CA2241389A1 - Methods of modulating muscle contraction - Google Patents

Abstract

Methods of modulating calcium independent smooth muscle contraction, and of modulating cardiac muscle contraction, by the administration of a PAK modulating agent are described. A method of treating a subject having a state characterized by smooth muscle contraction, (e.g., hypertension, asthma, irritable bowel syndrome, incontinence and menstrual cramps) or cardiac contractile dysfunction (e.g., heart failure and myocardial stunning) is also described. The method involves the administration, to a subject, of a therapeutically effective amount of a PAK modulating agent, e.g., a PAK inhibitor or stimulator. Assays, e.g., screening tests, to identify PAK modulating agents or agents that modulate muscle contraction are also provided.

Description

METHODS OF MODULATING MUSCLE CONTRACTION

Re/a~/ Applica~ions This application claims the benefit of priority of U.S. Provisional Patent Applications No. 60/050,478, filed June 23, 1997, and No. 60/ , filed June 16, 1998.
The contents of both these provisional applications are incorporated herein by reference.

Field of the Invention This invention relates to methods for mod~ ting muscle contraction, and particularly, to methods for mocl~ ting smooth muscle contraction in the absence of calcium, and for increasing the calcium sensilivily of both smooth muscle and cardiac muscle.

Background of the Invention p21 -activated kinase (PAK) proteins are serine/threonine protein kinases homologous to the yeast Ste20 kinase. Several distinct members of the PAK family have been identified in m~mm~ n cells, including PAK1, PAK2, PAK3, and PAK65. It has been shown that PAK proteins can be activated by the small Rho-family GTP-binding proteins Cdc42 and Racl, which are known to regulate assembly of the actin cytoskeletal structure (Zhang et al. 1995. J. Biol. Chem. 270:23934-36).
Smooth muscle, in contrast to striated muscle (cardiac and skeletal muscle), is capable of m~ L~ g force at low levels of intracellular calcium. One member of the small Rho-family of GTP-binding proteins, RhoA, has been shown recently to induce smooth muscle contraction independently of calcium. RhoA activates a serine/threonine kinase named ROK (RhoA associated kinase or pl60ROK) which can cause smooth muscle calcium-independent contraction. Cardiac muscle, on the other hand, requires increased levels of intracellular calcium for contraction.
Smooth muscle is found in blood vessels, the airways of the lungs, the gastro-intestin~l tract, the uterus and the urinary tract. The uncontrolled contraction of smooth muscle in such tissues is involved in states such as hypertension (a known risk factor for heart disease), asthma, irritable bowel syndrome, incontinence or menstrual cramps.
Hypertension or high blood ples~ulc, is the most common disease affecting the heart and blood vessels. Statistics indicate that hypertension afflicts one out of every five American adults. Asthma is a chronic disease characterized by airway hyperactivity, it occurs in 5-8%
ofthe U.S. population, and is an extraordinarily common cause of pulmonary hlll)~;"~ent Irritable bowel syndrome is a common syndrome characterized by frequently alt.qrn~ting con~ ation and ~ rrhç~ usually with abdominal pain. Often stress induced, it is also caused by such physical factors as spicy foods, lack of dietary fiber, and excessive caffeine con~ lion. Incontinence is the lack of voluntary control over micturition. In infants it is normal because neurons to the ç~tern~l sphincter muscle are not completely developed. In the adult it may occur as a result of unconsciousness, injury to the spinal nerves controlling the urinary bladder, irritation due to abnormal con.~tit~ tc in urine, disease of the urinary bladder and inability of the detrusor muscle to relax due to emotional stress. Menstrual cl~ll~ing is a painful spasmodic co~ action of the uterine muscles.
Abnormal contraction in cardiac muscle is the basis of many heart ~ e~ces. Heartfailure (HF) is a common heart disease caused mainly by coronary artery disease and hypertension. Because specific treatment is not possible for most patients, current therapies are designed to ameliorate the symptoms and/or forestall myocardial damage. This includes increasing myocardial contractility by increasing cardiac myofilament sensitivity to calcium (Nielsen, J.E. et al. 1995. J. Cardiovasc. Pharmacol. 26:S77-S84). Changes in calcium sel~iliYily also occur with acute disease states including myocardial s~mning Stunning is a reversible mild form of ischemic (no blood flow) damage, and may be in(luce~l byphysiological insult such as heart surgery. Severe i~rh~rnic damage occurs with myocardial infarction and results in changes to cardiac muscle function.
To date, treatments of the above mentioned states in smooth and cardiac muscle have not been completely effective.

Summa~y of the Invention The present invention is based, at least in part, on the discoveries that PAK induces contraction of smooth muscle in the absence of calcium and changes calcium sensitivity of cardiac muscle. These discoveries led to the methods of the invention.
By one aspect, the present invention provides methods for treating states char~ct~ri7ed by smooth muscle contraction. These states characterized by smooth muscle col-llaction include states characterized by i.,applupl;ate smooth muscle contraction. The tre~tm.ont can involve the reduction of or inhibition of hlal)pl~liate muscle contraction.
Examples of such states include hypertension, asthma, irritable bowel syndrome, incolllillence and menstrual cramps.
This invention pertains to a method of mod~ ting smooth muscle col.llaclion. Themethod involves ~llmini~tering a PAK mod~ ting agent, e.g., PAK inhibitor, to a subject such that modulation of smooth muscle collllaction occurs. A PAK inhibitor may be, for example, an agent which binds to, or blocks, either or both of the kinase domain of PAK
and the p2 1- (e.g., Cdc42 or Racl) binding domain of PAK, or the autophosphorylation sites of PAK. The smooth muscle is present preferably in a blood vessel, the airways of the lungs, the gastro-intestin~l tract, the uterus or the urinary tract.
Another aspect of the invention pertains to a method of treating a subject having a state characterized by smooth muscle co--ll~ction. The method involves a-lrnini~tering to a subject a thel~peulically effective amount of a PAK modlll~ting agent, e.g., PAK inhibitor, such that treatment of the state characterized by smooth muscle contraction occurs. In one embodiment, the smooth muscle has a high basal tone. In another embodiment, the state characterized by the contraction of smooth muscle involves abnormal or inapplupliate contraction of smooth muscle. In yet another embodiment, the state characterized by the contraction of smooth muscle involves abnormal or ir.al)plupliate relaxation of smooth muscle. Finally, the state char~cteri7ed by smooth muscle contraction is preferably hypertension, asthma, irritable bowel syndrome, incontinence or menstrual cramps.
Yet another aspect of the invention relates to a method of treating a subject having a heart condition associated with, or that could lead to, cardiac contractile dysfunction, for example, heart failure (HF). The method involves ~-lmini~tering to a subject a therapeutically effective amount of a PAK modnl~ting agent, e.g., a PAK stimulator or inhibitor, such that treatment of the heart condition occurs.
In yet another aspect, the invention provides assays, e.g., screening tests, to identify PAK or PAK mod~ ting agents. For example, screening tests of the invention can be used S to quantify the amount of PAK or PAK mRNA present, i.e., PAK ~ cssion, within smooth muscle or cardiac muscle. Insofar as PAK t;~ ession can be used as an indicator of a condition, such as hypertension or HF, screening tests according to the invention are therefore useful in ~c~es~ g, for example, the disease state of an individual.
Screening tests of the invention are also useful for detecting or identifying, for example, mod~ ting agents which are inhibitors, or alternatively, stimulators, of PAK
kinase ~lcs~ion or activity. In a ~rcrellcd embodiment, the screening assay identifies agents that modulate the kinase activity of a PAK protein. For example, the invention provides a method involving providing an indicator composition comprising a PAK protein having PAK kinase activity. The method further includes cont~ctin~ the indicatorcomposition with a test agent, and ~letPnnining the effect of the test agent on PAK kinase activity in the indicator composition to thereby identify a compound that modulates the kinase activity of a PAK protein. A statistically significant change, such as a decrease or increase, in the level of PAK kinase activity in the presence of the test agent (relative to what is detected in the absence of the test agent) is indicative of the test agent being a PAK
modlll~ting agent. The indicator composition can be, for example, a smooth muscle cell, a smooth muscle cell extract or a d~lergc~ skinned smooth muscle fiber bundle system.
Screening tests in accordance with the invention can involve identifying the phosphorylation state of downstream target proteins such as, for example, caldesmon (CAD) or a fragment thereof. Thus, in one embodiment, PAK kinase activity is ~eses.ce~l by measuring phosphorylation of CAD. In another embodiment, PAK~ kinase activity isassessed by measuring phosphorylation of the myosin light chain (LC20) or a fragment thereof. In still a further embodiment, PAK kinase activity is assessed by measuring phosphorylation of the calponin protein or a fragment thereof and/or another target protein, such as, for example, troponin I (or a fragment thereof) in cardiac muscle.

Other fealules and advantages of the invention will be a~pdl~lll from the following detailed description, and from the claims.

~riefD~sc,.plion ofthe Drawings Embodiments of the invention will now be described, by way of example, with reference to the accompanying drawings, wherein:
Figures lA and lB show recombinant Cdc42 (lA ) or Racl (lB), labelled with [35S]GTPyS, used in overlay assays of purified brain mPAK3 (1) and either intact (2) or Triton X-100 skinned (3) guinea pig taenia coli smooth muscle fibers.
Figure lC shows a western blot of a 12% SDS-polyacrylamide gel of intact (2) or skinned (3) guinea pig taenia coli and intact rat aorta (4) using an antibody raised against a synthetic peptide corresponding to residues 1-13 of mouse fibroblast mPAK3 (clone 3NT).
Figure 2 shows that constitutively active GST-mPAK3 induces Ca2+-independent contraction of guinea pig taenia coli Triton X-100 skinned muscle fibers without involving myosin light chain kinase (MLCK) or myosin light chain phosphatase (MLCP).
Figures 2A and 2B show typical isometric force tracings of Triton X-l 00 skinnedguinea pig taenia coli fibers in the presence of collsliluli~ely active GST-mPAK3 (2A), or the inactive mutant GST-mPAK3~297R (2B). Fibers were contracted (pCa 5.6 andtor 4.3) and relaxed (pCa 8.3) subsequent to incubation with GST-mPAK3 (bar) at pCa 8.3.
Figure 2C shows the high specificity of wortm~nnin for MLCK by its ability to inhibit in vitro phosphorylation of LC20 by MLCK (-) without inhibiting phosphorylation by GST-mPAK3 (-). 100% equals m~hllulll phosphorylation of LC20 (1 mole of phosphate per mole of protein) by either kinase in the absence of wortm~nnin Figure 2D shows a typical isometric force tracing of Triton X-100 skinned musclefibers in the presence of 1 mM wortm~nnin and constitutively acti~e GST-mPAK3. 1 mM
wortm~nnin was sufficient to elimin~te any Ca2+-dependent contraction (data not shown), and would inhibit any contribution of MLCK to the contraction induced by the presence of PAK.

Figure 3 shows uncoupling between LC20 phosphorylation and force development in skinned muscle fibers even though isolated myosin and LC20 can be phosphorylated at serine 19 by GST-mPAK3.
Figure 3A shows the time course of in vitro phosphorylation of intact isolated chicken gizzard smooth muscle myosin by GST-mPAK3 (* Indicates time at which SDS-PAGE and autoradiography samples were obtained). Isolated smooth muscle myosin phosphorylated in vitro by GST-mPAK3 was analyzed by autoradiography of a 12.5% SDS-polyacrylamide gel (left inset; G = Coomassie stained gel, A = colles~onding autoradiograph) and phospho-amino acid analysis (right inset).
Figure 3B shows time courses of in vitro phosphorylation of isolated smooth muscle LC20 in the presence (-) or absence (O) of GST-mPAK3 followed by the addition ofMLCK after 30 min (* Indicates time at which samples analyzed in Figure 3C were obtained).
Figure 3C shows tryptic peptide maps (left insets) and phospho-amino acid analysis (right inset) of isolated smooth muscle LC20 phosphorylated by MLCK or GST-mPAK3.
Figure 3D is a plot of smooth muscle LC20 phosphorylation (%) and the corresponding relative force (% of force relative to contracting conditions) obtained in the Triton X-100 skinned guinea pig taenia coli muscle fibers at pCa 4.3 (5, n = 5) and pCa 8.3 in the absence (4, n = 6) and presence of GST-mPAK3 (1-3, n = 5) after a 90 minute incubation at 25~C. Experiments were done using five guinea pig Triton X-100 skinned taenia coli preparations and three different GST-mPAK3 plc~alalions. Upper inset shows a typical western blot (using anti-LC20 mAb) of a one-dimensional isoelectric focusing gel (2 to 3 fibers per lane, 18). The ratio of phosphorylated (mono- and di-phosphorylated protein) to unphosphorylated LC20 was detPrmined by densitometry. Standard curves were done to ensure the exposures of the western blots were within a linear range. Note: GST-mPAK3 causes an increase in force with no increase in LC20 phosphorylation.
Figure 4 shows that PAK (4A and 4C) and ROK (4B and 4D) phosphorylate different proteins in skinned smooth muscle fibers. In vitro PAK-phosphorylated LC20 is included in 4A as a standard marker protein.

