Activation of Protein Formation and Cell Division by Bradykinin and Des-Argg-bradykinin*

We employed des-Argo-bradykinin to investigate the relation between bradykinin-induced prostaglandin (PG) synthesis and bradykinin-induced protein accumulation. In this feedback control system, bradykinin- induced PG synthesis limits bradykinin-induced protein production. At low concentration (5 X 10-’M), des- Args-bradykinin was significantly less active than bradykinin in stimulating the formation of prostaglan- dins by human fetal lung fibroblasts in culture. At high concentration (5 X M), bradykinin induced a 24% increase in protein formation, while des-Arg*-brady-kinin induced a 61% increase in collagen formation and an 80% increase in total protein accumulation. In the presence of indomethacin, bradykinin-induced protein formation was increased further, whereas des-Arg”-bradykinin-induced protein formation was un- changed. The bradykinin derivative increased the production of types I and 111 procollagens without affect-ing the distribution of procollagen types. The incor- poration of [’Hlthymidine into DNA in lung fibroblast cultures was increased 3-fold by des-Arg”-bradykinin alone or by bradykinin in combination with indomethacin. Des-Arg”-[Leu*]bradykinin inhibited the des- Ar8-bradykinin-induced protein formation and cell division. These data indicate that both bradykinin and des-Args-bradykinin stimulate protein formation and cell division; des-Arge-bradykinin alone stimulates protein formation and cell division

' The abbreviations used are: PGs, prostaglandins; DAL, des-Ar$-[Leu"]bradykinin. at higher concentrations. In addition, when bradykinin-induced PG synthesis was inhibited with indomethacin, the increase in protein formation induced by bradykinin was considerably higher. These data point to a feedback system in which PG synthesis limits the bradykinin-induced increase in protein production.
The effect of bradykinin in vivo may result in part from the generation of active metabolites. For example, cleavage of the C-terminal arginine residue results in the formation of the biologically active metabolite des-A&-bradykinin (4, 5 ) . Although this metabolite is less active than bradykinin in stimulating PG synthesis by endothelial cells in culture (6), certain vascular tissues appear to have surface receptors which are particularly activated by exposure to des-Argg-bradykinin (4, 5 ) . Our data indicate that des-Ar$-bradykinin is less active than bradykinin in stimulating PG synthesis by lung fibroblasts in culture; however, this bradykinin derivative dramatically stimulates protein formation and DNA synthesis.

MATERIALS AND METHODS
Human embryonic lung fibroblasts (IMR-90, Institute for Medical Research, Camden, NJ) were grown in Dulbecco's modified Eagle's medium with 0.37 g of sodium bicarbonate/100 ml, 10% fetal bovine serum, 100 units of penicillin/ml, 10 pg of streptomycin/ml, and 0.1 mM nonessential amino acids. The cells were maintained in a humidified 5% Con, 95% air incubator at 37 "C. The cell cultures were grown to confluence in Falcon P-35 dishes (prostaglandin determinations) or in Falcon P-60 dishes (protein determinations). After confluence was reached, the cells were placed into the quiescent state by reducing the serum content of the medium to 0.4%. These cells remain viable and can be restimulated to divide by increasing the serum content of the culture medium (7).
Prostaglandin Determination-Prostaglandins are synthesized within cells and rapidly extruded into the extracellular space; they are not stored (8). Prostaglandin production was determined by direct radioimmunoassay of the culture medium; preliminary extraction was not necessary (9). Production of prostacyclin and thromboxane Az was determined by assay for their degradative products, 6-keto-PGF1, and thromboxane BI, respectively. Antibodies to PGEt,B-keto-PGF,,, and thromboxane BP were developed and used in standard radioimmunoassay procedures (2,lO). The cross-reactivities of these antisera for the non-targeted PG did not exceed 4%.
