GB2441131A

GB2441131A - Modulation of binding of CK2a to NDPK

Info

Publication number: GB2441131A
Application number: GB0616841A
Authority: GB
Inventors: Anil Mehta; Russell Crawford
Original assignee: University of Dundee
Current assignee: University of Dundee
Priority date: 2006-08-24
Filing date: 2006-08-24
Publication date: 2008-02-27
Also published as: GB0616841D0

Abstract

Methods for identifying an agent which has the property of modulating the binding of CK2 a (casein kinase II) to an isoform of NDPK (nucleoside diphosphate kinase) are disclosed. Uses of the agents are also disclosed, including their use in the treatment or prevention of cancer. Assays for the binding of CK2 a to an isoform of NDPK or the distribution of CK2 a between isoforms of NDPK, and transgenic animals for use in such assays are also disclosed.

Description

I Agents that have the property of modulating the binding of CK2a to an

isoform of 2 NDPK, or a specific mutant thereof, and related products, methods and assays

S *5S*

4 Field of the invention

6 The invention relates to agents that have the property of modulating the binding of 7 CK2a to an isoform of NDPK, or a specific mutant thereof; methods of identifying: .. S...

8 agents which have the property of modulating the binding of CK2cz to an isoform of 9 NDPK, or a specific mutant thereof; and applications of such agents; and assays for *:* assessing the binding of CK2a to an isoform of NDPK, or a specific mutant thereof. *::

12 Background to the invention

14 Within this specification, and the claims appended hereto, references to NDPK-A, NDPK-B, and other isoforms of NDPK, CK2a, CK213, AMPK and AMPK ol include 16 homologues, such as orthologues and paralogues, thereof.

18 Within this specification, and the claims appended hereto, the term peptid& includes 19 a plurality of amino acids linked in a defined order by one or more peptide bonds, and does not imply an upper limit as to the number of amino acid residues, nor does it 21 imply that each amino acid residue is a naturally occurring proteinogenic residue, 22 except where this is clear from context, e.g. because a peptide is synthesised in vivo.

23 Where an amino acid residue of a variant, fragment, fusion or mutant of an isoform of I NDPK is referred to by a designation including a number, e.g. S120, this is a 2 reference to the number of the corresponding residue in the unmodified isoform of 3 NDPK and should not be construed as meaning that the referenced residue is at the 4 same numbered position in the variant, fragment, fusion or mutant of the isoform of NDPK.

7 A reference herein to a document should not be construed as an admission that the 8 document constitutes relevant prior art. All of the literature mentioned within this 9 specification is incorporated herein by reference.

11 Nucleoside diphosphate kinase (NDPK, NM23Iawd) has been recognised for over 12 forty years as a ubiquitous enzyme that catalyses the transfer of the y-phosphate of a 13 nucleoside triphosphate to a different nucleoside diphosphate (1), balancing different 14 cytosolic nucleotide pools. The most widely expressed isoforms of NDPK (NDPK-A and NDPK-B) regulate a diverse array of cellular events including growth and 16 development, tumour metastasis and transcriptional regulation (2, 3). Despite sharing *. S...

17 88% amino acid sequence identity, these two isoforms are reported to have distinct * 18 cellular functions (4). Accumulating evidence indicates that protein-protein 19 interactions modulate the specific molecular actions of NDPK (3). For example, NDPK is an in vivo substrate for protein kinase CK2, and NDPK phosphorylation at: .. S...

21 serine 120 (S 120) is inhibitory with respect to NDPK phosphotransfer activity (5). We *...

22 recently added to this notion by showing, firstly, that NDPK-A is a target for 23 phosphorylation by AMPK ((6) and (7)) and, secondly, that post-phosphorylation, 24 S122 phosphorylated NDPK-A reciprocally regulates AMPK al activity.

26 The AMP-activated protein kinase (AMPK) is a heterotrimeric protein with a 63 kDa 27 catalytic a subunit and two regulatory 13 and y subunits (38 and 36 kDa respectively), 28 each encoded by distinct genes (8, 9). AMPK is a sensor of cellular energy status, 29 responding to the cytosoiic AMP:ATP ratio equilibrated by the adenylate kinase reaction (10). Once activated by a rise in cellular [AMP], AMPK phosphorylates 31 several downstream substrates; the net effect switches off ATP-consuming pathways 32 (e.g. fatty acid synthesis and cholesterol synthesis) and switches on ATP-generating 33 pathways (e.g. fatty acid oxidation and glycolysis) (11). Our previous study 34 demonstrated that the AMPK al (but not the a2) catalytic subunit is associated with NDPK-A (but not NDPK-B) and engages in a process called "substrate channelling", 36 whereby ATP produced from GTP + ADP, by an NDPK-catalysed reaction, alters the 1 activity of AMPK al in an AMP-independent manner and irrespective of the 2 surrounding ATP concentration (6). In this reaction, AMPK al receives ATP inside a 3 complex of proteins shielded from the prevailing conditions in the surroundings.

Protein kinase CK2, formerly known as casein kinase 2, is a constitutively active 2a, 6 213 heterodimer (12). CK2 has a growing list of over 300 in vivo protein substrates and 7 plays an essential role in the cell (13, 14). CK2 phosphorylates NDPK in vivo (5) and 8 in this specification, we demonstrate that protein kinase CK2 also interacts with our 9 previously reported NDPK-A I AMPK al regulatory complex (6) under basal conditions. We show that CK2a and CK213 subunits bind to NDPK-A, NDPK-B and 11 AMPK, with CK2a binding the S120 region of NDPK-A and NDPK-B, dependent on 12 the status of AMPK activity. We recently demonstrated that the phospho-status of 13 S122, adjacent to the catalytic H118 (or equivalent) in all NDPK molecules, 14 determines whether NDPK-A is able to substrate channel, i.e. only when AMPK phosphorylates S122, does substrate channelling occur (7). Thereby, AMPK- 16 dependent phosphorylation of NDPK-A permits NDPK-A to regulate AMPK al. ...... S... 17.::

18 Here we show that when S122 is pre-phosphorylated by AMPK al, CK2a 19 translocates and binds to the closely related NDPK-B isoform. Our data provide novel * insights into the mechanism of interaction of the cytosolic NDPK-A / AMPK ci: .. S...

21 complex with CK2, where the association of CK2a with NDPK-B is an NDPK-A and 22 S122-dependent process. Thus, for the first time we show how NDPK-A and NDPK-B 23 relate to one another in the context of two other protein kinases that control many. S. 24 biological processes, including energy flux and oncogenesis.

26 This interaction was surprising as we have previously demonstrated repeatedly that 27 AMPK does not interact with NDPK-B. This interaction is also surprising because 28 CK2 has more than 300 known substrates and it would, therefore, have been thought 29 unlikely that a high proportion of CK2a which is displaced from binding to NDPK-A would translocate to another isoform of NDPK. Indeed, one might expect NDPK-B to 31 bind CK2a in a similar manner to NDPK-A. Furthermore, a mechanism involving free 32 CK2a is surprising as CK2 is a heterodimer.

34 Our findings are particularly relevant to cell proliferation and hence cancer and related syndromes. NDPKs act as metastatic suppressors I activators on certain 36 human tumours, such as breast carcinoma. Paradoxically, NDPK-A is also reported I to act as a metastasis promoter in other human tumours, including neuroblastoma.

2 The reasons for this are obscure. The presence of an S12OG mutation in NDPK-A 3 has been detected in 14 -30% of patients with advanced stages of neuroblastoma 4 (29) and it has been shown that an S12OG or S12OA mutation in NDPK-A causes metastasis by an unknown mechanism. Furthermore, NDPK has been implicated in 6 skin growth and wound healing where proliferation is important in keratinocyte 7 function (46).

9 Furthermore, it has previously been reported that NDPK-B induced activation of the oncogene, c-myc, by an unknown mechanism, leads to metastasis (35). Studies 11 have also linked an increase in "free" cellular CK2a amounts to various forms of 12 cancer, for example: (39), (40) and (41).

14 Our findings reveal a molecular mechanism which connects these different metastatic effectors. We have demonstrated that, where the binding of CK2o to NDPK-A is 16 disrupted, for example where the binding site of CK2a to NDPK-A is mutated (e.g. in.. S...

17 S12OA mutants), CK2a is translocated to NDPK-B (see Figure 6 below) and 18 phosphorylates NDPK-B (see Figure 5 below), which in turn creates an oncogenic 19 potential of the new complex.

IA S..

21 We thus propose that an agent which has the property of disrupting the binding of S.'.

22 CK2a to NDPK-B should be useful as a therapeutic agent for the treatment or 23 prevention of cell proliferative conditions, such as cancers, which result from.": 24 increased binding of CK2a to NDPK-B, for example due to a mutation in the binding site of NDPK-A for CK2a.

27 Similarly, an agent which has the property of promoting the binding of CK2a to 28 NDPK-A should be useful as a therapeutic agent for the treatment or prevention of 29 cell proliferative conditions, such as cancers which result from increased binding of CK2a to NDPK-B, or as a therapeutic agent for the treatment or prevention of cell 31 proliferative conditions, such as cancers, which result from increased amounts of 32 unbound CK2ci.

34 However, as mentioned above, NDPK-A has been shown in some circumstances to act as a metastasis promoter in some human tumours. It has also been previously 36 demonstrated (42) that in some circumstances, the activation of AMPK, for example I by metformin or in response to exercise, is linked to a reduced risk of cancer. Our 2 findings demonstrate that when S122 of NDPK-A is pre-phosphorylated by activated 3 AMPK al, CK2a translocates and binds to NDPK-B. Accordingly, an agent which 4 has the property of promoting the binding of CK2a to NDPK-A should be useful as a therapeutic agent for the treatment or prevention of cell proliferative conditions, such 6 as cancers, which result from increased binding of CK2a to NDPK-A, for example due 7 to a mutation in the binding site of NDPK-A, or NDPK-B, for CK2a. Such an agent 8 should also be useful for the treatment or prevention of cell proliferative conditions, 9 such as cancers, which result from increased amounts of unbound CK2a.

11 Similarly, an agent which has the property of disrupting the binding of CK2a to NDPK- 12 B should be useful as a therapeutic agent for the treatment or prevention of cell 13 proliferative conditions, such as cancers, which result from increased binding of 14 CK2a to NDPK-B.

16 Accordingly, it would be advantageous to identify agents which have the property of 0*SI 17 modulating the binding of CK2a to an isoform of NDPK, such as NDPK-A or NDPK-B. * 18 In this specification and the appended claims, the term "modulating the binding" 19 means disrupting or promoting the binding and related words, such as modulate and modulated, should be construed accordingly. * S... S...

