The present invention relates to substances which are useful in modifying cell-cell adhesion and in modifying the permeability of physiological barriers.

Cell-cell adhesion is of crucial importance for the development and maintenance of tissue structure. Furthermore, regulation of such adhesion plays a normal role in physiological situations such as tissue turnover. Aberrant control may contribute to the aetiology of pathologies such as cancer and inflammation.

Therefore, there has been considerable interest in the basic processes whereby cells adhere to each other and ways in which such processes may be regulated. In general, cell-cell adhesion involves proteins on neighbouring cells that bind to one another. The cytoplasmic domains of proteins actually involved in adhesion can also physically associate with other cytoplasmic proteins that may either play a mechanical role, such as providing links with other proteins, or provide a regulatory influence.

pl20 and plOO are members of the armadillo protein family. The armadillo repeats found in the members of this family appear to provide means whereby proteins can interact with others. pl20 and plOO, intracellular, cytoplasmic proteins, are now known to associate with cadherins, which are Ca ²⁺ -dependent adhesion molecules that contribute in an important way to cell-cell adhesion. Cadherins are important in control of cell

proliferation. If cadherins fail to function properly, cells can proliferate in an unregulated fashion and metastasize. Cadherins are also thought to be important components in controlling the permeability of physiological barriers i.e. cell tight junctions. Here, disruption of cadherin based cell-cell adhesion leads to increased tight junction permeability. This has raised the possibility that tight junctions, although potentially regulatable in themselves, may be regulated by the adhesiveness of the adherens junction. In turn, cadherin adhesiveness may be regulated by the phosphorylation state of associated catenins, including plOO and pl20.

The blood-brain barrier is an important example of a cell tight junction. It serves to separate the molecular, ionic and cellular environment of the blood from that of the brain. To a major degree, this separation is achieved by inter-endothelial tight junctions of high electrical resistance which greatly diminish paracellular flux. It is clear that the permeability of the tight junctions of the blood-brain barrier is not immutable. Rather, permeability appears to undergo dynamic regulation, but the way in which this is achieved is not fully understood.

In O95/13820 it is disclosed that tyrosine protein phosphorylation is a key regulator of the permeability of tight junctions in both epithelial and endothelial cells; tyrosine protein phosphorylation may therefore be manipulated to control the permeability of the blood- brain and other physiological barriers . Decreasing the degree of tyrosine protein phosphorylation reduces permeability of the blood-brain or other barrier,

whereas increasing the degree of tyrosine protein phosphorylation increases permeability. WO95/13820 also discloses that the proteins plOO and pl20 are believed to be substrates of tyrosine kinase. Further information regarding plOO and pl20 is provided in WO96/16170.

Although O95/13820 and WO96/16170 provide important information regarding the functioning of plOO and pl20 this information is not complete. Unexpectedly, the present inventors have now discovered that plOO and/or pl20 participate in a cycle which involves the phosphoryiation/dephosphorylation of serine/threonine residues on these proteins. Furthermore they have shown that this cycle is regulatable by agents which are known tumour promoters, inflammatory mediators and tight junction permeability modulators. Thus, agents which interfere with regulation of this cycle itself or pathways involved in its regulation could have application to a wide variety of medical situations.

According to the present invention there is provided an agent capable of inducing the phosphorylation of serine and/or threonine residues of plOO and/or pl20, for use in medicine.

The present invention also provides an agent capable of inducing the dephosphorylation of phosphorylated serine and/or threonine of plOO and/or pl20 residues, for use in medicine.

By using an agent as described above, the permeability of physiological barriers i.e. tight junction permeability could be modified and cell-cell adhesion

could also be modified.

By way of example, one way of utilising the present invention is to adjust the activity of protein kinase C. The present inventors have found that dephosphorylation of phosphorylated serine and/or threonine residues in pl20 and/or plOO can be induced by increasing protein kinase C activity (i.e. by using protein kinase C activators) , and that the phosphorylation of serine and/or threonine residues present in plOO and/or pl20 can be induced by decreasing protein kinase C activity

(i.e. by using protein kinase C inhibitors) .

Examples of protein kinase C activators are phorbol diesters, bryostatins 1 and 2, (-) indolactams V and (+) indolactam V, teleocidind, DHI

( [6- (N-Decylamino) -4-hydroxymethylindole] ) and ADMB ( [3- (N-Acetylamino) -5- (N-decyl-N-methylamino)benzyl alcohol]) , lipotoxin A4 and B4, mezerein, (-)-7- octylindolactam V, resiniferatoxin, thymeleatoxin.

Protein kinase C may also be activated by ligands that bind to receptors to generate diacylglyccrol . Examples of these are bombesin and other neuropeptides, platelet- derived growth factor, epidermal growth factor.

Examples of protein kinase C inhibitors are A3 ( [N- (2-Aminoethyl) -5-chloronaphthalene-l-sulfonamide] - this is optionally combined with HCl) , bisindolylmaleimide I (also known as GF 109203X) , chelerythrine chloride, G66976, G67874, H-7 ( [1- (5-isoquinolinesulfonyl) -2-methylpiperazine] - this is optionally combined with HCl) , hypericin, K-252a, b and c, melittin, phloretin,pseudohypericin, rottlerin, Ro 31-8220, Ro 32-0432, Y 333531, (-)balanol. Other

examples are givine in "Drug Delivery Today, 1996, vol 1 , pp438-447".

[The term "protein kinase C" , which is sometimes referred to as "PKC", is used herein to refer to a class of enzymes which catalyse the transfer of phosphate from ATP to the serine or threonine residues of polypeptides. Preferably this occurs in a specific manner so that other amino acid residues are not phosphorylated.

Activation of these enzymes can generally be inferred by assaying with MARCKS (a PKC specific substrate) . MARCKS is the term used for Myristoylated Alanine Rich C Kinase Substrate. This is a protein which was the first major PKC substrate identified. If increased phosphorylation of MARCKS is observed, it can usually be inferred that activation of PKC has occurred.

Inhibitors of PKC can be identified by their ability to prevent or reduce PKC activation.]

