www.elsevier.com/locate/brainresBrain Research 993 (2003) 172–176

Research report

D-chiro-Inositol enhances effects of hypothalamic toxin gold-thioglucose

Fumiko Isodaa,b, Laura Shiryc, Jeffrey Abergela,b, Geoffry Allanc, Charles Mobbsa,b,*

aNeurobiology of Aging Laboratories, Fishberg Center for Neurobiology, Mount Sinai School of Medicine, New York, NY 10029,USAbDepartment of Geriatrics, Mount Sinai School of Medicine, New York, NY 10029,USA

c Insmed Pharmaceuticals, Richmond, VA, USA

Accepted 8 September 2003

Abstract

D-chiro-Inositol (DCI) enhances reproductive function in insulin-resistant women with polycystic ovarian disease and enhances the effects

of insulin in the periphery, suggesting that this compound may act in part by sensitizing the hypothalamus to effects of insulin. Effects of

gold-thioglucose (GTG) to produce hypothalamic lesions and subsequent obesity are insulin-dependent, suggesting that responses to GTG

may be a marker of hypothalamic sensitivity to insulin. To assess these hypotheses, the present study assessed if DCI would enhance the

ability of a subthreshhold dose of GTG to produce hypothalamic lesions and subsequent obesity. At the subthreshhold dose used (0.4 mg/kg

i.p.), injection of GTG produced no subsequent effect on body weight compared to saline; similarly, at the dose of DCI used (10 mg/kg/day in

drinking water), DCI produced no effect on body weight. In contrast, when given to mice exposed to DCI, this dose of GTG produced

significant increase in body weight and evidence of an enhanced medial arcuate hypothalamic lesion.

D 2003 Elsevier B.V. All rights reserved.

Theme E: Endocrine and autonomic regulation

Topic: Neuroendocrine regulation: other

Keywords: Obesity; Insulin resistance; Polycystic ovarian syndrome; Hypothalamus

1. Introduction insulin into the area of the hypothalamus [17,26]; and (iv)

Women with polycystic ovarian syndrome (PCOS) ex-

hibit profound impairments in fertility and impaired

responses to insulin, but the mechanism linking insulin

resistance with reproductive function remains obscure.

One hypothesis linking these impairments is that PCOS

entails global insulin resistance, including reduced hypotha-

lamic responses to insulin, and the reproductive impair-

ments in PCOS arise secondary to impaired hypothalamic

responses to insulin. This hypothesis is supported by the

following observations: (i) PCOS clearly entails impair-

ments in hypothalamic regulation of neuroendocrine func-

tion [1,13,24] and these hypothalamic impairments appear

to be necessary for the observed reproductive impairments

[2]; (ii) hypothalamic neurons are sensitive to insulin [25];

(iii) gonadotropin secretion is stimulated by infusion of

0006-8993/$ - see front matter D 2003 Elsevier B.V. All rights reserved.

doi:10.1016/j.brainres.2003.09.008

* Corresponding author. Neurobiology of Aging Laboratories, Fishberg

Center for Neurobiology, Mount Sinai School of Medicine, 1 Gustave Levy

Pl., New York, NY 10029,USA. Tel.: +1-212-659-5929; fax: +1-212-849-

2510.

E-mail address: charles.mobbs@mssm.edu (C. Mobbs).

production of insulin resistance through global [8] or brain-

specific [7] genetic manipulation produces profound repro-

ductive impairments. Nevertheless, it has still not been

definitively proven that hypothalamic insulin resistance is

a primary cause of PCOS or if insulin resistance instead may

arise secondary to peripheral reproductive defects.

Insight into the mechanism underlying PCOS was pro-

vided by the unexpected discovery that administration of the

compound D-chiro-inositol (DCI) dramatically improved

reproductive function in women with PCOS while only

modestly improving whole-body insulin sensitivity [19].

