| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Carbohydrate and Lipid Metabolism Research Unit (M.T.v.d.M., N.J.C., G.P.S., I.H.B., I.P.G., B.I.J.), Departments of Medicine, Nuclear Medicine and Chemical Pathology, University of the Witwatersrand Medical School, Parktown, 2193, Johannesburg, South Africa; and Department of Internal Medicine (P.N.L.), University of Göteborg, Sahlgren’s Hospital, S-41345, Göteborg, Sweden
Address all correspondence and requests for reprints to: M. T. van der Merwe, M.D., Department of Endocrinology, Johannesburg General Hospital, Private Bag X39, Parktown, South Africa 2000. E-mail: 014jhp{at}chiron.wits.ac.za
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Ten BW and 10 WW were matched for the features listed in Table 1
. They were also matched for diet,
duration of obesity (8–10 yr), level of physical and habitual
activity, number of offspring (no more than 3), and socioeconomic
background. None of the women were postmenopausal or on an oral
contraceptive. All the patients gave informed consent, and the study
was approved by the Committee for Research on Human Subjects of the
University of the Witwatersrand.
|
Computed tomography (CT) scan measurements
In the week before their investigation, a five-level CT scan was done (Philips SR 7000, The Netherlands) as previously described (12). The women were in a fasted state. The scan parameters were: 1) 10-mm slice thickness; 2) 120 kilovolts; 3) 250 milliamperes; 4) 2 sec; and 5) 340–480 mm FOV, depending on the size of the subject. Photographic images were taken in resting expiration. The five levels were derived from two scanograms, and they included a CT scan slice at the level of the diaphragm, the umbilicus, L4–5 lumbar disc, the widest diameter of the pelvis and midthigh (the total distance from the iliac crest to acetabulae and from acetabulae to the knee joint, divided by 2). The first scanogram included the diaphragm, the umbilicus (marked with a metallic coin), iliac crest, and symphysis pubis, with a maximum length of 500 mm from the umbilicus. The second scanogram extended from the symphysis pubis to the knees. The widest parasagittal diameter was measured at the level of L4–5. The fat areas were calculated with a Region of Interest seeding program on a Philips Gyroview workstation (fat values were chosen between -30 HU and -130 HU). The areas of the sc and visceral fat were calculated separately, and the anatomical boundaries were as described by Kvist et al. (13, 14). Visceral fat was measured at the top three levels.
Body composition analyses
This was performed, first thing in the morning, with bioelectrical impedance analysis (12). The Bodystat machine (Bodytrach Pty. Ltd., South Africa) was used to assess the body composition for fat and muscle tissue (kg; %). The bioimpedance equations are based on the South African population as a whole.
Microdialysis and blood tests
The patients were studied, in the supine position, in a room kept at 25 C. One polyethylene catheter was inserted iv in the forearm and placed under a heating pad to ensure sampling of arterialized venous blood (15). At the onset of the study, we measured fasting plasma glucose, insulin, FFA, glycerol, and lactate levels. Two microdialysis catheters (0.4 x 30 mm, F4HPS, 5000-MW cut off, Fresenius, Germany) were inserted into sc adipose tissue, 5 cm lateral to the umbilicus and at the level of the midfemur. The insertion was done through a fine cannula (Jelco 16SDD, Criticon, S.A.) (16), and the inlet nylon tubing of the microdialysis catheter was connected to a precision pump (CMA/100, Sweden). No local anesthetic was required. For initial equilibration, the microdialysis catheters were perfused for 60 min, at a rate of 2.5 µL/min, with an isotonic saline solution containing 2.5 mmol/L glucose, 25 µmol/L glycerol, and 250 µmol/L lactate to prevent depletion of tissue concentrations (16).