22 It would also be advantageous to screen tissue from patients to assess whether * * 23 particular cell proliferative conditions or cancer types are associated with increased * 24 or reduced binding of CK2a to an isoform of NDPK, or to increased or reduced amounts of unbound CK2a. Alternatively, it would be advantageous to carry out tests 26 to assess whether a particular patient has, or may develop, a cell proliferative 27 conditions, such as cancer, associated with increased or reduced binding of CK2a to 28 an isoform of NDPK, or to increased or reduced amounts of unbound CK2a. Such a 29 test would help determine whether an agent which has the property of modulating the binding of CK2a to an isoform of NDPK, or disrupting the binding of CK2a to an 31 isoform of NDPK, was suitable for treating a particular type of cell proliferative 32 condition, such as cancer, or a tumour in a specific patient. Tests could be repeated 33 to assess whether a drug which modulated the binding of CK2a to an isoform of 34 NDPK was effective.

1 ummarv of the invention 3 According to a first aspect of the present invention there is provided a method of 4 identifying an agent which has the property of modulating the binding of CK2cz to an isoform of NDPK, or a specific mutant thereof, the method comprising selecting a test 6 agent and assessing whether the test agent has the property of modulating the 7 binding of CK2a to the isoform of NDPK, or specific mutant thereof.

9 The agent may have the property of modulating the binding of CK2cx to the isoform of NDPK, or specific mutant thereof, because it has the property of disrupting the 11 binding of CK2a to the isoform of NDPK, or specific mutant thereof, or because it has 12 the property of promoting the binding of CK2a to the isoform of NDPK, or specific 13 mutant thereof.

Preferably, the isoform of NDPK will be NDPK-A or NDPK-B. The isoform of NDPK 16 may be selected from a group comprising NDPK-A and NDPK-B. However, the *.

17 isoform of NDPK also be NDPK-C, NDPK-D, NDPK-E, NDPK-F, NDPK-G, NDPK-H 18 or NDPK-I. The isoform may be a hybrid of NDPK-A and NDPK-B as discussed in 19 (45) 20::::.

21 The method may be a method of identifying an agent which has the property of S...

22 modulating the binding of CK2a to an isoform of NDPK, the method comprising 23 selecting a test agent and assessing whether the test agent has the property of * *. : 24 modulating the binding of CK2a to the isoform of NDPK.

26 However, the method may be a method of identifying an agent which has the property 27 of modulating the binding of CK2a to a specific mutant of an isoform of NDPK, the 28 method comprising selecting a test agent and assessing whether the test agent has 29 the property of modulating the binding of CK2a to the specific mutant of the isoform of NDPK. By a "specific mutant" we mean a predetermined mutant which affects the 31 binding of CK2a to the isoform of NDPK in v/va, for example, a S120G or S12OA 32 mutant of NDPK-A. Specific mutants of an isoform of NDPK are perhaps more likely 33 to be used in procedures to identify agents which have the property of promoting the 34 binding of CK2a to a specific mutant, for example a specific mutant which may be a causative factor for a cell proliferative condition, such as cancer. However, there 36 could be circumstances where it was desirable to identify an agent which had the 1 property of disrupting the binding of CK2a to a specific mutant, for example a specific 2 mutant which may be a causative factor for a cell proliferative condition, such as 3 cancer. The specific mutant of the isoform of NDPK will typically differ from the wild 4 type isoform of NDPK by a single insertion, deletion or change of amino acid residue.

Agents which have the property of modulating the binding of CK2a to a specific 6 mutant of an isoform of NDPK will typically be useful in the treatment of diseases 7 where a causative factor is the presence of one of a range of mutations of the isoform 8 of NDPK.

The invention will now be further described with reference to the identification of an 11 agent which has the property of modulating the binding of CK2a to an isoform of 12 NDPK. However, it may also be applied correspondingly in relation to agents which 13 have the property of modulating the binding of CK2cz to a specific mutant of an 14 isoform of NDPK and, where this is the case, references below to the isoform of NDPK refer to the specific mutant of an isoform of NDPK.

17 The method may comprise assessing whether the test agent has the property of. : : 18 modulating the binding of CK2a to a plurality of isoforms of NDPK. The agent will 19 preferably have a specific mode of action. For example, the agent may have the property of modulating the binding of CK2a to NDPK-A, but not to NDPK-B, or vice 21 versa. Some modes of action will of their very nature modulate the binding of CK2a 22 to a plurality of isoforrns of NDPK, for example a test agent which binds to the NDPK S..

23 binding site of CK2a could be expected to modulate the binding of CK2a to all.. : 24 isoforms of NDPK which bind to the binding site.

26 Thus, the method may be a method of identifying an agent which has the property of 27 disrupting the binding of CK2ci to NDPK-B, the method comprising selecting a test 28 agent and assessing whether the test agent has the property of disrupting the binding 29 of CK2a to NDPK-B.

31 The method may be a method of identifying an agent which has the property of 32 promoting the binding of CK2a to NDPK-A, the method comprising selecting a test 33 agent and assessing whether the test agent has the property of promoting the binding 34 of CK2a to NDPK-A.

I The method may be a method of identifying an agent which has the property of 2 disrupting the binding of CK2a to NDPK-A, the method comprising selecting a test 3 agent and assessing whether the test agent has the property of disrupting the binding 4 of CK2o to NDPK-A.

6 The method may be a method of identifying an agent which has the property of 7 promoting the binding of CK2a to NDPK-B, the method comprising selecting a test 8 agent and assessing whether the test agent has the property of promoting the binding 9 of CK2a to NDPK-B.

11 When assessing whether the test agent has the property of modulating the binding of 12 CK2a to the isoform of NDPK, what is assessed is preferably the binding of CK2a, 13 which is not bound to CK213, to the isoform of NDPK. More preferably, what is 14 assessed is the binding of CK2a, which is not bound to CK213, to the isoform of NDPK, to which CK213 is bound. S... S...

17 The method may comprise identifying an agent which has the property of disrupting. * 18 the binding of CK2cx to NDPK-B and promoting the binding of CK2a to NDPK-A. The 19 effects of disrupting the binding of CK2cx to NDPK-B and promoting the binding of CK2a to NDPK-A would be synergistic in causing CK2a to be translocated from 21 NDPK-B to NDPK-A in vivo. The method may comprise assessing whether the test 22 agent both disrupts the binding of CK2a to NDPK-B and promotes the binding of 23 CK2a to NDPK-A. However, the method may comprise assessing only one of *: 24 whether the test agent both disrupts the binding of CK2a to NDPK-B and promotes the binding of CK2a to NDPK-A.

27 The method may comprise identifying an agent which has the property of disrupting 28 the binding of CK2a to NDPK-A and promoting the binding of CK2a to NDPK-B. The 29 effects of disrupting the binding of CK2a to NDPK-A and promoting the binding of CK2a to NDPK-B would be synergistic in causing CK2a to translocated from NDPK-A 31 to NDPK-B in vivo. The method may comprise assessing whether the test agent both 32 disrupts the binding of CK2a to NDPK-A and promotes the binding of CK2a to NDPK- 33 B. However, the method may comprise assessing only one of whether the test agent 34 both disrupts the binding of CK2a to NDPK-A and promotes the binding of CK2cz to NDPK-B.

1 The method may be a method of identifying an agent which has the property of 2 modulating the binding of CK2a to the isoform of NDPK, without directly modulating 3 the binding of CK213 to the isoform of NDPK. By udirectly modulating" in this context, 4 we mean to exclude indirect modulation of the binding of CK213 to the isoform of NDPK resulting from, for example, the action of the agent changing the amount of 6 free unbound CK2a arid thereby altering the equilibrium of CK23 between forming an 7 unbound dimer with CK2a and being bound to the isoform of NDPK. Thus, assessing 8 whether the test agent has the property of modulating the binding of CK2a to an 9 isoform of NDPK may be assessing whether the test agent has the property of modulating the binding of CK2a to the isoform of NDPK, without directly modulating 11 the binding of CK23 to the isoform of NDPK.

13 The method preferably aims to identify an agent which has the property of modulating 14 (i.e. disrupting or promoting) the binding of CK2ci to the isoform of NDPK in viva in a subject (typically a mammal and most preferably a human), or which can be used as S...

16 part of a process of designing or selecting an agent which has the property of * 17 modulating the binding of CK2a to the isoform of NDPK in viva in a subject.

19 For example, the method may aim to identify an agent which has the property of disrupting the binding of CK2a to the isoform of NDPK in vivo in a subject (typically a.: 21 mammal and most preferably a human), or which can be used as part of a process of S...

22 designing or selecting an agent which has the property of disrupting the binding of 23 CK2a to the isoform of NDPK in vivo in a subject. .. : Preferably, the method is a method of identifying an agent which may be used to 26 modulate the binding of CK2a to the isoform of NDPK in a subject in need thereof. In 27 the case of agents which disrupt the binding of CK2a to the isoform of NDPK, this 28 may be because, for example, the subject has a condition in which a greater than 29 normal amount (preferably double, three times, ten times or thirty times a normal amount) of CK2a binds to the isoform of NDPK, respectively in at least some cells of 31 the subject under at least some circumstances. In the case of agents which promote 32 the binding of CK2cx to the isoform of NDPK, this may be because, for example, the 33 subject has a condition in which a lesser than normal amount (preferably half, a third, 34 one tenth or a thirtieth of a normal amount) of CK2a binds to the isoform of NDPK, respectively in at least some cells of the subject under at least some circumstances, 36 perhaps because the subject, or some cells within the subject, such as tumour cells, 1 may express a mutated form of the, or another, isoform of NDPK, which mutation 2 results in increased or reduced binding of CK2a to the isoform of NDPK.

4 For example, the method may be a method of identifying an agent which may be used to disrupt the binding of CK2a to NDPK-B in a subject in need thereof, for 6 example because the subject has a condition in which a greater than normal amount 7 (preferably double, three times, ten times or thirty times a normal amount) of CK2a 8 binds to NDPK-B in at least some cells of the subject under at least some 9 circumstances. For example, the subject, or some cells within the subject, such as tumour cells, may express a mutated form of NDPK-A, which mutation reduces or 11 prevents the binding of CK2a to NDPK-A, such as a mutated form in which residue 12 120 of NDPK-A is glycine or alanine. Preferably, the said cells express only a 13 mutated form or mutated forms of NDPK-A, which mutation(s) reduce or prevent the 14 binding of CK2a to NDPK-A.

16 Preferably, the method is a method of identifying a substance which may be used in * 17 the treatment and/or prevention of a cell proliferative condition, such as cancer, for, : : 18 example rieuroblastoma. The method may be a method of identifying a substance 19 which may be used in the treatment and/or prevention of a neuroblastoma in which the cells of the neuroblastoma express a mutated form of NDPK-A which reduces or: * SS* **** 21 prevents the binding of CK2a to NDPK-A, such as a mutated form in which residue **** 22 120 of NDPK-A is glycine or alanine. The method may be a method of identifying a 23 substance which may be used in the treatment and/or prevention of psoriasis.