The present invention is however not limited to the use of activators/inhibitors of PKC since other agents which work independently of PKC but which can affect the level of phosphorylation of threonine and/or serine residues of plOO and/or pl20 can be used.

For example, lysophosphatidic acid (LPA) or histamine may be used to induce dephosphorylation of threonine and/or serine residues present in plOO and/or pl20 and agents which bind LPA or histamine or blockers of these agents may be used to block the effect of LPA or histamine.

This effect might also be achieved by the use of compositions which are hyperosmolar with respect to the physiological environment proximal to pi00 and/or pl20

(e.g. hyperosmolar solutions of sugars such as mannitol or arabinose) .

Hyperosmotic solutions are believed by some to open up the blood-brain barrier by causing the shrinking of brain endothelial cells which results in the mechanical opening of the endothelial tight junctions. However in view of the information provided herein, it is possible that hyperosmotic treatment triggers intracellular signalling processes leading to pl00/pl20 dephosphorylation which, as these are junctional proteins, could be responsible for the increase in tight junction permeability.

The present invention is therefore important in indicating that the use of hyperosmolar compositions could be avoided by using other agents to induce dephosphorylation of serine and/or threonine residues present in plOO and/or pl20. This is significant since the dangers associated with the use of hyperosmolar solutions can be avoided. (Hyperosmolar solutions such as mannitol solutions are currently used clinically.)

These dangers are twofold, firstly there is damage associated with rapid infusion of the hyperosmolar solution via the carotid or vertebral arteries. This includes arterial damage, deep venous thrombosis, pulmonary embolism, granulocytopenia, anaemia, sepsis, hemorrhagic cystitis and interstitial pneumonitis. Secondly, neurologic damage can follow blood-brain barrier opening with hyperosmolar treatment including visual changes, seizures and temporary neurological

deterioration.

Other agents useful in the present invention can be identified by screening, as described later.

Particular aspects of the present invention will now be disclosed in greater detail.

Physiolocrical Barrier Permeability

Agents of the present invention may be used to modify the permeability of physiological barriers.

Increasing Permeability

For example, agents capable of inducing the dephosphorylation of serine and/or threonine residues present in plOO and/or pl20 could be used to open physiological barriers temporarily e.g. by increasing the permeability of tight junctions of both endothelial and epithelial cell barriers. These agents are particularly useful in the case of the blood-brain barrier (an example of an endothelial barrier) , since they could allow drugs (i.e. therapeutic agents) which would normally not be able to cross this barrier to reach the brain.

The present invention is not however limited to drug delivery to the brain. It could be used for example in the delivery of drugs to other locations e.g. to tumours

(especially peripheral tumours) by loosening endothelial cell barriers .

It could also be used, for example, in nasal delivery of

drugs by loosening the nasal epithelial cell barrier. Indeed the present invention could be used for providing drug delivery across endothelial and epithelial cell barriers generally.

Where an agent of the present invention is used to facilitate drug delivery it may be administered together with a drug in a single composition.

Alternatively the agent and the drug may be administered separately. Sequential or simultaneous administration of the agent and the drug may be used.

The present invention is not limited to the use of particular drugs since it can be used to deliver a wide variety of drugs to target sites. However, it is anticipated that among the primary candidates for delivery by means of this aspect of the invention will be: anti-tumour compounds, such as methotrexate, adriamycin and cisplatin; growth factors, such as NGF, BDNF and CNTF, which are used to treat neurodegenerative disease; and neurotransmitter antagonists or agonists which do not penetrate the blood-brain barrier (such as certain NMDA receptor blockers) .

Another use of the present invention in increasing the permeability of physiological barriers is in opening pulmonary epithelial cell tight junctions. This can act to dilute the accumulation of mucous present in the lungs. It is also useful in aiding the administration of drugs to the lungs, e.g. where an inhaler is used.

b) Decreasing- Permeability

Agents capable of inducing the phosphorylation of unphosphorylated serine and/or threonine residues of plOO and/or pl20 could be used to decrease the permeability of physiological barriers. This could be achieved for example by decreasing the permeability of tight junctions of both endothelial and epithelial cell barriers. Such agents could be used, for example, in closing tight junctions after they have been opened or in preventing them from opening. In the case of the blood-brain barrier, it will generally be desirable to use these agents to close cell tight junctions in order to avoid possible damage to the brain after they have been opened to allow a drug to cross this barrier.

It is also desirable to close cell tight junctions in the blood-brain barrier for other purposes . For example this can be done: - a) to block or reduce the entry into the brain of lymphocytes which mediate an immune response (useful in treating multiple sclerosis) or of neutrophils which can damage neuronal cells after stroke; b) to prevent or reduce the entry of metastatic cancer cells into the brain (useful in treating cancer) ; and c) to prevent or reduce the risk of brain oedema eg. following stroke or traumatic head injury (useful in treating head injuries, oedema and stroke) .

Cell trafficking across endothelia (such as T cell migration across the blood-brain barrier) and epithelia

is believed to occur by several sequential steps. The trafficking cell initially binds to the endothelium or epithelium, at first loosely and then more tightly, migrates to the junctional region and is then believed to migrate through the junction. Clearly, tight junctions must be modulated during this process and it is possible that signalling changes involving pl00/pl20 serine/threonine phosphorylation may occur and be a necessary part of the transmigration process. Therefore blocking serine/threonine phosphorylation changes in plOO and pl20 could block transmigration.

The present invention could also be used to decrease the permeability of other physiological barriers. It could therefore be useful in treating oedema (even if this does not occur in the brain) . For example it could be used to treat vasogenic oedema in peripheral tissues by reducing the permeability of the vasculature (e.g. in treating high altitude pulmonary oedema) . Reducing the permeability of the vasculature is also useful in preventing or reducing metastases (whether in the brain or elsewhere) by preventing or reducing the ability of tumour cells to exit from the vasculature.

Another use of the present invention in respect of treating tumours is that it could be used to reduce the permeability of leaky junctions of the vasculature supplying solid tumours (i.e. to "tighten" these junctions) . This can reduce the uptake of nutrients by such cells.

