Bradykinin antagonists

Abstract

A bradykinin antagonist of the formulaX(BKA).sub.nwherein BKA is the peptide chain of a bradykinin antagonist peptide, X is a linking group and n is a whole number greater than 1. The BKA substituents may be the same or different. Monomeric antagonists of the formula X(BKA) are also disclosed.

Description

The present invention relates to bradykinin antagonists.

BACKGROUND OF THE INVENTION

Bradykinin (BK) is a nonapeptide (SEQ. ID. NO:1) (Arg.sup.1 -Pro.sup.2 -Pro.sup.3 -Gly.sup.4 -Phe.sup.5 -Ser.sup.6 -Pro.sup.7 -Phe.sup.8 -Arg.sup.9) which, along with lysyl-bradykinin (kallidin), is released from precursor kininogens by proteases termed kallikreins. Plasma kallikrein circulates as an inactive zymogen from which active kallikrein is released by Hageman factor. Tissue kallikrein appears to be located predominantly on the outer surface of epithelial cell membranes at sites thought to be involved in transcellular electrolyte transport.

Two major kinin precursor proteins, high molecular weight and low molecular weight kininogen, are synthesized in the liver, circulate in plasma, and are found in secretions such as urine and nasal fluid. High molecular weight kininogen is cleaved by plasma kallikrein, yielding bradykinin, or by tissue kallikrein, yielding kallidin. Low molecular weight kininogen, however, is a substrate only for tissue kallikrein. In addition, some conversion of kallidin to bradykinin may occur inasmuch as the amino terminal lysine residue of kallidin is removed by plasma aminopeptidases. Plasma half-lives for kinins are approximately 15 sec., with a single passage through the pulmonary vascular bed resulting in 80-90% destruction. The principle catabolic enzyme in vascular beds is the dipeptidyl carboxypeptidase kininase II or angiotensin-converting enzyme (ACE). A slower acting enzyme, kininase I, or carboxypeptidase N, which removes the carboxyl terminal Arg, circulates in plasma in great abundance. This suggests that it may be the more important catabolic enzyme physiologically. Des-Arg.sup.9 -bradykinin as well as des-Arg.sup.10 -kallidin formed by kininase I acting on bradykinin or kallidin, respectively, are active as bradykinin B1 receptor agonists but are relatively inactive at the more abundant bradykinin B2 receptor on which both bradykinin and kallidin are potent agonists.

Bradykinin is known to be one of the most potent naturally occurring stimulators of C-fiber afferents mediating pain. It also is a potent vasodilator, edema-producing agent, and stimulator of various vascular and nonvascular smooth muscles in tissues such as uterus, gut and bronchiole. The kinin/kininogen activation pathway has also been described as playing a pivotal role in a variety of physiological and pathophysiological processes, being one of the first systems to be activated in the inflammatory response and one of the most potent stimulators of: (i) phospholipase A.sub.2 and hence the generation of prostaglandins, thromboxanes and leukotrienes; and (ii) phospholipase C and thus the release of inositol phosphates and diacylglycerol. These effects are mediated predominantly via activation of bradykinin receptors of the B2 type.

Direct application of bradykinin to denuded skin or intra-arterial or visceral injection results in the sensation of pain in animals and in man. Kinin-like materials have been isolated from inflammatory sites produced by a variety of stimuli. In addition, bradykinin receptors have been localized to nociceptive peripheral nerve pathways and bradykinin has been demonstrated to stimulate central fibers mediating pain sensation. Bradykinin has also been shown to be capable of causing hyperalgesia in animal models of pain. See, Burch et al, "Bradykinin Receptor Antagonists", J Med Chem., 30:237-269 (1990) and Clark, W. G. "Kinins and the Peripheral and Central Nervous Systems", Handbook of Experimental Pharmacology, Vol. XXV: Bradykinin, kallidin, and kallikrein. Erdo, E. G. (ed.), 311-322 (1979).

These observations have led to considerable attention being focused on the use of bradykinin antagonists as analgesics. A number of studies have demonstrated that bradykinin antagonists are capable of blocking or ameliorating both pain as well as hyperalgesia in both animals and man. See, Ammons, W. S. et al, "Effects of intracardiac bradykinin on T.sub.2 -T.sub.5 medial spinothalamic cells", The American Physiological Society, 0363-6119 (1985); Clark, W. G., "Kinins and the Peripheral and Central Nervous Systems", Handbook of Experimental Pharmacology, Vol XXV: Bradykinin, kallidin, and kallikrein. Erdo, E. G. (ed.), 311-322 (1979); Costello, A. H. et al, "Suppression of carrageenan-induced hyperalgesia, hyperthermia and edema by a bradykinin antagonist", European Journal of Pharmacology, 171:259-263 (1989); Laneuville et al, "Bradykinin analogue blocks bradykinin-induced inhibition of a spinal nociceptive reflex in the rat", European Journal of Pharmacology, 137:281-285 (1987); Steranka et al, "Antinociceptive effects of bradykinin antagonists", European Journal of Pharmacology, 16:261-262 (1987); Steranka et al, "Bradykinin as a pain mediator:Receptors are localized to sensory neurons, and antagonists have analgesic actions", Neurobiology, 85:3245-3249 (1987).

Currently accepted therapeutic approaches to analgesia have significant limitations. While mild to moderate pain can be alleviated with the use of nonsteroidal anti-inflammatory drugs and other mild analgesics, severe pain such as that accompanying surgical procedures, burns and severe trauma requires the use of narcotic analgesics. These drugs carry the limitations of abuse potential, physical and psychological dependence, altered mental status and respiratory depression which significantly limit their usefulness.

Prior efforts in the field of bradykinin antagonists indicate that such antagonists can be useful in a variety of roles. These include use in the treatment of burns, perioperative pain, migraine and other forms of pain, shock, central nervous system injury, asthma, rhinitis, premature labor, inflammatory arthritis, inflammatory bowel disease, etc.

For example, Whalley et al, in Naunyn Schmiederberg's Arch. Pharmacol., 336:652-655 (1987) have demonstrated that bradykinin antagonists are capable of blocking bradykinin induced pain in a human blister base model. This suggests that topical application of such antagonists would be capable of inhibiting pain in burned skin, e.g. in severely burned patients in whom large doses of narcotics are required over long periods of time and for the local treatment of relatively minor burns or other forms of local skin injury.

The management of perioperative pain requires the use of adequate doses of narcotic analgesics to alleviate pain while not inducing excessive respiratory depression. Post-operative narcotic induced hypoventilation predisposes patients to collapse of segments of the lungs, a common cause of post-operative fever, and frequently delays discontinuation of mechanical ventilation. The availability of a potent non-narcotic parenteral analgesic could be a significant addition to the treatment of perioperative pain. While no currently available bradykinin antagonist has the appropriate pharmacodynamic profile to be used for the management of chronic pain, frequent dosing and continuous infusions are already commonly used by anesthesiologists and surgeons in the management of perioperative pain.

Several lines of evidence suggest that the kallikrein/kinin pathway may be involved in the initiation or amplification of vascular reactivity and sterile inflammation in migraine. See Back et al, "Determination of components of the kallikrein-kinin system in the cerebrospinal fluid of patients with various diseases", Res Clin. Stud. Headaches, 3:219-226 (1972). Because of the limited success of both prophylactic and non-narcotic therapeutic regimens for migraine as well as the potential for narcotic dependence in these patients, the use of bradykinin antagonists offers a highly desirable alternative approach to the therapy of migraine.

Bradykinin is produced during tissue injury and can be found in coronary sinus blood after experimental occlusion of the coronary arteries. In addition, when directly injected into the peritoneal cavity, bradykinin produces a visceral type of pain. See, Ness et al, "Visceral pain: a review of experimental studies", Pain, 41:167-234 (1990). While multiple other mediators are clearly involved in the production of pain and hyperalgesia in settings other than those described above, it is also believed that antagonists of bradykinin have a place in the alleviation of such forms of pain as well.

Shock related to bacterial infections is a major health problem. It is estimated that 400,000 cases of bacterial sepsis occur in the United States yearly, of those 200,000 progress to shock, and 50% of these patients die. Current therapy is supportive, with some suggestion in recent studies that monoclonal antibodies to Gram-negative endotoxin may have a positive effect on disease outcome. Mortality is still high, even in the face of this specific therapy, and a significant percentage of patients with sepsis are infected with gram-positive organisms which would not be amenable to anti-endotoxin therapy.