The method may be carried out as an experiment to determine or confirm whether a 26 specific test agent has the property of modulating (that is to say, disrupting or 27 promoting) the binding of CK2a to the isoform of NDPK. However, the method 28 preferably comprises the screening of a plurality of test agents to simultaneously 29 and/or sequentially assess whether they have the property of modulating the binding of CK2a to the isoform of NDPK.

32 The test agent may have the property of disrupting the binding of CK2a to the isoform 33 of NDPK, because it modifies CK2a, or the isoform of NDPK, to reduce or eliminate 34 the affinity of CK2a for the isoform, or vice versa. The test agent may be an allosteric inhibitor of the binding of CK2a to the isoform of NDPK, or vice versa. The test agent 36 may bind irreversibly to the binding site of CK2a for the isoform of NDPK, or vice I versa. The test agent may be an irreversible or suicide inhibitor of the binding of 2 CK2a to the isoform of NDPK, or vice versa.

4 The test agent may have the property of promoting the binding of CK2ci to the isoform of NDPK, because it modifies CK2a, or the isoform of NDPK, to enhance the affinity 6 of CK2a for the isoform of NDPK, or vice versa. The test agent may be an allosteric 7 promoter of the binding of CK2a to the isoform of NDPK, or vice versa. The test 8 agent may bind irreversibly to the binding site of CK2o for the isoform of NDPK, or 9 vice versa. The test agent may be an irreversible or suicide promoter of the binding of CK2a to the isoform of NDPK respectively, or vice versa.

12 The test agent may have the property of disrupting the binding of CK2a to the isoform 13 of NDPK because it competes with the binding of CK2a to the isoform of NDPK, for 14 example because it binds to the binding site for CK2a on the isoform of NDPK, or because it binds to the binding site for the isoform of NDPK on CK2a, or because it 16 provides a binding site for CK2a which competes with the binding site for CK2a on 17 the isoform of NDPK, or because it provides a binding site for the isoform of NDPK 18 which competes with the binding site for the isoform of NDPK on CK2a. The test 19 agent may be a competitive inhibitor, mixed inhibitor or non-competitive inhibitor of the binding of CK2a to the isoform of NDPK, or vice versa. The binding site of NDPK- 21 A, and of NDPK-B, for CK2o includes S120 of NDPK-A/NDPK-B which CK2a binds to S...

22 and phosphorylates. The binding site of NDPK-A, and of NDPK-B, for CK2a S...

23 comprises at least some of residues 11 ito 125 which are exposed in vivo. ..

The step of assessing whether a test agent has the property of modulating the 26 binding of CK2a to the isoform of NDPK may be carried out in si//co and may 27 comprise modelling or simulating the predicted interaction of CK2a and the isoform of 28 NDPK as appropriate with a test agent using a computer.

However, the step of assessing whether a test agent has the property of modulating 31 the binding of CK2a to the isoform of NDPK preferably comprises the step of 32 exposing the test agent to CK2a and/or the isoform of NDPK. In some 33 circumstances, the test agent will be exposed to more than one isoform of NDPK 34 (and typically also CK2a). In some circumstances, such as in viva and assays in live cell lines or using unpurified cell lysate, the test agent will be exposed to a plurality of 36 isoforms of NDPK (and typically also CK2a).

2 The test agent is preferably incubated with CK2a and/or the isoform of NDPK for a 3 period of time, preferably until interactions between the test agent and CK2a and/or 4 the isoform of NDPK have reached equilibrium. One option would be to expose a test agent to only one of CK2a and the isoform of NDPK and to determine whether the 6 test agent binds to or reacts with CK2a or the isoform of NDPK, as appropriate, in a 7 manner which suggests or demonstrates that the substance would have the property 8 of modulating the binding of CK2a to the isoform of NDPK (for example, because the 9 test agent binds the binding site for CK2a on the isoform of NDPK, or the binding site for the isoform of NDPK on CK2a). However, the test agent is preferably exposed to Ii both CK2a and the isoform of NDPK.

13 Thus, the assessment of whether the test agent has the property of modulating the 14 binding of CK2a to the isoform of NDPK may comprise the step of exposing the test agent to CK2a and the isoform of NDPK, and then determining whether the test agent S...

16 has modulated the binding of CK2a to NDPK. S. **.

1, S *** 1 S ** 18 The test agent should be exposed to CK2a, and the isoform of NDPK in 19 physiologically relevant conditions. The test agent may be exposed to CK2cz and the * isoform of NDPK in viva or in vitro. The test agent should preferably act specifically in: .: 21 modulating thebinding of CK2a to the isoform of NDPK. S... * S

22 *55 *SS* 23 Preferably, the test agent is exposed to CK2a and either the isoform of NDPK in a.. : 24 cell-based assay. For example, the test agent may be exposed to an animal cell culture in which CK2o and the isoform of NDPK are constituents of the cells. The 26 cells culture is preferably a mammalian cell culture and most preferably a human cell 27 culture, such as HeLa. The cell culture may comprise a human derived liver cell line, 28 such as 1-lep G2. The cells may then be lysed, for example using mechanical lysing 29 means or a detergent. One or more of CK2a, the isoform of NDPK, and the test agent may then be separated from the lysate for analysis. For example, CK2cz and 31 the isoform of NDPK respectively may be co-immunoprecipitated.

33 The test agent may be exposed to CK2a and the isoform of NDPK in a cell-free 34 assay. A cell-free assay may be carried out using purified proteins in a suitable buffer. Appropriate conditions are disclosed below. CK2a and the isoform of NDPK 36 may be purified from cell cultures. Purified recombinant proteins may be used.

1 Recombinant proteins can readily be prepared using standard molecular biology 2 methods, for example as disclosed in Mo/ecu/ar Cloning: A Laborato,y Manual (J.

3 Sanbrook and D. W. Russell, 3 Edition, 2000, published by Cold Spring Harbor 4 Laboratory Press). A cell-free assay may be carried out using an extract from an animal cell line, preferably an extract from the cytosols of animal cells, most 6 preferably an extract from the cytosols of human derived liver cells. Suitable 7 extraction protocols are disclosed in the experimental section below.

9 Whether the test agent has modulated the binding of CK2a to the isoform of NDPK, may be determined by carrying out a binding assay which provides a quantitative or 11 qualitative measure of the binding of CK2a to the isoform of NDPK.

13 For example, CK2a or the isoform of NDPK may be concentrated and/or separated 14 using a recognition molecule which specifically binds to one of CK2a or the isoform of NDPK (which are binding partners), such as an antibody to one of CK2a or the 16 isoform of NDPK, which recognition molecule is preferably attached to a support, 17 such as beads or a surface. The conditions should be such that the other of CK2a or 18 the isoform of NDPK remains complexed to the protein which is concentrated and/or 19 separated. A wash step may be used to wash off unbound materials, if required.

Thereafter, an assay may be performed to determine whether the other of CK2a or 21 the isoform of NDPK has been concentrated and/or separated along with its binding 22 partner. The assay may detect the presence of the other of CK2a or the isoform of 23 NDPK using a labelled biological recognition molecule specific to the other of CK2a..

24 or the isoform of NDPK. The labelled biological recognition molecule may be an enzyme-linked antibody and the assay to detect the presence of the other of CK2a or 26 the isoform of NDPK may be an enzyme-linked immunosorbent assay (ELISA), such 27 as a direct, indirect and/or sandwich and/or competitive ELISA. Western blot analysis 28 may also be used to assess the binding of CK2a to the isoform of NDPK. The assay 29 may be carried out using detectably labelled CK2a and/or isoform of NDPK to facilitate the assay. Other methods for measuring protein-protein binding are well 31 known to those skilled in the art, for example surface plasmon resonance, analytical 32 ultracentrifugation and isothermal titration calorimetry.

34 Another way to determine whether the other of CK2a or the isoform of NDPK has been concentrated and/or separated along with its binding partner is to carry out an 36 assay for an activity of CK2a or the isoform of NDPK. By "an activity" of a protein we 1 mean a reaction catalysed by that protein. For example, the method may comprise 2 the step of assaying a kinase activity of CK2a (or CK2 as a whole), or a kinase 3 activity of the isoform of NDPK, such as the conversion of nucleoside diphosphates to 4 nucleoside triphosphates. A kinase activity of CK2a (or CK2 as a whole) may be assayed by measuring the incorporation of 32P from a solution of [y32P] ATP into a 6 peptide substrate of CK2a (or CK2 as a whole), such as the specific synthetic peptide 7 substrate, "CK2-tide" [RRRADDSDDDDD] discussed below. This kinase activity of 8 CK2ci may vary dependent on whether CK2cz is bound to an isoform of NDPK. We 9 have previously demonstrated that under certain circumstances "substrate channelling" occurs whereby ATP produced from GTP + ADP, by an NDPK-catalysed 11 reaction, alters the activity of AMPK al in an AMP-independent manner and 12 irrespective of the surrounding AlP concentration (6). In this reaction, AMPK al 13 receives AlP inside a complex of proteins shielded from the prevailing conditions in 14 the surroundings. A similar mechanism may channel ATP to CK2a whilst it is bound to an isoform of NDPK meaning that the kinase activity of CK2a may vary dependent 16 on whether CK2a is bound to NDPK under at least some circumstances. The change..

17 in CK2u activity may provide a measure as to whether CK2o is bound to an isoform.* 18 of NDPK. A kinase activity of an isoform of NDPK (such as NDPK-B) may be 19 assayed by incubating NDPK-B with GTP and ADP and measuring ATP which is produced using firefly luciferase, which emits luminescence in proportion to the 21 concentration of AlP, and a luminometer. (ATP detection kits are available from 22 Sigma-Aldrich, Inc. of St. Louis, Missouri). 23.

24 Whether the test agent has modulated the binding of CK2a to the isoform of NDPK may be determined by carrying out a binding assay which provides a quantitative or 26 qualitative measure of a parameter related to the binding of CK2a to the isoform of 27 NDPK, for example the amount of unbound CK2a or isoform of NDPK or test agent, 28 or, in circumstances where the test agent disrupts the binding of CK2a to the isoform 29 of NDPK, by binding to either or both of CK2a and the isoform of NDPK, by assaying the binding of test agent to CK2cx or isoform of NDPK.

