- Email: [email protected]
Normalization of circulating leptin levels by fasting improves the reproductive function in obese OLETF female rats
Neuropeptides (2001) 35 (1), 45±49 ß 2001 Harcourt Publishers Ltd doi: 10.1054/npep.2000.0842, available online at http://www.idealibrary.com on
Normalization of circulating leptin levels by fasting improves the reproductive function in obese OLETF female rats H. Watanobe,1 M. Yoneda,2 A. Kohsaka,1 Y. Kakizaki,1 T. Suda,1 H. B. Schioth3 1
Third Department of Internal Medicine, Hirosaki University School of Medicine, Aomori, Japan Department of Gastroenterology, Dokkyo University School of Medicine, Mibu, Tochigi, Japan 3 Department of Neuroscience, Uppsala University, Sweden and 3Melacure Therapeutics, Uppsala, Sweden 2
SUMMARY In order to examine a possible detrimental effect of hyperleptinemia on the reproductive system, we examined whether a decrease in circulating leptin levels by fasting affects the estradiol/progesterone-induced luteinizing hormone (LH) and prolactin (PRL) surges in genetically obese OLETF (Otsuka-Long-Evans-TokushimaFatty) rats. Experiments were performed on both normally fed and 3-day starved groups from ovariectomized OLETF rats and their controls LETO (Long-Evans-Tokushima-Otsuka). Starved LETO rats, whose leptin levels were less than 0.5 ng/ml, did not show a significant surge of either LH or PRL. Normally fed OLETF rats, whose leptin levels were 9.7 + 1.8 ng/ml, showed a significant but small surge for both LH and PRL. Interestingly, starved OLETF rats, whose leptin levels (4.1 + 0.7 ng/ml) were similar to those in normally fed LETO rats (3.3 + 0.4 ng/ml), had significantly greater surges of both hormones than normally fed OLETF group. This study demonstrates for the first time that the normalization of circulating leptin levels in female OLETF rats augments the steroid-induced LH and PRL surges, and also suggests a deleterious effect of hyperleptinemia on the reproductive axis. ß 2001 Harcourt Publishers Ltd
INTRODUCTION In humans, the in¯uence of nutrition and body composition on puberty and the reproductive function has long been recognized (Frisch and McArthur, 1974; Frisch, 1980). The critical weight hypothesis of the development of puberty states that when body weight (BW) reaches a certain level, puberty takes place (Frisch and McArthur, 1974). In this context, leptin, a recently discovered adipocyte-derived hormone (Zhang et al., 1994; Pelleymounter et al., 1995), has been reported to positively regulate the reproductive axis in both humans and rodents, and is considered to serve as a critical metabolic
Received 23 August 2000 Accepted 20 September 2000 Correspondence to: Hajime Watanobe, M. D., Division of Internal Medicine, Center for Clinical Research, International University of Health and Welfare, 2600-1 Kitakanemaru, Otawara, Tochigi 324-8501, Japan. Fax: 81 287 24 1248; E-mail: [email protected]
signal linking adiposity and the reproductive system (Clarke and Henry, 1999). In apparent contradiction to this stimulatory effect of leptin on the endocrine axis, subnormal gonadal function has been demonstrated in obesity with hyperleptinemia. Obese women have an increased rate of gonadal dysfunction and infertility (Hartz et al., 1979). Also in men, serum testosterone levels are subnormal in obese males and return to normal with weight loss (Kley et al., 1979). In addition, there appears to be an inverse relationship between serum testosterone concentrations and BW (Amatruda et al., 1978; Kley et al., 1979). Similar ®ndings are observed in rodents as well. Impaired reproductive function has been reported in genetically obese strains of rats, such as the Zucker (Zucker and Zucker, 1961; Zucker and Zucker, 1962; Hemmes et al., 1978; Young et al., 1982; Whitaker et al., 1983; Doherty et al., 1985) and OLETF (Otsuka-Long-Evans-Tokushima-Fatty) (Shimizu et al., 1998) animals. Since it is known that both the Zucker and OLETF rats have hyperleptinemia (Rayner et al., 1997; 45
Watanobe et al.
Shimizu et al., 1998), an interesting possibility may thus arise that hyperleptinemia may act negatively on the reproductive function. Indeed, a recent study in humans demonstrated a detrimental effect of excess leptin on the reproductive system of girls (Bouvattier et al., 1998). As we are not aware of any previous study examining the effect of hyperleptinemia on the reproductive axis of genetically obese animals, in the present study we tested the in¯uence of fasting on the estradiol (E2)/progesterone (P)-induced luteinizing hormone (LH) and prolactin (PRL) surges in female OLETF rats, and compared the results with those from their controls LETO (Long-Evans-TokushimaOtsuka).