Collugen and Total Protein Production-Collagen and total protein production was evaluated by determining the amount of nondialyzable ["Clhydroxyproline (collagen) and ["Clproline (total protein) present in the culture medium and cell layer. The culture medium was replaced with serum-free, proline-free medium containing ascorbate (50 pg/ml). The cells were incubated for 1 h at 37 "C. The medium was replaced with serum-free medium containing 1 pCi/ml of [14C]proline (specific activity, 260 mCi/mmol; Schwarz/Mann). After 24 h, the media and the cell layers were lyophilized. The samples were placed in 6 N HCl and hydrolyzed for 24 h in uacm at 107 "C.
The hydrolysates were analyzed for ['4C]hydroxyproline and ["C] proline using a Beckman automatic amino acid analyzer fitted with a stream-splitting device and a fraction collector. The appropriate fractions were placed in scintillation fluid, and radioactivity was determined. To determine the nanomoles of hydroxyproline in the 9263

Bradykinin-and
Des-Ar$-bradykinin-induced Protein Formation hydrolysate, the amino acids were reacted with ninhydrin and assayed colorimetrically. Polyacrylamide Gel Electrophoresis and Autoradiography-Con-Ar2-bradykinin ( 5 X M) were labeled with ["C]proline for 24 h. fluent quiescent fibroblast cultures in the presence or absence of des-After the pulse-labeling period, a solution of protease inhibitors was immediately added to the pooled medium fractions to yield a final concentration of M p-hydroxymercuribenzoate, M phenylmethanesulfonyl fluoride, and 2 mM EDTA. The medium was then dialyzed against Hz0 at 4 "C to remove free [14C]proline and then lyophilized. Polyacrylamide gel electrophoresis of medium fractions in the presence or absence of dithiothreitol was performed on a 5% gel (11). Autoradiography was performed according to the method of Bonner and Laskey (12).
Release of Arachidonic Acid and Its Metabolites-Confluent quiescent fibroblasts were maintained in medium containing 0.4% serum for 2 days. These cultures were then labeled with [3H]arachidonic acid (specific activity, 867.4 Ci/mmol; New England Nuclear). After 12 h, the medium was removed and the cultures were rinsed twice with serum-free medium. The cultures were then incubated for 30 min with serum-free medium containing 2 mg/ml of fatty acid-free bovine serum albumin with no other additions as controls. Additional cultures received either 5 X M bradykinin or 5 X M des-Ar2-bradykinin. Radioactivity was determined in aliquots of the initial labeling medium, the medium and washes that were collected after 12 h, and the final incubation medium. The difference between the initial counts/min of [3H]arachidonic acid present in the culture medium and that remaining after 12 h was used to calculate percentage of [3H]arachidonate incorporated by the cells. The percentage released was determined by dividing the counts/min released into the culture medium by counts/min of [3H]arachidonate incorporated into the cell layer.
DNA Synthesis-DNA synthesis was evaluated by measuring [3H] thymidine incorporation into a trichloroacetic acid-insoluble cell fraction. The culture medium was replaced with serum-free medium containing 0.1 pCi/ml of [~nethyl-~HIthymidine (specific activity, 51 Ci/mmol; Amersham Corp.). After 24 h, the medium was removed and the cells were washed twice with saline. The cell layer was then exposed to 1 ml of 1% Triton X-100 and harvested into glass tubes. The suspension was centrifuged following the addition of 20% trichloroacetic acid. The trichloroacetic acid-precipitated material was solubilized by heating to 70 "C for 20 min, and radioactivity was determined by scintillation counting.
Statistics-Statistical evaluation of the data was carried out using a Student's t test for means of unequal size (13). Probability values <0.05 were considered significant.

