toxicity testing using the isolated in vitro perfused ovary

Reproductwe Toxwology, Vol 7, pp 63-68, 1993 0890-6238/93 $6 00 + 00 Ptanted m the U S A All rights reserved Copyright © 1993 Pergamon Press Ltd

• Female Reproductive Toxicity

TOXICITY TESTING USING THE ISOLATED IN VITRO PERFUSED OVARY

JOHN F. JARRELL,* MARGARET L. SEVCIK,* DAVID C. VILLENEUVE,~

and PER O. JANSON:~ *Department of Obstetrics and Gynaecology, University of Calgary, Calgary, Alberta, Canada; tHealth Protection Branch, Government of Canada, Ottawa, Ontario, Canada; *Department of Obstetrics and

Gynaecology, University of Goteborg, Goteborg, Sweden

Abstract m This artide discusses the use of in vitro perfusion techniques as a tool for toxicity testing in the ovary and how the rat ovary has been adapted for this purpose. A brief review of the development of in vitro ovarian perfusion is provided, focusing on steroidogensis and physiology of ovulation. Adaptation of this model for use as a toxicologic model is discussed in the context of other isolated organ models, (that is, liver, heart, lung). Surgical procedures, perfusate and criteria for viability are outlined. Advantages of this technique are highlighted including ability to administer high doses of drugs directly to intact organ devoid of other influences. Applications of this model are discussed and data from studies of glutathione depleted ovaries perfused with hexachlorobenzene (HCB) are presented. Increased oxygen consumption after addition of HCB is suggestive of a disordered respiratory metabolism and is an example of future markers of ovarian injury using this innovative technique.

Key Words ovary, perfuslon, toxicology, hexachlorobenzene, m vitro

INTRODUCTION

Isolated organ techniques have traditionally been used to ascertain normal organ function. Recently however, there has been a concerted effort to Iden- tify the common aspects of perfus~on of individual organ systems. Mehendale has recently provided an excellent summary of the basics of isolated organ perfusion for heart, lung, liver, kidney, brain, intes- tine, and pancreas (1). Ovarian perfusion was not included in this review. It has been suggested that there are many aspects of each system that are similar and that at least in terms of organ biochemistry, isolated organ perfusion should be considered a uni- fied techmque.

There are many advantages to the technique when it is apphed to toxicity testing The actual disposition of a chemical or endogenous material can be determined exactly. For example, isolated perfusions determined the uptake of glutathlone by the kidney (2). The assessment of the organ occurs

Address correspondence to John F Jarrell, Department of Onstetncs and Gynaecology, Umverslty of Calgary, Calgary, Alberta, Canada T2N 2T9

without disruption of anatomy, which is particularly relevant to organs systems such as the ovary in which growth factors are present. This may permit a more relevant approximation of the in VlVO state. Under certain circumstances, human organs can be perfused. The process permits use of either very small concentrations of precious material as well as higher concentrations than would ordinarily be possible through In vivo dosing. Many physiologic variables can be controlled that would otherwise be impossible to control.

The major disadvantages include the surgical dexterity required to establish the preparation, the fact that the integrity of the system is limited over time, and the fact that the perfused organ may not represent the actual state in vivo.

OVARIAN PERFUSION

The ovary has been the subject of in vitro perfuslon for many years (3). First reports of successful perfusions concentrated on steroidogenesis (4,5). Recently the physiology of ovarian function has been evaluated with respect to ovulation in the rabbit and the rat (6,7).

63

64 Reproductive Toxicology

Table 1. Viabili ty testing o f the perfused ovary

Procedure related Perfuslon rate Flow pressure relationship Visual examination H~stology and ultrastructure

Intermedmry metabohsm Energy state (0 2 consumption, respiratory quoUent, ATP

status, glucose utilization, lactate pyruvate ratios) SubceUular marker enzymes Ovarmn and perfusate biochemical measurements

Functional output Folhcular development Folhcular fluid Ovulation rate Oocyte chromosomal studxes Fertilization Pregnancy and progeny

The ovary is particularly suitable for this procedure because it can be isolated either as a separate organ in larger species (8) or as an en bloc dissection of the aorta, ovary, uterine horn, and vena cava, as in the rat (9).

