Biochimica et Biophysica Acta, 1131 (1992) 145-151 © 1992 Elsevier Science Publishers B.V. All rights reserved 0167-4781/92/$05.00
145
BBAEXP 92386
Differential regulation of kininogen gene expression by estrogen and progesterone in vivo Li-Mei Chen, Peter Chung, Steven Chao, Lee Chao and Julie Chao Department of Biochemistry and Molecular Biology, Medical University of South Carolina, Charleston, SC (USA) (Received 20 January 1992)
Key words: Kininogen; Kallikrein; Estrogen; Progesterone; Gene expression; (Rat)
Kininogens which have multifunctional domains, serve as the precursors of potent vasoactive kinin peptides and also function as cysteine proteinase inhibitors. Given its potential role in blood pressure homeostasis and inflammation, we have examined the regulation of rat kininogen gene expression by sex hormones in vivo. Our studies indicate a differential regulation of kininogen gene expression in rat liver by estrogen and progesterone. Northern and dot blot analysis using a rat low molecular weight kininogen cDNA probe show that kininogen mRNA levels in the liver of female rats are 4-fold higher than those in male rats. Ovariectomy results in a reduction of kininogen transcripts in the liver, while estradiol replacement of the ovariectomized rats increases kininogen mRNA levels. Similarly, Northern blot analysis using a kallikrein cDNA probe shows that estradiol treatment induces an increase of kallikrein gene expression in the kidney of the same animals. In contrast, progesterone treatment of the ovariectomized rats results in an increase in renal kallikrein mRNA levels while it reduces kininogen gene expression as compared to vehicle-treated ovariectomized animals. Immunoreactive kininogen levels in the serum, analyzed by a direct radioimmunoassay and Western blot, are increased by estradiol but slightly decreased by progesterone treatment. Western blot of serum proteins on a two-dimensional polyacrylamide gel reveals that in estradiol-treated ovariectomized rats, the levels of several 68 000 Da kininogens varying in charge are markedly higher than those in ovariectomized rats. The results indicate that estrogen is one of the determinants in regulating low molecular weight kininogen gene expression in vivo. The impact of estrogen-regulated kininogen expression on cardiovascular function awaits further investigation.
Introduction Kininogens are multifunctional proteins which serve as kinin precursors, cysteine proteinase inhibitors, blood coagulation cofactors and acute phase proteins [1,2]. Kinin peptides have a broad spectrum of biological activities including vasodilatation, vasoconstriction, smooth muscle contraction and relaxation, pain production and inflammation [3]. In mammals, three forms of kininogen have been identified as high molecular weight ( H M W ) kininogen, low molecular weight (LMW) kininogen, and T-kininogen which has been identified as an acute phase protein in the rat [4,5]. The major
Abbreviations: SDS, sodium dodecyl sulfate; PMSF, phenylmethylsulfonyl fluoride; aI-AT , al-antitrypsin; PAGE, polyacrylamide gel electrophoresis; DEP, diethyl pyrocarbonate; Ovex, ovariectomy; P, progesterone; E, estradiol; LMW kininogen, low molecular weight kininogen; HMW kininogen, high molecular weight kininogen. Correspondence to: J. Chao, Department of Biochemistry and Molecular Biology, Medical University of South Carolina, 171 Ashley Avenue, Charleston, SC 29425, USA.
site of kininogen synthesis is the liver. Kininogens, like al-antitrypsin [6] or kallikrein-binding proteins [7], are rapidly secreted from the liver into the circulation and are present at high concentrations in the blood. Our previous studies showed that kinin and kininogen levels are markedly increased during spinal cord injury and stroke [8]. Kininogens could potentially function as cysteine proteinase inhibitors which may limit the destructive action of lysosomal enzymes in inflammatory reactions. However, as a precursor of kinin peptides, kininogens could act as proinflammatory agents as well. In both animal models and human subjects, females appear to have less incidents of stroke, hypertension or other cardiovascular problems as compared to males (Ref. 9, and unpublished results). We and others have previously shown that female rats have higher immunoreactive kininogen levels in serum and tissues than male rats [10,11]. Using polyclonal and monoclonal antibodies to kininogen, we have found that L M W kininogen is distributed in various tissues including lung, kidney, brain, spinal cord, salivary gland and other tissues, and its expression appears to be induced during acute phase inflammation [10,12]. Re-
146 cent studies showed sex dimorphism of T-kininogen gene expression in several extrahepatic tissues including lung, brain, kidney and heart [13]. Furthermore, consensus DNA sequences for the binding of the glucocorticoid/estrogen receptor-steroid complexes have been identified in the T-kininogen gene from a rat hepatoma cell line [14]. In order to further understand the regulation and function of kininogens in vivo, we have developed a highly sensitive direct radioimmunoassay for T-kininogen [15] and isolated cDNA clones encoding rat and human LMW kininogens from expression cDNA libraries [16]. These protein and cDNA reagents were used as probes to study the regulation of kininogen gene expression by hormones and other factors. In the present studies, we showed that kininogen gene expression in rat liver is differentially regulated by estrogen and progesterone in the ovariectomized female rats as demonstrated by Northern blotting, Western blotting and by a direct radioimmunoassay. Materials and Methods
Animal treatment and RNA extraction Sprague-Dawley (200-250 g) female rats (Charles River) were either ovariectomized or sham-operated. 3 weeks after the surgery, rats began to receive subcutaneous injections of estradiol benzoate (50 p.g/kg body weight) or progesterone (5 m g / k g body weight) suspended in sesame oil every 48 h for 2 weeks. The control group rats were injected with the vehicle (sesame oil). Rats were anesthetized with pentobarbital (50 mg/kg). The heart was exposed, and 5 ml of blood were withdrawn. Heparin (100 units/rat) was injected into the left ventricle. After 30 s, the vena cava was cut and the circulation perfused via cardiac puncture with at least 30 ml of normal saline until tissues appeared blood-free. Liver and kidney were removed, minced and homogenized with a polytron (1 min) in GIT solution (4 M guanidine thiocyanate, 3 M sodium acetate (pH 6.0), 0.1 M 2-mercaptoethanol). The homogenate was ultracentrifuged (174 000 × g at 20°C for 21 h) over a cesium chloride gradient (5.7 M CsCI, 3 M sodium acetate (pH 6.0)). The resulting RNA pellet was dissolved in DEP-treated water and the RNA concentration was determined by absorbance at 260 nm in a spectrophotometer. Kininogen radioimmunoassay Purified T-kininogen (5 p~g) was labelled with ~zsI, using a lactoperoxidase method, and the labelled Tkininogen was separated from the reaction mixture with a Sephadex G-100 column as described previously [15]. In the antibody titration curve, T-kininogen antiserum dilutions in the assay buffer ranged from 1 : 1000 to 1 : 640 000. 100 #1 of ~25I-1abelled T-kininogen (10 000
cpm) and 100 pA of antibody diluted in the assay buffer were added to 200 ~1 assay buffer, bringing the final volume to 400 ~I, The assay mixtures were incubated at room temperature for 18 h. Antibody-bound Tkininogen was separated from free T-kininogen through centrifugation in an optimum combination of polyethylene glycol and bovine y-globulin. The final antiserum dilution was 1:480 000, and the standard ranged from 160 pg to 20 ng.
