WO2000014254A2 - Component of bromelain - Google Patents

Abstract

A fraction of bromelain has been separated into various components. The components responsible for the majority of the anti-ERK-2 and anti-IL-2 activity of bromelain have been identified. The active components contain various different proteins, some of which are novel.

Description

COMPONENT OF BROMELAIN

The present invention relates to a component of bromelain. In particular, the invention relates to the use of this bromelain component in medicine, particularly as an anti-cancer agent and an immunosuppressive agent capable of blocking T-cell responses.

Stem bromelain (bromelain) is the collective name for the proteolytic enzymes found in the tissues of the plant Bromeliaceae. It is a mixture of various moieties derived from the stem of the pineapple plant (Ananas comosus). Bromelain is known to contain at least five proteolytic enzymes but also non-proteolytic enzymes, including an acid phosphatase and a peroxidase; it may also contain amylase and cellulase activity. In addition, various other components are present.

Bromelain has previously been used in the treatment of a variety of conditions including inflammation and, in particular, it has been used in the treatment of diarrhoea. The use of bromelain in the treatment of infectious diarrhoea is described in WO-A-9301800, where it is suggested that bromelain works by destroying intestinal receptors for pathogens by proteo lysis, and in WO-A-8801506, which teaches that bromelain detaches pathogens from intestinal receptors.

Taussig et al, Planta Medica, 1985, 538-539 and Maurer et al, Planta Medica, 1988, 377-381 both suggest that bromelain may be of use in inhibiting tumour growth. US 5,223,406, DE-A-4302060 and JP-A-59225122 also teach the use of bromelain in the treatment of cancer. US 5,223,406 teaches that bromelain is capable of inducing tumour necrosis factor (TNF) while DE-A-4302060 teaches that bromelain can prevent metastasis by the structural modification of the tumour surface protein CD44.

In WO-A-9400147, various experiments were described which demonstrate that proteolytic enzymes and, in particular, bromelain, are capable of inhibiting secretion. The application also discloses that bromelain can reduce toxin binding activity and can inhibit the secretory effect of toxins such as heat labile toxin (LT) and cholera toxin (CT) and also toxins such as heat stable toxin (ST). This is in spite of the fact that ST has a very different mode of action from LT and CT. These observations were explained by the fact that one component of the bromelain mixture, stem bromelain protease, appears to be capable of modulating cyclic nucleotide pathways and this is discussed further in WO-A-9500169. In addition, bromelain has also been demonstrated to inhibit secretion caused by the calcium dependent pathway.

The present inventors have studied the varied biological effects of bromelain and, in particular, its effects in a well documented model of intracellular signal transduction, namely T cell receptor (TCR)/CD3 signalling and IL-2 production. Significant progress over recent years has led to the understanding of biochemical events which occur following TCR engagement (reviewed Cantrell, Annu. Rev. Immunol. 14, 259-274, (1996)), therefore TCR signalling provides an excellent model for elucidation of the effects of biologically active compounds. Effective T cell activation requires two signals. The first signal is generated by the TCR CD3 complex after engagement with antigen peptide presented by the major histocompatibility complex (MHC) expressed on antigen presenting cells (APC) (Cantrell, 1996). The second, costimulatory signal is generated by ligation of CD28 receptors on T cells with the B7 family of ligands on APC. A key element in the signalling pathway involved in transducing receptor-initiated signals to the nucleus is the family of mitogen-activated protein kinases (MAPk). The best studied of these kinases are the extracellular signal-regulated protein kinases ERK-1 and ERK-2 (also referred to as p44^MAPk and p42^MAPk, respectively). ER s are serine/threonine kinases that are activated when phosphorylated on tyrosine and threonine residues. In vitro, this activation is reversed if either residue is dephosphorylated. A relatively newly discovered member of the MAPk family are c-Jun NH₂-terminal kinases (JNKs) which exist as 46 kDa and 55 kDa forms that also require phosphorylation for activation. ERK activation is dependent on pSό¹^ and coupling of the TCR/CD3 complex to p21^Ras, with subsequent activation of the Raf-1/MEKl/ERK kinase cascade. INK activation also requires p21^Ras, as well as signals generated by the CD28 costimulatory receptor which activate GTP (guanosine triphosphate)-binding proteins (such as Racl or Cdc42) that induce the PAK/MEKK/SEK/JNK kinase cascade. Activated ERK phosphorylates Elk- 1, which in turn mediates induction of c-fos activity following phosphorylation of c-jun by JNK. Activated c-fos and c-jun combine to form the AP-1 protein required for IL-2 synthesis. The above events are summarised in Figure 1. All the above-mentioned signalling events require tyrosine phosphorylation, as inhibitors of protein tyrosine kinases (PTKs) inhibit many events associated with TCR stimulation, including T cell activation and IL-2 production.

The activation of T cells to produce interleukin 2 (IL-2) to drive T cell clonal expansion is an essential component of the immune response (Cantrell, 1996). The absence of this process can be fatal, as observed in people with T cell defects such as AIDS. However, the activation of T cells can also lead to detrimental consequences, such as autoimmune diseases and organ transplant rejection. Chronic autoimmune related disorders include rheumatoid arthritis, insulin-dependent diabetes mellitus and multiple sclerosis which arise because the body's immune system fails to distinguish "self from "non self and attacks normal tissue. In the case of organ transplants to replace diseased organs, rejection of the transplanted tissues may occur when T cells recognise the transplanted tissue as foreign and mount an attack against it. The continual activation of T cells may also exacerbate certain infectious diseases. For example, activation of T cells can lead to increased pathology of tuberculoid leprosy, schistosomiasis and visceral leishmaniasis.

Cyclosporin A, FK506 and Rapamycin are drugs currently used to block T cell responses and facilitate solid organ transplantation. These drugs prevent the production of IL-2, the major T-cell growth factor, by blocking T cell signal transduction. However, although these drugs are effective, cyclosporin A and FK506 are both toxic to kidneys and other organs. These compounds also impair the patient's defences against infections. Also treatment of patients with cyclosporin A is expensive. Therefore better and less expensive molecules are being sought.

In WO-A-9600082, we showed that bromelain could inhibit tyrosine phosphorylation and activation of ERK-2 in T cells stimulated via the TCR, or with combined phorbol ester plus calcium ionophore.

Because bromelain is a mixture of many different components, each of which has different activity, there are problems with its use as a pharmaceutical agent. In WO-A- 9838291 we described the isolation of various fractions of the bromelain mixture and showed that one of these fractions, designated CCS, can block T cell activation and thus may be of use as an immunosuppressive agent. Fraction CCS acts by blocking T cell signal transduction. Specifically, CCS inhibits tyrosine phosphorylation and activation of extracellular regulated kinase (ERK-2) and Raf-1, members of the mitogen activated protein kinase (MAP kinase) pathway (Avruch et al, (1994) J7RS, 19, 279-283). WO-A-9838291 also teaches that in association with decreased ERK-2 and Raf-1 activity, CCS inhibits interleukin 2 (IL-2) production by purified murine

CD4⁺ T cells and CD4⁺ T cell proliferation. Since Raf-1, is a well known proto- oncogene, CCS may also be of use as an anti-cancer agent. CCS may also prevent allergic reactions and treat parasitic infection.

Fraction CCS described in WO-A-9838291 is not a single protein but comprises ananain, comosain and various other components. We have now separated the CCS fraction into three subfractions and have identified various proteins within those subfractions.

Therefore, in a first aspect of the present invention there is provided a fraction of bromelain comprising: proteins of molecular weights 23.9kDa and 26.6 kDa when measured by SDS-PAGE or 23.459kDa and 23.6576kDa when measured by mass spectrometry; and

proteins having the following NH₂-terminal sequences:

Val Pro Gin Ser He Asp Trp Arg Asp Ser Gly Ala Val Thr Ser Arg Gly Glu Asn Pro Arg (SEQ ID NO: 1);

Val Pro Gin Ser He Asp Trp Arg Asn Tyr Gly Ala Val Thr Ser (SEQ ID NO: 2);

Val Pro Gin Ser He Asp Trp Arg Asp Ser Gly Ala Val Thr Ser Val Lys Asn Gin Gly Arg Xaa Gly Ser Xaa Xaa Ala Phe (SEQ ID NO: 3);

Asp Gly Ser Asn Asn Ala Arg Lys (SEQ ID NO: 4);

Val Pro Gin Ser He Asp Trp Arg Asp Ser Gly Ala Val Thr Ser Val Lys Asn Gin Gly Arg Asn Gly Ser Xaa He Ala Phe Xaa Ser Leu Xaa Xaa Xaa Pro (SEQ ID NO: 5); and

Val Pro Gin Ser He Asp Trp Arg Asp Tyr Gly Ala Val Thr Ser (SEQ ID NO: 6).

The protein fraction of the first aspect of the present invention has been designated CCS/P2 by the inventors and has been found to have a number of potentially useful activities. Firstly, we have found that it blocks ERK-2 phosphorylation, and therefore the MAP kinase cascade. In addition, it blocked IL-2 production and CD4⁺ T cell proliferation.

Some of the protein components of the CCS P2 fraction are known. The protein with the N-terminal sequence SEQ ID NO: 2 appears to be comosain, which has been described by Rowan et. al., (Biochemistry Journal 266 869-875 (1990)). The protein with the N- terminal sequence SEQ ID NO: 3 appears to be ananain, which was also described by Rowan et. al. (Arch. Biochem Biophys 267 (1) 262-270 (1988)) and the protein with N-terminal sequence SEQ ID NO: 6 seems to be stem bromelain protease (EC.3.4.22.32; Rowan et. al., Arch. Biochem Biophys 267 (1) 262-270 (1988) and Ritonja et al, (FEBS 247 (2) 419-424 (1989)). However, some of the protein components of this fraction are novel and, therefore, in a second aspect of the invention, there is provided:

a protein having a molecular weight of 26.6kDa as measured by SDS-PAGE and comprising the NH₂-terminal sequence:

Val Pro Gin Ser He Asp Trp Arg Asp Ser Gly Ala Val Thr Ser Arg Gly Glu Asn Pro Arg (SEQ ID NO: 1); or

a protein having a molecular weight of 23.9kDa as measured by SDS-PAGE and comprising the NH₂-terminal sequence: Asp Gly Ser Asn Asn Ala Arg Lys (SEQ ID NO: 4); or

a protein having a molecular weight of 23.9kDa as measured by SDS-PAGE and comprising the NH₂-terminal sequence:

Val Pro Gin Ser He Asp Trp Arg Asp Ser Gly Ala Val Thr Ser Val Lys Asn Gin Gly Arg Asn Gly Ser Xaa He Ala Phe Xaa Ser Leu Xaa Xaa Xaa Pro (SEQ ID NO: 5).

There was also some activity in another CCS fraction, designated CCS/P3 and therefore, in a further aspect of the invention, there is provided a fraction of bromelain comprising:

a major protein component of molecular weight 24.0kDa when measured by SDS- PAGE and proteins comprising the following NH₂-terminal sequences: Val Pro Gin Ser He Asp Trp Arg Asp Ser Gly Ala Val Thr Ser Xaa Xaa Gly Xaa Pro (SEQ ID NO: 7); and

Val Pro Gin Ser He Asp Trp Arg Asp Ser Gly Ala Val Thr Ser Xaa Lys Asn Gin Gly Arg (SEQ ID NO: 8).

