- Email: [email protected]
Effect of photoperiod on body weight and food intake of obese and lean Zucker rats
Life Sciences, Vol. 49, pp. 735-745 Printed in the U.S.A.
Pergamon Press
EFFECT OF PHOTOPERIOD ON BODY WEIGHT AND FOOD INTAKE OF OBESE AND LEAN ZUCKER RATS.
Lisa M. Larkin 1, B. J. Moore 1'3'4 Judith S. Stern 1-3, and Barbara A. Horwitz3.
Departments of Nutrition 1, Internal Medicine 2 and Animal Physiology 3, and The Food Intake Laboratory, University of California, Davis 95616; Weight Watchers International 4, 500 N. Broadway, Jericho, NY. 11753 (Received in final form July 8, 1991)
Summary Although the rat is usually not considered to be sensitive to photoperiod, under some experimental conditions photoperiod responses are unmasked. In addition, we have observed photoperiod-induced changes in body weight gain in lean and obese Zucker rats. In this experiment, body mass, food intake, body composition, brown adipose tissue (BAT) thermogenic state, and blood concentrations of corticosterone, insulin, and glucose were evaluated under one of two lighting conditions: a short (10 h light: 14 h dark) or a long (14 h light: 10 h dark) photoperiod. Plasma corticosterone and glucose concentrations measured under fasting conditions were unaffected by photoperiod in either genotype. The amount of BAT mitochondrial protein isolated was less in long photoperiod rats. BAT mitochondrial GDP binding was unaffected by photoperiod in the lean rats, but tended to be lower in long photoperiod obese rats than in short photoperiod obese rats. Although, photoperiod had no effect on daily food intake of rats exposed to the short versus long photoperiod, body mass was heaviest in obese rats raised in long photoperiod. Plasma insulin was increased in both lean and obese rats in long photoperiod. In addition, fat storage appeared to shift to internal depots in the lean rats exposed to long photoperiod. Our data demonstrate that photoperiod does have an effect on male Zucker rats with respect to body weight and fat distribution, with the obese rats being more sensitive to changes in photoperiod than the lean rats. 0024-3205/91 $3.00 + .00 Copyright (c) 1991 Pergamon Press plc
736
Photoperiod and Obesity
Vol. 49, No. i0, 1991
Photoperiod modifies fat accumulation, food intake, body composition, and body weight gain in a variety of mammals. These effects are pronounced in rodents such as Syrian hamsters and voles, whose reproductive cycle is sensitive to photoperiod (1). Rats are generally not thought to be as sensitive to changes in photoperiod as are the more "wild-type" rodents even though rats breed more successfully when placed in long photoperiod. Moreover, under some experimental conditions such as bulbectomy and pinealectomy, photoperiod responses are unmasked in rats (2,3). This study evaluated the effects of long versus short photoperiod on weight gain, body composition, and food intake as well as on several hormonal variables in Zucker lean and obese male rats. A pilot study from our laboratory indicated that pinealectomy increased weight gain without a change in food intake of lean but not obese Zucker rats. In addition, obese Zucker rats (fa/fa) lack the normal diurnal feeding pattern and periodicity in plasma corticosterone levels seen in most lean rats (4), adrenalectomy restores the feeding patterns to those observed in lean rats, and corticosterone replacement restores the patterns to those of sham-operated obese rats (5). These observations suggest that photoperiod may have a greater effect on the lean than on the obese rat. The present study negates this hypothesis. Materials and Methods Animals and Animal Care Five week old male Zucker obese (fa/fa) and lean (Fa/?) rats, were obtained from Charles River Laboratories, Inc. (Raleigh, NC). Three days after arrival, the animals were assigned to one of two photoperiods, either short (10 hrs light:14 hrs dark) or long (14 hrs light:10 hrs dark). The animals were separated to obtain two groups of obese rats and two groups of lean rats with equivalent initial body weights. The rats were allowed to adapt to the new photoperiod for four weeks before beginning the experiment. All animals were individually housed at 24-26°C in wire-bottom hanging cages and fed Simonsen's White diet (Gilroy, CA) and water ad libitum. Experimental Design From 9 to 14 wks of age, 24 hr food intake and body weight were measured 3 times a week. Prior to 14 wks of age, all rats were conditioned to the guillotine to reduce stress during sacrifice. At 14 wks of age, food was removed one h before lights off, and the rats were fasted overnight. Rats were sacrificed via decapitation the next morning, one h after lights on (lights on at 0600 for the long photoperiod group and 0800 for the short photoperiod group). At the time of sacrifice, trunk blood was collected; cervical, axillary, and scapular brown adipose tissue were removed and kept on ice for approximately 10 min prior to homogenization; white adipose tissue (epididymal, retroperitoneal, and mesenteric) was dissected, weighed, and replaced in the eviscerated carcass for determination of carcass composition.
Vol. 49, No. i0, 1991
Photoperlod and Obesity
737
GDP Binding Analysis The binding of GDP to freshly isolated brown fat mitochondria was determined as previously described by McDonald et al. (6), with slight modification of the mitochondrial isolation procedure. Minced BAT was placed in approximately 50 ml of isolation buffer (pH 7.2) composed of 0.25 M sucrose, 1 mM n-2-hydroxyethylpiperazine-N's-2ethanesulfonic acid (HEPES), and 0.2 mM ethylenediamine-tetraacetate (EDTA). This suspension was homogenized in a Kontes Duall glass-glass tissue grinder and centrifuged at 1200 x g for 10 min at 5°C (Sorvall RC-5B centrifuge). The supernatant was decanted and saved, and the pellet was resuspended in isolation buffer and centrifuged at 1200 x g for 10 min. The supernatants from both centrifugations were each spun at 20,000 x g for 14 min, and the pellets were resuspended in more isolation buffer, combined, and recentrifuged for 14 min at 16,318 x g. The resulting pellet was resuspended in reaction buffer and used in the GDP binding assay. In our hands, this isolation procedure routinely yields 30-35% recovery, with no differences between obese and lean rats. Net GDP bound was determined after correction for trapped extramitochondrial fluid. Specific binding was calculated by subtracting non-specific binding (i.e., binding in the presence of 100 uM ADP) from total binding (binding in the absence of ADP). Protein was determined by the method of Lowry et al. (7) using bovine serum albumin as a standard. Blood Analysis Trunk blood was collected in heparinized tubes, spun at 2000 x g for 20 min, and the resulting plasma was stored at -70°C for subsequent analysis of glucose, insulin, and corticosterone. Glucose was analyzed by the method of glucose oxidase (Beckman Glucose Analyzer 2, Beckman Instruments, Inc., Fullerton, CA). Insulin was analyzed using a modification of the single antibody/polyethylene glycol method of Desbuguois and Aurbach (8). Phosphate buffer (0.05 M) containing 0.4% human serum albumin (pH 7.4) (Cutter Biological, Berkely, CA), rat insulin standard (21.3 U/mg; Novo Biolabs, Wilton, CT), insulin antisera (porcine; ICN Diagnostics Division, Costa Mesa, CA), [1251]insulin (Amersham Corp., Arlington Heights, IL), and polyethlene glycol (MW 8,000; Sigma Chemical Co., St. Louis, MO) were used as previously described (9). Corticosterone was assayed using a [1251]-corticosterone radioimmunoassay for rats (ICN Diagnostics Division). Carcass Composition Following removal of brown adipose tissue, three white fat depots (epididymal, retroperitoneal, and mesenteric) were excised and weighed. The rat was then eviscerated and the white fat depots were returned to the carcass for subsequent carcass composition analysis using the method of Bell and Stern (10).