There are many ways in which the ovary can be assessed for vlablhty (Table 1). It should be noted that the same variables that are used to determine the viability of the organ are also used to measure toxic effects, which underscores the importance of standardized procedures for control and treatment groups. In larger animals in which perfusion can be under taken via the ovarian artery, both ovaries can be used simultaneously with one serving as a control. This is not possible with smaller animals in which the ana tomy of the arterial supply to the ovary prohibits perfuslon of the left ovary. Under these circumstances, it ts ideal to undertake simultaneous perfuslons of two antmals, control and treated

From a practical perspect ive one of the most common findings in a poor sample is that there is a low flow rate. In the adult rat ovary the flow rate is generally 0 8 to 1.2 mL/mln. With increasing injury, flow and the flow:pressure ratio decrease. Ovaries that are injured also become wslbly edematous. Hls- tologic and ultrastructural studies are possible but

Volume 7, Supplement 1, 1993

it ~s continuously stressed in the perfuslon literature that normal appearing tissue can be dysfunctional

Evaluating the energy state of the organ is readily accomplished by the measurement of ATP content of the perfused organ. ATP IS difficult to measure in the perfusate and consequent ly measures over the time of the perfus~on require assessment of ar ter ia l -venous differences of pO2, pCO2, pH, and oxygen consumption. This can be accomplished with access to a blood gas analyzer. Some cellular biochemical markers appear to increase normally during perfuslon, such as lactate dehydrogenase, lactate, and pyruvate As these markers can be found under " n o r m a l " circumstances, careful com- parison to treatment groups is recommended.

The functional output Is the most integrated measure of health and can relate to follicle forma- tion, oocyte function, and genetic integrity. Some perfuslons permit assessment of oocytes from a treated ovary to be evaluated in a host animal to determine if a toxic effect is conveyed by ovarian (and oocyte) exposure separately from the whole organism.

There is one unique aspect of ovarian perfuslon with respect to toxicity testing that relates to the fact that the individual ovary (Class I), the genetic products of induced ovulation (Class II), and the subsequent pregnancy and progeny (Class III) can be studied (Table 2). This is particularly relevant in light of the observations of Generoso and Katoh that the perlovulatory period is one m which there is particular susceptibility of the female mouse in terms of subsequent anomalies (10,11). Perfuslon during the perlovulatory period has recently been reviewed (6).

Species There are two major systems In use at present,

which involve the rodent and the rabbit (6,7). There are advantages and disadvantages to each system, as represented in the hterature, but both systems have demonstra ted utility m the toxicology field Perhaps one of the most sahent ~ssues is that at-

Table 2 Proposed classification of ovarian perfuslon

Class Ovarian status Variables Applications

I Normal PMSG Tissue and Biochemical markers, treated perfusate drug metabolism

II Timed ovulation Ovulation rate Oocyte chromosomes Follicular fluid Folhcular fired

kinetics III Timed ovulation Pregnancy rate Clomlphene toxicity

Progeny outcome

Toxicity testing m perfused ovary • J F JARRELL ET AL 65

tempts should be made wherever possible to take advantage of the perfusion of human ovaries.

Apparatus The system designed by Janson with two modi-

fications has been used exclusively for our experi- ments. It was initially designed for rabbit perfusion (12) but later adapted for use in the rat (13). The apparatus consists of a glass gas humidifier, oxygenator-reservoir, perfusion chamber, and bubble trap, all of which are water jacketed and maintained at 37 °C by an immersion circulator A roller pump is set at 100 mL/min in order to maintain a perfusion pressure of 50 to 70 mm Hg, which is measured by a simple mercury manometer. Approximately 99% of the perfusate bypasses the ovary and recirculates from the pump to the oxygenator-reservoir. One percent of the flow reaches the bubble trap, which prevent air emboli from entering the vascular spaces. Increases in perfusion rates can be effected by a valve in the shunt line.