(a) 40 Plasma LH (ng/ml)
46
30 20 10 0 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00
MATERIALS AND METHODS
Neuropeptides (2001) 35(1), 45±49
(b) Plasma PRL (ng/ml)
All of the following experiments were conducted in accordance with the Guidelines for Animal Experimentation, Hirosaki University. Female OLETF and LETO rats, 4 weeks of age, were obtained from Tokushima Research Institute, Otsuka Pharmaceutical (Tokushima, Japan). They were housed in an air-conditioned room with controlled lighting (light 08:00±20:00). The animals were given free access to laboratory chow and tap water, unless otherwise indicated. The rats were examined daily for vaginal opening from 30 days of age. When the animals were 7 weeks of age, we started taking vaginal smears by lavage for 3 weeks. At the age of about 10 weeks, both OLETF and LETO rats were bilaterally ovariectomized under light ether anesthesia, and subjected to experiments about 2 weeks later (at the age of about 12 weeks). At about 08:00 of the day 72 h before the experiment, we measured BW of every animal. Two days prior to the experiment, the animals were implanted with a jugular vein catheter ®lled with heparin solution under light ether anesthesia. At 10:00 on the same day, every rat was injected subcutaneously with 150 mg/kg BW of E2 benzoate (Mochida Pharmaceutical Co., Ltd., Tokyo, Japan). Experiments were performed on both normally fed and 3-day-starved groups from both OLETF and LETO strains. At about 08:00 on the day of the experiment, BW was measured and the jugular vein catheter was exteriorized for frequent blood sampling. At 09:00, 20 mg/kg BW of P (Mochida Pharmaceutical Co., Ltd.) was injected intramuscularly. Blood samples (200 ml) were collected every 30 min over a total period of 420 min (11:00±18:00). At 11:00, additional 200 ml of blood was drawn to measure leptin as well. To prevent the loss of circulating plasma volume, 0.9 % NaCl was injected intravenously immediately after each blood collection in the same volume as that drawn. The blood was collected in EDTA-2Na (2.5 mg/ml)-containing tubes, centrifuged, and the plasma was stored at ÿ 708C until assayed for leptin, LH, and PRL.
OLETF (normally fed) OLETF (3-day starved) LETO (normally fed) LETO (3-day starved)
300
200
100
0 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 Time of day (h) Fig. 1 Effects of 3-day starvation on steroid-induced LH and PRL surges in OLETF and LETO rats. The number of rats examined was 7±9 per group. , significantly different vs OLETF (normally fed) group. *, significantly different vs LETO (3-day starved) group. Where standard errors are not shown, they were smaller than the symbols. For further details, see text.
Plasma leptin levels were determined by a rat leptin RIA kit produced by Linco Research (St. Louis, MO, USA). The sensitivity of the assay was 0.5 ng/ml. LH and PRL levels were determined by RIA using the reagents kindly donated by Dr. A. F. Parlow (NIDDK). Rat LH-RP-3 and PRL-RP-3 were used as the standards. Sensitivity of the LH assay was 0.2 ng/ml, and that of PRL assay was 0.8 ng/ml. Both the intra- and interassay coef®cients of variation were less than 10 % in all the three assays. Results were expressed as the mean + S.E.M. One-way or two-way ANOVA followed by Scheffe's post hoc test was then used to analyze the data. Differences were considered signi®cant if P was smaller than 0.05. RESULTS OLETF rats showed a small but signi®cant delay in vaginal opening, i.e. at 40.5 + 0.8 days (n 20) in OLETF vs 35.4 + 0.7 days (n 20) in LETO. In addition, the estrous ß 2001 Harcourt Publishers Ltd
Normalization of circulating leptin levels by fasting improves the reproductive function
47
Table 1 BW and plasma leptin levels in the 4 experimental groups BW (g) 1)
Group
Number of rats
72 h before
Day of experiment2)
Percent change in BW
Plasma leptin3) (ng/ml)
OLETF (normally fed) OLETF (3-day starved) LETO (normally fed) LETO (3-day starved)
7 8 8 9
270 274 218 214
281 233 225 184
ÿ ÿ
9.7 + 1.8** 4.1 + 0.7 3.3 + 0.4 < 0.5
+ + + +
12* 13* 10 9
+ + + +
13* 10* 11 7
4 ( + 1) 15 ( + 3) 3 ( + 1) 14 ( + 3)
1) BW was measured at 08:00 of the day 72 h before the experiment. 2) BW was measured at 08:00 of the day of the experiment. 3) Leptin was measured in the samples obtained at 11:00 of the day of the experiment. *, statistically signi®cant vs LETO rats with the same treatment. **, statistically signi®cant vs the remaining 3 groups.
