Myometrial arginase activity pregnancy in the guinea pig

increases with advancing

Carl E Weine~; MD, b Richard G. Knowles, Phi)," Lewis D. Stegink, PhD,C Jolm Dawson, a and Salvador Moncada, MD, Phl)"

Beckenham, United Kingdom, and Iowa City, Iowa

OBJECTIVES: Arginase has been suggested to play an important role in cellular growth and development, particularly important to the fetus, by supplying L-ornithine for the synthesis of polyamines. The purpose of this investigation was to determine whether pregnancy alters myometrial arginase activity and whether estradiol was responsible for the change. STUDY DESIGN: Myometrium and kidney were obtained from nonpregnant and pregnant guinea pigs of known gestational age. Arginase activity was measured under physiologic conditions by the conversion carbon 14-labeled guanidino-L-arginine to carbon 14-labeled urea. The concentrations of the enzyme's substrate, L-arginine, and its principal metabolite, L-ornithine, were measured in myometrium from near-term pregnant animals by use of an amino acid analyzer. Finally, a group of random cycle guinea pigs received 500 Ixg/kg estradiol for 5 days before the myometrium was removed. RfiSUI.TS; Myometrial arginase activity in pregnant animals was more than double that of myometrium from nonpregnant animals by the time the first measurement was made at 0.14 gestation. It continued to rise, peaking at values >25-fold higher than the nonpregnant activity by 0.90 gestation. Arginase activity in the myometrium underlying the placental implantation site was >25 fold higher (p < 0.05) than myometrium from nonpregnant animals when first studied at 0.63 gestation and 10-fold higher than the contrNaterat fundal myometrium at the same time of gestation. Myometrial arginase activity in the sterile horn of six pregnant animals was half that of the horn containing one or more pups, but still five times higher than that of nonpregnant animals. Renal arginase activity also rose with advancing pregnancy, but the magnitude of the increase (up to 2-fold) was much smaller than that observed in either the fundal or placental implantation site myometrium. Estradiol had no significant effect on myometrial arginase activity. CONCLUSIONS" These studies demonstrate that pregnancy increases myometrial arginase activity and that the presence of placenta or fetus is necessary for the maximal effect. (AM J OBSTET GYNECOL 1996;174:779-82.)

Key words: Estradiol, myometrium, kidney, polyamines, pregnancy, guinea pig, L-arginine, L-ornithine

Arginase catalyzes the hydrolysis of L-arginine to L-or- nithine and urea) It is found predominantly in liver, where it functions as a part of the urea cycle, and in kidney. During pregnancy there is a shift in nitrogen

disposal away from the hepatic urea cycle to the ammonia cycle of the kidney. 2' ~ Although this shift could be related to alterations in the activities of urea cycle enzymes, it is

From the Wellcome Research Laboratories, ~ and The Perinatal Research Laboratory, Department of Obstetrics and Gynecology, b and the Division of Gastroenterology, Department of Pediatrics, C University of Iowa College of Medicine. Received for pub¢ication March 21, 1995; revised June 26, 1995; ac- cepted July 7, 1995. supported in part by United States Public Health Service grants No. HD22294 (C.t~W.), HL49041 (C.P.W.), and HL51735 (C.P..'VE). C.P. W. was a Wellcome Visiting Scientist while portions of this work were performed. Reprint requests: Perinatal Research Laboratory, Department of Obstet- rics and and Gynecology, University of Iowa College of Medicine, Iowa City, IA 52242. Copyright © 1996 t O, Mosby-Year Book, Inc. 0002-9378/96 $5.00+ 0 6/1/67800

unlikely to be the result of changes in arginase activity because pregnancy is not associated with a decrease in

hepatic arginase activity. *° Polyamines (e.g., spermidine, spermine, and putrescine), which are synthesized from arginase-derived L-ornithine, may have a role in cellular growth and development.: For example, arginase activity is increased in the lactating breast and spermidine is required for the synthesis of milk protein. 8' 9 The effect of

gestation on arginase activity in the uterus has not been studied, although arginase has been reported to be present in the myometrium of pregnant rats. ~° The obser-

vations that fetal mass increases rapidly during the second half of pregnancy and that a nonspecific inhibitor of arginase causes fetal growth deficiency in the rat may

indicate a progressive and essential demand for polyamines derived from arginine/ornithine. 1°

Limited information indicates that synthesis of argi- nase is actively regulated. For example, estradiol (2 lag) increases rat myometrial arginase activity by about 30% ]1 More recently, Corraliza et al. I2 demonstrated that both

779

780 Weiner et al. February 1996 Am J Obstet Gynecol

interleukin-4 and interleukin-10 induce arginase-specific messenger ribonucleic acid transcription. Both interleu- kin-10 and interleukin-4 are putative immunoprotectors of the conceptus. They are synthesized by decidua and amnion, 13 and their concentrations are increased in both the myometrium and the placental bed. 14' 15 The deple- tion of Th-2 cells, which produce these cytokines, is asso- ciated with fetal growth deficiency. 16 The purpose of the

current investigation was to determine whether preg- nancy alters myometrial arginase activity in the guinea pig

and, if so, to determine whether estradiol is responsible for the increase.