Western blot analyses of serum proteins on SDS-PAGE and two-dimensional gels Rat serum (1 #1) was separated using SDS-PAGE under reducing conditions with a 7.5-15% linear gradient gel containing 0.1% SDS [17]. Proteins were transferred to nitrocellulose and blotted with rabbit anti-rat kininogen antiserum followed by ~25I-kininogen similar to the procedures described previously [6]. Two-dimensional gel electrophoresis was performed according to O'Farrell [18]. Briefly, rat serum (10 M) was initially separated in the first-dimensional focusing system (pH 4-6) of the tube gel (10 cm × 2 mm). Second-dimensional electrophoresis was performed on SDS-PAGE as described above. The proteins were either stained with Coomassie brilliant blue or transferred to nitrocellulose for Western blot analysis using a bradykinindirected kininogen monoclonal antibody [12]. After blocking, the membrane was incubated with affinitypurified ~25I-labelled bradykinin-directed kininogen monoclonal antibody (approx. 500000 c p m / m l ) (1D m) for 2 h at room temperature in a buffer containing 0.15 M NaCI, 0.005 M EDTA, 0.05 M Tris-HCl (pH 7.4), 3% bovine serum albumin and 0.05% Nonidet P-40. The blots were washed, and the binding of the labelled bradykinin-directed antibody to kininogen was displayed visually by autoradiography. Immunological screening of a rat liver cDNA library for clones encoding rat kininogen A rat liver h g t l l cDNA library of approx. 2.4.107 P F U / m l (plaque forming units per ml) was plated in soft agarose with Escherichia coli strain Y1090 as the host cell. Nitrocellulose filters soaked in 10 mM IPTG (isopropyl /3-i>thiogalactoside) were used for plaque lifting according to the procedures of Young and Davis [19]. Filters were washed in TBST (50 mM Tris (pH 8.0), 150 mM NaCI and 0.05% Tween-20) three times for 10 rain each and then blocked for 1 h at 30°C with B L O T T O (5% non-fat dry milk, 0.01 M sodium phosphate (pH 7.4), 0.14 M NaCI, 1 mM PMSF, 1 /xg/ml thimerosal, 0.02% NaN 3 and 0.01% antiform A). After being blocked, the filters were incubated with rabbit anti-rat kininogen antiserum (1 : 500 in BLOTTO) for 1 h at 30°C with gentle shaking [15]. The filters were then washed three times in B L O T T O for 10 min each and incubated for 90 min with [125I]kininogen (approx.
147 250000 c p m / m l ) in BLOTTO. After being washed three times with B L O T T O and once with phosphatebuffered saline (0.01 M sodium phosphate/0.14 M NaCl (pH 7.4)), the filters were air-dried and exposed to Kodak X-Omat film. Purified rat T-kininogen at concentrations of 1.5, 0.15 and 0.015 /~g was spotted onto nitrocellulose as positive controls and processed as above. The positive clone which contained the largest cDNA insert (about 1.2 kb) was subcloned into the M13 mpl9 sequencing vector and sequenced as described previously [20]. Sequence analysis showed that the cDNA clone encodes LMW K-kininogen, but it is 38 nucleotides short of a full-length cDNA at its 5' end. The LMW K-kininogen cDNA was used as a probe in the Northern and dot blot analysis.
cDNA probe preparations Double-stranded tissue kallikrein cDNA insert, about 530 bp in length [21], was prepared from M13 mpl8 DNA by restriction digestion (EcoRI/BamH1) followed by agarose gel electrophoresis and electroelution. a~-Antitrypsin cDNA was prepared as described previously [6]. Kininogen, kallikrein and a~-antitrypsin cDNA probes were labelled with [a-S2p]dATP (New England Nuclear Research Products, Boston, MA), using a nick-translation kit (Bethesda Research Laboratories, Bethesda, MD) according to the procedures recommended by the supplier. Unincorporated label was removed by spin-column (G-50), and the specific activity of each probe was around ( 1 - 2 ) . 108 cpm//~g DNA.
Northern blot analysis Total RNAs from liver or kidney were denatured in a solution containing 1 x Mops buffer (5 x Mops buffer = 0.1 M Mops (pH 7.0), 20 mM sodium acetate, and 0.5 mM EDTA), 50% formamide and 2.2 M formaldehyde at 65°C for 15 min. Denatured RNA samples were resolved in a 1.5% agarose gel containing 2.2 M formaldehyde in 1 X Mops buffer. The gel was run at 100 V with constant voltage for 2 h. RNA was visualized by ethidium bromide staining and was transferred to Immobilon-N membrane according to the protocols suggested by the manufacturer. The blot was baked at 80°C for 2 h and prehybridized with a solution containing 6 x SSC (1 x SSC = 0.15 M NaC1, 0.015 M sodium citrate), 5 x D e n h a r d t ' s solution (50 X Denhardt's solution = 1% Ficol, 1% polyvinylpyrrolidone and 1% bovine serum albumin), 0.5% SDS and 100 /zg/ml herring sperm DNA, at 60°C for 2 h. Nick-translated cDNA probe was added to the solution at a final concentration of 3. l0 s c p m / m l and hybridization was carried out for 12 h at 60°C. The blot was washed to a final stringency of 0.5 X SSC/0.1% SDS at 60°C and the autoradiography was carried out with intensifying screen at -80°C.
Kininogen
(x1-AT --Male --Female
1
2
3
4
1
2
3
4
Fig. 1. Dot blot analysis of kininogen and al-antitrypsin mRNA in the liver of male and female Sprague-Dawley rats. Left: RNAs probed with a kininogen cDNA. Right: RNAs probed with an al-antitrypsin cDNA. Total liver RNA from male or female rats: lane 1, 10 p~g; lane 2, 5 #g; lane 3, 2.5 ~g; lane 4, 1.25 p~g.
Dot blot hybridization Aliquots of total RNA were diluted by 2-fold in series with DEP-treated water (10 to 1.25 ~g), mixed with formamide, formaldehyde and 20 x SSC, and incubated at 65°C for 15 min. The denatured RNA solution was applied to the slots of a Minifold (Schleicher and Schfill) and slowly filtered onto nitrocellulose membranes. The wells of the Minifold were rinsed with 0.2 ml of 10 x SSC, and the nitrocellulose filters were then removed and baked for 2 h at 80°C in a vacuum oven. Prehybridization and hybridization of the membrane filters were carried out as described above for Northern blot. Autoradiographs (Kodak XAR-5 film) were obtained and quantitated by densitometry.