Figures 4A and 4B are autoradiographs of a 12% SDS-polyacrylamide gel analysis showing that phosphorylation of LC20 by GST-ROK in the absence of calcium is similar to phosphorylation of LC20 by MLCK at pCa 4.3.
Figures 4C and 4D are autoradiographs of a 10% SDS-polyacrylamide gel analysis showing that caldesmon is the only protein phosphorylated by GST-mPAK3 that is not phosphorylated by GST-ROK in the ~kinne,~ muscle fibers. These autoradiographs were generated from the same blots used for western blotting with the anti-caldesmon or anti-desmin antibodies.
Figure 4E shows the time course of in vitro phosphorylation of isolated chicken gizzard h-caldesmon (hCAD) by GST-mPAK3 (-) or GST- ROK (O).
Figure 4F shows an autoradiograph of a 12 % SDS-polyacrylamide gel of LC20, C-te.rmin~l fragment of caldesmon ~CAD39), and intact chicken gizzard h-caldesmon phosphorylated in vitro by GST-mPAK3.
Figure 5 shows a relationship between phosphorylation of c~ldecmon by PAK and the binding of caldesmon and actin. The data translate into an approximately two-fold reduction in the affinity of caldesmon for actin-ll~ polllyosin.
Figure 6 is a graph showing that PAK-phosphorylated caldesmon is a poor inhibitor of myosin ATPase activity. Thus, phosphorylation of caldesmon tli~inhibits caldesmon which allows actin-tropomyosin to interact with myosin, incleasillg ATPase activity and reSultinp~ in contraction.
Figure 7 is a schematic representation of a possible PAK pathway responsible forsmooth muscle contraction.
Figure 8 shows the results of a gel overlay assay in which GTP y35S-Cdc42 bound to a ~ 62kDa protein in porcine (P), dog (D), and rat (R) cardiac muscle, confirming the presence of PAK or a PAK homologue in cardiac muscle (C3, control--brain PAK; Sm, guinea pig smooth muscle; Rs, rat skinned muscle).
Figures 9A, B, and C show autoradiographs of in vitro phosphorylation of myosin isolated from various sources (C, cardiac; Sk, skeletal muscle; Sm, smooth muscle; P or Pl, platelet; D, Dictyostelium). LC20 (about 20 kDa) is phosphorylated to various extents by PAK (Figure 9A), MLCK (Figure 9B), and ROK (Figure 9C). Cardiac muscle LC20 is phosphorylated by MLCK and ROK, but poorly by PAK.
Figure 10 shows the time course of phosphorylation of cardiac LC20 by GST-ROK, up to 0.5 mol phosphate per mol protein.
Figure 11 shows the results of a phospho-amino acid analysis for cardiac LC20, in which it can be seen that it is the same serine that is phosphorylated by both MLCK and GST-ROK.
Figures 1 2A, B, and C show tryptic peptide maps produced from cardiac LC20 phosphorylated by MLCK, ROK, and MLCK and ROK, respectively.
Figure 13 is an autoradiograph showing in vitro phosphorylation of cardiac troponin I (TnI). Cardiac Tn complex (TnI, TnT, and TnC) was phosphorylated by GST-mPAK in the presence of 32P-ATP, and TnI isolated by reversed phase HPLC.
Figures 14A and 14B show that the force produced by Triton X-100 ~inned cardiac muscle fiber bundles increases in the presence of GST-mPAK3, at calcium concentrations that produce subm~xim~l contraction.
Figure 15 is an exemplary trace showing increased calcium sensitivity of cardiacmuscle contraction in the presence of GST-mPAK.

De~ailed D~s~ ion According to one aspect of the present invention there is provided a method of modnl~ting smooth muscle contraction, e.g., calcium independent. The method involves ~llmini~tP~ring a PAK mod~ ting agent to a subject such that modulation of smooth muscle contraction occurs.
The language "smooth muscle" is intP.n~led to include smooth muscle sensitive to the PAK mod~ ting agents of the present invention. Smooth muscle-is sensitive to a PAK
mod~ ting agent if the agent modulates the contraction of the smooth muscle. Examples of smooth muscle include smooth muscle of a blood vessel, the airways of the lungs, the gastro-intestinal tract, the uterus, and the urinary tract.
As used herein, the language "modlll~ting smooth muscle contraction" is intended to include the capacity to inhibit or stimulate smooth muscle contraction to various levels, e.g., which allows for the treatment of targeted states. The language is also intended to include the inducement of relaxation of smooth rnùscle, e.g., total relaxation, and the contraction of, smooth muscle which is in relaxed state and it is desired to have the muscle in a more - . . ~ . .
contracted state, e.g., the sphincter in esophageal reflux.
According to another aspect of the present invention, there is provided a method of mod~ ting cardiac muscle contraction by ch~nging the calcium sensitivity of cardiac muscle. The method involves ~-lmini.ct~o.ring a PAK mod~ ting agent to a subject such that a change in calcium sensitivity of cardiac muscle, and hence modulation of contraction of cardiac muscle, occurs. A PAK modul~ting agent is an agent which énhances, or reduces, the ~ ression of PAK or a PAK homologue present in cardiac muscle,~or an agent which stim~ tes or inhibits the activity of PAK or a PAK homologue in cardiac muscle.
As used herein, the ~ s~ion "change in calcium sensitivity of cardiac muscle"
means a change in the amount (intracellular concentration) of calcium necessary to elicit -cardiac muscle contraction of a given.level of force. The change may be an increase in calcium sensitivity, such that a lower calcium concentration, relative to a normal physiological cardiac calcium concentràtion, will elicit cardiac muscle contraction with a given level of force. Alternatively, the change may be a decrease in calcium sensitivity, such that a higher calcium concentration, relative to a normal physiological cardiac calcium concentration, will elicit cardiac muscle contraction with a given level of force. A normal physiological cardiac calcium concentration refers to the concentration of calciurn normally present in cardiac muscle of an individual not subjected to the methods of the present invention.
As used herein, the terms "subject" and "individual" are intended to include ~nim~
preferably m~mm~ , most preferably hllm~n~. In a pref~lled embodiment, the subject is a primate. In an even more preferred embodiment, the primate is a human.
As used herein, the language "PAK modul~hng agent" includes;an agent which interacts with a p21-activated kinase and affects its activity. The modul~ting agent can be an inhibitor or a potentiator. As used herein, the language "PAK inhibitor" is intended to include agents which inhibit a PAK kinase. Examples of agents that inhibit serine/threonine kinases that may accordingly inhibit PAK, include Staurosporin, PD098059, Genistein, g Sm~e, it is o~en difflcnlt to achielre i~trar~ c4~ce~ n~s ofthc an~sense m~ .~O;G.,1 to S~IY1J~ cl~t;o~ on ~dO~ .31nRNAS, apl~f~~ proach utilize~s a ~ ant DNA co~ct in whic~ thc ~ e ~oe oligonucleotidG is placed undOE
the çon~ol of a ~trong pol m ~ pol II l~,u~ot~ . I~e use of such a eo-~ll u~i to t~ r~;~t targetsmoo1~musclecell6inthepadent~villresultin~heL~n~r~ril~;o~.of~ufficientP-~~~ t~
of ~ingle s~ded RNA~ ~at vnll form ~ple .. -lary base pairs un~h the ~ -~Ao~ -~.uc PAK ~ipts and thereby prevent tr~slation of the PAK mRN~ ~or example, a vec~or can be i~troduced in vi~o swh t~at it is taken up by a ~mooth mu~cle ~ell and direct~ the of ~n ~tisense R~ S7~ch a ~ector call rema~n e~yJl~al or become cl~Q~o~ y inte~ted, as long a~ it can be t~-e~ to l~lu~hP~ ~he desired anti~--ceRNA. Such vectors a~ he co~h~t~d by re~4mh~ t DNA tr~hnslogy .-~G!hod~
in thc art. Vectors can be ~!~cmid, Yiral, or o~s known in the a~t, used for re~ Dtiol~ ~md g~o~ ncelLc. lhl, ~si~lofthe~e.lceonr4~:ngthe~nt~6~n~eRNA
c~n be by any ~ ote- known ~n the att to act in mam~n, p~ Iy hum~n smooth lS muscle cells. S~ r~ tl - ~ can be ~ VGi~lP~ ;ve. Such prolnot~R ~nclude but are not limited to; tho $V40 early pro~,t,r ~e~ion (1 t~ rlet al. 1981. Nats~re 290:304 310), the p.olu~t,,r c~ ~t~;"rA ~n the 3' long t ~ P~l repeat of Rou~ 6aI~Oma ~
(Yamamoto e~ aL 1980. Cell 22:787-797), thc he~pes ~..Ji.l;l.~ kiP;~e p~omoter (WaE~ner et al. 19~1 Proc. Nat~. Acad. Sci. US~ 78:1441-1445), the regulatory 5 ~ of the~1~ts11OIh;.. ~ ~t~ ~ et aL 1982. ~a~re 296:39~2)~ etc. Any ype of p~d, co~m:d, YAC or vi~l lrector c~n be use~ to pr~are ehc ~ .k;~t I~NA co~ wllid can be introduced directly into ehe eiOO~ue site, e.g,, ~lhe g~ t~.
Ril~e - ~le~ ~i~ed to c~lydc~lly cleave PA~ niRNA ~..~ can ~lso be ~aea to preverle ~l~tion of PAK mRNA and ~O;~ of PAK (See, e.~, PCI
~5 ~ I Publication W0 90/11364, plll~1iChf~ October ~, l9gO; Sarver ~ ul. 1990.
Science 247:12~2-12~5). ~Yhile rib~,~ that cleave r~NA at site sp~fic l.40 s~nt -~ ca~ be u8ud ~ d~sb~yPA~K ~ ~he use of ha~mr ~- - Ad n~nes ~
prefoned. ~ hP~Ti ~ es oleave ~A~ at lne~tjQn.~ t~P~ by ~bng re~ons ~atfo~cc~ lel~P-~1~yba$epE~irswiththetargetm~NA. Theso1elequ~r~ 1isth~tthe target m~:NA h~lre ~he ~llo~ing s -~Ih-~re oftwo ~a~es; 5~-U~-3'. The cou~hu~liull and ~0-.10.1998 1~629 I~B456853 re ~ ed with the phosphorylatable Ser/Thr replaced by another amino acid that is not phosphorylatable, such that the peptide acts as a "dead-end" pseudosubstrate and inhibits PAK. Preferably, amino acids used to replace the Ser/Thr are Asp/Glu or Ala. Since PAK
can not phosphorylate such peptides, they act as competitive inhibitors. Suitable synthe~i7e~1 peptides include about 10 to 40 amino acids, typically about 12 to 20 amino acids.
Peptide inhibitors of PAK can thus be synthesized based on PAK substrates, either exogenous (LC20 and CAD) or troponin I, or intramolecular (i.e., autophosphorylation/
autoinhibitory sites). As well, peptide inhibitors of PAK can be peptides that compete for Cdc42 or Racl binding sites of PAK. In any case, peptide inhibitors of PAK are non-phosphorylatable analogues or pseudosubstrates of PAK substrates.
Kinases regulated by autophosphorylation may contain several amino acid residuesin addition to the autoinhibitory sequence that are important for ~ulJsl~le recognition (Johnson, L.N. et al. 1996. Cell 85:149-158; Kemp, B.E. et al. 1994. Trends in Biochem.
Sci. 19:440-444). For instance, an important basic amino acid is sitll~te~l between the autoinhibitory region and the catalytic domain. In the PAK sequence there are two possible Lys residues that meet this criterion. Thus, the synthetic peptides can also include either or both of the two possible candidate residues. In these peptides, Cys is preferably substituted with Ala to minimi7e oxidation problems and disulfide bridge formation. Also, the N- and C-termini of all peptides are preferably acetylated and amidated, respectively, to elimin~te effects from extra charges. This is especially preferred for small peptides, particularly as the ~inillluln functional sequence is approached.
The ",il~i",um phosphorylatable amino acid sequence of a synthesized peptide that is actively phosphorylated by PAK can be determined using a series of N- and C-t~nnin~l truncation analogues. For example, an N-terminal truncation peptide series based on the PAK sequence (Tuazon, P.T. et al. 1997. Biochemistry 36:16059-16064; Manser, E. et al.
1997. Mol. Cell Biol. 17: 1129-1143; Bagrodia, S. et al. 1995. J. Biol. Chem. 270:22731 -22737) can be established as follows:

peptide 1. SVKLTDFGFAAQITPEQSKRST*MVGTPY

peptide 2. GF_AQITPEQSKRST*MVGTPY
peptide 3. PEQSKRST*MVGTPY
peptide 4. SKRST*MVGTPY

_ is an Ala substituted for Cys in the native sequence. K is a Lys (i.e., a basic residue) as discussed above. T* is the targeted Thr phospho-amino acid.