cycles of OLETF rats were longer and less regular than those of LETO rats. The mean cycle length between 7±10 weeks of age was signi®cantly longer in OLETF [5.8 + 0.3 days (n 20)] than in LETO [4.3 + 0.1 days (n 20)]. This lengthening of the estrous cycle was generally due to prolongation of diestrus. Table 1 shows the data of BW and plasma leptin levels in the 4 groups examined in this study. The percent reduction in BW after fasting was similar between 3-day-starved OLETF and 3-day starved LETO groups. With respect to plasma leptin concentrations, normally fed OLETF group had a 2.9 times higher level of the hormone than normally fed LETO group. Leptin levels in the 3-day starved OLETF group were less than half those in the normally fed OLETF group, and statistically indistinguishable from the levels in normally fed LETO group. Plasma leptin in 3-day-starved LETO group was below the assay sensitivity. Fig.1 shows the temporal changes in plasma LH and PRL in all 4 groups. With respect to LH levels [Fig.1 (a)], the normally fed LETO group showed signi®cantly higher levels of the hormone (LH surge) between 14:00±18:00 than at 11:00. In agreement with our previous data obtained from female Wistar rats (Kohsaka et al., 1999; Watanobe et al., 1999a; Watanobe et al., 1999b), the 3 day starvation imposed on LETO rats completely abolished LH surge. However LH data in OLETF rats formed a marked contrast to those in LETO rats. During the period of 17:00±18:00, the plasma LH levels in the normally fed OLETF group were signi®cantly higher than those in the 3-day-starved LETO group, and formed a small surge-like secretion of LH. The magnitude of LH surge in the OLETF (normally fed) group was markedly smaller than that in the LETO (normally fed) group. Even so, it is very interesting to note that the 3-day starvation loaded on OLETF rats did not suppress, but signi®cantly augmented LH surge. Although the magnitude of LH surge in the OLETF (3-day-starved) group was still signi®cantly smaller than that in normally fed LETO group during the period of 16:00±18:00, the surge magnitude of the former group signi®cantly exceeded that of OLETF (normally fed) group between 15:30±18:00. ß 2001 Harcourt Publishers Ltd
A similar ®nding was also observed for PRL surge [Fig.1 (b)]. In agreement with our previous observations in female Wistar rats (Kohsaka et al., 1999; Watanobe et al., 1999a; Watanobe et al., 1999b), the 3-day-starved LETO group did not show a signi®cant PRL surge, while normally fed LETO group exhibited a robust surge of the hormone. Similar to that observed for LH surge, effects of altered nutritional states on PRL surge contrasted between OLETF and LETO rats. Although the OLETF (normally fed) group had a signi®cant PRL surge, its magnitude was markedly smaller than that in normally fed LETO group. The magnitude of PRL surge in the 3-day-starved OLETF group was signi®cantly greater than that in the normally fed OLETF group, even though the surge magnitude of the former group was still signi®cantly smaller than that of normally fed LETO group between 15:30±18:00. DISCUSSION In this study, we examined whether alterations in circulating leptin levels affect the reproductive function of female OLETF rats, by employing the E2/P-induced LH and PRL surges as an indicator. Firstly, prior to ovariectomy we checked the postnatal day of vaginal opening and estrous cycles of OLETF rats in comparison to their controls LETO. We found that in OLETF rats the vaginal opening was signi®cantly delayed and the estrous cycles were signi®cantly longer. These results suggest impaired reproductive function of female OLETF rats as compared to LETO rats. Similarly delayed vaginal opening and longer estrous cycles were also reported for female Zucker rats (Whitaker et al., 1983). The present data of LH and PRL surges were even more interesting. The effects of starvation on the hormonal surges in LETO rats were essentially the same as those in Wistar rats which we described in our previous studies (Kohsaka et al., 1999; Watanobe et al., 1999a; Watanobe et al., 1999b). By contrast, in OLETF rats the effects of altered nutritional states on the LH and PRL surges were the reverse of those in LETO rats. As compared to the magnitudes of hormonal surges in the LETO (normally Neuropeptides (2001) 35(1), 45±49
48
Watanobe et al.
fed) group, the surges were very small in the OLETF (normally fed) group. However, these surges in OLETF rats were signi®cantly increased in magnitude after 3 days' starvation. Plasma leptin level in 3-day-starved OLETF group was 42% of that in normally fed OLETF group, and was statistically indistinguishable from that in LETO (normally fed) group. These results strongly suggest that in female OLETF rats the hyperleptinemia deriving from obesity may exert a deleterious effect on the reproductive axis. Impaired reproductive function has previously been reported in Zucker (Zucker and Zucker, 1961; Zucker and Zucker, 1962; Hemmes et al., 1978; Young et al., 1982; Whitaker et al., 1983; Doherty et al., 1985) and OLETF (Shimizu et al., 1998) rats. For both strains, lower serum testosterone levels than in controls were reported in males (Young et al., 1982; Shimizu et al., 1998). For Zucker rats, food restriction was reported to increase the incidence of fertility in males (Zucker and Zucker, 1962). However, the present study is the ®rst to demonstrate in female rats of a genetically obese strain that the normalization of circulating leptin levels by starvation can signi®cantly reinstate the E2/P-induced LH and PRL surges. Preexisting literature suggests the existence also in humans of a deleterious effect of obesity on the reproductive axis (Amatruda et al., 1978; Hartz et al., 1979; Kley et al., 1979; Bouvattier et al., 1998). Among the reports, the recent study of Bouvattier et al. (1998) appears to be especially important. They found that there were signi®cant negative correlations between gonadotropin-releasing hormone-induced gonadotropin responses and both the body mass index and circulating leptin levels. Among the data in this study, it is worth noting that the magnitudes of LH and PRL