In vivo calibration for the microdialysis catheters
The equilibration calibration technique has been described in detail before (16). Two 10-min samples were collected for each perfusion concentration; and post oral glucose tolerance test (post-OGTT), the samples were collected at 15-min intervals at the level of the final concentration. The net change for the glycerol and lactate concentrations in the outgoing dialysate were calculated. The linear relationship between the concentration of glycerol and lactate in the ingoing perfusate and the net change of these concentrations in the outgoing dialysate could then be used to calculate the absolute interstitial concentrations for these substances, by means of regression analyses, by determining the point of no net influx of substrate into perfusate (16).
OGTT
A standard 75-g OGTT was performed at approximately 1400 h. Arterialized postprandial blood samples were taken at 4 time points over the next 2 h.
sc blood flow
Blood flow was measured to determine lactate and glycerol
release/100 g adipose tissue·min. An injection of 3–6 mol/L
Bq of 133 Xenon in 0.1 mL sterile saline was given 5
mm under the skin in the sc adipose tissue, contralateral to the two
microdialysis sites. After 60 min, xenon clearance was monitored at 20
frames/min for a period of 180 min, of which 120 min were post OGTT.
The energy window was set at 81 keV ± 10%. Initial counts/site
were approximately 10,000 in the 3-min missing period, to enable
reliable measurements for 3–4 h. The Elscint 415 ECT 
camera was
used for the acquisition of the data, with the detector placed at
30–50 cm from the 133Xenon depot (frame size: 128 x
128 pixels; zoom factor: 1; collimator: 3). The counts were plotted on
a logarithmic-linear diagram, as a function of time. Adipose tissue
blood flow (ATBF) was then calculated with the equation: ATBF =
·k·100, where is the tissue-to-blood partition coefficient
for xenon at equilibrium, and k is the slope of the washout curve.
Experimental values of k were estimated by performing least-squares
regression analysis. A mean value of 10.0 mL/g was used for the
partition coefficient (17). In addition, this value of 10.0 mL/g was
chosen for both the abdominal and femoral areas in both groups of women
(18), because no difference in cell size was documented during in
vitro cell studies in these four areas (submitted data). Control
experiments have previously shown a 24% mean relative error of
133Xe clearance, measured at two corresponding sites
(17).
Biochemical analyses
Glucose was measured using an enzymatic colorimetric method
(GOD-PAP, Boehringer Mannheim, Mannheim, Germany), and insulin levels
using an enzyme-amplified sensitivity immunoassay (EASIA-Medgenix
Diagnostics, Belgium; normal range 13.8–172.5 pmol/L). FFA
levels were measured using the Half-micro test (Boehringer
Mannheim). Fasting bloods were also analyzed for progresterone
(Amerlite Progesterone Assay, Amersham, United
Kingdom; enhanced luminescence, with normal range 5.84–96.0 nmol/L in
luteal phase), free-testosterone (Enzymun-Test, Boehringer Mannheim
Immunodiagnostics; enzyme-linked immunosorbent assay, with normal range
0–2.1 nmol/L), and sex hormone-binding globulin (SHBG:
125I, Farmos Diagnostica, Finland; immunoradiometric
assay kit with normal range 20–120 nmol/L). In addition, we assayed
epinephrine, norepinephrine, and dopamine levels in the postabsorptive
state. These samples were taken 1 h after the insertion of the iv
cannula, when the subjects were recumbent and relaxed. Samples were
placed on ice instantly and were centrifuged within 10 min
(high-performance liquid chromotography, with electrochemical
detection, was used for analyses; and normal ranges are as follows:
NE, 100–800 pg/L; E, 10–80 pg/L; Dopamine, 10–150
pg/L. The glycerol content in adipose tissue dialysate and plasma were
determined using a radiometric phosphorous method (19). Glycerokinase
was obtained from Boehringer Mannheim, and (32p)ATP was
from New England Nuclear (Boston, MA). Plasma lactate was analyzed
within 14 days, according to a fluorometric method described by Loomis
(20). The specimens were centrifuged within the hour, and the serum was
stored at -70 C until analyzed in a single batch to avoid interassay
variation.