Statistical analys& Data are expressed as mean _+ S.E. Differences were assessed by analysis of variance. Differences were considered significant at a level of P < 0.05. Results
Regulation of kininogen gene expression by sex hormones Kininogen gene expression in rat liver was analyzed by dot blot and Northern blot analysis using a kininogen cDNA probe. Fig. 1 shows that the levels of kininogen m R N A in the liver of female rats are 4-fold higher than those in male rats as determined by densitometric scanning of the autoradiogram in dot blot analysis (left). The results are consistent with Northern blot analysis showing a 3-fold higher levels of LMW kininogen m R N A in female rat liver as compared to those in the males (data not shown). When the same blot was stripped and reblotted with an ch-antitrypsin cDNA probe as a control experiment [6], in contrast to kininogen, the transcripts of al-antitrypsin gene in male rats are about 8-fold higher than those of females (Fig. 1, right). Fig. 2 shows the effect of estrogen and progesterone on kininogen (left) and kallikrein (right) gene expression as analyzed by Northern blot hybridization. Ovariectomy of female rats results in a
148 38% decrease of kininogen mRNA levels in the liver (Fig. 2, left, lane 2). Estradiol replacement in ovariectomized females restored kininogen mRNA levels to those of sham-operated rats (Fig. 2, left, lane 4). However, progesterone treatment of the ovariectomized rats results in a slight decrease of the kininogen gene expression (Fig 2, left, lane 3). The results suggest a differential regulation of kininogen gene expression by estradiol vs. progesterone. Similarly, ovariectomy of female rats results in a decrease of tissue kallikrein mRNA levels in the kidney and estradiol replacement in ovariectomized females restored renal kallikrein gene expression to the levels of sham-operated rats (Fig. 2, right, lane 4). Progesterone treatment of the ovariectomized female rats has a greater effect than estradiol and results in increased renal kallikrein mRNA levels which are higher than those of estradioltreated ovariectomized rats. (Fig. 2, right, lane 3). The results show that estradiol up-regulates both hepatic kininogen and renal kallikrein gene expression in vivo
Liver Kininogen
Kidney Kallikrein
while progesterone only up-regulates the synthesis of kallikrein in the kidney.
Hormonal effect on kininogen levels Immunoreactive kininogen levels in the serum of sham-operated, ovariectomized, estradiol- and progesterone-treated rats were analyzed by a direct radioimmunoassay developed for T-kininogen and by Western blot. Fig. 3A shows that ovariectomy of female rats results in a 43% decrease of kininogen content in the serum (from 1214_+ 159 Ixg to 700_+ 155 /xg kininog e n / m l serum, means ± S.E., n = 5, P < 0.05). Estradim replacement in ovariectomized females restored kininogen content to a level not significiantly different from that of sham-operated rats (from 700_+ 155 to 1260 _+ 164 txg kininogen/ml serum, means _+ S.E., n = 5, P < 0.05). However, progesterone treatment results in a further decrease of kininogen levels in the serum (from 700 _+ 155 to 571 _+ 36 # g kininogen/ml serum). These results are consistent with Western blot analysis showing an up-regulation of kininogen level by estradiol (Fig. 3B, lane 3) and down-regulation by progesterone (Fig. 3B, lane 4). In Fig. 3B, only one single band with a molecular mass about 68000 Da in each lane was blotted by the T-kininogen polyclonal antibodies. Western blot analysis of serum kininogens in two-dimensional gels Fig. 4 shows Western blot analysis of kininogens in the serum of vehicle-treated and estradiol-treated ovariectomized rats using a kinin-directed kininogen monoclonal antibody. Serum samples were run in twodimensional gels. Total serum proteins were first separated by isoelectric focusing in a pH 4-6 first-dimensional system and then by second-dimensional SDSPAGE. After electrophoretic transfer of the proteins onto nitrocellulose, kininogens were identified by their ability to bind to an ~251-1abelled bradykinin-directed kininogen monoclonal antibody. Several 68000 Da kininogens can be identified in the two-dimensional gel. In estradiol-treated ovariectomized rats, the levels of several 68 000 Da kininogens varying in charge were markedly higher than those in ovariectomized rats (Fig. 4). The results indicated that low molecular weight kininogen synthesis is up-regulated by estradiol in vivo. Discussion
1
2
3
4
1
2
3
4
Fig. 2. Northern blot analysis of kininogen and kallikrein m R N A s in the female Sprague-Dawley rats. Left: Liver R N A s (10 txg) probed with a kininogen cDNA. Right: kidney R N A s (50 txg) probed with a tissue kallikrein cDNA. Lane 1, sham-operated rat; lane 2, ovariectomized rat; lane 3, progesterone-treated ovariectomized rat; lane 4, estradiol-treated ovariectomized rat.
The present studies show that kininogen gene expression is differentially regulated by estrogen and progesterone in vivo. Kininogen mRNA levels were determined by dot blot and Northern blot analyses using a kininogen cDNA probe. Immunoreactive kininogen levels in the serum were determined by a direct radioimmunoassay for T-kininogen (Fig. 3A) and by Western blotting using a T-kininogen polyclonal
149
M.W. 200K--
A 1500
97K-E 2 Q) (n
68K-1000
43K--
r-.
o c
500
25K--
::1.
Sham
Ovex
Ovex+E
1
Ovex+P
2
3
4
Fig. 3. (A) Effects of estradiol and progesterone on kininogen levels in serum measured by a kininogen radioimmunoassay. Sham, sham-operated rat. Ovex, vehicle-treated ovariectomized rat. Ovex+ E, estradiol-treated ovariectomized rat. Ovex+ P, progesterone-treated ovariectomized rat. (B) Effects of estradiol and progesterone on kininogen levels in serum analyzed by immunoblot. Rat serum (1 /~1) was electrophoresed in a linear gradient (7.5-15%) polyacrylamide gel containing 0.1% SDS and transferred to nitrocellulose. The blot was incubated with kininogen polyclonal antibodies. Lane 1, sham-operated rat; lane 2, ovariectomized rat; lane 3, estradiol-treated ovariectomized rat; lane 4, progesterone-treated ovariectomized rat.
antibody (Fig. 3B) and a bradykinin-directed kininogen monoclonal antibody in two-dimensional gel electrophoresis (Fig. 4). We show that ovariectomy of female rats results in a reduction of both kininogen
mRNA and immunoreactive kininogen levels, and that estradiol replacement restores kininogen content to that of sham-operated female rats. Progesterone replacement further reduces both kininogen mRNA and
ESTRADIOL TREATED
VEHICLE TREATED
tit
a. a ¢~
+ Acid
= Base Isoelectric
focusing
Acid
= Base
pH 4--6
Fig. 4. Western blot analysis of kininogens in vehicle treated- and estradiol-treated female rat serum in two-dimensional gels. Rat serum (10 #l) was first separated by isoelectric focusing with pH 4-6 system in first-dimension and then by second-dimensional SDS-PAGE. Proteins were transferred to nitrocellulose and the blot was incubated with 125I-labelled bradykinin-directed kininogen monoclonal antibody.