RESULTS
The effects of des-Arg-bradykinin and bradykinin on PG production by lung fibroblasts are shown in Table  I. Des-Argg-bradykinin is significantly less active than bradykinin in stimulating PG synthesis by lung fibroblasts in culture. At a concentration of 5 X M, des-A$-bradykinin stimulated a small increase in PGE2, PGF2,, and thromboxane A2 synthesis. Des-Argg-bradykinin did not stimulate PG synthesis when used a t a concentration of 5 X M or lower. In contrast, bradykinin at a concentration of 5 X lo-' M stimulated a marked increase in the synthesis of PGE2, prostacyclin, PGF2,, and thromboxane A2. This concentration of bradykinin was previously shown to induce maximal PG synthesis in these cells (2). The increase in PG synthesis induced by bradykinin or des-Argg-bradykinin was inhibited by the addition of indomethacin (5 X 1O"j M).
Bradykinin and des-Ar2-bradykinin were evaluated for their effects on the release of fatty acids in cultures prelabeled with [3H]arachidonic acid. The release of radioactivity from such cultures is shown in Table 11. Bradykinin stimulated a 2-fold increase in release of radioactivity into the culture medium at 5 X 1O"j M, whereas des-Ar2-bradykinin at 5 X 1O"j M did not induce a detectable increase in release.
We compared the effect of bradykinin and des-Argg-bradykinin on the formation of nondialyzable [14C]hydroxypro-line (collagen) and nondialyzable ['4C]proline (total protein) by lung fibroblasts in culture (Fig. 1). In these studies, bradykinin (5 X M) stimulated a 24% increase in total protein production but induced no change in collagen production. The addition of bradykinin together with indomethacin resulted in a 51% increase in collagen formation and a 98% increase in total protein formation, confirming our previous results (2). In contrast, des-Arg-bradykinin alone at 5 X M stimulated a 61% increase in collagen and an 80% increase in total protein accumulation. The presence of indomethacin did not affect the level of des-A$-bradykinin-induced collagen or total protein formation.
To ensure that des-Argg-bradykinin-induced increase in [ 14C] hydroxyproline reflected an increase in collagen formation and not a change in amino acid pool size, we measured the nanomoles of nondialyzable hydroxyproline in the culture medium. The total hydroxyproline level increased from 8.34 The amount and distribution of procollagens present in the medium were determined by autoradiograms of polyacrylamide gels (Fig. 2). Densitometric scans of the autoradiograms indicate that procollagen type I11 accounts for 8% of the total procollagen synthesized by the fibroblasts in both untreated and des-Ar2-bradykinin-treated cultures. The densitometric scans also indicated that des-Argg-bradykinin induced a 2fold increase in the total amount of procollagens synthesized.
Bradykinin and des-Ar2-bradykinin were examined for their effects on cell division. This was determined by evaluating the incorporation of [3H]thymidine into DNA in fibroblast cultures (Fig. 3). In these cell cultures, bradykinin alone did not stimulate an increase in [3H]thymidine incorporation into DNA. However, when prostaglandin synthesis was inhibited by indomethacin, bradykinin induced a %fold increase in [3H]thymidine incorporation into DNA. In contrast, 5 X 1O"j M des-A$-bradykinin alone stimulated a 3-fold increase in [3H]thymidine incorporation. The increase in cell division induced by des-Argg-bradykinin was not affected by the presence of indomethacin. In separate experiments, at 1 x M, des-Argg-bradykinin induced maximal [3H]thymidine incorporation into DNA.
Des-Argg-[Leu'lbradykinin competitively inhibits the effects of des-Argg-bradykinin on smooth muscle preparations (4). The effect of DAL on bradykinin-induced PG synthesis is shown in Table 111. DAL alone, at 5 x M, did not stimulate a detectable increase in PG synthesis, and the presence of DAL did not inhibit bradykinin-induced PG synthesis.
We assessed the effect of DAL on des-Ar2-bradykinininduced cell division (Fig. 4).  (Table IV). However, the addition of DAL resulted in a concentration-dependent decrease in des-A&-bradykinin-induced collagen and total protein formation. The presence of DAL at 5 X M inhibited des-Argg-bradykinininduced stimulation of total protein formation. This inhibi-

5). DAL was less effective in inhibiting the increase in collagen
and total protein formation induced by bradykinin or the combination of bradykinin and indomethacin (Fig. 5).

