglitazone and rosiglitazone in patients with
type 2 diabetes and dyslipidemia. Diabetes Care 28:1547–1554
9. Nagashima K, Lopez C, Donovan D, Ngai C, Fontanez N, Bensadoun A,
Fruchart-Najib J, Holleran S, Cohn JS, Ramakrishnan R, Ginsberg HN 2005
Effects of the PPAR
␥ agonist pioglitazone on lipoprotein metabolism in pa-
tients with type 2 diabetes mellitus. J Clin Invest 115:1323–1332
10. Majali KA, Cooper MB, Staels B, Luc G, Taskinen MR, Betteridge DJ 2006
The effect of sensitisation to insulin with pioglitazone on fasting and post-
prandial lipid metabolism, lipoprotein modification by lipases, and lipid trans-
fer activities in type 2 diabetic patients. Diabetologia 1–11
11. Nolan JJ, Jones NP, Patwardhan R, Deacon LF 2000 Rosiglitazone taken once
daily provides effective glycaemic control in patients with type 2 diabetes
mellitus. Diabet Med 17:287–294
12. Chappuis B, Braun M, Stettler C, Allemann S, Diem P, Lumb PJ, Wierzbicki
AS, James R, Christ ER 2007 Differential effect of pioglitazone (PGZ) and
rosiglitazone (RGZ) on postprandial glucose and lipid metabolism in patients
with type 2 diabetes mellitus: a prospective, randomized crossover study. Di-
abetes Metab Res Rev 23:392–399
13. Taghibiglou C, Carpentier A, Van Iderstine SC, Chen B, Rudy D, Aiton A,
Lewis GF, Adeli K 2000 Mechanisms of hepatic very low density lipoprotein
overproduction in insulin resistance. Evidence for enhanced lipoprotein as-
sembly, reduced intracellular ApoB degradation, and increased microsomal
triglyceride transfer protein in a fructose-fed hamster model. J Biol Chem
275:8416 – 8425
14. Haidari M, Leung N, Mahbub F, Uffelman KD, Kohen-Avramoglu R, Lewis
GF, Adeli K 2002 Fasting and postprandial overproduction of intestinally
derived lipoproteins in an animal model of insulin resistance. Evidence that
chronic fructose feeding in the hamster is accompanied by enhanced intestinal
de novo lipogenesis and ApoB48-containing lipoprotein overproduction.
J Biol Chem 277:31646 –31655
15. Carpentier A, Taghibiglou C, Leung N, Szeto L, Van Iderstine SC, Uffelman
KD, Buckingham R, Adeli K, Lewis GF 2002 Ameliorated hepatic insulin
resistance is associated with normalization of microsomal triglyceride transfer
protein expression and reduction in very low density lipoprotein assembly and
secretion in the fructose-fed hamster. J Biol Chem 277:28795–28802
16. Leung N, Naples M, Uffelman K, Szeto L, Adeli K, Lewis GF 2004 Rosigli-
tazone improves intestinal lipoprotein overproduction in the fat-fed Syrian
Golden hamster, an animal model of nutritionally-induced insulin resistance.
Atherosclerosis 174:235–241
17. Lewis GF, Uffelman K, Naples M, Szeto L, Haidari M, Adeli K 2005 Intestinal
lipoprotein overproduction, a newly recognized component of insulin resis-
tance, is ameliorated by the insulin sensitizer rosiglitazone: studies in the fruc-
tose-fed Syrian golden hamster. Endocrinology 146:247–255
18. Cohn JS, Wagner DA, Cohn SD, Millar JS, Schaefer EJ 1990 Measurement of
very low density and low density lipoprotein apolipoprotein (Apo) B-100 and
high density lipoprotein Apo A-I production in human subjects using deuter-
ated leucine. Effect of fasting and feeding. J Clin Invest 85:804 – 811
19. Batal R, Tremblay M, Barrett PH, Jacques H, Fredenrich A, Mamer O, Dav-
ignon J, Cohn JS 2000 Plasma kinetics of apoC-III and apoE in normolipidemic
and hypertriglyceridemic subjects. J Lipid Res 41:706 –718
20. Batal R, Tremblay M, Krimbou L, Mamer O, Davignon J, Genest Jr J, Cohn
JS 1998 Familial HDL deficiency characterized by hypercatabolism of mature
apoA-I but not proapoA-I. Arterioscler Thromb Vasc Biol 18:655– 664
21. Tremblay AJ, Lamarche B, Ruel I, Hogue JC, Bergeron J, Gagne´ C, Couture
P 2004 Lack of evidence for reduced plasma apo B48 catabolism in patients
with heterozygous familial hypercholesterolemia carrying the same null LDL
receptor gene mutation. Atherosclerosis 172:367–373
22. Duez H, Lamarche B, Uffelman K, Valero R, Cohn J, Lewis GF 2006 Hyper-
insulinemia is associated with increased production rate of intestinal apoB-
48-containing lipoproteins in humans. Arterioscler Thromb Vasc Biol
26:1357–1363
23. Reardon MF, Poapst ME, Uffelman KD, Steiner G 1981 Improved method for
quantitation of B apoprotein in plasma lipoproteins by electroimmunoassay.