32 The binding of test agent to CK2a or the isoform of NDPK can be measured by 33 methods which are equivalent to those discussed above for assaying the binding of 34 CK2a to the isoform of NDPK. Unbound CK2a or isoform of NDPK or test agent may be assayed by removing bound materials using a recognition molecule which 36 specifically binds one of CK2a, the isoform of NDPK or the test agent, and assaying I the amount of CK2a, isoform of NDPK, or test agent which is unbound in the resulting 2 supernatant. Preferably, the isoform of NDPK and molecules which are bound to the 3 isoform of NDPK, such as CK2a or test agent, are removed using a recognition 4 molecule which specifically binds the isoform of NDPK. The recognition molecule may be an antibody to the isoform of NDPK, and molecules which are bound to the 6 isoform of NDPK may be removed by immunoprecipitation. The amount of CK2o or 7 test agent left unbound can then be assayed, for example using an enzyme-linked 8 immunosorbent assay (ELISA), such as a direct or indirect and/or sandwich and/or 9 competition ELISA.

11 Particularly in the case of cell-free assays, one of CK2a, the isoform of NDPK or the 12 test agent may be immobilised on a surface or bead prior to the exposure of the test 13 agent to CK2cx, and the isoform of NDPK. The binding of another of CK2a, the 14 isoform of NDPK, or the test agent to the immobilised substance can then be measured, for example, using surface plasmon resonance or an ELISA.

17 A labelled form of CK2a, isoform of NDPK, or test substance may be used to facilitate * : : 18 the binding assay. For example, CK2a, the isoform of NDPK, or the test substance 19 may be radiolabelled, labelled with a luminescent, fluorescent or chemiluminescent material, or an enzyme which catalyses a detectable, and preferably quantifiable 21 reaction. CK2a and the isoform of NDPK, or the test substance and CK2a or the ** 22 isoform of NDPK, may be labelled with labels which participate in fluorescence *** 23 resonance energy transfer when the labelled substances are bound to each other. ... : Whether the test agent has modulated the binding of CK2cz to the isoform of NDPK 26 may be determined by carrying out a qualitative or quantitative activity assay which 27 measures an activity of CK2a or the isoform of NDPK, which varies dependent on the 28 binding of CK2a to the isoform of NDPK, i.e. where the binding of CK2a to the 29 isoform of NDPK affects the kinetics of a reaction catalysed by CK2a or the isoform of NDPK.

32 It should be borne in bind that, when CK2a binds an isoform of NDPK, this should 33 affect one or more activities of both CK2a and the isoform of NDPK. Some activities 34 of the isoform of NDPK may be activated by the binding of CK2a to the isoform of NDPK, for example the downstream activation of c-myc may be activated by the I binding of CK2a to NDPK-B; others may be inhibited, for example the interconversion 2 of nucleoside diphosphates and nucleoside triphosphates.

4 For example, the method may comprise the step of assaying a kinase activity of CK2a (or CK2 as a whole) or a kinase activity of the isoform of NDPK, such as the 6 conversion of nucleoside diphosphates to nucleoside triphosphates, as discussed 7 above. The rate of phosphorylation of "CK2-tide" [RRRADDSDDDDDJ or another 8 substrate by CK2, or the rate of conversion of nucleoside diphosphates to nucleoside 9 triphosphates by an isoform of NDPK are each sensitive to the binding of CK2cz to the isoformofNDPK.

12 The measured activity of CK2a or the isoform of NDPK, which varies dependent on 13 the binding of CK2ci to the isoform of NDPK, may be an activity of a kinase which 14 forms a complex with CK2a or the isoform of NDPK, or which is activated by a complex comprising CK2a or the isoform of NDPK, provided that that activity varies 16 dependent on the binding of CK2cx to the isoform of NDPK. For example, CK23 forms 17 a complex with NDPK-A and NDPK-B and perhaps other isoforms of NDPK in vivo.

18 CK213 is known to be promiscuous, i.e. to bind to and/or activate many different 19 kinases. We hypothesise that the kinase activity of CK2t3, or an activity of kinases * upon which CK23 acts, may vary dependent on the binding of CK2ci to an isoform of 21 NDPK. Any such activity could be measured.

I

23 Whether the test agent has modulated the binding of CK2a to the isoform of NDPK.. : 24 may be determined by assaying the phosphorylation of NDPK-B because CK2a phosphorylates NDPK-B. The phosphorylation of NDPK-B can be measured using 26 an anti-phosphoserine antibody, for example in an ELISA.

28 In a preferred embodiment, the test agent is exposed to intracellular CK2a and the 29 isoform of NDPK in a cell-based assay. The cells are then lysed and CK2a, or preferably the isoform of NDPK, is then separated (e.g. immunoprecipitated) using an 31 antibody to CK2cx, or preferably an antibody to the isoform of NDPK, respectively.

32 The resulting precipitate is then assayed for an activity of the isoform of NDPK (for 33 example, the conversion of nucleoside diphosphates to nucleoside triphosphates) or 34 CK2a (for example, the phosphorylation of the synthetic peptide substrate "CK2-tide"). In one embodiment, an antibody to CK2a is used to immunoprecipitate the 36 isoform of NDPK which is then assayed using an assay for the activity of the isoform I of NDPK. In another embodiment, an antibody to the isoform of NDPK is used to 2 immunoprecipitate the isoform of NDPK, which is then assayed for the activity of the 3 isoform of the NDPK.

Whether the test agent has modulated the binding of CK2a to the isoform of NDPK 6 may be determined by carrying out an assay of the phosphorylation of the isoform of 7 NDPK, for example, using an anti-phosphoserine antibody. A reduction in the 8 amount of phosphorylated isoform of NDPK compared to a control sample indicates 9 that the binding of CK2a to the isoform has been disrupted, where as an increase indicates the binding of CK2a to the isoform has been promoted.

12 The test agent may be introduced into an experimental animal, such as a non-human 13 mammal, for example a rodent or rabbit. The experimental animal may be genetically 14 altered so that a greater amount of CK2a binds the isoform of NDPK, under basal conditions or in response to a specific stimulus, than would be the case in equivalent * VI.

16 circumstances in the corresponding wild type animal. For the purposes of identifying 17 agents which have the property of disrupting the binding of CK2a to NDPK-B or 18 promoting the binding of CK2a to NDPK-A, the animal may be genetically modified so 19 that it expresses a variant of NDPK-A which comprises a mutation that reduces or obviates the binding of CK2a to NDPK-A (for example, where residue 120 of NDPK-A 21 is glycine or alanine). The experimental animal may be genetically modified so that it 22 does not express NDPK-A, for example, it may be an NDPK-A knockout non-human 1* *.*.

23 mammal (e.g. a mouse). The preparation of an NDPK-A knockout mouse is..

24 disclosed in Arnaud-Dabernat, S. et al., 2003 (22). An NDPK-B knockout mouse or mutant may be used for the purposes of identifying agents which have the property of 26 disrupting the binding of CK2a to NDPK-A or promoting the binding of CK2a to 27 NDPK-B.

29 The test agent may be introduced into the experimental animal. Where the test agent is a peptide, DNA or RNA coding for the test agent may be introduced into the test 31 animal using a suitable gene therapy vector which causes at least some cells of the 32 test animal to express the test agent. In order to assess whether the test agent has 33 modulated the binding of CK2a to the isoform of NDPK, cells from the experimental 34 animal may be lysed using mechanical lysing means or a detergent.

I In the case of experiments carried out in vivo in order to assess whether the test 2 agent has disrupted the binding of CK2a to the isoform of NDPK, the progress of a 3 disease linked to the excess binding of CK2a to the isoform of NDPK may be 4 monitored. For example, the progress or development of a cell proliferative condition, such as cancer (e.g. a neuroblastoma) linked to the excess binding of CK2a to 6 NDPK-B may be monitored. In the case of experiments carried out in vivo in order to 7 assess whether the test agent has promoted the binding of CK2a to the isoform of 8 NDPK, the progress of a disease linked to sub-normal binding of CK2o to the isoform 9 of NDPK may be monitored. For example, the progress or development of a cell proliferative, such as cancer (e.g. a neuroblastoma) linked to the sub-normal binding 11 of CK2a to NDPK-A may be monitored.

13 A change in the rate of proliferation of cancer, or tumour cell motility, or cell 14 proliferation or the rate of metastasis, or a proportion of a group of experimental animals which develop a cell proliferative condition, such as cancer, linked to the S...

16 excess or sub-normal binding of CK2a to the isoform of NDPK may be evidence that 17 the test agent has modulated the binding of CK2a to the isoform of NDPK.

S

19 A measurement to determine whether a test agent has modulated the binding of S *S CK2a to the isoform of NDPK may be interpreted with reference to a control in which 21 no test agent is present and CK2a binds the isoform of NDPK. A difference between 22 a qualitative or quantitative measurement from a binding or activity assay subsequent 23 to or during the exposure of test agent to CK2a and the isoform of NDPK, and the 24 said control may be indicative that the test agent has modulated the binding of CK2a to the isoform of NDPK.

27 The test agent may be introduced to CK2a and the isoform of NDPK in the presence 28 of another isoform of NDPK (indeed, this will usually be the case in an in viva or cell- 29 based assay). For example the test agent may be introduced to CK2a and both NDPK-A and NDPK-B.

32 The method may include the step of comparing the binding of CK2a to NDPK-B in the 33 presence of the test agent in circumstances where substantially no NDPK-A 34 molecules, or a relatively low proportion of NDPK-A molecules, are phosphorylated at S122, to the binding of CK2a to NDPK-B in the presence of the test agent in 36 circumstances where substantially all NDPK-A molecules, or a relatively high 1 proportion of NDPK-A molecules are phosphorylated at S122. Thus, the method may 2 comprise the step of phosphorylating NDPK-A at S122, for example using activated 3 AMPK. Methods for phosphorylating NDPK-A at S122 using activated AMPK are 4 discussed below.

6 We have demonstrated that, in normal circumstances /n vivo, CK2a binds to NDPK-A 7 if S122 of NDPK-A is unphosphorylated and translocates to NDPK-B in response to 8 the phosphorylation of NDPK-A at S122 by activated AMPK. Accordingly, a 9 therapeutic agent which disrupted the binding of CK2cz to NDPK-B should preferably not disrupt the binding of CK2a to NDPK-A, or disrupt it to only a limited extent.

11 Similarly, a therapeutic agent which disrupted the binding of CK2a to NDPK-A should 12 preferably not disrupt the binding of CK2a to NDPK-B, or disrupt it to only a limited 13 extent.