Synthetic peptide inhibitors or modulators It has been shown that PAKl phosphorylates synthetic peptides with sequence (R/K)-(R/K)-X-S (Tuazon et al., 1997 Biochemistry 36:16059). All of the possiblephosphorylation sites we have identified include this sequence, either conserved or semi-conserved. However, many other kinases including MLCK and PKA have the same or similar amino acid consensus sequence. To increase specificity of peptide inhibitors for PAK rather than other kinases including PKA and MLCK, an extension of the amino acid sequence past (R/K)-(R/K)-X-S/T is p,ere"ed.
In the first autoinhibitory region, one or more positively charged amino acid residues (Arg, R or Lys, K) upstream (towards the N-terminus) of the phosphorylated Ser or Thr are believed to play a role. Therefore, these are included in the initial screening of the peptides for TnI and caldesmon.
TnI:
As reported herein, skeletal TnI is not phosphorylated, and this is supported by lack of the (R/K)-(R/K)-X-S/T motif in its amino acid sequence. There are three regions of cardiac TnI which contain this motif. They are conserved belwee" rat and human cardiac TnI. These regions termin~te with amino acid residues Ser 22, 23, 38, and 164 of human cardiac TnI. Ser 22 and 23 are phosphorylated by protein kinase A (PKA) during ~-adrenergic activation.
For cardiac TnI, there is an additional R or K situated upstream in relatively close proximity to two of these four Ser in human. The two regions are:
YRAYATEPHAKKKS* (residues 25-38); and RISADAMMQALLGARAKES* (residues 147-164).
However, for Ser 22, 23 (Ser 21, 22 in rat) there is no upstream R or K present, indicating either that neither of these Ser are phosphorylated, or (more likely) that the additional positively charged residue is not critical.
However, Ser 22 and 23 of cardiac TnI are believed to represent a different conformationally prefelled binding site for PAK, since there are three proline (Pro, P) residues situated at positions 11, 13, 15 and 17. It is likely that these induce a unique structure in TnI that is plef~lled by PAK (AAREPRPAPAPIRRRSS* residues 7-22 of human cardiac TnI). With myosin light chain LC20, there are also Pro situated at -5 u~ ll of the target Ser, and u~slle~ll Pro occur at several of the autophosphorylation sites for PAKl and 3.

Caldesmon:
Caldesmon has several possible PAK phosphorylation sites near the C-termim.s of chicken gizzard h-CAD. Several of these sites are not present in human fibroblast caldesmon, but may be present in the smooth muscle isoform of human h-CAD. Based on the chicken gizzard h-CAD sequence, there are KEAKVEAKKES* (good match, residues423-433), PFKCFSPKGS*S* (weakly conserved, residues 592-602), PAPKPS (residues 718-723) and KVTATGKKS* (residues 751-759). Two ofthese sites arephosphorylated by PAK. Two tryptic fragments of PAK phosphorylated chicken gizzard h-CAD have been isolated and amino acid sequencing of such will verify the actual phosphorylation sites.
Note that these phosphorylation sites contain either a positively charged amino acid (K in these cases) or several prolines starting at -3 or -4 positions.

Autoinhibitory re~ion of PAK l/PAK 65 Importantly, the autoinhibito~y region of PAK65 or PAKl has been deteTmined to be near or at SKRSMVGTPY (site 1) and SVDPVPAPVGDS*HVDGAAK (site 2). (Note that several amino acids were mi~in~ from site 1 in Benner et al., 1995. J. Biol. Chem. 270:
21121 -21128, but this error was corrected in a subsequent report of the cDNA sequence, Manser et al., 1997.) Site 1 is rapidly phosphorylated and site 2 is required for activation of PAK1. Site 1 is highly conserved in all PAK isoforms but only the extreme N-t~rmin~l part of site 2 is. There are no basic amino acids within site 2 (which is part of the "preferred"
consensus sequence later described by Tuazon et al., 1997). There is a series of PXPXP at position -4 or -5, and it is believed that those proline residues provide unique identification for PAK. .
Further work on PAKa or PAKl (Manser et al., 1997. Mol. Cell. Biology 17:1129) showed that there are up to 7 autophosphorylation sites. Two of these are the same as shown by Benner et al. Several of these sites are conserved between the 3 isoforms of PAK
(including PAK~ which is PAK3). For PAK3, sites A, E and F are conserved; site C which is KYMS in PAK1 is changed to KYLSin PAK3; and site G (catalytic site) which is TTPPKRST in PAKl is changed to SGAKRST in PAK3; while sites B and D are mis~ing the key residues.
It is expected that PAKl is a good model for PAK3 with respect to specificity oractivity at the catalytic site, and differences between substrates that the two kinases can phosphorylate may be minim~l That is, the difference between specificity of PAKl and PAK3 likely lies in another domain, such as SH domains, and in the actual cellular localization of the various isoforms of PAK. This means that the amino acid residues near the sites of phosphorylation of substrates for PAKl and PAK3 are expected to be similar, if not the same. So, even if PAK3 does not contain all of the autophosphorylation sites that are present in PAKl, the consensus amino acid sequences for PAKl or PAK3 are expected to be the same. Therefore, the amino acid sequences of PAK1 phosphorylation sites were employed in designing amino acid sequences for peptide substrates for PAK3.

Peptide with conformationally plefellc;d PAK sites For determining the any important aspects of phosphorylation sites for PAK whereupstream Pro residues likely provide an unique conformation, we compared the amino acid sequences. For smooth muscle MLCK, the consensus phosphorylation site is (K/R2,X)-XI 2-K/R3-X23-R-X2-S-N-V-F (Kennelly et al. 1991. J. Biol. Chem. 266:15555) and does not have any proline residues, even though there are numerous positively charged amino acid residues near R andlor K. It is believed that there is a unique amino acid sequence/conformation for PAK that is not seen with other kinases of interest.

PAPPMRNTS* (site A autophosphorylation) PRPEHTKS* (site E autophosphorylation) TTPPKRST* (site G autophosphorylation)catalytic site TEPHAKKKS* (residues 25-38 TnI) PRPAPAPIRRRS*S* residues 7-22 of human cTnI
PFKCFSPKGSS* (weakly conserved, residues 592-602 h-CAD) PAPKPS* (residues 718-723 h-CAD) Deterrnin~tion of the TnI and caldesmon phosphorylation sites permits an amino acid consensus sequence for this series of peptides to be det~nnined. These are expected to be similar to PXPPXl -2(R/K)(IVK)XS*. By ch~nging the number of residues (range from 1 -3) and amino acid composition of the X position, the optimal peptide substrate for PAK
can be ~letennined. Thus, the phosphorylatable Ser can be substituted with Asp, Ala and Gly to produce peptides expected to be inhibitors.

Optimization of peptide sequence The amino acid residues initially screened above are first truncated from the N-termim-s to determine the .~ il,llllll sequence required for substrate activity by PAK. Then, any residues other than K, R or P which may be important are substituted with other amino acid residues. Conservative changes in amino acid sequence such as R to K (both being positively charged) or V to I or L (all being hydrophobic) allow for a number of peptides with the similar affinity and efficacy for PAK inhibition or modulation. Thus there is expected to be a family of peptides which have certain key amino acid residues, but also other less critical amino acid residues, or additional residues which may be widely substituted.
Finally, the use of synthetic peptides is often the starting point for the development of peptide mimetics or drugs. The parts of the peptide or peptide mimetic which are in contact with PAK in 3 dimensional space so as to provide the desired biological function are critical for this purpose. Thus, a family of peptides with the same efficacy for PAK should have specific and critical moieties (e.g., carbon group, proline "kink", positive charge, or the like) located in space at the same sites, based on the structure and chemical nature of the natural substrate. Determining the structure of this family of peptides provides the pre~l~ed structure of the PAK binding/phosphorylation site and ligands which interact with said site.
To assess the specificity of PAK for the synthesized peptides, the peptides should also be tested as possible inhibitors and/or substrates for other kinases which are considered to have a regulatory role in muscle; in particular, MLCK, MAPK (Upstate Biotechnologies), ROK, and PKA (protein kinase A, active catalytic domain, Sigma). In addition, the peptides should be tested against other PAK family members including myosin I heavy chain kinase. PAK autoinhibitory peptides are expected to exhibit different activities with PAK and MAPK and PKA since their consensus sequences differ (Johnson, L.N. et al.
1996. Cell 85:149-158). Because both MLCK and PAK can phosphorylate smooth muscle LC20, there may be some cross specificity within their consensus sequences, and optimization of peptide inhibitors may require peptide library scl~el~ g. In addition, some substrate peptides are expected to be phosphorylated by other members of the PAK family while inhibitory peptides act as inhibitors. Since there are only two amino acids within the concenclls sequences of myosin I heavy chain kinase (a member of PAK family) that match with PAK, the peptides based on PAK sequence (as described above) should be lessefficient with other members of the PAK family. Nevertheless, specific amino acid ~ul)~lilulions can be carried out on a synthesi7e~ peptide, using methods well known in the art such as solid phase synthesis (Merrifield. 1964. J. Am. Chem. Assoc. 85: 2149-2154), or ~ synthesis in homogeneous solution (Houbenwcyl. 1987. Methods of Organic Chemistry, ed.
E. Wansch. Vol. 15 I and II. Thieme, Stuttgart), to optimize the specificity of PAK for the peptide, and to elimin~te any cross-phosphorylation with other known Ser/Thr kinases including MLCK and other members of the PAK family.
Synthesized peptides can be used to prepare PAK modulators such as PAK
inhibitors where they are good substrates or competitors for PAK. Synthesized peptides may thus be tested for their utility as substrates for PAK by, for example, det~"llini~-g the phosphorylation kinetics of the peptides at various concentrations of PAK with 32P-ATP.
The level of 32p incorporated into the peptide can be quantified following separation of the peptide from free ATP by NO-P81 paper assays or gel filtration (Sephadex G-10).
Similarly, standard kinetic assays can be performed to assess the effectiveness of PAK inhibitors. The more effective the peptide inhibitor is, the less phosphorylation (less 32p label) there will be of LC20. Peptide inhibitors can also be tested against CAD and TnI.
Various concentrations of peptide inhibitors can be tested on Triton X-l 00 skinned smooth muscle and cardiac muscle fibers in the presence of exogenously supplied PAK to determine the optimal inhibitor concentration for inhibition of contraction. With these studies, the levels of LC20, CAD, TnI, and any other target protein phosphorylation should preferably also be monitored.
Peptide inhibitors of PAK can be ~ssessecl in smooth muscle in vivo using different strategies. Briefly, the peptides can be perfused into reversibly permeabilized smooth muscle and alpha-toxin or ,~-escin skinned muscle fibers, and micro-injected into isolated myocytes. In contrast to Triton-skinned muscle fibers, these muscle plepalalions preserve signal transduction pathways and retain essential cytosolic mediators, m~king them suitable for ~ses~ment of PAK function, and hence modulator function, in vivo.
Reversible permeabilization of intact smooth muscle tissue uses high EDTA/ATP inthe extracellular solution and is used frequently to load living tissue with proteins and compounds less than 100 kDa. Alpha toxin produces uniformly small holes in the plasma membrane which allow molecules under 1 kDa to enter. Because ~-escin skinned fibers allow entrance of large proteins (<100 kDa), the fibers slowly run down due to slow loss of cytoplasmic proteins (including PAK) after several stim~ tions (Otto, B. et al. 1996. J.
Physiol. 496:317-329). This limits the number of contractions that can be carried out by a single fiber. While either of these procedures is suitable, the use of a large peptide modulator will dictate the use of the ~-escin skinned fiber procedure.
For in vivo ~ses.~i",ent of peptide inhibitors of PAK, it is necess~y to activate PAK.
While the exact stimulation required to activate PAK, other than PDGF in platelets and hyperosmotic shock in myocytes, is not known, various agonists known to 1) activate the Rac/Cdc42 pathway and 2) cause calcium-spn~iti7~tion~ ranging from PDFG to GTP~S, may activate PAK. These agonists can be pre-screened for their ability to activate PAK
and/or phosphorylate its target proteins. Activation of PAK can then be determined by loading alpha-toxin skinned fibers with 32p and monitoring autophosphorylation of PAK
(and incorporation of 32p into PAK). Following various stimulations, the fibers can be analyzed by SDS-PAGE and the level of 32p incorporated into the band migrating at the molecular weight of PAK and/or target proteins determined. PAK can also be immunoprecipitated from stimulated tissue for direct analysis, using, for example, the antibody previously used to demonstrate that PAK is present in smooth muscle (Van Eyk, J.E. et al. 1998. J. Biol. Chem. submitted). This antibody has high affinity and specificity for PAK and is suitably used to identify the PAK band and for immllnl)plecipitation.
Also, with the intact preparation, intercellular calcium can be measured using the photoprotein aqueorin, which can be loaded into the smooth muscle cells. This allows measurement of calcium levels during stimul~tion with agonist (Rembold, C.M. et al. 1988.
Circ. Res. 63:593-603). With either skinned fiber plep~lion, calcium levels are controlled by the bathing soIution. This method can therefore be use to demonstrate a reduction in calcium level fluctuation in the presence of a peptide inhibitor of PAK.
As would be understood by a person of ordinary skill in the art, a variety of amino acid substitutions may be made to the above-identified peptides while preserving their structure and protein (PAK) binding activity. Conservative substitutions are described in the patent literature, as, for example, in U.S. Pat. No. 5,264,558. It is thus expected, for example, that interchange among non-polar aliphatic neutral amino acids, Gly, Ala, Pro, Val and Ile would be possible. Likewise, substitutions among the polar aliphatic neutral amino acids, Ser, Thr, Met, Asn and Gln could possibly be made. Substitutions among the charged acidic amino acids, Asp and Glu, could probably be made, as could substitutions among the charged basic amino acids, Lys and Arg. Substitutions among the aromatic amino acids, including Phe, His, Trp and Tyr would also likely be possible. These sorts of substitutions are well known to those skilled in the art. Other substitutions might well be possible. All such variants are encompassed by the scope of the invention. It would be expected that the greater the sequence similarity of the peptide to the related region of the native protein, the greater the degree of the desired biological activity.