A further use of the present invention is in treating inflammation. This could be achieved by reducing leukocyte migration to peripheral tissues through

endothelial cell tight junctions. By reducing the permeability of such junctions leukocyte migration to peripheral tissues can be reduced. The present invention is not however limited to treating inflammation of peripheral tissues. It could be used to treat other inflammatory conditions. For example, it could be used to treat intestinal inflammation. Neutrophil migration across intestinal epithelia can be a cause of intestinal inflammation. By reducing the permeability of epithelial cell junctions intestinal inflammation could therefore be reduced.

A yet further use of the present invention is in treating gastric ulcers. Gastric ulcers can be exacerbated by "loose" tight junctions. Thus decreasing the permeability of tight junctions of gastric epithelial cells could be used in the alleviation of gastric ulcers.

From the foregoing passages it is clear that the present invention could be used to treat a wide variety of different disorders by reducing the permeability of cell junctions in order to prevent cells which can cause disorders from reaching sites where such disorders can arise. The present invention could therefore generally be useful in preventing or reducing aberrant cell trafficking.

Cell-Cell Adhesion Properties

Agents of the present invention could also be used to modify cell-cell adhesion.

For example, such agents could be used to induce the

phosphorylation of serine and/or threonine residues of pl20 and/or plOO in order to increase the tendency of cells to adhere to one another.

This could, for example, prevent or slow down the loss of contact inhibition which is seen in the development of cancerous cells . Thus the loss of growth control which is associated with loss of contact inhibition could be prevented or reduced. Such agents could also prevent or slow down the increased cell motility, migration and metastasis, which occurs in the development of cancers.

Medicaments

The agents of the present invention (which could modify the permeability of physiological barriers and which could modify cell-cell adhesion properties) could be used in the preparation of medicaments for use in human or in veterinary medicine.

These medicaments could be used for the treatment of an existing condition or for prophylactic treatment and the word "treatment" is used herein to cover both of these alternatives.

An agent for use in the present invention may be combined with a pharmaceutically acceptable carrier to form a pharmaceutically acceptable composition. This pharmaceutical composition will usually be sterile. It may be in any suitable form (depending upon the desired method of administering it to a patient) . For example, it may be provided in unit dosage or multiple dosage form. It may be provided as a derivative. Thus

pharmaceutically acceptable salts, esters or other derivative forms may be used, provided that activity is retained.

Pharmaceutical compositions within the scope of the present invention may be adapted for administration by any appropriate route, for example by the oral

(including buccal or sublingual) , rectal, nasal, topical

(including buccal, sublingual or transdermal) , vaginal or parenteral (including subcutaneous, intramuscular, intravenous or intradermal) route. Such compositions may be prepared by any method known in the art of pharmacy, for example by admixing the active ingredient with a carrier under sterile conditions.

Screens

Without being bound by theory, a possible scheme is shown in Figure 1, which will be described later.

In any event the present inventors have established that regulation of the degree of phosphorylation of serine/threonine present in pl20/pl00 could be used to modify the permeability of physiological barriers and also to modify cell-cell adhesion properties. This finding was not predictable prior to the present invention and is significant in establishing the important regulatory role of pl00/pl20.

[plOO and pl20 are described in greater detail in WO95/13820 and WO96/16170, where tyrosine phosphorylation is also discussed. They are proteins with molecular weights of about lOOkDa and about 120kDa respectively, and are associated with cadherin and

cadherin complexes present in endothelial and epithelial cells. ]

In view of this important finding, the present invention could be used to provide a screen for pharmaceutically active compounds.

For example, plOO and/or pl20 could be used to screen for substances capable of increasing the permeability of physiological barriers/reducing cell-cell adhesion.

This could be done by screening for substances which can induce the dephosphorylation of phosphorylated threonine and/or serine residues present in plOO and/or pl20 (preferably under physiological conditions) .

Alternatively plOO and/or pl20 could be used to screen for substances capable of decreasing the permeability of physiological barriers/increasing cell-cell adhesion. This could be done by screening for substances which can induce the phosphorylation of threonine and/or serine residues present in plOO and/or pl20.

Research Uses

The agents disclosed herein are not limited to being used in medical treatment or in screening since they are generally useful for research into cell-cell adhesion and into physiological barriers (eg. cell junctions, such as cell tight junctions) . In particular, these agents are useful for studying the proteins plOO and pl20 and for developing more accurate models of the functions of these proteins. The agents disclosed herein are especially useful in investigating the effects induced by phosphorylation/dephosphorylation of

threonine and/or serine residues present in plOO and/or pl20.

Diagnostic Uses

Agents of the present invention may be used in diagnosis since they can be used to alter the permeability of physiological barriers in order to allow substances which are useful in aiding a diagnosis to reach a desired location. This may be a location which they would not otherwise reach via a particular route of administration. Thus, for example, substances useful in diagnosis which do not normally cross the blood-brain barrier may be allowed to cross this barrier via the present invention. The present invention can also be used to facilitate the entry of diagnostic substances into tumours.

Any suitable diagnostic substances may be used. Antibodies, antibody fragments, lectins or other molecules having high binding specificity may be linked to diagnostic substances in order to target them to a desired site. (Such molecules can also be linked to drugs to target pharmaceutically active substances to a desired site.) For example cancer cells can be targeted.

Preferred diagnostic substances are useful in imaging. For example, they may be useful in providing brain scans .

In the foregoing discussions , references to " increasing" or "decreasing" the permeability of physiological barriers and to " increasing" or "decreasing" cell-cell

adhesion are made in the context of administering agents of the present invention to a patient. These shall be taken to include not only absolute increases/decreases in permeability or adhesion but also the increases/decreases in permeability or adhesion relative to the situation which would arise if a patient were left untreated (rather than being treated with an agent for use in the present invention) .

The present invention will now be described by way of illustration only, with reference to the accompanying figures.

