Plucenru(1991), 12, 143-151

The Activities of Thiol Proteases in the Rat Visceral Yolk Sac Increase During Late Gestation

JOHN D. GRUBB”, THOMAS R. KOSZALKb, JOSEPH J. DRABICK’ & ROBERT M. METRIONEd Department of Biochemistry and Molecular Biology ThomasJ@rson University, Philadelphia, P,4 19107, US4 u Current address: Department of Microbiology and Immunology, lJniver@ ofRochesterMedica1 Center, Rochester, IVYI 4642, IX4 b Current address: Research Department, Division ofDa~elopnwntu1 Biology, Alfred I. DuPont Institute, Wilmington, DE 19803, I LT.4 ’Current address: Infectious Disease Service, Fkzlter Reed .4nq lVedical Center, Washington, DC 20.307-5000, US4 d To whom correspondence should be addressed

Paper accepted 2510.1990

SUMMARY while the rat YKS has been shown to possess an active lysosomal proteolytic system, there are no published reports on the identity oftheseproteases nor on their changes in activity during the latter half ofgestation. We have used speciJ;c synthetic substrates to show that cathepsins B, L and H arepresent in this organ from days 12.5 to 20.5 of gestation. Cathepsins B and L exhibit a marked inmease in activity beginning on day 15.5 of gestation. By days 19.5-20.5, cathepsin B activity is increased tenfold over that observed on day 12.5. The activity of cathepsin L may be elevated on day 12.5, decreases more than halfby day 14.5 and then increasesfourfald by day 20.5. The activity of cathepsin H does not change throughout this period nor do the cathepsins exhibit marked changes in activity in the placenta during this same period or in the PYS from days 12.5 to 14.5 ofgestation. These results indicate a specific increase in VKS cathepsin B and L activities late in gestation. These enzymes may be involved in meeting the nutritional needs of the emb yo and/or in the degenerative changes which may occur in the VYS and PYSprior to parturition. Studies on the degradation of rat serum albumin by extracts of day 19.5 VKS indicate that cathepsin L may be the quantitatively most important protease in late gestation. 0143-4004/91/020143

+ 09 $05.00/O

IQ 1991 Baillike Tindall I,td

144

Placenta (1991), Vol. 12

INTRODUCTION Considerable evidence has been accumulated to indicate that the visceral yolk sac (VYS) is of critical importance during postimplantation development of the rodent embryo. During midgestation the VYS digests proteins captured by pinocytosis to free amino acids which are incorporated into embryonic protein, presumably after entering the embryonic circulation via the vitelline vasculature (Freeman, Beck and Lloyd, 1981; Freeman and Lloyd, 1983a). Results obtained from in vitro and in vivo experiments using leupeptin, an inhibitor of the lysosomal cysteine proteases, cathepsins B and L, indicate that one or both of these proteases may be responsible for the degradation ofprotein following pinocytosis by the VYS. Injection of leupeptin into pregnant rats on day 8.5 or 9.5 of gestation or culturing of day 9.5 conceptuses in the presence of leupeptin for 48 h resulted in growth retardation and abnormal development, and changes occurred in the VYS which are similar to those seen in storage diseases (Beck and Lowy, 1982; Freeman and Lloyd, 1983b). These abnormalities are probably due to deprivation of the embryo of its supply of amino acids at a crucial period of development. Freeman and Lloyd (1983b) h ave shown that when day 11.5 conceptuses were cultured in the presence of leupeptin for 6 h, [3H]-leucine-labelled serum proteins accumulated within the yolk sac but were not degraded to their constituent amino acids, resulting in a decreased flow of amino acids to the embryo. Knowles et al (1981) have presented evidence suggesting that cysteine proteases may also be active during the latter part of gestation. They found leupeptin to inhibit the degradation of 1251-labelled bovine serum albumin during in vitro culture of intact day 17.5 VYS. In contrast, pepstatin, an inhibitor of the aspartic protease cathepsin D, has been observed to have no effect in vitro on the development of day 9.5 conceptuses (Freeman and Brown, 1985) or on the proteolytic breakdown of ‘251-labelled bovine serum albumin by intact day 17.5 VYS (Knowles et al, 1981). In a preliminary report, Metrione, Koszalka and Drabick (1983) have shown that cathepsin B-like activity increases in the VYS during the latter half of gestation. There have been no other reports of changes in protease activities in this extraembryonic membrane during this time of gestation, We have therefore undertaken to determine the specific activities of cathepsins B, L and H in the VYS in comparison to those in the parietal yolk sac (PYS) and placenta from days 11.5 to 21.5 of gestation using specific peptide substrates.