Preferably, a therapeutic agent which disrupts the binding of CK2a to NDPK-B, or to 16 NDPK-A respectively, should reduce the amount of CK2a bound to NDPK-A, or to 17 NDPK-B respectively, by less than 25%, more preferably less than 10%. It may well 18 increase the amount of CK2ci bound to NDPK-A, or to NDPK-B respectively. S ** S. S *5S*

If the test agent competes with NDPK-B to bind to CK2ci, it should preferably have a 21 greateraffinityforCK2athan NDPK-B (forexample, byafactorof atleast2, 3,10,30 22 or preferably 100) but a lower affinity for CK2a than NDPK-A in which S122 is not 23 phosphorylated (for example, by a factor of at least 2, 3, 10, 30 or 100). If the test 24 agent disrupts the binding of CK2cz to NDPK-B by binding the CK2a binding site on NDPK-B, it should have a lower affinity for the CK2a binding site on NDPK-A which is 26 not phosphorylated at S122 then for the CK2a binding site on NDPK-B, for example 27 byafactorof 10, l000r 1000.

29 Thus, the method may comprise assessing whether the test agent has the property of modulating the binding of CK2a to another isoform of NDPK. For example, where the 31 method is a method of identifying an agent which has the property of modulating the 32 binding of CK2o to NDPK-B, the method may further comprise assessing whether the 33 test agent has the property of modulating the binding of CK2a to NDPK-A. Where the 34 method is a method of identifying an agent which has the property of modulating the binding of CK2a to NDPK-A, the method may further comprise assessing whether the 36 test agent has the property of modulating the binding of CK2a to NDPK-B.

2 The method may be a method of identifying an agent which has the property of 3 modulating the binding of CK2o to the isoform of NDPK (e.g. disrupting the binding of 4 CK2a to NDPK-B) in physiologically relevant conditions, whilst minimising modulation (e.g. disruption) of the binding of CK2a to another isoform of NDPK (e.g. NDPK-A).

7 The method may further comprise assessing whether the test agent modulates (for 8 example, inhibits or activates) one or more activities (such as an activity of a kinase, 9 which it would not be desirable to modulate in vivo), for the purpose of identifying an agent which has the property of modulating the binding of CK2ci to an isoform of 11 NDPK whilst minimising side effects due to the undesirable modulation of said activity 12 or activities.

14 The test agent may comprise or consist of a peptide. The test agent may be a synthetic chemical entity. The test agent may be selected from a library of such test * *..

16 agents created for the purpose of screening. * :: 18 Examples of peptides which could be screened in order to assess whether they have 19 the property of modulating the binding of CK2a to an isoform of NDPK include: *. S...

fragments of an isoform of NDPK (for example, fragments of NDPK-A or NDPK-B or a 21 hybrid of NDPK-A and NDPK-B), or variants thereof. For example, fragments of *.

22 NDPK-B, or variants thereof, may be screened to assess whether they have the.: 23 property of modulating the binding of CK2a to NDPK-B, for example, by competing 24 with NDPK-B to bind CK2a.

26 The sequence of human NDPK-B, Genbank accession number P22392, is: 28 1 manlertfia ikpdgvqrgl vgeiikrfeq kgfrlvamkf lraseehlkq 29 51 hyidlkdrpf fpglvkymns gpwamvweg lnvvktgrvm lgetnpadsk 101 pgtirgdfci qvgrniihgs dsvksaekei slwfkpeelv dykscahdwv 31 151 ye 33 The sequence of human NDPK-A, Genbank accession number P15531, is: 1 mancertfia ikpdgvqrgl vgeiikrfeq kgfrlvglkf mqasedllke 36 51 hyvdlkdrpf faglvkymhs gpwamvweg lnvvktgrvm lgetnpadsk 1 101 pgtirgdfci qvgrniihgs dsvesaekei glwfhpeelv dytscaqnwi 2 151 ye 4 CK2a binds and phosphorylates S122 of NDPK-A and NDPK-B. The difference between NDPK-A and NDPK-B, in the exposed region from amino acids 111 to 125 6 which binds CK2cz, is that NDPK-B comprises lysine at residue 124, instead of 7 glutamic acid.

9 Thus, test agents which could be screened in order to assess whether they have the property of disrupting the binding of CK2a to NDPK- B include: 12 (i) peptides comprising 8, 10, 12, 16, 20, 24, or more than 24 and preferably no more 13 than 50, or no more than 100, consecutive residues of NDPK-B including Lys124, or 14 variants thereof including variants with 1, 2, 3, 4, 5, 6 or 7 differences from the corresponding sequence of NDPK-B within 10 residues inclusive of Lys124; and......

17 (ii) peptides comprising or consisting of residues 111 to 125 of NDPK-B (i.e. 18 QVGRNIIHGSDSVKS), or variants thereof including variants with 1, 2, 3,4, 5, 6 or 7 19 differences from the corresponding region of NDPK-B, having fewer than 150, : ..

preferably fewer than 100, more preferably fewer than 50, and most preferably fewer 21 than 25 amino acid residues, S..., LL. ..

23 In each case, a difference comprises the deletion of an amino acid residue, the 24 insertion of an amino acid residue (whether the insertion is of a proteinogenic amino acid residue or a non-proteinogenic amino acid residue) or a substitution of one 26 amino acid residue for another (whether the substitution is conservative or preferably 27 non-conservative and where the substituted amino acid residue may be a non- 28 proteinogenic amino acid residue). Such peptides can readily be prepared by those 29 skilled in the art. Non-proteinogenic amino acid residues can be prepared by automatic peptide synthesis, such services being widely available. Peptides may 31 comprise detectable labels (e.g. N-terminal fluoresceiri) or reactive moieties which 32 can be used to concentrate or immobilise the peptide, e.g. N-terminal biotin. Peptide 33 test agents, particularly for use in cell-based assays or animal experiments, may 34 comprise modifications to prolong their lifetime within a cell or animal, for example, an amino terminal capping group (e.g. an acetyl group or carbobenzyl) and/or a carboxy 36 terminal capping group (e.g. an amide group). Variants may also comprise 1 peptidomimetics of the described peptides, being synthetic chemicals designed 2 starting from a specific peptide to have substantially the same structural and/or 3 functional characteristics but improved pharmacological properties, such as 4 resistance to peptidases in vitro. Preferably, the residue which is equivalent to Lys124 is lysine.

7 We hypothesize that, once CK2a has bound NDPK-B, the resulting complex 8 translocates to the cell nucleus where it directly or indirectly activates c-myc.

9 Accordingly, the fragment of NDPK-B, or a variant and/or fusion thereof, is preferably selected so that a complex between the said fragment or variant and/or fusion thereof 11 and CK2o is not translocated to the cell nucleus, or is translocated to a lesser extent 12 than a complex between CK2a and NDPK-B, and the method may include assessing 13 whether the fragment of NDPK-B, or a variant and/or fusion thereof, is trarislocated to 14 the cell nucleus in vivo and, if so, comparing the amount of the fragment of NDPK-B, or variant and/or fusion thereof, which is translocated to the nucleus with a control *. S...

16 using wild type NDPK-B.

18 Corresponding test agents which could be screened in order to assess whether they * 19 have the property of disrupting the binding of CK2a to NDPK-A would preferably: .. S...

comprise equivalent fragments of NDPK-A, or variants thereof, and the residue which 21 is equivalent to residue 124 of NDPK-A would preferably be glutamic acid.

22 * : 23 The invention also extends to assessing whether the test agent has the property of 24 modulating the binding of CK2a to an isoform of NDPK by exposing the test agent to a variant, fragment and/or fusion of CK2a and/or a variant, fragment and/or fusion of 26 the isoform of NDPK.

28 The term variant, fragment and/or fusion of a protein includes variants of the protein, 29 fragments of the protein and variants of fragments of the protein, and fusions of any of these. A variant of CK2a or an isoform of NDPK may differ from the wild type 31 sequence by the addition, insertion, deletion or substitution of one or more amino 32 acid residues, but should retain the property of binding to the isoform of NDPK, and, 33 at least in the case of NDPK-A or NDPK-B, preferably also phosphorylate the isoform 34 of NDPK at S120 (in the case of CK2a) or binding to CK2a (in the case of an isoform of NDPK). To be useful in the screening method of the present invention, a fragment 36 of CK2a should include the binding site of CK2a for the isoform of NDPK, and a I fragment of an isoform of NDPK should include the binding site of the isoform of 2 NDPKforCK2a.

4 CK2a and/or the isoform of NDPK, or variants or fragments thereof may be fused to other peptides or chemical entities, for example, peptides or chemical entities which 6 can be used for purification (e.g. a polyhistidine tail which can be used to purify or 7 concentrate the peptide or a complex thereof by virtue of its affinity for nickel, or 8 biotin) or as labels (e.g. cyan fluorescent protein or yellow fluorescent protein for use 9 in fluorescence resonance energy transfer, or an enzyme, such as glucose oxidase, or a fluorescent dye). I'

12 The invention extends in a second aspect to the use of an isoform of NDPK and 13 CK2a, or variants, fragments and/or fusions thereof, in a screening assay to identify 14 agents which have the property of modulating the binding of the isoform of NDPK to CK2a. .... *...

17 The agents may have the property of either promoting or disrupting the binding of the 18 isoform of NDPK to CK2a. The isoform of NDPK is preferably NDPK-A or NDPK-B. * in. S. S...

Preferably, free CK2a (or a variant, fragment and/or fusion thereof) which is not in the *5S* 21 form of a dimer with CK23 is used. It may, however, be important for CK2 to be 22 present complexed to the isoform of NDPK. Suitable test agents for screening are * 23 discussed above.

A third aspect of the invention provides a kit comprising an isoform of NDPK and 26 CK2a, or variants, fragments and/or fusions thereof, and one or more test agents 27 suitable for screening to assess whether they have the property of modulating the 28 binding of the isoform of NDPK to CK2a.

The kit may be suitable for screening whether the test agent has the property of either 31 promoting or disrupting the binding of the isoformof NDPK to CK2a. The isoform of 32 NDPK is preferably NDPK-A or NDPK-B.

34 Preferably, the kit comprises CK2a which is not in the form of a dimer with CK2F3.

Preferably, the kit does not include CK213. Suitable test agents for screening are 36 discussed above.

2 The invention extends in a fourth aspect to the use of an experimental animal, which 3 has been genetically altered so that a greater or lower amount of CK2a binds an 4 isoform of NDPK under basal conditions or in response to a specific stimulus than would be the case in equivalent circumstances in the corresponding wild type animal, 6 and a test agent, in an experiment to assess whether the test agent has the property 7 of modulating the binding of CK2a to the isoform of NDPK.