surges in the OLETF (3-day starved) group were still signi®cantly smaller than those in LETO (normally fed) group, even though the plasma leptin levels in these two groups were statistically the same. Although we have no clear explanation for this difference, two possibilities may be offered. First, the treatment of 3-day starvation might have been too short for the endocrine hypothalamus of OLETF rats to escape from the detrimental effect of hyperleptinemia. It is very likely that the impairment of gonadal function in obesity ensues as a result of sustained hyperleptinemia in both humans and rodents. Thus, maintenance of normal leptin levels for a longer period of time might have been necessary for OLETF rats to achieve a complete normalization of the steroid-induced LH and PRL surges. Another possible mechanism is that the 3-day starvation might have been too severe a protocol to reduce circulating leptin levels. It is likely that the fasting treatment decreases not only leptin but also other metabolic ingredients. In this context, Schneider et al. (1998) recently reported an important ®nding from using hamsters that the alleged stimulatory Neuropeptides (2001) 35(1), 45±49
action of leptin on the reproductive axis is indirect and requires oxidation of metabolic fuels such as glucose and fatty acid. Thus, instead of the acute 3-day starvation, food restriction for a longer period which is expected to cause a slow and steady decline in leptin levels without exhausting the reservoir of metabolic fuels, might have provided OLETF rats with a normal expression of LH and PRL surges. At any rate, it is an important question whether hyperleptinemia is the sole factor mediating the impaired reproductive function in obesity or if other metabolic factor(s) is/are also involved. Further detailed studies are necessary to clarify this interesting issue. In summary, this study demonstrated for the ®rst time that the normalization by fasting of circulating leptin levels in obese OLETF female rats signi®cantly augments the E2/P-induced LH and PRL surges. The present results further support the deleterious effect of hyperleptinemia on the reproductive function. ACKNOWLEDGEMENTS We wish to thank the National Hormone and Pituitary Program of NIDDK and Dr. A. F. Parlow for the generous donation of reagents for rat LH and PRL RIAs. REFERENCES Amatruda JM, Harman SM, Pourmotabbed G, Lockwood DH (1978) Depressed plasma testosterone and fractional binding of testosterone in obese males. J Clin Endocrinol Metab 47: 268±271. Bouvattier C, Lahlou N, Roger M, Bougneres P (1998) Hyperleptinaemia is associated with impaired gonadotropin responses to GnRH during late puberty in obese girls, not boys. Eur J Endocrinol 138: 653±658. Clarke IJ, Henry BA (1999) Leptin and reproduction. Rev Reprod 4: 48±55. Doherty PC, Baum MJ, Finkelstein JA (1985) Evidence of incomplete behavioral sexual differentiation in obese male Zucker rats. Physiol Behav 34: 177±179. Frisch RE, McArthur JW (1974) Menstrual cycles: fatness as a determinant of minimum weight for height necessary for their maintenance or onset. Science 185: 949±951. Frisch RE (1980) Pubertal adipose tissue: is it necessary for normal sexual maturation? Evidence from the rat and human female. Fed Proc 39: 2395±2400. Hartz AJ, Barboriak PN, Wong A, Katayama KP, Rimm AA (1979) The association of obesity with infertility and related menstrual abnormalities in women. Int J Obes 3: 57±73. Hemmes RB, Hubsch S, Pack HM (1978) High dosage of testosterone propionate increases litter production of the genetically obese male Zucker rat. Proc Soc Exp Biol Med 159: 424±427. Kley HK, Solbach HG, Mckinnan JC, Kruskemper HL (1979) Testosterone decrease and oestrogen increase in male patients with obesity. Acta Endocrinol 91: 553±563. Kohsaka A, Watanobe H, Kakizaki Y, Habu S, Suda T (1999) A significant role of leptin in the generation of steroid-induced luteinizing hormone and prolactin surges in female rats. Biochem Biophys Res Commun 254: 578±581. Pelleymounter MA, Cullen MJ, Baker MB, Hecht R, Winters D, Boone T, Collins F (1995) Effects of the obese gene product on body weight regulation in ob/ob mice. Science 269: 540±543.
ß 2001 Harcourt Publishers Ltd
Normalization of circulating leptin levels by fasting improves the reproductive function
Rayner DV, Dalgliesh GD, Duncan JS, Hardie LJ, Hoggard N, Trayhurn P (1997) Postnatal development of the ob gene system: elevated leptin levels in suckling fa/fa rats. Am J Physiol 273: R446±R450. Schneider JE, Goldman MD, Tang S, Bean B, Ji H, Friedman MI (1998) Leptin indirectly affects estrous cycles by increasing metabolic fuel oxidation. Horm Behav 33: 217±228. Shimizu H, Ohtani K-I, Uehara Y, Abe Y, Takahashi H, Tsuchiya T, Sato N, Ibuki Y, Mori M (1998) Orchiectomy and response to testosterone in the development of obesity in young Otsuka-Long-EvansTokushima Fatty (OLETF) rats. Int J Obes Relat Metab Disord 22: 318±324. Watanobe H, SchioÈth HB, Wikberg JE S, Suda T (1999a) The melanocortin 4 receptor mediates leptin stimulation of luteinizing hormone and prolactin surges in steroid-primed ovariectomized rats. Biochem. Biophys. Res. Commun 257: 860±864. Watanobe H, Suda T, Wikberg JE S, Schioth HB (1999b) Evidence that physiological levels of circulating leptin exert a stimulatory effect on
ß 2001 Harcourt Publishers Ltd
49
luteinizing hormone and prolactin surges in rats. Biochem Biophys Res Commun 263: 162±165. Whitaker EM, Shaw MA, Herrey GR (1983) Plasm oestradiol-17 beta and testosterone concentrations as possible causes of the infertility of congenitally obese Zucker rats. J Endocrinol 99: 485±490. Young RA, Frink R, Longcope C (1982) Serum testosterone and gonadotropins in the genetically obese male Zucker rat. Endocrinology 111: 977±981. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM (1994) Positional cloning of the mouse obese gene and its human homologue. Nature 372: 425±432. Zucker LM, Zucker TF (1961) Fatty, a new mutation in the rat. J Hered 52: 275±278. Zucker TF, Zucker LM (1962) Hereditary obesity in the rat associated with high serum fat and cholesterol. Proc Soc Exp Biol Med 110: 165±171.