Figure 1

This shows a scheme of how protein kinase C (PKC) activation could result in dephosphorylation of pi 00 and pi 20. In order to bring about dephosphorylation of pl00/pl20, PKC activation, as elicited by phorbol 12,13-dibutyrate (PDB) must lead to net kinase inhibition and/or activation of phosphatases. Activated kinase is depicted as "kinase*". Since the effect on pl00/pl20 is rapid and can easily be reversed it is clear that pl00/pl20 is capable of cycling between phosphorylated and lesser phosphorylated forms, subject to the action of pl00/pl20 kinas ) and pl00/pl20 phosphatase(s). In resting cells, pl20* is phosphorylated on serine and threonine, the level of phosphate being maintained by the opposing actions of a serina- hreonine phosphatase. and corresponding kinase. PKC may act directly, or indirectly to inhibit the kinase. leading to pl20 dephosphorylation (as represented here). Alternatively, PKC could somehow promote the action of the serine/threonine phosphatase, which would also cause dephosphorylation of pl20. This contrasts with the tyrosine phosphorylation cycle of pl20; in resting cells there is little or no tyrosine phosphorylation of pl00/pl20. Following activation of src, or perhaps other related kinases, pi 20 becomes phosphorylated on tyrosine. In endothelial cells, lysophosphatidic acid (LPA), histamine and mannitol treatment also cause pl00/pl20 dephosphorylation but, based on experiments using protein kinase C inhibitors, seem to act independently of protein kinase C.

NB: This scheme can also apply to plOO, but for simplicity only pl20 has been shown.

Figure 2

The migration of the cellular proteins plOO and pl20 during electrophoresis is increased following PDB- treatment of MDCK I cells. PDB (phorbol-12,13 hT3utyrate) is a pharmacological activator of protein kinase C. MDCK I cells are a kidney derived epithelial cell line, with particularly well developed junctional complexes. Cells treated for 30 minutes with either 200 nM PDB (+) or vehicle (-) were then lysed into Laemmli sample buffer, and separated by electrophoresis using 6% acrylamide gels. Proteins were transferred to nitrocellulose and probed with antibodies to pl00/pl20, and then reprobed with E- cadherin and paxillin antibodies. PDB treatment causes pi 00 and p 120 to migrate as faster, tighter bands (A). In contrast PDB treatment has no effect on E-cadherin (B), and induces an upward band shift of paxillin (C) due to its phosphoiylatioa

Figure 3 pi 00 and pl20 electrophoretic mobility is affected by PDB in a number of different epithelial cell lines. LLC-PKi, (a porcine kidney cell line), MDCK strain II cells and gut-derived Caco-2 cells were treated with or without 200 nM PDB for 30 minutes, lysed into Laemmli sample buffer, separated by electrophoresis and analysed by immunoblotting. Migration of pi 00 and pl20 from LLC-PKi and MDCK II cells increased markedly in response to phorbol ester. The effect was less striking, but still present in Caco-2 cells.

Figure 4 ^'

The mobility of pi 00 and pi 20 during electrophoresis is also altered by PDB-treatment in endothelial cells. The pl00/pl20 proteins from human umbilical vein endothelial cells (HUVECs), bovine aortic endothelial cells (BAECs) and primary pig brain endothelial cells (PBECs) were all affected by PDB- treatment Cells were treated with or without 200 nM PDB for 30 minutes, lysed into Laemmli sample buffer, separated by electrophoresis and analysed by immunoblotting with antibodies recognising pi 00 and pi 20. In all cases, the mobility of pi 00 and pi 20 increased Thus the effect of PKC activation (via PDB) on p 100/p 120 mobility is a widespread phenomenoα

FigureS

Pretreatment of MDCK I cells with the specific PKC inhibitors bisindolylmaleimide I or Ro 31-8425 for 5 minutes prior to PDB addition completely abolished the pl 0/pl20 band shift (Example 4 A). An inactive member of the bisindolylmaleimide class, bisindolylmaleimide V, was unable to block the PDB- induced band shift. Analysis of dose-dependence showed that PDB was effective at concentrations between 30 and 100 nM (Example 4B), corresponding to the concentrations required to activate PKC. Also, an inactive sterεoisomer of PDB, 4α-PDB, had no effect on pl00/pl20 mobility (Example 4 . The effect of PDB could also be obtained by a cell permeant diacylglycerol analogue, 1,2- dioctanoylglycerol (DiCg) (Example 4D). These results demonstrate that the pi 00/120 band shift occurs as a response to PKC activation by PDB. This rules out the possibility that the pl00/pl20 mobility shift seen in response to PDB was due to activation (by PDB) of some unknown pathway. Experimental details: A. MDCK I cells were pretreated with either bisindolylmaleimide I (2.5 μM), an inactive analogue of bisindolylmaleimide I, bisindolylmaleimide V (2.5 μM) or Ro 31-8425 (5 μM) for 5 minutes, and then exposed to 200 nM PDB for 30 minutes. Cell lysates were analysed by electrophoresis, and immunoblotted with pl00/pl20 antibodies. B. PDB Dose dependence. MDCK I cells were incubated with a range of PDB concentrations from 1 to 200 nM for 30 minutes. pl00/pl20 mobility increased with increasing PDB concentration, with maximal affect at 1 0 nM PDB. C. A stereoisomer of PDB, 4α-PDB does not activate PKC, and has no effect of pl00/pl20 mobility. 4α-PDB was added at 200 nM for 30 minutes. D. DiCg, a cell permeant diacylglycerol analogue induces a pl00/pl20 mobility shift. MDCK I cells were treated with 0.5 mM DiC ₈ for 10 minutes.

Figure 6

Time dependence and reversibility of PDB effect MDCK I cells were treated with 200 nM PDB for the times indicated then lysed into Laemmli sample buffer, separated by electrophoresis, and immunoblotted Reversibility was demonstrated by treating cells with 200 nM PDB for 30 minutes, then adding 2.5 μM bisindolylmaleimide I to inhibit PKC (still in the presence of PDB) for the times indicated Maximal effect of PDB on pl00/pl20 in seen between 1 and 5 minutes, and can be reversed by

bisindolylmaleimide I with equal rapidity. Addition of the inactive bisindolylmaleimide V did not reverse the pl00/p'120 band shift (data not shown). These data show that interconversion of plOO and pl20 from slower to faster migrating forms is dynamic, and clearly argues against the possibility that the increased p 100/p 120 mobility is due to pro teoiysis.