One of the protein components of the CCS/P3 fraction is novel and, therefore, in a further aspect of the invention, there is provided a protein having a molecular weight of 24.0kDa as measured by SDS-PAGE and comprising the NH₂-terminal sequence: Val Pro Gin Ser He Asp Trp Arg Asp Ser Gly Ala Val Thr Ser Xaa Xaa Gly Xaa Pro (SEQ ID NO: 7).

The protein with the N-terminal sequence of SEQ ID NO:8 seems to be ananain.

Fraction CCS/P3 has been fractionated into further protein fractions designated as CCS/P3/S1, CCS/P3/S2, CCS/P3/S3, CCS/P3/S4, CCS/P3/S5 and CCS/P3/S6. The separation of these protein fractions is described in the examples below. The protein fractions were tested for biological activity and it was found that fractions CCS/P3/S1 and CCS/P3 P6 have a number of potentially useful activities. CCS/P3/S1 inhibits the production of interleukin 2, while CCS/P3/S6 has been shown to inhibit the phosphorylation of ERK-2 and therefore block activation of the mitogen activated protein kinase cascade.

N-terminal sequencing of the proteins in the active fractions CCS/P3/S1 and CCS/P3/S6 was carried out. CCS P3/S1 comprises a major protein of 26 kDa when measured by SDS-PAGE, has a high specific activity towards the Z-Arg-Arg-pNA substrate and comprises a protein having the following NH₂-terminal sequence; Val Pro Gin Ser He Asp Trp Arg Asp (or Asn) Tyr (or Ser) Gly Ala Val Thr Ser Val Lys Asn Gin Asn Pro (SEQ ID NO: 9).

This protein appears to be either stem bromelain protease or the protein which appears on the DDBJ/EMBL/GenBank database as accession No. D38532.

CCS/P3/S6 comprises a major protein of 24 kDa when measured by SDS-PAGE, has a high specific activity towards the BZ-Phe-Val-Arg-pNA substrate and comprises proteins having the following NH₂-terminal sequences;

Val Pro Gin Ser He Asp Tφ Arg Asp Ser Gly Ala Val Thr Ser Val Lys (or Gly or Asn) Pro Gin (or He or Asn) Gly Arg (SEQ ID NO: 10) and

Val Pro Gin Ser He Asp Tφ Arg Asp Ser Gly Ala Val Thr Ser (SEQ ID NO: 11).

CCS/P3/S1 and CCS/P3/S6 fractions of bromelain form further aspects of the present invention.

One of the proteins in CCS/P3/S6 is novel and therefore, in a yet another aspect of the invention, there is provided a protein having a molecular weight of 24 kda as measured by SDS-PAGE, having a substrate specificity for BZ-Phe-Val-Arg-pNA substrate and comprising the NH₂-terminal sequence:

Val Pro Gin Ser He Asp Tφ Arg Asp Ser (or Tyr) Gly Ala Val Thr Ser Val Lys (or Gly or Asn) Pro Gin (or He or Asn) Gly Arg (SEQ ID NO: 10).

The other protein in CCS/P3/S6, with N-terminal sequence SEQ ID NO: 11 has a sequence and enzymatic properties which correspond to those of ananain. The proteins of the present invention may be obtained by purification of pineapple stem extracts as described herein. However, it may be more convenient to obtain them by expression of nucleic acids encoding them. Therefore, in a further aspect, the present invention also provides nucleic acids encoding the novel proteins mentioned above.

The nucleic sequence encoding the proteins of the present invention may be obtained once the proteins are isolated and purified by techniques such as reverse transciptase polymerase chain reaction. Those skilled in the art will be well aware of this technique and of how to obtain the primers for amplification.

The nucleic acids may then be inserted into a suitable vector, for example a plasmid, cosmid or phage, which may then be used to transform a host so that expression of the required protein occurs in the host. These techniques are well known to those skilled in the art, who would be able both to determine the nucleic acid sequence encoding the protein and to express it to obtain the protein.

Although in WO-A-9724138, we stated that proteases in general are capable of decreasing MAP kinase activation, we have now found that this is not the case as trypsin does not abrogate T cell signalling and, indeed, in other studies, has been shown to increase MAPk activation (Belham et al, 1996, Biochem. J, 320, 939-946). Thrombin, a protease involved in the blood coagulation cascade, has also been shown to increase MAP kinase activation (Vouret-Craviari et al, 1993, Biochem. J., 289, 209-214). The inventors have now also shown that other proteases contained within the crude bromelain mixture do not block the activation of the MAP kinase pathway.

It is possible that the effects of CCS/P2 on the MAP kinase pathway in T cells are mediated by specific proteolytic effects at the cell surface. It is known that bromelain cleaves the CD45 RA isoform and selectively removes other surface molecules from human PBMCs. Bromelain also partially removes CD4 from T cell surfaces. Since CD45 and CD4 play an obligate stimulatory role in TCR-mediated T cell activation, CCS P2 may interfere with TCR signalling by affecting these molecules. Although the importance of CD45 and CD4 is well recognised for TCR-initiated signal transduction, it is possible to bypass their requirements for T cell activation by the use of phorbol ester and calcium ionophore. Use of combined phorbol esters and ionophore restores normal function to T cells which have been made refractory to TCR stimulation by the use of tyrosine kinase inhibitors or which are CD45 or p56^Lck deficient.

However, in the present study, the inventors have shown that normal function is not restored to T cells pre-treated with CCS/P2 when they are treated with PMA plus ionophore. The inhibitory effect of CCS/P2 on ERK-2 is thus not thought to be mediated via effects on CD45 or CD4 on T cells. CCS P2 possibly affects an as yet unidentified surface molecule, which, in turn, affects the MAP kinase pathway. The inhibitory effect of CCS P2 on cytokine production is thus not thought to be mediated via its effects on CD45 or CD4 on T cells. The inhibitory effect of CCS/P2 on T cell signal transduction was not because of toxicity of the compound, since CCS/P2 did not affect T-cell viability.

The subfraction CCS/P3/S6 described above has been shown to have a similar effect on ERK-2, and therefore on the MAP kinase pathway, as CCS/P2. CCS/P3 also has this effect, although to a lesser extent than either CCS/P2 or CCS/P3/S9. The subfraction CCS/P3/S1 does not affect ERK-2 but, suφrisingly, it has been shown to inhibit the production of IL-2.

Since we have shown that the CCS/P2 and CCS/P3/S6 components of the CCS fraction from crude bromelain block activation of the MAP kinase pathway and block T cell activation, CCS/P2 and CCS/P3/S6 may be of use in the treatment of T cell- mediated diseases. The crude CCS/P3 component has lower activity but will also have the same use.

In addition to its importance for IL-2 production and T cell activation, the MAP kinase pathway is also important for the production of growth factors such as epidermal growth factor (EGF), platelet derived growth factor (PGDF) and insulin-like growth factor (IGF). CCS/P2, CCS/P3/S6 and CCS/P3 will therefore block the production of these, and other, growth factors and the production of other cytokines such as IL-4, IFN-γ, GM-GSF and many more.

Thus, in summary, CCS/P2, CCS/P3/S6 and, to a lesser extent, CCS/P3 are all able to inhibit the MAP kinase pathway. As a consequence of this, they are also able to inhibit the activation of T cells and the production of growth factors and cytokines. These uses of CCS P2, CCS/P3/S6 and CCS/P3 form a further aspect of the invention.

CCS/P2, CCS/P3/S6 and, to a lesser extent, CCS/P3 will therefore be of use in a method for the treatment of diseases or conditions mediated by the activation of the MAP kinase pathway, the activation of T cells or the production of growth factors or cytokines, the method comprising administering to a patient in need of such treatment an effective amount of CCS P2, CCS/P3/S6 or CCS/P3.

CCS/P2, CCS/P3 and CCS/P3/S6 may also be of use for the treatment of cancer. The anti-tumour mechanism of action of CCS/P2 and CCS/P3/S6 remains to be determined but seems likely to be a result of the blocking of activation of the ERK-2 pathway.

As mentioned earlier, MAP kinase activation is dependent on p21^Ras and Raf-1, which are important oncogenes. p21^Ras and Raf-1 proteins help to relay signals from growth factor receptors on the surface of cells to MAP kinases to stimulate cell proliferation or differentiation. Oncogenic (or mutant) p21^Ras or Raf-1 genes produce defective proteins that have acquired independence from externally supplied growth factors and, at the same time, may no longer respond to external growth-inhibitory signals. Mutant p21^Ras or Raf-1 proteins are thus persistently hyperactive and their unbridled catalytic activity has a detrimental effect on the control of cell growth. Oncogenic p21^Ras or Raf-1 genes therefore promote cancer and tumour formation by disrupting the normal controls on cell proliferation and differentiation. Approximately 30% of human cancers have mutations in a p21^Ras gene.

Given that signals transmitted by p21^Ras and Raf-1 can be blocked via MAP kinase, CCS P2 and CCS/P3/S6 would be expected to block cancer and tumour growth. The protein of the present invention would therefore be useful for treating many different types of cancer including solid cancers such as ovarian, colon, breast or lung cancer and melanoma as well as non-solid tumours and leukaemia.

Thus, in a further aspect of the invention, there is provided CCS/P2, CCS/P3/S6 and crude CCS/P3 for use in the treatment or prevention of cancer, and, in particular, those types of cancer listed above.

There is also provided the use of CCS/P2, CCS/P3/S6 and crude CCS P3 in the preparation of an agent for the treatment or prevention of cancer, and, in particular, those types of cancer listed above.

The continual activation of T cells during chronic disease can also lead to pathological consequences, as can be found in certain chronic parasitic infections, such as chronic granulatomas diseases such as tuberculoid leprosy, schistosomiasis and visceral leishmaniasis. Furthermore, the invasion of parasites and pathogens, and their subsequent survival in cells, is dependent on these organisms utilising host cell signalling pathways (Bliska et al, 1993, Cell, 73, 903-920). For example, Salmonella has been demonstrated to phosphorylate MAP kinase, which allows for the bacteria to become endocytosed by macrophages (Galan et al, 1992, Nature, 357, 588-589). The bacteria then proliferate and destroy the cell. Because CCS/P2, CCS/P3 and CCS/P3/S6 have been shown to modify host signalling pathways, and, in particular, to inhibit MAP kinase, another of its potential applications could be to inhibit invasion by parasites and pathogens and their survival in cells.

Although it is less active than CCS/P2, the crude fraction CCS/P3 also has biological activity and will therefore have the same uses as CCS/P2.

In our earlier application WO-A-9600082 we discussed the inhibition of the MAP kinase cascade by crude bromelain. However, at that time, we were not able to determine which component of the crude bromelain mixture was responsible for this activity although we speculated that it might be stem bromelain protease. We have now discovered that in addition to blocking cyclic nucleotide pathways, stem bromelain protease does have some activity against the MAP kinase pathway. However, it is far less effective in blocking the MAP kinase cascade than the CCS/P2 mixture of the present invention or CCS/P3/S6. Indeed, we have now found that CCS/P2 is in the region of ten orders of magnitude more active than stem bromelain protease in blocking MAP kinase activation.