Methods

Nonpregnant and timed-pregnant mixed breed and

Harfley guinea pigs of similar chronologic age were pur- chased from commercial breeders. Pregnancy could be confirmed visually as early as 9 days after copulation (0.14

of gestation, 2 days after implantation). The guinea pig was selected for several reasons. The sex hormonal pro- files of both the estrous cycle and pregnancy of the guinea pig are quite similar to those of m a n ) w' 18 Also,

among nonprimates the guinea pig has the placenta most comparable to that of the h u m a n ) 9 Finally, the rapid fetal

growth rate and the high fetal/maternal weight ratio demonstrates the need for an efficient supply of amino acids to the growing fetus) 9

Arginase activity. The animals were terminally anesthe- tized (pentobarbital 120 mg/kg intraperitoneally), and

both the fundal myometrium opposite the placenta and that immediately under the placental implantation site

were excised. Samples were immediately frozen in liquid nitrogen and stored at -70 ° C until studied. To ensure that

an adequate weight of tissue confined to the placental im- plantation site was obtained, placental site specimens were not taken until 0.60 gestation. Biopsy specimens were taken from sterile uterine horns, when present, for com- parison to the pup-containing horn. Because the kidney was known to be rich in arginase, 2° renal cortical tissue was also excised from nonpregnant and pregnant animals for

measurement of renal arginase activity. The frozen tissue was mechanically homogenized

(Ystral) at 0 ° C in three volumes of buffer containing 320

mmol /L sucrose, 50 mmol /L Tris, 1 mmol /L ethylenedi- aminetetraacetic acid, 1 mmol /L DL-dithiothreitol, 100

lag/ml phenylmethylsulphonyl fluoride, 10 pg /ml leu- peptin, 10 pg /ml soybean trypsin inhibitor, and 2 pg /ml aprotinin brought to a pH 7.0 at 20 ° C with hydrochloric acid. The crude homogenate was centrifuged at 0 ° C at 15,000g for 20 minutes, and the pellet was discarded. A preliminary study demonstrated the stability of arginase in both the stored tissue and homogenate for up to 1 year when frozen at -70 ° C.

The measurement ofarginase activitywas based the con- version of carbon 14-labeled guanidino-L-arginine

(Amersham, Arlington Heights, Ill.; 53 ~lCi/lamol ) to 14C-

labeled urea at 37 ° C because of a modification of the method of Ruegg and Russell21 Briefly, 20 111 of homoge-

nate was added to 100 pl of a HEPES (N-2-hydroxyeth- ylpiperazine-N-2-ethanesulfonic acid) buffer (pH 7.5) containing 100 tamol/L L-arginine and 12.5 I~mol/L mag- nesium chloride. Preliminary study demonstrated linear enzyme kinetics over a 1-hour incubation period. There-

fore the reaction mixture was incubated for 1 hour at 37 ° C before the reaction was stopped by 0.5 ml of 250 mmol /L acetic acid containing excess cold urea (7 mol/L) and L- arginine (10 mmol/L). Any labeled L-arginine (or other metabolite of L-arginine other than urea) remaining was bound to the H+ form of Dowex 50 (Sigma, St. Louis). After a 10-minute centrifugation (5000g), the supernatant was aspirated and counted in a liquid scintillator. Results are expressed in nanomoles per minute of urea formed per

gram of tissue. Intraassay and interassay variation were each <8%.

Effect of pregnancy on L-arginine and L-ornithine. Myo- metrium obtained from near-term pregnant (across from

the placenta) ( n= 5 ) and nonpregnant ( n= 5 ) guinea pigs was homogenized with a Polytron homogenizer in 5 ml /gm tissue weight of a 5% solution (wt/vol) of sul- fosalicylic acid (Sigma). The homogenized samples were placed in an ice bath while the other samples were pre- pared. The homogenate was centrifuged for 20 minutes at 10,000g to remove precipitated protein. The superna- tam solution was ultrafiltered with a Centrieon-10 cen- trifugal filter assembly (Amicon, Beverly, Mass.); amino-

ethylcysteine (Sigma) was added as the internal standard. The filtrate was either analyzed immediately on a Beck- man 6300 amino acid analyzer (Beckman Instruments,

Palo Alto, Calif.) coupled with ChromPerfect (Littau, Switzerland) software or stored at -70 ° C until analyzed. Amino acid content was expressed in micromoles per gram of tissue.