The apparatus has been modified to include a calibrated flow meter, which is located in the line entering the bubble trap for constant flow rate measurements. Syringe ports directly adjacent to the ovary have also been added to the system to allow arterial and venous samples to be withdrawn. These ports are used for all collections and are particularly relevant for perfusate gas analysis, which is accom- phshed with an ILS 1301 Blood Gas Analyzer.

Perfusion medium The medium used is 80 mL of M199 with Earle's

salts; NaHCO3 (0.026 M) is added to buffer to neutral pH. The perfusate also contains 0.4% BSA, gentamicin sulfate (50 /zg/mL), heparin sulfate (0.2 IU/mL), and insulin (0.02 IU/mL). Gentamicin has not been shown to interfere with steroidogenesis and insulin is required for steroidogenests.

The perfusate is critical to the dosing of the organ. In most circumstances, it is ideal to perfuse for at least two hours to permit equilibrium, and after two hours the chemical is added to the reservoir. The chemical is placed in solution with the greatest volume available and immediately begins to enter the shunt and the organ. Sampling of the perfusate is usually done each hour and an equal volume of perfusate is replaced with each sampling.

Surgwal procedure The s u r g i c a l removal of an en bloc dissection

of the ovary is performed with sodium pentothal (50 mg/kg body weight). Heparin sulfate is injected intraperitonealy prior to surgery to prevent clotting

(200 IU). With the animal lying on its back, the limbs are fixed in extension on a surgical board. The abdomen is cleansed with alcohol and a midline incision is made from pubis to xiphisternum. The caudal mesentenc artery is ligated. The distal colon is ligated and resected. The intestines are kept warm with saline soaked gauze. Iliolumbar arteries and veins are hgated and severed. The left renal and suprarenal vessels are hgated and severed. The right kidney is stripped of its capsule, the renal vessels are hgated and the entire kidney is removed. The right uterine horn is ligated and severed from the rest of the uterus. The aorta and vena cava are dissected free of the posterior wall and ties are placed cranial to the junction of the ovarian artery and vein. Both the aorta and vena cava are then cannulated with anglocaths superior to the inferior bifurcation of the aorta and vena cava. The cannulas are secured by ligatures and the use of winged catheters. The dorsal lumbar arteries are then ligated and severed. The specimen is dissected free of the posterior wall, with all its ligatures intact, and removed.

To ensure vascular integrity of the specimen, heparlnized saline, warmed to 37 °C is hand perfused with a 1 cc syringe into the ovary. Perfuslon of approximately 4 mL of saline ensures that most of the blood has been cleared from the specimen. When surgery is successful, the organ will blanch and there will be no leakage from any of the ties. The specimen is quickly attached to the perfusion apparatus and the perfusion begun.

APPLICATIONS

In addition to a growing list of publications related to the physiology of ovarian function there is a growing number of publications relevant to reproductive toxicology. Calcium and magnesium concentrations in the perfusate have been shown to influence the time to ovulation. Exclusion of calcium and magnesium result in rapid follicular development and ovulation of immature oocytes. Under these conditions, hCG could effect oocyte maturation (14). Similar findings have been identified for certain inhlbitors. Histamine and histamine blockade have been studied in isolated rabbit perfuslons (15). Histamine (100 ng/mL) can induce ovulation in this model. Ovulation is blocked by the coincident administration of clmetldlne (10 p,g/mL) but not chlorphenirimine (66.7/zg/mL). Of note, the oocytes released in response to histamine are immature and this response can be corrected by hCG administration (15). PGF2~ can induce ovulation when added to a perfuslon system at 1, 10, and I00 ng/mL. The

66 Reproductive Toxicology Volume 7, Supplement 1, 1993

effectiveness is greater at higher doses but 1s less than that observed with hCG administration alone. The rate of ovulation could not be inhibited with antihistamines (16) but the administration of mdo- methacln does inhibit hCG stimulated ovulation without changes in steroidogenesis (17) as well as inhibiting FSH induced ovulation in the rat ovary (18)

Perfusions of ovaries with puromycin and cyclohexamlde have shown that inhibition of protein synthesis is associated with inhibited ovulation and that de novo protein synthesis is important late in the ovulatory process (19).