The inventors have also shown that although CCS/P3/S1 does not affect the MAP kinase pathway, it does block IL-2 production, which is important for T cell activation. Therefore it will also be of use in the treatment of diseases or conditions mediated by activation of T-cells and the production of growth factors or cytokines and this is a further aspect of the present invention.

The activation of the MAP kinase pathway in T cells to produce IL-2 and drive T cell clonal expansion is an essential component of the immune response. CCS/P2, CCS/P3, CCS/P3/S1 and CCS/P3/S6 inhibit this process and are therefore immunosuppressants. The absence of this process can have fatal consequences, as can be observed in people with AIDS or genetic mutations which result in T cell defects. However, the activation of T cells can also lead to detrimental consequences. For example, if autoreactive T cells are activated, autoimmune diseases can result. As a consequence of their immunosuppressive activity, CCS/P2, CCS/P3, CCS/P3/S1 and CCS/P3/S6 are likely to be of use in the treatment of autoimmune diseases such as rheumatoid arthritis, type-1 diabetes mellitus, multiple sclerosis, Crohn's disease and lupus.

Also, the activation of T cells specific for engrafted tissue can lead to graft or transplant rejection and so CCS/P2, CCS/P3, CCS/P3/S1 and CCS/P3/S6 may also be of use in preventing this.

The activation of allergen-specific T cells can cause allergic reactions. Inflammatory cytokines and other cellular products, such as histamine, are released from cells following exposure to allergens. The release of histamine and inflammatory cytokines involves the MAP kinase pathway and so blocking of the MAP kinase pathway with CCS/P2, CCS/P3 or CCS/P3/S6 or preventing the activation of T cells using CCS/P3/S1 is likely to be an effective treatment for allergies.

In addition, CCS/P2, CCS/P3, CCS/P3/S 1 and CCS/P3/S6 are all likely to be of use in the prevention of toxic shock and other diseases mediated by over production of bacterial endotoxins. Toxic shock is mediated by the production of lipopolysaccharides (LPS) from gram-negative bacteria. LPS triggers the production of TNF-α and interleukin- 1 via activation of the MAP kinase pathway in macrophages. The secretion of these cytokines elicits a cascade of cytokine production from other cells of the immune system (including T cells), which leads to leucocytosis, shock, intravascular coagulation and death.

A further use for CCS/P2, CCS/P3, CCS/P3/S1 and CCS/P3/S6 is in the prevention of programmed cell death (apoptosis). This is a special event whereby cells are stimulated to destroy their own DNA and die. It is an essential event in most immune responses (to prevent the accumulation of too many cells), but can also have immunosuppressive consequences in some instances, such as in HIV infection and ageing so that too many cells die and there are insufficient left to combat infection (Perandones et al, 1993, J. Immunol, 151, 3521-3529). Because the initiation of apoptosis is dependent on specific cell signalling events, including activation of the MAP kinase pathway, CCS/P2, CCS/P3, CCS/P3/S1 and CCS/P3/S6 are likely to be effective in blocking apoptosis.

The protein components of CCS/P2 and CCS/P3 have been partially identified. It is not yet clear which of the individual components is responsible for the biological activity of CCS/P2 and CCS/P3 but the active proteins can be used in any of the ways described above.

One component of CCS/P2 and CCS/P3/S1 appears to be comosain and it is possible that one of the novel proteins in combination with comosain may be responsible for the biological activity of CCS/P2 and/or CCS/P3/S1.

Similarly, ananain is a component of CCS/P2, CCS/P3 and CCS/P3/S6 and so it is possible that one of the novel proteins in combination with ananain may be responsible for the activity of these fractions.

In yet another aspect of the invention, therefore, there is provided a protein having the N-terminal sequence of SEQ ID NO: 1 , SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 7 or SEQ ID NO: 10 in combination with ananain and/or comosain for use in medicine.

Another component of CCS/P2, CCS/P3/S1 and CCS/P3/S6 appears to be stem bromelain protease or the protein which appears on the DDJB/EMBL/GenBank database as accession No: D38532. Therefore, it is possible that one of these proteins in combination with ananain and/or comosain and/or one of the novel proteins may be responsible for the activity of the fractions.

CCS/P2, CCS/P3, CCS/P3/S1, CCS/P3/S6 or one of the individual components will usually be formulated before administration to patients and so, in a further aspect of the invention there is provided a pharmaceutical or veterinary composition comprising CCS/P2, CCS/P3, CCS/P3/S1, CCS/P3/S6 or a protein of amino terminal sequence of SEQ ID NO: 1, 4, 5, 7 or 10, optionally combined with comosain and/or ananain, together with a pharmaceutically or veterinarily acceptable excipient or adjuvant.

The substance of the invention may be administered by a variety of routes including enteral, for example oral, nasal, buccal, topical or anal administration or parenteral administration, for example by the intravenous, subcutaneous, intramuscular or intraperitoneal routes.

In many cases, the oral route may be preferred as this is often the route which patients find most acceptable. The oral route may be particularly useful if many doses of the protein are required.

When oral administration is chosen, it may be desirable to formulate the bromelain fraction or protein of the invention in an enteric-coated preparation in order to assist its survival through the stomach. Alternatively, another orally administrable dosage form may be used, for example a syrup, elixir or a hard or soft gelatin capsule, either of which may be enteric coated.

However, under certain circumstances, it may more convenient to use a parenteral route. For parenteral administration, the protein may be formulated in distilled water or another pharmaceutically acceptable solvent or suspending agent. A suitable dose of the bromelain fraction or protein of the invention to be administered to a patient may be determined by the clinician. However, as a guide, a suitable dose may be from about 0.5 to 20 mg per kg of body weight. It is expected that in most cases, the dose will be from about 1 to 15 mg per kg of body weight and preferably from 1 to 10 mg per kg of body weight. For a man having a weight of about 70 kg, a typical dose would therefore be from about 70 to 700 mg.

The invention will now be further described with reference to the following examples and to the drawings in which:

FIGURE 1 is a diagrammatic representation of signal transduction events associated with T cell activation that leads to IL-2 production.

FIGURE 2 is an ultra violet elution profile that shows the fraction CCS of crude stem bromelain after cation exchange chromatography on SP Sepharose high performance media.

FIGURE 3 is an ultra violet chromatogram of the CCS/PI, CCS/P2 and CCS/P3 subcomponents of CCS obtained after Gel Permeation Chromatography. Peak 1 is CCS/PI, Peak 2 is CCS/P2, Peak 3 is CCS/P3.

FIGURE 4 is a graph showing that CCS/P2 inhibits ERK-2 phosphorylation in T cells. GA15 were treated with CCS or CCS subfractions (CCS/PI, CCS/P2, CCS/P3, 25 μg/ml) for 30 min, washed and then stimulated with PMA plus ionophore for 5 mins. Cells were then lysed and postnuclear supernatants were subjected to SDS-PAGE and immunoblotted with phospho-specific ERK antibody. Phosphorylation of ERK-2 was quantitated by densitometry of bands revealed by chemo luminescence. FIGURE 5 is a pair of plots showing that CCS/P2 decreases proliferation of purified CD4⁺ T cells. T cells were treated with CCS/P2 (25 μg/ml), washed, then cultured in either media alone, or with immobilised anti-CD3ε mAb plus soluble anti-CD28 mAb. Proliferation was determined by the incoφoration of ³H- thymidine. CD4⁺ T cells cultured in the absence of mAb (stimuli) did not proliferate.

FIGURE 6 is a pair of plots showing that CCS/P2 decreases IL-2 production in purified CD4+ T cells. T cells were treated with CCS and CCS subfractions (25 μg/ml), washed, then cultured in either media alone or with immobilised anti-CD3ε mAb and soluble anti-CD28 mAb. IL-2 production was determined by the CTL-L assay (Gillis et al, 1978, J. Immunol. 120, 2027-2032.). CD4⁺ T cells cultured in the absence of mAb (stimuli) did not produce any detectable IL-2.

FIGURE 7 is an SDS-PAGE gel of CCS/P2 and CCS/P3 subcomponents obtained after Gel Permeation Chromatography.

FIGURE 8 is an ultra violet chromatogram of the CCS/PI, CCS/P2 and CCS/P3 subcomponents of CCS obtained using a modified method of Gel Permeation chromatography. Pool No: 1 is CCS/PI, Pool No: 2 is CCS/P2 and Pool No: 3 is CCS/P3.

FIGURE 9 is a 12% SDS-PAGE gel of CCS/P2 and CCS/P3 subcomponents obtained using the modified Gel permeation Chromatography method.

FIGURE 10 is an ultra violet chromatogram of the CCS/P3/S1, CCS/P3/S2, CCS/P3/S3, CCS/P3/S4, CCS P3/S5, CCS/P3/S6 subcomponents of CCS/P3 obtained using a Superdex 75™ HR 10/30 column. Pool 1 is CCS/P3/S1, Pool 2 is CCS/P3/S2, Pool 3 is CCS/P3/S3, Pool 4 is CCS/P3/S4, Pool 5 is CCS/P3/S5, and Pool 6 is CCS P3/S6. FIGURE 11 - is a 12% SDS-PAGE gel of CCS/P3/S1 to CCS/P3/S6 subcomponents obtained using a Superdex 75™ HR 10/30 column.

FIGURE 12 is a graph showing that CCS and CCS/P3/S6, but not stem bromelain protease (SBP) or CCS/P3/S1, inhibits ERK-2 phosphorylation in T cells. GA15 cells were treated with Fraction CCS, SBP, CCS/P3/S1 or CCS/P3/S6 (20 μg/ml) for 30 min, washed and then stimulated with PMA plus ionophore for 5 mins. Control GA15 cells were treated with an equal amount of the same buffer that the CCS samples were suspended in. Cells were then lysed and postnuclear supernatants were subjected to SDS-PAGE and immunoblotted with phospho-specific ERK antibody. Phosphorylation of ERK-2 was quantitated by densitometry of bands revealed by chemoluminescence.

FIGURE 13 shows a graph which compares ERK activity versus the specific enzyme activity of different CCS fractions. A) ERK activity versus Bz-Phe-Val-Arg-pNA activity; B) ERK activity versus Arg-Arg activity. Data show that ERK inhibitory activity correlates with Bz-Phe-Val-Arg activity.

FIGURE 14 is a pair of plots showing that CCS/P3/S1 decreases IL-2 production in purified CD4+ T cells. T cells were treated with CCS and CCS/P3/S1 and CCS/P3/S6 subfractions (20 μg/ml), washed, then cultured in either media alone or with immobilised anti-CD3ε mAb and soluble anti-CD28 mAb. IL-2 production was determined by the CTL-L assay (Gillis et al, 1978, J. Immunol. 120, 2027-2032.). CD4⁺ T cells cultured in the absence of mAb (stimuli) did not produce any detectable IL-2.