Multiple studies have suggested a role for the kallikrein/kinin system in the production of shock associated with endotoxin. See, Aasen et al, "Plasma Kallikrein Activity and Prekallikrein Levels during Endotoxin Shock in Dogs", Eur. Surg., 10:50-62 (1977); Aasen et al "Plasma Kallikrein-Kinin System in Septicemia", Arch Surg., 118:343-346 (1983); Katori et al, "Evidence for the involvement of a plasma kallikrein/kinin system in the immediate hypotension produced by endotoxin in anaesthetized rats", Br. J. Pharmacol., 98:1383-1391 (1989); and Marceau et al, "Pharmacology of Kinins: Their Relevance to Tissue Injury and Inflammation", Gen. Pharmacol., 14: 209-229 (1982). Recent studies using newly available bradykinin antagonists have demonstrated in animal models that these compounds can profoundly affect the progress of endotoxic shock. Weipert, et al., "Attenuation of Arterial Blood Pressure Fall in Endotoxin Shock in the Rat Using the Competitive Bradykinin Antagonist Lys-Lys-[Hyp.sup.2, Thi.sup.5.8, D-Phe.sup.7 ]-BK", Brit J. Pharm., 94, 282-284, (1988). Less data is available regarding the role of bradykinin and other mediators in the production of septic shock due to Gram-positive organisms. However, it appears likely that similar mechanisms are involved. Shock secondary to trauma, while frequently due to blood loss, is also accompanied by activation of the kallikrein/kinin system. See, Haberland, "The Role of Kininogenases, Kinin Formation and Kininogenase Inhibitor in Post Traumatic shock and Related Conditions", Klinische Woochen-schrift, 56:325-331 (1978).

Numerous studies have also demonstrated significant levels of activity of the kallikrein/kinin system in the brain. Both kallikrein and bradykinin dilate cerebral vessels in animal models of CNS injury. See, Ellis et al, "Inhibition of Bradykinin- and Kallikrein-Induced Cerebral Arteriolar Dilation by Specific Bradykinin Antagonist", Stroke, 18.:792-795 (1987) and Kamitani et al, "Evidence for a Possible Role of the Brain Kallikrein-Kinin System in the Modulation of the Cerebral Circulation", Circ Res. 57:545-552 (1985). Bradykinin antagonists have also been shown to reduce cerebral edema in animals after brain trauma. Based on these data, it is believed that bradykinin antagonists should be useful in the management of stroke and head trauma.

Other studies have demonstrated that bradykinin receptors are present in the lung, that bradykinin can cause bronchoconstriction in both animals and man and that a heightened sensitivity to the bronchoconstrictive effect of bradykinin is present in asthmatics. Some studies have been able to demonstrate inhibition of both bradykinin and allergen induced bronchoconstriction in animal models using bradykinin antagonists. These studies indicate a potential role for the use of bradykinin antagonists as clinical agents in the treatment of asthma. See, Barnes, "Inflammatory Mediator Receptors and Asthma", Am. Rev. Respir. Dis., 135:S26-S31 (1987); Burch et al, "Bradykinin Receptor Antagonists", J. Med. Chem., 30:237-269 (1990); Fuller et al, "Bradykinin-induced Bronchoconstriction in Humans", Am. Rev. Respir. Dis., 135:176-180 (1987); Jin et al, "Inhibition of bradykinin-induced bronchoconstriction in the guinea-pig a synthetic B.sub.2 receptor antagonist", Br. J. Pharmacol., 9.7:598-602 (1989) and Polosa et al, "Contribution of histamine and prostanoids to bronchoconstriction provoked by inhaled bradykinin in atopic asthma", Allergy, 45:174-182 (1990). Bradykinin has also been implicated in the production of symptoms in both allergic and viral rhinitis. These studies include demonstration of both kallikrein and bradykinin in nasal lavage fluids and that levels of these substances correlate well with symptoms of rhinitis. See, Baumgarten et al, "Concentrations of Glandular Kallikrein in Human Nasal Secretions Increase During Experimentally Induced Allergic Rhinitis", J. Immunology, 137:1323-1328 (1986); Jin et al, "Inhibition of bradykinin-induced bronchoconstriction in the guinea-pig by a synthetic B.sub.2 receptor antagonist", Br. J. Pharmacol., 97:598-602 (1989) and Proud et al, "Nasal Provocation with Bradykinin induces Symptoms of Rhinitis and a Sore Throat", Am. Rev. Respir. Dis., 137:613-616 (1988)

In addition, studies have demonstrated that bradykinin itself can cause symptoms of rhinitis.

Stewart and Vavrek in "Chemistry of Peptide Bradykinin Antagonists", Bradykinin Antagonists: Basic and Chemical Research, R. M. Burch (Ed.), pages 51-96 (1991) discuss peptide bradykinin antagonists and their possible use against effects of bradykinin. A great deal of research effort has been expended towards developing such antagonists with improved properties. However, notwithstanding extensive efforts to find such improved bradykinin antagonists, there still remains a need for more effective bradykinin antagonists.

The two major problems with presently available bradykinin antagonists are their low levels of potency and their extremely short durations of activity. Accordingly, important objectives of the present invention include the provision of novel bradykinin antagonist peptides which are characterized by increased potency and duration of action. Other objects will also be hereinafter evident.

SUMMARY OF THE INVENTION

The invention is based, in one important embodiment, on the finding that compounds comprising two or more bradykinin antagonist (BKA) peptides which are chemically linked together to form a dimer or higher oligomer ("mer") demonstrate greater potency and/or duration of action than the parent bradykinin antagonist peptide itself.

Broadly speaking, the compounds of the invention can be illustrated by the formula:

X(BKA).sub.n (I) where BKA is the nucleus of any bradykinin antagonist peptide, X is a linking group and n is a whole number greater than 1. Preferably n is 2, thereby providing a dimer. However, the compound may include more than two BKA substituents, i.e. the compound may be a trimer or even a higher "mer", up to the limit permitted by the nature of the linking group.

The BKA substituents may be the same throughout so as to provide, in a sense, a homogenous compound. On the other hand, it may be preferred in some situations that the compound include different BKA substituents, i.e. the compound may be heterogeneous with respect to the BKA substituents. Both homogeneous and heterogenous embodiments are contemplated by the present invention as discussed more fully below.

The concept of providing dimers of pharmaceutically active materials to improve such characteristics as metabolic stability, selectivity and receptor binding has previously been described for other systems. However, the literature has not disclosed bradykinin antagonist dimers or higher "mers" as contemplated herein. Furthermore, prior dimerization efforts have, in large measure, favored dimerizing from one or both ends of the present compounds whereas it has been found, as part of the present invention, that dimerizing various bradykinin antagonists from an external position using either the free alpha amino or carboxyl groups at the termini of these ligands does not appear to enhance, and in most cases reduces, the activity of these agents. This is in marked contrast to the prior art involving other dimerized systems wherein dimerizing from one or both of the termini appears to be the method of choice.

A further embodiment of the invention provides a novel group of "monomeric" compounds of the formula:

X(BKA) (II) i.e. compounds of formula (I) where n is 1 and the modifier X functions to improve the antagonist properties of the parent bradykinin antagonist peptide represented by the BKA substituent. This improvement may be shown by, for example, increased potency and/or duration of action.

Thus, in brief, the invention is concerned with bradykinin antagonist peptides of formula (I), which may be homogeneous or heterogeneous dimers or higher "mers", or of the formula (II) where a bradykinin antagonist is modified to include the group X to provide products of improved antagonist properties.

DETAILED DESCRIPTION OF THE INVENTION

Numerous bradykinin antagonist peptides are known in the art and any of these may be used for present purposes to provide the BK substituents. One of the most potent bradykinin antagonists In vitro is the peptide having the formula:

dARG.sup.0 -Arg.sup.1 -Pro.sup.2 -Hyp.sup.3 -Gly.sup.4 -Phe.sup.5 -Ser.sup.6 -(d)Phe.sup.7 -Leu.sup.8 -Arg.sup.9 See Regoli et al, Trends in Pharmacological Science, 11:156-161 (1990). This peptide is referred to herein for convenience as CP-0088 and, for purposes of the present invention, it may be used as such for linking with the same or different peptide having bradykinin antagonist activity to provide antagonists as represented by formula (I).