Estimation of glycerol and lactate release
According to Fick’s principle, arterialized venous plasma (A)
and capillary venous plasma concentrations of lactate and glycerol (V)
and plasma flow rate (Q) were entered into the equation: (V-A) x
Q x (1-Hematocrit x 10-2) (µmol/100
g·min). Conversion of interstitial (I) to venous (V) lactate and
glycerol levels were performed by the equation: V = I-A x
(1-e -PS/Q) + A, where PS is the permeability surface
product area (adopted to be 
5 mL/100 g/min for glycerol and
lactate) (21).
Statistical analysis
Data are expressed as means ± SEM. Significance for multiple samples was established by one-way ANOVA. Statistical significance of parametric data was assessed by Student’s paired and unpaired t test. Nonparametric data (e.g. blood flow, tissue levels, and release values) were assessed by signed rank test for paired samples. Absolute interstitial levels were calculated by linear regression analysis with the least-squares method, and linear correlations were tested with Pearsons correlation coefficient. Results were deemed significant at P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
)The women were preselected so as to have no significant differences in their clinical, anthropometric, or body composition measurements. However, the WW have substantially more visceral fat, compared with the BW, as measured by CT scan (152 ± 2.0 vs. 110 ± 5.0 cm2; P < 0.05).
Postabsorptive biochemical measurements (Table 2)
The BW had higher mean postabsorptive plasma FFA and glycerol levels and lower lactate and plasma insulin. The remaining fasting biochemical data showed no significant differences between the two groups of women.
|
During the 75-g OGTT, the glucose levels in the BW were higher at
1 h (8.8 ± 0.8 vs. 5.3 ± 0.4 mmol/L;
P < 0.005) and 2 h (7.6 ± 0.5
vs. 4.4 ± 0.3 mmol/L; P < 0.002) but
not at fasting (4.7 ± 0.2 vs. 4.5 ± 0.3 mmol/L)
(Fig. 1). The insulin values were lower
in the BW at 1 h (328 ± 56 vs. 630 ± 75
pmol/L; P < 0.005) but not at 2 h [344 ±
50 vs. 340 ± 36 pmol/L; P = NS (not
significant)] (Fig. 1). C-peptide, with its longer half-life (compared
with insulin), remained lower in the BW at fasting (420 ± 97
vs. 1700 ± 100 pmol/L; P < 0.0001),
at 1 h (1375 ± 450 vs. 6001 ± 333 pmol/L;
P < 0.0001), and at 2 h (1270 ± 340
vs. 4170 ± 366 pmol/L; P < 0.0001).
At 30 min post OGTT, there was a correlation between the lower visceral
fat mass and insulin levels in the BW (r = 0.6; P
< 0.01). The postprandial FFA, glycerol, and lactate levels are shown
in Fig. 2. By 120 min, the FFA levels had
decreased to a level of 321 µmol/L in the BW (P <
0.005) and to a level of 141 µmol/L in the WW (P <
0.0001). The plasma glycerols decreased from 79.3 to 35.2 µmol/L in
BW (P < 0.005) and from 60.4 to 26.3 µmol/L in the
WW (P < 0.05) (Fig. 2). By contrast, the plasma
lactate levels increased significantly in the BW only, and this
increase was already significant by 1 h after the onset of the
OGTT (1918 µmol/L) (Fig. 2). By 2 h post-OGTT, a 42% increase
in plasma lactate levels was still present in the BW (1768
µmol/L).
|
|
)The WW began with a significantly higher blood flow rate in the postabsorptive state in both the abdominal area and the femoral area. However, the WW showed no vasodilatory response to the glucose load. By contrast, the BW showed a significant vasodilatory effect on the glucose load, and they increased their blood flow in both regions, but more so in the femoral region.
|
During the postabsorptive state, the interstitial glycerol levels
were higher in the BW, compared with WW, for both the abdominal region
(183 ± 6.4 vs. 155 ± 4.4 µmol/L;
P < 0.002) and femoral region (188 ± 8.6
vs. 155 ± 4.1 µmol/L; P < 0.002)
(Fig. 3). The higher lipolytic rate in
the BW vs. WW was further supported by the higher glycerol
release (µmol/kg·min) in the abdominal area (1.15 ± 0.17
vs. 0.65 ± 0.03; P < 0.01), as well
as femoral area (1.55 ± 0.19 vs. 0.72 ± 0.05;
P < 0.001). In addition, we noted a regional
difference in the BW, with a higher lipolytic rate observed in the
femoral, compared with the abdominal area (P < 0.009).