Figure 7

PDB induces dephosphorylation of p 100 and p 120. We investigated the possibility that the altered pattern of migration during electrophoresis of pi 00 and pi 20 following treatment with PDB is due to a change in the phosphorylation state of these proteins. MDCK I cells were metabolically labelled with [ ³² P]orthophosphate. treated with or without 200 nM PDB, and the plOO and pl20 isolated by immunoprecipitaήon with antibodies recognising plOO and pl20. This immunoprecipitation was carried out under buffer conditions which removed any associated proteins from pi 00 and pi 20. Protein was separated by electrophoresis. transferred to nitrocellulose filters, and exposed to film (Example 6A). The greater the amount of radioactive phosphate incorporated into the protein, the greater the signal on the film. Thus, the level of phosphate present in pi 00 and pi 20 from untreated cells could be compared to that from PDB-treated cells by densitometric scanning of the resulting a toradiograph. To normalise for slight differences in protein loading, the same filters were then probed with pl00/pl20 antibody, and the level of pl00/pl20 protein determined by densitometry. Thus, the amount of phosphate per unit protein could be calculated Panel B shows the results of the scanning of the autoradiographs shown in panel A, (arbitrary units). 'P _n ' indicates "phosphate level normalised for protein'. In this experiment, approximately 40% of the phosphate in pl00/pl20 was lost following PDB treatment The results from seven independent sets of such experiments showed that, following PDB treatment phosphate content of pl00/pl20 was reduced to 58 ± 11% (mean ± s.d) of that in untreated cells. The dephosphorylation of plOO and pl20 is specific, since there was no change in the phosphorylation of E-cadherin immunoprecipitated from the same cell lysates, or on the phosphorylation state of proteins in whole cell extracts (data not shown). This rules out the extremely unlikely possibility that the effect on pl00/pl20 was an artefact due to non-specific effects of PDB on labelling.

Panel C. shows analysis of the phosphoamino acid (PAA) content of pl00/pl20. Following phosphate labelling, immunoprecipitation and visualisation of pl00/pl20 as in panel A above, the levels of pl00/pl20 protein were determined by immunoblotting. Thus it could be ascertained that approximately equal amounts of protein from PDB-treated and untreated cells were used for the PAA analysis. The pl00/pl20 bands were excised and the proteins hydrolysed into their constituent amino acids by boiling in 5.7 M HCl. PAAs were then separated by two-dimensional electrophoresis, and detected by autoradiography. pl00/pl20 from control cells contains mainly phosphoserine (S), some phosphothreonine (T) and no detectable phosphotyrosine (Y). Following PDB treatment both phosphoserine and phosphothreonine levels are reduced Densitometric scanning of the autoradiographs of the PAA analysis revealed a reduction in phosphate signal of approximately 40% (data not shown),

agreeing with the values obtained from the whole protein phosphate labelling experiments.

Figure 8

In order to test the hypothesis that a cycle of pl00/pl20 dephosphorylation exists MDCK I cells were treated with kinase or phosphatase inhibitors to try and perturb this cycle. Thus, incubation for 60 minutes with 100 nM staurosporine. a broad range kinase inhibitor, induced a pl00/pl20 band shift similar to that seen in response to PDB (panel A). This would be consistent with inhibition of a pl00/pl20 kinase. Addition of a more specific staurosporine derivative, KT 5926 (at a concentration of 1 μM) also induced a pl00/pl20 band shift (panel A), suggesting these inhibitors may indeed act directly on the pl00/pl20 kinase.

Addition of the serine threonine phosphatase inhibitors calyculin A (100 nM) or okadaic acid (1 μM) to MDCK I cells from 60 minutes had the opposite effect of pl00/pl20 mobility: induction of an upward band shut consistent with increased phosphorylation (panel B). Thus it would appear that in resting cells, a certain level of pl00/pl20 phosphorylation is maintained by a balance of the opposing effects ofkinase and phosphatase. Inhibition of the kinase tips the balance in favour of the phosphatase, causing net pl00/pl20 dephosphorylatioa In contrast, inhibition of the phosphatase leads to predominant kinase activity, manifested by hyperphosphorylation of pl00/pl20 (see Example 1). We propose that PKC acts to perturb this cycle, perhaps by phosphorylating and inhibiting a pl00/pl20 kinase. Alternatively, PKC could act to increase the activity of the pl00/pl20 phosphatase.

Figure 9

It has been shown that although some of the pl00/pl20 in cells is associated with adherens junctions, there is also a large pool that is not complexed with cadherins (at least by immunoprecipitation analysis). In order to determine which of these pl00/pl20 pools are subject to dephosphorylatioa MDCK I cells were treated with PDB, extracted using Triton lysis buffer, which preserves protein-protein interactions, and immunoprecipitated with E-cadherin antibodies. This procedure served to isolate only that portion of the cellular pl00/pl20 that associates with the cadherins. Following electrophoresis, and transfer to nitrocellulose, the pi 00 and pi 20 present in E-cadherin immunoprecipitates were detected by immunoblotting (Example 9 A). It can be seen that the E-cadherin-associated pl00/pl20 is dephosphorylated in response to PDB. Immunoblot analysis of the non-E-cadherin-associatedpl0O/pl20 revealed that this too was subject to dephosphorylation (data not shown). There is no obvious change in the amount of plOO and pl20 present in E-cadherin immunoprecipitates after addition of PDB. In additioa irnmunofluorescence studies in MDCK I cells revealed no change in the localisation of pi 00 and pl20 (Example 9 B,C). Panel (B) shows pl00/pl20 in control cells, (C) shows pl00/pl20 in PDB- treated cells (Bar 20 μM). In both cases, pi 00/pl20 is shown to be present at regions of cell-cell contact