MATERIALS

AND

METHODS

Pregnant rats derived from the Wistar strain were used in all experiments. The morning on which sperm were found in the vaginal contents was considered to be day 0.5 of gestation. (Z-Arg-Arg-AMC), Z-PhenylZ-Arginine-Arginine-7-amino-4-methylcoumarin alanine-Arginine-AMC (Z-Phe-Arg-AMC), Arginine-AMC (Arg-AMC), Z-Phenyl(Z-Phe-Phe-CHNz) and 7-amino-4-methylalanine-Phenylalanine-diazomethane coumarin (AMC) were purchased from Enzyme Systems Products, Livermore, CA. The four peptide substrates were stored as concentrated stock solutions in dimethylsulphoxide at 4°C and diluted to the appropriate concentration before use. The benzyloxycarbonyl Llysine thiobenzylester (Z-Lys thiobenzylester) was a gift from Dr Elliot Shaw. Preparation of organs and enzyme extracts For each day of gestation studied, rats were killed by cervical dislocation and the uteri isolated and placed in cold PBS. The conceptuses consisting of embryos, extraembryonic membranes

Gnrbb c’tal: kYS Cuthepsin Activities

145

and chorioallantoic placentae were removed from the uterus. The capsular portion of the PYS, if present, was isolated (Clark et al, 1975) and the VYS dissected away from the chorioallantoic placenta, consisting of the maternal and fetal units. The VYS was removed from around the embryo and the amnion separated from the VYS. The VYS, PYS and placenta were washed in cold PBS to remove blood and extraneous tissue and stored at -70°C until used, up to 1 month. The three organs were isolated in parallel from each conceptus on days 12.5-14.5 of gestation. Since the PYS begins to disintegrate on or about day 15.5 of gestation, only the VYS and placenta were isolated on days 15.5-20.5 of gestation. Due to the small size of the VYS from day 12.5 to day 15.5 the number of conceptuses indicated below, for each litter, was divided into two pools. Each pool was then extracted and assayed in triplicate. Thus, for the first litter on day 14.5, six conceptuses out of 15 present were used. These six conceptuses were divided into two pools of three each. Each pool was assayed in triplicate. This same pooling was repeated for the other two litters studied on da!14.5. From day 17.5 to day 20.5 the organs were sufficiently large that pooling was not necessar) These organs were assayed individually. Thus, for the first litter on day 19.5 each of six out of the total of 16 conceptuses was extracted and assayed individually. The number of conceptuses whose organs were extracted and assayed out of the total number of conceptuses in the litter are listed below for each day of gestation: Day Day Day Day Day Day Day

12.5: 13.5: 14.5: 15.5: 17.5: 19.5: 20.5:

16/16, 16/16, 12/12. 8/16, 10/15, 10/19. 6/15,9/14, S/12. 6/13,6/11, S/16. 4/17, 4/17, 4/13. 6/16, 6/13, 6/12. 2/12, 3/16, 3/15.

In order to ensure that an efficient extraction procedure was used, in subsequent experiments we undertook to determine the effect of a variety of extraction conditions on cathepsin B activity. Three buffers were used: 0.1 M sodium acetate, pH 4.5; 0.1 M sodium phosphate, pH 6.0; and 0.1 M Tris-HCl, pH 8.5. All buffers contained 0.15 XI NaCl and 1 rnhl EDTA and were compared to unbuffered 0.15 M NaCl/l rn.V EDTA. None of the buffers was more efficient at extraction than the unbuffered salt solution. Furthermore, even after storage of the extracts at -20X, the activities of all three cathepsins were the same as determined immediately after extraction, indicating that the proteases were not being inactivated by high pH. Neither the presence of high salt concentration (1 \I NaCl), detergent (Tween 20, 0.1 per cent) or sonication improved the extraction. The level of cathepsin B was slightly higher in extracts prepared after freezing than in those where extraction was performed immediately after isolation. However, three additional freeze-tha\l cycles did not result in any further increases in extracted cathepsin B activity. Also, cathepsin B was stable in whole tissue stored at -70°C for at least 4 weeks after tissue isolation. On the basis of these results organ tissue was stored at -70°C after isolation and extracted with 0.15 $1 NaCl, 1 rnM EDTA. The crude enzyme extract was prepared from frozen tissue by homogenizing with ten volumes (1 ml/l00 mg wet tissue) of 0.15 M NaCl, 1 rnM EDTA at 4°C with a Dual1 Teflon or glass homogenizer. After stirring for 1 h at 4”C, the sample was centrifuged at 12 000 g for 10 min at 4°C. The supernatant was retained and stored at -20°C.