In addition, modulators, such as inhibitors, of PAK ~ression or activity useful in the method of the invention include PAK specific antisense nucleic acids, ribozymes and intracellular antibodies. Antisense approaches involve the design of oligonucleotides (either DNA or RNA) that are complementary to PAK mRNA. The ~nticence oligonucleotides will bind to the PAK mRNA transcripts and prevent translation. ~ltem~tively, the antisense oligonucleotide may bind to DNA of a PAK gene, such as, for example, a regulatory element. Absolute complementarity, although preferred, is not required. A sequence "complementary" to a portion of a nucleic acid, as referred to herein, means a sequence having sufficient complementarity to be able to hybridize with the RNA, forming a stable duplex. In the case of double-stranded ~nticen.ce nucleic acids, a single strand of the duplex DNA may thus be tested, or triplex formation may be assayed. The ability to hybridize will depend on both the degree of complementarity and the length of the antisense nucleic acid.
Generally, the longer the hybridizing nucleic acid, the more base micm~tches with an RNA
it may contain and still form a stable duplex (or triplex, as the case may be). One skilled in the art can asc~l~in a tolerable degree of mi.cm~tch by use of standard procedures to detemmine the melting point of the hybridized complex.
Oligonucleotides that are complementary to the 5' end of the PAK message, e.g., the S' untr~ncl~tecl sequence up to and including the AUG initiation codon, should work most efficiently at inhibiting translation. However, sequences complementary to the 3' untr~ncl~tecl sequences of the PAK mRNA may also be used. (Wagner, R. 1994. Nature 372:333). Therefore, oligonucleotides complementary to either the 5' or 3' untr~ncl~ted, non-coding regions of a PAK gene could be used in an ~ntic~nce approach to ir~ibit translation of endogenous PAK mRNA. Oligonucleotides complementary to the 5' untr~ncl~ted region of the mRNA should include the complement of the AUG start codon.
.Anticçnce oligonucleotides complementary to mRNA coding regions are generally less efficient inhibitors of translation but could be used in accordance with the invention.
Whether designed to hybridize to any of the afore-mentioned regions of PAK nucleic acid, ~nti.c~nce nucleic acids should be at least about six nucleotides in length, and are preferably oligonucleotides ranging from 6 to about 50 nucleotides in length. In certain embodiments, the oligonucleotide is at least about 10 nucleotides, at least about 17 nucleotides, at least about 25 nucleotides, or at least about 50 nucleotides.
Regardless of the choice of target sequence, it is ple~el,ed that in vitro studies are first performed to quantitate the ability of the ~nti~en~e oligonucleotide to inhibit PAK gene t;~res~ion and therefore smooth muscle contraction, or cardiac muscle calcium sensitivity.
It is preferred that these studies utilize controls that distinguish between antisense gene inhibition and nonspecific biological effects of oligonucleotides. It is also plef~lled that these studies compare levels of the target RNA or protein with that of an intern~l control RNA or protein. Results obtained using the antisense oligonucleotide can be colllpaled with those obtained using a control oligonucleotide. It is pler~lled that the controloligonucleotide is of approximately the same length as the test oligonucleotide and that the nucleotide sequence of the oligonucleotide differs from the ~nti~. n~e sequence no more than is necessary to prevent specific hybridization to the target sequence.
The oligonucleotides can be DNA or RNA or chimeric llli~UleS or derivatives or modified versions thereof, single-stranded or double-stranded. The oligonucleotide can be modified at the base moiety, sugar moiety, or phosphate backbone, for example, to improve stability of the molecule, hybridization, etc. The oligonucleotide may include other appended groups such agents facilitating transport across the cell membrane (see, e.g., Letsinger et al. 1989. Proc. NatL Acad. Sci. U.S.A. 86:6553-6556; T em~itre et al. 1987.
Proc. Natl. Acad. Sci. U.S.A. 84:648-652; PCT Publication No. W088/09810, published December 15, 1988).
While antisense nucleotides complementary to the PAK coding region sequence could be used, those complementary to the transcribed untr~n~l~te~l region are most l~lere"ed.
The antisense molecules can be delivered to smooth or cardiac muscle cells whichexpress the PAK in vivo or in vitro. A number of methods have been developed fordelivering antisense DNA or RNA to cells; e.g., antisense molecules can be injected directly into the tissue site, or modified antisense molecules, designed to target the desired cells (e.g., antisense linked to peptides or antibodies that specifically bind receptors or antigens expressed on the smooth muscle cell surface) can be ~timini~tered systematically.

Since it is often difficult to achieve intracellular concentrations of the antisense molecule sufficient to ~u~pless translation on endogenous mRNAs, a prere,led approach utilizes a recombinant DNA construct in which the antisense oligonucleotide is placed under the control of a strong pol III or pol II promoter. The use of such a construct to transfect, for example, target smooth muscle cells in the patient will result in the transcription of sufficient amounts of single stranded RNAs that will form complementary base pairs with the endogenous PAK ll~lsc,i~ and thereby prevent translation of the PAK mRNA. For example, a vector can be introduced in vivo such that it is taken up by a smooth or cardiac muscle cell and directs the transcription of an antisense RNA. Such a vector can remain episomal or become chromosomally integrated, as long as it can be transcribed to produce the desired antisense RNA. Such vectors can be constructed by recombinant DNA
technology methods standard in the art. Vectors can be plasmid, viral, or others known in the art, used for replication and e~lession in m~mm~ n cells. Expression of the sequence encoding the antisense RNA can be by any promoter known in the art to act in m~mm~ n, pref~ably human, smooth muscle or cardiac muscle cells. Such promoters can be inducible or constitutive. Such promoters include but are not limited to: the SV40 early promoter region (Bernoist et al. 1981. Nature 290:304-310), the promoter contained in the 3' long tçrmin~l repeat of Rous sarcoma virus (Yamamoto et al. 1980. Cell 22:787-797), the herpes thymidine kinase promoter (Wagner et al. 1981 Proc. Natl. Acad. Sci. U.S.A. 78:1441-1445), the regulatory sequences of the metallothionein gene (Brinster et al. 1982. Nature 296:39-42), etc. Any type of plasmid, cosmid, YAC or viral vector can be used to prepare the recombinant DNA construct which can be introduced directly into the tissue site; e.g., the gastro-intestinal tract.
Ribozyme molecules designed to catalytically cleave PAK mRNA transcripts can also be used to prevent translation of PAK mRNA and ~ res~ion of PAK. (See, e.g., PCT
lntern~tional Publication WO 90/11364, published October 4, 1990; Sarver et al. 1990.
Science 247:1222-1225). While ribozymes that cleave mRNA at site specific recognition sequences can be used to destroy PAK mRNAs, the use of hammerhead ribozymes is prefellcd. Hammerhead ribozymes cleave mRNAs at locations dlctated by fl~nking regions that form complementary base pairs with the target mRNA. The sole requirement is that the target mRNA have the following sequence of two bases: 5'-UG-3'. The construction and production of harnmerhead ribozymes is well known in the art and is described more fully in Haseloffet al. 1988. Nature 334:585-S91. There are numerous potential hammerhead ribozyme cleavage sites within the nucleotide sequence of human PAK cDNA. Preferably the ribozyme is engineered so that the cleavage recognition site is located near the 5' end of the PAK mRNA; i.e., to increase efficiency and minimi~e the intracellular accumulation of non-functional mRNA transcripts.
Ribozymes of usefulness in the present invention also include RNA
endoribonucleases (hereinafter "Cech-type ribozymes") such as the one which occurs naturally in Tetrahymena thermophila (known as the IVS, or L-l9 IVS RNA) and which has been extensively described by Thomas Cech and collaborators (Zaug, et al. 1984. Science 224:574-578; Zaug et al. 1986. Science 231 :470-475, Zaug et al. 1986. Nature 324:429-433; published International patent application No. WO 88/04300 by University Patents Inc.; Been et al. 1986. Cell 47:207-216). The Cech-type riboz.vmes have an eight base pair active site which hybridizes to a target RNA sequence whereafter cleavage of the target RNA takes place.
As in the ~nti~, n~e approach, the ribozymes can be composed of modified oligonucleotides (e.g., for improved stability, targeting, etc.) and should be delivered to smooth muscle or cardiac muscle cells which express a PAK in vivo. A preferred method of delivery involves using a DNA construct "encoding" the ribozyme under the control of a strong constitutive pol III or pol II promoter, so that transfected cells will produce sufficient quantities of the ribozyme to destroy endogenous PAK mRNA and inhibit translation.
Because ribozymes unlike antisense molecules, are catalytic, a lower intracellular concentration is required for efficiency.
~lt~ tively, endogenous PAK gene t;A~lession can be reduced by targeting deoxyribonucleotide sequences complementary to the regulatory region of the PAK gene (i.e., the PAK promoter and/or enhancers) to form triple helical structures that prevent seli~tion of the PAK gene in target muscle cells in the body. (See generally, Helene, C.
1991. Anticancer Drug Des. 6:569-84; Helene, C., et al. 1992. Ann, N. Y. Acad. Sci. 660:27-36; and Maher, L.J. 1992 Bioassays 14:807-15).

Nucleic acid molecules to be used in triple helix formation for the inhibition of transcription are preferably single stranded and composed of deoxyribonucleotides. The base composition of these oligonucleotides should promote triple helix formation via Hoogsteen base pairing rules, which generally require sizable stretches of either purines or pyrimidines to be present on one strand of a duplex. Nucleotide sequences may bepyrimidine-based, which will result in TAT and CGC triplets across the three associated strands of the resulting triple helix. The pyrimidine-rich molecules provide base complementarity to a purine-rich region of a single strand of the duplex in a parallel orientation to that strand. In addition, nucleic acid molecules may be chosen that are purine-rich, for example, co~t~ g a stretch of G residues. These molecules will form a triple helix with a DNA duplex that is rich in GC pairs, in which the majority of the purine residues are located on a single strand of the targeted duplex, resulting in CGC triplets across the three strands in the triplex.
Alternatively, the potential sequences that can be targeted for triple helix formation may be increased by creating a so called "switchback" nucleic acid molecule. Switchback molecules are synth~si7e~1 in an altern~tin~ 5'-3', 3'-5' manner, such that they base pair with first one strand of a duplex and then the other, elimin~ting the necessity for a sizable stretch of either purines or pyrimidines to be present on one strand of a duplex.
In yet another embodiment, the antisense oligonucleotide is an a-anomeric oligonucleotide. An a-anomeric oligonucleotide forms specific double-stranded hybrids with complementary RNA in which, contrary to the usual ~-units, the strands run parallel to each other (Gautier et al. 1987. Nucl. Acids Res. 15:6625-6641). The oligonucleotide is a 2'-0-methylribonucleotide (Inoue et al. 1987. Nucl. Acids Res. 15:6131-6148), or a chimeric RNA-DNA analogue (Inoue et al. 1987. FEBS Lett. 215:327-330).
In one embodiment, a PAK inhibitor is an antibody which binds to either PAK (i.e., an anti-PAK antibody) or a PAK downskeam element (i.e., an anti-caldesmon antibody) to substantially block or inhibit an interaction between PAK and a PAK dowll~lle~ll element (i.e., blocking antibodies). Additionally, antibodies to PAK or a PAK downskeam element can be produced by conventional techniques. Antibodies can be polyclonal, or more preferably, are monoclonal. Polyclonal and monoclonal antibodies can be prepared by standard techniques known in the art. For example, a m~mm~l, (e.g., a mouse, h~m~t~r, or rabbit) can be immunized with an antigen (i.e., PAK or a PAK downstream element), for example with purified protein, recombinant protein, or peptide fragments thereof, or with a cell which expresses the antigen (e.g., expresses PAK or a PAK downstream element on the cell surface) to elicit an antibody response against the antigen in the m~mm~l ~lt~.rn~tively, tissue or a whole organ which expresses the antigen can be used to elicit antibodies. The progress of immuni7~tion can be monitored by detection of antibody titers in plasma or serum. Standard ELISA or other immnn~assay can be used with the antigen to assess the levels of antibodies. Following i~ i7~tion, antisera can be obtained and, if desired, polyclonal antibodies isolated from the sera. To produce monoclonal antibodies, antibody producing cells (lymphocytes) can be harvested from an immuni7ed animal and fused with myeloma cells by standard somatic cell fusion procedures thus immortalizing these cells and yielding hybridoma cells. Such techniques are well known in the art. For example, the hybridoma technique originally developed by Kohler and Milstein (1975.
Nature 256:495-497) as well as other techniques such as the hu~nan B-cell hybridoma technique (Kozbar et al. 1983. Immunol. Today 4:72), and the EBV-hybridoma technique to produce human monoclonal antibodies (Cole et al. 1985. Monoclonal Antibodies in Cancer Therapy Allen R. Bliss, Inc., pages 77-96) can be used. Hybridoma cells can be screened immunochemically for production of antibodies specifically reactive with the antigen and monoclonal antibodies isolated.
Another method of generating specific antibodies, or antibody fragments, reactive against PAK or a PAK downstream element is to screen ~ ession libraries encodingimmunt~globulin genes, or portions thereof, ~ c;ssed in bacteria with the antigen (or a portion thereof). For example, complete Fab fragments, VH regions, Fy regions and single chain antibodies can be expressed in bacteria using phage t;~ression libraries. See for example Ward et al. 1989. Nature 341:544-546; Huse et al. 1989. Science 246:1275-1281;
and McCafferty et al. 1990. Nature 348:552-554. ~ltern~tively, the SCID-hu mouse can be used to produce antibodies, or fragments thereof.