DISCUSSION
In this report, we examine the hypothesis that bradykininrelated peptides stimulate protein accumulation through a cellular mechanism different from that involving PG production. It is known that bradykinin activates phospholipase Az and perhaps other lipases which in turn liberate arachidonic acid from lipid stores (1, 14, 15). The arachidonate is then converted to PGs by a series of enzymes. Our data (Tables I and 11) and that of others (6) indicate that des-Arg-bradykinin is significantly less active than bradykinin in stimulating the release of fatty acids and the synthesis of prostaglandins. The C-terminal arginine residue appears to be required for optimal bradykinin-induced PG synthesis.
Protein formation induced by bradykinin may well be mediated through a mechanism that does not involve the release of fatty acids. In order to evaluate this possibility, we examined the ability of des-Arg-bradykinin to activate collagen and total protein formation by fibroblasts. Although des-A&-bradykinin (5 X M) only minimally stimulated PG synthesis, this bradykinin metabolite induced an 80% increase in protein accumulation and a 61% increase in collagen formation; the peptide equally increased the production of both type I and type 111 procollagens.
We previously demonstrated (2) that bradykinin-induced protein formation was markedly increased by the presence of indomethacin (inhibition of PG synthesis). These data indicate that bradykinin-induced PG synthesis provides a feedback control system for protein accumulation, particularly collagen. In the present studles, we found that des-A$bradykinin alone increased collagen and total protein accumulation to levels similar to that produced by the combination of bradykinin and indomethacin. Since des-Arg-bradykinin only minimally stimulates PG synthesis, protein formation induced by des-Arg-bradykinin was not limited by PG feedback control. Indeed, the presence of indomethacin did not affect the level of des-Arg-bradykinin-induced collagen and total protein formation (Fig. 1).
The increase in total protein production induced by both bradykinin and des-Arg-bradykinin was localized primarily in the cell layer fraction. This suggested that the increase in total production might be associated with an increase in cell division. We examined the effect of bradykinin and des-Argbradykinin on cell division by evaluating the incorporation of [3H]thymidine into DNA (Fig. 3). Des-A$-bradykinin a t 5 X loT6 M stimulated a %fold increase in the incorporation of [3H]thymidine into DNA. In the presence of indomethacin, bradykinin itself stimulated a 3-fold increase in DNA synthesis. Other workers have reported that bradykinin has mitogenic properties when used at high concentrations. Whitfield and associates (16) indicated that bradykinin at 1 PM stimulated DNA synthesis in rat thymocyte cultures. Furthermore, Lys-bradykinin has been reported to induce DNA synthesis perhaps by stimulating Na' influx (17).
PG synthesis induced by bradykinin appeared to inhibit the bradykinin-induced [3H]thymidine incorporation in DNA. Therefore, bradykinin-induced PG synthesis limits both protein production and cell division. Indeed, the addition of PGE, to fibroblast cultures is known to inhibit both protein formation and cell growth (18, 19). In contrast, des-Arg-bradykinin at 5 X M only minimally stimulates PG synthesis, and the presence of indomethacin did not affect des-Ar2-  hradykinin-induced cell division and protein formation. Des-Arg"-bradykinin is a octapeptide with a phenylalanine residue in the C-terminal position. Substitution of the terminal phenylalanine with a leucine residue forms the hradvkinin analogue des-Arg"-[ Leu"]hradykinin.
This molecule did not stimulate total protein formation or cell division. However, the presence of DAL inhibited the des-Arg"-hradykinininduced increase in protein formation and cell division (Table  IV and Fig. 4). The dose-concentration relation for DAL induced inhibition suggest that des-A&-hradykinin and DAL competitively hind to a membrane receptor. Taken together, these results suggest the presence of two distinct systems activated by hradykinin.