Clin Chem 27:892– 895
24. Karpe F, Hamsten A, Uffelman K, Steiner G 1996 Apolipoprotein B-48. Meth-
ods Enzymol 263:95–104
25. Taskinen MR 2005 Type 2 diabetes as a lipid disorder. Curr Mol Med 5:297–
308
26. Patel J, Anderson RJ, Rappaport EB 1999 Rosiglitazone monotherapy im-
proves glycaemic control in patients with type 2 diabetes: a twelve-week, ran-
domized, placebo-controlled study. Diabetes Obes Metab 1:165–172
27. Fonseca V, Rosenstock J, Patwardhan R, Salzman A 2000 Effect of metformin
and rosiglitazone combination therapy in patients with type 2 diabetes mel-
litus: a randomized controlled trial. JAMA 283:1695–1702
28. Raskin P, Rendell M, Riddle MC, Dole JF, Freed MI, Rosenstock J 2001 A
randomized trial of rosiglitazone therapy in patients with inadequately con-
trolled insulin-treated type 2 diabetes. Diabetes Care 24:1226 –1232
29. Samaha FK, Szapary PO, Iqbal N, Williams MM, Bloedon LT, Kochar A,
Wolfe ML, Rader DJ 2006 Effects of rosiglitazone on lipids, adipokines, and
inflammatory markers in nondiabetic patients with low high-density lipopro-
tein cholesterol and metabolic syndrome. Arterioscler Thromb Vasc Biol 26:
624 – 630
30. Tan GD, Fielding BA, Currie JM, Humphreys SM, De´sage M, Frayn KN,
Laville M, Vidal H, Karpe F 2005 The effects of rosiglitazone on fatty acid and
triglyceride metabolism in type 2 diabetes. Diabetologia 48:83–95
1728
Duez et al.
Rosiglitazone and Lipoprotein Metabolism
J Clin Endocrinol Metab, May 2008, 93(5):1722–1729
31. van Wijk JP, de Koning EJ, Castro CM, Rabelink TJ 2005 Rosiglitazone
improves postprandial triglyceride and free fatty acid metabolism in type 2
diabetes. Diabetes Care 28:844 – 849
32. James AP, Watts GF, Mamo JC 2005 The effect of metformin and rosiglitazone
on postprandial lipid metabolism in obese insulin-resistant subjects. Diabetes
Obes Metab 7:381–389
33. Sakamoto J, Kimura H, Moriyama S, Odaka H, Momose Y, Sugiyama Y,
Sawada H 2000 Activation of human peroxisome proliferator-activated re-
ceptor (PPAR) subtypes by pioglitazone. Biochem Biophys Res Commun 278:
704 –711
34. Ginsberg HN, Le NA, Gibson JC 1985 Regulation of the production and
catabolism of plasma low density lipoproteins in hypertriglyceridemic sub-
jects. Effect of weight loss. J Clin Invest 75:614 – 623
35. Riches FM, Watts GF, Hua J, Stewart GR, Naoumova RP, Barrett PH 1999
Reduction in visceral adipose tissue is associated with improvement in apoli-
poprotein B-100 metabolism in obese men. J Clin Endocrinol Metab 84:2854 –
2861
36. Alam S, Stolinski M, Pentecost C, Boroujerdi MA, Jones RH, Sonksen PH,
Umpleby AM 2004 The effect of a six-month exercise program on very low-
density lipoprotein apolipoprotein B secretion in type 2 diabetes. J Clin En-
docrinol Metab 89:688 – 694
37. Lewis GF 1997 Fatty acid regulation of very low density lipoprotein produc-
tion. Curr Opin Lipidol 8:146 –153
38. Mayerson AB, Hundal RS, Dufour S, Lebon V, Befroy D, Cline GW, Enocks-
son S, Inzucchi SE, Shulman GI, Petersen KF 2002 The effects of rosiglitazone
on insulin sensitivity, lipolysis, and hepatic and skeletal muscle triglyceride
content in patients with type 2 diabetes. Diabetes 51:797– 802
39. Bandyopadhyay GK, Yu JG, Ofrecio J, Olefsky JM 2006 Increased malonyl-
CoA levels in muscle from obese and type 2 diabetic subjects lead to decreased
fatty acid oxidation and increased lipogenesis; thiazolidinedione treatment
reverses these defects. Diabetes 55:2277–2285
40. Li AC, Glass CK 2004 PPAR- and LXR-dependent pathways controlling lipid
metabolism and the development of atherosclerosis. J Lipid Res 45:2161–2173
41. Zhang YL, Hernandez-Ono A, Siri P, Weisberg S, Conlon D, Graham MJ,
Crooke RM, Huang LS, Ginsberg HN 2006 Aberrant hepatic expression of
PPAR
␥2 stimulates hepatic lipogenesis in a mouse model of obesity, insulin
resistance, dyslipidemia, and hepatic steatosis. J Biol Chem 281:37603–
37615
42. Levy JC, Matthews DR, Hermans MP 1998 Correct homeostasis model as-
sessment (HOMA) evaluation uses the computer program. Diabetes Care 21:
2191–2192
J Clin Endocrinol Metab, May 2008, 93(5):1722–1729
jcem.endojournals.org
1729
Wyszukiwarka
Podobne podstrony:
1722, Opis bibliograficzny w bibliografii załącznikowej
1722
1722
1722
1722
Ведомости 1722 09 29
1722
Ведомости 1722 12 31 2
Hobby 1722; sztućce i pudełko łyżki
więcej podobnych podstron