As used herein, the term "antibody" is intended to include fragments thereof which retain a desired functional property, e.g., the ability to inhibit an interaction between PAK
and a PAK downstream element. Antibodies can be fragmented using conventional techniques and the fragments screened for utility in the same manner as described above for whole antibodies. For example, F(ab')2 fragments can be generated by treating antibody with pepsin. The reslllting F(ab')2 fragment can be treated to reduce disulfide bridges to produce Fab' fragments.
The term "antibody" is further intended to include derivatives thereof which retain a desired functional property, e.g., the ability to inhibit an interaction between PAK and a PAK dowllsllealll element. Antibody derivatives include chimeric molecules, hllm~ni7ed molecules, molecules with reduced effector functions and bispecific molecules. An antibody, or fragment thereof, produced in a non-human subject can be recognized to varying degrees as foreign when the antibody is ~flmini.~tered to a human subject and an immune response against the antibody may be generated in the subject. One approach for minimi7ing or elimin~ting this problem is to produce chimeric or hum~ni7ed antibody derivatives, i.e., antibody molecules comprising portions which are derived from non-human antibodies and portions which are derived from human antibodies. Chimeric antibody molecules can include, for example, the variable region from an antibody of a mouse, rat, or other species, with human constant regions. A variety of approaches for m~king chimeric antibodies have been described. See, for example, Morrison et al. 1985.
Proc. Natl. Acad. Sci. U.S.A. 81:6851; Takeda et al. 1985. Nature 314:452: Cabilly et al.
U.S. Patent No. 4,816,567; Boss et al. U.S. Patent No. 4,816,397; Tanaguchi et al.
European Patent Publication EP171496; European Patent Publication 0173494, United Kingdom Patent GB 2177096B. In a further modification, hum~ni7ed antibodies have only the hypervariable domains of the variable region of non-human origin and have other parts of the variable region of the antibody, especially the conserved framework regions of the antigen-binding domain, of human origin. Such hllm~ni7ed antibodies can be made by any of several techniques known in the art, (e.g., Teng et al. 1983. Proc. Natl. Acad. Sci. U.S.A.
80: 7308-7312; Kozbor et al. 1983. Immunology Today 4:727; Olsson et al. 1982. Meth.
Enzymol. 92:3-16), and are preferably made according to the teachings of PCT Publication 3 or EP 0239400. Hum~ni~ed antibodies can be commercially produced by, for example, Scotgen Limited, 2 Holly Road, Twickenham, Middlesex, Great Britain.
For use therapeutically, it is also preferred that an antibody prep~lion be unable to fix complement or induce other effector functions. Complement fixation can be pl~ven~ed by deletion of the Fc portion of the antibody, by using an antibody isotype which is not capable of fixing complement, or, less preferably, by using a complement fixing antibody in conjunction with a drug which inhibits complement fixation. ~lt~rn~tively, amino acid residues within the Fc region which are important for activating complement (see e.g., Tan et al. 1990. Proc. Natl. Acad. Sci. USA 87:162-166; Duncan et al. 1988. Nature 332:738-740) can be mut~tecl to reduce or elimin~te the complement-activating ability of an intact antibody. Likewise, amino acids residues within the Fc region which are important for binding ofthe Fc region to Fc receptors (see e.g., Canfield, S.M. et al. 1991. J. Exp. Med.
173:1483-1491; Lund, J. et al. 1991. J. Immunol. 147:2657-2662) can also be m~ltated to reduce or elimin~te Fc receptor binding if an intact antibody is to be used.
An antibody which binds to PAK or a PAK downstream element can be assessed for blocking or inhibitory activity by conventional techniques. For example, the ability of the antibody to block phosphorylation of caldesmon can be determined.
PAK potentiators are those agents capable of potenti~ting the activity of a PAK
kinase. PAK mod~ tin~ agents including inhibitors and potentiators can be readily identified using the screening methods described below and adapted to therapeutic uses.
PAK modlll~ting agents can be used alone or in conjunction with other agents which effect PAK kinase activity. For example, PAK mod~ ting agents can be used with calcium to further enhance or increase contraction of a smooth or cardiac muscle, e.g., involved in states desired to be treated. In some embodiments, the smooth or cardiac muscle contraction involved in the method is considered to be calcium sensitive and the PAK
modlll~ting agent(s) of the present invention enhance the calcium's ability to induce or modulate smooth or cardiac muscle contraction, e.g., the enhancement can be observed by det~ i l-g the force per concentration of calcium.
Another aspect of the invention pertains to a method of treating a subject having a state characterized by smooth muscle contraction. The method involves arlmini~tering to a subject a therapeutically effective amount of a PAK mod~ ting agent, e.g., a PAK inhibitor, such that treatment of the state characterized by smooth muscle conkaction occurs. In one embodiment, the state is characterized by the contraction of smooth muscle having a high basal tone. In another embodiment, the state characterized by the contraction of smooth muscle involves a state characterized by abnormal or inappropliate contraction of smooth muscle. In yet another embodiment, the state characterized by the contraction of smooth muscle involves abnormal or ina~plopliate relaxation of smooth muscle. In yet another embodiment, the treatment of the state involves the reduction or inhibition of inappropl;ate smooth muscle contraction. Examples of states include hypertension, asthma, irritable bowel syndrome, incontinence, and menstrual cramps.
Another aspect of the invention pertains to a method of treating a subject having a state characterized by cardiac contractile dysfunction. For example, the state characterized by cardiac contractile dysfunction can be a state having decreased cardiac contraction in response to calcium (decreased calcium-s~nsilivily). The method involves ~-lmini.stering to a subject a therapeutically effective amount of a PAK modnl~ting agent, e.g., a PAK
stim~ tor, such that increased calcium sellsilivily occurs. As well, the method also provides for the ~(1mini.~tration of PAK or the catalytic domain of PAK directly to heart muscle. In another embodiment, the state is characterized by increased cardiac contraction in response to calcium (increased calcium-sellsilivily). The method involves ~drninistl~ring to a subject a therapeutically effective amount of a PAK modnl~tin~ agent, e.g., a PAK
inhibitor, such that decreased calcium-sensitivity occurs. In yet another embodiment, the state characterized by the contraction of cardiac muscle involves abnormal or inapplopl;ate contraction or relaxation of cardiac muscle. In yet another embodiment, the treatment of the state involves the modulation of inappr~pl;ate cardiac contraction or relaxation. Examples of characterized by cardiac contractile dysfunction include myocardial stllnning, myocardial infarction, myocardial myopathies and heart failure. Clinical conditions which require inotropic intervention can use modulation of PAK and its effect on Ca-sensitivity of cardiac muscle.
As used herein, the term "state" is art recognized and includes a disorder, disease or state characterized by the contraction of smooth or cardiac muscle.

As used herein, the term "hypertension" is art recognized and includes the state in which excessive smooth muscle contraction of a blood vessel occurs which results in hypertension in a subject.
As used herein, the term "asthma" is art recognized and includes the state in which excessive smooth muscle contraction of the airways in the lungs of a subject occurs.
As used herein, the term "incontinence" is art recognized and includes the state in which excessive smooth muscle contraction of the urinary tract occurs.
As used herein, the term "irritable bowel syndrome" is art recognized and includes the state in which excessive smooth muscle contraction of the gastro-intestinal tract occurs.
As used herein, the term "menstrual cramps" is art recognized and includes the state in which excessive smooth muscle contraction of the uterus occurs.
As used herein, the term "cardiac contractile dysfunction" is art recognized andincludes the states in which myocardial contractility is either insufficient or excessive.
For the purposes of this invention, the term "therapeutically effective amount"
includes an amount effective, at dosages and for periods of time llecess~ y, to achieve the desired level of smooth or cardiac muscle contraction. A therapeutically effective amount of a PAK mod~ ting agent, e.g., a PAK inhibitor, may vary according to factors such as the disease state, age, and weight of the individual, and the ability of the PAK mod~ ting agent to elicit a desired level of muscle contraction or calcium sensilivily in the subject. Dosage regimens may be adjusted to provide the o~hllu~ therapeutic response. A the,~;ulically effective amount is also one in which any toxic or detrimental effects of the PAK
mod~ ting agent are outweighed by the therapeutically beneficial effects. It is to be noted that dosage values may vary with the severity of the state to be alleviated. It is to be further understood that for any particular subject, specific dosage regimens should be adjusted over time according to the individual need and the professional judgment of the person ~lmini~tloring or supervising the atlmini~tration of the PAK mod~ ting agent.
A PAK mod~ ting agent, e.g., PAK inhibitor, can be ~rlmini~tered to a subject by a variety of methods known in the art. The PAK modnl~tin~ agent can be provided in a manner such that it can be taken up by the cell or in a manner such that it can be converted to a form that can be readily taken up by the cell. In various embodiments, the PAK mod~ ting agent is a-1minict~red in a formulation suitable for inll~vellous, intraperitoneal, subcutaneous, intl~luscular, intravaginal, topical, transdermal or oral atlminictration. In a prefe,l~d embodiment, the PAK mod~ ting agent is a-lmini.ctered in a time release formulation (also referred to as a sustained-release formulation), for example in a composition which includes a slow release polymer, or a composition suitable for depot injection. The PAK mo~ ting agent can be prepared with carriers that will protect the inhibitor against rapid release, such as a controlled release formulation, including implants, tr~ncderm~l patches, and microencapsulated delivery systems. Biodegradable, biocompatible polymers can be used, such as ethylene vinyl acetate, polyanhydrides, polyglycolic acid, collagen, polyorthoesters, polylactic acid and polylactic, polyglycolic copolymers (PLG). Many methods for the preparation of such formulations are patented or generally known to those skilled in the art.
See, e.g, Sustained and Controlled Release Drug Delivery Systems, J.R. Robinson, ed., Marcel Dekker, Inc., New York, 1978. Particularly plefelled formulations include controlled-release compositions such as are known in the art for the a-lmini.ctration of leuprolide, e.g, microcapsules (U.S. Patents 4,652,441 and 4,917,893), injectable formulations (U.S. Patent 4,849,228), lactic acid-glycolic acid copolymers useful in making microcapsules or injectable formulations (U.S. Patents 4,677,191 and 4,728,721), and sustained-release compositions for water-soluble polypeptides (U.S. Patent 4,675,189).
When al)plopl;ately formulated, a PAK mod~ ting agent may be orally a-lminictered, for example, with an inert diluent or an ~ccimil~ble edible carrier. The PAK
modnl~ting agent may also be enclosed in a hard or soft shell gelatin capsule, col,~lessed into tablets, or incorporated directly into the subject's diet. For oral therapeutic a-lmini.ctration, the PAK mod~ ting agent may be incorporated with excipients and used in the form of ingestible tablets, buccal tablets, troches, capsules, elixirs, suspensions, syrups, wafers, and the like. The percentage of the PAK modlll~ting agent in the compositions and preparations may, of course, be varied. The amount of the PAK modlllating agent in such therapeutically useful compositions is such that a suitable dosage will be obtained.
To aflminicter a PAK modlll~ting agent by other than parenteral a-lminictration, it may be necessary to coat the compound with, or co-a~minicter the compound with, a m~t~ri~l to prevent its inactivation. For example, the PAK modlll~ting agent may be a-lmini~tered to a subject in an a~plopliate carrier, for example, liposomes, or a diluent.
Pharmaceutically acceptable diluents include saline and aqueous buffer solutions.
Liposomes include water-in-oil-in-water CGF emulsions as well as conventional liposomes (Strejan et al. 1984. J. Neuroimmunol. 7:27). Pharmaceutically acceptable carriers include sterile aqueous solutions or dispersions and sterile powders for the extemporaneous preparation of sterile injectable solutions or dispersion. The use of such media and agents for ph~rrn~seutically active substances is well known in the art. Except insofar as any conventional media or agent is inco~ )alible with the active compound, use thereof in the pharmaceutical compositions of the invention is contemplated. Supplementary active compounds can also be incorporated into the compositions.
Therapeutic compositions typically must be sterile and stable under the conditions of m~nllf~cture and storage. The composition can be formulated as a solution, microemulsion, liposome, or other ordered structure suitable to high drug concentration. The carrier can be a solvent or dispersion medium co~ t~g~ for example, water, ethanol, polyol (for example, glycerol, propylene glycol, and liquid polyethylene glycol, and the like), and suitable lures thereof. The proper fluidity can be m~int~ined, for example, by the use of a coating such as lecithin, by the m~int~n~nce of the required particle size in the case of dispersion and by the use of surfactants. In many cases, it will be prerel~ble to include isotonic agents, for example, sugars, polyalcohols such as mannitol, sorbitol, or sodium chloride in the composition. Prolonged absorption of the injectable compositions can be brought about by including in the composition an agent which delays absorption, for example, monostearate salts and gelatin.
Sterile injectable solutions can be prepared by incorporating the PAK mo~ tin~
agent in the required amount in an al,plop~iate solvent with one or a combination of ingredients enumerated above, as required, followed by filtered sterili7~tion. Generally, dispersions are prepared by incorporating the PAK modnl~ting agent into a sterile vehicle which contaills a basic dispersion medium and the required other ingredients from those enumerated above. In the case of sterile powders for the preparation of sterile injectable solutions, the plefelled methods of preparation are vacuum drying and freeze-drying which yields a powder of the active ingredient plus any additional desired ingredient from a previously sterile-filtered solution thereof.
Dosage regimens may be adjusted to provide the optimum therapeutic response. Forexample, a single bolus may be ~-lmini~tered, several divided doses may be ~llmini~tered over time or the dose may be proportionally reduced or increased as indicated by the exigencies ofthe thel~;ulic situation. It is especially advantageous to form~ te parenteral compositions in dosage unit form for ease of ~(lmini~tration and uniformity of dosage.
Dosage unit form as used herein refers to physically discrete units suited as unitary dosages for the m~mm~ subjects to be treated; each unit collL~ g a predetermined quantity of active compound calculated to produce the desired the~culic effect in association with the required pharmaceutical carrier. The specification for the dosage unit forms of the invention are dictated by and directly dependent on (a) the unique char~ct~i.ctics of the PAK
mod~ ting agent and the particular therapeutic effect to be achieved, and (b) the limitations inherent in the art of compounding such a PAK mod~ ting agent for the treatment of states describedherein.
In yet another embodiment, the invention provides a method for identifying PAK
mod~ ting agents. For example, mod~ ting agents which are inhibitors, or alternatively, stimlll~tQrs~ of PAK kinase activity can be identified. The method involves providing an indicator composition comprising a PAK protein having PAK kinase activity, contacting the indicator composition with a test agent, and determinin~ the effect of the test agent on PAK
kinase activity in the indicator composition to thereby identify a compound that modulates the kinase activity of a PAK protein. In a pler~;lled embodiment, the screening assay identifies agents that modulate the kinase activity of a PAK protein, for example, peptides or peptide mimetics. A statistically significant change, such as a decrease or an increase, in the level of PAK kinase activity in the presence of the test agent (relative to what is detected in the absence of the test agent) is indicative of the test agent being a PAK mod~ ting agent.
Morgan and co-workers (Morgan et al. 1989. In: Annual Reports in Medicinal Chemistry. Ed.: Vinick, F.J. Academic Press, San Diego, CA, pp. 243-252.) define peptide mimetics as "structures which serve as appropl;ate substitutes for peptides in interactions with receptors and enzymes. The mimetic must possess not only affinity but also efficacy and substrate function. As used herein, the term 'peptide mimetic" is used in a manner con.ci.ct~nt with the above art-recognized definition. That is, a peptide mimetic exhibits function(s) of a particular peptide, without restriction of structure. Peptide mimetics of the invention, e.g., analogues of structural motifs such as the PAK autophosphorylation site or the substrate binding site of PAK, may include amino acid residues or other chemical moieties which provide the desired functional characteristics.
As used herein, the term "indicator composition" is int~ndecl to include any composition that can be used to screen and identify PAK modnl~ting agents. The indicator composition can be, for example, a smooth muscle cell, a cardiac myocyte, a smooth muscle cell or cardiac muscle cell exkact, or a detergent-skinned smooth or cardiac muscle fiber bundle system. Methods for the preparation of intact smooth muscle cells or extracts from such cells are well known in the art and previously described (Glukhova et al. 1987. Tissue Cell 19:657-63; Childs et al. 1992. J. Biol. Chem. 267:22853-9, 1992; White et al. 1996. J.
Biol.Chem. 271:15008-17). Methodsforple~ gTriton-skinnedsmoothandcardiac muscle fiber bundles are also known in the art (Strauss et al. 1992. Am. J. Physiol.
262:1437-45; Van Eyk, J.E. et al. 1998. Circ. Res. 82:261-271). Bovine cardiac and rabbit skeletal troponin can be purified as outlined in Ingraham R.H. et al., 1988. Biochemistry 27:5891-5898.
As used herein, the language "modulates PAK kinase activity" is art-recognized and is inten(led to include the capacity to inhibit or stimulate PAK kinase activity, e.g., the ability of the PAK kinase to phosphorylate its substrates. The modulation can be complete inhibitor or partial inhibition. The modulation of PAK kinase activity include, modulation to the extent necessary or sufficient to treat the states described herein.
As used herein the term "test agent" is intt~.nded to include an agent that modulates the kinase activity of PAK. Such agents can be, for example, a drug, an antibiotic, an enzyme, a chemical compound, a ~ e of chemical compounds, a biological macromolecule, and analogues thereof.
As used herein the term "PAK protein" is inten~1ed to include a protein that belongs in the family of PAK serine/threonine protein kinases and has the ability to induce smooth muscle contraction, e.g., in the absence of calcium, or to change the calcium sensitivity of cardiac muscle. These include m~mm~ n isoforms identified, e.g., PAKl, PAK2, PAK3, or lower eucaryotic isoforms, such as the yeast Ste20 (Leberter et al., EMBO J. 11 :4805- 13, 1992) or the Dictyostelium single-headed myosin I heavy chain kinases (Wu et al., J. Biol.
Chem. 271:31787-90, 1996).
In one embodiment, PAK kinase activity is assessed by measuring phosphorylation of the caldesmon protein. In another embodiment, PAK kinase activity is assessed by measuring phosphorylation of the myosin light chain (LC20). In still a further embodiment, PAK kinase activity is ~sessed by me~ ring phosphorylation of the calponin protein. In further embodiments, PAK kinase activity is ~.sessed by me~llring phosphorylation of TnI
or any target protein in cardiac muscle. Several methods are known in the art and readily available for determinin~ the activity of candidate PAK inhibitors or stimulators against PAK kinase on intact cells, cell extracts or skinned smooth muscle and cardiac muscle fibers. Phosphorylation of the PAK kinase cellular substrates (e.g., caldesmon, myosin light chain (LC20), calponin, TnI) can be measured using antiphosphoserine antibodies or phosphopeptide fing~,ylhlt~, methods well known in the art and described in the "Examples" section.
In many drug screening programs which test libraries of mod~ tinf~ agents and natural extracts, high throughput assays are desirable in order to m~ximi7e the number of modlll~ting agents surveyed in a given period of time. Assays which are performed in cell-free systems, such as may be derived with smooth muscle and cardiac muscle cell extracts, are often ~.,e~lled as "primary" screens in that they can be generated to permit rapid development and relatively easy detection of an alteration in a molecular target which is mediated by a test modnl~ting agent. Moreover, the effects of cellular toxicity and/or bioavailability of the test mocllll~ting agent can be generally ignored in the in vitro system, the assay instead being focussed primarily on the effect of the test agent on the PAK kinase as may be manifest in an alteration of phosphorylation of downstream elements.
The efficacy of the test agent can be assessed by generating dose response curves from data obtained using various concentrations of the test mod~ ting agent. Moreover, a control assay can also be performed to provide a baseline for comparison. In the control assay, the indicator composition comprising the PAK kinase is incubated in the absence of a test agent.
The present invention is further illustrated by the following example, which should not be construed as limiting. The contents of all references, pending patent applications, and published patents, cited throughout this application (including the "Background"
Section) are hereby expressly incorporated by reference.