Neuropeptides (2001) 35(1), 45±49
Normalization of circulating leptin levels by fasting improves the reproductive function in obese OLETF female rats H. Watanobe,1 M. Yoneda,2 A. Kohsaka,1 Y. Kakizaki,1 T. Suda,1 H. B. Schioth3 1
Third Department of Internal Medicine, Hirosaki University School of Medicine, Aomori, Japan Department of Gastroenterology, Dokkyo University School of Medicine, Mibu, Tochigi, Japan 3 Department of Neuroscience, Uppsala University, Sweden and 3Melacure Therapeutics, Uppsala, Sweden 2
SUMMARY In order to examine a possible detrimental effect of hyperleptinemia on the reproductive system, we examined whether a decrease in circulating leptin levels by fasting affects the estradiol/progesterone-induced luteinizing hormone (LH) and prolactin (PRL) surges in genetically obese OLETF (Otsuka-Long-Evans-TokushimaFatty) rats. Experiments were performed on both normally fed and 3-day starved groups from ovariectomized OLETF rats and their controls LETO (Long-Evans-Tokushima-Otsuka). Starved LETO rats, whose leptin levels were less than 0.5 ng/ml, did not show a significant surge of either LH or PRL. Normally fed OLETF rats, whose leptin levels were 9.7 + 1.8 ng/ml, showed a significant but small surge for both LH and PRL. Interestingly, starved OLETF rats, whose leptin levels (4.1 + 0.7 ng/ml) were similar to those in normally fed LETO rats (3.3 + 0.4 ng/ml), had significantly greater surges of both hormones than normally fed OLETF group. This study demonstrates for the first time that the normalization of circulating leptin levels in female OLETF rats augments the steroid-induced LH and PRL surges, and also suggests a deleterious effect of hyperleptinemia on the reproductive axis. ß 2001 Harcourt Publishers Ltd
INTRODUCTION In humans, the in¯uence of nutrition and body composition on puberty and the reproductive function has long been recognized (Frisch and McArthur, 1974; Frisch, 1980). The critical weight hypothesis of the development of puberty states that when body weight (BW) reaches a certain level, puberty takes place (Frisch and McArthur, 1974). In this context, leptin, a recently discovered adipocyte-derived hormone (Zhang et al., 1994; Pelleymounter et al., 1995), has been reported to positively regulate the reproductive axis in both humans and rodents, and is considered to serve as a critical metabolic
Received 23 August 2000 Accepted 20 September 2000 Correspondence to: Hajime Watanobe, M. D., Division of Internal Medicine, Center for Clinical Research, International University of Health and Welfare, 2600-1 Kitakanemaru, Otawara, Tochigi 324-8501, Japan. Fax: 81 287 24 1248; E-mail: [email protected]
signal linking adiposity and the reproductive system (Clarke and Henry, 1999). In apparent contradiction to this stimulatory effect of leptin on the endocrine axis, subnormal gonadal function has been demonstrated in obesity with hyperleptinemia. Obese women have an increased rate of gonadal dysfunction and infertility (Hartz et al., 1979). Also in men, serum testosterone levels are subnormal in obese males and return to normal with weight loss (Kley et al., 1979). In addition, there appears to be an inverse relationship between serum testosterone concentrations and BW (Amatruda et al., 1978; Kley et al., 1979). Similar ®ndings are observed in rodents as well. Impaired reproductive function has been reported in genetically obese strains of rats, such as the Zucker (Zucker and Zucker, 1961; Zucker and Zucker, 1962; Hemmes et al., 1978; Young et al., 1982; Whitaker et al., 1983; Doherty et al., 1985) and OLETF (Otsuka-Long-Evans-Tokushima-Fatty) (Shimizu et al., 1998) animals. Since it is known that both the Zucker and OLETF rats have hyperleptinemia (Rayner et al., 1997; 45
Watanobe et al.
Shimizu et al., 1998), an interesting possibility may thus arise that hyperleptinemia may act negatively on the reproductive function. Indeed, a recent study in humans demonstrated a detrimental effect of excess leptin on the reproductive system of girls (Bouvattier et al., 1998). As we are not aware of any previous study examining the effect of hyperleptinemia on the reproductive axis of genetically obese animals, in the present study we tested the in¯uence of fasting on the estradiol (E2)/progesterone (P)-induced luteinizing hormone (LH) and prolactin (PRL) surges in female OLETF rats, and compared the results with those from their controls LETO (Long-Evans-TokushimaOtsuka).