146

Placenta (1991), CX. I2

Enzyme assays The thiol esterase assay was according to Metrione (198 1) using Z-Lys thiobenzylester as the substrate and immobilized gluthathione as the sulphydryl activator. A unit of activity is the amount of enzyme which liberates 1 mmol of benzyl mercaptan/min. The activities of individual cathepsins were determined following Kirschke, et al (1983) using synthetic peptide substrates coupled to the fluorescent leaving group, 7-amino-4-methylcoumarin. The respective cathepsin activities are expressed as unitdmg protein, where 1 unit of enzyme activity is equal to the release of 1 nmol of aminomethylcoumarin/min. The protein concentration of each extract assayed was determined according to Lowry et al (195 1). As shown in Table 1, the protein content and total activities of cathepsins B and I, increase 43-, 312- and 93-fold, respectively, in the VYS from days 13 to 21 of gestation. Therefore, in order to represent changes in the concentration of enzymatic activity (i.e. increased synthesis of enzymes, decreased destruction of enzyme, etc.), throughout this report we will present data as specific activity (activity mg total extracted protein). RESULTS Initial studies were performed at various times late in gestation using the thioesterase assay. These studies indicated that there is an approximately eightfold increase in proteolytic potential in the VYS during this period, but no corresponding increase in the PYS (not obtained after day 14.5) or the chorioallantoic placenta, although a low, unchanging activity was detected in both of these organs. The activity at day 18 was dependent on the presence of dithiothreitol, a sulphydryl activator, and inhibited by iodoacetate, a sulphydryl specific inhibitor (Table 2). The molecular weight of this enzyme was determined by gel filtration and found to be 3 1 kDa. While these experiments were useful in determining that the proteolytic components of the VYS preferentially increase late in gestation, and that this activity was due to sulphydryl activated proteases, the substrate employed was unable to distinguish between the main two such enzymes, cathepsins B and L. Cathepsin B is known to catalyse the hydrolysis of this substrate (Bajkowski and Frankfater, 1975) and cathepsin L catalyses the hydrolysis of ZLys-4-nitrobenzylester (Barrett and Kirschke, 1981) and would probably utilize Z-Lysthiobenzylester as well. Therefore, subsequent experiments were undertaken, using the Z-Arg-Arg-AMC (cathepsin B) and Z-Phe-Arg-AMC specific fluorogenic substrates, (cathepsin L). Z-Arg-Arg-AMC is highly specific for cathepsin B and is not cleaved by Table 1. Total cytoplasmic protein and cathepsin B or L activity in the VYS from days 13 to 21 of gestation. Cathepsin

activityh

Day of gestation

Protein”

B

L.

12.5 13.5 14.5 15.5 17.5 19.5 20.5

0.098 0.129 0.490 1.405 2.540 4.199 4.219

0.123 0.250 0.897 4.538 17.65 48.21 38.44

1.023 0.995 2.489 8.964 36.04 79.87 95.48

’ mg protein/organ. ‘units of activity/organ.