One system acti- he mediated through a different receptor than that involved with PG synthesis. This is suggested by the ability of des-Arg"-hradykinin to stimulate protein formation and cell division without stimulating the release of arachidonic acid. Effect of des-Argg-[Leu8]bradykinin on bradykinin-induced stimulation of prostaglandin production All cells were maintained in Dulbecco's modified Eagle's medium containing 0.4% bovine fetal serum for 3 days.
The medium was removed and replaced with 2 ml of serum-free medium without additions as controls. Additional Moreover, DAL inhibited the increase in cell division and protein production by des-Arg-bradykinin without inhibiting the bradykinin-induced PG synthesis. However, DAL was less effective in inhibiting the increase in collagen and protein formation induced by bradykinin or the combination of bradykinin and indomethacin. Thus, it is not yet certain whether bradykinin and des-A&-bradykinin activate a single receptor for cell division and protein accumulation.
Studies by others also indicate that bradykinin has multiple effects on the same tissue (4, 20). These actions may result from the activation of separate membrane receptors. For example, bradykinin and its metabolite des-Arg-bradykinin have a different pattern of activity on smooth muscle prepa-  rations. Bradykinin stimulates the contraction of guinea pig ileum, whereas des-Argg-bradykinin is more effective in stimulating the contraction of rabbit aorta. These data suggest that bradykinin has at least two membrane receptors. Receptor binding studies employing [3H]bradykinin indicate a dissociation constant of 5 X M for membranes derived from the guinea pig ileum (21). The binding constant for des-Ar$" bradykinin on isolated membranes is unknown. Studies of intact rabbit mesenteric vein indicate a binding constant of approximately 1-1.6 X M (22,23). The relation between these smooth muscle receptors and bradykinin-induced prostaglandin synthesis and protein production in cell culture is not known.
Similar to the intact bradykinin molecule, platelet-derived growth factor and certain lymphokines increase collagen production, cell division, and PG synthesis by fibroblasts in culture (24)(25)(26). These effectors are larger molecules than bradykinin. It is unclear whether these effectors also function through two or more activation systems.
Bradykinin is degraded to an inactive metabolite by the angiotensin-converting enzyme, which removes the C-terminal dipeptide Phe8-Ar$ (4). This enzyme is present on the luminal surface of endothelial cells (27). A decrease in angiotensin-converting enzyme activity appears to occur in several animal models of fibrogenic lung injury, perhaps as a result of endothelial cell damage and tissue hypoxia (28)(29)(30)(31). The Cultures were confluent and maintained in medium containing 0.4% bovine fetal serum for 2 days. The medium was replaced with fresh serum-free medium containing ["Clproline. Control cultures (C) received no additions. All effectors were added at 5 X 10" M. A, as indicated, cultures received des-Ar2-bradykinin (DA) alone or together with DAL and indomethacin (INDO). B, additional cultures received bradykinin (BK) alone or together with DAL and indomethacin. After 24 h, medium and cell layer were harvested, and nondialyzable ["Clproline was determined. The data are expressed as mean f 1 S. E. ( n = 4). The asterisk denotes significant difference from bradykinin or des-Ar2-bradykinin alone (p < 0.05). The double asterisk denotes significant difference from bradykinin plus indomethacin or des-Ar2-bradykinin plus indomethacin (p < 0.05). decrease in angiotensin-converting enzyme activity may favor the conversion of bradykinin t o des-Arg-bradykinin by plasma carboxypeptidases (5). Des-Arg-bradykinin may have physiologic importance in the complex inflammatory process that results in cell proliferation and excess connective tissue accumulation in the lung. Our studies suggest that the presence of des-Arg-bradykinin could contribute to the fibrotic process by stimulating cell division and collagen production without inducing PG synthesis.