EXAMPLE 1 - Purification of CCS subfractions a. Materials. Bromelain (E.C 3.4.22.4; proteolytic activity, 1,541 nmol/min/mg; Cortecs Batch Ref: 24077) was purchased from Polyamine Coφoration (Taiwan). SP Sepharose HP, Pharmalyte 3-10, PD 10 desalting columns, and Hitrap Q were obtained from Pharmacia Biotech. All other reagents were of analytical grade and obtained from either Sigma Chemical Co. or British Drug House.

b. Method. CCS was prepared from crude stem bromelain (Cortecs Batch ref. QC24077) as previously described (WO-A-9838291), but with minor modifications. Briefly, crude stem bromelain was fractionated on a SP Sepharose HP Cation exchange column. The column was prepared by packing 200 ml of media into an XK 50/20 column (Pharmacia Biotech) and equilibrated with 5 column volumes of acetate buffer (20 mM, pH 5.0) containing 1 mM disodium EDTA on a Gradifrac chromatography system (Pharmacia) at 48.8 ml/min. 48.8 ml of crude stem bromelain solution (30 mg/ml) was loaded onto the column. The unbound protein was collected and the column was then washed with 243 ml of acetate buffer. Proteins bound to the column, were eluted with a linear gradient of 0 to 0.65 M NaCl in acetate buffer over 12 column volumes. Fractions (15 ml) were collected throughout the gradient. The CCS fraction, being the last double peak from the column was collected. CCS fractions were pooled from 10 different column runs and analysed for protein content, proteolytic activity and by SDS-PAGE.

Pooled fraction CCS (60 ml) was desalted by the use of PD10 columns equlibrated in acetate buffer. The desalted CCS (82 ml) was then concentrated by loading samples (at 1 ml/min) onto a Q Sepharose Hitrap column (1 ml) equilibrated in acetate buffer. Bound CCS protein was eluted stepwise by addition of 3.5 ml of acetate buffer containing 1.0 M NaCl. Concentrated CCS was then further purified by gel permeation chromatography (GPC). GPC was performed on a Biocad Perseptive Biosystems Chromatographic Workstation using a Biosep SEC S2000 column (3000 x 7.8 mm, 5μ:Phenomenex, U.K) equlibrated with 5 column volumes of acetate buffer at a flow rate of 1 ml/min. 500 μl of concentrated CCS was loaded onto the column. Proteins were eluted over 2 column volumes and 0.5 ml fractions were collected. The proteolytic activity and protein content was determined on CCS subfractions pooled from six runs. Samples were then transported to Imperial College, London, on ice whereupon they were immediately aliquoted and stored at -80°C until required for in vitro studies as described below.

c. Results. Figure 2 shows a typical U.V. chromatogram of crude stem bromelain following cation exchange chromatography on SP Sepharose High Performance media. The major peaks are designated as CCS through to CCZ as previously described in WO-A-9838291.

Fraction CCS was further fractionated into subfractions by gel permeation chromatography. Three distinct peaks were obtained and named as indicated in Table 1 and are shown in Figure 3.

Table 1 Summary of pooled CCS subfractions fractions from Gel Permeation

Chromatography

EXAMPLE 2 - Effect of CCS/PI, CCS/P2 and CCS/P3 on activation of ERK-2. Phorbol ester and ionophore stimulation of T cells act synergistically to reproduce many features of TCR stimulation such as IL-2 secretion, IL-2 receptor expression, and T cell proliferation (Truneh et al., 1985, Nature 313, 318-320; Rayter et al., 1992, EMBO 11, 4549-4556). Phorbol esters can mimic antigen receptor triggering and bypass TCR-induced protein tyrosine kinases to activate ERK-2 by a direct agonist

Ras action on PKC and p21 . Calcium ionophore A23187 induces increased intracellular release of Ca^ ^" and therefore mimics the action of inositol 1,4,5-trisphosphate (IP3).

Phorbol esters and ionophore however, stimulate PKC pathways that are not controlled by the TCR (Izquierdo et al, 1992, Mol. Cell. Biol. 12, 3305-3312) suggesting separate intracellular pathways within T cells that regulate T cell function.

We have previously shown that Fraction CCS blocks tyrosine phosphorylation of ERK-2 following stimulation of T cells with combined PMA plus calcium ionophore (WO-A-9838291). We therefore investigated which sub-fraction of CCS could block ERK-2 signalling via the TCR-independent pathway by examining its effect on PMA and ionophore-induced tyrosine phosphorylation.

a. Materials. Rabbit polyclonal phospho-specific MAPk IgG which recognises tyrosine phosphorylated ERK-1 and ERK-2 kinases were from New England BioLabs (Hitchin, Hertfordshire, UK). Goat anti-rabbit IgG Ab conjugated to horse radish peroxidase (HRP) were from BioRad (Hemel Hemstead, Hertfordshire, UK).

The T cell hybridoma GA15 was a generous gift from B. Fox (ImmuLogic Pharmaceutical Coφoration, Boston, MA). GA15 was generated by fusing the thymoma BW5147 with the Th2 clone F4 specific for KLH in association with I-Ab, and were maintained as previously described (Fox, 1993, Int. Immunol. 5, 323-330). GA15 exhibit a ThO cell phenotype as they produce IL-2, IL-4 and IFN-γ following stimulation with crosslinked anti-CD3ε mAb (Fox, 1993, Int. Immunol. 5, 323-330). b. Method. Cells (2 x 10⁷) suspended in RPMI 1640 were treated with

CCS, CCS/PI, CCS/P2 and CCS/P3 (0 to 25 μg/ml) diluted in acetate buffer (20 mM, pH 5) containing EDTA (1 mM) for 30 min at 37°C. Mock treated cells were treated with an equal volume of acetate/EDTA buffer. At high concentrations of CCS (25 μg/ml), CCS/P2 and CCS/P3 cell aggregation occurred, as noted previously in studies with crude bromelain (WO-A-9838291). Following treatment, cell aggregates were gently dispersed by washing the cells 3 times and then resuspending in fresh RPMI. Control T cells were treated in an identical manner.

T cells were stimulated with combined phorbol 12-myristate 13-acetate (PMA) (20 ng/ml) and calcium ionophore A23187 (1 μM) for 5 minutes. Stimulation was terminated by the addition of ice-cold lysis buffer (25 mM Tris, pH 7.4, 75 mM NaCl, 2 mM EDTA, 0.5% Triton X-100, 2 mM sodium orthovanadate, 10 mM sodium fluoride, 10 mM sodium pyrophosphate, 74 μg/ml leupeptin, 740 μM PMSF and 74 μg/ml aprotinin) for 30 min with continual rotation at 4°C. Lysates were clarified (14,000 x g for 10 min) and an equal volume of 2 X SDS-PAGE sample buffer (50 mM Tris, pH 7, 700 mM 2-ME, 50% (v/v) glycerol, 2% (w/v) SDS, 0.01% (w/v) bromophenol blue) was added to postnuclear supernatants. Proteins were solubilised at

. . 6 100°C for 5 min and samples containing 1 x 10 cell equivalents were resolved by SDS-PAGE.

Separated proteins were transferred to nitrocellulose membranes (Bio-Rad) which were then blocked with 5% (w/v) bovine serum albumin (Sigma, fraction V; BSA), 0.1% Nonidet P-40 in Tris-buffered saline (170 mM NaCl and 50mM Tris, pH 7.4; TBS). Immunoblots were incubated with phospho-specific MAPk IgG overnight at 4°C followed by goat anti-rabbit IgG Ab conjugated to horse radish peroxidase for 1 hr at room temperature. Antibodies were diluted in antibody dilution buffer comprised of 0.5%) (w/v) BSA, 0.1% (v/v) Tween-20 in TBS. Following each incubation step, membranes were washed extensively with 0.1% Tween-20 in TBS. Immunoreactivity was determined using the ECL chemiluminescence detection system (Amersham Coφ., Arlington Heights, IL).

c. Results Stimulation of T cells with combined ionophore and PMA induced tyrosine phosphorylation of ERK-2 as expected. CCS pre-treatment reduced tyrosine phosphorylation of ERK-2 consistent with our results obtained earlier (WO- A-9838291). Of the CCS subfractions tested, CCS subfraction P2 (CCS/P2) had the most consistent anti-ERK-2 activity (Fig 4). In some experiments, CCS/P3 also displayed some activity, albeit at a much lower level than that observed with CCS/P2. This data suggests that CCS/P2 contains the active component which blocks ERK-2 activation. The reduced amount of activity observed in CCS/P3 suggests that some of the active component may be within this fraction.

EXAMPLE 3 - Effect of CCS/PI, CCS/P2 and CCS/P3 on Interleukin 2 production and CD4+ T cell proliferation

Ras

Activation of p21 , Raf-1, MEK-1 and ERKs are essential for induction of IL-2 transcription in T cells (Izquierdo et al., 1993, J. Exp. Med. 178, 1199). IL-2 is the major autocrine T cell growth factor which induces proliferation of T cells. Therefore the ability of CCS/P2 to block ERK activation, suggests that it would be expected to inhibit IL-2 production and T cell proliferation. We therefore investigated whether CCS/P2 could effect a functional outcome of T cell signalling, namely IL-2 production and T cell proliferation in highly purified murine CD4⁺ T cells.

a. Materials and Methods. Anti-CD3ε-chain mAb (145-2C11) and anti-

CD28 mAb (PV-1) were purchased from Pharmingen (San Diego, CA) and goat anti- hamster IgG Ab was from Sigma (Dorset, UK). Splenocytes were isolated from female BALB/c mice (6-8 weeks old), as previously described (PCT/GB95/01501). Highly purified CD4⁺ T cells were isolated from splenocytes using magnetic activated cell sorting (MACS).

Purified CD4⁺ T cells diluted in RPMI were treated with CCS subfractions (10 μg/ml and 25 μg/ml) or diluent alone at 37°C for 30 min, washed in fresh RPMI and then resuspended in culture medium. For measurements of IL-2 production, T cells were stimulated with immobilised anti-CD3ε (4 μg/ml) and soluble anti-CD28 mAb (10 μg/ml). Anti-CD3ε mAb, diluted in PBS, was immobilised to 24-well, flat bottom, microculture plates (Corning, Corning, NY) by incubation for 16 hours at 4°C. Wells were then washed three times in PBS prior to adding triplicate cultures of purified

CD4+ T cells (2.5 - 5 x 10⁶ cells per well). Cells were then incubated at 37°C in humidified 5% CO₂ for 24 h. IL-2 levels in the culture supernatant were measured using the CTL-L bioassay (Gillis et al, 1978, J. Immunol. 120, 2027-2032).

For measurements of T cell proliferation, cells were treated with CCS subfractions as above, and then similarly stimulated with immobilised anti-CD3ε (4 μg/ml) and soluble anti-CD28 mAb (10 μg/ml). Cells were then cultured in 96 well, flat-bottom plates (Nunc) at 10^ cells per well for 36 h. Cultures were pulsed with 0.5 μCi of [³H]TdR 12 h prior to harvesting onto glass fibre filters.

b. Results CCS/P2 (10 and 25 μg/ml) inhibited CD4⁺ T cell proliferation when the ERK pathway was stimulated with anti-CD3ε mAb (Fig 5). CCS/PI and CCS/P3 had a small effect on proliferation, which was significantly less than the effect of CCS/P2. CCS/P2 also reduced IL-2 production (25 μg/ml), whilst CCS/PI and CCS P3 had no effect on IL-2 production (Fig 6). This data clearly indicate that CCS/P2 blocks T cell activation. IL-2 production and proliferation were dependent on cell stimulation with anti-TCR antibodies as no cytokine was detected in cells cultured in tissue culture media alone (Fig 5 and 6). EXAMPLE 4. SDS-PAGE on fractionated samples. a. Materials. Broad range molecular weight markers, kaliedoscope markers, and Sequi-Blot™ PVDF membranes were obtained from Biorad. Precast 12 and 14 % Tris/Glycine acrylamide gels, 4-20% Tris/Glycine acrylamide gradient gels and running buffers (Mes nuPAGE and Tris/Glycine/SDS continuous) were obtained from Novex. E-64 (L-trans-Epoxysuccinyl-leucylamido(4-guanidino)butane, a selective cysteine protease inhibitor (Barrett et al, 1982, Biochem. J. 201, 189-198), was from Sigma. All other reagents were of analytical grade and obtained from either Sigma Chemical Co. or British Drug House.

b. Method. Samples were prepared for electrophoresis by first inhibiting enzyme activity with E-64 in which 50 μl of sample (0.2 mg/ml) was mixed with 1 μl of E-64 (1 mM). The samples were then mixed in SDS-PAGE sample buffer (62.5 mM Tris-HCl, pH 6.8, containing 10% (v/v) glycerol, 2% (w/v) SDS, 160 mM dithiothreitol and 0.02% w/v bromophenol blue) and heated to 95°C for 5 min in a water bath.