DCI was originally isolated as a potential mediator of

insulin action [14,22], and insulin increases plasma levels

of DCI by 8.8-fold [23]. One hypothesis to explain these

results would be that DCI produces more pronounced

effects on hypothalamic sensitivity to insulin than on

peripheral sensitivity to insulin. One way to test this

hypothesis would be to assess if DCI enhances hypotha-

lamic-specific processes that are insulin-dependent. One

such process is the effect of gold-thioglucose (GTG), which

at low doses produces a hypothalamic lesion (followed by

obesity) that is completely dependent on insulin [12]. The

F. Isoda et al. / Brain Research 993 (2003) 172–176 173

magnitude of GTG-induced hypothalamic lesions has been

used to assess effects of estrogen on hypothalamic processes

[28]. We therefore hypothesized that if DCI enhances

hypothalamic sensitivity to insulin, then DCI should also

enhance the effect of a subthreshold dose of GTG to

produce body weight increase and reduce expression of

genes expressed in the GTG-sensitive hypothalamic field. In

the present study, we found that this was indeed the case.

Fig. 1. Total body weight gain 6 months after injection of 0.4 mg/kg b.w.

GTG (+) or saline (�) in mice that had previously been treated with DCI 10

mg/kg/day in the drinking water (+) or drinking water only (�). Different

letters indicate differences as indicated by ANOVA followed by the Fisher

PLSD test ( p< 0.05).

2. Materials and methods

We had previously observed that male CBA mice are

more sensitive to GTG than other strains of mice and thus

GTG produces more reliable obesity especially at lower

doses compared to other strains [3,18]. Therefore, we

carried out a dose–response curve of GTG on body weight

and observed that a dose of 0.6 mg/kg b.w. GTG (i.p.)

would produce a significant body weight gain over saline-

injected controls, but a dose of 0.4 mg/kg dose would not

produce any effect on body weight in this strain. Thus, we

reasoned that if DCI increased the effectiveness of GTG by

only 50%, this would be detectable if GTG were adminis-

tered at a dose of 0.4 mg/kg b.w. Therefore, adult (6–

8 weeks) male CBA mice were obtained from Jackson

Laboratories and maintained at a 12/12 light/dark cycle with

ad lib access to food (Purina Lab chow) and water. At 12

weeks of age, mice were placed on DCI (10 mg/kg/day b.w.

in drinking water, a dose obtained by extrapolating from

effective doses in humans [19]) or water alone for 2 weeks

before a single injection of GTG. After 2 weeks of exposure

to DCI (at 9–11 weeks of age), mice were given a single i.p.

injection of GTG, 0.4 mg/kg. After GTG injection, mice

were no longer exposed to DCI so that subsequent changes

in body weight would be attributable only to the acute

effects DCI on sensitivity to GTG, not subsequent effects of

DCI on body weight. Body weight was monitored for 6

months after injection, then mice were sacrificed and

hypothalamic gene expression was determined by in situ

hybridization, as previously described [3].

3. Results and discussion

As shown in Fig. 1, over the 6 months after GTG or

saline was administered, all groups gained about 6 g, except

the group administered both GTG and DCI, which gained

about 10 g, a significant 80% enhancement of body weight

gain ( p < 0.05). Total body weight, blood glucose and

epididymal fad pad weight taken at the time of sacrifice

(again, 6 months after the GTG injection) showed a similar

trends, though these effects were not statistically significant

due to variability.

It was anticipated that this effect of DCI would be

mediated by a potentiation of GTG-induced hypothalamic

damage by DCI. Because the dose of GTG was near

threshold, no significant lesion was visible 6 months after

injection (Fig. 2A). Nevertheless, hypothalamic levels of

agouti-related peptide (AGRP) mRNA were significantly

reduced by GTG in the presence of DCI, but not in the

absence of DCI (Fig. 2B and C). This result is informative

because AGRP is produced in the medial aspect of the

arcuate nucleus, which lies on the outer edge of the GTG

lesion [4]; thus, the expression of AGRP might be a very

sensitive indicator of the size of the GTG lesion. On the

other hand, as shown in Fig. 3, the reduction of POMC

mRNA by GTG was not influenced by DCI. It is notable that

GTG reduced POMC mRNA equally in the presence or

absence of DCI, but only lesions produced in the presence of

DCI lead to obesity. This result tends to argue against a key

role of POMC in mediating the obesity-producing effects of

the GTG lesion, in contrast to our previous conclusions [4].