150 immunoreactive kininogen levels. The results indicated that estrogen and progesterone exhibit a differential effect in regulating kininogen gene expression in vivo. Since rat K- and T-kininogens share a high degree of nucleic acid sequence identity [25] and their cDNAs have cross-reactivity between each other, we could not precisely describe which species of the mRNA is regulated by estrogen in the Northern blot using the 1.2 kb LMW K-kininogen cDNA probe. Furthermore, although the kininogen cDNA probe covers the aminoterminal portion preceding the bradykinin moiety, the HMW kininogen mRNA is not detected in the Northern blot (Fig. 2, left). The reason could be that HMW kininogen is expressed in low abundance in the total liver RNA. In the control experiment, Northern blot hybridization showed that both estrogen and progesterone up-regulate renal kallikrein synthesis (Fig. 2, right). Regulation of renal kallikrein gene expression by sex hormones has recently been shown by Madeddu et al. [22]. Plasma kininogen may play important roles in the estrous cycle, pregnancy, parturition and puerperium of rats and in the human menstrual cycle. Since the ovary is an important part of a very complex endocrine system and bears multifunctional characteristics, ovariectomy not only decreases the level of estradiol but also causes a series of changes in the homeostasis of organisms. Consequently, other hormones and factors may also contribute to the regulation of kininogen during this process. Our present work showed that estrogen but not progesterone replacement in the ovariectomized rats increases LMW kininogen synthesis at both mRNA and protein levels. The findings are consistent with those by others showing that estrogen regulates rat T-kininogen but not HMW kininogen levels in the plasma [23]. Similarly the expression of angiotensinogen, another vasoactive peptide precursor, has also been shown to be controlled by estrogen [24]. The physiological significance of regulation of kininogen and angiotensinogen gene expression by estrogen has yet to be established. Analysis of Western blots from a two-dimensional gel using a bradykinin-directed kininogen monoclonal antibody reveals that, in the serum of estradiol-treated ovariectomized rats, the concentrations of several kininogens, similar in molecular weights but different in isoelectric points, are higher than those in the serum of ovariectomized females. Since T- and K-kininogens share a high degree of sequence identity [25], polyclonal antibody against T-kininogen and monoclonal anti-bradykinin-directed kininogen antibody can not distinguish these two LMW kininogens in the radioimmunoassay or immunoblot. Further, the reported molecular weights of rat low molecular weight kininogens (T- and K-kininogen) are similar [26]. The varying charges of these kininogens may reflect either a differ-
ence in carbohydrate content or a small number of amino acid substitutions. Liver is the major site of kininogen gene expression. However, kininogen content in the liver is very low. Comparison of the specific activities of kininogen content in the liver and serum (both expressed as /~g kininogen per mg protein) indicates that kininogen levels in the liver extracts are about 1% of those in the serum (data not shown). The results indicate that kininogen is a secretory protein and is secreted into the blood stream rapidly after its synthesis. Low levels of kininogen gene expression have been identified in other tissues including lung, kidney, spinal cord, adrenal gland, brain, and heart [12,13]. In addition, using immunohistochemical techniques, we and others have recently localized kininogen in rat hypothalamus, spinal cord and brain stem [27,28]. The findings that kininogen levels are regulated by estrogen in vivo and are increased during spinal cord trauma, brain injury or other tissues, suggest that the estrogen-regulated expression of kininogen may play potential protective roles in the cardiovascular system. This notion is supported by the observations that male subjects have more frequent incidents of stroke and hypertension than female subjects [9]. The impact of female sex hormones in the regulation of kininogen gene expression awaits further investigation.
Acknowledgement This work was supported by the National Institute of Health Grant HL 29397.
References 1 Mflller-Esterl, W., lwanaga, S. and Nakanishi, S. (1986) Trends Biochem. Sci. 11,336-339. 2 Miiller-Esterl, W. (1987) Seminars in Thrombosis and Hemostasis. Vol. 13 (No. 1), pp. 115-126, Thieme Medical Publishers, New York. 3 Johnson, A.R. (1979) In Bradyki~in, Kallidin and Kallikrein. (Erd6s, E.G., ed.), Pharmacology 25, Suppl., pp. 357 399, Springer-Verlag, Berlin. 4 Kato, H., Nagasawa, S. and lwanaga, S. (1981) Methods Enzymol. 80, 172-198. 5 Barlas, A., Okamoto, H. and Greenbaum, L.M. (1985) Biochem. Biophys. Res. Commun. 129, 280-286. 6 Chao, S., Chai, K.X., Chao, L. and Chao, J. (1990) Biochemistry 29, 323-329. 7 Chao, J., Chai, K.X., Chen, L.M., Xiong, W., Chao, S., Cheryl, W.M., Wang, L., Lu, H.S. and Chao, L. (1990) J. Biol. Chem. 265, 16394-16401. 8 Xu, J., Hsu, C.Y., Junker, H., Chao, S., Hogan, E.L. and Chao, J. (1991) J. Neurochem. 57, 975-980. 9 Von Eiff, A.W., Gogolin, E., Jacobs, V. and Neus, H. (1986) Clin. Exp. Hyper. (A) 8, 577-581. 10 Chao, J., Chao, S., Xiong, W., Chen, L.M., Swain, C. and Chao, L. (1989) Adv. Exp. Med. Biol. 247, 297-303. 11 Oh-ishi, S., Hayashi, I., Kusunoki, A., Nagashima, Y., Hayashi, M., Yamaki, K., Utsunomiya, I. and Yamasu, A. (1988) Biochem. Biophys. Res. Commun. 150, 1069-1076.
151 12 Chao, J., Swain, C., Chao, S., Xiong, W. and Chao, L. (1988) Biochim. Biophys. Acta 964, 329-339. 13 Mann, E.A. and Lingrel, J.B. (1991) Biochem. Biophys. Res. Commun. 174, 417-423. 14 Anderson, K.P. and Lingrel, J.B. (1989) Nucleic Acids Res. 17, 2835-2848. 15 Chao, S., Chao, L. and Chao, J. (1989) Biochim. Biophys. Acta 991,447-483. 16 Chao, S., Chao, J. and Chao, L. (1990) Methods in Nucleic Acids Research, Chapter 15, pp. 307-320, CRC Press, Boca Raton. 17 Chao, J., Woodley, C., Chao, L. and Margolius, H.S. (1983) J. Biol. Chem. 258, 15173-15178. 18 O'Farrell, P.H. (1975) J. Biol. Chem. 250, 4007-4021. 19 Young, R.A. and Davis, R.W. (1983) Proc. Natl. Acad. Sci. USA 80, 1194-1198. 20 Chao, S., Chao, L. and Chao, J. (1989) BioTechniques 7, 68-71.
21 Gerald, W.L., Chao, J. and Chao, J. (1986) Biochim. Biophys. Acta. 886, 1-14. 22 Madeddu, P., Glorioso, N., Maioli, M., Demontis, M.P., Varoni, M.V., Anania, V., Xiong, W., Chai, K. and Chao, J. (1991) J. Hypertens. 9 (Suppl 6), $244-$245. 23 Bouhnik, J., Savoie, F., Michaud, A., Baussant, T., Alhenc-Gelas, F. and Corvol, P. (1989) Life Sci. 44, 1859-1866. 24 Dzau, V.J. and Hermann, H.C. (1982) Life Sci. 30, 577-584. 25 Furuto-Kato, S., Matsumoto, A., Kitamura, N. and Nakamishi, S. (1985) J. Biol. Chem. 260, 12054-12059. 26 Enjyoji, K.I., Kato, H., Hayashi, I., Oh-ishi, S. and lwanaga, S. (1988) J. Biol. Chem. 263, 965-972. 27 Li, Z.H., Xu, J., Chao, J. and Hogan, E.L. (1992) FASEB J. 6, A1622, abstract. 28 Richoux, J.P., Gelly, J.L., Bouhnik, J., Baussant, T., Alhenc-Gelas, F., Grignon, G. and Corvol, P. (1991) Histochemistry 96, 229-243.