Broad range SDS-PAGE molecular weight standards diluted (1 :20) in SDS-PAGE sample buffer were treated similarly and run with the samples. Continuous or gradient gels were run on an Xcell II Mini-Cell™ (Novex) or mini Protean II™ (BioRad) electrophoresis system at 200 V until the dye front reached the end of the gel (30 to 45 min). After electrophoresis, separated proteins were stained overnight with orbital mixing in either Gelcode™ blue colloidal Coomassie stain (Pierce) or Coomassie blue R20 stain (BioRad). Gels were destained, to obtain a clear background, in a solution of 25%) (v/v) methanol and 10% (v/v) acetic acid. c. Results The purity of the CCS/P2 and CCS/P3 samples as shown by

SDS-PAGE are in Figure 7. The calculated molecular weights of the bands are described in Table 2.

Table 2 Summary of the molecular weights of proteins found in biologically active CCS/P2 and CCS/P3 subfractions as determined by SDS- PAGE.

In CCS/P2 the molecular weights of the major protein species are 23.9 and 26.6 kDa. The major protein present in CCS/P3 is of molecular weight of approximately 24.0 kDa. Proteins of this size have been reported previously for cysteine proteinases isolated from stem bromelain by other authors (Rowan and Buttle, 1994, Methods in Enzymology 244, 555-568). The similar size of 23.9 kDa in CCS/P2 and the 24.0 kDa protein in CCS/P3 suggests that these proteins might be the same species.

The molecular weights of the separated protein species in the CCS/P2 and CCS/P3 is consistent with our earlier report where we noted that the CCS fraction of bromelain is comprised of three major protein bands of 15.07, 25.9 and 27.45 WO-A-9838291). The slight difference in the reported molecular weights may be because of the different amounts of protein loaded on the gels, or may reflect differences in the precision of the analysis of the gels.

EXAMPLE 5 Protease and Protein Assay. a. Materials and Method. Protease activity was determined as previously described using the substrates Z-Arg-Arg-pNA and Bz-Phe-Val-Arg-pNA (Napper et al, 1994, Biochem. J. 301, 727-735) purchased from Bachem (U.K.) Ltd. These substrates were selected since they have been used previously for investigating the substrate specificity of proteinases isolated from pineapple stems. Z-Arg-Arg-pNA has been shown to be a good substrate for stem bromelain proteinase and comosain (Napper et al, 1994, Biochem. J. 301, 727-735), whilst Bz-Phe-Val-Arg-pNA is a good substrate for acidic fruit bromelain protease and ananain (Rowan et al, 1990, Biochem. J. 266, 869-875).

Protein was measured using a DC protein assay kit supplied by Bio-Rad based on the method of Lowry et al. (1951, J. Biol. Chem. 193, 265-275). Protein samples were compared to bovine serum albumin standards (0 to 1.0 mg/ml) prepared in either acetate buffer (20 mM, pH 5.0) which contained 1 mM disodium salt EDTA, acetate buffer plus NaCl (1 M), or Milli Q, as appropriate.

b. Results. The proteolytic activity against the synthetic peptides Z-Arg-

Arg-pNA and Bz-Phe-Val-Arg-pNA and the protein content of individual fractions are shown in Table 3.

Table 3 Calculated protein content, Bz-Phe-Val-Arg-pNA and Z-Arg-Arg-pNA activity of crude bromelain fractions and CCS subfractions.

1216.82 388.95

0

498.3

50.3

The main Z-Arg-Arg-pNA activity was found in the three major peaks (Fraction CCW, CCY and CCU) that correspond to stem bromelain proteinase (SBP). Fraction CCS, the later eluting fraction comprised of ananain, comasain and other unidentified components, has the highest Bz-Phe-Val-Arg-pNA activity. Fraction CCX also had moderate Bz-Phe-Val-Arg-pNA activity.

The CCS subfraction, designated CCS/PI does not have any activity against either of the substrates tested, suggesting that this protein is not a proteolytic enzyme. CCS/P2 reacted against both substrates, although it appears to have a higher specific activity against Bz-Phe-Val-Arg-pNA. CCS/P3 also has activity against both substrates, but the activities are much lower than CCS/P2. Also, CCS/P3 clearly has a higher preference for the Bz-Phe-Val-Arg-pNA substrate. Based on literature reports, ananian and fruit bromelain have a higher preference for the Bz-Phe-Val-Arg-pNA substrate, whereas comosain has a preference for Z-Arg-Arg-pNA (Napper et al, 1994, Biochem. J. 301, 727-735). Based on the substrate preferences obtained for CCS subfractions, the data indicate that CCS/P3 predominantly contains the enzyme ananain. CCS/P2 appears may also contain ananain as well as the enzyme comosain.

EXAMPLE 6 - Mass Spectrometrv

Since CCS/P2 was found to contain the highest biological activity, further analysis of this fraction was undertaken. a. Method. CCS/P2 (2000 pmoles) obtained from gel permeation chromotography was desalted by dialysis overnight against MilliQ water and then concentrated by a centrifugal evaporator to 1 mg/ml. The samples were then sent for mass spectrometry analysis at the Michael Barber Centre at UMIST. Samples were diluted in 50% (v/v) acetonitrile containing 0.1% (v/v) formic acid and analysed using a VG Quattro tandem quadruple spectrometer. The electrospray ion source utilised an infusion rate of 2 μl/min. The instrument was calibrated as per the manufacturer's instructions and the spectra obtained was processed by a MassLynx data system on the "MaxEnt" iterative alogorithm.

b. Results. A summary of the molecular weight species found by mass spectrometry of CCS/P2 are shown in Table 4.

Table 4 Summary of the molecular weights of species proteins found in

CCS/P2 as determined by mass spectrometry.

Earlier SDS-PAGE analysis of CCS/P2 only revealed two major protein species of 23.9 and 26.6 kDa. However mass spectrometry, a more precise technique for determining molecular weights, revealed five species in CCS/P2. The two lowest Mr were of 5,667.2 Da and 5,767.4 Da. These proteins are probably members of the cysteine proteinase inhibitors found in stem bromelain extracts (Hatano et al, 1998, J. Biochem. (Tokyo), 124, 457-461). The two species of 23,459 Da and 23,657.6 Da are most likely enzymes, which may correspond to the 23.9 and 26.6 kDa proteinases revealed by SDS-PAGE. The difference in molecular weight reported between the two techniques is most likely caused by the differences in the sensitivitiy of the two techniques.

Several different molecular masses for bromelain enzymes have been reported. The Mr of ananain has been reported to range from 23,411 to 23,478 (Lee et al, 1997, Biochem J. 327, 199-202; Napper et al, 1994, Biochem. J. 301, 727-735). The different masses of ananain may be attributed to the presence of alternative active and inactive forms (Napper et al, 1994, Biochem. J. 301, 727-735). Two active forms of ananain have been reported, one of which has an Mr of 23,420. The inactive forms have a Mr of 23,452, 23,499 and 23,580. The inactive forms are thought to arise from oxidation of the active site thiol which accounts for its increase in mass relative to the active form (i.e. Cys-SH is converted into Cys-SO₂H) (Napper et al, 1994, Biochem. J. 301, 727-735). Other minor structural variations, such as glycosylation of the enzyme may explain the differences in mass. Other authors report that the Mr of ananain is 23,478 which closely resembles the theoretical mass of 23,464 Da based on the amino acid sequence (Lee et al, 1997, Biochem J. 327, 199-202). The Mr of the F9 protein, putatively identified as ananain, is reported to be 23,427 Da (Harrach et al, 1995, J. Protein. Chem. 14, 41-52; Napper et al, 1994, Biochem. J. 301, 727-735).

Various forms of comosain have also been described. These have a Mr of 24,509 and 23,569 (Napper et al, 1994, Biochem. J. 301, 727-735). The Mr of stem bromelain protease has been reported to be 24,395 (Napper et al, 1994, Biochem. J. 301, 727- 735). This value closely resembles the Mr of F4 protease reported to be 24,397. F4 and F5 protease (Mr of 24,472 Da), most likely correspond to stem bromelain proteinase, since the NH₂-terminal sequence of these enzymes are identical (Harrach et al, 1995, J. Protein. Chem. 14, 41-52).

The two species of 23,459 and 23,657.6 Da in CCS/P2 do not precisely match the Mr of ananain, comosain or any other bromelain species previously published. It is possible that the proteins in CCS/P2 represent novel proteases or, alternatively, may represent previously unidentified alternate forms of ananain, comosain or stem bromelain protease.

EXAMPLE 8 - Purification of CCS/P2 and CCS/P3 subfractions - modified method

We next used a modified method to that described in EXAMPLE 1 to obtain the CCS/P2 and CCS/P3 subfractions. This modified method was used to try to further resolve the different proteins in the two CCS subfractions.

a. Method. CCS was prepared from crude stem bromelain (Cortecs Batch ref. QC24077) as described In EXAMPLE 1, but with the following modifications. Briefly, crude stem bromelain was fractionated at 40 ml/min (instead of 48.8 ml/min) on a SP Sepharose HP Cation exchange column. The buffer used for fractionation contained 0.1 mM disodium EDTA (instead of ImM disodium EDTA).

Pooled fraction CCS was concentrated to 10 - 25 mg/ml by the use of Vivaspin™ units (4 ml or 15 ml capacity units; Vivascience Ltd.) with a 5,000 da molecular weight cut off (instead of buffer exchange and concentration using PD10 columns and Q- Sepharose HP). Concentrated CCS was then further purified by gel permeation chromatography (GPC) using acetate buffer that contained 20mM actetate, 1 mM disodium EDTA, 0.25 M NaCl, pH 5.0) (instead of acetate buffer that did not contain NaCl). 0.1 ml to 0.2 ml of concentrated CCS was loaded onto the GPC column instead of 0.5 ml. Proteins were eluted over 2 column volumes and 0.2 ml fractions were collected instead of 0.5 ml fractions.