738
Photoperiod and Obesity
Vol. 49, No. i0, 1991
Statistics Analysis of variance (ANOVA) was used to evaluate data using a factorial 2 x 2 design (photoperiod versus genotype), except for the analysis of blood data where a one way ANOVA was used because the variances between genotypes differed. Differences were considered to be significant at p < 0.05. When the ANOVA was significant, individual means were evaluated using the Newman Keurs post hoc test. Results Body Weight Initial body weights at five wks of age were: long photoperiod lean rats (LL), 94.0 + 3.8 g; short photoperiod lean rat (SL), 93.4 + 3.7 g; long photoperiod obese rats (LO), 144.0 + 3.4 g; and short photoperiod obese rats (SO), 144.0 + 3.4 g. After four wks adaptation to the experimental photoperiods, the LO rats were heavier than the SO animals (Fig. 1). This pattern continued throughout the remainder of the experiment (p < 0.001) (Fig. 1).
600 m E
2
Ot
LO(N:10) SO(N=IO)
~
500 400
IO) "6
300
~ "0 0 ¢)
200
100
0
\
~ \~I 5
\
\
LL (N=IO) SL(N=9)
~
I
10
,
!
11
,
I
12
,
I
13
,
I
14
Week of Experiment
Fig. 1 Weekly body weights in lean and obese male rats in either short or long photoperiod. Values are means + SEM; SEM bars are within the mean symbol. *LO values differ significantly (p < 0.05) from SO values.
Vol. 49, No. I0, 1991
Photoperiod and Obesity
739
Food Intake Food intake measurements began after the rats had been in their respective photoperiods for a total of 4 wks. Although there was no significant effect of photoperiod on daily food consumption, there was an effect of genotype (Fig. 2; p < 0.0001). During the first two weeks of measurements, the obese animals ate 40% more per day than did their lean counterparts. This difference then decreased, becoming nonsignificant by .the end of the measurement period (Fig. 2). Photoperiod differentially affected the lean versus obese rats. In long photoperiod, the obese rats consistently ate more than did the lean rats, but daily differences never reached significance (p=0.07) (Fig. 2). However, total food consumed during the 5 wks of this experiment was significantly greater in the obese vs lean rats, and there was an interaction of genotype with photoperiod: LL, 293.9 + 72.8 g; SL 324.9 _+39.1 g; LO, 404.9 + 43.4 g; SO, 359.1 + 37.9 g. This interaction was not significant when total food consumption was expressed in a body massindependent fashion: LL, 6.2 + 0.6 g/kg°67; SL, 6.5 + 0.8 g/kg°67; LO, 5.8 + 0.6 g/kg°67; SO, 5.5 + 0.6 g/kg °'67.
40
>, (u "o oi
-
30 ~Lr
.......
4)
m r.
.~...-
. . . - 4-r ~
==
.....
. .......
=
'~'" ,~
" ""
......
"''O
--~
&
20
LO (N:10)
SO (N=10) SL (N=9)
LL (N=IO)
i
"0 0 0
10
0
"\ \
I
,
I
10
i
I
11
,
12
Week
of
I
13
*
I
14
Experiment
Fig. 2 Daily food intake in lean and obese male rats in either short or long photoperiod. Kcal/gram = 3.00. Values are means + SEM; SEM bars are ~r
within the mean symbol. values.
LO values differ significantly (p < 0.05) from SO
740
Photoperiod and Obesity
Vol. 49, No. i0, 1991
Carcass Composition Analysis Genotype significantly altered carcass composition, with the obese animals having heavier carcass weights, higher percent carcass fat, but lower percent water (p < 0.0001) than their lean counterparts. However, lean body mass, protein, and ash of the LO animals did not significantly differ from that in LL animals (Table I). The obese animals were sensitive to the different photoperiods, with percent water and grams of fat being the only variables unaffected by photoperiod. The LO animals had greater carcass weight, lean body mass, protein, ash, water, and percent body fat than did the SO rats. Although lean rats showed similar trends, the differences due to photoperiod did not reach significance except for carcass protein which was higher in tong photoperiod animals (Table I).
TABLE I Carcass Composition
LEAN Photoperiod n
Long 9
OBESE Short 10
Long 10
Short 10
Body mass, g
369.2_+13.1 a
352.2+ 8.5 a
575.4+ 9.2 b
523.3-+12.5 c
Carcass mass, g
304.2+ 9.5 a
288.8+ 9.0 a
485.4-+ 8.3 b
436.7+11.9 c
LBM, g
272.3+ 5.4 a
252.2+ 9.8 a
269.3+ 6.3 a
223.0-+ 7.6 b
Protein, g
262.1+ 5.5 a
236.2+ 8.6 b
258.4+ 6.1a
214.1-+ 7.7 c
Ash, g Fat,
12.0+ 0.27 a
12.5+ 0.92 a
39.9+ 2.4 a
34.0+ 2.9 a
216.1+ 5.4 b
213.8+ 8.4 b
13.8+ 0.7 a
11.9+ 1.0 a
44.5+ 0.9 b
48.9+ 1.2 c
g
188.1+ 4.0 a
173.3+ 7.5 a
156.6_+4.1 b
137.6+ 5.6 C
%
60.2_+ 0.5 a
60.5_+ 0.8 a
32.3_+0.7 b
31.5_+ 1.0 b
69.2_+ 0.2 a
68.6_+ 0.4 a
58.2+ 1.0 b
61.7_+ 1.1 c
g % Water,
Water/LBM x 100
10.9+ 0.29 a
8.9+ 0.12 b
Values are means + SEM. LBM, lean body mass; n, number of animals in groups. Values with different superscripts are significantly different from each other at p < 0.05.