145
BBAEXP 92386
Differential regulation of kininogen gene expression by estrogen and progesterone in vivo Li-Mei Chen, Peter Chung, Steven Chao, Lee Chao and Julie Chao Department of Biochemistry and Molecular Biology, Medical University of South Carolina, Charleston, SC (USA) (Received 20 January 1992)
Key words: Kininogen; Kallikrein; Estrogen; Progesterone; Gene expression; (Rat)
Kininogens which have multifunctional domains, serve as the precursors of potent vasoactive kinin peptides and also function as cysteine proteinase inhibitors. Given its potential role in blood pressure homeostasis and inflammation, we have examined the regulation of rat kininogen gene expression by sex hormones in vivo. Our studies indicate a differential regulation of kininogen gene expression in rat liver by estrogen and progesterone. Northern and dot blot analysis using a rat low molecular weight kininogen cDNA probe show that kininogen mRNA levels in the liver of female rats are 4-fold higher than those in male rats. Ovariectomy results in a reduction of kininogen transcripts in the liver, while estradiol replacement of the ovariectomized rats increases kininogen mRNA levels. Similarly, Northern blot analysis using a kallikrein cDNA probe shows that estradiol treatment induces an increase of kallikrein gene expression in the kidney of the same animals. In contrast, progesterone treatment of the ovariectomized rats results in an increase in renal kallikrein mRNA levels while it reduces kininogen gene expression as compared to vehicle-treated ovariectomized animals. Immunoreactive kininogen levels in the serum, analyzed by a direct radioimmunoassay and Western blot, are increased by estradiol but slightly decreased by progesterone treatment. Western blot of serum proteins on a two-dimensional polyacrylamide gel reveals that in estradiol-treated ovariectomized rats, the levels of several 68 000 Da kininogens varying in charge are markedly higher than those in ovariectomized rats. The results indicate that estrogen is one of the determinants in regulating low molecular weight kininogen gene expression in vivo. The impact of estrogen-regulated kininogen expression on cardiovascular function awaits further investigation.
Introduction Kininogens are multifunctional proteins which serve as kinin precursors, cysteine proteinase inhibitors, blood coagulation cofactors and acute phase proteins [1,2]. Kinin peptides have a broad spectrum of biological activities including vasodilatation, vasoconstriction, smooth muscle contraction and relaxation, pain production and inflammation [3]. In mammals, three forms of kininogen have been identified as high molecular weight ( H M W ) kininogen, low molecular weight (LMW) kininogen, and T-kininogen which has been identified as an acute phase protein in the rat [4,5]. The major
Abbreviations: SDS, sodium dodecyl sulfate; PMSF, phenylmethylsulfonyl fluoride; aI-AT , al-antitrypsin; PAGE, polyacrylamide gel electrophoresis; DEP, diethyl pyrocarbonate; Ovex, ovariectomy; P, progesterone; E, estradiol; LMW kininogen, low molecular weight kininogen; HMW kininogen, high molecular weight kininogen. Correspondence to: J. Chao, Department of Biochemistry and Molecular Biology, Medical University of South Carolina, 171 Ashley Avenue, Charleston, SC 29425, USA.
site of kininogen synthesis is the liver. Kininogens, like al-antitrypsin [6] or kallikrein-binding proteins [7], are rapidly secreted from the liver into the circulation and are present at high concentrations in the blood. Our previous studies showed that kinin and kininogen levels are markedly increased during spinal cord injury and stroke [8]. Kininogens could potentially function as cysteine proteinase inhibitors which may limit the destructive action of lysosomal enzymes in inflammatory reactions. However, as a precursor of kinin peptides, kininogens could act as proinflammatory agents as well. In both animal models and human subjects, females appear to have less incidents of stroke, hypertension or other cardiovascular problems as compared to males (Ref. 9, and unpublished results). We and others have previously shown that female rats have higher immunoreactive kininogen levels in serum and tissues than male rats [10,11]. Using polyclonal and monoclonal antibodies to kininogen, we have found that L M W kininogen is distributed in various tissues including lung, kidney, brain, spinal cord, salivary gland and other tissues, and its expression appears to be induced during acute phase inflammation [10,12]. Re-
146 cent studies showed sex dimorphism of T-kininogen gene expression in several extrahepatic tissues including lung, brain, kidney and heart [13]. Furthermore, consensus DNA sequences for the binding of the glucocorticoid/estrogen receptor-steroid complexes have been identified in the T-kininogen gene from a rat hepatoma cell line [14]. In order to further understand the regulation and function of kininogens in vivo, we have developed a highly sensitive direct radioimmunoassay for T-kininogen [15] and isolated cDNA clones encoding rat and human LMW kininogens from expression cDNA libraries [16]. These protein and cDNA reagents were used as probes to study the regulation of kininogen gene expression by hormones and other factors. In the present studies, we showed that kininogen gene expression in rat liver is differentially regulated by estrogen and progesterone in the ovariectomized female rats as demonstrated by Northern blotting, Western blotting and by a direct radioimmunoassay. Materials and Methods
Animal treatment and RNA extraction Sprague-Dawley (200-250 g) female rats (Charles River) were either ovariectomized or sham-operated. 3 weeks after the surgery, rats began to receive subcutaneous injections of estradiol benzoate (50 p.g/kg body weight) or progesterone (5 m g / k g body weight) suspended in sesame oil every 48 h for 2 weeks. The control group rats were injected with the vehicle (sesame oil). Rats were anesthetized with pentobarbital (50 mg/kg). The heart was exposed, and 5 ml of blood were withdrawn. Heparin (100 units/rat) was injected into the left ventricle. After 30 s, the vena cava was cut and the circulation perfused via cardiac puncture with at least 30 ml of normal saline until tissues appeared blood-free. Liver and kidney were removed, minced and homogenized with a polytron (1 min) in GIT solution (4 M guanidine thiocyanate, 3 M sodium acetate (pH 6.0), 0.1 M 2-mercaptoethanol). The homogenate was ultracentrifuged (174 000 × g at 20°C for 21 h) over a cesium chloride gradient (5.7 M CsCI, 3 M sodium acetate (pH 6.0)). The resulting RNA pellet was dissolved in DEP-treated water and the RNA concentration was determined by absorbance at 260 nm in a spectrophotometer. Kininogen radioimmunoassay Purified T-kininogen (5 p~g) was labelled with ~zsI, using a lactoperoxidase method, and the labelled Tkininogen was separated from the reaction mixture with a Sephadex G-100 column as described previously [15]. In the antibody titration curve, T-kininogen antiserum dilutions in the assay buffer ranged from 1 : 1000 to 1 : 640 000. 100 #1 of ~25I-1abelled T-kininogen (10 000
cpm) and 100 pA of antibody diluted in the assay buffer were added to 200 ~1 assay buffer, bringing the final volume to 400 ~I, The assay mixtures were incubated at room temperature for 18 h. Antibody-bound Tkininogen was separated from free T-kininogen through centrifugation in an optimum combination of polyethylene glycol and bovine y-globulin. The final antiserum dilution was 1:480 000, and the standard ranged from 160 pg to 20 ng.