Protein fractions obtained were subjected to SDS-PAGE analysis, protein and enzyme analysis as described previously in EXAMPLE 4 and EXAMPLE 5. b. Results A typical U.V. chromatogram obtained using the above- mentioned modifications is shown in FIGURE 8. Three distinct peaks were obtained which were named as CCS/Pl, CCS/P2 and CCS/P3, as previously. These peaks are further described in Table 5. The elution profile obtained using the modified method is substantially different to the elution profile obtained using the method described in EXAMPLE 1 (Compare FIGURE 3 with FIGURE 8). The difference in the elution profile may be caused by an increase in the ionic strength of the buffer due to the addition of NaCl. The change in ionic strength of the buffer may have altered the ionic interactions between protein species present and the gel matrix.

Samples of CCS/Pl, CCS/P2 and CCS/P3 were transported to Imperial College,

London, on dry ice whereupon they were thawed, aliquoted and stored at -80°C until they were tested for anti-ERK activity as described in EXAMPLE 2.

Table 5 Summary of pooled CCS subfractions fractions from Gel Permeation

Chromatography (modified method).

The purity of the CCS/P2 and CCS/P3 samples obtained by the modified method as shown by SDS-PAGE are in Figure 9. The approximate molecular weights of the bands are described in Table 6. Table 6 Summary of the approximate molecular weights of proteins found in CCS subfractions CCS/P2 and CCS/P3 obtained by the modified method as determined by SDS-PAGE.

In CCS/Pl the molecular weight of the major protein species is 13 kda, as previously observed using the method described in EXAMPLE 1. Using the modified method to obtain CCS/P2, the molecular weights of the major protein species obtained are approximately 13 kDa, 24, 26 and 28 kDa. In CCS/P3, the major protein species obtained are 24 and 26 kDa.

Earlier using the method described in EXAMPLE 1, the CCS/P2 fraction did not contain the 13 kda band. The 28 kda band obtained using the modified method may be the 26.6 kda band previously obtained in EXAMPLE 1 or a new protein species not previously revealed under the conditions employed in EXAMPLE 1. Earlier CCS/P3 predominantly contained a 24 kDa protein as the major protein species. Using the modified method, an additional 26 kda protein species became apparent. Therefore, it appears that the modifications to the method of preparing CCS subfractions alters the composition of the major protein species present in the CCS/P2 and CCS/P3 fractions (compare Table 2 and Table 6).

The proteolytic activity against the synthetic peptides Z-Arg-Arg-pNA and Bz-Phe- Val-Arg-pNA and the protein content of CCS/Pl, CCS/P2 and CCS/P3 fractions obtained by the modified method are shown in Table 7. Table 7 Calculated protein content, Bz-Phe-Val-Arg-pNA and Z-Arg-Arg-pNA activity of CCS subfractions.

CCS/Pl obtained by the modified method has negligible proteolytic activity against both substrates as previously observed in EXAMPLE 1. CCS/P2 reacted against both the Bz-Phe-Val-Arg-pNA and Z-Arg-Arg-pNA substrates. However, in contrast to CCS/P2 obtained earlier (as prepared in EXAMPLE 1) which had a preference for the Bz-Phe-Val-Arg-pNA substrate, CCS/P2 obtained by the modified method had a preference for the Z-Arg-Arg substrate (compare Table 3 and Table 7).

CCS/P3 obtained by the modified method also had a different profile from the CCS/P3 obtained in EXAMPLE 1. Earlier, CCS/P3 had very low specific activity against Z- Arg-Arg-pNA, but moderate activity against the Bz- Phe-Val-Arg-pNA substrate. CCS/P3 prepared by the modified method now has a much higher total specific activity than CCS/P3 prepared earlier and has a higher specific activity than CCS/P2. CCS/P3 prepared by the modified method still retains its preference for the Bz-Phe- Val-Arg-pNA substrate.

The differences observed in the specific activities of CCS/P2 and CCS/P3 fractions is consistent with changes in the composition of the major protein species present in the CCS/P2 and CCS/P3 fractions observed by SDS-PAGE.

To determine whether the CCS subfractions prepared by the modified method were biologically active, they were tested for anti-ERK activity as described in EXAMPLE 2. CCS/P2 and CCS/P3 (at 20 μg/ml) prepared by the modified method both had anti- ERK activity (data not shown). CCS/Pl did not display any anti-ERK activity. These results are consistent with that obtained earlier for samples prepared as described in EXAMPLE 1.

EXAMPLE 9 - Purification of CCS/P3/S1 to CCS/P3/S6 subfractions We next sought to identify the active component in the CCS subfractions. We chose to sub-fractionate CCS/P3 prepared as described in EXAMPLE 8, since this method yielded material that had high specific protease activities that was also biologically active. This material also contained substantially lower amounts of the 13 kDa, non- biologically active component than CCS/P2.

a. Method. Fraction CCS/P3 prepared as described in EXAMPLE 8 was subjected to sub-fractionation as follows. CCS/P3 was first concentrated to 7.5 to 20 mg/ml using 0.5 ml or 4 ml Vivaspin™ units (Molecular weight cut off is 5,000 da; Vivascience Ltd). Gel permeation chromatography was then performed using a Pharmacia Superdex™ 75 HR 10/30 column equilibrated with 3 column volumes of acetate buffer (100 mM sodium acetate, 1 mM di-sodium EDTA, 0.25 M NaCl, pH 5.0) at a flow rate of 0.5 ml/min. 0.1 to 0.2 ml of concentrated CCS/P3 was used for each column run and fractions (0.5 ml) were collected and pooled as described in Table 8.

Each pooled fraction was analysed for protein content, SDS-PAGE and proteolytic activity as described in EXAMPLE 4 and 5. Samples were also sent to Imperial College on dry ice, whereupon they were thawed, aliquoted and stored at -70°C until required for testing for anti-ERK activity and effects on IL-2 production and CD4+ T cell proliferation as described in EXAMPLE 2 and 3. b. Results A typical U.V. chromatogram obtained using the Superdex™ 75 HR 10/30 column is shown in FIGURE 10. The proteins eluted were pooled into 6 different fractions as described in Table 8. The purity of the pooled samples as shown by SDS-PAGE are shown in Figure 11. The approximate molecular weights of the bands are described in Table 9. The specific activities of the different fractions are as shown in Table 10.

Table 8 Summary of pooled CCS/P3 subfractions fractions from Superdex 75

HR 10/30 column Gel Permeation Chromatography.

Table 9 Summary of the approximate molecular weights of proteins found in CCS/P3 subfractions CCS/P3/S1 to CCS/P3/S6 as determined by SDS-PAGE.

Table 10 - Calculated protein content, Bz-Phe-Val-Arg-pNA and Z-Arg-Arg- pNA activity of CCS/P3/S1 to CCS/P3/S6 subfractions.

CCS/P3/S1 and CCS/P3/S2 fractions react against both the Bz-Phe-Val-Arg-pNA and Z-Arg-Arg-pNA substrates, but have a much higher preference for the Z-Arg-Arg substrate. Fractions CCS/P3/S3 to CCS/P3/S4 also react with both substrates, but have a higher preference for the Bz-Phe-Val-Arg-pNA substrate. This change in preference for the different substrate indicates a difference in the relative proportion of different proteases present within the fractions. This is reflected in the change in the molecular weight species present in the different CCS/P3/S subfractions. In CCS/P3/S1 and CCS/P3/S2 there is a higher proportion of the 26 kda protein, and a higher preference for the Z-Arg-Arg-pNA substrate. In CCS/P3/S3 and CCS/P3/S4 the 24 kda protein is present, and these fractions have a preference for the Bz-Phe-Val- Arg-pNA substrate. Together this data indicate that the 26 kda protein has a preference for the Z-Arg-Arg-pNA substrate, while the 24 kda protein has a preference for the Bz-Phe-Val-Arg-pNA substrate. This is further supported by the observation that Fractions CCS/P3/S5 and CCS/P3/S6 both have a high specific activity against the Bz-Phe-Val-Arg-pNA substrate and negligible activity against the Z-Arg-Arg-pNA substrate and these fractions predominantly contain the 24 kda protein. As mentioned earlier in EXAMPLE 5, the literature reports that ananain and fruit bromelain have a higher preference for the Bz-Phe-Val-Arg-pNA substrate, whereas comosain and stem bromelain protease have a preference for Z-Arg-Arg-pNA (Napper et al, 1994, Biochem. J. 301, 727-735). Based on the substrate preferences obtained for CCS/P3/S subfractions, the data indicate that CCS/P3/S1 to S2 predominantly contains the enzyme comosain and/or stem bromelain protease. Fraction CCS/P3/S3 and S4 contain a mixture of ananain as well as comosain or stem bromelain protease. Fraction CCS/P3/S5 and CCS/P3/S6 predominantly contain ananain.

Earlier we showed in EXAMPLE 2 that Fraction CCS/P2 and CCS/P3 which contained a mixture of ananain and comosain blocked tyrosine phosphorylation of ERK-2 following stimulation of T cells with combined PMA plus calcium ionophore. We also showed that CCS/P2 could block CD4+ T cell proliferation and IL-2 production. Since CCS/P2 and CCS/P3 both contain a mixture of ananain and comosain, both with different substrate specificities, we wished to test which specific activity was responsible for the anti-ERK and inhibition of T cell responses. We therefore selected from the CCS/P3 subfraction, the materials with the highest Z-Arg- Arg-pNA activity (and purest 26 kda protein; CCS/P3/S1) and the highest Bz-Phe- Val-Arg activity (and purest 24 kda protein; CCS/P3/S6)

Stimulation of T cells with combined calcium ionophore and PMA induced tyrosine phosphorylation of ERK-2 as expected. Crude fraction CCS pre-treatment reduced tyrosine phosphorylation of ERK-2 consistent with our results obtained earlier (WO- A-9838291). CCS/P3/S6 displayed anti-ERK activity, while CCS/P3/S1 did not. See Figure 12. This data suggests that the active component within Fraction CCS that blocks ERK-2 activation resides within CCS/P3/S6. Since CCS/P3/S6 has a very high specific activity against Bz-Phe-Val-Arg-pNA, and negligible activity against Z-Arg- Arg-pNA, this data suggest that anti-ERK activity correlates with Bz-Phe-Val-Arg- pNA activity and the 24 kda protein. A plot comparing ERK inhibitory activity of various batches of Fraction CCS materials with different Z-Arg-Arg-pNA and Bz-Phe- Val-Arg-pNA, would support this view. Figure 13 shows that anti-ERK activity of various CCS batches correlates with Bz-Phe-Val-Arg activity, and not Z-Arg-Arg- pNA activity.

We next tested whether CCS/P3/S1 or CCS/P3/S6 would block IL-2 production. Suφrisingly, we saw that CCS/P3/S1, but not CCS/P3/S6 blocked IL-2 production. See FIGURE 14. This result is quite unexpected, since it is well known in the scientific literature that ERK-2 is required for IL-2 production. Since CCS/P3/S6 blocked ERK-2 it would be expected to also block IL-2 production. This data suggests that T cells have an ability to bypass their requirement for ERK-2 in their production of IL-2 by activating alternate pathways. It is interesting to note that CCS/P3/P1 can inhibit IL-2 production, but not by blocking ERK-2. The mechanism by which CCS/P3/P1 acts to block IL-2 is not known, but may involve inhibition of any of the steps in the IL-2 production pathway such as those illustrated in Figure 1. For example, CCS/P3/P1 may act to block calcineurin, a well known target of the immunosuppressant drug, cyclosporin A. CCS/P3/P6 may also act to inhibit protein kinase C (PKC) the INK pathway, or other currently unidentified T cell signalling pathways.