Es~adiol supplementation. Nonpregnant random- cycle animals were given estradiol (500 tag/kg/day intra- peritoneally) for 5 days (n = 4). This dose of estradiol was selected to provided a circulating concentration of estra- diol similar to that found at term in the guinea pig. 22

Analyses. The results were presented as their mean +

1 SEM. The first day of pregnancy was defined as the day of copulation. Analyses included the Student t test, anal- ysis of variance with a Tukey's test for multiple compari- sons to identify differences among groups, Kruskal-Wallis one-way analysis of variance, and linear correlation. A p value <0.05 was considered to indicate either a signifi- cant correlation or difference among means.

Results

Effect of pregnancy on uterine arginase activity Myometrium. Myometrial arginase activity was more

than double that of myometrium from the nonpregnant

Vo]ume 174, Number 2 Weiner et al. 781 Am ] Obstet Gynecol

Arginase Activity

nmol /min/g

35 ~-

30

25

20

15

10

5

0

- O - Fundus

~ 7 I t I i I NP .20 .40 .60 .80 1.00

Gestation

Fig. 1. Myometrial arginase activity (nanomoles of urea per minute per gram) rises during pregnancy in both myometrium opposite placental implantation site (fundus, open circle) and underlying placenta (implantation, closed squares), x axis, Frac- tion of completed gestation. Each time point is represented by mean _+ 1 SEM of at least four animals.

animal by the first measu remen t at 0.14 gestation (Fig. 1).

Activity then p la teaued unti l 0.63 gestation, after which

there was a dramatic increase in activity peaking by 0.90

gestation at >25 times the activity in myomet r ium f rom

n o n p r e g n a n t animals. Al though there appeared to be a

small peak in activity be tween 0.25 and 0.4 gestation, the

activity level was no t significantly di f ferent f rom 0.45 to

0.6 gestation (p = 0.13). Arginase activity at the placental implanta t ion site was

>25-fold h igher than in n o n p r e g n a n t myomet r ium when

first s tudied at 0.63 gestation and 10-fold h igher than its

contralateral fundal myometr ium. Arginase activity at the

placental implanta t ion site activity r ema ined high unti l

the end of pregnancy.

A sterile u ter ine h o r n was discovered in six p regnan t

animals near t e rm (>55 days). Myometrial arginase activ-

ity was half that of the h o r n conta in ing one or more pups

(6.2 _+ 2 n m o l / m i n per gram vs 12.6 + 5 n m o l / m i n per

gram, p < 0.04) but five times h igher than in myomet r ium

f rom n o n p r e g n a n t animals (n = 10) (6.2 + 2 n m o l / m i n

per gram vs 1.11 _+ 0.5 n m o l / m i n per gram, p < 0.003).

Kidney. There was a positive corre la t ion between gesta-

t ional age and renal arginase activity ( r=0 .60 ,

p = 0.0004) (Fig. 2). However, the magni tude of the in-

crease was m u c h smaller than that observed in the myo-

metr ium. By 0.70 gestation renal arginase was only twice

that observed in the n o n p r e g n a n t animal ( n = 6 )

(p < 0.02). Effect o f pregnancy on uter ine L-arg in ine and L-orni-

thine. Pregnancy had no effect on the concent ra t ion of

L-arginine (0.61 + 0.1 vs 0.45 +-- 0.1 l~mol /gm in controls)

but significantly increased the concent ra t ion of its ar-

ginase metabol i te o rn i th ine (0 .66+ 0.2 vs 0.29 + 0.02

u m o l / g m , p = 0.02, in tissue f rom pregnan t and nonpreg-

Arginase Activity

nrnol Urea/min/g

12

10

8

6

4

2

0

n = 31 T h, Z/I

T

- - - - - . J _ z \ /

I I I I , L _ _ L . L - - - - . NP .20 .40 .60 .80 100

Gestation

Fig. 2. Renal arginase activity (nanomoles of urea formed per minute per gram) rises during pregnancy, x axis, Fraction of completed gestation, n= 31 animals. Each time point is repre- sented by the mean -+ 1 SEM of at least four animals.

nant animals, respectively). As a result, the ratio of sub-

strate to metabol i te was significantly reduced dur ing

pregnancy (0.93 _+ 0.08 vs 1.57 + 0.2 in controls, p = 0.01).