9 The experiment is typically for assessing whether the test agent has the property of either promoting or disrupting the binding of the isoform of NDPK to CK2a. The 11 isoform of NDPK is preferably NDPK-A or NDPK-B. The genetic alteration which 12 causes a greater or lower amount of CK2ci to bind to a first isoform of NDPK may be 13 an alteration in a second isoform of NDPK. For example, an experimental animal 14 may be genetically altered so that a greater amount of CK2a binds NDPK-B, by altering expressed NDPK-A, for example to S12OG or S12OA. 16.:

17 Optional features of the experimental animal and test agent are described above.

18 Preferably, the amount of CK2cz which binds the isoform under basal conditions or in 19 response to a specific stimulus is more than 3, 10, 30 or 100 times the amount which: ..

is bound under equivalent circumstances in the corresponding wild type animal, or S...

21 less than a third, a tenth, a thirtieth or a hundredth the amount which is bound under 22 equivalent circumstances in the corresponding wild type animal. *:::: 24 A fifth aspect of the invention provides a method of developing a pharmaceutical product, comprising identifying one or more agents using the method of the first 26 aspect and modifying the one or more identified agents to optimise their 27 pharmaceutical properties. The invention also extends to a pharmaceutical product 28 developed by this method.

Thus, an identified agent may be used as a pharmaceutical composition in 31 combination with a suitable carrier, excipient or stabiliser. However, the identified 32 agent may be a lead compound used to develop another agent suitable for use as the 33 active ingredient in a pharmaceutical composition. The process of developing a 34 suitable agent may include applying rational drug design techniques known to those skilled in the art and the screening of variants of the identified agent. Agents may be 36 prepared and tested which differ from an identified agent in terms of their activity to I modulate the binding of CK2a to the isoform of NDPK, and in terms of other 2 pharmaceutically important properties, such as their half-life in vivo, or associated 3 side effects.

The method may further comprise the step of formulating the agent into a 6 pharmaceutically composition with a pharmaceutically acceptable carrier, excipient or 7 stabiliser. The formulation will depend on the intended route of administration, for 8 example whether it is to be administered by a topical, enteral or parenteral route.

In the case of an agent which consists of a peptide comprising proteinogenic amino 11 acids, the method extends to preparing a gene therapy vector comprising DNA or 12 RNA encoding the identified or modified agent and which is operable, when 13 introduced into a subject, to cause at least some cells of the subject to synthesise the 14 agent. S...

S S

16 Gene therapy vectors and methods for using gene therapy vectors to induce the 17 production of a specified protein in a patient are well known to those skilled in the art 18 and are reviewed in Gardlik et al., Med. Sci. Monit, 2005; 11(4): RAI1O-121, which is 19 incorporated herein by reference. : .. * SSS

21 The pharmaceutical product may be for use in the treatment and/or prevention of a 22 cell proliferative condition, such as cancer (e.g. neuroblastoma) or psoriasis. The *:: 23 pharmaceutical product may be for use in the treatment and/or prevention of a cell 24 proliferative condition, such as cancer, which results from increased or reduced binding of CK2a to the isoform of NDPK. For example, the pharmaceutical product 26 may be for use in the treatment and/or prevention of a cell proliferative condition, 27 such as a cancer, which results from increased binding of CK2a to NDPK-A due to a 28 mutation in the binding site of NDPK-A for CK2a. (e.g. a mutation in which residue 29 120 of NDPK-A is glycine or alanine).

31 According to a sixth aspect of the invention, there is provided an agent identified by 32 the method of the first aspect of the present invention or developed by the method of 33 the fifth aspect of the present invention, for use in therapy.

Where the agent is a peptide, the invention also extends to a gene therapy vector 36 which is operable, when introduced into a host, to cause a cell within a host to I express a peptide agent identified by the method of the first aspect of the invention or 2 developed by the method of the fifth aspect of the present invention. Accordingly, the 3 gene therapy vector typically comprises DNA or RNA encoding a peptide agent 4 identified by the method of the first aspect of the invention.

6 According to a seventh aspect of the present invention there is provided a 7 pharmaceutically acceptable agent which has the property of modulating the binding 8 of CK2a to the isoform of NDPK in viva. The agent may either disrupt or promote the 9 binding of CK2a to the isoform of NDPK and should preferably do so specifically.

The pharmaceutically acceptable agent may be combined with a pharmaceutically 11 acceptable carrier, excipient or stabiliser. The invention also extends to a gene 12 therapy vector which, when introduced into a host, causes cells with the host to 13 express a peptide which has the property of modulating the binding of CK2a to the 14 isoform of NDPK in viva. S... I * S...

16 The agent may be an agent identified by the method of the first aspect of the 17 invention or developed by the method of the fifth aspect of the invention. The agent 18 (or the peptide expressed in a host by virtue of the action of the gene therapy vector) * 19 may be a fragment of an isoform of NDPK (which need not be the same isoform), or a: *.

variant and/or fusion thereof, which binds to CK2a in vi vo and which has the property * I..

21 of disrupting the binding of CK2a to the isoform of NDPK by competing with the 22 isoform of NDPK to bind CK2ci. The fragment of an isoform of NDPK, or a variant.: 23 and/or fusion thereof, may correspond to one of the fragments of NDPK-A or NDPK- 24 B, or variants and/or fusions thereof, which are described as potential test agents in relation to the first aspect of the invention above.

27 According to an eighth aspect of the invention there is provided a method of 28 modulating the binding of CK2a to an isoform of NDPK in a patient in need thereof, 29 the method comprising administering an effective amount of an agent which has the property of modulating the binding of CK2o to the isoform of NDPK, or an effective 31 amount of a gene therapy vector which causes cells within a patient to express an 32 effective amount of a peptide agent which has the property of modulating the binding 33 of CK2a to the isoform of NDPK.

Preferably, the agent is an agent which was identified by the method of the first 36 aspect of the invention or developed by the method of the fifth aspect of the invention.

1 The agent (or the peptide expressed in a host by virtue of the action of the gene 2 therapy vector) may be a fragment of an isoform of NDPK (which need not be the 3 same isoform), or a variant and/or fusion thereof, which binds to CK2a in vivo and 4 which has the property of disrupting the binding of CK2a to the isoform of NDPK by competing with the isoform of NDPK to bind CK2a. The fragment of an isoform of 6 NDPK, or a variant and/or fusion thereof, may correspond to one of the fragments of 7 NDPK-A or NDPK-B, or variants and/or fusions thereof, which are described as 8 potential test agents in relation to the first aspect of the invention above.

The modulating of the binding of CK2cx to an isoform of NDPK may be promotion or 11 disruption of the binding of CK2a to an isoform of NDPK. The isoform of NDPK is 12 preferably NDPK-A or NDPK-B.

14 The agent may have the property of disrupting the binding of CK2a to the isoform of NDPK, because it modifies CK2a, or the isoform of NDPK, to reduce or eliminate the S...

16 affinity of CK2a for the isoform, or vice versa. The agent may be an allosteric 17 inhibitor of the binding of CK2a to the isoform of NDPK, or vice versa. The agent may 18 bind irreversibly to the binding site of CK2a for the isoform of NDPK, or vice versa.

19 The agent may be an irreversible or suicide inhibitor of the binding of CK2o to the isoform of NDPK, or vice versa.

S S

LI

22 The agent may have the property of promoting the binding of CK2a to the isoform of * : 23 NDPK, because it modifies CK2a, or the isoform of NDPK, to enhance the affinity of 24 CK2a for the isoform of NDPK, or vice versa. The agent may be an allosteric promoter of the binding of CK2a to the isoform of NDPK, or vice versa. The agent 26 may bind irreversibly to the binding site of CK2a for the isoform of NDPK, or vice 27 versa. The agent may be an irreversible or suicide promoter of the binding of CK2a 28 to the isoform of NDPK respectively, or vice versa.

The agent may have the property of disrupting the binding of CK2a to the isoform of 31 NDPK because it competes with the binding of CK2a to the isoform of NDPK, for 32 example because it binds to the binding site for CK2a on the isoform of NDPK, or 33 because it binds to the binding site for the isoform of NDPK on CK2cz, or because it 34 provides a binding site for CK2a which competes with the binding site for CK2a on the isoform of NDPK, or because it provides a binding site for the isoform of NDPK 36 which competes with the binding site for the isoform of NDPK on CK2a. The agent I may be a competitive inhibitor, mixed inhibitor or non-competitive inhibitor of the 2 binding of CK2a to the isoform of NDPK, or vice versa. The binding site of NDPK-A, 3 and of NDPK-B, for CK2a includes S120 of NDPK-AJNDPK-B which CK2a binds to 4 and phosphorylates. The binding site of NDPK-A, and of NDPK-B, for CK2Q comprises at least some of residues 111 to 125 which are exposed fri vivo.

7 The agent may comprise a fragment of the, or an isoform of NDPK, or a variant 8 and/or fusion thereof which binds to CK2a. The agent may comprise a gene therapy 9 vector which, when introduced into a host, causes the expression of a fragment of the, or an isoform of NDPK, or a variant and/or fusion thereof which binds to CK2cx.

12 Preferably, the fragment of the, or an isoform of NDPK, will not be operable to 13 catalyse one or more activities of the isoform of NDPK, to reduce any side effects.

Preferably, the method is a method of treating and/or preventing a cell proliferative 16 condition, such as a cancer (for example, neuroblastoma) or psoriasis. Preferably, 17 the method is a method of treating and/or preventing a cell proliferative condition, 18 such as a cancer, which results from increased or reduced binding of CK2a to the 19 isoform of NDPK, for example from increased binding of CK2a to NDPK-B, for: .. S...

example due to a mutation in the binding site of NDPK-A for CK2a. (e.g. a mutation in S...

21 which residue 120 of NDPK-A is glycine or alanine).

22.. : 23 In a ninth aspect, the invention extends to the use of an agent which modulates the 24 binding of CK2a to an isoform of NDPK in the manufacture of a medicament for treating and/or preventing a condition associated with increased or reduced binding 26 of CK2a to the or an isoform of NDPK.

28 The agent may promote or disrupt the binding of CK2ci to an isoform of NDPK. An 29 agent which disrupts the binding of CK2a to the isoform of NDPK would typically be used in the manufacture of a medicament for treating and/or preventing a condition 31 associated with increased binding of CK2a to the isoform of NDPK. An agent which 32 promotes the binding of CK2a to the isoform of NDPK would typically be used in the 33 manufacture of a medicament for treating and/or preventing a condition associated 34 with reduced binding of CK2a to the isoform of NDPK. However, the agent may be used in the manufacture of a medicament for treating and/or preventing a condition 36 associated with increased binding of CK2a to a different isoform of NDPK. For 1 example, an agent which promotes the binding of CK2a to NDPK-A, or to NDPK-B, 2 may be used in the manufacture of a medicament for treating and/or preventing a 3 condition associated with increased binding of CK2a to NDPK-B, or to NDPK-A 4 respectively.