(a) 40 Plasma LH (ng/ml)
46
30 20 10 0 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00
MATERIALS AND METHODS
Neuropeptides (2001) 35(1), 45±49
(b) Plasma PRL (ng/ml)
All of the following experiments were conducted in accordance with the Guidelines for Animal Experimentation, Hirosaki University. Female OLETF and LETO rats, 4 weeks of age, were obtained from Tokushima Research Institute, Otsuka Pharmaceutical (Tokushima, Japan). They were housed in an air-conditioned room with controlled lighting (light 08:00±20:00). The animals were given free access to laboratory chow and tap water, unless otherwise indicated. The rats were examined daily for vaginal opening from 30 days of age. When the animals were 7 weeks of age, we started taking vaginal smears by lavage for 3 weeks. At the age of about 10 weeks, both OLETF and LETO rats were bilaterally ovariectomized under light ether anesthesia, and subjected to experiments about 2 weeks later (at the age of about 12 weeks). At about 08:00 of the day 72 h before the experiment, we measured BW of every animal. Two days prior to the experiment, the animals were implanted with a jugular vein catheter ®lled with heparin solution under light ether anesthesia. At 10:00 on the same day, every rat was injected subcutaneously with 150 mg/kg BW of E2 benzoate (Mochida Pharmaceutical Co., Ltd., Tokyo, Japan). Experiments were performed on both normally fed and 3-day-starved groups from both OLETF and LETO strains. At about 08:00 on the day of the experiment, BW was measured and the jugular vein catheter was exteriorized for frequent blood sampling. At 09:00, 20 mg/kg BW of P (Mochida Pharmaceutical Co., Ltd.) was injected intramuscularly. Blood samples (200 ml) were collected every 30 min over a total period of 420 min (11:00±18:00). At 11:00, additional 200 ml of blood was drawn to measure leptin as well. To prevent the loss of circulating plasma volume, 0.9 % NaCl was injected intravenously immediately after each blood collection in the same volume as that drawn. The blood was collected in EDTA-2Na (2.5 mg/ml)-containing tubes, centrifuged, and the plasma was stored at ÿ 708C until assayed for leptin, LH, and PRL.
OLETF (normally fed) OLETF (3-day starved) LETO (normally fed) LETO (3-day starved)
300
200
100
0 11:00 12:00 13:00 14:00 15:00 16:00 17:00 18:00 Time of day (h) Fig. 1 Effects of 3-day starvation on steroid-induced LH and PRL surges in OLETF and LETO rats. The number of rats examined was 7±9 per group. , significantly different vs OLETF (normally fed) group. *, significantly different vs LETO (3-day starved) group. Where standard errors are not shown, they were smaller than the symbols. For further details, see text.
Plasma leptin levels were determined by a rat leptin RIA kit produced by Linco Research (St. Louis, MO, USA). The sensitivity of the assay was 0.5 ng/ml. LH and PRL levels were determined by RIA using the reagents kindly donated by Dr. A. F. Parlow (NIDDK). Rat LH-RP-3 and PRL-RP-3 were used as the standards. Sensitivity of the LH assay was 0.2 ng/ml, and that of PRL assay was 0.8 ng/ml. Both the intra- and interassay coef®cients of variation were less than 10 % in all the three assays. Results were expressed as the mean + S.E.M. One-way or two-way ANOVA followed by Scheffe's post hoc test was then used to analyze the data. Differences were considered signi®cant if P was smaller than 0.05. RESULTS OLETF rats showed a small but signi®cant delay in vaginal opening, i.e. at 40.5 + 0.8 days (n 20) in OLETF vs 35.4 + 0.7 days (n 20) in LETO. In addition, the estrous ß 2001 Harcourt Publishers Ltd
Normalization of circulating leptin levels by fasting improves the reproductive function
47
Table 1 BW and plasma leptin levels in the 4 experimental groups BW (g) 1)
Group
Number of rats
72 h before
Day of experiment2)
Percent change in BW
Plasma leptin3) (ng/ml)
OLETF (normally fed) OLETF (3-day starved) LETO (normally fed) LETO (3-day starved)
7 8 8 9
270 274 218 214
281 233 225 184
ÿ ÿ
9.7 + 1.8** 4.1 + 0.7 3.3 + 0.4 < 0.5
+ + + +
12* 13* 10 9
+ + + +
13* 10* 11 7
4 ( + 1) 15 ( + 3) 3 ( + 1) 14 ( + 3)
1) BW was measured at 08:00 of the day 72 h before the experiment. 2) BW was measured at 08:00 of the day of the experiment. 3) Leptin was measured in the samples obtained at 11:00 of the day of the experiment. *, statistically signi®cant vs LETO rats with the same treatment. **, statistically signi®cant vs the remaining 3 groups.