During the postprandial period, the interstitial glycerol levels and
glycerol release remained higher in the BW for a period of 2 h
(Fig. 3). By 60 min, the regional difference first observed in the BW
was no longer apparent. After the ingestion of 75 g glucose, an
increase in interstitial lactate levels was seen, after 30 min in the
BW, reaching a maximum concentration after 90 min (P <
0.007; Wilcoxon) (Fig. 4).
Although the interstitial lactate level increased in the WW post OGTT,
it did not reach statistical significance (Fig. 4). Although no
regional differences were observed in either of the groups, the
interstitial lactate levels were significantly higher than the plasma
levels in both groups, but more so in the BW. During the postabsorptive
state, the interstitial-arterial difference was higher in the BW in
both the abdominal area (1580 ± 176 µmol/L) and femoral area
(1520 ± 175 µmol/L). After 30 min post OGTT, there was a rapid
drop in these differences in the BW (Fig. 5). The interstitial-arterial differences
were also higher in the BW, compared with WW (in whom there was no
significant change during the postprandial period, compared with the
postabsorptive state) (Fig. 5). There was a higher sc adipose tissue
lactate release in the BW, compared with the WW, in the postabsorptive
state (abdominal: 7.8 ± 0.9 vs. 2.4 ± 0.3
µmol/kg·min, P < 0.0001; femoral: 9.1 ± 0.9
vs. 2.1 ± 0.3 µmol/kg·min, P <
0.0001) (Fig 6). The higher lactate release in the BW
was maintained in the postprandial period for 2 h; and by 120 min,
the lactate release was higher in the femoral area, compared with the
abdominal area, in the BW. (Fig. 6).
|
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
In South Africa, it has previously been shown that the BW have less visceral fat than WW for a comparable BMI (12). This was confirmed during this study. In addition, we found a correlation between the lower visceral fat mass and insulin levels in the BW at 30 min post-OGTT. It is known that visceral fat mass, measured by CT scanning, accounts for a significantly greater percentage variance of the cardiovascular risk profile when compared with total body fat mass (22). Furthermore, visceral fat mass also has a significant correlation with both higher fasting and postprandial insulin levels (23, 24). The higher incidence of ischemic heart disease amongst obese WW may well be associated with their larger amount of visceral fat, and perhaps with their greater degree of hyperinsulinemia (25). Furthermore, several investigators have shown that FFA, in the presence of hypoinsulinemia, will lead to a lower cardiovascular disease morbidity when compared with the combination of elevated FFA and hyperinsulinemia (26, 27). This may be partly responsible for the lower incidence of ischemic heart disease in our BW. The more androgenic profile that can be associated with visceral obesity in obese women (28, 29) was not apparent from the sex-steroid levels or SHBG levels documented in our WW. However, this does not exclude the possibility of ethnic differences in the regional tissue distribution or number of sex-steroid receptors (5, 30). Furthermore, it should also be noted that the WHR amongst the subjects presently investigated was not particularly high.
The relative insulinopenic state of the obese BW might reflect a reduction in the ß-cell mass previously observed in black Southern Africans (31, 32). In support of our findings, Osei et al. (33) have recently reported that the postglucose serum insulin concentrations are significantly greater in the black Americans when compared with native Nigerians. A putative reduction in the ß-cell mass could be linked to transient periods of protein-energy malnutrition during infancy or in utero (34, 35). These observations become even more important in light of recent evidence that differentiated human islet cells have only very limited proliferative capacity (36). In the presence of a higher secretion of insulin, the WW could dispose of their glucose load more effectively.