6 The medicament may be for treating and/or preventing a cell proliferative condition, 7 such as a cancer (e.g. neuroblastoma) or psoriasis. Preferably, the medicament is 8 for treating and/or preventing a cell proliferative condition, such as a cancer, which 9 results from increased binding of CK2cx to NDPK-B, for example due to a mutation in the binding site of NDPK-A for CK2a. (e.g. a mutation in which residue 120 of NDPK- 11 A is glycine or alanine).

13 The agent may be identified by the method of the first aspect of the invention or 14 developed by the method of the fifth aspect of the invention. The agent (or the peptide expressed in a host by virtue of the action of the gene therapy vector) may be 16 a fragment of an isoform of NDPK (which need not be the same isoform), or a variant 17 and/or fusion thereof, which binds to CK2cx in vivo and which has the property of 18 disrupting the binding of CK2a to the isoform of NDPK by competing with the isoform 19 of NDPK to bind CK2a. The fragment of an isoform of NDPK, or a variant and/or: ..

fusion thereof, may correspond to one of the fragments of NDPK-A or NDPK-B, or 21 variants and/or fusions thereof, which are described as potential test agents in 22 relation to the first aspect of the invention above. * ,** : 24 According to a tenth aspect of the invention, there is provided a method of assessing whether a patient has an amount of CK2a bound to an isoform of NDPK, which differs 26 from a reference figure, the method comprising taking a sample of one or more cells 27 from the patient, lysing the one or more cells and measuring the amount of CK2ci 28 bound to the isoform of NDPK, in the lysed cells.

The tenth aspect of the invention also extends to assessing whether a patient has an 31 amount of CK2a bound to a mutant of the isoform of NDPK. It may be advantageous 32 to use a test which provides a measurement if any of a range of mutants of the 33 isoform of NDPK is present. If specific recognition molecules, such as antibodies 34 specific to the isoform of NDPK are used, the mutants may all be measured provided the mutations do not affect the recognition site on the isoform of NDPK of the specific 36 recognition molecules or any other activities of the isoform of NDPK which are I relevant to the test. However, the mutant may be a specific mutant and the test may 2 measure the amount of CK2o bound to the specific mutant of the isoform of NDPK 3 and compare this with a reference figure.

The method may be a method of assessing whether a patient has, or may develop, a 6 cell proliferative condition, such as a cancer, which should respond to treatment using 7 an agent which modulates the binding of CK2a to the isoform of NDPK. The 8 modulation may be in increase or decrease in the amount of binding of CK2a to the 9 isoform of NDPK.

11 The isoform of NDPK is preferably NDPK-A or NDPK-B.

13 The method may be repeated to determine whether a course of treatment using an 14 agent which modulates the binding of CK2o to the isoform of NDPK is working. The course of treatment may be part of a clinical trial. *:: ::* S *** L'.J * * 17 It may also be valuable to assess the distribution of CK2a, in cells taken from a 18 patient, between two or more of the states of (i) being corn plexed with a first isoform S 19 of NDPK (typically NDPK-B), (ii) being complexed with a second isoform of NDPK (typically NDPK-A), and (iii) being dissociated from both isoforms of NDPK (typically S...

21 NDPK-A and NDPK-B). This information may be helpful to a medical practitioner in 22 reaching a diagnosis concerning a patient who may be suffering from or who may 23 develop a disease associated with a higher than normal amount of CK2ci binding to 24 NOPK-B, or who may develop a disease associated with a higher than normal amount of CK2a binding to NDPK-A, or a higher than normal amount of CK2o free 26 within their cells. The disease may be a cell proliferative condition, such as cancer or 27 psoriasis. This information may be helpful when screening a population to identify 28 individuals who may be at risk of a disease associated with an increased or reduced 29 amount of CK2a bound to an isoform of CK2a, for example a higher or lower than normal amount of CK2a binding to NDPK-B (for example a cell proliferative condition 31 such as cancer), or a higher or lower than normal amount of CK2a free within their 32 cells, or when carrying out statistical studies. This information may be helpful when 33 assessing whether a patient may respond to treatment or benefit from prophylaxis 34 using an agent which modulates the binding of CK2a to NDPK-A or NDPK-B.

1 Accordingly, the invention extends in an eleventh aspect to a method of assessing 2 the distribution of CK2a, in cells within a patient, the method comprising the steps of 3 taking a sample of cells from a patient, lysing the cells, and assaying two or more of 4 (i) the amount of CK2a, which is bound to a first isoform of NDPK (typically NDPK-B), (ii) the amount of CK2a which is bound to a second isoform of NDPK (typically 6 NDPK-A), and (iii) the amount of CK2ci which is not bound to either isoform of NDPK.

8 It may be that the cells within a patient have a mutant form of the first and/or second 9 isoforms of NDPK and the eleventh aspect of the invention extends to assessing the distribution of CK2a between normal or mutant forms of the first and second isoforms 11 of NDPK. It may be advantageous to use a test which provides a method of 12 assessing the distribution of CK2a between a range of mutants of the isoforms of 13 NDPK. If specific recognition molecules, such as antibodies specific to the isoform of 14 NDPK are used, the method should work provided that the mutations do not affect the recognition site on the isoform of NDPK of the specific recognition molecules or any * 16 other activities of the isoform of NDPK which are relevant to the test. However, a:: 17 mutant may be a specific mutant and the test may comprise assaying the amount of 18 CK2a which is bound to a specific mutant of the first and/or second isoform of NDPK.

Ir. S. S. S S...

In order to assess the distribution of CK2a in the cytoplasm of cells, lysis (for S...

21 example, using mechanical lysing means or a detergent) will generally be sufficient to 22 enable access to the proteins which are the subject of the assay. The method may 23 however comprise extracting proteins from another region of the cells, for example, 24 within the nuclei, in order to assess the distribution of CK2a in that region of cells.

26 The amount of CK2a which is complexed with the first and/or second isoforms of 27 NDPK can be assessed by adapting the methods discussed above for providing a 28 quantitative or qualitative measure of the binding of CK2a to NDPK-B.

For example, complexes of CK2a to NDPK-A and/or NDPK-B may be concentrated 31 and/or separated using a recognition molecule (such as an antibody) which 32 specifically binds CK2a, in which case complexes of CK2a with NDPK-A and/or 33 NDPK-B will both be concentrated and/or separated together. NDPK-A and/or 34 NDPK-B which has been concentrated and/or separated with CK2a can then be assayed, for example using an ELISA (such as a direct, indirect and/or sandwich 36 and/or competition ELISA). CK2a may be immobilised using an antibody which binds 1 specifically to CK2a and two discernibly detectably labelled antibodies for NDPK-A 2 and/or NDPK-B (e.g. an antibody which binds specifically to NDPK-A labelled with a 3 fluorescent moiety of one colour, and an antibody which binds specifically to NDPK-B 4 labelled with a fluorescent moiety of another colour) may be used to simultaneously assay the amount of each of NDPK-A and NDPK-B which was bound to CK2a. Other 6 techniques which could be used to assay the binding of CK2a to NDPK-A and/or 7 NDPK-B include Western blotting.

9 Alternatively, complexes of CK2a to NDPK-A or NDPK-B may be concentrated and/or separated using recognition molecules (such as antibodies) which specifically bind 11 NDPK-A or recognition molecules which specifically bind to NDPK-B, thereby 12 enabling complexes of CK2a with only one of NDPK-A or NDPK-B to be concentrated 13 and/or separated. The amount of CK2cz which has bound NDPK-A or NDPK-B, as 14 appropriate, can then be assayed, for example by using a recognition molecule which specifically binds CK2cx (for example using a direct, indirect and/or sandwich and/or.

16 competition ELISA, or a Western blot), or an assay for an activity of CK2ci, for.*:: 17 example, by measuring the incorporation of 32P from a solution of [y32P] AlP into a 18 peptide substrate of CK2a such as the peptide the specific synthetic peptide 19 substrate, "CK2-tide" [RRRADDSDDDDD] discussed below. S...

21 The amount of free CK2a which is not bound to NDPK-A or NDPK-B can be assayed 22 by concentrating or separating both NDPK-A and NDPK-B from a cell lysate and 23 assaying the amount of CK2a which is left in the resulting supernatant, for example 24 using a recognition molecule which specifically binds CK2ci or an assay for an activity of CK2a. NDPK- A and NDPK-B may be removed using separate recognition 26 molecules which are specific for each of NDPK-A and NDPK-B (for example, 27 antibodies to NDPK-A and antibodies to NDPK-B) or recognition molecules which 28 bind both NDPK-A and NDPK-B (for example, antibodies which bind to both NDPK-A 29 and NDPK-B).

31 Preferably, an individual lysate is analysed to determine the amount of CK2a which 32 has bound NDPK-A, and the amount of CK2a which has bound NDPK-B, and the 33 amount of CK2a which has not bound NDPK-A or NDPK-B.

Equivalent assays may be used to assay the binding of CK2a to other isoforms of 36 NDPK.

2 The assay may be compared with one or more controls. The assay may be repeated, 3 for example to determine whether a drug which modulates the binding of CK2a to an 4 isoform of NDPK is working.

6 The invention also extends to synergistic uses of more than one therapeutic agent.

7 The binding of CK2cz to NDPK-A or NDPK-B is inhibitory of the nucleoside 8 phosphotransfer activity of NDPK (5). Where a causative factor of a disease, such as 9 a cell proliferative condition (e.g. cancer), is the increased binding of CK2a to an isoform of NDPK (e.g. NDPK-B), and hence increased inhibition of the isoform of 11 NDPK, a therapeutic agent which disrupts the binding of CK2a to the isoform of 12 NDPK, or which promotes the binding of CK2a to another isoform of NDPK (e.g. 13 NDPK-A), may be used synergistically with a therapeutic agent which inhibits CK2a 14 as both therapeutic agents will cause a reduction in the inhibition of the isoform of

NDPK g,-. S *4

17 We have demonstrated that herein that when S122 is pre-phosphorylated by AMPK *( I 18 cxl in v/va, CK2a translocates from NDPK-A and binds to the closely related NDPK-B: 19 isoform. Thus, a therapeutic agent which disrupts the binding of CK2a to NDPK-B or which promotes the binding of CK2a to NDPK-A, or a mutant of NDPK-A found in a **$S 21 particular patient, could be used synergistically with an inhibitor of AMPK activity. :. 22 "

23 Thus, a twelfth aspect of the invention provides the use of an inhibitor of CK2a, and 24 an agent which has the property of modulating the binding of CK2a to an isoform of NDPK, or a specific mutant thereof, in the treatment and/or prevention of a disease 26 associated with modulated binding of CK2cx to the isoform of NDPK, or specific 27 mutant thereof. The inhibitor of CK2a, and the agent which has the property of 28 modulating the binding of CK2a to the isoform of NDPK, or specific mutant thereof, 29 should have synergistic effects. Thus, the disease may be associated with increased binding of CK2a to the isoforrn of NDPK (e.g. NDPK-B), and the agent which has the 31 property of modulating the binding of CK2a to the isoform of NDPK may have the 32 property of disrupting the binding of CK2a to the isoform of NDPK (e.g. NDPK-B), or 33 the agent may have the property of promoting the binding of CK2a to a second 34 isoform of NDPK (e.g. NDPK-A).