Vol. 49, No. I0, 1991
Photoperiod and Obesity
741
Fat Distribution When fat deposition was expressed in terms of percent of carcass mass, there was no significant difference due to photoperiod in the individual depots (epididymal, retroperitoneal, mesenteric) or in the total of all three depots (Table II). However, there was a significant difference (p < 0.0001) due to genotype, with the genetically obese rats having more fat in all three depots. Internal fat distribution was estimated by dividing the sum of the mass of the mesenteric, retroperitoneal, and epididymal depots by total carcass fat: the higher the ratio, the more the internal distribution of body fat. The LL rats had relatively more internal fat (ratio = 0.35) than did the SL (ratio = 0.29) or SO rats (ratio = 0.26) (Table II). Although the LO rats tended to have more internal fat (ratio = 0.28) than the SO rats, this difference was not statistically significant.
TABLE II White Adipose Tissue Mass (grams)
LEAN Photoperiod n
Long 9
OBESE Sho~ 10
Long 10
Sho~ 10
EPI
5.65+0.37 a
4.24+0.45 a
18.40+0.68 b
16.27+0.77 c
RP
4.23+0.39 a
3.44+0.32 a
26.78+1.33 b
24.23+1.55 b
MES
3.37+0.34 a
2.51+_0.23a
14.95+0.75 b
13.99+_O.45b
TOTAL
13.25+1.06 a
10.18+0.96 a
60.14+2.31b
54.49+2.24 c
TOTAL
3.69_+O.13a
3.52+0.09 a
5.75_+0.09b
5.23_+0.12c
0.35+0.1 la
0.29+0.01b
0.28_+0.01bc
0.26+0.01c
100g BW EPI+RP+MES CARC
Values are means + SEM. EPI, epididymal; RP, retroperitoneal; MES, mesenteric; EPI+RP+MES=TOTAL; CARC, carcass fat; n = number of rats/group. Values with different superscripts are significantly different from each other at p < 0.05.
742
Photoperiod and Obesity
Vol. 49, No. i0, 1991
TABLE III Plasma Corticosterone, Glucose, and Insulin.
LEAN
Corticosterone (nmol/L) Glucose (mmol/L) Insulin (pmol/L)
OBESE
Long
Short
Long
(4) 150.6+53.5
(4) 203.6+75.9
(4) 310.4+237.3
Short
(5) 206.9_+65.7
(9)
(10)
(10)
(10)
7.1 + 0.3
6.6 + 0.2
6.9 + 0.3
7.1 + 0.3
(9)
(9)
(10)
(9)
140 + 12 a
1542 + 174
239 + 22
988 + 151a
Values are means + SEM. The number of rats per group is given in parenthesis. Values with different superscripts indicate that short photoperiod values are significantly different from long photoperiod values in the same genotype. The (n) is smaller for corticosterone because of technical difficulties.
Plasma Insulin, Glucose, and Corticosterone There were no significant differences in plasma levels of glucose and corticosterone with either photoperiod or genotype (Table III). However, as expected, plasma insulin was significantly elevated in obese vs lean rats. Moreover, the one way ANOVA performed separately on lean and obese genotypes demonstrated that long photoperiod rats had significantly higher plasma insulin levels than did the short photoperiod rats (p < 0.03 for obese rats and p < 0.001 for lean rats). Brown Adipose Tissue and GDP Binding SO animals had 12.6% less BAT mass than did the LO rats (Table IV). There was no such difference in the lean animals. When BAT mass was expressed in terms of kg body mass, differences between LO and SO rats disappeared, but the SL rats had higher values than did the LL group. Although BAT mass was greater in the obese vs lean animals, the obese brown fat depots were very pale, indicating large amounts of stored triglycerides. A better indication of differences in BAT oxidative function is mitochondrial protein and GDP binding (Table IV).
Vol. 49. No. I0, 1991
Photoperlod and Obesity
743
TABLE IV Brown Adipose Tissue Mass and GDP Binding OBESE
LEAN Photoperiod n
Long
Short
Long
Short
9
10
10
10
BAT mass g
0.88+0.06 a
1.10_+0.06 a
5.76+0.28 b
5.03+0.25 c
g/kg BW
2.37_+0.09a
3.11+0.13 b
10.0+0.32 c
9.66+0.37 c
mitochondrial protein (mg)
12.6+0.7 a
17.8+1.0 b
3.7_+0.9 c
GDP binding (nmol/mg mito prot)
0.15+0.01
0.12_+0.02
Recovered
0.08_+0.02
7.0_+0.8 d
0.11+0.03
Values are means + SEM. BAT, brown adipose tissue; GDP, guanosine 5'diphosphate; BW, body weight; n, number of animals in group. In a given row, values with different letter superscripts differ significantly (p < 0.05).
The amount of mitochondrial protein recovered showed a significant effect of photoperiod (p < 0.0001) and of genotype (p < 0.0001). Short photoperiod animals had more recoverable BAT mitochondrial protein than did long photoperiod rats, and the lean rats had more than their obese counterparts. The specific activity of GDP binding (nmol/mg mitochondrial protein) was not altered by photoperiod or genotype (Table IV).