Figure 10

Human umbilical vein endothelial cells were incubated with histamine (His), pyrilamine (Pyr) or cimetidine (Cim) alone or in combination As indicated cells were preincubated in the absence or presence of cimetidine (10 μM) or pyrilamine (10 μM) for 10 minutes and then treated with or without histamine (10 μM) for a further 10 minutes. The cells were extracted and protein analysed by gel electrophoresis and immunblotting. Histamine alone induced an increased mobility of plOO and pl20. Pyrilamine, a Hi blocker, and cimetidine, a H ₂ block alone had no effect but pyrilamine blocked the histamine effect As histamine is an agent well known to be responsible for increased tight junction permeability in endothelial cells, these data raising the possibility that the pl00/pl20 mobility shift could be causally related to such an increase. The Hi-mediated effect of histamine indicates that its effect in some way is mediated by activation of phosphoinosital lipid-specific phospholipase C. Like the effect of phorbol dibutyrate, these data indicate that histamine stimulates dephosphorylation of pl00/pl20. In the following examples, increase mobility of pl00/pl20 means decreased phosphorylatioa

Figure 11

Human umbilical vein endothelial cells were incubated with histamine for times indicated The cells were extracted and protein analyzed by gel electrophoresis and immunoblotting. Flistamine induced a rapid increase in the mobility of plOO andpl20.

Figure 12

Human umbilical vein endothelial cells were incubated in the absence (Cont) or presence of either histamine (10 μM), lysophosphatidic acid (LPA; 10 μM) or ATP (100 μM) for 15 minutes. The cells were extracted and protein analyzed by gel electrophoresis and immunblotting. Histamine, lysophosphatidic acid and ATP all produce an increase in the mobility of plOO and pl20. These diverse agents are know to bind to cell surface receptors to trigger intracellular signalling changes, probably in all cases mediated by activation of phosphoinositol lipid-specific phospholipase C.

Figure 13

Human umbilical vein endothelial cells were preincubated for 15 minutes with or without the protein kinase C inhibitor bisindolylmaleimide (Bis; 2 μM). The cells were then further treated for 5 minutes with or without histamine (His; 10 μM) or phorbol dibutyrate (PDB; 200 nM). The cells were extracted and protein analyzed by gel electrophoresis and immunblotting. As before, histamine and phorbol dibutyrate treatment resulted in an increased mobility of pl20 and plOO. The effect of phorbol dibutyrate was blocked by bisindolylmaleimide, whereas that of histamine was not so strongly affected suggesting that histamine mainly acts independently of protein kinase C activatioa

Figure 14

Brain endothelial cells from pig were incubated with or without lysophosphatidic acid (LPA; 10 μM for 30 minutes). The cells were extracted and protein analyzed by gel electrophoresis and immunblotting. Each lane represents protein extracted from an individual Transwell. LPA treatment resulted in an increased mobility of pi 20 and pi 00. LPA also increases tight junction permeability in these cells (Schulze et al., submitted for publication), raising the possibility that the p 100/p 120 mobility shift could be causally related to such an increase. Further experiments have shown that the LPA effect on pl00/pl20 can be seen within a few minutes of addition to the cells.

Figure 15

Brain endothelial cells from pig were preincubated for 10 minutes with or without the protein kinase C inhibitor biandolylrnaleimide (2.5 μM). The cells were then treated with or without lysophosphatidic acid (LPA; 10 μM for 70 minutes). The cells were extracted and protein analyzed by gel electrophoresis and immunblotting. As before, LPA treatment resulted in an increased mobility of p 120 and p 100 but this was not blocked by bisindolylmaleimide, suggesting that LPA acts independently of protein kinase C activatioa Furthermore, bisindolylmaleimide did not block the ability of lysophosphatidic acid to increase tight junction permeability in these cells (results not shown).

Figure 16 (a) Hyperosmolar mannitol causes a drop in pig brain endothelial cell transcellular electrical resistance. Pig brain endothelial cell cultures were treated with 0.2-0.4 M mannitol for up to 6 hours and transcellular electrical resistance recorded The presence of a measurable electrical resistance across a monolayer of cells indicates that there is very little ionic flux across the monolayer, this is only possible when well developed tight junctions are present between cell-cell contacts in the culture. A drop in transcellular electrical resistance represents opening of these cell-cell contacts to ions and the decrease in transcellular electrical resistance in response to hyperosmolar mannitol indicates that hyperosmolarity opens tight junctions. Mean measurements ± standard deviation of 3 Transwells are shown Mean initial transcellular electrical resistance was 815 ± 69 Ω.cm ² (n=12). (b) The drop in transcellular electrical resistance caused by hyperosmolar mannitol treatment is clearly seen within 20 seconds. Pig brain endothelial cell cultures were treated with 0.8M mannitol and transcellular electrical resistance measurements recorded over 2 minutes. Mean measurements ± standard deviation of 3 Transwells are shown (c) The drop in transcellular electrical resistance caused by hyperosmolar mannitol treatment is fully reversible. Pig brain endothelial cell cultures were treated with 0, 0.4M or 0.8M mannitol. After 2 minutes the Transwells were returned to control medium and transcellular electrical resistance measurements were taken every minute for the following 18 minutes. In another experiment, recovery was also shown to be possible after treatment with 0.8M mannitol for 30 minutes (not shown). Mean measurements ± standard deviation of 3 Transwells are shown Mean initial transcellular electrical

resistance was 419 ± 56 Ω.cm ² (n=9).