cycles of OLETF rats were longer and less regular than those of LETO rats. The mean cycle length between 7±10 weeks of age was signi®cantly longer in OLETF [5.8 + 0.3 days (n 20)] than in LETO [4.3 + 0.1 days (n 20)]. This lengthening of the estrous cycle was generally due to prolongation of diestrus. Table 1 shows the data of BW and plasma leptin levels in the 4 groups examined in this study. The percent reduction in BW after fasting was similar between 3-day-starved OLETF and 3-day starved LETO groups. With respect to plasma leptin concentrations, normally fed OLETF group had a 2.9 times higher level of the hormone than normally fed LETO group. Leptin levels in the 3-day starved OLETF group were less than half those in the normally fed OLETF group, and statistically indistinguishable from the levels in normally fed LETO group. Plasma leptin in 3-day-starved LETO group was below the assay sensitivity. Fig.1 shows the temporal changes in plasma LH and PRL in all 4 groups. With respect to LH levels [Fig.1 (a)], the normally fed LETO group showed signi®cantly higher levels of the hormone (LH surge) between 14:00±18:00 than at 11:00. In agreement with our previous data obtained from female Wistar rats (Kohsaka et al., 1999; Watanobe et al., 1999a; Watanobe et al., 1999b), the 3 day starvation imposed on LETO rats completely abolished LH surge. However LH data in OLETF rats formed a marked contrast to those in LETO rats. During the period of 17:00±18:00, the plasma LH levels in the normally fed OLETF group were signi®cantly higher than those in the 3-day-starved LETO group, and formed a small surge-like secretion of LH. The magnitude of LH surge in the OLETF (normally fed) group was markedly smaller than that in the LETO (normally fed) group. Even so, it is very interesting to note that the 3-day starvation loaded on OLETF rats did not suppress, but signi®cantly augmented LH surge. Although the magnitude of LH surge in the OLETF (3-day-starved) group was still signi®cantly smaller than that in normally fed LETO group during the period of 16:00±18:00, the surge magnitude of the former group signi®cantly exceeded that of OLETF (normally fed) group between 15:30±18:00. ß 2001 Harcourt Publishers Ltd
A similar ®nding was also observed for PRL surge [Fig.1 (b)]. In agreement with our previous observations in female Wistar rats (Kohsaka et al., 1999; Watanobe et al., 1999a; Watanobe et al., 1999b), the 3-day-starved LETO group did not show a signi®cant PRL surge, while normally fed LETO group exhibited a robust surge of the hormone. Similar to that observed for LH surge, effects of altered nutritional states on PRL surge contrasted between OLETF and LETO rats. Although the OLETF (normally fed) group had a signi®cant PRL surge, its magnitude was markedly smaller than that in normally fed LETO group. The magnitude of PRL surge in the 3-day-starved OLETF group was signi®cantly greater than that in the normally fed OLETF group, even though the surge magnitude of the former group was still signi®cantly smaller than that of normally fed LETO group between 15:30±18:00. DISCUSSION In this study, we examined whether alterations in circulating leptin levels affect the reproductive function of female OLETF rats, by employing the E2/P-induced LH and PRL surges as an indicator. Firstly, prior to ovariectomy we checked the postnatal day of vaginal opening and estrous cycles of OLETF rats in comparison to their controls LETO. We found that in OLETF rats the vaginal opening was signi®cantly delayed and the estrous cycles were signi®cantly longer. These results suggest impaired reproductive function of female OLETF rats as compared to LETO rats. Similarly delayed vaginal opening and longer estrous cycles were also reported for female Zucker rats (Whitaker et al., 1983). The present data of LH and PRL surges were even more interesting. The effects of starvation on the hormonal surges in LETO rats were essentially the same as those in Wistar rats which we described in our previous studies (Kohsaka et al., 1999; Watanobe et al., 1999a; Watanobe et al., 1999b). By contrast, in OLETF rats the effects of altered nutritional states on the LH and PRL surges were the reverse of those in LETO rats. As compared to the magnitudes of hormonal surges in the LETO (normally Neuropeptides (2001) 35(1), 45±49
48
Watanobe et al.
fed) group, the surges were very small in the OLETF (normally fed) group. However, these surges in OLETF rats were signi®cantly increased in magnitude after 3 days' starvation. Plasma leptin level in 3-day-starved OLETF group was 42% of that in normally fed OLETF group, and was statistically indistinguishable from that in LETO (normally fed) group. These results strongly suggest that in female OLETF rats the hyperleptinemia deriving from obesity may exert a deleterious effect on the reproductive axis. Impaired reproductive function has previously been reported in Zucker (Zucker and Zucker, 1961; Zucker and Zucker, 1962; Hemmes et al., 1978; Young et al., 1982; Whitaker et al., 1983; Doherty et al., 1985) and OLETF (Shimizu et al., 1998) rats. For both strains, lower serum testosterone levels than in controls were reported in males (Young et al., 1982; Shimizu et al., 1998). For Zucker rats, food restriction was reported to increase the incidence of fertility in males (Zucker and Zucker, 1962). However, the present study is the ®rst to demonstrate in female rats of a genetically obese strain that the normalization of circulating leptin levels by starvation can signi®cantly reinstate the E2/P-induced LH and PRL surges. Preexisting literature suggests the existence also in humans of a deleterious effect of obesity on the reproductive axis (Amatruda et al., 1978; Hartz et al., 1979; Kley et al., 1979; Bouvattier et al., 1998). Among the reports, the recent study of Bouvattier et al. (1998) appears to be especially important. They found that there were signi®cant negative correlations between gonadotropin-releasing hormone-induced gonadotropin responses and both the body mass index and circulating leptin levels. Among the data in this study, it is worth noting that the magnitudes of LH and PRL surges in the OLETF (3-day starved) group were still signi®cantly smaller than those in LETO (normally fed) group, even though the plasma leptin levels in these two groups were statistically the same. Although we have no clear explanation for this difference, two possibilities may be offered. First, the treatment of 3-day starvation might have been too short for the endocrine hypothalamus of OLETF rats to escape from the detrimental effect of hyperleptinemia. It is very likely that the impairment of gonadal function in obesity ensues as a result of sustained hyperleptinemia in both humans and rodents. Thus, maintenance of normal leptin levels for a longer period of time might have been necessary for OLETF rats to achieve a complete normalization of the steroid-induced LH and PRL surges. Another possible mechanism is that the 3-day starvation might have been too severe a protocol to reduce circulating leptin levels. It is likely that the fasting treatment decreases not only leptin but also other metabolic ingredients. In this context, Schneider et al. (1998) recently reported an important ®nding from using hamsters that the alleged stimulatory Neuropeptides (2001) 35(1), 45±49
action of leptin on the reproductive axis is indirect and requires oxidation of metabolic fuels such as glucose and fatty acid. Thus, instead of the acute 3-day starvation, food restriction for a longer period which is expected to cause a slow and steady decline in leptin levels without exhausting the reservoir of metabolic fuels, might have provided OLETF rats with a normal expression of LH and PRL surges. At any rate, it is an important question whether hyperleptinemia is the sole factor mediating the impaired reproductive function in obesity or if other metabolic factor(s) is/are also involved. Further detailed studies are necessary to clarify this interesting issue. In summary, this study demonstrated for the ®rst time that the normalization by fasting of circulating leptin levels in obese OLETF female rats signi®cantly augments the E2/P-induced LH and PRL surges. The present results further support the deleterious effect of hyperleptinemia on the reproductive function. ACKNOWLEDGEMENTS We wish to thank the National Hormone and Pituitary Program of NIDDK and Dr. A. F. Parlow for the generous donation of reagents for rat LH and PRL RIAs. REFERENCES Amatruda JM, Harman SM, Pourmotabbed G, Lockwood DH (1978) Depressed plasma testosterone and fractional binding of testosterone in obese males. J Clin Endocrinol Metab 47: 268±271. Bouvattier C, Lahlou N, Roger M, Bougneres P (1998) Hyperleptinaemia is associated with impaired gonadotropin responses to GnRH during late puberty in obese girls, not boys. Eur J Endocrinol 138: 653±658. Clarke IJ, Henry BA (1999) Leptin and reproduction. Rev Reprod 4: 48±55. Doherty PC, Baum MJ, Finkelstein JA (1985) Evidence of incomplete behavioral sexual differentiation in obese male Zucker rats. Physiol Behav 34: 177±179. Frisch RE, McArthur JW (1974) Menstrual cycles: fatness as a determinant of minimum weight for height necessary for their maintenance or onset. Science 185: 949±951. Frisch RE (1980) Pubertal adipose tissue: is it necessary for normal sexual maturation? Evidence from the rat and human female. Fed Proc 39: 2395±2400. Hartz AJ, Barboriak PN, Wong A, Katayama KP, Rimm AA (1979) The association of obesity with infertility and related menstrual abnormalities in women. Int J Obes 3: 57±73. Hemmes RB, Hubsch S, Pack HM (1978) High dosage of testosterone propionate increases litter production of the genetically obese male Zucker rat. Proc Soc Exp Biol Med 159: 424±427. Kley HK, Solbach HG, Mckinnan JC, Kruskemper HL (1979) Testosterone decrease and oestrogen increase in male patients with obesity. Acta Endocrinol 91: 553±563. Kohsaka A, Watanobe H, Kakizaki Y, Habu S, Suda T (1999) A significant role of leptin in the generation of steroid-induced luteinizing hormone and prolactin surges in female rats. Biochem Biophys Res Commun 254: 578±581. Pelleymounter MA, Cullen MJ, Baker MB, Hecht R, Winters D, Boone T, Collins F (1995) Effects of the obese gene product on body weight regulation in ob/ob mice. Science 269: 540±543.
ß 2001 Harcourt Publishers Ltd
Normalization of circulating leptin levels by fasting improves the reproductive function
Rayner DV, Dalgliesh GD, Duncan JS, Hardie LJ, Hoggard N, Trayhurn P (1997) Postnatal development of the ob gene system: elevated leptin levels in suckling fa/fa rats. Am J Physiol 273: R446±R450. Schneider JE, Goldman MD, Tang S, Bean B, Ji H, Friedman MI (1998) Leptin indirectly affects estrous cycles by increasing metabolic fuel oxidation. Horm Behav 33: 217±228. Shimizu H, Ohtani K-I, Uehara Y, Abe Y, Takahashi H, Tsuchiya T, Sato N, Ibuki Y, Mori M (1998) Orchiectomy and response to testosterone in the development of obesity in young Otsuka-Long-EvansTokushima Fatty (OLETF) rats. Int J Obes Relat Metab Disord 22: 318±324. Watanobe H, SchioÈth HB, Wikberg JE S, Suda T (1999a) The melanocortin 4 receptor mediates leptin stimulation of luteinizing hormone and prolactin surges in steroid-primed ovariectomized rats. Biochem. Biophys. Res. Commun 257: 860±864. Watanobe H, Suda T, Wikberg JE S, Schioth HB (1999b) Evidence that physiological levels of circulating leptin exert a stimulatory effect on
ß 2001 Harcourt Publishers Ltd
49
luteinizing hormone and prolactin surges in rats. Biochem Biophys Res Commun 263: 162±165. Whitaker EM, Shaw MA, Herrey GR (1983) Plasm oestradiol-17 beta and testosterone concentrations as possible causes of the infertility of congenitally obese Zucker rats. J Endocrinol 99: 485±490. Young RA, Frink R, Longcope C (1982) Serum testosterone and gonadotropins in the genetically obese male Zucker rat. Endocrinology 111: 977±981. Zhang Y, Proenca R, Maffei M, Barone M, Leopold L, Friedman JM (1994) Positional cloning of the mouse obese gene and its human homologue. Nature 372: 425±432. Zucker LM, Zucker TF (1961) Fatty, a new mutation in the rat. J Hered 52: 275±278. Zucker TF, Zucker LM (1962) Hereditary obesity in the rat associated with high serum fat and cholesterol. Proc Soc Exp Biol Med 110: 165±171.
Neuropeptides (2001) 35(1), 45±49