Western blot analyses of serum proteins on SDS-PAGE and two-dimensional gels Rat serum (1 #1) was separated using SDS-PAGE under reducing conditions with a 7.5-15% linear gradient gel containing 0.1% SDS [17]. Proteins were transferred to nitrocellulose and blotted with rabbit anti-rat kininogen antiserum followed by ~25I-kininogen similar to the procedures described previously [6]. Two-dimensional gel electrophoresis was performed according to O'Farrell [18]. Briefly, rat serum (10 M) was initially separated in the first-dimensional focusing system (pH 4-6) of the tube gel (10 cm × 2 mm). Second-dimensional electrophoresis was performed on SDS-PAGE as described above. The proteins were either stained with Coomassie brilliant blue or transferred to nitrocellulose for Western blot analysis using a bradykinindirected kininogen monoclonal antibody [12]. After blocking, the membrane was incubated with affinitypurified ~25I-labelled bradykinin-directed kininogen monoclonal antibody (approx. 500000 c p m / m l ) (1D m) for 2 h at room temperature in a buffer containing 0.15 M NaCI, 0.005 M EDTA, 0.05 M Tris-HCl (pH 7.4), 3% bovine serum albumin and 0.05% Nonidet P-40. The blots were washed, and the binding of the labelled bradykinin-directed antibody to kininogen was displayed visually by autoradiography. Immunological screening of a rat liver cDNA library for clones encoding rat kininogen A rat liver h g t l l cDNA library of approx. 2.4.107 P F U / m l (plaque forming units per ml) was plated in soft agarose with Escherichia coli strain Y1090 as the host cell. Nitrocellulose filters soaked in 10 mM IPTG (isopropyl /3-i>thiogalactoside) were used for plaque lifting according to the procedures of Young and Davis [19]. Filters were washed in TBST (50 mM Tris (pH 8.0), 150 mM NaCI and 0.05% Tween-20) three times for 10 rain each and then blocked for 1 h at 30°C with B L O T T O (5% non-fat dry milk, 0.01 M sodium phosphate (pH 7.4), 0.14 M NaCI, 1 mM PMSF, 1 /xg/ml thimerosal, 0.02% NaN 3 and 0.01% antiform A). After being blocked, the filters were incubated with rabbit anti-rat kininogen antiserum (1 : 500 in BLOTTO) for 1 h at 30°C with gentle shaking [15]. The filters were then washed three times in B L O T T O for 10 min each and incubated for 90 min with [125I]kininogen (approx.
147 250000 c p m / m l ) in BLOTTO. After being washed three times with B L O T T O and once with phosphatebuffered saline (0.01 M sodium phosphate/0.14 M NaCl (pH 7.4)), the filters were air-dried and exposed to Kodak X-Omat film. Purified rat T-kininogen at concentrations of 1.5, 0.15 and 0.015 /~g was spotted onto nitrocellulose as positive controls and processed as above. The positive clone which contained the largest cDNA insert (about 1.2 kb) was subcloned into the M13 mpl9 sequencing vector and sequenced as described previously [20]. Sequence analysis showed that the cDNA clone encodes LMW K-kininogen, but it is 38 nucleotides short of a full-length cDNA at its 5' end. The LMW K-kininogen cDNA was used as a probe in the Northern and dot blot analysis.
cDNA probe preparations Double-stranded tissue kallikrein cDNA insert, about 530 bp in length [21], was prepared from M13 mpl8 DNA by restriction digestion (EcoRI/BamH1) followed by agarose gel electrophoresis and electroelution. a~-Antitrypsin cDNA was prepared as described previously [6]. Kininogen, kallikrein and a~-antitrypsin cDNA probes were labelled with [a-S2p]dATP (New England Nuclear Research Products, Boston, MA), using a nick-translation kit (Bethesda Research Laboratories, Bethesda, MD) according to the procedures recommended by the supplier. Unincorporated label was removed by spin-column (G-50), and the specific activity of each probe was around ( 1 - 2 ) . 108 cpm//~g DNA.
Northern blot analysis Total RNAs from liver or kidney were denatured in a solution containing 1 x Mops buffer (5 x Mops buffer = 0.1 M Mops (pH 7.0), 20 mM sodium acetate, and 0.5 mM EDTA), 50% formamide and 2.2 M formaldehyde at 65°C for 15 min. Denatured RNA samples were resolved in a 1.5% agarose gel containing 2.2 M formaldehyde in 1 X Mops buffer. The gel was run at 100 V with constant voltage for 2 h. RNA was visualized by ethidium bromide staining and was transferred to Immobilon-N membrane according to the protocols suggested by the manufacturer. The blot was baked at 80°C for 2 h and prehybridized with a solution containing 6 x SSC (1 x SSC = 0.15 M NaC1, 0.015 M sodium citrate), 5 x D e n h a r d t ' s solution (50 X Denhardt's solution = 1% Ficol, 1% polyvinylpyrrolidone and 1% bovine serum albumin), 0.5% SDS and 100 /zg/ml herring sperm DNA, at 60°C for 2 h. Nick-translated cDNA probe was added to the solution at a final concentration of 3. l0 s c p m / m l and hybridization was carried out for 12 h at 60°C. The blot was washed to a final stringency of 0.5 X SSC/0.1% SDS at 60°C and the autoradiography was carried out with intensifying screen at -80°C.
Kininogen
(x1-AT --Male --Female
1
2
3
4
1
2
3
4
Fig. 1. Dot blot analysis of kininogen and al-antitrypsin mRNA in the liver of male and female Sprague-Dawley rats. Left: RNAs probed with a kininogen cDNA. Right: RNAs probed with an al-antitrypsin cDNA. Total liver RNA from male or female rats: lane 1, 10 p~g; lane 2, 5 #g; lane 3, 2.5 ~g; lane 4, 1.25 p~g.
Dot blot hybridization Aliquots of total RNA were diluted by 2-fold in series with DEP-treated water (10 to 1.25 ~g), mixed with formamide, formaldehyde and 20 x SSC, and incubated at 65°C for 15 min. The denatured RNA solution was applied to the slots of a Minifold (Schleicher and Schfill) and slowly filtered onto nitrocellulose membranes. The wells of the Minifold were rinsed with 0.2 ml of 10 x SSC, and the nitrocellulose filters were then removed and baked for 2 h at 80°C in a vacuum oven. Prehybridization and hybridization of the membrane filters were carried out as described above for Northern blot. Autoradiographs (Kodak XAR-5 film) were obtained and quantitated by densitometry.