Reduction m the availability of oxygen free ra&- cals has been undertaken with the perfusion of su- peroxide &smutase alone and with catalase, which was assocmted with a reduction m the number of follicles ovulating (20). The relationship to free radi- cals has been indirectly studied by the measurement of glutathione in the perfusate Although it is not known If the ovary takes up glutathione as is the case for the kidney, it is definitely released from the ovary and the concentrations are slightly less than those reported for the liver (21) Concentrations of glutathlone in the perfusate differ during the estrous cycle and are highest during estrus (22). The levels are further increased by administration of PMSG into the medium of ovaries in estrus but not m metestrus (23). Glutathione release can effectively be reduced by repeated injections of buthlonine sul- foxlmme, an inhibitor of the 3J-glutamyl synthetase activity, the rate limiting enzyme of glutathlone synthesis.

One of the more intriguing studies has been the effects of clomiphene citrate on ovarian function It Is known that clomlphene reduces the number of implantations and the effect occurs only with administration of the chemical before ovulation and not after ovulatory. A direct effect of clomiphene has been shown to occur m the Isolated rabbit perfusion with respect to ovum maturation and oocyte degeneration, which occurred in the absence of hCG. Ad- diUon of estradiol into the perfusate reversed ovum degeneration (24) A similar pattern was identified in early embryonic development. Clomlphene citrate was found to exert an antlestrogenic effect on the intrafollicular oocyte that altered postfertilization development and was reversible with estradlol (25) This system has been expended to the Class III type of perfusion by the transfer of embryos fertilized after collection as part of the process of in vitro perfuslon. Under these circumstances, the postfer- tdlzation alteration m embryonic development trans- lated into a reduced number of progeny m the host

animals receiving the embryos This would appear to be mediated at least in part by an antlestrogenlc process because the coadministration of estrogen in the original perfusate improved progeny outcome subsequently (26,27)

Adrenoreceptor function has been evaluated in the perfused rabbit ovary. Stimulation with terbuta- line or noradrenahne in the presence of a-receptor blockade with phenoxybenzamme mcreased the number of gonadotropln-lnduced ovulations whde perfusion with phentolamlne and phenoxybenza- mine alone reduced ovulation (28).

We have further used the in vitro perfusion system to evaluate the effects of administration of hexachlorobenzene (HCB) on the intermediary metabohsm of the ovary. Adult cycling rats were treated with either saline or BSO (2 nmoles/kg) every six hours on day of proestrus and ovaries prepared for perfuslon on the following day of estrus. After a two hour period of equilibration the animals were perfused with either HCB (1 /xg/mL) dissolved in dlmethyl sulphoxlde (DMSO) or DMSO alone The

PERCENT CHANGE FROM BASELINE OXYGEN

CONSUMPTION AFTER TREATMENT WITH

DMSO OR DMSO + HCB O SAL- DMSO • SAL- HCB

I n B S O - DMSO 90 F - • BSO- HCB

o= 3o

. . .

lad

-6O I I f I

0 1 2 :3 4

TIME (HOURS)

Fig 1. Twenty-four female cycling rats were treated on day of proestrus with saline or BSO and their ovaries were perfused the following day when in estrus. The lnmal 2 hours of perfuslon were an eqmllbnum period followed by the addition of HCB (1/xg//zL DMSO/mL of perfusate) or DMSO alone (l/xL/mL ofperfusate) Oxygen consump- tmn was measured and converted to a percent change from the basehne levels for 4 addmonal hours

Toxicity testing in perfused ovary • J F JARRELL ET AL 67

CONCENTRATION OF LACTATE DEHYDROGENASE

AFTER TREATMENT WITH DMSO OR DMSO + HCB,

CORRECTED TO T=O

a relatively new technology but has the advantage of being based upon excellent foundations in the physiology of ovarian function, and offers the possi- bility of identifying new markers of ovarian injury.