Since POMC mRNA is produced in the center of the GTG

field, this molecular marker, similar to GTG, may not

provide a sensitive and precise indicator of hypothalamic

damage as a marker (viz., AGRP) that is expressed on the

outer edge of the lesion. It should also be noted that, since

GTG also produces lesions in the brainstem [27], we cannot

exclude a role for these structures in mediating effects of

DCI on GTG-induced obesity. Furthermore, the mechanism

by which the obesity develops is unclear, though possibly

involves the production of hyperphagia, which is observed

after exposures to higher doses of GTG [4]. However, in the

present study, we did not measure food intake.

Taken together, these studies clearly demonstrate that

DCI enhances body weight gain produced by a subthreshold

Fig. 2. Hypothalamic effects of DCI and GTG, 6 months after injection of 0.4 mg/kg b.w. GTG (+) or saline (�) in mice that had previously been treated with

DCI 10 mg/kg/day in the drinking water (+) or drinking water only (�). (A) Cresyl violet stain indicating no apparent lesion due to the low dose of GTG used

and the 6-month delay between injection and sacrifice. (B) Micrograph of film indicating AGRP signal in arcuate nucleus after by in situ hybridization using

the AGRP probe. (C) Quantification of relative AGRP mRNA (in arbritrary density units). Different letters indicate differences as indicated by ANOVA

followed by the Fisher PLSD test ( p< 0.05).

F. Isoda et al. / Brain Research 993 (2003) 172–176174

dose of GTG, and that this effect appears to be mediated by

enhanced hypothalamic damage, as reflected particularly by

reduction in AGRP mRNA. Whether DCI generally in-

creases the size of the GTG lesion (not strongly supported

by the POMC mRNA data, though these data may not be

highly sensitive to size of the lesion) or whether DCI more

specifically enhances the effect of GTG on specific neuronal

populations (especially those in the medial aspect of the

arcuate nucleus) remains to be determined. In any case, the

results clearly support the hypothesis that, to the extent that

the GTG lesion reflects an insulin-sensitive process involv-

ing a glucose transport mechanism, this process is particu-

larly enhanced by DCI.

Although effects of GTG, and enhancement of those

effects by DCI, are ultimately manifest in disturbances of

neural function leading to obesity, several lines of evidence

suggest that these effects of GTG may be indirect, and that

the direct effect of GTG and DCI may actually be on glial

cells. First, whereas little evidence supports a role for insulin

in stimulating neuronal glucose uptake, considerable evi-

dence suggests that insulin may directly stimulate glucose

uptake and metabolsm in glial cells [9,10]. Second, glial

toxins attenuate the size of GTG lesions [27] and responses

to 2-deoxglucose [29]. Third, GTG lesions are attenuated by

blockers of glucose transport [5] and are insulin-dependent

[11]. Fourth, the first morphological evidence of the GTG

lesion is observed at the blood–brain barrier, which includes

glia [16]; only later is evidence of neuronal damage ob-

served [6]. Finally, high-capacity glucose transporters, while

observed in the hypothalamus [15,20], are probably

Fig. 3. Relative pro-opiomelanocorticotropin (POMC) mRNA (in arbritrary

density units), as determined by in situ hybridization, 6 months after

injection of 0.4 mg/kg b.w. GTG (+) or saline (�) in mice that had

previously been treated with DCI 10 mg/kg/day in the drinking water (+) or

drinking water only (�). Different letters indicate differences as indicated

by ANOVA followed by the Fisher PLSD test ( p< 0.05).

F. Isoda et al. / Brain Research 993 (2003) 172–176 175

expressed primarily at much higher levels in glia than in

neurons [16]. These data suggest that the physiological and

molecular responses observed in the present study may be

due to enhancement of glial insulin sensitivity by DCI,

leading to increased glial uptake of GTG. This hypothesis is

particularly intriguing in view of studies suggesting that

hypothalamic glia may play an important role in the

regulation of reproductive function [21]. These observations

raise the possibility that DCI may influence symptoms of

polycystic ovarian syndrome through action on hypotha-

lamic glial cells.