While CP-0088 is used herein to illustrate various aspects of the invention, it needs to be emphasized that the invention contemplates the use of any available or known bradykinin antagonist peptide for present purposes. For example, a wide variety of modifications have been proposed for bradykinin antagonist peptides in the recent patent literature and these can be used to provide the BKA component of the present products. See, for example, EP-A-0334244 (Procter and Gamble) which discloses nona- and larger bradykinin antagonist peptides in which certain amino acid residues are modified. EP-A-0370453 (Hoechst) and WO 89/01780 and WO 89/01781 (Stewart et al) also describe bradykinin antagonist peptides. None of these patent publications appears to show homogeneous or heterogenous dimers or higher "mers", or the linker modified monomers as contemplated herein. However, as noted, the peptides of these publications can be used in the practice of the present invention. Accordingly, while CP-0088 is used for purposes of illustration, the invention is not to be viewed as limited thereto.

According to one embodiment of the invention, the linker X is advantageously joined to the parent peptide antagonist through a cysteine (Cys) sulfhydryl group in the peptide chain. This may require initially modifying the starting peptide so that it includes a Cys group in the appropriate position in the peptide chain. The invention is illustrated using various Cys derivatives of CP-0088 where Cys replaces Ser or some other amino acid in the CP-0088 peptide chain. It has been found that Cys conveniently provides for linking to an appropriate linker X through the sulfhydryl group. According to this aspect of the invention, it is preferred to incorporate the cysteine residue at the 6-position of CP-0088. This position has been previously regarded as relatively unimportant. See Stewart, JM and Vavrek, RJ, (1991), "Chemistry of peptide bradykinin antagonists", in Bradykinin Antagonists: Basic and Clinical Research, R. M. Burch (ed.), Marcel Dekkar Inc. NY, pp. 51-96. However, an important aspect of the present invention is based on the finding that, in contrast to this generally accepted view of the 6 position, this specific locus is important for both dimerization and monomer modification in providing optimum properties for the resulting product. It will, however, be appreciated that the invention is not limited to modifications in the 6-position.

As shown hereinafter, any one of a wide variety of linkers X may be used for present purposes. This group functions to chemically join together two or more peptides which are themselves BK antagonists. However, the linker can itself also contribute to improve the overall antagonist properties as shown hereinafter. Preferably the linker is a flexible, linear group which is readily reactive at two or more sites with the BK antagonist peptide. Preferred linkers include bismaleimido alkanes (BMA) such as bismaleimido hexane (BMH). Reaction of such a linker can occur between the maleimido groups and the sulfhydryl moiety of a cysteine residue of the peptide chain to provide an S-succinimido derivative. For example, a 1,6-bis-S-succinimidohexane (BSH) dimer can be prepared by reacting 2 equivalents of Cys containing peptide antagonist and 1 equivalent of bis-maleimido hexane (BMH). Other suitable linking groups X are described later.

One of the preferred dimer antagonists of formula (I) according to the invention is identified herein, for ease of reference, as CP-0127. This compound has been found to have significantly greater potency and half-life than any other presently known bradykinin antagonist, including CP-0088 which, as noted earlier, is considered to be one of the most potent bradykinin antagonists known in the art. The improved antagonist properties of CP-0127 are quite unexpected in that they represent more than what might be thought of as the normal additive effect of combining two or more peptide antagonists.

Compound CP-0127, which demonstrates all the characteristics required for pharmaceutical application as a bradykinin antagonist, e.g. the compound is suitable for parenteral use in the treatment of acute indications such as septic shock and perioperative pain, may be structurally illustrated as follows: ##STR1##

As will be evident, CP-0127 is a homogeneous dimer in the sense that the two peptide chains attached to the linker are the same. The compound comprises the reaction product of two equivalents of the Cys.sup.6 derivative of the bradykinin antagonist CP-0088, i.e. CP-0088 where Ser in the 6-position has been replaced by cysteine (Cys), Joined together by reaction between the sulfhydryl moiety of Cys and bis-maleimido hexane (BMH). The resultant product is a 1,6-bis-S-succinimido-hexane (BSH), Cys.sup.6 dimer of CP-0088.

The cysteine-derivative (Cys.sup.6) of CP-0088 can be prepared by conventional means. This Cys.sup.6 derivative, which is identified herein as CP-0126 for ease of reference, may then be reacted with the desired bis-maleimide. In the case of CP-0127, where the bis-maleimide is bismaleimide hexane (BMH), the dimer (CP-0127) is obtained according to the following reaction scheme: ##STR2##

The foregoing illustrates only one possible way of preparing the products of formula (I). In an alternative method, the linker may be joined to a fragment of the peptide, another peptide fragment joined to the other end of the linker and the balance of the respective peptides (BKA) completed by adding whatever other peptide fragments might be required. Those in the art will recognize that whatever process steps are used, precautions need to be taken to protect various groups, notably amino and carboxy termini of the peptides or fragments thereof, from undesired reaction.

While BMH provides a preferred linking group X, other bis-maleimide alkanes (BMA), for example, any bis-maleimides where the alkane contains from 1-12 carbons or more may be used. The alkane group may also be substituted or interrupted by, for example, carbonyl and/or amino groups. While the length and nature of the alkyl chain in the BMA may be varied as noted, it appears that as the length increases from 1-12 carbons, the pA.sub.2 does not vary substantially. However, the inhibition of recovery increases. Additionally, as the alkyl chain length increases beyond 8 carbon atoms, partial agonism may be noted. Generally speaking, when using a BMA linker with CP-0126 as the BK component, the highest PA.sub.2 values are obtained with alkyl chain lengths in the range of 6 to 9 carbon atoms and it appears that, of such linkers, the best results are obtained using BMH (bis-maleimido hexane) to provide the indicated BSH linkage. However, it will be recognized that effective results can be obtained using other types of linkers. Thus, it is not essential that the group X include the succinimido substituent provided the modification used is effective to link together the BKA peptide substituents without undesirably affecting the antagonist properties.

Representative alternatives to BMH as the linking group are structurally shown below with the resulting dimer designation as used later herein, shown to the left of the linker structure, it being understood that, in each instance, the BKA component of the dimer is the same as in CP-0126.

__________________________________________________________________________DIMER LINKER__________________________________________________________________________CP-0162 ##STR3##CP-0163 ##STR4##CP-0164 ##STR5##CP-0165 ##STR6##CP-0166 ##STR7##CP-0170 ##STR8##CP-172 ##STR9##CP-0176 ##STR10##CP-0177 ##STR11##__________________________________________________________________________

It will be recognized from the foregoing that the compounds of formulas (I) and (II) according to the invention may be prepared using conventionally available procedures. The antagonist peptide portions are readily available or may be prepared in fragments using solid or solution phase synthesizing techniques. Known procedures may be used to provide a Cys-residue in the position desired in the peptide chain and conventional chemistry is involved in adding the linker modification and any subsequent-"mer" formation.

BRIEF DESCRIPTION OF THE DRAWING

The invention is illustrated in the following examples and the related FIGS. 1A, 1B, 2A, 2B, 2C, 3, 4, 5A, 5B and 6 which graphically illustrate test results obtained with CP-0127.

EXAMPLE 1

Preparation of CP-0127

The peptide CP-0126 for use in dimerization with bismaleimidohexane (BMH) to produce CP-0127 was generated by solid phase peptide synthesis using a standard stepwise elongation of the peptide chain common to the peptide field (J. M. Stewart and J. D. Young, Solid Phase Peptide Synthesis, Pierce Chemical Co., 1984). Briefly stated, solid phase peptide synthesis begins with N.sup..varies. -deprotection of the amino acid attached to the synthesis resin. This step is followed by neutralization and washing of the deprotected amino acid-containing resin which prepares it to react with the next amino acid, itself activated by dicyclohexylcarbodiimide and 1-hydroxybenzotriazole to facilitate the formation of the peptide bond --CO--NH--). A subsequent washing of the now (di)peptide-containing resin is then followed by the same series of events which are continued until the desired peptide has been produced. The finished peptide is then cleaved off the resin under conditions which simultaneously remove all of the individual amino acid side-chain protecting groups.

All reagents used in the synthesis described herein were obtained from standard commercial sources.

Solid phase peptide synthesis was carried out by manual methods using bubbling nitrogen gas as the agitation source for mixing. N.sup..varies. -tert-butyloxycarbonyl (N-t-BOC) group was used as the temporary protecting group in all the peptide couplings. More specifically, the following resin and protected amino acids were used in a sequential manner: N-t-BOC-L-Arg (Tos) -PAM resin, N-t-BOC-L-Leu, N-t-BOC-D-Phe, N-t-BOC-L-Cys(4-Meb), N-t-BOC-L-Phe, N-T-BOC-Gly, N-t-BOC-L-Hyp (Bzl), N-t-BOC-L-Pro, N-t-BOC-L-Arg(Tos), N-t-BOC-D-Arg(Tos).