147

Cnrbb l’t al: I’YS Cathepsin .4ctizities

cathepsins L or H (Barrett and Kirschke, 1981). While Z-Phe-Arg-AMC is cleaved by cathepsins B and L, the peptidyl diazomethane Z-Phe-Phe-CHNz has been reported to inhibit cathepsin L 100 per cent while it has little effect on cathepsin B (Kirschke and Shaw, 1981). Therefore, cathepsin L activity is determined as the difference in the amount of substrate cleaved in the absence and presence of 1 rnM Z-Phe-Phe-CHNZ. The specificity of each cathepsin assay is suggested by results from gel filtration and ion exchange chromatography of crude extracts from day 19.5 VYS. Only a single peak of activity against Z-Arg-Arg-AMC and Z-Phe-Arg-AMC eluted from a gel filtration column and was in the reported molecular weight range of rat cathepsins B and L, 26-30 kDa (Kirschke et al, 1980). Another day 20 VYS extract was prepared in 50 rnM sodium acetate (PH 5.0) and applied to a CM-Sephadex column. Activity against either substrate only eluted upon application of a sodium chloride gradient. One peak of activity against Z-Arg-Arg-AMC eluted at 150 rn%l NaCl and two peaks of activity against Z-Phe-Arg-AMC eluted at 150 and 300 m,\,l. The first peak of activity was only inhibited 1 per cent by Z-Phe-Phe-CHNZ, whereas the second was inhibited 100 per cent. These results suggest that the first peak of activity is cathepsin B and the second is cathepsin L (Bando, Kominami and Katunuma, 1986). Furthermore, the fractionation of VYS extracts on two different types of columns indicates the absence of significant amounts of other proteases capable of hydrolysing these peptide substrates. From days 12.5 to 14.5 only slight increases in cathepsin B specific activity, with Z-ArgArg- IMC as the substrate, were observed in the VYS (Figures 1 and 2). From day 15.5 until the end of gestation, the specific activity of cathepsin B increased markedly. By days 19.520.5, when the specific activity appeared to reach a maximum, it had elevated more than tenfold over that observed on day 12.5. The specific activity of cathepsin L also exhibited a similar increase during late gestation. Cathepsin L activity may be slightly elevated on day 12.5 and then may decrease to about one-half by day 14.5 after which it increases until da! 20.5 when it is elevated about fourfold over that found at day 14.5. Cathepsin activities in the VYS varied between days to a much greater extent than was observed in the PYS or the placenta (Figure 1). From days 12.5 to 14.5 of gestation, the activity of cathepsin B was similar in all three tissues (Figure 2). However, beginning on day 15.5. cathepsin B activity increased sharply in the VYS whereas in the placenta it remained low and perhaps even decreased slightly on days 17.5-20.5. By day 19.5 of gestation, the activity of cathepsin B was 27-fold greater in the VYS than in the placenta (Figure 2). The specific activity of cathepsin L appeared to always be higher in the VYS than in the PYS or placenta (Figure 3). From days 12.5 to 15.5 cathepsin L activity was always at least threefold Table 2. Sulphydryl nature of the \rYS protease. An extract of day 21 \YS was passed through a Sephadex G-75 column (I.5 x 95 cm) in 0.155 \I NaCl. Fractions of 2.3 ml were

collected and enzyme activity determined using the thiol esterase assay with immobilized glutathione as activator. The fraction with the highest activity was used as a source of enzyme in order to test for the sulphydryl requirement units/ml

no activator with activator 1 \I iodoacetic acid (no activator) 1 \I iodoacetic acid (with activator)

1.32 9.77 :

-

148

Plam ta (1991), Vol. 12

0

I2

I4

16 Day

of

I8

?O

gestotlon

Figure 1. Variation in activities of cathepsins B (O-@), L (A-A) and H ($-4) in crude enzyme extracts of VYS from days 12.5 to 20.5 of gestation. Cathepsin activities are expressed as units/mg protein, where 1 unit is equal to the release of 1 nmol of aminomethyl-coumarin/min at 37°C. Each point represents the mean + s.d. of triplicate determinations made on homogenates prepared by pooled (days 12.5-15.5) or individual organs (days 17.5-20.5) f rom each ofthree pregnant rats. The number of pools assayed for days 12.5-15.5 was six. The number of individual conceptuses assayed for each day from 17.5 to 20.5 days was: day 17.5, 12; day 19.5, 18; day 20.5, 9.