Discussion Rats are generally not thought to be as sensitive to photoperiod as the more "wildtype" rodents such as Syrian hamsters and voles (1). Nonetheless, under some experimental conditions such as bulbectomy and pinealectomy, photoperiod responses are unmasked in rats (2,3). Studies looking at the effects of photoperiod on metabolic variables in Syrian hamsters have shown increased carcass fat after 10 wks of short (10L:14D) versus long photoperiod (14L:10D) (11). Earlier, Bartness and Wade (1) had found greater body weight gain and carcass lipid content in short (8L:16D) vs long (16L:8D) photoperiod Syrian hamsters without any significant differences in food intake. Their data indicated that short photoperiod Syrian hamsters were more energy efficient
744
Photoperiod and Obesity
Vol. 49, No. i0, 1991
than were the long photoperiod animals. In the present study, male obese Zucker rats exposed to 9 wks in short photoperiod (10L:14D) consistently gained less lean body mass than did rats housed in long photoperiod (14L:10D) despite the fact that food intake did not differ between the two groups. Absolute fat was comparable in short photoperiod obese and long photoperiod obese rats, although percent body fat was higher in the former. These photoperiod-induced differences in composition of gain may reflect altered testosterone levels/function in the obese Zucker rats (12) as has been reported in Sprague-Dawley rats (2). Thus, genetically obese Zucker rats exposed to long photoperiod appeared to be more energy efficient than rats exposed to short photoperiod. In both lean and obese rats, photoperiod had no effect on brown fat GDP binding per mg of mitochondrial protein (an in vitro index of the tissue's thermogenic state at the time of sacrifice). However, because the amount of mitochondrial protein recovered was greater in short vs long photoperiod obese rats, total GDP binding would be higher in the short photoperiod animals. Although this could contribute to the lower weight gain of the short photoperiod obese rats, it is not consistent with their lower lean body mass and higher percent body fat. Thus the effect of photoperiod of body composition in the obese rats is unlikely to be due to altered brown fat thermogenesis. Among other factors that may have contributed to the photoperiod-induced alteration in metabolic efficiency are differential spontaneous activity levels in the two photoperiods and effects of meal feeding. It is possible that the long photoperiod animals were less active than the short photoperiod rats because they spent less time in their active (nocturnal) period and presumably more time sleeping. There is evidence that increased activity is associated with lower plasma insulin concentrations (13). The fact that insulin levels were higher in both lean and obese long photoperiod groups supports the concept that these rats were less active. Meal feeding or the effects of consuming the same amount of calories in a shorter period of time generally results in increased fat gain, with decreased carcass protein and water and enhanced sensitivity to insulin (14,15). Considering the fact that rats generally consume 70-85% of their food in the dark phase one would expect to see a compression of meal intervals in the tong photoperiod animals. Therefore, one might also expect to see meal feeding effects in these rats -- specifically, increased carcass fat and decreased carcass protein and water. We did not observe this. In fact, carcass protein was significantly greater in the long versus short photoperiod groups. Recent studies have shown that there is an association between distribution of fat depots and other variables associated with obesity including hyperinsulinemia (16-18). The relative amount of internal fat, represented by the sum of mesenteric, retroperitoneal and epididymal depots divided by total carcass fat, increased with exposure to long photoperiod in the lean rats (Table II). Increased internal fat deposition in humans has major health implications, including insulin resistance, hyper-insulinemia, and hypertension, changes which are also seen with obesity (16-18). Under conditions in which this altered distribution of fat occurred in the long photoperiod lean rats, plasma insulin levels were also increased. Thus, in lean rats, long photoperiod evoked
Vol. 49, No. I0, 1991
Photoperiod and Obesity
745
hyperinsulinemia, distribution of fat to internal depots, and subsequent body weight gain. Although insulin levels were higher in long vs short photoperiod obese rats, there was no significant effect on body fat distribution, due perhaps to the insulin resistance in the obese genotype. In conclusion, male Zucker rats appear to be photoperiod sensitive at least with respect to a variety of factors related to energy balance and energy efficiency. Moreover, the obese Zucker rats appear to be more responsive to photoperiod than the lean animals. This is reflected in the more pronounced photoperiod-induced changes in carcass composition in the obese versus the lean rats. Finally, the observation that long photoperiod results in higher plasma insulin in obese and lean rats and increased internal/external fat in lean rats deserves further study. Acknowledgements The authors thank Sue Hansen, Johnson Lai, and Ahn Le for their technical assistance. This research was supported in part by National Institute of Health Grants DK18899, DK32907 and DK35747, and National Science Foundation Grant DCB 8421183. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.
T.J. BARTNESS and G.N. WADE, Endocrinology 114492-498 (1984). R.J. NELSON and I. ZUCKER, Neuroendocrinology 3_.22266-271 (1981). E.P. WALLEN and F.W. TUREK, Nature 289402-404 (1981). M.J. GIBSON, A.S. LIOI-FA, and D.T. KRIEGER, Neuropeptides 1349-362 (1981). M.R. FREEDMAN, T.W. CASTONGUAY, and J.S. STERN, Am. J. Physiol. 249 R584R594 (1985). R.B. McDONALD, J.S. HAMILTON, J.S. STERN,and B.A. HORWlTZ, Am. J. Physiol. 254 R462 (1988). O.H. LOWRY, N.J. ROSEBROUGH, A.L. FARR, and R.J. RANDAL, J.Biol.Chem. 193 265-275 (1951). B. DESBUQUOIS and G.D. AURBACH, J. Clin. Endocr 3__33732(1971). J.L. WALBERG, P.A. MOLE, and J.S. STERN, Am. J. Clin. Nutr. 3_.88879-883 (1983). G.E. BELL and J.S. STERN, Growth 4_~163-80(1977). K.S. KOI-I-, B.J. MOORE, L.M. FOURNIER, and B.A. HORWlTZ, Am. J. Physiol. 256 R174-R180 (1989). E.S. EDMONDS, S.K. DALLIE, and B.W. THYACHUMNAR, 2-7891-897 (1982). E.A. APPLEGATE and J.S. STERN, Metabolism 36709-714 (1987). C.D. BERDANIER, Nutr. Rpts. Int. 1_!1:6517-524 (1975). C. COHN and D. JOSEPH, J. Nutrition 9__6694-100 (1968). D.J. EVANS, R. MURRAY, and A.H. KlSSEBAH, J. Clin. Invest. 7_4_41515-1525 (1984). A.H. KlSSEBAH, N. Vydelingum, R. Murray, etal., J. Clin. Endocrinol. Metab. 5_44:2 254-260 (1982). A.N. PEERIS, M.F. STRUVE, and A.H. KISSEBAH, Int. J. Obesity 1_~1581-589(1987).