Figure 17

(a) Hyperosmolar mannitol treatment induces an increase in mobility in p 1 OO/p 120. Pig brain endothelial cell cultures were exposed to 0.4M mannitol for 10 minutes then cells were extracted protein separated by electrophoresis and pl00/pl20 detected by immunoblotting. Lanes 1, 3, 5 control cells with iso- osmotic medium; lanes 2, , 6 mannitol treated (b) The increase in mobility in pl00/pl20 induced by hyperosmolar mannitol is seen within 5 minutes and maintained for 30 minutes. Following hyperosmotic treatment cells were extracted at the times indicated protein separated by electrophoresis and pl00/pl20 detected by immunoblotting. Lane 1; control (no mannitol), lanes 2, 3, 4, 5, 6 and 7: respectively 5, 10, 15, 20, 25, 30 minutes treatment with 0.8M mannitol. In a separate experiment the increase inpl00/pl20 mobility was clearly detected as early as 30 seconds after addition of hyperosmolar mannitol (not shown), (c) The increase in pi 00/p 120 mobility in response to hyperosmolar mannitol is completely reversed 30 minutes after mannitol is removed Following treatment as indicated cells were extracted protein separated by electrophoresis and pl00/pl20 detected by immunoblotting. Lane 1; no mannitol, lane 2; 5 minutes treatment with 0.8M mannitol, lane 3; 5 minutes in 0.8M mannitol followed 30 minutes recovery in iso-osmotic medium, lane 4; 35 minutes treatment with 0.8M mannitol. (d) The protein kinase C inhibitor bisindolylmaleimide does not alter the pig brain endothelial cell response to hyperosmolar treatment Hyperosmolar sorbitol was added to pig brain endothelial cell cultures for 10 minutes, causing both a substantial drop in transcellular electrical resistance (from 677 to 73 Ω.cm ) and an increase in mobility of pl00/pl20. Following pre-incubation with 2.5 μM bisindolylmaleimide, transcellular electrical resistance dropped to a similar extent (from 643 to 93 Ω.cm ) in response to 10 minutes treatment with 0.3M sorbitol. Likewise, the increase in mobility of pl00/pl20 in response 0.3M sorbitol was not prevented by inhibition or protein kinase C. Following treatment as indicated cells were extracted protein separated by electrophoresis and pl00/pl20 detected by immunoblotting. Lane 1; control, lane 2; 10 minutes 0.3M sorbitol, lane 3; 15 minutes 2.5 μM bisindolylmaleimide, lane 4; pretreatment for 5 minutes with 2.5 μM bisindolylmaleimide followed by addition of 0.3M sorbitol.

MATERIALS AND METHODS Some but not all of the materials and methods have been described in detail before (see: Rubin et al., 1991: Staddon et al. 1995a,b; Schulze et al., 1997), but those of particular relevance to the present application are described again here.

Chemicals Phorbol- 12, 13-dibutyrate (PDB), bisindolylmaleimide I, bismdolylmaleimide V, KT5926, staurosporine (Calbiochem) and 4α-PDB (LC Laboratories), were made up as stock solutions at appropriate concentrations in DMSO, and added to cells at a 1:1000 dilutioa DiCg (Molecular Probes, Inc.) was

made up as 50 mM stock in DMSO. and added at 1:100, to give a final concentration of 0.5 raM. Ro 31- 8425 was provided by Eisai Ltd Okadaic acid (Calbiochem) was made up as 100 μM stock in DMSO, and added at 1 : 100. Calyculin A (Calbiochem) was made up as 10 μM stock in PBS. and added to cells at 1 : 100. In all cases, appropriate vehicle controls were performed

Cells

MDCK strain I and II cells were provided by Barry Gumbiner (Memorial Sloan-Kettering Cancer Center, NY). LLC-PKi and Caco-2 cells were obtained from the European Collection of Animal Cell Cultures. All cells were maintained at 37°C under 5% CQ ₂ in humidified air. MDCK strain I and II cells were maintained in MEM containing 10% FCS, 100 U/ml penicillia 100 μgml streptomycin and 2 mM L-glutamine. Caco-2 cells were maintained in MEM containing 15% FCS, 1% non-essential amino acids and 1 μgml bovine insulin (Sigma 1-6634). LLC-PKi cells were grown in Ml 99 (Gibco), 10% FCS, 100 U/ml penicillia 100 μg/ml streptomycin and 2 mM L-glutamine. For whole cell lysate analysis, 10 cells were plated on 6.5 mm diameter, 0.4 μm, polycarbonate Transwell filters from Costar, in 025 ml of medium. The basolateral chamber contained 0.75 ml of medium. For immunoprecipitation, 5x10 ⁵ cells were plated onto 24 mm Transwell filters, in 2 ml of medium, with 3 ml of medium in the lower chamber. Cells were used 4-7 days after plating. Tissue culture materials were from Gibco. Porcine brain endothelial cells were isolated and grown by modifications of a previously published procedure (Rubin et al., 1991 ). Brain capillary fragments were plated on dishes coated with rat tail collagen and human fibronectia The cultures were fed every other day with growth medium composed of 50% astrocyte-conditioned medium and 50% Dulbecco ^' s Modified Eagle Medium containing 10% plasma-derived serum, 125 μg/ml heparin, antibiotics and glutamine. They were incubated in a humidified atmosphere of 10% CO_. at 37°C. When the endothelial cells were approximately 50% confluent (after 4-6 days in culture), they were trypsinized and plated onto collagen- coated polycarbonate Transwell filters (6.5 mm diameter, 0.4 μm pore size, tissue culture treated Costar Corp., Cambridge, MA, US ) in the same medium (250 μl apically, 750 μl basolaterally). After 2-3 days on filters, the medium was switched to a 1:1 mixture of medium and a Dulbecco's Modified Eagle Medium-based chemically defined medium ('N2', Rubin et al., 1991; Bottenstein and Sato, 1979). The final concentration of plasma-derived serum was 5%. Human umbilical vein endothelial cells were from Clonetics.

Resistance measurements

Transcellular electrical resistance (TER) measurements were taken with a pair of fixed current-passing, voltage measuring electrodes (World Precision Instruments, Stevenage, Hertfordshire, U.K). TER values are expressed in Ω.cm ² and were corrected for the resistance of cell-free filters.