Statistical analys& Data are expressed as mean _+ S.E. Differences were assessed by analysis of variance. Differences were considered significant at a level of P < 0.05. Results
Regulation of kininogen gene expression by sex hormones Kininogen gene expression in rat liver was analyzed by dot blot and Northern blot analysis using a kininogen cDNA probe. Fig. 1 shows that the levels of kininogen m R N A in the liver of female rats are 4-fold higher than those in male rats as determined by densitometric scanning of the autoradiogram in dot blot analysis (left). The results are consistent with Northern blot analysis showing a 3-fold higher levels of LMW kininogen m R N A in female rat liver as compared to those in the males (data not shown). When the same blot was stripped and reblotted with an ch-antitrypsin cDNA probe as a control experiment [6], in contrast to kininogen, the transcripts of al-antitrypsin gene in male rats are about 8-fold higher than those of females (Fig. 1, right). Fig. 2 shows the effect of estrogen and progesterone on kininogen (left) and kallikrein (right) gene expression as analyzed by Northern blot hybridization. Ovariectomy of female rats results in a
148 38% decrease of kininogen mRNA levels in the liver (Fig. 2, left, lane 2). Estradiol replacement in ovariectomized females restored kininogen mRNA levels to those of sham-operated rats (Fig. 2, left, lane 4). However, progesterone treatment of the ovariectomized rats results in a slight decrease of the kininogen gene expression (Fig 2, left, lane 3). The results suggest a differential regulation of kininogen gene expression by estradiol vs. progesterone. Similarly, ovariectomy of female rats results in a decrease of tissue kallikrein mRNA levels in the kidney and estradiol replacement in ovariectomized females restored renal kallikrein gene expression to the levels of sham-operated rats (Fig. 2, right, lane 4). Progesterone treatment of the ovariectomized female rats has a greater effect than estradiol and results in increased renal kallikrein mRNA levels which are higher than those of estradioltreated ovariectomized rats. (Fig. 2, right, lane 3). The results show that estradiol up-regulates both hepatic kininogen and renal kallikrein gene expression in vivo
Liver Kininogen
Kidney Kallikrein
while progesterone only up-regulates the synthesis of kallikrein in the kidney.
Hormonal effect on kininogen levels Immunoreactive kininogen levels in the serum of sham-operated, ovariectomized, estradiol- and progesterone-treated rats were analyzed by a direct radioimmunoassay developed for T-kininogen and by Western blot. Fig. 3A shows that ovariectomy of female rats results in a 43% decrease of kininogen content in the serum (from 1214_+ 159 Ixg to 700_+ 155 /xg kininog e n / m l serum, means ± S.E., n = 5, P < 0.05). Estradim replacement in ovariectomized females restored kininogen content to a level not significiantly different from that of sham-operated rats (from 700_+ 155 to 1260 _+ 164 txg kininogen/ml serum, means _+ S.E., n = 5, P < 0.05). However, progesterone treatment results in a further decrease of kininogen levels in the serum (from 700 _+ 155 to 571 _+ 36 # g kininogen/ml serum). These results are consistent with Western blot analysis showing an up-regulation of kininogen level by estradiol (Fig. 3B, lane 3) and down-regulation by progesterone (Fig. 3B, lane 4). In Fig. 3B, only one single band with a molecular mass about 68000 Da in each lane was blotted by the T-kininogen polyclonal antibodies. Western blot analysis of serum kininogens in two-dimensional gels Fig. 4 shows Western blot analysis of kininogens in the serum of vehicle-treated and estradiol-treated ovariectomized rats using a kinin-directed kininogen monoclonal antibody. Serum samples were run in twodimensional gels. Total serum proteins were first separated by isoelectric focusing in a pH 4-6 first-dimensional system and then by second-dimensional SDSPAGE. After electrophoretic transfer of the proteins onto nitrocellulose, kininogens were identified by their ability to bind to an ~251-1abelled bradykinin-directed kininogen monoclonal antibody. Several 68000 Da kininogens can be identified in the two-dimensional gel. In estradiol-treated ovariectomized rats, the levels of several 68 000 Da kininogens varying in charge were markedly higher than those in ovariectomized rats (Fig. 4). The results indicated that low molecular weight kininogen synthesis is up-regulated by estradiol in vivo. Discussion
1
2
3
4
1
2
3
4
Fig. 2. Northern blot analysis of kininogen and kallikrein m R N A s in the female Sprague-Dawley rats. Left: Liver R N A s (10 txg) probed with a kininogen cDNA. Right: kidney R N A s (50 txg) probed with a tissue kallikrein cDNA. Lane 1, sham-operated rat; lane 2, ovariectomized rat; lane 3, progesterone-treated ovariectomized rat; lane 4, estradiol-treated ovariectomized rat.
The present studies show that kininogen gene expression is differentially regulated by estrogen and progesterone in vivo. Kininogen mRNA levels were determined by dot blot and Northern blot analyses using a kininogen cDNA probe. Immunoreactive kininogen levels in the serum were determined by a direct radioimmunoassay for T-kininogen (Fig. 3A) and by Western blotting using a T-kininogen polyclonal
149
M.W. 200K--
A 1500
97K-E 2 Q) (n
68K-1000
43K--
r-.
o c
500
25K--
::1.
Sham
Ovex
Ovex+E
1
Ovex+P
2
3
4
Fig. 3. (A) Effects of estradiol and progesterone on kininogen levels in serum measured by a kininogen radioimmunoassay. Sham, sham-operated rat. Ovex, vehicle-treated ovariectomized rat. Ovex+ E, estradiol-treated ovariectomized rat. Ovex+ P, progesterone-treated ovariectomized rat. (B) Effects of estradiol and progesterone on kininogen levels in serum analyzed by immunoblot. Rat serum (1 /~1) was electrophoresed in a linear gradient (7.5-15%) polyacrylamide gel containing 0.1% SDS and transferred to nitrocellulose. The blot was incubated with kininogen polyclonal antibodies. Lane 1, sham-operated rat; lane 2, ovariectomized rat; lane 3, estradiol-treated ovariectomized rat; lane 4, progesterone-treated ovariectomized rat.
antibody (Fig. 3B) and a bradykinin-directed kininogen monoclonal antibody in two-dimensional gel electrophoresis (Fig. 4). We show that ovariectomy of female rats results in a reduction of both kininogen
mRNA and immunoreactive kininogen levels, and that estradiol replacement restores kininogen content to that of sham-operated female rats. Progesterone replacement further reduces both kininogen mRNA and
ESTRADIOL TREATED
VEHICLE TREATED
tit
a. a ¢~
+ Acid
= Base Isoelectric
focusing
Acid
= Base
pH 4--6
Fig. 4. Western blot analysis of kininogens in vehicle treated- and estradiol-treated female rat serum in two-dimensional gels. Rat serum (10 #l) was first separated by isoelectric focusing with pH 4-6 system in first-dimension and then by second-dimensional SDS-PAGE. Proteins were transferred to nitrocellulose and the blot was incubated with 125I-labelled bradykinin-directed kininogen monoclonal antibody.