Finished peptidyl-resin was dried in vacuo and then placed in the reactor of a Peninsula Laboratories Type I HF apparatus. CP-0126 was cleaved from the resin using standard HF procedures without any carbocation scavengers. After HF removal in vacuo, the resin was washed with diethyl ether and the peptide was extracted from the resin with deionized water.

Two equivalents of the resulting peptide CP-0126 was allowed to react with 1 equivalent (0.5 molar ratio) of bismaleimidohexane (Pierce) in 0.05M ammonium bicarbonate buffer (pH 8.4) and stirred for 1 hour. The reaction mixture was then concentrated in vacuo and the resulting concentrate redissolved in 0.1M ammonium bicarbonate (pH 5). The solution was loaded in toto onto a column of sulfopropyl (SP) Sephadex (Sigma) and the column eluted with a pH 5-9 gradient over one hour period. The fractions containing the peptide dimers were then combined, concentrated in vacuo, redissolved in deionized water and lyophilized.

The final peptide dimer CP-0127 is assessed for purity by reversed phase HPLC, amino acid analysis, peptide sequence analysis, and Electro-Spray mass spectrometry.

EXAMPLE 2

Compound CP-0127 may also be prepared in the following manner wherein several separate peptide fragments are prepared, the linking group (BSH) is added to join two fragments and the resulting fragment dimer is then completed by adding the other required peptide fragments.

More specifically, the process involves preparation of the following separate peptide fragments:

(1) Phe-Cys-dPhe (2) Leu-Arg (3) dArg-Arg-Pro (4) HPro-Gly Fragments (1) and (2) are joined together to form: (5) NH.sub.2 -Phe-Cys-dPhe-Leu-Arg-OFm where OFm is fluorenylmethyl. Two molecules of the intermediate (5) may then be joined together through their respective Cys-groups using BMH to give: ##STR12##

Fragments (3 ) and (4 ) are joined together to form

(7) (BOC)-dArg-Arg-Pro-HPro-Gly-OH where BOC is t-butyloxycarbonyl and thereafter two molecules of (7) are Joined to (6) to provide: ##STR13## followed by removal of the end protective groups to give the dimer CP-0127.

This procedure is more specifically outlined hereinafter:

Fragment (4) is prepared as follows:

N-T-BOC-Leu (16.68 g, 67 mmole) and Arg-OFm.2 HCl (29.68 g, 70 mmole) are dissolved in 140 mL of N,N-dimethylformamide and then 16 mL of diphenylphosphoryl azide (20.43 g, 74 mmole) and 25 mL of diisopropylethylamine (18.55 g, 143 mmole) are added. The reaction proceeds at room temperature until complete by thin layer chromatography (n-butanol:acetic:water 4:1:1) on silica gel.

When complete, the reaction mixture is concentrated in vacuo until about 10% of the original volume is left. The resulting slurry is dissolved in ethyl acetate (150 mL) and washed in a 250 mL separatory funnel with 10% sodium bicarbonate (3.times.100 mL), water (2.times.100 mL), 0.1M hydrochloric acid (3.times.100 mL) and water (2.times.100 mL). The final organic layer is then dried over sodium sulfate and concentrated to dryness in vacuo.

At this point, the solid or oil is dissolved in a small amount of ethyl acetate and crystallized using hexane. The solid is filtered by vacuum filtration and dried in vacuo.

The N-t-BOC group is removed using 4N HCl in dioxane (100 mL) for 1 hour at room temperature. The solvent is then removed in vacuo and the resulting solid (or oil) dissolved in a small amount of methanol (20 mL). Crystallization is induced by the addition of ethyl ether and the resulting solid is isolated by vacuum filtration, washed with ethyl ether (2.times.50 mL) and dried in vacuo.

Fragment 2 is prepared using the same procedure. Fragments 1 and 3 are done in an analogous fashion except that an additional amino acid is added. At this point, the N-t-BOC group is left on fragments 1 and 3 and the ester groups are removed as described below.

The esters (methyl or ethyl) are removed by dissolving the peptide in 100 mL methanol and adding 1.5 equivalents of sodium hydroxide; the reaction mixture then being allowed to stir for 1 hour at 0.degree. C. The reaction mixture is neutralized with 1N HCl and concentrated in vacuo to a volume of ca. 30 mL. Water (100 mL) is added and then 1N HCl until the solution pH reaches ca. 1. The peptide is extracted from the aqueous solution with ethyl acetate (2.times.100 mL) and the combined ethyl acetate is dried over sodium sulfate, filtered and concentrated in vacuo. The resulting oil is dissolved in a small volume of ethyl acetate and crystallized using hexane.

Fragments (1) and (2) are coupled together and worked up in the same manner as described above to give the N-terminal pentapeptide. Fragments (3) and (4) are coupled together and worked up in the same manner as described above to give the C-terminal pentapeptide.

The C-terminal pentapeptide (10 g) is deprotected using 100 mL of 4N HCl in dioxane for 1 hour. The dioxane is removed in vacuo to yield an oil which is dissolved in 20 mL of methanol and treated with excess ethyl ether to produce a white powder. The powder thus produced is dissolved in 90 mL of water and then 10 mL of methanol is added followed by 30 mL of acetic acid saturated with iodine (I.sub.2). Stirring is then continued for 1 hour. The reaction mixture is extracted with two equal volumes of chloroform and the water layer lyophilized. The resulting off-white powder is dissolved in 20 mL of water and treated with activated Reductacryl resin (CALBIOCHEM) for 20 minutes. The Reductacryl is removed by filtration and the resin then washed with 20 mL of 0.1M ammonium bicarbonate buffer (pH 8). The combined filtrates containing the peptide are treated with 1.6 g of bis-maleimidohexane (Pierce) in 3 mL of N,N-dimethylformamide and stirred for 2 hours. The reaction mixture is then concentrated in vacuo and the resulting concentrate redissolved in 50 mL of 0.1M ammonium bicarbonate buffer (pH 5). The solution is loaded in toto onto a 50 g column of sulfopropyl (SP) Sephadex (Sigma) and the column eluted with a pH 5-9 gradient over a one hour period at 3 mL per minute. The fractions containing the peptide are combined, concentrated in vacuo, redissolved in 100 mL of water and lyophilized.

At this point, 2 equivalents of the N-terminal peptide is coupled to 1 equivalent of the C-terminal dimer peptide using the same protocol as described above for fragment 4.

Deprotection of the final peptide is achieved by dissolving 5 g of the peptide in 75 mL of N,N-dimethylformamide containing 30% piperidine and then stirring for 20 minutes. The solvents are removed in vacuo and the resulting oil dissolved in 100 mL of water. The pH is then lowered to ca. 1 with 1N HCl and the product extracted into ethyl acetate (3.times.100 mL). The combined ethyl acetate is removed in vacuo to yield an oil which is then dissolved in 4N HCl in dioxane and stirred for 1 hour. The dioxane is removed in vacuo and the resulting oil dissolved in 30 mL of methanol. Excess ethyl ether is finally added to yield the product as a white powder.

Final purification of the peptide dimer is carried out by dissolving 5 g in 0.1N ammonium bicarbonate buffer (pH 5), loading the solution onto a 50 g column of sulfopropyl (SP) Sephadex (Sigma) and running a gradient from pH 5 to pH 9 over a 1 hour period while collecting 10 mL fractions. Each fraction is assessed by analytical HPLC and pure fractions are combined and lyophilized 3 times with water.

The final product is assessed for purity by reverse phase HPLC, amino acid analysis, peptide sequence analysis, and Electro-Spray mass spectrometry.

In carrying out the above-described synthesis, it is to be noted that the amino group of each amino acid is protected with N-tert-Butoxycarbonyl (N-t-BOC) or the like as needed. The carboxyl group of the amino acids is also protected, in this case with an ester function (e.g., ethyl, methyl or 9-fluorenemethyl ester). The quanidino side chain of arginine (Arg) is protected as the hydrochloride (HCl) salt and the sulfur atom of cysteine (Cys) is protected using the acetamidomethyl group (Acm).