Cell lysis and Immunnoprecipitation

Whole cell lysates were prepared by rapidly replacing the culture medium in 6.5 mm with hot laemmli sample buffer, followed by heating at 100°C for 5 minutes. Immunoprecipitations were carried out at 4°C. Confluent monolayers of cells were washed rapidly with ice cold PBS (containing 0.9 mM CaCl ₂ and 0.5 mM MgCl ₂ ) before addition of 0.5 ml of the appropriate buffer. The filters were then gently scraped and the lysate was removed and centrifuged at 14000 x g. The supernatant was precleared with 10% (w/v) protein A-Sepharose in lysis buffer for 30-60 minutes. Appropriate primary antibodies were added to the lysate, and after 2-3 hours the immunecomplex was isolated using rabbit anti-mouse secondary antibodies (Jackson Immunoresearch Labs Inc., West Grove PA) and protein A-Sepharose. The beads were washed five times with lysis buffer, and the bound protein was solubilized in SDS- sample buffer, followed by heating at 100°C for 5 minutes.

For Immunoprecipitations, Triton lysis buffer or TDS lysis buffer was used Triton lysis buffer (pH adjusted to 7.4 using NaOH) contained 25 mM Hepes, 1 % Triton X- 100.2 mMEDTA, 0.1 MNaCl, 25 mM NaF, 1 mM vanadate, and protease inhibitors. TDS lysis buffer (pH 7.4) comprised Triton lysis buffer supplemented with 0.5% sodium deoxycholate and 0.2%> SDS. The protease inhibitors used in each case were; leupeptin, 10 μg/ml; α2-rnacroglobulir_, 0.1 U/ml; soybean trypsin inhibitor, 10 μg/ml and PMSF, 1 mM. Antibodies recognising E-cadherin, β-catenin and pi 20 were obtained from Transduction Labs. (Lexington, KY).

Electrophoresis and immunoblotting

Cell extracts or immunoprecipitates in SDS sample buffer were resolved by SDS-PAGE. The slab gels were then equilibrated in buffer comprising; 48 mM Tris, 39 mM glycine, 20% methanol and 0.03% SDS. Proteins were transferred to nitrocellulose filters (Hybond ECL; Amersham), which were Ponceau S stained and then blocked for 2 hours at 37°C with 1% BSA in PBS containing 0.05% Tween-20. Filters were then probed with p 120 antibody, at a dilution of 1 :2500. After washing, filters were incubated with horseradish peroxidase conjugated anti-mouse antibody (Amersham), at a dilution of 1:5000. The filters were then extensively washed in PBS/0.5% Tween-20, and immunoreactive bands were detected by enhanced chemiluminescence (Amersham) following the manufacturers instructions.

Phosphate labelling

MDCK I cells grown to confluence on 24 mm filters were incubated overnight in phosphate-free MEM, containing 0.5% serum (dialysed against 0.9% NaCl and 10 mM Hepes (pH 7.5)) and 100 μCi/ml (Amersham). Cells were treated with 200 nM PDB for 30 minutes, then lysed into TDS buffer, and immunoprecipitated with pl20 antibodies as described above. Following transfer of proteins to nitrocellulose, the filters were exposed overnight to film, and the resulting autoradiograph of [ ³² P]-labelling scanned by densitometiy using a Bio-Rad Imaging Densitometer GS-670. Filters were

then probed with pl20 antibodies, and the resulting ECL exposure scanned in the same way, to quantify protein loading.

For phosphoamino acid (PAA) analysis, MDCK I cells were labelled as above with [ ³² P]orthophosphate, and the pi 00 and pi 20 extracted from PDB-treated or untreated cells by immunoprecipitation (as above). Following transfer to Immobilon P, and immunoblotting to assess recovery of protein, the area of the filter containing plOO and pl20 was excised The proteins were hydrolysed at 110°C for one hour in 5.7 M HCl to release PAAs. Following lyophilisation, two dimensional PAA separation and detection was carried out as previously described (Boyle etal., 1991).

Immunofluorεscence staining

Cells were fixed at room temperature for 20 minutes in 3% paraformaldehyde made up in PBS containing 0.5 mM CaCl ₂ and 0.5 mM MgSO ₄ . Fixed cells were washed and then permeabilized by incubation with 0.5%ι Triton X- 100 in PBS for 10 minutes. After washing, the cells were incubated for 30 minutes in PBS containing 10% calf serum and 0.1 M lysine,pH 7.4. Incubation with pl20 antibody (0.5 μg/ml) was in PBS containing 10% calf serum for 1 hour. After washing, the cells were then incubated for 60 minutes with a 1:100 dilution of FITC-coηjugated anti-mouse IgG (Jackson Immunoresearch Labs), in PBS coritainirig 10% calf serum. After washing, the filters were mounted with Citifluor (Citifluor Products, Canterbury, UK) and examined using a Nikon Microphot-FXA fluorescence microscope fitted with 40 x and 60 x objectives. Photographs were taken using Kodak T-Max film (400 ASA).

REFERENCES

Boyle WJ. Van der Geer P and Hunter T., Phosphopeptide

Mapping and Phosphoamino acid Analysis by Two- Dimensional Separation on Thin-Layer Cellulose Plates. Methods in Enzy ology 1991: 201: 110-149. Laemmli UK cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 1970: 227: 680-685. Schulze C, Smales C, Rubin LL and Staddon JM.

Lysophosphatidic acid increases light junction permeability in brain endothelial cells. Journal of Neurochemistry 1997: 68: 991-1000. Staddon J.M. , Herrenknecht K. , Smales C, and Rubin L.L., Evidence that tyrosine phosphorylation may increase tight junction permeability. Journal of Cell Science 1995a: 108: 609-619. Staddon J.M., Smales C, Schulze C. , Esch F.S., and Rubin L.L., pl20, a pl20-related protein (plOO) , and the cadherin/catenin complex. Journal of Cell Biology 1995b: 130: 369-381. Rubin L.L., Hall D.E., Porter S., Barbu K. , Cannon H.C., Horner H.C., Janatpour M. , Lia C.W. , Manning K. , Morales K. , Tanner L.I., Tomaselli K.J., and Bard F.A. cell culture model of the blood-brain barrier. Journal of Cell Biology 1991: 115: 1725-1735.