F.~ ~le Methods Protein Plepa,alions Intact smooth muscle myosin, LC20, myosin light chain kinase (MLCK) and caldesmon were purified from chicken gizzard (Hathaway, D.R. et al. 1983. Anal. Biochem.
135:37-43; Ikebe, M. et al. 1985. J. Biol. Chem. 260:10027-10031; Bretscher, A. 1984. J.
Biol. Chem. 259:12873-12880) while PAK3 was isolated from rat brain (Wu, C. et al. 1995.
J. Biol. Chem. 270:25070-25078). Recombinant caldesmon fragments (CAD39 and CAD40) were prepared and purified as previously described (Novy, R.E. et al. 1993. Cell Motil. Cytosk. 26:248-261). Recombinant Cdc42 and Racl were expressed and purified as described in Manser, E. et al. (1994. Nature 367:40-46). Plasmid pGST-mPAK3 encoding mouse fibroblast mPAK3 fused to GST in pGEX-KG was expressed and the fusion protein purified according to Manser, E. et al. (1994. Nature 367:40-46). Plasmid pDK-mPAK3K297R encoding an inactive mPAK3 mutant with lysine amino acid residue 297 was mllt~ted to arginine 297 (mPAK3~297R cDNA) and fused to GST by subcloning a BamHl fragment of PDK-mPAK3K297R into pGEX-4T3. The GST-mPAK3 fusion proteins were expressed in E. coli. The catalytic subunit of recombinant ROK (Rho-kinase) was ~y~essed as a GST fusion protein in baculovirus (GST-ROK; 5). All recombinant kinases were purified on a glutathione-Sepharose affinity column (Amersham Pharmacia), concentrated (Amicon) and dialyzed against 10 mM imidazole, pH 7Ø As reported previously (Bagrodia, S. et al. 1995. J. Biol. Chem. 270:27995-27998), GST-mPAK3 is susceptible to degradation, leading to different activities for each preparation. In order to ensure cl n.ci~tency, activities ofthe various GST-rnPAK3 preparations were standardized against CA 0224l389 l998-06-23 myelin basic protein phosphorylation. GST-mPAK was used in the skinned fiber assay within 4 days but could m~int~in sufficient activity for in vitro phosphorylation analysis if stored frozen in 50% glycerol. Various preparations of GST-mPAK3 (~0.5 to 5 ~Lg/ml of active GST-mPAK3) were able to induce Ca2+-independent contractions in skinned smooth muscle fibers ranging from 26.1 to 80.9% of CaZ+-dependent contraction.

Preparation of Deter~ent-skinned Smooth Muscle Fibers Triton-X100 skinned fibers from adult guinea pig taenia coli were prepared as previously described (Strauss, J.D. et al. 1992. Am. J. Physiol. 262:C1437-1445; Strauss, J.D. et al. 1993. Methods: A Companion to Methods in Enzymology 5:281-290). The skinned fiber bundles were stored at -20~C in a solution co~ g 50% glycerol (v/v), 4 mM EGTA, 1 rnM sodium a_ide, 7.5 mM ATP, and 20 mM imi(l~7ole, pH 6.7 and used within one month. Thin fiber bundles (~100 llm in diameter) were mounted on an AME
801, Sensonor, Norway) force tr~n~dllc~r for analysis (26,27). "Relaxing" solution eonsisted of 10 mM m~gn~sium chloride, 1 mM sodium a_ide, 7.5 mM disodium ATP, 4mM EGTA, and 20 mM imidazole (pH 6.7), 10 rnM sodium phosphocreatine and 10 U/mlereatine phosphokinase (ionie strength of 110 mM, 2mM free Mg2+, and 7.2 mM MgATP).
"Contraeting" solution consisted of relaxing solution supplemented to 0.1 mM calmodulin and 4.0 mM ealeium ehloride (pCa = 4.3). The solutions used to bathe the fibers eontained either GST-mPAK3, GST-mPAK3K29'R, GST-ROK or an equivalent amount of 10 mM
imidazole buffer used in the final dialysis during p~ lion of PAK and ROK.

Mechanical Measurements of Smooth Muscle Small smooth muscle fiber bundles (5.0 mm in length and 100-200 llm in diameter)were dissected from the muscle strip and mounted for isometric force measurements as previously described (Strauss, et al., 1992). To induce contraction, the strips were incubated in the contracting buffer co.~ g Ca2~(pCa = 4.5) and 1 ~lM calmodulin. Fibers generated between 0.5 and 2.5 milliNewtons (mN) of force within 2 to 10 minutes (1.5 x 105N/M2).
To study the effect of exogenous GST-mPAK3 and GST-mPAK3K29'R on force generation of the skinned fibers, ~ 5 ~lg/ml (55 nM) of each protein was included in the relaxing buffer.
To demonstrate the phosphorylation of target protein in skinned smooth muscle fibers with GST-mPAK3, skinned fibers were incubated in the relaxation buffer COIlL~ either GST-mPAK3 or GST-mPAK3~29'R and 1 mM ~ 32P[ATP] (1 mCi per ml). After 60 min of incubation at 25~C, the fiber was washed with the relaxation buffer to remove ATP. The washed fiber was analyzed by 2-dimensional isoelectric focusing/SDS-PAGE as described (Nixon, G.F. et al. 1995. J. Physiol. 487:283-289), lD SDS-PAGE, or lD isoelectric focusing on a gel (Strauss, J.D. et al. 1992). Subsequently, autoradiography andcorresponding western blot analyses indicated that c~lde~mon and desmin were phosphorylated.

Cardiac Skinned Fiber Bundle Experiments Hearts from rat and dissected trabecula were placed in ice cold relaxing buffer co~-t~ -g 0.1 mmoVl EGTA, 2 mmoVl Mg2+, 79.2 mmoVl potassium chloride, 5.0 mmoVlMgATP2-, 12 mmoVl creatine phosphate, and 20 mmoVl MOPS, pH 7.0 (ionic strength, 150 mmoVl), plus a protease inhibitor cocktail (50 ~moVl phenylmethylsulfonyl fluoride, 3.6 ~moVl leupeptin, and 2.1,umoVl pepstatin A). Fiber bundles (-100,um in diameter) from the trabeculae were glued to a force tr~n~dllcer (AME 801, Sensonor, Norway) at one end and to a fixed post attached to a mic~ lator. The fibers were skinned in relaxing buffer collL~ g 10 IU/ml creatine kinase and 1% Triton X-100 for 30 to 45 mimltes The fibers were transferred to relaxing buffer co,lL~ g 10 IU/ml creatine kinase, and the sarcomere lengths were set at 2.2,um on the basis of the laser diffraction pattern. Isometric pCa-force relations were determined by bathing the skinned fiber bundles sequentially in solutions (10 mmoVl EGTA, 2 mmoVl Mg2+, 79.2 mmoVl pola~,siu.ll chloride, 5.0 mmoVl MgATP2-, 12 mmoVl creatine phosph~te, 10 IU/ml creatine kinase, and 20 mmoVl MOPS, pH 7.0 [ionic strength, 150 mmoVl]) that contained increasing concentrations of calcium chloride to achieve pCa values from 8.0 to 4.5. Fibers were contracted at pCa 4.3 (m;lxi~ response), 5.5, or 6.25 (subm~xim~l calcium concentration) and relaxed several times, then placed in a solution corlL~;Ili~lg ~ 5,~g/ml GST-mPAK3 ( ~ 55 nM) at various calcium concentrations.