The higher fasting plasma FFA levels in the BW are most likely the result of the more active lipolysis caused by a reduction of the antilipolytic effect of insulin. This is further supported by the higher fasting and postprandial plasma glycerol levels observed in the BW, as well as their higher interstitial sc adipose tissue glycerol concentrations and glycerol release. Accordingly, Joffe et al. (37) have previously documented markedly elevated FFA levels in patients with destruction of their ß-cell mass caused by chronic pancreatitis. There is also the possibility of an acquired tissue insulin resistance, secondary to chronically elevated FFA levels. Systemically elevated FFA levels may induce insulin resistance in both muscle and liver (38) by increasing the flux of fructose-6-phosphate into the hexosamine pathways (39), as well as by decreasing insulin-stimulated glucose uptake by reducing the activity of protein kinases that act downstream from the insulin receptor, particularly mitogen-activated-protein kinase (40). Recent studies performed by Buthelezi et al.2 on the obese BW and WW, indicated a higher isoproterenol-induced lipolytic activity in BW, even in the presence of a maximum concentration of insulin. They also documented a greater degree of insulin resistance in the BW during euglycemic hyperinsulinemic clamp studies1. Thus, it is possible that the obese BW have acquired insulin resistance at both the level of the adipocyte and muscle tissue. The effect of FFA, to cause diminished ß-cell glucose oxidation, is largely exerted through inhibition of pyruvate dehydrogenase enzyme activity, and this is mediated through an increase in pyruvate dehydrogenase kinase activity (41). Furthermore, recent data suggested that chronically elevated FFA levels will cause an additional longitudinal impairment of ß-cell function in humans (42, 43).
Though the higher rate of ATBF in the BW during the postprandial period may have enhanced their glycerol release, it did not explain the regional differences in glycerol release observed in the BW. The higher femoral blood flow rate became significantly apparent by 1 h post-OGTT but was not present during the postabsorptive state. It was during the postabsorptive period, and for up to 1 h postprandially, that the higher glycerol release in the gluteo-femoral area of the BW was significant. One possible explanation for this regional difference in adipose tissue glycerol release in the BW could be a difference in the insulin receptor activity (26, 27).
During the postprandial period, the brisker ATBF observed in the BW may have been influenced by the higher degree of sc lactate release induced by the glucose load (44, 45). In the WW, the oral glucose load was not sufficient to induce an increased blood flow, and the rate remained unchanged. The reason for the higher gluteo-femoral ATBF in the BW by 1 h post-OGTT, as compared with the abdominal area, was not clear from the present study and will need some further investigation. It is known that insulin, apart from its effects on glucose uptake and metabolism, also stimulates limb blood flow through vasodilation (46), and that the effect may be diminished in obesity. The fact that the BW had a lower ATBF, compared with the WW before oral glucose, but not after the oral glucose load, suggests that glucose metabolism and lactate release may have been a more important determinant of post-OGTT ATBF than the interstitial insulin level.