Pergamon Press
EFFECT OF PHOTOPERIOD ON BODY WEIGHT AND FOOD INTAKE OF OBESE AND LEAN ZUCKER RATS.
Lisa M. Larkin 1, B. J. Moore 1'3'4 Judith S. Stern 1-3, and Barbara A. Horwitz3.
Departments of Nutrition 1, Internal Medicine 2 and Animal Physiology 3, and The Food Intake Laboratory, University of California, Davis 95616; Weight Watchers International 4, 500 N. Broadway, Jericho, NY. 11753 (Received in final form July 8, 1991)
Summary Although the rat is usually not considered to be sensitive to photoperiod, under some experimental conditions photoperiod responses are unmasked. In addition, we have observed photoperiod-induced changes in body weight gain in lean and obese Zucker rats. In this experiment, body mass, food intake, body composition, brown adipose tissue (BAT) thermogenic state, and blood concentrations of corticosterone, insulin, and glucose were evaluated under one of two lighting conditions: a short (10 h light: 14 h dark) or a long (14 h light: 10 h dark) photoperiod. Plasma corticosterone and glucose concentrations measured under fasting conditions were unaffected by photoperiod in either genotype. The amount of BAT mitochondrial protein isolated was less in long photoperiod rats. BAT mitochondrial GDP binding was unaffected by photoperiod in the lean rats, but tended to be lower in long photoperiod obese rats than in short photoperiod obese rats. Although, photoperiod had no effect on daily food intake of rats exposed to the short versus long photoperiod, body mass was heaviest in obese rats raised in long photoperiod. Plasma insulin was increased in both lean and obese rats in long photoperiod. In addition, fat storage appeared to shift to internal depots in the lean rats exposed to long photoperiod. Our data demonstrate that photoperiod does have an effect on male Zucker rats with respect to body weight and fat distribution, with the obese rats being more sensitive to changes in photoperiod than the lean rats. 0024-3205/91 $3.00 + .00 Copyright (c) 1991 Pergamon Press plc
736
Photoperiod and Obesity
Vol. 49, No. i0, 1991
Photoperiod modifies fat accumulation, food intake, body composition, and body weight gain in a variety of mammals. These effects are pronounced in rodents such as Syrian hamsters and voles, whose reproductive cycle is sensitive to photoperiod (1). Rats are generally not thought to be as sensitive to changes in photoperiod as are the more "wild-type" rodents even though rats breed more successfully when placed in long photoperiod. Moreover, under some experimental conditions such as bulbectomy and pinealectomy, photoperiod responses are unmasked in rats (2,3). This study evaluated the effects of long versus short photoperiod on weight gain, body composition, and food intake as well as on several hormonal variables in Zucker lean and obese male rats. A pilot study from our laboratory indicated that pinealectomy increased weight gain without a change in food intake of lean but not obese Zucker rats. In addition, obese Zucker rats (fa/fa) lack the normal diurnal feeding pattern and periodicity in plasma corticosterone levels seen in most lean rats (4), adrenalectomy restores the feeding patterns to those observed in lean rats, and corticosterone replacement restores the patterns to those of sham-operated obese rats (5). These observations suggest that photoperiod may have a greater effect on the lean than on the obese rat. The present study negates this hypothesis. Materials and Methods Animals and Animal Care Five week old male Zucker obese (fa/fa) and lean (Fa/?) rats, were obtained from Charles River Laboratories, Inc. (Raleigh, NC). Three days after arrival, the animals were assigned to one of two photoperiods, either short (10 hrs light:14 hrs dark) or long (14 hrs light:10 hrs dark). The animals were separated to obtain two groups of obese rats and two groups of lean rats with equivalent initial body weights. The rats were allowed to adapt to the new photoperiod for four weeks before beginning the experiment. All animals were individually housed at 24-26°C in wire-bottom hanging cages and fed Simonsen's White diet (Gilroy, CA) and water ad libitum. Experimental Design From 9 to 14 wks of age, 24 hr food intake and body weight were measured 3 times a week. Prior to 14 wks of age, all rats were conditioned to the guillotine to reduce stress during sacrifice. At 14 wks of age, food was removed one h before lights off, and the rats were fasted overnight. Rats were sacrificed via decapitation the next morning, one h after lights on (lights on at 0600 for the long photoperiod group and 0800 for the short photoperiod group). At the time of sacrifice, trunk blood was collected; cervical, axillary, and scapular brown adipose tissue were removed and kept on ice for approximately 10 min prior to homogenization; white adipose tissue (epididymal, retroperitoneal, and mesenteric) was dissected, weighed, and replaced in the eviscerated carcass for determination of carcass composition.
Vol. 49, No. i0, 1991
Photoperlod and Obesity
737
GDP Binding Analysis The binding of GDP to freshly isolated brown fat mitochondria was determined as previously described by McDonald et al. (6), with slight modification of the mitochondrial isolation procedure. Minced BAT was placed in approximately 50 ml of isolation buffer (pH 7.2) composed of 0.25 M sucrose, 1 mM n-2-hydroxyethylpiperazine-N's-2ethanesulfonic acid (HEPES), and 0.2 mM ethylenediamine-tetraacetate (EDTA). This suspension was homogenized in a Kontes Duall glass-glass tissue grinder and centrifuged at 1200 x g for 10 min at 5°C (Sorvall RC-5B centrifuge). The supernatant was decanted and saved, and the pellet was resuspended in isolation buffer and centrifuged at 1200 x g for 10 min. The supernatants from both centrifugations were each spun at 20,000 x g for 14 min, and the pellets were resuspended in more isolation buffer, combined, and recentrifuged for 14 min at 16,318 x g. The resulting pellet was resuspended in reaction buffer and used in the GDP binding assay. In our hands, this isolation procedure routinely yields 30-35% recovery, with no differences between obese and lean rats. Net GDP bound was determined after correction for trapped extramitochondrial fluid. Specific binding was calculated by subtracting non-specific binding (i.e., binding in the presence of 100 uM ADP) from total binding (binding in the absence of ADP). Protein was determined by the method of Lowry et al. (7) using bovine serum albumin as a standard. Blood Analysis Trunk blood was collected in heparinized tubes, spun at 2000 x g for 20 min, and the resulting plasma was stored at -70°C for subsequent analysis of glucose, insulin, and corticosterone. Glucose was analyzed by the method of glucose oxidase (Beckman Glucose Analyzer 2, Beckman Instruments, Inc., Fullerton, CA). Insulin was analyzed using a modification of the single antibody/polyethylene glycol method of Desbuguois and Aurbach (8). Phosphate buffer (0.05 M) containing 0.4% human serum albumin (pH 7.4) (Cutter Biological, Berkely, CA), rat insulin standard (21.3 U/mg; Novo Biolabs, Wilton, CT), insulin antisera (porcine; ICN Diagnostics Division, Costa Mesa, CA), [1251]insulin (Amersham Corp., Arlington Heights, IL), and polyethlene glycol (MW 8,000; Sigma Chemical Co., St. Louis, MO) were used as previously described (9). Corticosterone was assayed using a [1251]-corticosterone radioimmunoassay for rats (ICN Diagnostics Division). Carcass Composition Following removal of brown adipose tissue, three white fat depots (epididymal, retroperitoneal, and mesenteric) were excised and weighed. The rat was then eviscerated and the white fat depots were returned to the carcass for subsequent carcass composition analysis using the method of Bell and Stern (10).