Gel Overlay Assay Gel overlay assay was performed as outlined in Manser, E. et al. 1996 (Methods in Ezymology 256:215-227). Intact and ~l~inne~ smooth muscle (ileum, aorta) and cardiac muscle samples were analyzed by 10% SDS- PAGE supplemented with 10% glycerol, 5 mM m~gn~ium chloride and 1 mM dithiothreitol. Following transfer to nitrocellulose the proteins were denatured by incubating in a solution of 6M guanidine HCl, 50 mM zinc chloride, SmM magnesium chloride, 25 mM MES pH 6.5 and 0.05% Triton X-100 (30 min at 4~ C), then renatured by incubating three times with 50 mM sodium chloride, 2.5 mM
dithiothreitol, 25 mM MES, pH 6.5, 1.25 mM m~gnesium chloride, 50 rnM zinc chloride, 1% bovine serum albumin and 0.05% Triton X-100 (2 hours at 4~C). Purified recombinant human Cdc42 and Racl were labelled with S35-GTPyS prior to probing of the nitrocellulose blots.

Protein Phosphorylation In vitro phosphorylation of intact smooth, skeletal, cardiac, platelet, and Dictyostelium myosin and isolated LC20, caldesmon, and caldesmon fragme~tc, and cardiac and skeletal Tn were carried out at 25~C in 10 rnM Tris, pH 7.0, 50 mM sodium chloride, lmM [y32P]ATP (5 x 105 cpm per nmol) or at 37~C in the same buffer except co~ g 150 mM sodium chloride. Aliquots of the reaction llliXlU~ were analyzed for protein phosphorylation using Whatman P81 paper and SDS-gel electrophoresis/autoradiography as previously described (Childs, T.J. et al. 1992. J. Biol. Chem. 267:22853-22859). In the case of cardiac and skeletal Tn, the three Tn subunits (TnI, TnT, and Tn~C) were separated by reversed phase HPLC. The individual subunits were analyzed for incorporation of 32p (i.e., phosphorylation by PAK) by SDS-PAGE and autoradiography and direct counting.
Phosphorylated amino acids were identified by thin-layer electrophoresis after partial hydrolysis of the phosphorylated proteins in 6 N hydrochloric acid as previously described (Childs, T.J. et al. 1992. J. Biol. Chem. 267:22853-22859; Mak, A.S. et al. 1991. J. Biol.

Chem. 266:19971-19975; Mak, A.S. et al. 1991. J. Biol. Chem. 266:6678-6681). Two-dimensional tryptic peptide maps were produced as previously described (Childs, T.J. et al.
1992. J. Biol. Chem. 267:22853-22859; Mak, A.S. et al. 1991. J. Biol. Chem. 266:19971-19975; Mak, A.S. et al. 1991. J. Biol. Chem. 266:6678-6681).
s GST-mPAK3 and GST-ROK Phosphorylation Assay in Skinned Smooth Muscle Fibers Smooth muscle skinned fibers mounted on a "U-shaped" pin were incubated at 25~C
under various conditions using the same conditions as in the skinned fiber assays, except, when required, assay buffers contained lmM [~32P]-ATP (0.25 mCi per ml) instead of 7.2 mM ATP. After 90 min incubation, fibers were submerged in ice cold 15% trichloroacetic acid and 2mM inorganic phosphate followed by acetone, to inactivate the kinase/phosph~t~es This ensures preseNation of the phosphorylation levels. Fibers were stored at -20~C until analysis. When required, autoradiographs were prepared from the same blots used for western blotting. Phosphoimages of the blots were developed using the Storm phosphorimager 820 (Molecular Dynamics, Sunnyvale, CA) or directly on film (X-Omat Blue XB-l, Kodak).

Actin-Tropomyosin Bindin~ Assay To quantitate the amount of smooth muscle caldesmon and smooth muscle ll~o-nyosin bound to the skeletal actin thin filament (> 95% amino acid homology to smooth muscle actin), centrifugation studies were carried out in binding buffer con~i~ting of 20 mM Tris-HC1 pH 7.2, 10 mM KCl, 5 mM MgCl2 and 1 mM DTT. The concentration of actin was 5 nmole/200 ~ 25 ~lM) and the mole ratio of actin to tropomyosin was 7:2.
Increasing quantities of unphosphorylated or phosphorylated caldesmon were added to the actin-tropomyosin l~ u~e. Protein concentrations were fletermined by the Lowry method or by absorbance at 280 nm. Samples were spun for 30 min at approximately 95,000 rpm, resulting in the pelleting of over 95% of the actin. The obtained pellets were dissolved in 100 ~11 of 0.05% aqueous trifluroacetic acid (TFA) and injected on a Zorbax C8 SB300 reversed-phase column (4.6 mm x 250 mm) on a Varian HPLC system (Star series). The various proteins were eluted using a 2% B/min linear A/B gradient with a 5 min initial isocratic hold, where solvent A is 0.05% aqueous TFA and solvent B is 0.05% TFA in acetonitrile. The flow rate was 1 mVmin. The peak areas were detçrmined at 210 nm and converted to nanomoles using a standard curve obtained for each protein. The amount of protein pelleted in the absence of actin was subtracted from the amount pelleted in the presence of actin. The experiments were done in triplicate and the standard deviation calculated.

Actin-Tropomyosin Activated ATPase Activity of Smooth Muscle Subfragment 1 (Sl) Reactions consisted of 10 ~IM actin, 2 IlM tropomyosin, 0.5 ~lM Sl, and variableconcentrations of unphosphorylated and phosphorylated c~l-lesmon. The buffer contained 40 mM Tris-HCI pH 7.8, 50 mM KCI, 5 mM MgCl2, 1 mM DTT at 37~C. Reactions were performed in a 96 well microtitre plate, triggered by the addition of 4 mM ATP. The phosphate assay was performed according to the method of Chifflet, S. et al. 1988. (Anal.
Biochem. 168:1-4). Briefly, reactions were termin~ted by the addition of an equal volume (100 )lL) of reagent which contained 6% (w/v) SDS, 3% (w/v) ascorbic acid, lM HCI, 0.5%
(w/v) ammonium molybdate. The mixture was incubated for 5 mimltes prior to the addition of a second reagent which contained 2% sodium citrate, 2% sodium m-arsenite, and 2%
acetic acid. Blue colour developed quickly over 5 min1ltes Microtitre plates were read at 650 nm on an E-max ELISA plate reader (Molecular Dynamics, Sunnyvale CA). Phosphate was determined by comr~ring experimental values to standard phosphate values obtained through dilution of phosphate-buffered saline (PBS).

Western Blot Analyses 10 and 12% SDS-PAGE and western blot analysis were carried out as described (Van Eyk, J. et al. 1998. Cir. Res. 82:261-271). Preparations of skinned muscle fibers as described above were homogenized in sample buffer prior to analysis. Detection of the PAK homologue in smooth muscle was achieved using an antibody raised against a synthetic peptide corresponding to 13 residues at the N-t~rmin~l end of mouse fibroblast mPAK3 (NT3 MAb, gift from S. Pelech, Kinetek Inc., Vancouver, Canada, dilution 1 :200).

Caldesmon, desmin and LC20 were detected using the following antibodies: clone hHCD
(dilution 1 :2000, Sigma), clone DEU-10 (dilution 1: 100, Sigma) and clone MY-21 (dilution 1 :200, Sigma), respectively. For quantification of LC20 phosphorylation, skinnçd muscle fibers (2-3 fibers/lane) were subjected to one-~limen~ional isoelectric focusing (Strauss, J.D.
et al. 1992. Am. J. Physiol. 262:C1437-1445), separating unphosphorylated LC20 from mono- and di-phosphorylated LC20. The ratio of phosphorylated to total LC20 was quantified by densitometry (Sigma Gel).
Vi~ tion of blots was carried out as follows. After the nitrocellulose membrane (blot) was blocked, it was washed several times in PBS-T to remove traces of blocking solution. The blot was then probed with primary antibody and it was rocked slowly for one hour. Antibody was prepared in PBS-T with 0.5% blocking solution (Boehringer Marmheim chemil~lminescence blotting substrate POD) and an antibody concentration of between 1 to 5 ug/ml was used. Primary antibody was removed and the blot was washed several times in PBS-T. Next, a secondary antibody was added and the solution rocked slowly for one hour. Secondary antibody was prepared in a similar fashion as was the primary antibody. Secondary antibody was then removed and the blot washed several times in PBS-T. After the final wash, 0.1 M Tris-HCI, pH 9.5 was added to the blot and it was rocked slowly for 5 min. The Tris-HCI was removed and 1.0 ml of chemilllminescence reagent (CDP-STAR ch~mihlminescence reagent, NEN Life Science Products) was added to the blot, so as to evenly cover the blot, which was then kept still for 5 min. The blot was removed with forceps and excess chemiluminescence reagent was removed from the blot by wiping on the side of the container. After the excess chemihlmin~sc~nce reagent was removed, the blot was placed face down on waxed paper and wrapped, ensuring that no air bubbles were present between the blot and the waxed paper. The blot was then exposed to film.

