Aldose reductase inhibition prevents galactose-induced ovarian dysfunction in the Sprague-Dawley rat William R. Meyer, MD,"' b Michael B. Doyle, MD," Jamie A. Grifo, MD, PhD," Kenneth J. Lipetz, PhD,b Peter J. Oates, PhD,c Alan H. DeCherney, MD," and Michael P. Diamond, MD" New Haven and Groton, Connecticut, and Bethesda, Maryland OBJECTIVE: Our objective was to determine whether impaired ovarian function induced by short-term creation of a galactosemic state in the rat might be prevented by the coadministration of an aldose reductase inhibitor. STUDY DESIGN: Prepubertal Sprague-Dawley rats were fed four different diets including (1) control, (2) 40% galactose, (3) 40% galactose and an aldose reductase inhibitor, and (4) an aldose reductase inhibitor with the control diet. Percentage germinal vesicle breakdown, postovulatory oocyte quantities, hormonal parameters, ovarian histologic evaluation, and ovarian galactitol concentrations were determined. RESULTS: The galactose-fed animals (group 2) had decreased germinal vesicle breakdown (47%) versus control (69%, p < 0.05). Galactose-exposed animals had significantly decreased quantities of postovulatory eggs (6.4 per animal) after menotropin ovarian stimulation in comparison with controls (14.1, P < 0.01). In rats exposed to high dietary levels of galactose (group 2) ovarian galactitol concentrations were Significantly higher (protein 42.12 fLmol/gm versus 0.0 for controls, p < 0.005). When galactose-fed animals received the aldose reductase inhibitor, ovarian accumulation of galactitol was significantly reduced and the observed detrimental effects on the oocyte were prevented. CONCLUSION: Galactitol accumulation or metabolic flux through aldose reductase in galactosemic rodents may be involved in the demonstrated ovarian dysfunction. (AM J OasTET GVNECOl 1992; 167: 1837-43.) Key words: Galactitol, aldose reductase inhibitor, sorbinil, Sprague-Dawley rats Galactose is converted into usable glucose in the liver by three enzymatic reactions. Classic galactosemia, inherited as an autosomal recessive trait, results from the absence or decreased activity of one of these enzymes, galactose-I-phosphate uridyl transferase. In galactosemia excess galactose is reduced to galactitol by an aldose reductase. It has been hypothesized that the accumulation of galactose metabolites, including galactitol, may contribute to the clinical manifestations of galactosemia. Afflicted patients have signs and symptoms consistent with hepatic dysfunction, neurologic impairment, cataract formation, and ovarian failure. The origin of the pathophysiologic ovarian insult is unknown, but premature menopause occurs in homozygotic galactosemic women in spite of dietary galactose restriction. 1 Additionally, women who are heterozygotic From the Yale-New Haven Hospital, Yale University School of Medicine, a Montgomery Infertility Institute, b and the Central Research Division, Pfizer, Inc.' Presented in part at the Thirty-eighth Annual Meeting of the Society for Gynecologic Investigation. San Antonio, Texas, March 20-23, 1991. Reprint requests: William R. Meyer, MD. 10215 Fernwood Road, Suite 303, Bethesda, MD 20817. 6/6/41907 for the trait may experience both an increased incidence of infertility and an earlier menopause! It has been shown that rodents ingesting diets highly saturated with galactose mimic many of the observed clinical manifestations of galactosemic women, including decreased oocyte production." The objective of this study was to determine whether impaired ovarian function in the "galactosemic" rodent might be ameliorated by reducing galactitol accumulation or by preventing abnormal metabolic redox flux through aldose reductase by the coadministration of an aldose reductase inhibitor. Material and methods One hundred sixteen prepubertal female SpragueDawley rats were housed in group cages and maintained on a 12-hour-light and 12-hour-dark schedule under standardized conditions with full access to specific diets and water. Guidelines for care and treatment of animals as preapproved by the animal care divisions of Yale University School of Medicine and Pfizer, Inc., were adhered to. Animals