EXAMPLE 10 - NH₂-terminal sequencing of proteins.

To further characterise the major protein bands obtained in CCS/P2 and CCS/P3, we conducted NH₂-terminal sequencing.

a. Method. CCS/P2, CCS/P3, CCS/P3/S1 and CCS/P3/S6 were subjected to

SDS-PAGE as follows. Continuous SDS-PAGE slab gels were prepared by the method of Laemmli (1970, Nature. 227, 680-685). Stacking and separating gel concentrations were 5% and 15% acrylamide, respectively. Samples to be analysed, were boiled for 5 minutes in an equal volume of a sample buffer consisting of Tris- HC1 (50 mM, pH 7), 2% (w/v) SDS, 50% (v/v) glycerol, 0.02% (w/v) bromophenol blue and 10%) (v/v) 12 M 2-mercaptoethanol. After electrophoresis, separated proteins were electro-blotted in Tris/glycine/methanol buffer [Tris (50 mM), glycine (40 mM), SDS (0.04% w/v), methanol (20% v/v)] onto PVDF membrane using a Bio- Rad transfer apparatus at 20 V for 12 hours.

The membrane was stained with 0.025% w/v Coomassie blue R-250 dissolved in 40 % (v/v) methanol for 5 min, followed by destaining in 50 % (v/v) methanol. The membrane was air-dried between filter paper at room temperature and the proteins in the 23 to 25 kda region were sent for NH₂-terminal amino acid sequencing at Liveφool University, Durham University and Leicester University. Briefly, the protein bands on the membrane were excised and placed in the upper cartridge of the sequencer. NH -terminal amino acid analysis of CCS proteins were determined by Edman degradation using a gas phase sequencer (Applied Biosystems), equipped with an on-line phenylthiohydantion amino acid analyser.

b. Results The amino acid sequences obtained from CCS/P2 and CCS/P3 are shown in Table 11.

Table 11 NH2-Terminal sequence of CCS/P2 and CCS/P3 proteins obtained by Biosep SEC S2000 column chromatography (EXAMPLE 1) and CCS/P3/S1 and CCS/P3/S6 proteins obtained by Superdex fractionation (EXAMPLE 9) and separated by SDS-PAGE.

Table 12 shows the NH₂-terminal amino acids of comosain (Napper et al, 1994, Biochem. J. 301, 727-735), ananain (Lee et al, 1997, Biochem J. 327, 199-202), CCX2 (PCT/GB98/00592), CCZ (PCT/GB98/00591), stem bromelain protease (Ritonja et al, 1989, FEBS Lett. 247, 419-424), F4, F5 and F9 (Harrach et al, 1995, J. Protein. Chem. 14, 41-52) which derive from either the fruit or the stem of the pineapple plant (Ananus comosus). The DDBJ/EMBL/GenBank or the NCBI (GenPept) accession numbers for each protein are listed as well as whether the enzymes derived from the stem or the fruit of the pineapple plant.

All published proteins NH₂-terminal sequences share many homologies. Of the stem enzymes, ananain differs by 4 out of 29 amino acids, and comosain by 2 out of 20 acids, when compared to stem bromelain protease. Ananain and comosain both differ by 6 out of 20 amino acids when compared to CCZ and 4 out of 20 amino acids when compared to CCX2. CCZ differs by 8 out of 21 amino acids when compared to stem bromelain protease, and 5 out of 20 compared to CCX2. Comosain differs by 2 amino acids to ananain. The NH2 -terminal sequences of the F4 and F5 proteases match identically with stem bromelain protease (pl4518). A comparison of the published chromatograms of F4, F9 and stem bromelain protease (Harrach et al, 1995, J. Protein. Chem. 14, 41-52; Rowan et al, 1990, Biochem. J. 266, 869-875) suggest that F4, F9 and stem bromelain protease are the same enzymes. The NH2 -terminal sequence of the F9 protease matches ananain and has been reported to be the same as ananain (Harrach et al, 1995, J. Protein. Chem. 14, 41-52).

Likewise for the enzymes that derive from the stem, the NH -terminal sequences of enzymes that derive from the fruit are similar to each other. Two fruit enzymes (accession no: d 14057 and d 14058) are identical to each other. The remaining fruit enzymes only differ from each other in 4 out of 25 amino acids. In general, the NH - terminal sequences of proteases that derive from stem are more similar to each other, than the NH -terminal sequences of proteases that derive from fruit. Similarly, the fruit enzymes are more similar to each than stem. There are however, two exceptions; the NH₂- terminal sequence of D38532 (fruit) matches identically with pi 4518 (stem), and the NH₂- terminal sequence of CCX2 (stem) matches with dl4059 (fruit). In the case of CCX2, despite the similarity in NH - terminal sequence to dl4059, the remainder of the amino acid sequence do not match identically, suggesting that CCX2 and dl4059 are distinct proteins. Of all the fruit and stem enzyme NH₂-terminal sequences, the CCZ enzyme is the most divergent.

It is clear that the proteinases described in the fruit and stem of the pineapple plant are closely related. However despite small differences in the amino acid sequences, the proteins are all distinct, showing divergence from each other. The difference between two proteinases become more obvious when the complete amino acid sequences are compared, and when structural and kinetic analysis of proteins are conducted. For example, although the NH₂-terminal sequence of ananain only differs by 2 out of 21 amino acids when compared to stem bromelain protease (95%ι homology), the complete amino acid sequence shows that they are only possess 77% identity (Lee et al, 1997, Biochem J. 327, 199-202). The difference between ananain and stem bromelain protease is further reflected when their substrate specificity is compared. For example, despite the 95%> identity in NH2 -terminal sequence, ananain shows a preference for the substrate Bz-Phe-Val-Arg-pNA, whilst stem bromelain protease has a preference for Z-Arg-Arg-pNA, a different substrate (Napper et al, 1994, Biochem. J. 301, 727-735). Both enzymes also differ in their isolelectric point, molecular weight and biological activity (Napper et al, 1994, Biochem. J. 301, 727-735; WO-A- 9838291).

Similarly, when the NH - terminal sequence of comosain is compared with stem bromelain protease, it only differs in 2 out of 20 amino acids. However, comosain differs in its inhibitor specificity, it is structurally different and is immuno logically distinct. These physico-chemical differences indicate that comosain is a distinct enzyme from ananain and stem bromelain protease.

Of the NH₂-terminal sequences obtained from CCS/P2, SEQ ID NOs: 2 and 3 match comosain and ananain respectively and SEQ ID NO:6 matches either stem bromelain protease or a fruit bromelain protease having the accession number d38532 in the DDBJ/EMBL/GenBank database. SEQ ID NO: 1, 4 and 5, however, do not match any of the NH₂-terminal sequences previously reported. Of the sequences from CCS/P3, SEQ ID NO: 8 matches ananain and SEQ ID NO: 7 is novel. The first 15 amino acids of SEQ ID NOs: 1, 5 and 7 all match ananain, however the remaining amino acids do not match any of the previously reported stem or fruit bromelain sequences. In SEQ ID NO: 1, amino acids 16 to 21 are different. In SEQ ID NO: 5, the amino acids 22, 25, 31 and 35 are different. In SEQ ID NO: 7, the amino acids 18 and 20 are different from previously published sequences.

Of the NH₂-terminal sequences obtained from CCS/P3/S1, SEQ ID No: 9 matches stem bromelain protease or a fruit bromelain enzyme with the database accession number d38532. If the Asp is replaced with a Asn at position 9, then SEQ ID No: 9 matches comosain. If the Tyr at position 10 is replaced with Ser, then SEQ ID No: 9 resembles ananain.

Of the NH₂-terminal sequences obtained from CCS/P3/S6, SEQ ID No: 10 resembles ananain until position 19, where there is a Pro. If the Ser at position is 10 is replaced with a Tyr, then SEQ ID No: 10 resembles stem bromelain protease or CCX2 or the fruit bromelain enzyme d38532 up until position 19. Since the N-terminal sequence does not identically match, ananain, stem broemlain protease, CCX2 or fruit bromelain d38532, it is likely that SEQ ID No: 10 is novel. Since the protease in CCS/P3/S6 has a very high substrate specificity for Bz-Phe-Val-Arg-pNA substrate, then it is most likely not to be stem bromelain protease, but may represent a novel form of stem bromelain protease. The different amino acids are indicated in bold in Table 11. As previously mentioned, despite many enzymes present in the pineapple plant having very similar NH₂ terminal sequences (some only differing by two amino acids), these enzymes are structurally and kinetically different. They also have different biological activities. This would suggest that because SEQ ID No: 1, 5, 7 and 10 have different amino acid sequences from any previously reported, they are novel proteins. SEQ ID No:4 does not match any protein previously reported on the DDBJ/EMBL/GenBank or the NCBI (GenPept) databases and is also a novel protein.

Table 12. NH -Terminal sequence of ananain, comosain, stem bromelain protease, CCX2, and CCZ proteins.

DISCUSSION

Bromelain is a crude proteolytic extract obtained from pineapple stems. Previously, we isolated a fraction from bromelain, termed CCS, which comprises of two enzymes called ananain and comosain, and other as yet unidentified components. We showed that CCS, but not stem bromelain protease, CCX2 or CCZ, could block ERK-2 activation. In addition, CCS blocked IL-2 production and CD4+ T cell proliferation. CCS however, did not affect splenocyte proliferation, suggesting that CCS has a selective mode of action. The inhibitory effect of CCS on ERK-2 was dependent on its proteolytic activity since E-64, a selective cysteine protease inhibitor, could abrogate the effect of CCS. In the present study, we further fractionated CCS to determine the active component within the CCS mixture which was responsible for blocking ERK-2 activity and T cell activation.

CCS subfraction P2 (CCS/P2), but not CCS/Pl, blocked ERK-2 activation, IL-2 production and T cell proliferation. CCS/P3 had a marginal effect on ERK-2 activation and proliferation, however did not affect IL-2 production. The data suggest that CCS/P2 contains the active component responsible for blocking T cell activation. Mass spectrometry of CCS/P2 revealed that it comprised of 5 distinct entities. Three molecules were low molecular weight proteins and probably are members of the bromelain family of protease inhibitors. Two molecules were 23,459 and 23,657.6 da. Based on the presence of enzyme activity in CCS/P2, it is most likely that these high molecular weight species are enzymes. NH2-terminal sequencing of the high molecular weight proteins revealed the presence of 6 different amino acid sequences. Given the high degree of similarity between the bromelain enzymes, it might be possible that several enzymes may have the same molecular mass therefore explaining why more than two protein species were revealed by NH2 -terminal sequence analysis.

Based on NH2 -terminal sequence analysis, comosain (SEQ ID No: 2), stem bromelain proteinase (SEQ ID NO: 6) and ananain (SEQ ID No: 3) appear to be present in CCS/P2. The other proteins (SEQ ID NO: 1 and 5) appear to be ananain-like, however, do not identically match the amino acid sequence of ananain, therefore suggesting that they are novel enzymes. SEQ ID NO: 4 is completely novel and does not match any previously described protein listed on any data base. Another ananain- like enzyme (SEQ ID NO: 7) is also present in CCS/P3.