EXAMPLE 3

Compound CP-0127, prepared as described in Example 1, was compared against the Cys.sup.6 -modified peptide CP-0126 and the parent peptide CP-0088 for biological activity using tests which are well-recognized in the art. Thus in vitro studies were performed on standard assay systems, namely, guinea-pig ileum (GPI), rat uterus (RU) and rabbit jugular vein (RJV). It is well known that each of these tissues possesses a bradykinin receptor which is of the B2 class but each is of a different subtype. The potency of each compound was assessed as a pA.sub.2 value according to the method of Arunlakshana & Schild, Br. J. Pharmacol. Chemother., 14:48-58 (1959). The PA.sub.2 can be defined as the negative logarithm of the molar concentration of the antagonist in the presence of which the potency of the agonist is reduced two-fold, i.e. twice the amount of agonist is required to produce a given response in the presence of, than in the absence of, the antagonist. A high potency of antagonism is indicated by a high PA2 value, i.e., a low concentration of antagonist is effective.

The percentage recovery (% Rec) was also determined for compounds with respect to each assay. Percentage recovery is a measure of the duration of action, the lower the percentage recovery value, the greater the duration of action of the compound.

The pA.sub.2 and percent recovery (% Rec) values obtained for CP-0088, CP-0126 and CP-0127 on the GPI, RU and RJV assays are shown in Table A:

TABLE A______________________________________COM- GPI RU RJVPOUND pA.sub.2 % REC pA.sub.2 % REC pA.sub.2 % REC______________________________________CP-0088 6.6 100 7.4 100 8.5 90CP-0126 6.4 100 7.1 100 8.8 60CP-0127 7.7 100 8.8 50 10.5 10______________________________________

It can be seen that in each tissue, the parent peptide CP-0088 is approximately equipotent with the Cys.sup.6 modified peptide (CP-0126). In contrast, the dimer CP-0127 is significantly more potent than either CP-0126 or CP-0088. The increases in potency for CP-0127 are significantly higher than CP-0126 and CP-0088 in each tissue with the greatest increase being shown on the RJV assay. The increases in potency for CP-0127 vs. CP-0126 on the GPI and RU were less marked but still substantial.

It will also be noted that CP-0127 showed a significantly lower % Rec on the RU and RJV assays than CP-0088 and CP-0126. The % Rec for CP-0126 on the RJV assay was also markedly better than that obtained with CP-0088.

With respect to duration of activity, it is noted that after exposure to the highest concentration of antagonist, each tissue was washed extensively and then assessed for recovery from the antagonist effect. FIGS. 1A and 1B represent the concentration effect curves to bradykinin (BK) on rat uterus in vitro using (1A) CP-0126 and (1B) CP-0127. The differences in the BK recovery curves should be noted. As can be seen from FIG. 1A, the concentration effect curve to BK after exposure to CP-0126 almost returns to that seen in the control. In contrast, with CP-0127, there was still significant antagonism of bradykinin even after repeated washings (FIG. 1B). This is an indication that CP-0127 has a longer duration of activity than CP-0126.

EXAMPLE 4

Three in vivo models were used for studies to compare the effects of CP-0126 and CP-0127. These were (a) rabbit skin vascular permeability; (b) bradykinin-induced hypotension in the rat; and (c) lipopolysaccharide-induced hypotension in the anesthetized rat (CP-0127 only) as follows:

(a) Rabbit Skin Vascular Permeability

In these experiments, two protocols were used. The first protocol used the classical assay in which bradykinin was injected intradermally either alone or in combination with the compound under investigation following intravenous administration of Evans Blue dye. Thirty minutes after intradermal injection of bradykinin with and without the compound, the animal was killed, its skin removed and the area of "blueing" (i.e., vascular permeability or edema) measured. The results using this protocol are shown in FIG. 2A. It can be seen that CP-0126 is more potent than CP-0127 using this protocol.

The second protocol involved pre-injecting the test compound either 15 or 30 minutes before injecting bradykinin in the same site. The results are shown in FIGS. 2B and 2C, respectively. It is clear that CP-0127 has totally inhibited responses to bradykinin, whereas CP-0126 is significantly less effective.

(b) Rat Blood Pressure

A comparison of the antagonist potency of CP-0126 and CP-0127 was made in anesthetized rats. Following calculation of the hypotension dose-response curve to bradykinin in these animals, the minimum dose producing maximal reduction in blood pressure was repeated at approximately three minute intervals before, during and after an intravenous infusion of either CP-0126 or CP-0127 at 25 ug kg.sup.-1 min.sup.-1. FIG. 3 shows the results obtained in terms of the percentage change in mean arterial blood pressure in the rat to bradykinin (4.times.10.sup.-9 moles) before, during and after an intravenous infusion of either CP-0126 or CP-0127 at 25 .mu.g kg.sup.-1 min.sup.-1. It can be seen from FIG. 3 that CP-0127 almost totally inhibited responses to this large dose of bradykinin whereas CP-0126 was almost totally ineffective. After stopping the infusion of CP-0127, it can be seen that the responses to bradykinin returned slowly being fully recovered after approximately 20 minutes.

(c) Lipopolysaccharide (LPS) Induced Hypotension in the Anesthetized Rat

LPS-induced hypotension in the rat is a standard animal model of septic shock. In this model, anesthetized rats are injected intravenously with LPS from Eschericha coli at 15 mg kg.sup.-1. This produces an immediate and profound drop (50-60% change) in mean arterial blood pressure (FIG. 4). When CP-0127 was infused at 50 .mu.g/kg/min intravenously 60 minutes before, during and after the injection of LPS there was a significant attenuation of the immediate drop in blood pressure which then rapidly recovered back to normal. Normotension was then sustained for the duration of the experiment (approximately 3-4 hours). This was in contrast to the control group in which profound hypotension was sustained. In fact, 3 animals died within 20 minutes of LPS injection. See FIG. 4 which illustrates the percentage change in mean arterial blood pressure in the anaesthetized rat to an intravenous bolus injection of lipopolysaccharide (LPS) 15 mg kg.sup.-1.

The data referred to above shows that the dimer bradykinin antagonist (CP-0127) is significantly more potent than the Cys.sup.6 monomer antagonist (CP-0126) in both in vitro and in vivo models of bradykinin activity using standard assay systems to evaluate the efficacy of the antagonists against a specific stimulus (i.e., bradykinin) and against a non-specific stimulus (LPS) in a recognized animal model of septic shock.

The in vitro data demonstrate exceptional potency of CP-0127 on each assay tissue, the most marked activity being seen on the rabbit jugular vein. As noted, compound CP-0088 was chosen from the literature for test purposes because it was the most potent (assessed by standard pA.sub.2 determination) bradykinin antagonist in classical in vitro assay systems reported. Using the same criteria (pA.sub.2 determination), the potency of CP-0127 appears to surpass most if not all of the previous bradykinin antagonists that have been reported, particularly on rabbit jugular vein.

The highest pA.sub.2 value for CP-0127 was obtained on a vascular preparation, the rabbit jugular vein and it appears that this class of bradykinin antagonist should be suitable for treating vascular processes where bradykinin is involved. Thus, when viewed in the light of bradykinin's role in inflammation, the effects of CP-0127 on permeability/edema and vasodilation are of particular relevance. The data from the skin vascular permeability studies are of interest since CP-0126 was found to be more effective than CP-0127 when both antagonists were co-injected with bradykinin. However, when CP-0127 and CP-0126 were pre-injected either 15 to 30 minutes before challenging with bradykinin, CP-0127 totally inhibited responses to bradykinin whereas CP-0126 was much less effective than when co-injected with bradykinin. The reason for this difference is not understood but it may be that the dimer requires more time for equilibration with the receptor and once bound is more difficult to displace compared with the monomer. In addition, it appears that CP-0127 may be more metabolically stable than CP-0126.

As noted earlier, there is strong evidence that bradykinin is significantly involved in septic shock due to gram negative bacterial infection. In this connection, it is to be noted that the present dimer (CP-0127) totally reversed the response to LPS at a dose that effectively blocked hypotensive responses to maximal doses of bradykinin in the anesthetized rat. Although CP-0126 has not been assessed in the LPS septic shock model, it is believed that, at the same dose, it would be much less effective than CP-0127 considering the lack of effect of this compound (CP-0126) against high doses of bradykinin.

Thus, in brief, from the in vitro and in vivo data provided herein, there are significant quantitative differences between the monomeric (CP-0126) and dimeric (CP-0127) bradykinin antagonists with respect to absolute potency, resistance to wash-off and duration of action.