Effect o f estradiol on myometr ia l arginase activity. A

5 day course of estradiol had no significant effect on

myometr ia l arginase activity, the n o n p r e g n a n t values

being 1.12 _+ 0.50 n m o l / m i n per gram, n = 10, 1.40 + 0.25

n m o l / m i n per gram, n = 6, for the estradiol t reated (one-

tailed test, p = 0.56).

C o m m e n t

Pregnancy dramatically increases myometr ia l arginase

activity. The greatest increase occurs in the section of

myomet r ium directly underlying the placenta site. The

significant increase (more than twofold) in myometr ia l

L-ornithine concent ra t ion is consistent with the increased

arginase activity. The magni tude of the increase in myo-

metr ial arginase activity is remarkable consider ing that it

rises f rom a fraction of that found in the kidney of the

n o n p r e g n a n t animal to reach a level manyfold higher.

The re is no obvious explanat ion why our findings differ

f rom those of Mendez et al. 1° who repor ted that arginase

activity in the rat is lower at the placental implantat ion

site than in nondecidual ized myometr ium. However, the

rat has a placenta different f rom that of e i ther the human

or gu inea pig.

Arginase activity in myomet r ium removed f rom the

sterile horn of p regnan t animals is also increased com-

pared with myomet r ium f rom n o n p r e g n a n t animals.

Thus e i ther the stimulus or the cell responsible for the

increase in arginase activity circulates but has its greatest

u ter ine concent ra t ion in the region of the placenta. A sex

h o r m o n e such as estradiol would seem a possible candi-

date because it is known to induce a variety of enzyme

activities, inc luding ano the r L-arginine-util izing enzyme,

nitric oxide synthase. 23 However, we found that a 5-day

course of estradiol at a dose selected to mimic the circu-

lating concent ra t ion at the end of the guinea pig preg-

782 Weiner et al. February 1996 Am J Obstet Gynecol

nancy 22 did not significantly increase myometr ial arginase

activity. This result is consistent with those of Freeman

and Williams, ~I who repor t ed a small (30%) increase in

rat myometr ia l arginase activity after estradiol t rea tment

(2 p g / d a y for 7 days). It is clear that the effect of estradiol

(25% to 30%) in e i ther the rat or the guinea pig is too

small to account for the effect of pregnancy on myome-

trial arginase activity (25-fold). Al though the effect o f

pregnancy could be media ted by ano ther sex ho rmone ,

Lamers and Mooren 5 have repor t ed that ne i ther testos-

t e rone no r p roges te rone alter hepat ic arginase activity in

castrated juveni le rats.

Ano the r possible stimulus for the increase in myome-

trial arginase activity is induct ion by Th-2 cytokines. ~2

These growth factors have a short hal f life in vivo and must

exer t their effect proximally to the site of product ion. In-

terleukin-10 is p roduced by the amnion and decidua, 14' ~5

and the placenta is part iculary rich in in ter leukin-4-pro-

ducing cells. 13 Thus it is reasonable to speculate that the

high level of myometr ia l arginase activity reflects interleu-

kin-10 and interleukin-4 synthesis by the decidua and pla-

centa, respectively. This explanat ion would also be consis-

tent with the h igher levels ofarginase activity at the placen-

tal implanta t ion site. The increase in arginase in the sterile

h o r n dur ing pregnancy could ref lect sequestrat ion of

Th-2 cells released into the materna l circulation. Because

e i ther the inhibi t ion of arginase I° or the deple t ion of cyto-

k ine-producing cells is associated with fetal growth deft-

ciency, 16 arginase may play a role in supplying orn i th ine or

polyamines to the fetus. This hypothesis would be consis-

tent with the observat ion that in the rat dur ing pregnancy

uter ine orn i th ine decarboxylase activity increases. 24

O u r observations conf i rm findings r epor t ed by Reme-

sar et al., 4 who no ted that renal arginase activity more

than doubles dur ing pregnancy. This observat ion is also

consistent with the hypothesis that the agent responsible

for the induct ion of arginase in the myomet r ium dur ing

pregnancy circulates.

In conclusion, we have demons t ra ted that myometr ia l

arginase activity is increased manyfold dur ing pregnancy:

The role of myometr ia l arginase is unknown, but on the

basis of the pr ior work cited, we hypothesize that arginase

has a role supplying the rapidly growing fetus with

polyamines by orn i th ine decarboxylation.