738
Photoperiod and Obesity
Vol. 49, No. i0, 1991
Statistics Analysis of variance (ANOVA) was used to evaluate data using a factorial 2 x 2 design (photoperiod versus genotype), except for the analysis of blood data where a one way ANOVA was used because the variances between genotypes differed. Differences were considered to be significant at p < 0.05. When the ANOVA was significant, individual means were evaluated using the Newman Keurs post hoc test. Results Body Weight Initial body weights at five wks of age were: long photoperiod lean rats (LL), 94.0 + 3.8 g; short photoperiod lean rat (SL), 93.4 + 3.7 g; long photoperiod obese rats (LO), 144.0 + 3.4 g; and short photoperiod obese rats (SO), 144.0 + 3.4 g. After four wks adaptation to the experimental photoperiods, the LO rats were heavier than the SO animals (Fig. 1). This pattern continued throughout the remainder of the experiment (p < 0.001) (Fig. 1).
600 m E
2
Ot
LO(N:10) SO(N=IO)
~
500 400
IO) "6
300
~ "0 0 ¢)
200
100
0
\
~ \~I 5
\
\
LL (N=IO) SL(N=9)
~
I
10
,
!
11
,
I
12
,
I
13
,
I
14
Week of Experiment
Fig. 1 Weekly body weights in lean and obese male rats in either short or long photoperiod. Values are means + SEM; SEM bars are within the mean symbol. *LO values differ significantly (p < 0.05) from SO values.
Vol. 49, No. I0, 1991
Photoperiod and Obesity
739
Food Intake Food intake measurements began after the rats had been in their respective photoperiods for a total of 4 wks. Although there was no significant effect of photoperiod on daily food consumption, there was an effect of genotype (Fig. 2; p < 0.0001). During the first two weeks of measurements, the obese animals ate 40% more per day than did their lean counterparts. This difference then decreased, becoming nonsignificant by .the end of the measurement period (Fig. 2). Photoperiod differentially affected the lean versus obese rats. In long photoperiod, the obese rats consistently ate more than did the lean rats, but daily differences never reached significance (p=0.07) (Fig. 2). However, total food consumed during the 5 wks of this experiment was significantly greater in the obese vs lean rats, and there was an interaction of genotype with photoperiod: LL, 293.9 + 72.8 g; SL 324.9 _+39.1 g; LO, 404.9 + 43.4 g; SO, 359.1 + 37.9 g. This interaction was not significant when total food consumption was expressed in a body massindependent fashion: LL, 6.2 + 0.6 g/kg°67; SL, 6.5 + 0.8 g/kg°67; LO, 5.8 + 0.6 g/kg°67; SO, 5.5 + 0.6 g/kg °'67.
40
>, (u "o oi
-
30 ~Lr
.......
4)
m r.
.~...-
. . . - 4-r ~
==
.....
. .......
=
'~'" ,~
" ""
......
"''O
--~
&
20
LO (N:10)
SO (N=10) SL (N=9)
LL (N=IO)
i
"0 0 0
10
0
"\ \
I
,
I
10
i
I
11
,
12
Week
of
I
13
*
I
14
Experiment
Fig. 2 Daily food intake in lean and obese male rats in either short or long photoperiod. Kcal/gram = 3.00. Values are means + SEM; SEM bars are ~r
within the mean symbol. values.
LO values differ significantly (p < 0.05) from SO
740
Photoperiod and Obesity
Vol. 49, No. i0, 1991
Carcass Composition Analysis Genotype significantly altered carcass composition, with the obese animals having heavier carcass weights, higher percent carcass fat, but lower percent water (p < 0.0001) than their lean counterparts. However, lean body mass, protein, and ash of the LO animals did not significantly differ from that in LL animals (Table I). The obese animals were sensitive to the different photoperiods, with percent water and grams of fat being the only variables unaffected by photoperiod. The LO animals had greater carcass weight, lean body mass, protein, ash, water, and percent body fat than did the SO rats. Although lean rats showed similar trends, the differences due to photoperiod did not reach significance except for carcass protein which was higher in tong photoperiod animals (Table I).
TABLE I Carcass Composition
LEAN Photoperiod n
Long 9
OBESE Short 10
Long 10
Short 10
Body mass, g
369.2_+13.1 a
352.2+ 8.5 a
575.4+ 9.2 b
523.3-+12.5 c
Carcass mass, g
304.2+ 9.5 a
288.8+ 9.0 a
485.4-+ 8.3 b
436.7+11.9 c
LBM, g
272.3+ 5.4 a
252.2+ 9.8 a
269.3+ 6.3 a
223.0-+ 7.6 b
Protein, g
262.1+ 5.5 a
236.2+ 8.6 b
258.4+ 6.1a
214.1-+ 7.7 c
Ash, g Fat,
12.0+ 0.27 a
12.5+ 0.92 a
39.9+ 2.4 a
34.0+ 2.9 a
216.1+ 5.4 b
213.8+ 8.4 b
13.8+ 0.7 a
11.9+ 1.0 a
44.5+ 0.9 b
48.9+ 1.2 c
g
188.1+ 4.0 a
173.3+ 7.5 a
156.6_+4.1 b
137.6+ 5.6 C
%
60.2_+ 0.5 a
60.5_+ 0.8 a
32.3_+0.7 b
31.5_+ 1.0 b
69.2_+ 0.2 a
68.6_+ 0.4 a
58.2+ 1.0 b
61.7_+ 1.1 c
g % Water,
Water/LBM x 100
10.9+ 0.29 a
8.9+ 0.12 b
Values are means + SEM. LBM, lean body mass; n, number of animals in groups. Values with different superscripts are significantly different from each other at p < 0.05.