In considering the bioavailability of CP-0127, it has been demonstrated, using a single dose (10 mg/kg) given subcutaneously (S.C.) that the blood pressure and pain responses to BK can be totally blocked for at least 2 hours (the duration of the experiments). Using anaesthetized rats, dose-response curves to BK were constructed and the ED.sub.50 identified (see FIG. 5a). The ED.sub.50 was then injected before and at 15 minute intervals after a single S.C. injection of CP-0127. From FIG. 5b, it can be seen that the response to BK is totally blocked for the duration of the experiment (2 hours).

In a separate set of experiments, pain in response to the intra-arterial injection of BK was assessed in conscious rats which had previously (48 hours before) been implanted with an indwelling cannula into the carotid artery. Bradykinin (20 nmoles/kg) was injected intra-arterially which produced a characteristic contralateral paw elevation, and head and body rotation, followed by a period of quiescence. The response was quantified as a ranking scale of 0-5 reflecting the degree of behavioral response observed.

The data shown in FIG. 6 demonstrate that a single S.C. injection of CP-0127 at 10 mg/kg totally abolished the response to BK with inhibition lasting for the duration of the experiment (approximately 2 hours). Controls showed a stable response pattern during the entire experimental period.

EXAMPLE 5

This example illustrates the effect of introducing the Cys-moiety and consequently the linker into different positions along the chain of CP-0088. The various compounds identified in Table B below were prepared following the procedure of Example 1 with the Cys moiety introduced into the parent CP-0088 at the position indicated followed by reaction between the sulfhydryl of the Cys moiety and BMH to provide dimers as indicated. The Cys-modified monomers and the resultant dimers were tested for activity on guinea pig ileum in vitro on the standard tissue bath/strain gauge system as before. The results obtained in terms of pA.sub.2, together with those for CP-0088, CP-0126 and CP-0127, are given in Table B. The references to "L" and "D" are used to show that optical isomers were involved as indicated.

TABLE B__________________________________________________________________________COMPOUND DESCRIPTION pA.sub.2 GUINEA PIGNUMBER COMPOSITION REFERENCE ILEUM__________________________________________________________________________CP-0088 ##STR14## Monomer .apprxeq.6.6CP-0126 L-CYS.sup.6 Monomer 6.3CP-0127 L-CYS.sup.6 Dimer .apprxeq.7.7CP-0140 L-CYS.sup.0 Monomer .apprxeq.6.5CP-0152 L-CYS.sup.0 Dimer .apprxeq.6.5CP-0149 D-CYS.sup.0 Monomer .apprxeq.6.3CP-0161 D-CYS.sup.0 Dimer .apprxeq.6.3CP-0141 L-CYS.sup.1 Monomer .apprxeq.6.0CP-0153 L-CYS.sup.1 Dimer .apprxeq.7.7CP-0142 L-CYS.sup.2 Monomer .apprxeq.5.7CP-0154 L-CYS.sup.2 Dimer .apprxeq.7.6CP-0143 L-CYS.sup.3 Monomer .apprxeq.6.5CP-0155 L-CYS.sup.3 Dimer .apprxeq.7.1CP-0136 L-CYS.sup.4 Monomer InactiveCP-0137 L-CYS.sup.4 Dimer InactiveCP-0173 D-CYS.sup.4 Monomer InactiveCP-0203 D-CYS.sup.4 Dimer InactiveCP-0144 L-CYS.sup.3 Monomer .apprxeq.6.3CP-0156 L-CYS.sup.3 Dimer .apprxeq.7.9CP-0145 L-CYS.sup.7 Monomer InactiveCP-0157 L-CYS.sup.7 Dimer InactiveCP-0146 D-CYS.sup.7 Monomer InactiveCP-0158 D-CYS.sup.7 Dimer InactiveCP-0147 L-CYS.sup.8 Monomer InactiveCP-0159 L-CYS.sup.8 Dimer InactiveCP-0148 L-CYS.sup.9 Monomer InactiveCP-0160 L-CYS.sup.9 Dimer Inactive__________________________________________________________________________

The above results indicate that the rank order of potency of the dimers shown in Table B, listed by linking position, is:

5.gtoreq.6.gtoreq.1.gtoreq.2>3>0>>>4, 7, 8, 9 with the 4, 7, 8 and 9 position dimers inactive. Table B further shows that all active dimers with Cys in the 1, 2, 3, 5 and 6 positions were more potent than the corresponding monomer. This is true for both the L- and D- isomeric forms where determined.

EXAMPLE 6

The pA.sub.2 and % recovery values on the rat uterus model of bradykinin activity were determined for the various monomers and dimers referred to in Example 5 and compared with the corresponding values obtained for reference compound CP-0088, CP-0127 and the parent Cys-.sup.6 monomer (CP-0126). The results are shown below in Table C.

TABLE C__________________________________________________________________________ pA.sub.2 pA.sub.2 % REC % RECMONOMER DIMER MONOMER DIMER MONOMER DIMER__________________________________________________________________________CP-0088 -- 7.4 -- 100 --CP-0126 (L-CYS.sup.6) CP-0127 7.1 8.8 100 50CP-0144 (L-CYS.sup.5) CP-0156 7.2 8.1 95 75CP-0136 (L-CYS.sup.4) CP-0137 Inactive Inactive -- --CP-0173 (D-CYS.sup.4) CP-0203 Inactive Inactive -- --CP-0143 (L-CYS.sup.3) CP-0155 P.A.* 6.2 -- 100CP-0142 (L-CYS.sup.2) CP-0154 6.4 7.2 80 90CP-0148 (L-CYS.sup.1) CP-0153 6.7 7.7 100 100CP-0140 (L-CYS.sup.0) CP-0152 7.0 7.9 50 50CP-0149 (D-CYS.sup.0) CP-0161 7.1 6.5 75 100CP-0145 (L-CYS.sup.7) CP-0157 Inactive Inactive -- --CP-0146 (D-CYS.sup.7) CP-0158 Inactive Inactive -- --CP-0147 (L-CYS.sup.8) CP-0159 Inactive Inactive -- --CP-0148 (L-CYS.sup.9) CP-0160 Inactive Inactive -- --__________________________________________________________________________ P.A. = Partial Agonist

While the relative potencies shown in Table C for the rat uterus model were somewhat different from those obtained in Example 5 on the guinea pig ileum, it is noteworthy that the compounds based on the Cys modification in the 4, 7, 8 and 9 positions of CP-0088 were again inactive. Additionally, the rat uterus results indicate a preference for the 6-position modification although it will be recognized that other positions may well be preferred with other linkages and modifications depending, for example, on the receptor types or subtypes involved.

EXAMPLE 7

While the data generated from the homodimer CP-0127 indicate that this compound is a significant improvement over other compounds in the literature, it is important to realize that the effects of dimerization are not a unique aspect of the specific linker chosen, BMH. Listed below (Table D) are a series of dimers that were derived from the base compound CP-0126 and a variety of bis-maleimido alkane linkers. These data clearly indicate that various bis-maleimido alkanes can be used to form effective bradykinin antagonist dimers. The selection of a preferred compound will depend on the intended application based upon the specific pharmacodynamic qualities desired.

TABLE D______________________________________EFFECT OF LINKER LENGTH ##STR15###of CARBON COMPOUND pA.sub.2 % RECATOMS NO. (UTERUS) (40 MIN)______________________________________n = 2 CP-0162 8.3 90n = 3 CP-0172 8.6 90n = 4 CP-0209 8.2 50n = 6 CP-0127 8.8 50n = 8 CP-0211 8.4 25n = 9 CP-0229 9.0 0-10 n = 10 CP-0230 8.4 0 n = 12 CP-0166 8.2 0*______________________________________ *IRREVERSIBLE AT 80 MINUTES

Example 8

The following compounds according to the invention were also prepared as modifications of CP-0126: ##STR16## represents CP-0126. These compounds were prepared by reacting CP-0126, i.e. ##STR17## with the indicated linker compound essentially in the manner of Example 1.

As will be appreciated, the listed compounds include modified peptide monomers (CP-0174, -0175, -0139); dimers (CP-0138, -0163 and -0170) and the trimers (CP-0164 and -0176).