We also used a modified method to produce biologically active CCS/P3. Subfractionation of CCS/P3 into different components revealed two fractions that were biologically active. CCS/P3/S1 was shown to block interleukin 2 production, but did not affect ERK-2 activation. This data suggests that CCS/P3/S1 blocks cytokine production via an ERK-2-independent pathway. CCS/P3/S6 blocked ERK-2 activation, but did not affect cytokine production. Therefore, this data suggest that alternate pathways exist in T cells that may be used to produce IL-2. Together, the data shows that there are two active components in Fraction CCS and CCS/P3 that both block T cell activation, but act in different ways. It is possible that the two components may act together to synergise or enhance the ability of CCS to inhibit T cell activation.

CCS/P3/S1 contains one or two different components. SEQ ID No: 9 matches stem bromelain protease, CCX2 and fruit bromelain, but based on CCS/P3/Sl 's preference for Z-Arg-Arg-pNA, SEQ ID No: 9 is most likely to be stem bromelain, since CCX2 and fruit bromelain have a preference for the Bz-Phe-Val-Arg substrate. SEQ ID No: 9 may also be comosain, since comsain has a preference for the Z-Arg-Arg-pNa substrate.

CCS/P3/S6 contains two different components. SEQ ID No: 10 resembles ananain up to position 18, and may represent a novel form of ananain. If the Ser at position 10 is replaced by a Tyr, then SEQ ID NO: 10 resembles stem bromelain protease, or fruit bromelain (d38532) up to position 18, and may also represent novel forms of these enzymes. SEQ ID No: 11 matches ananain. In earlier studies with CCS, we have shown that use of a selective enzyme inhibitor abrogates the effect of CCS on ERK-2 activation. These data suggest that the active component in CCS is an enzyme. We propose that there are at least two active components in CCS. These may be comosain, ananain or one of the novel ananain- like enzymes. These molecules seem to act by different mechanisms and maywork singularly or synergistically to block T-cell activation.

In summary, we have further separated the active bromelain fraction CCS into different subfractions and have identified several active fractions, designated CCS/P2,

CCS/P3, CCS/P3/S1 and CCS/P3/S6. CCS/P2 blocks ERK-2 activation, T cell proliferation and IL-2 production. CCS/P3 and CCS/P3/S6 bloke ERK-2 activation, while CCS/P3/S1 blocks IL-2 production. These fractions may therefore be effective as anti-cancer agents or as immunosuppressive agents. Several novel proteins were identified in the bromelain fractions.

Claims

1. A fraction of bromelain comprising proteins of molecular weights 23.9kDa and

26.6 kDa when measured by SDS-PAGE or 23.459kDa and 23.6576kDa when measured by mass spectrometry; and proteins having the following NH₂-terminal sequences:

Val Pro Gin Ser He Asp Trp Arg Asp Ser Gly Ala Val Thr Ser Arg Gly Glu Asn Pro Arg (SEQ ID NO: 1);

Val Pro Gin Ser He Asp Trp Arg Asn Tyr Gly Ala Val Thr Ser (SEQ ID NO: 2);

Val Pro Gin Ser He Asp Trp Arg Asp Ser Gly Ala Val Thr Ser Val Lys Asn Gin Gly Arg Xaa Gly Ser Xaa Xaa Ala Phe (SEQ ID NO: 3);

Asp Gly Ser Asn Asn Ala Arg Lys (SEQ ID NO: 4);

Val Pro Gin Ser He Asp Trp Arg Asp Ser Gly Ala Val Thr Ser Val Lys Asn Gin Gly Arg Asn Gly Ser Xaa He Ala Phe Xaa Ser Leu Xaa Xaa Xaa Pro (SEQ ID NO: 5); and

Val Pro Gin Ser He Asp Trp Arg Asp Tyr Gly Ala Val Thr Ser (SEQ ID NO: 6).

2. A protein having a molecular weight of 26.6kDa as measured by SDS-PAGE and comprising the NH₂-terminal sequence: Val Pro Gin Ser He Asp Trp Arg Asp Ser Gly Ala Val Thr Ser Arg Gly Glu Asn Pro

Arg (SEQ ID NO: l); or

a protein having a molecular weight of 23.9kDa as measured by SDS-PAGE and comprising the NH₂-terminal sequence: Asp Gly Ser Asn Asn Ala Arg Lys (SEQ ID NO: 4); or a protein having a molecular weight of 23.9kDa as measured by SDS-PAGE and comprising the NH₂-terminal sequence:

Val Pro Gin Ser He Asp Tφ Arg Asp Ser Gly Ala Val Thr Ser Val Lys Asn Gin Gly Arg Asn Gly Ser Xaa He Ala Phe Xaa Ser Leu Xaa Xaa Xaa Pro (SEQ ID NO: 5).

3. A fraction of bromelain comprising a major protein component of molecular weight 24.0kDa when measured by SDS-PAGE and proteins comprising the following NH₂-terminal sequences: Val Pro Gin Ser He Asp T╧å Arg Asp Ser Gly Ala Val Thr Ser Xaa Xaa Gly Xaa Pro

(SEQ ID NO: 7); and

Val Pro Gin Ser He Asp Tφ Arg Asp Ser Gly Ala Val Thr Ser Xaa Lys Asn Gin Gly Arg (SEQ ID NO: 8).

4. A protein having a molecular weight of 24.0kDa as measured by SDS-PAGE and comprising the N ^-terminal sequence:

Val Pro Gin Ser He Asp Tφ Arg Asp Ser Gly Ala Val Thr Ser Xaa Xaa Gly Xaa Pro (SEQ ID NO: 7).

5. A fraction of bromelain which comprises a major protein of 26 kDa when measured by SDS-PAGE, has a high specific activity towards the Z-Arg-Arg-pNA substrate and comprises a protein having the following NH₂-terminal sequence;

Val Pro Gin Ser He Asp Tφ Arg Asp (or Asn) Tyr (or Ser) Gly Ala Val Thr Ser Val

Lys Asn Gin Asn Pro (SEQ ID NO: 9).

6. A fraction of bromelain which comprises a major protein of 24 kDa when measured by SDS-PAGE, has a high specific activity towards the BZ-Phe-Val-Arg- pNA substrate and comprises proteins having the following NH₂-terminal sequences; Val Pro Gin Ser He Asp T╧å Arg Asp Ser Gly Ala Val Thr Ser Val Lys (or Gly or Asn) Pro Gin (or He or Asn) Gly Arg (SEQ ID NO: 10) and

Val Pro Gin Ser He Asp Tφ Arg Asp Ser Gly Ala Val Thr Ser (SEQ ID NO: 11).

7. A protein having a molecular weight of 24 kda as measured by SDS-PAGE, having a substrate specificity for BZ-Phe-Val-Arg-pNA substrate and comprising the NH₂-terminal sequence:

Val Pro Gin Ser He Asp Tφ Arg Asp Ser (or Tyr) Gly Ala Val Thr Ser Val Lys (or Gly or Asn) Pro Gin (or He or Asn) Gly Arg (SEQ ID NO: 10).

8. A nucleic acid encoding a protein as claimed in any one of claims 2, 4 or 7.

9. A fraction of bromelain or a protein as claimed in any one of claims 1 to 7 for use in medicine.

10. A fraction of bromelain as claimed in any one of claims 1, 3 or 6 or a protein as claimed in any one of claims 2, 4 or 7 for use in inhibiting the MAP kinase pathway, the activation of T cells and the production of growth factors and cytokines.

11. A fraction of bromelain as claimed in any one of claims 1, 3 or 6 or a protein as claimed in any one of claims 2, 4 or 7 for use in the treatment of diseases or conditions mediated by the activation of the MAP kinase pathway, the activation of T cells or the production of growth factors or cytokines.

12. A fraction of bromelain as claimed in any one of claims 1, 3 or 6 or a protein as claimed in any one of claims 2, 4 or 7 for use in the treatment of cancer.

13. A fraction of bromelain as claimed in any one of claims 1, 3 or 6 or a protein as claimed in any one of claims 2, 4 or 7 for use in the treatment of parasite or pathogen infections.

14. A fraction of bromelain as claimed in claim 5 for inhibiting T cell activation and the production of growth factors or cytokines.

15. A fraction of bromelain as claimed in claim 5 for use in the treatment of diseases or conditions mediated by T cell activation or the production of growth factors or cytokines.

16. A fraction of bromelain as claimed in any one of claims 1, 3, 5 or 6 or a protein as claimed in any one of claims 2, 4 or 7 for use as an immunosuppressant.

17. A fraction of bromelain as claimed in any one of claims 1, 3, 5 or 6 or a protein as claimed in any one of claims 2, 4 or 7 for use in the treatment or prevention of autoimmune diseases, graft or transplant rejection by a host, allergic reactions, toxic shock or apoptosis.

18. A protein as claimed in any one of claims 2, 4 or 7, in combination with ananain and/or comosain and/or stem bromelain protease and/or the protein which appears on the DDJB/EMBL/GenBank database as accession No. 38532 for use in medicine.

19. The use of a fraction of bromelain as claimed in any one of claims 1, 3 or 6 or a protein as claimed in any one of claims 2, 4 or 7 in the preparation of an agent for the treatment of diseases or conditions mediated by the activation of the MAP kinase pathway, the activation of T cells or the production of growth factors or cytokines.

20. The use of a fraction of bromelain as claimed in any one of claims 1, 3 or 6 or a protein as claimed in any one of claims 2, 4 or 7 in the preparation of an agent for the treatment of cancer.

21. The use of a fraction of bromelain as claimed in any one of claims 1, 3 or 6 or a protein as claimed in any one of claims 2, 4 or 7 in the preparation of an agent for the treatment of parasite or pathogen infections.

22. The use of a fraction of bromelain as claimed in claim 5 in the preparation of an agent for inhibiting T cell activation and the production of growth factors or cytokines.

23. The use of a fraction of bromelain as claimed in claim 5 for the preparation of an agent for the treatment of diseases or conditions mediated by T cell activation or the production of growth factors or cytokines.

24. The use of a fraction of bromelain as claimed in any one of claims 1, 3, 5 or 6 or a protein as claimed in any one of claims 2, 4 or 7 for the preparation of an immunosuppressant.

25. The use of a fraction of bromelain as claimed in any one of claims 1, 3, 5 or 6 or a protein as claimed in any one of claims 2, 4 or 7 in the preparation of an agent for the treatment or prevention of autoimmune diseases, graft or transplant rejection by a host, allergic reactions, toxic shock or apoptosis.

26. A pharmaceutical or veterinary composition comprising a fraction of bromelain as claimed in any one of claims 1, 3, 5 or 6 or a protein as claimed in any one of claims 2, 4 or 7 optionally combined with comosain and/or ananain, together with a pharmaceutically or veterinarily acceptable excipient or adjuvant.

27. A pharmaceutical or veterinary composition as claimed in claim 26, which is adapted for enteral, for example oral, nasal, buccal, topical or anal administration.

28. A pharmaceutical or veterinary composition as claimed in claim 26, which is adapted for parenteral, for example intravenous, subcutaneous, intramuscular or intraperitoneal administration.

29. A pharmaceutical or veterinary compostion as claimed in claim 26 which is adapted for oral administration and which is enterically coated.

30. A process for the preparation of a protein as claimed in any one of claims 2, 4 or

7, the process comprising expressing a nucleic acid as claimed in claim 8.