Vol. 49, No. I0, 1991
Photoperiod and Obesity
741
Fat Distribution When fat deposition was expressed in terms of percent of carcass mass, there was no significant difference due to photoperiod in the individual depots (epididymal, retroperitoneal, mesenteric) or in the total of all three depots (Table II). However, there was a significant difference (p < 0.0001) due to genotype, with the genetically obese rats having more fat in all three depots. Internal fat distribution was estimated by dividing the sum of the mass of the mesenteric, retroperitoneal, and epididymal depots by total carcass fat: the higher the ratio, the more the internal distribution of body fat. The LL rats had relatively more internal fat (ratio = 0.35) than did the SL (ratio = 0.29) or SO rats (ratio = 0.26) (Table II). Although the LO rats tended to have more internal fat (ratio = 0.28) than the SO rats, this difference was not statistically significant.
TABLE II White Adipose Tissue Mass (grams)
LEAN Photoperiod n
Long 9
OBESE Sho~ 10
Long 10
Sho~ 10
EPI
5.65+0.37 a
4.24+0.45 a
18.40+0.68 b
16.27+0.77 c
RP
4.23+0.39 a
3.44+0.32 a
26.78+1.33 b
24.23+1.55 b
MES
3.37+0.34 a
2.51+_0.23a
14.95+0.75 b
13.99+_O.45b
TOTAL
13.25+1.06 a
10.18+0.96 a
60.14+2.31b
54.49+2.24 c
TOTAL
3.69_+O.13a
3.52+0.09 a
5.75_+0.09b
5.23_+0.12c
0.35+0.1 la
0.29+0.01b
0.28_+0.01bc
0.26+0.01c
100g BW EPI+RP+MES CARC
Values are means + SEM. EPI, epididymal; RP, retroperitoneal; MES, mesenteric; EPI+RP+MES=TOTAL; CARC, carcass fat; n = number of rats/group. Values with different superscripts are significantly different from each other at p < 0.05.
742
Photoperiod and Obesity
Vol. 49, No. i0, 1991
TABLE III Plasma Corticosterone, Glucose, and Insulin.
LEAN
Corticosterone (nmol/L) Glucose (mmol/L) Insulin (pmol/L)
OBESE
Long
Short
Long
(4) 150.6+53.5
(4) 203.6+75.9
(4) 310.4+237.3
Short
(5) 206.9_+65.7
(9)
(10)
(10)
(10)
7.1 + 0.3
6.6 + 0.2
6.9 + 0.3
7.1 + 0.3
(9)
(9)
(10)
(9)
140 + 12 a
1542 + 174
239 + 22
988 + 151a
Values are means + SEM. The number of rats per group is given in parenthesis. Values with different superscripts indicate that short photoperiod values are significantly different from long photoperiod values in the same genotype. The (n) is smaller for corticosterone because of technical difficulties.
Plasma Insulin, Glucose, and Corticosterone There were no significant differences in plasma levels of glucose and corticosterone with either photoperiod or genotype (Table III). However, as expected, plasma insulin was significantly elevated in obese vs lean rats. Moreover, the one way ANOVA performed separately on lean and obese genotypes demonstrated that long photoperiod rats had significantly higher plasma insulin levels than did the short photoperiod rats (p < 0.03 for obese rats and p < 0.001 for lean rats). Brown Adipose Tissue and GDP Binding SO animals had 12.6% less BAT mass than did the LO rats (Table IV). There was no such difference in the lean animals. When BAT mass was expressed in terms of kg body mass, differences between LO and SO rats disappeared, but the SL rats had higher values than did the LL group. Although BAT mass was greater in the obese vs lean animals, the obese brown fat depots were very pale, indicating large amounts of stored triglycerides. A better indication of differences in BAT oxidative function is mitochondrial protein and GDP binding (Table IV).
Vol. 49. No. I0, 1991
Photoperlod and Obesity
743
TABLE IV Brown Adipose Tissue Mass and GDP Binding OBESE
LEAN Photoperiod n
Long
Short
Long
Short
9
10
10
10
BAT mass g
0.88+0.06 a
1.10_+0.06 a
5.76+0.28 b
5.03+0.25 c
g/kg BW
2.37_+0.09a
3.11+0.13 b
10.0+0.32 c
9.66+0.37 c
mitochondrial protein (mg)
12.6+0.7 a
17.8+1.0 b
3.7_+0.9 c
GDP binding (nmol/mg mito prot)
0.15+0.01
0.12_+0.02
Recovered
0.08_+0.02
7.0_+0.8 d
0.11+0.03
Values are means + SEM. BAT, brown adipose tissue; GDP, guanosine 5'diphosphate; BW, body weight; n, number of animals in group. In a given row, values with different letter superscripts differ significantly (p < 0.05).
The amount of mitochondrial protein recovered showed a significant effect of photoperiod (p < 0.0001) and of genotype (p < 0.0001). Short photoperiod animals had more recoverable BAT mitochondrial protein than did long photoperiod rats, and the lean rats had more than their obese counterparts. The specific activity of GDP binding (nmol/mg mitochondrial protein) was not altered by photoperiod or genotype (Table IV).