began their diets between 21 and 24 days of age, immediately after weaning. They were maintained on the diet for 17 days. 1837 1838 Meyer et al. The animals were divided into the following groups: The control group (group 1) received standard rat laboratory chow pellets (United States BioChemical, Cleveland); group 2 received rat laboratory chow pellets with a 40% galactose addition as a caloric replacement4 ; group 3 received a diet similar to that of group 2 except for the addition of the aldose reductase inhibitor sorbinil (Pfizer, Central Research Division, Groton, Conn.) at a concentation of 2 gmlS kg of rat chow'; the diet of group 4 was identical to that of the control group with the exception of added sorbinil at 2 gmlS kg feed concentration. After 14 days on their specific diets, each of the rats in the study received an intramuscular injection of S IU of human menopausal gonadotropin. The animals were continued on their respective diets, and they were weighed daily. Oocyte maturational index, as assessed by germinal vesicle breakdown, was determined in rats 48 hours after their human menopausal gonadotropin injections. Rats were killed by decapitation, and ovaries were dissected free and placed in phosphate-buffered saline solution. During examination under a dissecting microscope 10 large preovulatory follicles were punctured with a stainless steel needle, and the oocytes were collected in phosphate-buffered saline solution at room temperature (21 C). Each oocyte-cumulus complex was denuded from its corna-~muls investments by repeated passage through a narrow-bore, hand-drawn glass pipette. Denuded oocytes were inspected 2 hours later. Oocytes that retained a germinal vesicle were considered to show inhibition of maturation. Examinations of oocytes were made by an independent observer. Oocyte maturation was expressed as a percentage of germinal vesicle breakdown. 5 The remaining animals in each respective group received 10 IU of human chorionic gonadotropin approximately S6 hours after the human menopausal gonadotropin injection. The human chorionic gonadotropin was used to enhance ovulation. Fifty-two of the remaining animals were killed by decapitation 69 to 72 hours after the human menopausal gonadotropin il1iection, and their respective ovaries and oviducts were surgically isolated. The ovaries were separated by sharp dissection. The dissected-free ovaries were embedded in paraffin and sections 2 J.Lm thick were stained with hematoxylin and eosin. Tissue sections were histologically examined for alterations in ovarian stroma and its respective follicles. A clinical pathologist made subjective comparisons between stromal and follicular alterations among groups in a blinded fashion. The oviducts were examined and flushed under the dissecting microscope (original magnification x 120) in phosphate-buffered saline solution. Flushing of the oviduct was performed with phosphate-buffered saline 0 December 1992 Am J Obstet Gynecol solution mixed with O.S% bovine serum albumin, which allowed easy visualization of oocyte-cumulus complexes. The oocytes were counted individually and quantified for each animal. The remaining 32 animals were killed individually by an intraperitoneal injection of a pentobarbital solution. At this time a vena caval puncture was performed to obtain approximately 2 ml of blood. Serum was separated from coagulated blood by centrifugation and stored at - 20 0 C. Progesterone and estradiol serum concentrations were determined with the Coat-A-Count radioimmunoassay (Diagnostic Products, Los Angeles). The progesterone assay has a sensitivity of 0.08 ng/ml, intraassay coefficient of variation precision of < 10%, and interassay coefficient of variation precision of < 10%. The estradiol assay has a sensitivity of 8 pg/ml and intraassay and interassay coefficient of variation precisions of < 10%. Specimens for each assay were run in duplicate and by the same individual. Ovaries were dissected free, weighed, and analyzed for ovarian myoinositol and galactitol concentrations by gas chromatography-mass spectometry.6 Statistical analysis included performing X2 methods for comparison of germinal