REFERENCES

1. Pegg AE, McCann PP. Polyamine metabolism and function. AmJ Physiol 1982;243:C212-21.

2. Herrera E, Knopp RH, Freinkel N. Carbohydrate metabo- lism in pregnancy, VI: plasma fuels, insulin, liver composi- tion, gluconeogenesis and nitrogen metabolism during late gestation in the fed and fasted rats. J Clin Invest 1969;48: 2260-72.

3. Metzger BE, Agnoli FS, Freinkel N. Effect of sex and preg- nancy on formation of urea and ammonia during gluconeo- genesis in perfused rat livers. Horm Metab Res 1970;2:367-8.

4. Remesar X, Arola LI, Palou A, Alemany M. Arginase activity during pregnancy and lactation. H0rm Metab Res 1984;16: 468-70.

5. Lamers WH, Mooren PG. Role of sex steroid hormones in the normal and glucocorticoid hormone-induced evolution of carbamoylphosphate synthase (ammonia) and arginase activity in rat liver ontogenesis. Biol Neonate 1981 ;40:78-90.

6. Roberge A, Charbonnean R, Berlinguet L. Variation of the enzymes of the urea cycle and asparatate transcarbamoylase in the liver of pregnant rats. CanJ Biochem 1967;45:1371-5.

7. Heby O. Role of polyamines in the control of cell prolifera- tion and differentiation. Differentiation 1981;14:1-20.

8. Oka T, Perry JW. Arginase affects lactogenesis through its influence on the biosynthesis of spermidine. Nature 1974; 250:660-1.

9. Mezl VA, Knox WE. Metabolism of arginine in lactating rat mammary gland. BiochemJ 1977;164:105-13.

10. Mendez JD, Yanez R, Wong C, Hicks JJ. Uterine arginase inhibition affect the rat embryonic development. Contra- ception 1986;33:59%604.

11. Freeman J~, Williams MA. Extrogenic induction of uterine arginase activity. Horm Metab Res 1971;3:352.

12. Corraliza IM, Soler G, Eichmann K, Modolell M. Arginase induction by suppressors of nitric oxide synthesis (IL-4, IL-10 and PGE 2 in murine bone-marrow-derived macropha- ges. Biochem Biophys Res Commun 1995;206:667-73.

13. Lin H, Mosmann TR, Guilbert L, Tuntipopipat S, Wegmann TG. Synthesis of T helper 2-type cytokines at the maternal- fetal interface. J I/nmunol 1993;151:4562-73.

14. Delassus S, Coutinho GC, Saucier C, Darche S, Kourilsky P. Differential cytokine expression in maternal blood and pla- centa during murine gestation. J Immunol 1994;152:2411- 20.

15. Athanassakis-Vassiliadis I. Lymphokine production by de- cidual cells in allogeneic and syngeneic murine pregnancy. Cytokine 1993;5:354-61.

16. Chaouat G, Menu E, Athanassakis I, Wegmann TG. Maternal T cells regulate placental size and fetal survival. Reg Immu- nol 1988;1:143-8.

17. Challis JRG, Heap RB, Illingworth DV. Concentrations of oestrogen and progesterone in the plasma of nonpregnant, pregnant and lactating guinea pigs. J Endocrinol 1971;51: 333-45.

18. Thapar M, Kumari GL, Shrivastav TG, Pandey PK. Hor- monal control of implantation in guinea pigs. Steroids 1988; 52:85-108.

19. Martensson L. The pregnant rabbit, guinea pig, sheep and rhesus monkey as models in reproductive physiology. Eur J Obstet Gynecol Reprod Biol 1984;18:169-82.

20. Spector EB, Rice SCH, Moedjono S, Bernard B, Cederbaum SD. Biochemical properties of arginase in human adult and fetal tissues. Biochem Med 1982;28:165-75.

21. Ruegg UT, Russell AS. A rapid and sensitive assay for argi- nase. Anal Biochem 1980;102:206-12.

22. Fallgren B, Bjellin L, Edvinsson L. Effect of pregnancy and sex steroids on alpha 1-adrenoceptor mechanisms in the guinea pig uterine vascular bed. Pharmacol Toxico11988;63: 375-81.

23. Weiner CP, Lizasoain I, Baylis S, Knowles RG, Charles I, Moncada S. Induction of calcium-dependent nitric oxide synthases by sex hormones. Proc Natl Acad Sci U S A 1994; 91:5212-6.

24. Heald PJ. Changes in ornithine decarboxylase during early implantation in the rat. Biol Reprod 1979;20:1195-9.