In the postabsorptive state, glucose enters the blood almost exclusively from the liver. Some of this glucose arises from glycogen breakdown, and some from gluconeogenesis, of which lactate will constitute approximately 25 mg/min (47). The stimulus for this gluconeogenesis would be mainly a decreased insulin/glucagon ratio. Measurements done with microdialysis on sc adipose tissue have shown that adipose tissue is a significant source of lactate production (48) and that it is proportional to the adipose tissue mass of obese subjects (44). One of the implications of the adipose tissue release of lactate, not only during fasting but also after a glucose load (48), is that adipose tissue may use more glucose in the postprandial state than originally proposed (49). This may be particularly true in BW, in whom we have demonstrated a higher adipose tissue lactate release and higher interstitial levels, compared with the WW. Osei et al. (50, 51) have previously demonstrated that although the insulin sensitivity index was reduced in West African Ghanaians with IGT, glucose-dependent-glucose disposal (Sg) remained intact to serve as a compensatory mechanism for hyperglycemia. They believed this preservation of Sg to be ethnicity dependent. It may be through this preservation of Sg that an increased amount of lactate is released, both during fasting and after the glucose load. Henry et al. (52) recently demonstrated that either hyperinsulinemia or hyperglycemia will stimulate glucose conversion into lactate in adipocytes. Indirect evidence also exists to suggest that an enhanced lactate release from the adipose tissue in obese subjects in the postabsorptive state would be related to insulin-resistance (53). We have reason to believe that this is the case, because we have shown the BW to be more resistant in vivo than the WW2. The higher ATBF in the WW during the postabsorptive state makes the higher lactate release in the BW during fasting even more significant. The present study does not allow for an exact estimate of total body lactate appearance. However, based on previous lactate appearance studies (54), and assuming that the mean lactate release estimated over the 2-h post-OGTT persisted for 4 h, and that visceral fat was at least as efficient in lactate release, we calculated that the total body fat would convert approximately 5.5% of the ingested glucose to lactate in the BW vs. the 2.2% converted in WW. This may have important metabolic implications, because lactate is an important gluconeogenic precursor during fasting (55).
Finally, the decrease in the interstitial-arterial lactate difference in the BW during the postprandial period indicates an additional source of lactate production, other than sc adipose tissue. This source is most likely to be muscle (in insulin-resistant states, the amount of glucose stored as glycogen will be reduced, and the amount released as lactate and alanine will be increased (56).
In summary, we have demonstrated both a higher glycerol and lactate release from the sc adipose tissue of our relatively insulinopenic urban BW. Thus, although the concept of a so-called benign obesity may be feasible for African female adolescents, this does not seem to be the case for the older urban BW, in whom obesity may play an important role in the pathogenesis of diabetes.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Buthelezi EP, van der Merwe M-T, Lönnroth
PN, Gray IP, Crowther NJ. Ethnic differences in the sensitivity of
adipocyte lipolytic activity to insulin.
Received March 27, 1998.
Revised July 9, 1998.
Accepted July 29, 1998.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  K. Danadian, V. Lewy, J. J. Janosky, and S. Arslanian Lipolysis in African-American Children: Is It a Metabolic Risk Factor Predisposing to Obesity? J. Clin. Endocrinol. Metab., July 1, 2001; 86(7): 3022 - 3026. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.-T. van der Merwe, G. P. Schlaphoff, N. J. Crowther, I. H. Boyd, I. P. Gray, B. I. Joffe, and P. N. Lonnroth Lactate and Glycerol Release from Adipose Tissue in Lean, Obese, and Diabetic Women from South Africa J. Clin. Endocrinol. Metab., July 1, 2001; 86(7): 3296 - 3303. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Punyadeera, M-T. van der Merwe, N. J. Crowther, M. Toman, G. P. Schlaphoff, and I. P. Gray Ethnic differences in lipid metabolism in two groups of obese South African women J. Lipid Res., May 1, 2001; 42(5): 760 - 767. [Abstract] [Full Text]
|
|
|
|
 |
|
 P. Rheeder, R.P. Stolk, and D.E. Grobbee Ethnic differences in C-peptide levels and anti-GAD antibodies in South African patients with diabetic ketoacidosis QJM, January 1, 2001; 94(1): 39 - 43. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
,
P < 0.0001).
, BW; 
, WW; Student’s unpaired
t test: #, P < 0.05; ##,
P < 0.005. ANOVA, compared with time 0: *,
P < 0.05; **, P < 0.005; ***,
P < 0.0001.
, WW); *, P < 0.002; **,
P < 0.001; ***, P < 0.0001.
Release (right panel): femoral area (
, BW; 
, WW); #, P < 0.01;
##, P < 0.001.
, plasma lactate; •,
abdominal; 
, plasma
lactate; 