70O

o ,F~&,L- DMSO soo • SAL-HCB

T , . . , n B S O - D M S O I 500 • B S O - H C B ~ )

/ 1

t~ 300

O C V , , , , , , o 1 2 3 4 5 6

TIME (HOURS) Ftg. 2. The concentraUon of lactate dehydrogenase (Umts/ L) m the perfusate of in wt ro perfused ovaries after perfusion with D M S O or D M S O + HCB is presented. Twenty- four female rats were pre t reated with saline or BSO on the day of proest rus (4 doses of 2nmol/kg BW). The following day, when m estrus, the ovaries were r emoved and perfused for 6 hours. Concent ra t ions were correc ted to the t = 0 level to account for blood m the specimen.

administration of BSO significantly reduced ovarian glutathlone levels to 56% of control levels demon- strating the potency of this dosing regime (data not shown). Oxygen consumption of the ovaries was measured throughout the perfusion, as well as lactate dehydrogenase, creatme kinase, and the lactate:pyruvate ratio in the perfusate. The animals pretreated with saline (glutathione content intact) had a reduction in oxygen consumption after perfusion with HCB (data shown in Figure I). Animals pretreated with BSO (glutathione content depleted) demonstrated a marked increase m oxygen consumption one hour following addition of HCB. In both groups there was a significant increase in lactate dehydrogenase release into the perfusate (Figure 2). There were no changes in arterial-venous pH, pCO2, perfusate creatine kinase, or lactate:pyruvate ratios.

SUMMARY

Assessment of the ovary as a target of toxic injury remains a priority for reproductive toxicolo- gists and is emerging as a priority for regulators. Use of in vitro perfusion as a tool for such assessment is

REFERENCES

1 Mehendale HM Apphcation of isolated organ techniques m toxicology In Hayes AW, ed Principles and Methods of Toxicology. 2nd ed New York Raven Press, 1989 699-740

2. Fonteles MC, Pdhon D J, Jeske AH, Llebach FH. ExtracUon of glutath]one by the isolated perfused rabbit kidney J Surg Res 1976,21169 169-74

3. Carrel A, Lmdberg CA Culture of whole organs Sc]ence 1935,81 621-3

4 Werthessen NT A technique of organ culture for protracted metabohsm studies Endocnnology 1949,44 109

5 Romanoff EB, Pmcus G Studies of the isolated perfused ovary methods and examples of application Endocrinology 1962,71 752-5

6 Wallach EE, Atlas SJ Studies of the penovulatory interval m the m vitro perfused ovary Adv Exp Med B]ol 1987,219 179-201

7 Ahren K, Janson PO, Selstam G Perfuslon-ofovanes m vitro and m vwo In Proceedings of the Karohnska Symposia on Research Methods m Reproductwe Endocrinology 1971 285-309

8 Wallach EE, Wright KH, Hamada Y Investigation of mam- mahan ovulation w~th an m vitro perfused rabbit ovary preparation Am J Obstet Gynecol 1978,132 728-38

9 Brannstrom M, Johansson BM, Sogn J, Janson PO Charac- terization of an m vitro perfused rat ovary model ovulation rate, oocyte maturation, sterotdogenesls and influence of PMSG priming Acta Physlol Scand 1987,130'107-11

10 Generoso WM, Rutledge JC, Cmn KT, Hughes LA, Braden BD Exposure to female mice to ethylene oxide within hours after matmg leads to fetal malformation and death Mutat Res 1987,176 269-74

11 Katoh M, Cachelro NLA, Cornett CV, Cain KT, Rutledge JC, Generoso WM Fetal anomahes produced subsequent to treatment of zygotes with ethylene oxide or ethyl methanesul- fonate are not hkely due to the usual genetic causes Mutat Res 1989,210 337-44

12 Janson PO, LeMalre WJ, Kallfelt B, et al The study of ovulation m the isolated perfused rabbtt ovary I Methodol- ogy and pattern of steroidogenests Blol Reprod 1982, 26 456-65