References

[1] M. Barontini, M.C. Garcia-Rudaz, J.D. Veldhuis, Mechanisms of hy-

pothalamic–pituitary–gonadal disruption in polycystic ovarian syn-

drome, Arch. Med. Res. 32 (2001) 544–552.

[2] S.L. Berga, T.L. Daniels, Can polycystic ovary syndrome exist with-

out concomitant hypothalamic dysfunction? Semin. Reprod. Endocri-

nol. 15 (1997) 169–175.

[3] H.T. Bergen, N. Monkman, C.V. Mobbs, Injection with gold thioglu-

cose impairs sensitivity to glucose: evidence that glucose-responsive

neurons are important for long-term regulation of body weight, Brain

Res. 734 (1996) 332–336.

[4] H.T. Bergen, T.M. Mizuno, J. Taylor, C.V. Mobbs, Hyperphagia and

weight gain after gold-thioglucose: relation to hypothalamic neuro-

peptide Y and proopiomelanocortin, Endocrinology 139 (1998)

4483–4488.

[5] D.F. Brown, J.M. Viles, Systemic phlorizin prevents gold thioglucose

necrosis in the ventromedial hypothalamus, Brain Res. Bull. 8 (1982)

347–351.

[6] D.F. Brown, J.M. Viles, An ultrastructural study of gold thioglucose

lesion formation in the ventromedial hypothalamus: subnecrotic gold

thioglucose dose, J. Neuropathol. Exp. Neurol. 45 (1986) 20–27.

[7] J.C. Bruning, D. Gautam, D.J. Burks, J. Gillette, M. Schubert, P.C.

Orban, R. Klein, W. Krone, D. Muller-Wieland, C.R. Kahn, Role of

brain insulin receptor in control of body weight and reproduction,

Science 289 (2000) 2122–2125.

[8] D.J. Burks, J.F. de Mora, M. Schubert, D.J. Withers, M.G. Myers,

H.H. Towery, S.L. Altamuro, C.L. Flint, M.F. White, IRS-2 pathways

integrate female reproduction and energy homeostasis, Nature 407

(2000) 377–382.

[9] D.W. Clarke, F.T. Boyd Jr., M.S. Kappy, M.K. Raizada, Insulin

binds to specific receptors and stimulates 2-deoxy-D-glucose uptake

in cultured glial cells from rat brain, J. Biol. Chem. 259 (1984)

11672–11675.

[10] D.W. Clarke, F.T. Boyd Jr., M.S. Kappy, M.K. Raizada, Insulin stim-

ulates macromolecular synthesis in cultured glial cells from rat brain,

Am. J. Physiol. 249 (1985) C484–C489.

[11] A.F. Debons, I. Krimsky, A. From, H. Pattinian, Diabetes-induced

resistance of ventromedial hypothalamus to damage by gold thio-

glucose: reversal by adrenalectomy, Endocrinology 95 (1974)

1636–1641.

[12] A.F. Debons, I. Krimsky, M.L. Maayan, K. Fani, F.A. Jemenez, Gold

thioglucose obesity syndrome, Fed. Proc. 36 (1977) 143–147.

[13] A. Lanzone, M. Guido, M. Ciampelli, A.M. Fulghesu, V. Pavone, C.

Proto, A. Caruso, S. Mancuso, Evidence of a disturbance of the

hypothalamic–pituitary–adrenal axis in polycystic ovary syndrome:

effect of naloxone, Clin. Endocrinol. (Oxf.) 45 (1996) 73–77.

[14] J. Larner, L.C. Huang, C.F. Schwartz, A.S. Oswald, T.Y. Shen, M.

Kinter, G.Z. Tang, K. Zeller, Rat liver insulin mediator which

stimulates pyruvate dehydrogenase phosphate contains galactos-

amine and D-chiroinositol, Biochem. Biophys. Res. Commun.

151 (1988) 1416–1426.

[15] C. Leloup, M. Arluison, N. Lepetit, N. Cartier, P. Marfaing-Jallat, P.