Table E compares the pA, values obtained on testing the above compounds for activity in the guinea pig ileum (GPI) and rat uterus tests with CP-0126 and CP-0127:

TABLE E______________________________________ GPI RAT UTERUSCOMPOUND pA.sub.2 pA.sub.2 % RECOVERED______________________________________CP-0126 .apprxeq.6.3 7.1 100CP-0127 .apprxeq.7.7 8.8 50CP-0174 .apprxeq.7.2 8.4 50CP-0175 .apprxeq.7.2 8.5 30CP-0139 .apprxeq.7.1 7.9 100CP-0138 .apprxeq.6.2 7.4 100CP-0163 .apprxeq.8.2 7.6 100CP-0164 .apprxeq.7.5 7.6 50CP-0170 .apprxeq.6.3 7.0 100CP-0176 .apprxeq.7.0 8.6 50______________________________________ The data given in Tables D and E shows the following: 1. Reacting the cysteine in the "6" position with free maleimide thereby converting it into an S-succinimide cysteine improves potency (compare CP-0126 with CP-0139, -0174, -0175). However, it has also been found that monomers as a class behave differently to dimers as a class with respect to their resistance to "wash-off" regardless of their structure. Monomers can be removed and the dose/response profile of the tissue returns to pre-treatment status with one exchange of buffer while dimers require multiple washings to reverse their activity and return the system to baseline. This aspect of monomer vs. dimer activity in in vitro models of bradykinin antagonism is indicative of prolonged duration of action for the dimers in vivo. 2. Improved potency with dimerization does not appear to be dramatically influenced by the linker as long as the linker is linear and flexible. However, a more rigid or constrained linker (CP-0170) does not improve potency with respect to the corresponding monomer (CP-0126). To underscore this point, the disulfide dimer (CP-0138), which is probably the most constrained and rigid of the dimers tested, is also relatively less active. In fact, it is no different in inhibitory activity than the CP-0126 monomer in this system. 3. Trimerization (CP-0164 and CP-0176) does not appear to improve potency beyond that which was obtained by dimerization (CP-0127, -0163 and -0166).

EXAMPLE 9

To show the generalizability of this concept with respect to the ligand, another representative bradykinin antagonist (CP-0181) was modified by introducing Cys in the 6-position and dimerizing with BMH to form a BSH dimer (CP-0215):

CP-0181

DArg.sup.0 -Arg.sup.1 -Pro.sup.2 -Hyp.sup.3 -Gly.sup.4 -Phe.sup.5 -Ser.sup.6 -DPhe.sup.7 -Phe.sup.8 -Arg.sup.9

The pA.sub.2 values on the rat uterus assay and % recovery were determined for this BSH dimer and compared with CP-88 and CP-0127. The results are shown below in Table F and it will be noted that improved pA.sub.2 results and % recovery were realized for both dimers when compared with their corresponding reference monomers:

TABLE F__________________________________________________________________________ RAT UTERUS %COMPOUND pA.sub.2 RECOVERY__________________________________________________________________________CP-0088 7.4 100dARG.sup.0 --ARG.sup.1 --PRO.sup.2 --HYP.sup.3 --GLY.sup.4 --PHE.sup.5--SER.sup.6 --dPHE.sup.7 --LEU--ARGCP-0127 8.8 50BSH(CYS.sup.6) DimerCP-0181 7.0 100dARG.sup.0 --ARG.sup.1 --PRO.sup.2 --HYP.sup.3 --GLY.sup.4 --PHE.sup.5--SER.sup.6 --dPHE.sup.7 --PHE.sup.8 --ARG.sup.9CP-0215 8.2 40BSH(CYS.sup.6) Dimer__________________________________________________________________________

The foregoing results (Examples 3-9) show that the dimers are more effective than either the parent or modified antagonist and that trimers, while better than the monomers, do not show improvement over dimers. It is also believed that the best results are obtained using reactive bismaleimide rings which are themselves preferably joined by simple straight chain hydrocarbon, as the peptide linking group. The resulting introduction of two chiral centers to give optically different compounds does not seem to impact undesirably on the antagonist activity.

It is noted that the dimerization of peptide agonists and antagonists in order to increase potency and/or duration of action is recognized in the pharmaceutical art. See Caporale et al, Proc. 1Oth American Peptide Symp., Pierce Chemical Co., Rockford, Ill. 449-451 (1988) and Rosenblatt et al, European Patent Application No. EP 293130A2. Thus, dimerization of peptide agonists has been disclosed for enkephalins/endorphins (Shimohigashi, Y., et. al., BBRC, 146, 1109-1115, 1987); substance P (Higuchi, Y., et. al., E.J.P., 160, 413-416, 1989); bradykinin (Vavrek, R. and Stewart, J. Proc. 8th Amer. Pept. Symp., 381-384, 1983); neurokinin A & B, (Kodama, H., et. al., E.J.P., 151, 317-320, 1988); insulin (Roth, R. A., et. al., FEBS, 170, 360-364, 1984) and atrial natriuretic peptide (Chino, N., et. al., BBRC, 141, 665-672, 1986). Dimerization of antagonists has been shown for parathyroid hormone (Caproale, L. H., et. al., Proc. 10th Amer. Pept. Symp., 449-451, 1987). However, there does not appear to have been any prior work in this respect with bradykinin antagonists as disclosed herein.

It will be appreciated from the foregoing that various modifications may be made without departing from the spirit and scope of the invention. For example, while the bradykinin antagonist designated as CP-0088 has been modified by inserting Cys at various positions in the preparation of the illustrated dimers, any parent antagonist may be used to form useful dimers or the like without any modification or with a different type of modification. The linker used to form the compounds of the invention may also be varied, as noted earlier. As a further illustration of this aspect of the invention, it is noted that the linker may comprise dextran which may be modified randomly along the length of the dextran molecule so as to ultimately possess reactive maleimide residues as shown below: ##STR18##

The resulting maleimide-substituted dextran can then be reacted with a free sulfhydryl containing peptide in a fashion analogous to that previously described at varying molar ratios to yield peptide-dextran conjugates of various substitution densities and overall size.

As noted easrlier, an important aspect of the invention is the provision of hetero-dimers or higher hetero- "mers" where different bradykinin antagonists (BDA) are used. This "hetero" embodiment makes it possible to design dimers, for example, which are effective against two or more different bradykinin receptors, e.g. B.sub.1 and B.sub.2 receptors. These two receptors appear to be the most abundant and apparently have the greatest distribution in various tissues. In general, B.sub.1 rec ptor ligands (both agonists and antagonists) are formed when the C-terminal arginine is removed from the corresponding B.sub.2 receptor ligand by either circulating or tissue associated carboxypeptidases.

Examples of this conversion from B.sub.2 to B.sub.1, selectivity are seen with the conversion of bradykinin to des-Arg.sup.9 -bradykinin and kallidin to des-Arg.sup.10 -kallidin. This suggests that it would be preferable for a general BK antagonist to contain both B.sub.1 and B.sub.2 receptor antagonist activity. On the other hand, there may be situations where the antagonist should be selectively effective against B.sub.1 or B.sub.2 receptors alone.

The distribution and relative importance of these two receptor systems varies from tissue to tissue and species to species. In addition, various pathophysiologic processes can alter the distribution and importance of these two receptor systems over time within the same animal. In other words, the receptor distribution in the naive animal may be considerably different from the receptor distribution in that same animal after a pathophysiologic process (sepsis for example), has gone on for some time (see Marceua, F. et al, in "Pharmacology of Kinins: Their Relevance to Tissue Injury and Inflammation", General Pharmacology, 14, pp 209-229). As a result, it is difficult to know what type of antagonist is most appropriate for any given application and relying on the in vivo conversion of a B.sub.2 antagonist to a B.sub.1 antagonist may not be feasible in certain circumstances. The provision of heterodimers according to the invention offers the possibility of dealing with such situations where, for example, both B.sub.1 and B.sub.2 receptors are involved.

Representative heterodimers according to the invention, based on the indicated B, and B, antagonist monomers are listed below: ##STR19## By using these heterodimers, it is possible for each "side" of the dimer to act on its respective receptor, i.e., the B.sub.1 antagonist side will block the B.sub.1, receptors and the B2 antagonist will block the B.sub.2 receptors.

Example 10

To test the above hypothesis, several heterodimers were assessed for their antagonist activity against BK in rat uterus (B.sub.2 - receptor) or against des-Arg.sup.9 -BK (B.sub.1 -receptor selective agonist) on rabbit aorta (B.sub.1 -receptor) in vitro. Potency in the rat uterus assay was assessed as described above and PA.sub.2 values determined. These are shown in Table G. Potency in rabbit aorta was assessed in the following way: Concentration effect curves were constructed to des-Arg.sup.9 -BK 1 hour and 3 hours after setting up the preparations. At time 5 hours, a single concentration of (10.sup.-7 M) of des-Arg.sup.9 -BK was added to produce a sustained contradiction. Each antagonist was then added, in a cumulative fashion, on top of the contraction and the negative log of the molar concentration of the antagonist causing a 50% reversal (IC.sub.50) of the concentration measured. These data are also shown in Table G.