vesicle breakdown. Analysis of variance was used for comparison of oocyte quantity, serum estradiol and progesterone levels, animal and ovarian weights, and intraovarian galactitol and myoinositol concentrations. Data are expressed as mean ± SEM. Significance was defined as p < O.OS. Results As depicted in Fig. 1, galactose exposure significantly reduced germinal vesicle breakdown from a control percentage of 69% to 47% (p < O.OS). With the supplementation of aldose reductase inhibitor, germinal vesicle breakdown was similar to normal rates (76%). Excess dietary galactose had an adverse effect on ovulation, with a drop from 14.1 ± 1.6 oocytes per rat in the controls to 6.4 ± l.0 In galactose-fed animals (p < 0.01). Ovulation rates in aldose reductase-supplemented galactose-fed rats were indistinguishable from rates in controls, 16.9 ± 3.1 oocytes per rat (Fig. 2). Ovarian galactitol concentrations in the control (group 1) and control plus sorbinil-fed (group 4) animals were below detectable levels (Table I). Animals exposed to a 40% galactose diet (group 2) had significantly elevated ovarian galactitollevels. The addition of the aldose reductase inhibitor (group 3) reduced ovarian galactitol levels. Estradiol and progesterone parameters are depicted in Table II. Galactose exposure lowered progesterone levels but only to a significant extent when the galactose diet included aldose reductase inhibitor. Otherwise, no other hormonal effect was detected by various diet exposure. In comparison with age-matched controls, rats fed Galactitol Volume 167 Number 6 ~ Io 80 w m 60 1839 c a: w .... u in ~ 40 20 GAL GALIARI ARI GROUP Fig. 1. Maturity index, depicted as percentage germinal vesicle breakdown, in individual oocytes obtained from individual rats in various groups. GAL, Galactose; ARI, aldose reductase inhibitor. Asterisk, p < 0.05. 20 o GAL GALlARt ARt GROUP Fig. 2. Gonadotropin-stimulated ovulation rates in various groups of rats. GAL, Galactose; ARI, aldose reductase inhibitor. Asterisk, p < 0.01. Table I. Ovarian galactitol and myoinositol concentration in each experimental group (mean ± SEM) (N = 32) Galactitol (I-LmoVgm protein) Myoinositol (I-LIDOVgm protein) Normal Galactose only Galactose plus aldolase reductase inhibitor Aldolase reductase inhibitor 0.0 27.5 ± 2.0 42.1 ± 6.4* 27.1 ± 1.6 3.4 ± 0.4 37.9 ± 7.6t 0.0 51.2 ± 7.4* *P < 0.005. tp < 0.05. 40% galactose gained less weight (Table 3). Histologic review of the ovaries demonstrated comparable numbers of small, medium, and large follicular complexes in all groups. However, karyorrhexis was most significant in corpora luteal granulosa cells in the galactose-exposed animals. Intercellular granulosa cell organization was poorest in this group. Control animals and aldose reductase inhibitor-induced only exposed animals had 1840 Meyer et al. December 1992 Am J Obstet Gynecol Fig. 3. Hematoxylin and eosin-stained ovarian tissue from control group. (Original magnification x 100.) Table II. Estradiol and progesterone serum levels in each experimental group (mean ± SEM) (N = 32) Estradiol (pg/ml) Progesterone (ng/ml) Normal Galactose only Galactose plus aldolase reductase inhibitor Aldolase reductase inhibitor 14.4 ± 4.1 24.2 ± 6.9 11.3 ± 2.1 12.7 ± 5.0 12.8 ± 2.2 2.9 ± 1.0* 15.4 ± 2.5 17.5 ± 4.0t *p < 0.005 as compared with normal value. tp < 0.05 as compared with value for galactose plus aldolase reductase inhibitor. Table III. Animal (grams) and ovarian weights (milligrams) of rat in each experimental group (mean ± SEM (N = 32) Animal Ovary Normal Galactose only Galactose plus aldolase reductase inhibitor Aldolase reductase inhibitor 140.3 ± 4.3 3.5 ± 0.2 122.0 ± 3.2* 2.7 ± 0.2* 113.9 ± 2.4* 2.1 ± 0.2* 136.6 ± 2.7 3.8 ± 0.2 *p < 0.05. normal granulosa cellularity and organization. The addition of the aldose reductase inhibitor in the galactosefed animals lessened the majority of these histologic aberrations observed in luteal granulosa cells (Figs. 3 through 6). Comment Galactosemia caused by lack of galactose-I-phosphate uridyl transferase affects up to one in every 63,000 human newborn infants. The disease is associated with liver dysfunction, cataract