13 Sogn JH, Curry TE Jr, Brannstrom M, et al Inhlbmon of folhcle-sttmulatmg hormone-reduced ovulation by mdometh- acre m the perfused rat ovary Prostaglandms and FSH- reduced ovulations m vitro Blol Reprod 1987;36 536-42

14 Kobayashl Y, Kitm H, Santmh R, Wright KH, Wallach EE Influence of calcmm and magnesium depnvaUon on ovulation and ovum maturation m the perfused rabbit ovary Btol Re- prod 1984,31 287-95

15 Kobayashl Y, Wright KH, Santulh R, Kltai H, Wallach EE Effect of histamine and histamine blockers on the ovulatory process m the m vitro perfused rabbit ovary Blol Reprod 1983,28 385-92

16 Kital H, Kobayashl Y, Santulh R, Wright KH, Wallach EE The relationship between prostaglandms and histamine in the ovulatory process as determined with the m vitro perfused rabbit ovary Fertd Stenl 1985,43 646-51

17 Katz E, Dharmarajan AM, Sueoka K, Ghodganonkar RB, Dubm NH, Wallach EE Effects of systemic administration of mdomethacm on ovulation, lutemlzaUon, and steroidogen- esls in the rabbit ovary Am J Obstet Gynecol 1989, 161 1361-5

68 Reproductive Toxicology Volume 7, Supplement 1, 1993

18 LeMmre WJ, Koos RD, Clark MR, et al Studies on the role of prostaglandms m ovulation using the isolated perfused ovary as a model Adv Prostaglandln Thromboxane Leuko- trlene Res 1985,15 589-91

19 Brannstrom M, Boberg BM, Tornell J, Janson PO, Ahren K Effects oflnh~bators ofprotem synthesas on the ovulatory process of the perfused rat ovary J Reprod Fertll 1989,85 451-9

20 Mayazakl T, Sueoka K, Dharmarajan AM, Atlas S J, Bulkley GB, Wallach EE Effect of inhibition of oxygen free radical on ovulation and progesterone productmn by the an vatro perfused rabbat ovary [abstract] J Reprod Fertll 1991, 91 207-12

21 Sles H Reduced and oxldazed glutathmne efflux from hver In Sakamoto Y, ed Glutathlone storage, transport and turn- over m mammals Tokyo Japan Science Socaety, 1983 63 -88

22 Clague N, Sevclk M, Stuart G, Brannstrom M, Janson PO, Jarrell J The effect of estrous cycle and buthlonlne sulfoxl- mine on glutathlone release from the m vitro perfused rat ovary Reprod Toxacol 1992,6 533-9

23 Jarrell J, Sevclk M, Stuart G Regulation of total ovarian glutathlone content m the rat Reprod Toxlcol 1992,6 133-5

24 Yoshlmura Y, Kltal H, Santulh R, Wright K, Wallach EE Direct ovarian effect of clomlphene citrate m the rabbit Fertll Sterll 1985,42 471-6

25 Yoshimura Y, Hosol Y, AtLas S J, Adachl T, Santulh R, Wal- lach EE Estradlol reverses the hmltmg effects of clomlphene citrate on early embryomc development in the m v~tro perfused rabbit ovary Fertll Sterll 1987,48 1030-5

26 Yoshlmura Y, Hosol Y, Atlas SJ, Dharmarajan AM, Adachl T, Wallach EE Effect of the exposure of mtrafolhcular oocytes to clomlphene citrate on pregnancy outcome m the rabbit Fertil Sterll 1988,50 153-8

27 Yoshtmura Y, Hosol Y, Atlas S J, Wallach EE Effect of clomlphene c~trate on m vitro ovulated ova Fertll Sterll 1986,45 800-4

28 Schmldt G, Owman C, Sjobarg NO, Walles B Influence of adrenoreceptor agomsts and antagonists on ovulation m the rabbat ovary perfused m vitro J Auton Pharmacol 1985,5 241-50

toxicity testing using the isolated in vitro perfused ovary

Documents