Results Smooth Muscle Overlay assays with Cdc42 and Racl indicate that [3sS]GTP~S-Cdc42 bound to bands of 62 and 65 kDa in extracts of guinea pig taenia coli smooth muscle while [3sS]GTPrS-Racl detected a single band of 62 kDa (Figure lA & B). An antibody raised against the N-tennin~l 13 amino acid residues of mouse fibroblast mPAK3 reacted with a protein of 62 kDa in guinea pig smooth muscle and a protein of the same molecular mass in rat aorta (Figure 1 C). These results indicate that smooth muscle contains one, and possibly two, PAK isoforms (Figure lA). PAK was absent from Triton-skinned smooth muscle fibers (Figure lA & B) suggesting that, like ROK (Kureishi, Y. et al. 1997. J. Biol. Chem.
272:12257-12260), PAK is either a cytoplasmic or membrane-bound enzyme.
Triton-skinned guinea pig taenia coli smooth muscle fibers were in(l~lced to contract in a Ca2+-independent manner when incubated in the presence of recombinant, constitutively active GST-mPAK3 (Figure 2A). The force induced by GST-mPAK3 (~5 llg/ml; 55 nM) in relaxing buffer (pCa <8.0) reached a ms. x i ., ,l..,, level equivalent to 62 + 12% (n = 10) of that achieved by addition of a calcium-co.,ls.;..i..g activation solution (pCa 4.3). Under the same conditions, a kinase-dead PAK mutant, GST-mPAKK297R, was unable to induce force in the absence of Ca2+ (Figure 2B). In previous studies, Ca2+-independent smooth musclecontraction has been in~ cecl through the use of unregulated forms of MLCK (Somlyo, A.P.
et al. 1994. Nature 372:231-236; Stull, J.T. et al. 1991. Hypertension 17:723-732), addition of phosphatase inhibitors (e.g., Strauss, J.D. et al. 1992. Am. J. Physiol. 262:C1437-C1445;
Paul, R.J. et al. 1987. Prog. Clin. and Bio. Res. 245:319-332; Trinlcle-Mulcahy, L. et al.
1995. J. Biol. Chem. 270:18191-18194) or most recently by another Rho-Family GTPase-dependent kinase, ROK (Kureishi, Y. et al. 1997. J. Biol. Chem. 272:12257-12260). In all cases the degree of smooth muscle contraction correlates with an increase in the level of LC20 phosphorylation. In the case of ROK, contraction is promoted by the direct phosphorylation of LC20 on serine 19 (Amano M, et al. 1996. J. Biol. Chem. 271 :20246-20249) in addition to the phosphorylation and inhibition of myosin light chain phosphatase (MLCP) (Kimura, K., et al. 1996. Science 273:245-248). ~
This dual effect of ROK was demonstrated by the use of wortm~nnin which is a potent inhibitor of MLCK but does not affect the activity of either ROK (Kureishi, Y. et al.
1997. J. Biol. Chem. 272:12257-12260) or PAK (Figure 2C). The addition of a con~liluliv~ly active GST-ROK catalytic domain to Triton-skinned smooth muscle fibers produces a wolL~ nnin-sensitive contraction at pCa 6.5 (Ca2+-dependent contraction) as well as a wortm~nnin-insensitive contraction at pCa <8.0 (Ca2+-independent contraction) (Kureishi, Y. et al. 1997. J. Biol. Chem. 272:12257-12260). On the other hand, wortm~nnin at a concentration of 1 mM had little effect on the contraction of smooth muscle induced by GST-mPAK3 at low calcium (Figure 2D), even though this concentration is sufficient to completely inhibit MLCK-dependent contraction at elevated Ca2+ (data not shown). These results demonstrate that the Ca2+-independent contraction promoted by PAK occurs without a requirement for MLCK activity. Furthermore, it seems unlikely that PAK promotes contraction by inhibiting myosin light chain phosphatase (MLCP) since the Ca2+-independent contractions achieved with phosphatase inhibitors are invariably dependent on MLCK activity and are abolished by MLCK inhibitors (e.g., Strauss, J.D. et al. 1992. Am. J.
Physiol. 262:C1437-C1445; Paul, R.J. et al. 1987. Prog Clin and Bio Res. 245:319-332;
Trinkle-Mulcahy, L. et al. 1995. J. BioL Chem. 270:18191-18194). Thus, PAK most likely works by direct phosphorylation of a contractile protein rather than ~lt~rin~ either MLCK or MLCP.
These results prompted an investigation into whether PAK directly phosphorylatesLC20 and thus achieves contraction in a "traditional" manner. In vitro analysis shows that GST-mPAK3 phosphorylates intact chicken gizzard smooth muscle myosin to 2 moles of phosphate per mole (Figure 3A). Phosphate is incorporated only into a single serine residue of LC20 (Figure 3A). Furthermore, MLCK was unable to phosphorylate LC20 following GST-mPAK3 treatment (Figure 3B), indicating that PAK and MLCK both phosphorylateserine 19. Indeed, identical two-dimensional tryptic phosphopeptide maps were obtained from LC20 phosphorylated by either MLCK or GST-mPAK3 (Figure 3C). These results predict that PAK, like MLCK and ROK, promotes smooth muscle force generation by increasing LC20 phosphorylation levels. However, under conditions where GST-mPAK3 induced Triton-skinned smooth muscle fibers to contract with ~70% of the m~im~1 force obtained in the presence of Ca2+, no significant increase in the level of LC20 phosphorylation was observed (Figure 3D). In fact, the level LC20 phosphorylation rem~inecl similar to the level of relaxed fibers (absence of GST-mPAK3 and calcium, Figure 3D lane 4) even as force induced by GST-mPAK3 increased from 26 to 70% (Figure 3D, lanes 1-3). The uncoupling between LC20 phosphorylation and force generation implies that PAK does not direct or indirectly activate myosin, but must employ an alternative and novel mechanism to contract the skinned muscle fibers.
To begin to define the molecular basis of PAK-induced contraction, it is critical to identify the proteins phosphorylated by mPAK3 in the skinned smooth muscle fibers.
Protein substrates for mPAK3 were labeled with 32p under conditions where GST-mPAK3 produces ~70% of m~im~l Ca2+-dependent force (Figure 4). Gel electrophoresis analysis of the proteins labelled during a PAK-in~ ce(l contraction were then performed. Two proteins, with approximate molecular weights of 58 and 145 kDa, were more highly phosphorylated in the presence than the absence of GST-mPAK3 (Figure 4A & C, compare lanes 2 and 3).
The 58 and 145 kDa proteins were identified by western blot analysis as desmin and c~l(lç~mon, respectively. Little if any, phosphorylation of LC20 was detected in the fibers co~ cled by GST-rnPAK3.
GST-ROK, the GST-fusion protein of the constitutively active catalytic domain ofROK (Kureishi, Y. et al. 1997. J. Biol. Chem. 272: 12257-12260), caused Triton-skinned smooth muscle fibers to contract in a Ca2+-independent manner with up to 80% of m~l~im~l force (Kureishi, Y. et al. 1997. J. Biol. Chem. 272: 12257-12260). Under these conditions, the major proteins phosphorylated in the skinned fibers by GST-ROK were LC20, desmin and two proteins of greater than 158 kDa (Figure 4B & D). Clearly, neither of these high molecular weight pl~teills are caldesmon (Figure 4D). One is most likely the catalytic domain of MLCP, which is known to be phosphorylated by ROK in vitro and has an approximate molecular weight of 158 kDa by SDS PAGE.
In a colllp~;son of the protein substrates for ROK and PAK under conditions where GST-ROK and GST-mPAK3 induce similar amount of Ca2+-independent force (79.5 vs 71.1 %, respectively), GST-ROK incorporated more phosphate into LC20 than did GST-mPAK3. Phosphorylation of LC20 by GST-ROK in the absence of calcium is similar to that by MLCK at pCa 4.3 (compale Figure 4B lanes 3 and 4). As well, GST-ROK did not phosphorylate caldesmon, which is the main substrate for GST-mPAK3 (compare Figure 4C
lane 2 and Figure 4D lane 3). In vitro phosphorylation studies confirm that chicken gizzard h-caldesmon is a better substrate for GST-rnPAK3 than for GST-ROK (Figure 4E). GST-mPAK3 phosphorylated hCAD to 2 moles of phosphate per mole of protein. The C-t~rmin~l domain of fibroblast l-caldesmon (corresponding to chicken gizzard caldesmon residues 458-756) is a substrate for GST-mPAK3 (Figure 4F) but no phosphorylation of the N-t.ormin~l c~l-le~mnn domain was observed (data not shown). The C-t~.nnin--s ofcaldesmon contains multiple binding sites for actin, tropomyosin and calmodulin.Caldesmon inhibits the actin-activated Mg2+-ATPase of myosin (review see Chalovich, J.M. 1992. Pharmacol. Ther. 55:95-148) and has been suggested to provide a basal resting inhibition of vascular tone. The force of contraction of Triton-skinned smooth muscle fibers increases upon the partial extraction of c~l-lesmon (Malmqvist, U. et al. 1996.
Pfluger Archiv. 432:241-247) or decreases due to competitive binding of a 20-kDa actin-binding fragment of c~l(le~mon (Pfitzer, G. et al. 1993. Proc. Natl. Acad. Sci. 90:5904-5908). As well, a synthetic peptide of an actin binding region of caldesmon increases force of ~-escin skinned arterial muscle fibers at low concentrations of CaZ+ (Katsuyama, H. et al.
1992. J. Biol. Chem. 267:14555-14558), probably by competing with endogenous c~ldesmon for the actin fil~ment This could potentially revive caltle~mon inhibition. This knowledge, taken together with the results of the present studies, led to a further investigation of whether a reduction in caldesmon interaction with actin would increase force generation resulting in contraction. Thus, the binding assay study described above investigate(l the affinity of unphosphorylated and PAK-phosphorylated caldesmon for the actin thin filament. The results, shown in Figure 5, translate to an approximately two-fold reduction in the affinity of calcle~mon for actin in the unphosphorylated and phosphorylated states. The data indicate that the mech~ni~m of caldesmon inhibition of contraction is not related to the displacement of caldesmon from the actin thin filament, but probably a subtle movement of c~lde~mon on the actin fil~ment.
Even with a small change in actin binding affinity, phosphorylation of caldesmon by PAK has a large effect on caldesmon function. As shown in Figure 6, smooth muscle myosin ATPase activity was inhibited when caldesmon was unphosphorylated, whereas inhibition was clearly released in the presence of PAK-phosphorylated c~lde~mQn. Thus, there is an increase in interaction between actin and myosin resulting in increased ATPase activity. Increased ATPase activity is positively correlated with increased force.

A possible PAK pathway responsible for contraction is shown schematieally in Figure 7.

Cardiac Muscle Studies first investig~ted the presence of PAK in eardiac muscle. In a gel overlay assay, GTP y35S-Cdc42 bound to a ~ 62kDa protein in porcine, dog, and rat eardiae musele.
As shown in Figure 8, this protein eomigrated with brain purified PAK and the PAK
homologue from smooth muscle. The presence of PAK or a PAK homologue in eardiae musele was thus eonfirmed.
The phosphorylation of eardiae musele proteins by PAK and ROK was investigated using isolated eardiae myofibrils and assoeiated proteins. PAK was ineffieient at phosphorylating LC20 when associated with isolated cardiae and skeletal myosin in the presenee of, and hence while phosphorylating, myosin from smooth muscle and platelets (Figure 9A). In contrast, ROK phosphorylated LC20 in all of the various myosins tested (Figure 9C). ROK phosphorylated cardiac LC20 up to 0.5 mol phosphate per mol protein, and was equally effective at phosphorylating LC20 as part of the myofibrils (Figure 10) or as an isolated protein (Figure 9C).
Smooth musele MLCK, a kinase that speeifieally phosphorylates LC20 reslllting insmooth muscle contraction, also phosphorylated cardiac LC20 (Figure 9B), but at a low level of 0.2 mol phosphate per mol protein. The addition of both ROK and MLCK did not inerease the level of phosphorylation of eardiac LC20, indicating that the two kinases phosphorylate the same amino aeid. This is supported by the fact that it is the sarne serine of eardiae LC20 that is phosphorylated by either kinase (serine 16; Figure 11), and the resulting phosphorylated eardiae LC20s produee the same tryptic peptide map (Figure 12).
Skeletal and cardiac muscle is regulated by the troponin (T~n) eomplex whieh is eomposed of TnI, TnT, and TnC. In the absenee of ealeium, TnI binds to actin and inhibits the actin-myosin interaetion, resulting in relaxation. In the presenee of ealeium, TnI binds to TnC resulting in contraction. PAK is able to phosphorylate eardiae, but not skeletal Tn, and the autoradiograph of Figure 13 indieates that TnI is the predominant troponin phosphorylated by PAK in the eardiae Tn eomplex.

Triton X-100 skinned cardiac muscle fibers were bathed in GST-mPAK in the absence (pCa 8.5) and presence of calcium at ma~inlulll (pCa 4.5) as well as at subm~im~l calcium concentrations. As shown in Figure 14A, PAK had no effect in the absence of calcium and little effect (~ 3%) at maximum concentration in the force developed by the skinned fibers. However, PAK caused up to a 174% (range, 25.6 to 441.7%, n = 4) increase in force at subm~im~l calcium concentration (pCa 6.25), and at pCa 5.5, the average increase in force was 105% (range, 25.7 to 185.7%, n = 4)(Figure 14B). A repres~ e trace of skinned cardiac muscle fiber contraction is shown in Figure 15.

Equivalents Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. Such equivalents are int~nded to be encompassed by the following claims.

Claims (33)

1. A method of modulating smooth muscle contraction, comprising administering a PAK modulating agent to a subject such that modulation of smooth muscle contraction occurs.

2. The method of claim 1, wherein the contraction is calcium independent.

3. The method of claim 1 wherein the PAK modulating agent is a PAK
inhibitor.

4. The method of claim 1, wherein the smooth muscle is present in a blood vessel.

5. The method of claim 1, wherein the smooth muscle is present in the airways of the lungs.

6. The method of claim 1, wherein the smooth muscle is present in the gastro-intestinal tract.

7. The method of claim 1, wherein the smooth muscle is present in the urinary tract or uterus.

8. A method of treating a subject having a state characterized by the contraction of smooth muscle, comprising administering to the subject a therapeutically effective amount of a PAK modulating agent such that treatment of the state characterized by contraction of smooth muscle occurs.

9. The method of claim 8 wherein the modulating agent is a PAK inhibitor.

10. The method of claim 8, wherein the state is characterized by contraction of smooth muscle having high basal tone.

11. The method of claim 8, wherein the state characterized by the contraction ofsmooth muscle involves abnormal or inappropriate contraction of smooth muscle.

12. The method of claim 8, wherein the state characterized by the contraction ofsmooth muscle involves abnormal or inappropriate relaxation of smooth muscle.

13. The method of claim 8, wherein the state characterized by contraction of smooth muscle is selected from the group consisting of hypertension and asthma.

14. The method of claim 8, wherein the state characterized by contraction of smooth muscle is selected from the group consisting of incontinence, irritable bowel syndrome and menstrual cramps.

15. A method for identifying an agent that modulates the kinase activity of a PAK protein, comprising:
providing an indicator composition comprising a PAK protein having PAK kinase activity;
contacting the indicator composition with a test agent; and determining the effect of the test agent on PAK kinase activity in the indicatorcomposition to thereby identify a compound that modulates the kinase activity of a PAK
protein.

16. The method of claim 15, wherein the indicator composition comprises a smooth muscle cell.

17. The method of claim 15, wherein the indicator composition comprises a smooth muscle cell extract.

18. The method of claim 15, wherein the indicator composition comprises a detergent-skinned smooth muscle fiber bundle system.

19. The method of claim 15, wherein PAK kinase activity is assessed by measuring phosphorylation of caldesmon or a fragment thereof.

20. The method of claim 15, wherein PAK kinase activity is assessed by measuring phosphorylation of myosin light chain (LC20) or a fragment thereof.

21. The method of claim 15, wherein PAK kinase activity is assessed by measuring phosphorylation of calponin or a fragment thereof.

22. A method of modulating cardiac muscle contraction, comprising administering a PAK modulating agent to a subject such that modulation of cardiac muscle contraction occurs.

23. The method of claim 22, wherein the PAK modulating agent increases calcium sensitivity of cardiac muscle.

24. The method of claim 22 wherein the PAK modulating agent is a PAK
stimulator.

25. The method of claim 22, wherein the PAK modulating agent decreases calcium sensitivity of cardiac muscle.

26. The method of claim 25 wherein the PAK modulating agent is a PAK
inhibitor.

27. A method of treating a subject having a state characterized by cardiac contractile dysfunction, comprising administering to the subject a therapeutically effective amount of a PAK modulating agent such that treatment of the state characterized by cardiac contractile dysfunction occurs.

28. The method of claim 27 wherein the modulating agent is a PAK stimulator.

29. The method of claim 27 wherein the modulating agent is a PAK inhibitor.

30. The method of claim 15, wherein the indicator composition comprises a cardiac muscle cell.

31. The method of claim 15, wherein the indicator composition comprises a cardiac muscle cell extract.

32. The method of claim 15, wherein the indicator composition comprises a detergent-skinned cardiac muscle fiber bundle system.

33. The method of claim 15, wherein PAK kinase activity is assessed by measuring phosphorylation of troponin I (TnI) or a fragment thereof.