16

18

20

Day of gestation

PYS (O-O) and Figure 2. Variations in cathepsin B activi in crude enzyme extracts of VYS (O-O), placenta (0-O) from days 12.5 to 20.5 or gestation. See Figure 1 legend for definition of enzyme units and for explanation of data points. Each point represents the mean ?I s.d. of triplicate determinations made on homogenates prepared by pooled (days 12.5-15.5) or individual or ans (days 17.5-20.5) from each of three pregnant rats. The number of 001s assayed for days 12.5-1 4 .5 was six. The number of individual conceptuses assayed for each day Prom 17.5 to 20.5 days was: day 17.5, 12; day 19.5,18; da! 20.5, 9. in WS. From days 17.5 to 20.5 of gestation, cathepsin L activity appeared to increase in the VYS such that by day 20.5 it was 40-fold higher in the VYS than in the placenta (Figure

higher

3). The specific activity of cathepsin H in the VYS, PYS and placenta did not change during this period. The thiol esterase activity peak obtained by passing an extract of day 17.5 VYS through a Sephadex G-75 gel filtration column was tested for activity with two aminopeptidase substrates, L-leucine-p-nitroanilide and L-arginine methylester. Activity against either substrate was not detectable, confirming the lack of measurable cathepsin H in this preparation. Furthermore, there was no change in the hydrolysis of the specific cathepsin H

149

Grrrbb tv al: CYS Cathepsin Activities 30

16

18

20

Day of gestation

Figure 3. Variation in cathepsin L activity in crude enzyme extracts of W’S (0-O) PYS (O-Cl) and placenta (0-V) from days 12.5 to 20.5 of gestation. See Figure 1 legend for definition of enzyme units and for explanation of data points. Each point re resents the mean f s.d. of triplicate determinations made on homogenates prepared by pooled (days P2.5-15.5) or individual organs (days 17.5-20.5) from each of three pregnant rats. The number of 001s assayed for days 12.5-15.5 was six. The number of individual conceptuses assayed for each day Prom 17.5 to 20.5 days was: day 17.5,12; day 19.5,18; day 20.5, 9.

substrate, Arg-Ah4C, by extracts of these organs throughout the latter half of gestation (Figure 1 and data not shown). When extracts from day 20 VYS were tested for proteolytic activity against reductively methylated [3H]-labelled rat serum albumin (Rice and Means, 1971) a pH optimum of 4 was observed. This activity was inhibited completely by leupeptin, iodoacetic acid, iodacetamide and antipain, but only 40 per cent by N-ethylmaleimide, and 23 per cent by pepstatin, when all inhibitors were tested at 1 mM. Interestingly, the cathepsin L specific inhibitor, Z-Phe-PheCHNZ, inhibited the protease activity 72 per cent. This result would tend to confirm the concept that, in the VYS, late in gestation, cathepsin L is quantitatively the most important enzyme involved in the degradation of rat serum albumin and that cathepsin B and possibly cathepsin D have secondary roles.

DISCUSSION The results reported here indicate for the first time that catbepsins B and L increase markedly during late gestation in the rat VYS. This observation is consistent with the findings of Livesey and Williams (1979) and Knowles et al (198 1) who, utilizing day 17.5 rat yolk sacs, found active pinocytosis and degradation of both radiolabelled endogenous and exogenous proteins. However, these workers did not determine quantitative relationships as a function of gestational age. Our findings also confirm that protein degradation is inhibited in the VYS by leupeptin, a microbial inhibitor (Aoyagi and Umezawa, 1975) which is effective against cathepsins B and L but is relatively ineffective against cathepsin I-I. Using fluorescence histochemistry, we have also shown that the activity of cathepsin B increases primarily in VYS endodermal cells on days 16.5 and 19.5 of gestation (unpubl. obs.). Although there is good evidence that the VYS of the early postimplantation embryo is important in the rat in supplying nutrients, in the form of amino acids derived from pinocytosed protein (Freeman, Beck and Lloyd, 1981), little is known about the nutritional

Placenta (1991), Vol. I/?