It is clear from these data that each part of the heterodimer can act independently from the other on its respective receptor. The potential for such compounds in the treatment of disorders in which both BK.sub.1 - and BK.sub.2 -receptors are believed to play a role is clear.

TABLE G______________________________________COMPOUND RAT UTERUS RABBIT AORTANUMBER pA.sub.2 IC.sub.50______________________________________CP-0272 8.5 5.4 .+-. 0.2CP-0273 8.1 5.5 .+-. 0.1CP-0290 NOT TESTEDCP-0291 8.2 6.2 .+-. 0.1______________________________________

As evident, the invention contemplates the use of a wide variety of linkers X, with or without Cys-modification of the parent bradykinin antagonist peptide (BKA). A further alternative to linkers of the type referred to above, which makes it possible to avoid the cysteine modification as described for the BSH type of compounds described earlier, is shown below: ##STR20## n=2-12 m=1-12

*Chirality can be equivalent to either "D" or "L" amino acids R/R'" can be any combination of protecting groups such as N-t-Boc, Fmoc, NPYS R'/R" can be any combination of protecting groups such as a methyl ester, ethyl ester or benzyl ester

The general syntheses of the linkers of type I and II alternative to the cysteine modifications may be prepared in the following manner. Several separate fragments consisting of organic linkers and properly protected amino acids are prepared and joined together. The resulting fragments are then cyclized intramolecularly to the general hydantoin structures using the condition of base catalyzed cyclization.

More specifically, the synthesis of type I linker involves the preparation of the following separate fragments: ##STR21## Fragments (1), (2) and (3) are joined together to form: ##STR22##

Fragment (4) is then reacted with ethyl chloroformate and followed by base catalyzed cyclization to yield the desired type I alternative linker.

The synthesis of type II linker involves the preparation of the following separate fragments: ##STR23##

Fragments (5), (6) and (7) are joined to form: ##STR24##

Fragment (8) is then reacted with ethyl chloroformate and followed by base catalyzed cyclization to yield the desired type II alternative linker.

Once synthesized, the desired linkers from either Group I or II can be condensed with the appropriate peptide fragments to form the desired hetero- or homo-dimer. Illustrated below is a schematic representation of such a synthesis: ##STR25##

The above synthesis illustrates the production of a heterodimer wherein each of the reactive groups of the linker are capable of being differentially deprotected so that all four fragments of the heterodimer can be added separately.

It will be appreciated that the type of linker shown above permits "offsetting" the ligands, i.e. differentially coupling them to the linker. For example, it is possible to make a ligand that is linked from the "6" position on one side and the "5" position on the other if this is deemed desirable for pharmacological reasons.

Other examples of alternative fragments that can be used for hetero- or homo-dimer synthesis include are listed below.

FRAGMENTS A/B:

B.sub.1 SELECTIVE - a) NH.sub.2 -dPhe-Leu

b) NH.sub.2 -dThi-Leu c) NH.sub.2 -Pro-Leu d) NH.sub.2 -d(Tic) -Oic e) NH.sub.2 -Pro-Ile

V.sub.2 SELECTIVE - f) NH.sub.2 -dPhe-Leu-Arg

g) NH.sub.2 -dThi-Leu-Arg h) NH.sub.2 -d(Tic)-Oic-Arg i) NH.sub.2 -dPhe-Thi-Arg j) NH.sub.2 -dPhe-Phe-Arg k) NH.sub.2 -dPhe-Ile-Arg FRAGMENTS C/D: .alpha.) d(Arg)-Arg-Pro-Hyp-Gly-Phe-CO.sub.2 H .beta.) d(Arg)-Arg-Pro-Hyp-Gly-Thi-CO.sub.2 H .tau.) Lys-Arg-Pro-Pro-Gly-Phe-CO.sub.2 H .delta.) Arg-Pro-Pro-Gly-Phe-CO.sub.2 H Thi=Thienylalanine

It will also be appreciated that while reference has been made to the B.sub.1 /B.sub.2 antagonist as a heterodimer, it is just as possible for one to incorporate this linker into a heterodimer that blocks two completely different receptors such as B.sub.2, bradykinin receptors on one side and neurokinin-1 (Substance P) receptors on the other. This type of heterodimer could be particularly useful in controlling pain without the use of opiates.

While data showing the effectiveness of the linker-modified peptide monomer per se has been presented above, the effect of simply modifying the peptide monomer with the linker is further illustrated in Table H which sets out pA.sub.2 (rat uterus) and % recovery values obtained with the listed modified peptides.

TABLE H__________________________________________________________________________EFFECT OF MONOMER MODIFICATIONSCOMPOUND RAT UTERUS %NUMBER COMPOUND pA.sub.2 RECOVERY__________________________________________________________________________CP-0126 ##STR26## 7.1 100CP-0139 ##STR27## 7.9 100CP-0264 ##STR28## 7.5 100CP-0257 ##STR29## 8.4 80CP-0174/0175 ##STR30## 8.8 50CP-0256 ##STR31## 9.2 0CP-0266 ##STR32## 8.2 70CP-0174 ##STR33## 8.4 50CP-0175 ##STR34## 8.5 30__________________________________________________________________________

The data in Table H indicates that, in general, the PA.sub.2 of a given monomer increases with increasing alkyl chain length and correlates well with the corresponding dimers. The data also show that S-(N-alkylsuccinimide)-cysteine modified monomers are more potent than the S-(N-hexylacetamide)-cysteine modified monomer. Additionally, the results indicate that racemic mixtures and optical isomers are both functionally useful with the possibility that racemic mixtures may be preferred in offering a heterogeneity of receptor action.

While Table H indicates that linker-modified monomers are themselves effective and suitable for use according to the invention, it will be usually preferred to use dimers or higher "mers" because the latter, general speaking, show an even greater degree of activity over the unmodified peptide monomer, than the corresponding linker-modified monomer. In any case, i.e. with the linker-modified monomer and the corresponding dimer, the best overall results were obtained with the modification in the Cys.sup.6 position.

The BK antagonists of the invention may be used in the form of conventional pharmaceutical compositions comprising the antagonist and a pharmaceutically acceptable carrier. Such compositions may be adapted for topical, oral, aerosolized, intramuscular, subcutaneous, or intravenous administration. The amount of antagonist present in such compositions will range from, for example, about 0.001 to 90.0% by weight depending on the application and mode of administration although more or less of the active component may be used. Conventional dosages will vary considerably on the basis of the intended application and mode of administration, e.g. 0.1 to 100 micrograms per kg body weight per minute are contemplated for use in the context of the septic shock.

__________________________________________________________________________SEQUENCE LISTING(1) GENERAL INFORMATION:(iii) NUMBER OF SEQUENCES: 1(2) INFORMATION FOR SEQ ID NO:1:(i) SEQUENCE CHARACTERISTICS:(A) LENGTH: 9 amino acids(B) TYPE: amino acid(D) TOPOLOGY: linear(xi) SEQUENCE DESCRIPTION: SEQ ID NO:1:ArgProProGlyPheSerProPheAr g15

The scope of the invention is defined in the following claims wherein:

Claims

1. A bradykinin antagonist of the formula

X(BKA).sub.2

where BKA is a peptide chain selected from the group consisting of dArg.sup.0 -Arg.sup.1 -Pro.sup.2 -Hyp.sup.3 -Gly.sup.4 -Phe.sup.5 -Cys.sup.6 -dPhe.sup.7 -Leu.sup.8 -Arg.sup.9 and dArg.sup.0 -Arg.sup.1 -Pro.sup.2 -Hyp.sup.3 -Gly.sup.4 -Phe.sup.5 -Cys.sup.6 -dPhe.sup.7 -Phe.sup.8 -Arg.sup.9 and X is a linking group joined to each peptide chain through the Cys-position, selected from the group consisting of ##STR35## wherein R.sup.1 is (CH.sub.2).sub.n with n being a whole number of from 1-12.

2. A bradykinin antagonist according to claim 1, wherein n is 6.

3. A bradykinin antagonist according to claim 2, wherein each BKA is dArg.sup.0 -Arg.sup.1 -Pro.sup.2 -Hyp.sup.3 -Gly.sup.4 -Phe.sup.5 -Cys.sup.6 -dPhe.sup.7 -Leu.sup.8 -Arg.sup.9.