Ferre, L. Penicaud, Glucose transporter 2 (GLUT 2): expression in

specific brain nuclei, Brain Res. 638 (1994) 221–226.

[16] P.J. Magistretti, L. Pellerin, D.L. Rothman, R.G. Shulman, Energy on

demand, Science 283 (1999) 496–497.

[17] D.W. Miller, D. Blache, G.B. Martin, The role of intracerebral insulin

in the effect of nutrition on gonadotrophin secretion in mature male

sheep, J. Endocrinol. 147 (1995) 321–329.

[18] T.M. Mizuno, S.P. Kleopoulos, H. Bergen, J.L. Roberts, C.A. Priest,

C.V. Mobbs, Gold thioglucose-induced obesity is associated with

decreased hypothalamic pro-opiomelanocortin (POMC) mRNA and

elevated glial fibrillary acidic protein (GFAP) mRNA, Endocr. Soc.

Abstr. 80 (1998) 431–432.

[19] J.E. Nestler, D.J. Jakubowicz, P. Reamer, R.D. Gunn, G. Allan, Ov-

ulatory and metabolic effects of D-chiro-inositol in the polycystic

ovary syndrome, N. Engl. J. Med. 340 (1999) 1314–1320.

[20] C. Ngarmukos, E.L. Baur, A.K. Kumagai, Co-localization of GLUT1

and GLUT4 in the blood–brain barrier of the rat ventromedial hypo-

thalamus, Brain Res. 900 (2001) 1–8.

[21] S.R. Ojeda, Y.J. Ma, B.J. Lee, V. Prevot, Glia-to-neuron signaling and

the neuroendocrine control of female puberty, Recent Prog. Horm.

Res. 55 (2000) 197–223 (discussion 223–4).

[22] H.K. Ortmeyer, N.L. Bodkin, K. Lilley, J. Larner, B.C. Hansen,

Chiroinositol deficiency and insulin resistance: I. Urinary excretion

rate of chiroinositol is directly associated with insulin resistance in

spontaneously diabetic rhesus monkeys, Endocrinology 132 (1993)

640–645.

[23] R.E. Ostlund Jr., J.B. McGill, I. Herskowitz, D.M. Kipnis, J.V. San-

tiago, W.R. Sherman, D-chiro-Inositol metabolism in diabetes melli-

tus, Proc. Natl. Acad. Sci. U. S. A. 90 (1993) 9988–9992.

[24] C.L. Pastor, M.L. Griffin-Korf, J.A. Aloi, W.S. Evans, J.C. Mar-

shall, Polycystic ovary syndrome: evidence for reduced sensitivity

of the gonadotropin-releasing hormone pulse generator to inhibition

by estradiol and progesterone, J. Clin. Endocrinol. Metab. 83

(1998) 582–590.

F. Isoda et al. / Brain Research 993 (2003) 172–176176

[25] D. Spanswick, M.A. Smith, S. Mirshamsi, V.H. Routh, M.L.

Ashford, Insulin activates ATP-sensitive K+ channels in hypo-

thalamic neurons of lean, but not obese rats, Nat. Neurosci. 3

(2000) 757–758.

[26] T. Tanaka, S. Nagatani, D.C. Bucholtz, S. Ohkura, H. Tsukamura, K.

Maeda, D.L. Foster, Central action of insulin regulates pulsatile lutei-

nizing hormone secretion in the diabetic sheep model, Biol. Reprod.

62 (2000) 1256–1261.

[27] J.K. Young, The glial drug methionine sulfoximine reduces goldthio-

glucose lesions in mice, Brain Res. Bull. 22 (1989) 929–936.

[28] J.K. Young, D.M. Nance, R.A. Gorski, Effects of estrogen upon

hypothalamic vulnerability to goldthioglucose in mice, Brain Res.

Bull. 3 (1978) 231–235.

[29] J.K. Young, J.H. Baker, M.I. Montes, The brain response to 2-deoxy

glucose is blocked by a glial drug, Pharmacol. Biochem. Behav. 67

(2000) 233–239.