150

role of this extraembryonic membrane late in gestation. Thus, our observation that enzymes known to be involved in the breakdown of proteins in the VYS increase markedly in activity as gestation progresses is of considerable interest. Since the rat VYS continues to be active pinocytically on day 17.5 of gestation (Livesey and Williams, 1979), and perhaps even later, it is conceivable that the nutritional function of this organ continues to assume an important role in development even after the chorioallantoic placenta becomes functional. It is also of interest that during the period just before birth or during birth, the VYS ruptures. Cathepsins B and L may participate, directly or indirectly, in weaking this structure, thus promoting its destruction. In order for the cathepsins to be active in this role it is likely that they would have to be released from the cells of origin. It is noteworthy that Al-Zaid, Gumaa and Bow-Resli (1989) have shown that cathepsin B and G activities increase in rat amniotic fluid during the last third of gestation. They concluded that cathepsins may participate in the destabilization of fetal membranes and, therefore, may contribute to their rupture. It would be of interest to compare cathepsin B and G activities in the rat amnion and VYS to determine which tissue is the more likely source of activity in the amniotic fluid. Both proteases have been shown to be released from cells in culture as active immature forms (Mort, Recklies and Poole, 1984; Gal and Gottesman, 1986). Furthermore, both of these enzymes have been shown to be capable of degrading fibrillar collagen (Kirschke et al, 1982; Maciewicz et al, 1987) and could possibly also degrade the extracellular matrix and basement membranes found in these tissues. We have observed that cathepsin L, but not cathepsin B, appears to decrease in activity between days 12.5 and 14.5. Therefore, it would be of interest to determine if cathepsin L has an activity maximum prior to day 12.5. This period of development is of interest because the proteolytic digestion of exogenous protein by the VYS is of particular importance in supplying amino acids to the embryo for use in protein synthesis during organogenesis. Freeman, Beck and Lloyd (1981) and Freeman and Lloyd (1983a) cultured day 9 conceptuses in vitro for 48 h in the presence of radiolabelled rat serum proteins and found that the proteins were taken up by the VYS, degraded to their constituent amino acids and passed on to the embryo. Furthermore, Beckman et al (1990) have shown that most of the supply of amino acids is supplied to the embryo by this process as compared to the transport of free amino acids by the VYS. When the conceptuses were cultured in the presence of leupeptin, radiolabelled protein accumulated in the VYS and dysmorphogenesis resulted. The results presented here would appear to support the implication made by Freeman and Lloyd (1983b) that cathepsin L may be quantitatively the most important protease in the VYS during midgestation, although these authors could not differentiate between cathepsins B and L with the techniques they employed.

ACKNOWLEDGEMENTS The authors thank Carole Andrews for excellent technical grant HD-18396.

assistance.

This research

was supported

in part by NIH

REFERENCES Al-Zaid, N. S., Gumaa, K. A. & Bow-Resli, M. N. (1989) Changes in amniotic fluid cathepsins with gestational age.Jounzal ofDevelopmental Ph.ysiology, 12, 273-275. Aoyagi, T. & Umezawa, H. (1975) Structures and activities of protease inhibitors of microbial origin. In Proteases and Biological Control (Ed.) Reich, E., Rifkin, D. & Shaw, E. pp. 429-454. New York: Cold Spring Harbor.

Grnbb -the rat visceral yolk sac yields amino Jcids for synthesis of embryonic protein.3obumal ofEmbryology and Experimentul~~lorph~)l~~~~, 73, 307-315. Freeman, S. J. &Lloyd, J. B. (1983b) Inhibition ofproteolysis in rat yolk sac as a cause of teratogenesis. Effects of leupeptin in ritro and br zizo. 3oumnlofEmbryology and Experimental .Morphology, 78, 183-I 93. Freeman, S. J. & Brown, N. A. (1985) Comparative effects of cathepsin inhibitors on rat embryonic development irr t‘l/r~. Evidence that cathepsin D is unimportant in the proteolptic function of yolk sac.3cmmal ~$‘Enrhr)~okq~~ em/ E.pcrimental .l~Jorpholqq, 86, 27 1-281. Freenmn, S. J., Beck, F. & Lloyd, J. B. (1981) The role of the visceral yolk sac in mediating protein utilizatilm h! rat embryos cultured itr vitro. J’ownal ofEmbryo1off: and Experimental .Morphology, 6, 223-234. Gal, S. & Gottesman, M. M. (1986) The major excreted protein of transformed tibroblasts is an a&able acidprorease._~~~olrrmzlofBi&gicul Chemistry, 261, 1760-1765. Kirschke, H. & Shaw, E. (1981) Rapid inactivation ofcathepsin L. by Z-Phe-PheCHN2 and Z-Phe-.4la-CX\7. Bio