formation, mental retardation, and death in the newborn if galactose restriction is not instituted. Yet, in spite of dietary galactose restriction, galactosemic women are frequently afflicted with gonadal dysfunction and premature menopause. Even women identified as carriers who demonstrate reduced transferase activity have been characterized as subfertile! Because gonadal dysfunc- Galactitol Volume 167 l\umber (; 1841 Fig. 4. Hematoxylin and eosin-stained ovarian tissue from galactose-fed group. Arrow, Area of severe karyorrhexis and cellular disorganization. (Original magnification x 100.) Fig. 5. Hematoxylin and eosin-stained ovarian tissue from galactose and aldose reductase inhibitorexposed animals. Arrow, Area of minimal karyorrhexis. (Original magnification x 100.) tion occurs only in women, the physiologic or anatomic insult in galactosemia is suggested to occur not in the hypothalamic-pituitary-gonadal axis but rather within the ovary itselC The enzyme aldose reductase converts sugars into their respective sugar alcohols or polyols. In the case of galactosemia, excess galactitol accumulates. Galactitol is metabolized poorly with poor cell membrane penetrance. Galactitol accumulates in viable cells, which may cause osmotic imbalance and cellular swelling, which in 1842 Meyer et al. December 1992 Am J Obstet Gynecol Fig. 6. Hematoxylin and eosin-stained ovarian tissue from aldose reductase inhibitor only-exposed group. (Original magnification x 100.) N~P' ) GALACTITOL GALATOSE UDP-G1C) GALACTOS:1 UDP-Gal GLUCOSE - 1 - 1 - PHOSATE~ PHOSPHATE UDPGlc TCA }athWay Fig. 7. Pathway of galactose metabolism. UDP, Uridine diphosphate; Glc, glucose; Gal, galactose; TCA, tricarboxycyclic acid. turn may alter cellular ion permeability and result in cell dysfunction. Galactitol tends to accumulate in high concentration in nerve tissue, the lens of the eye, and the heart of animals exposed to high galactose diets. However, in galactose-fed animals ovarian concentrations of galactitol exceed those levels found in the pancreas, liver, spleen, and kidney." Increased ovarian levels, in comparison with those of the kidney, are pertinent because altered vascular filtration function in the kidney is an observed affliction of galactosemic animals. 9 The treatment of galactosemic rats with potent inhibitors of aldose reductase decreases the accumulation of galactitol, and sorbinil has been shown to ameliorate the ill effects on various organ systems, such as the formation of cataracts in the lens, defective healing in scraped corneas, decreased conduction velocity in motor nerves, and basement membrane thickening in retinal capillaries'" In·12 As a consequence, we hypothesized that excess flux through aldose reductase or galactitol accumulation may result in ovarian dysfunction in galactosemic animals rather than excess galactose-I-phosphate or inadequate uridine diphosphate galactose production, as suggested by other investigators (Fig. 7).13. 14 However, direct measurements of these metabolites during aldose reductase inhibitor exposure would be necessary to definitively exclude these as the responsible toxic metabolites in galactosemic animals. Galactitol Volume 167 Number 6 1843 REFERENCES Specifically, excess galactose exposure in animals has been reported to result in decreased oocyte production in both natural and gonadotropin-stimulated cycles. 3 • 15 The current study has confirmed these quantitative findings along with extending the prior observations to demonstrate significant developmental delay in individual oocytes (as assessed by germinal vesicle breakdown). Increased intraovarian galactitol concentrations support the premise that this metabolite may be linked to the qualitative and quantitative defects in rat oocytes. Most supportive is the fact that when galactitol concentrations are prevented from rising within the ovary because of the action of the aldose reductase inhibitor, both postovulatory egg numbers and the degree of egg maturation approach normal values. These changes occur in spite of a decrease in serum progesterone levels. The observation of lowered progesterone levels in the galactose and galactose-aldose