150 immunoreactive kininogen levels. The results indicated that estrogen and progesterone exhibit a differential effect in regulating kininogen gene expression in vivo. Since rat K- and T-kininogens share a high degree of nucleic acid sequence identity [25] and their cDNAs have cross-reactivity between each other, we could not precisely describe which species of the mRNA is regulated by estrogen in the Northern blot using the 1.2 kb LMW K-kininogen cDNA probe. Furthermore, although the kininogen cDNA probe covers the aminoterminal portion preceding the bradykinin moiety, the HMW kininogen mRNA is not detected in the Northern blot (Fig. 2, left). The reason could be that HMW kininogen is expressed in low abundance in the total liver RNA. In the control experiment, Northern blot hybridization showed that both estrogen and progesterone up-regulate renal kallikrein synthesis (Fig. 2, right). Regulation of renal kallikrein gene expression by sex hormones has recently been shown by Madeddu et al. [22]. Plasma kininogen may play important roles in the estrous cycle, pregnancy, parturition and puerperium of rats and in the human menstrual cycle. Since the ovary is an important part of a very complex endocrine system and bears multifunctional characteristics, ovariectomy not only decreases the level of estradiol but also causes a series of changes in the homeostasis of organisms. Consequently, other hormones and factors may also contribute to the regulation of kininogen during this process. Our present work showed that estrogen but not progesterone replacement in the ovariectomized rats increases LMW kininogen synthesis at both mRNA and protein levels. The findings are consistent with those by others showing that estrogen regulates rat T-kininogen but not HMW kininogen levels in the plasma [23]. Similarly the expression of angiotensinogen, another vasoactive peptide precursor, has also been shown to be controlled by estrogen [24]. The physiological significance of regulation of kininogen and angiotensinogen gene expression by estrogen has yet to be established. Analysis of Western blots from a two-dimensional gel using a bradykinin-directed kininogen monoclonal antibody reveals that, in the serum of estradiol-treated ovariectomized rats, the concentrations of several kininogens, similar in molecular weights but different in isoelectric points, are higher than those in the serum of ovariectomized females. Since T- and K-kininogens share a high degree of sequence identity [25], polyclonal antibody against T-kininogen and monoclonal anti-bradykinin-directed kininogen antibody can not distinguish these two LMW kininogens in the radioimmunoassay or immunoblot. Further, the reported molecular weights of rat low molecular weight kininogens (T- and K-kininogen) are similar [26]. The varying charges of these kininogens may reflect either a differ-
ence in carbohydrate content or a small number of amino acid substitutions. Liver is the major site of kininogen gene expression. However, kininogen content in the liver is very low. Comparison of the specific activities of kininogen content in the liver and serum (both expressed as /~g kininogen per mg protein) indicates that kininogen levels in the liver extracts are about 1% of those in the serum (data not shown). The results indicate that kininogen is a secretory protein and is secreted into the blood stream rapidly after its synthesis. Low levels of kininogen gene expression have been identified in other tissues including lung, kidney, spinal cord, adrenal gland, brain, and heart [12,13]. In addition, using immunohistochemical techniques, we and others have recently localized kininogen in rat hypothalamus, spinal cord and brain stem [27,28]. The findings that kininogen levels are regulated by estrogen in vivo and are increased during spinal cord trauma, brain injury or other tissues, suggest that the estrogen-regulated expression of kininogen may play potential protective roles in the cardiovascular system. This notion is supported by the observations that male subjects have more frequent incidents of stroke and hypertension than female subjects [9]. The impact of female sex hormones in the regulation of kininogen gene expression awaits further investigation.
Acknowledgement This work was supported by the National Institute of Health Grant HL 29397.
References 1 Mflller-Esterl, W., lwanaga, S. and Nakanishi, S. (1986) Trends Biochem. Sci. 11,336-339. 2 Miiller-Esterl, W. (1987) Seminars in Thrombosis and Hemostasis. Vol. 13 (No. 1), pp. 115-126, Thieme Medical Publishers, New York. 3 Johnson, A.R. (1979) In Bradyki~in, Kallidin and Kallikrein. (Erd6s, E.G., ed.), Pharmacology 25, Suppl., pp. 357 399, Springer-Verlag, Berlin. 4 Kato, H., Nagasawa, S. and lwanaga, S. (1981) Methods Enzymol. 80, 172-198. 5 Barlas, A., Okamoto, H. and Greenbaum, L.M. (1985) Biochem. Biophys. Res. Commun. 129, 280-286. 6 Chao, S., Chai, K.X., Chao, L. and Chao, J. (1990) Biochemistry 29, 323-329. 7 Chao, J., Chai, K.X., Chen, L.M., Xiong, W., Chao, S., Cheryl, W.M., Wang, L., Lu, H.S. and Chao, L. (1990) J. Biol. Chem. 265, 16394-16401. 8 Xu, J., Hsu, C.Y., Junker, H., Chao, S., Hogan, E.L. and Chao, J. (1991) J. Neurochem. 57, 975-980. 9 Von Eiff, A.W., Gogolin, E., Jacobs, V. and Neus, H. (1986) Clin. Exp. Hyper. (A) 8, 577-581. 10 Chao, J., Chao, S., Xiong, W., Chen, L.M., Swain, C. and Chao, L. (1989) Adv. Exp. Med. Biol. 247, 297-303. 11 Oh-ishi, S., Hayashi, I., Kusunoki, A., Nagashima, Y., Hayashi, M., Yamaki, K., Utsunomiya, I. and Yamasu, A. (1988) Biochem. Biophys. Res. Commun. 150, 1069-1076.
151 12 Chao, J., Swain, C., Chao, S., Xiong, W. and Chao, L. (1988) Biochim. Biophys. Acta 964, 329-339. 13 Mann, E.A. and Lingrel, J.B. (1991) Biochem. Biophys. Res. Commun. 174, 417-423. 14 Anderson, K.P. and Lingrel, J.B. (1989) Nucleic Acids Res. 17, 2835-2848. 15 Chao, S., Chao, L. and Chao, J. (1989) Biochim. Biophys. Acta 991,447-483. 16 Chao, S., Chao, J. and Chao, L. (1990) Methods in Nucleic Acids Research, Chapter 15, pp. 307-320, CRC Press, Boca Raton. 17 Chao, J., Woodley, C., Chao, L. and Margolius, H.S. (1983) J. Biol. Chem. 258, 15173-15178. 18 O'Farrell, P.H. (1975) J. Biol. Chem. 250, 4007-4021. 19 Young, R.A. and Davis, R.W. (1983) Proc. Natl. Acad. Sci. USA 80, 1194-1198. 20 Chao, S., Chao, L. and Chao, J. (1989) BioTechniques 7, 68-71.
21 Gerald, W.L., Chao, J. and Chao, J. (1986) Biochim. Biophys. Acta. 886, 1-14. 22 Madeddu, P., Glorioso, N., Maioli, M., Demontis, M.P., Varoni, M.V., Anania, V., Xiong, W., Chai, K. and Chao, J. (1991) J. Hypertens. 9 (Suppl 6), $244-$245. 23 Bouhnik, J., Savoie, F., Michaud, A., Baussant, T., Alhenc-Gelas, F. and Corvol, P. (1989) Life Sci. 44, 1859-1866. 24 Dzau, V.J. and Hermann, H.C. (1982) Life Sci. 30, 577-584. 25 Furuto-Kato, S., Matsumoto, A., Kitamura, N. and Nakamishi, S. (1985) J. Biol. Chem. 260, 12054-12059. 26 Enjyoji, K.I., Kato, H., Hayashi, I., Oh-ishi, S. and lwanaga, S. (1988) J. Biol. Chem. 263, 965-972. 27 Li, Z.H., Xu, J., Chao, J. and Hogan, E.L. (1992) FASEB J. 6, A1622, abstract. 28 Richoux, J.P., Gelly, J.L., Bouhnik, J., Baussant, T., Alhenc-Gelas, F., Grignon, G. and Corvol, P. (1991) Histochemistry 96, 229-243.