Discussion Rats are generally not thought to be as sensitive to photoperiod as the more "wildtype" rodents such as Syrian hamsters and voles (1). Nonetheless, under some experimental conditions such as bulbectomy and pinealectomy, photoperiod responses are unmasked in rats (2,3). Studies looking at the effects of photoperiod on metabolic variables in Syrian hamsters have shown increased carcass fat after 10 wks of short (10L:14D) versus long photoperiod (14L:10D) (11). Earlier, Bartness and Wade (1) had found greater body weight gain and carcass lipid content in short (8L:16D) vs long (16L:8D) photoperiod Syrian hamsters without any significant differences in food intake. Their data indicated that short photoperiod Syrian hamsters were more energy efficient
744
Photoperiod and Obesity
Vol. 49, No. i0, 1991
than were the long photoperiod animals. In the present study, male obese Zucker rats exposed to 9 wks in short photoperiod (10L:14D) consistently gained less lean body mass than did rats housed in long photoperiod (14L:10D) despite the fact that food intake did not differ between the two groups. Absolute fat was comparable in short photoperiod obese and long photoperiod obese rats, although percent body fat was higher in the former. These photoperiod-induced differences in composition of gain may reflect altered testosterone levels/function in the obese Zucker rats (12) as has been reported in Sprague-Dawley rats (2). Thus, genetically obese Zucker rats exposed to long photoperiod appeared to be more energy efficient than rats exposed to short photoperiod. In both lean and obese rats, photoperiod had no effect on brown fat GDP binding per mg of mitochondrial protein (an in vitro index of the tissue's thermogenic state at the time of sacrifice). However, because the amount of mitochondrial protein recovered was greater in short vs long photoperiod obese rats, total GDP binding would be higher in the short photoperiod animals. Although this could contribute to the lower weight gain of the short photoperiod obese rats, it is not consistent with their lower lean body mass and higher percent body fat. Thus the effect of photoperiod of body composition in the obese rats is unlikely to be due to altered brown fat thermogenesis. Among other factors that may have contributed to the photoperiod-induced alteration in metabolic efficiency are differential spontaneous activity levels in the two photoperiods and effects of meal feeding. It is possible that the long photoperiod animals were less active than the short photoperiod rats because they spent less time in their active (nocturnal) period and presumably more time sleeping. There is evidence that increased activity is associated with lower plasma insulin concentrations (13). The fact that insulin levels were higher in both lean and obese long photoperiod groups supports the concept that these rats were less active. Meal feeding or the effects of consuming the same amount of calories in a shorter period of time generally results in increased fat gain, with decreased carcass protein and water and enhanced sensitivity to insulin (14,15). Considering the fact that rats generally consume 70-85% of their food in the dark phase one would expect to see a compression of meal intervals in the tong photoperiod animals. Therefore, one might also expect to see meal feeding effects in these rats -- specifically, increased carcass fat and decreased carcass protein and water. We did not observe this. In fact, carcass protein was significantly greater in the long versus short photoperiod groups. Recent studies have shown that there is an association between distribution of fat depots and other variables associated with obesity including hyperinsulinemia (16-18). The relative amount of internal fat, represented by the sum of mesenteric, retroperitoneal and epididymal depots divided by total carcass fat, increased with exposure to long photoperiod in the lean rats (Table II). Increased internal fat deposition in humans has major health implications, including insulin resistance, hyper-insulinemia, and hypertension, changes which are also seen with obesity (16-18). Under conditions in which this altered distribution of fat occurred in the long photoperiod lean rats, plasma insulin levels were also increased. Thus, in lean rats, long photoperiod evoked
Vol. 49, No. I0, 1991
Photoperiod and Obesity
745
hyperinsulinemia, distribution of fat to internal depots, and subsequent body weight gain. Although insulin levels were higher in long vs short photoperiod obese rats, there was no significant effect on body fat distribution, due perhaps to the insulin resistance in the obese genotype. In conclusion, male Zucker rats appear to be photoperiod sensitive at least with respect to a variety of factors related to energy balance and energy efficiency. Moreover, the obese Zucker rats appear to be more responsive to photoperiod than the lean animals. This is reflected in the more pronounced photoperiod-induced changes in carcass composition in the obese versus the lean rats. Finally, the observation that long photoperiod results in higher plasma insulin in obese and lean rats and increased internal/external fat in lean rats deserves further study. Acknowledgements The authors thank Sue Hansen, Johnson Lai, and Ahn Le for their technical assistance. This research was supported in part by National Institute of Health Grants DK18899, DK32907 and DK35747, and National Science Foundation Grant DCB 8421183. References 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.
T.J. BARTNESS and G.N. WADE, Endocrinology 114492-498 (1984). R.J. NELSON and I. ZUCKER, Neuroendocrinology 3_.22266-271 (1981). E.P. WALLEN and F.W. TUREK, Nature 289402-404 (1981). M.J. GIBSON, A.S. LIOI-FA, and D.T. KRIEGER, Neuropeptides 1349-362 (1981). M.R. FREEDMAN, T.W. CASTONGUAY, and J.S. STERN, Am. J. Physiol. 249 R584R594 (1985). R.B. McDONALD, J.S. HAMILTON, J.S. STERN,and B.A. HORWlTZ, Am. J. Physiol. 254 R462 (1988). O.H. LOWRY, N.J. ROSEBROUGH, A.L. FARR, and R.J. RANDAL, J.Biol.Chem. 193 265-275 (1951). B. DESBUQUOIS and G.D. AURBACH, J. Clin. Endocr 3__33732(1971). J.L. WALBERG, P.A. MOLE, and J.S. STERN, Am. J. Clin. Nutr. 3_.88879-883 (1983). G.E. BELL and J.S. STERN, Growth 4_~163-80(1977). K.S. KOI-I-, B.J. MOORE, L.M. FOURNIER, and B.A. HORWlTZ, Am. J. Physiol. 256 R174-R180 (1989). E.S. EDMONDS, S.K. DALLIE, and B.W. THYACHUMNAR, 2-7891-897 (1982). E.A. APPLEGATE and J.S. STERN, Metabolism 36709-714 (1987). C.D. BERDANIER, Nutr. Rpts. Int. 1_!1:6517-524 (1975). C. COHN and D. JOSEPH, J. Nutrition 9__6694-100 (1968). D.J. EVANS, R. MURRAY, and A.H. KlSSEBAH, J. Clin. Invest. 7_4_41515-1525 (1984). A.H. KlSSEBAH, N. Vydelingum, R. Murray, etal., J. Clin. Endocrinol. Metab. 5_44:2 254-260 (1982). A.N. PEERIS, M.F. STRUVE, and A.H. KISSEBAH, Int. J. Obesity 1_~1581-589(1987).