reductase inhibitor fed animals might be expected because of the noted histologic alteration in luteal granulosa cell organization. Similar to the studies of Stewart et al. 16 and others l7 involving galactosemic and diabetic animals, our study confirms that the lack of a decrease in tissue myoinositol levels, in spite of an increase in polyol levels, appears inconsequential to the degree of organ dysfunction. In addition, the diminished weight of both the animal and the ovary that occurs with galactose exposure was not a contributing factor in ovarian dysfunction. In conclusion, our evidence indicates that ovarian dysfunction in galactosemic rats is linked to the metabolism of galactose through aldose reductase. It should be noted that this conclusion is based on data accumulated in the galactose-exposed rodent. In humans with classic galactosemia the absence of ovarian galactose-lphosphate uridyl transferase activity may result in metabolic abnormalities distinct from those found in galactose-fed animals with normal transferase activity. I" We thank Dr. Joseph R. Williamson, Department of Pathology, Washington University School of Medicine, St. Louis, Missouri, for analysis of ovarian polyollevels; Dr. Arona Kumar, Department of Pathology, Suburban Hospital, Bethesda, Maryland; and Craig A. Ellery, Pfizer, Inc., for excellent technical assistance. 1. Kaufman FR, Kogut MD, Donnell GN, Goebelsmann U. March C, Koch R. Hypergonadotropic hypogonadism in female patients with galactosemia. N Engl J Med 1981; 304:994-8. 2. Cramer DW, Harlow BL, Barbieri RL, Ng WG. GalactoseI-phosphate uridyl transferase activity associated with age at menopause and reproductive history. Fertil Steri11989; 51:609-15. 3. Swartz WJ, Mattison DR. Galactose inhibition of ovulation in mice. Fertil Steril 1988;49:522-6. 4. Robison WG, Kador PF, Kinoshita JH. Retinal capillaries: basement membrane thickening by galactosemia prevented with aldose reductase inhibitor. Science 1983;221: 1177-8. 5. Pellicer A, Parmer TG, Stoane JM, Behrman HR. Desensitization to follicle-stimulating hormone in cumulus cells is coincident with hormone induction of oocyte maturation in the rat follicle. Mol Cell Endocrinol 1989;64: 179-88. 6. Eades DM, WilliamsonJR, Sherman WR. Rapid analysis of sorbitol, galactitol, mannitol, and myoinositol mixtures from biological sources. J Chromatogr 1989;490:1-8. 7. Kaufman FR, Donnell GN, Roe TF, Kogut MD. Gonadal function in patients with galactosemia. J Inherit Metab Dis 1986;9: 140-6. 8. Stewart MA, Kurien MM, Sherman WR, Cotlier EV. Inositol changes in nerve and lens of galactose-fed rats. J Neurochem 1968;15:941-6. 9. Pugliese G, Tilton RG, Speedy A, et at. Vascular filtration function in galactose-fed versus diabetic rats: the role of polyol pathway activity. Metabolism 1990;39:690-7. 10. DatiIes M, Fukui H, Kuwabara T, KinoshitaJH. Galactose cataract prevention with sorbinil, an aldose reductase inhibitor: a light microscopic study. Invest Ophthalmol Vis Sci 1982;22:174-9. 11. Fukushi S, Meroli LO, Tanaka M, Datiles M, KinoshitaJH. Reepithelialization of denuded corneas in diabetic rats. Exp Eye Res 1980;31:611-21. 12. Vue DK, Hanwell MA, Satchell PM, Turtle JR. The effect of aldose reductase inhibition on motor nerve conduction velocity in diabetic rats. Diabetes 1982;31:789-94. 13. Schwarz V. The value of galactose phosphate determinations in the treatment of galactosemia. Arch Dis Child 1960;35:428-32. 14. Xu YK, Ng WG, Kaufman FR, Lobo RA, Donnell GN. Galactose metabolism in human ovarian tissue. Pediatr Res 1989;25:151-5. 15. Chen YT, Mattison DR, Feigenbaum L, Fukui H, Schulman JD. Reduction in oocyte number following prenatal exposure to a diet high in galactose. Science 1981;214: 1145-7. 16. Stewart MA, Sherman WR, Kurien MM, Moonsammy GI, Wisgerhof M. Polyol accumulations in nervous tissue of rats with experimental diabetes and galactosemia. J Neurochem 1967;14:1057-61. 17. Loy A, Lurie KG, Ghosh A, Wilson JM, MacGregor CG, Matchinsky FM. Diabetes and the myoinositol paradox. Diabetes 1990;39: 1305-12. 18. Williams CA, Macdonald I. Metabolic defects of dietary galactose. World Rev Nutr Diet 1982;39:23-52.