36 The isoform of NDPK is preferably NDPK-A or NDPK-B.

2 Where the disease is associated with increased binding of CK2a to NDPK-B, or a 3 specific mutant thereof, the use may also comprise use of an agent which inhibits 4 AMPK. The agent which inhibits AMPK should preferably inhibit the al subunit of AMPK and may selectively inhibit AMPK olin preference to AMPK a2.

7 The disease is preferably a cell proliferative condition, such as a cancer, in which a 8 causative factor is increased or reduced binding of CK2a to the isoform of NDPK.

9 The disease may be neuroblastoma. The disease may be a cell proliferative condition, such as a cancer (e.g. a neuroblastoma), in which affected cells express a 11 mutated form of NDPK-A which reduces or prevents the binding of CK2a to NDPK-A, 12 such as a mutated form in which residue 120 of NDPK-A is glycine or alanine.

14 The inhibitor of CK2cx may act on CK2 dimers. The inhibitor of CK2cz should have the property of reducing the inhibition of NDPK phosphotransfer activity by CK2a.

16 Glitazones (thiazolidinediones) are agonists of peroxisome proliferator activated 17 receptor v and are used in the treatment of diabetes. Glitazones are described in 18 (43), the contents of which are herein incorporated by reference. It has recently been 19 demonstrated that troglitazone suppresses CK2 activity (44). Thus, the inhibitor of * CK2a is preferably a glitazone. The glitazone may be one of ciglitazone, englitazone, 21 troglitazone, rosiglitazone, pioglitazone or a pharmaceutically acceptable form such 22 as a salt, thereof. The inhibitor of CK2a may be a peptide which is expressed within 23 the cells of the patient by introducing into the patient a gene therapy vector which 24 comprises DNA or RNA encoding the inhibitor of CK2a and which is operable, when introduced into a subject, to cause at least some cells of the subject to synthesise the 26 said inhibitor of CK2a.

28 Preferably, the agent which has the property of modulating the binding of CK2a to the 29 isoform of NDPK, or specific mutant thereof, is an agent which was identified by the method of the first aspect of the invention or developed by the method of the fifth 31 aspect of the invention. The agent (or the peptide expressed in a host by virtue of the 32 action of the gene therapy vector) may be a fragment of an isoform of NDPK (which 33 need not be the same isoform), or a variant and/or fusion thereof, which binds to 34 CK2a in viva and which has the property ofdisrupting the binding of CK2a to the isoform of NDPK by competing with the isoform of NDPK to bind CK2a. The 36 fragment of an isoform of NDPK, or a variant and/or fusion thereof, may correspond I to one of the fragments of NDPK-A or NDPK-B, or variants and/or fusions thereof, 2 which are described as potential test agents in relation to the first aspect of the 3 invention above. Where the agent which has the property of modulating the binding 4 of CK2a to the isoform of NDPK, or specific mutant thereof, consists of a peptide comprising proteinogenic amino acids, the agent may be expressed within the patient 6 by introducing to the patient a gene therapy vector comprising DNA or RNA encoding 7 the agent which has the property of modulating binding of CK2a to the isoform of 8 NDPK, or specific mutant thereof, wherein the gene therapy vector is operable, when 9 introduced into a subject, to cause at least some cells of the subject to synthesise the said agent.

12 The inhibitor of AMPK may be compound C (6-[4-(2-Piperidin-1 -yl-ethoxy)-phenyl)]-3- 13 pyridin-4-yl-pyrrazolo[1,5-a]-pyrimidine), available from AMD Biosciences I Merck 14 Biosciences. The inhibitor of AMPK may be expressed in vivo by introducing into the patient a gene therapy vector comprising DNA or RNA encoding an agent which 16 inhibits AMPK and which, when introduced into a patient, causes at least some cells 17 within the patient to express the inhibitor of AMPK. The inhibitor of AMPK may, for * 18 example, be an antisense mRNA transcript which reduces or blocks translation of 19 AMPK mRNA, or may be an AMPK mutant which functions as a dominant negative. * .* 21 Where the agent which has the property of modulating binding of CK2a to the isoform *** 22 of NDPK, or specific mutant thereof, consists of a peptide comprising proteinogenic 23 amino acids, the agent may be expressed within the patient by introducing to the 24 patient a gene therapy vector comprising DNA or RNA encoding the agent which has the property of modulating binding of CK2a to the isoform of NDPK, or specific 26 mutant thereof, wherein the gene therapy vector is operable, when introduced into a 27 subject, to cause at least some cells of the subject to synthesise the said agent.

29 The (I) inhibitor of CK2o, (ii) the agent which has the property of modulating the binding of CK2cz to an isoform of NDPK, or a specific mutant thereof, and optionally 31 (iii) the inhibitor of AMPK, where relevant, may be delivered separately or in any 32 combination of two or all three agents. Each may be delivered in a pharmaceutically 33 acceptable form, in combination with a pharmaceutically acceptable carrier, excipient 34 or stabiliser and thus, the invention also extends to a composition comprising an inhibitor of CK2a and an agent which has the property of modulating the binding of 36 CK2a to an isoform of NDPK, or a specific mutant thereof. The composition may I further comprise a pharmaceutically acceptable carrier. The composition may further 2 comprise an inhibitor of AMPK.

4 Brief Description of the Drawings

6 An example embodiment of the invention will now be illustrated with reference to the 7 following Figures: 9 FIGURE 1: The NDPK-A I AMPK al complex interacts with protein kinase CK2 11 In all Western blot figures, each bead-linked precipitating antibody is shown at the top 12 of the blot and the relevant isoform-specific probing antibody at the bottom. Figure 1, 13 left panel: SDS-PAGE / Western blotting of CK2, NDPK-A and AMPK al 14 immunoprecipitates from an equal mixture (5 pg each) of four purified recombinant proteins: CK2a, NDPK-A, AMPK al and Kir4.2 (non-specificity control) probed 16 reciprocally as labelled. Figure 1, right panel: equivalent experiment but using 17 immunoprecipitates from a cell extract in a human derived liver cell line, Hep G2 as 18 input, using Kir4.2 as a non-specificity control (+CT, last lane). In both a recombinant.:.

19 reconstituted system and a native cytosolic extract, protein kinase CK2 is present in the cytosolic NDPK-A I AMPK al complex. Representative of n>3. * S S...

22 FIGURE 2: Which subunits of NDPK / AMPK interact with protein kinase CK2? * * * * . 24 Figure 2. uDDer left panel: A screen of 6 different peptides used as substrates for either AMPK (closed bars) or protein kinase, CK2 (open bars). The first four peptides 26 (NDPKpep 1, 2, 3 and 4) correspond to exposed regions of NDPKA (corresponding to 27 amino acids: 49-62, 92-105, 111-125 and 139-152 respectively) alongside relevant 28 kinase specificity substrate peptide controls; SAMS and CK2-tide. Purified, active rat 29 liver AMPK ci phosphorylated SAMS peptide, NDPKpep 3 and NDPKpep 4, whilst active recombinant CK2 was only able to phosphorylate CK2-tide and NDPKpep 3.

32 Figure 2. lower left panel: Dot blots of each of the four NDPK peptides spotted onto 33 nitrocellulose and overlaid with pure, recombinant CK2a, AMPK ci and a2 catalytic 34 subunits as labelled (left). AMPK ci and CK2a both selectively bind to the NDPKpep 3 region of NDPK-A.

I Figure 2, ftojt panel: Binding matrix of each of the NDPK and AMPK recombinant 2 subunits spotted onto nitrocellulose and overlaid with either recombinant CK2a or 3 CK213 as labelled. We observe that CK213 is capable of binding both isoforms of 4 NDPK (A and B) and both isoforms of AMPK 13(1 and 2). CK2a was observed to bind both NDPK isoforms (A and B) and selectively, the AMPK y2 isoform.

7 FIGURE 3: Investigating the S122-dependent interaction of CK2 with NDPK-A I 8 AMPKa1 Figure 3, left panel: Multiple kinase assays (+ AMPK or + CK2 specific protein 11 kinases as indicated) with recombinant wild-type NDPK-A as target. In bars 2 and 4, 12 NDPK-A was pre-phosphorylated for 10 minutes with the other kinase prior to 13 conducting the kinase assays (curly brackets). Prior phosphorylation of NDPK-A with 14 CK2 had no effect on the ability of AMPK to phosphorylate NDPKA (second bar, left panel) however, prior phosphorylation of NDPK-A with AMPK abolished the ability of * a.

16 CK2 to phosphorylate NDPK-A (last bar). * :: 18 Figure 3, upper right panel: A substrate channelling assay using recombinant NDPK- 19 A and CK2a added to purified, active rat liver AMPK and assayed in the presence of GTP, ADP or GTP + ADP (from which NDPK-A makes ATP as reported in (6).

21 Alternatively, we applied AlP GDP. Samples were taken for Western analysis and * *S** 22 immunoprecipitated using an NDPK-A antibody, probing for the presence of AMPK 23 al and CK2a (Figure 3, upper and lower blot right panels, respectively). Whenever 24 AMPK is active (substrate channelling), CK2 a is no longer in complex with NDPK-A / AMPK al (lane 3, lower blot) whereas under "substrate steal" conditions, CK2a is 26 present as it is under basal conditions and with GTP or ADP alone.

28 FIGURE 4: Activation of AMPK al results in exclusion of CK2a from the cytosolic 29 NDPK-A / AMPK al complex in a cell system 31 Figure 4A: Hep G2 cells were treated for 20 minutes with the foHowing compounds as 32 follows: 10 MM 4,5,6,7-tetrabromobenzotriazole (TBB), 1pM PKAi, 100 nM PKCi, 10 33 pM spermidine (Sp), 2 mM metformin (Met) and 2 mM phenformin (Phen). Cell 34 cytosol was extracted and immunoprecipitated using either NDPK-A or CK2a antibodies as labelled. Precipitates were probed for acetyl CoA carboxylase 1 36 (ACC1), AMPK al, CK2a, CK213 or NDPK-A. When cells were treated with AMPK 1 activators (metformin / phenformin), CK2a was lost from the complex (lanes 5 and 6).

2 CK2t3 was found to be present under all conditions tested and its association with 3 NDPK-A I AMPK ci complex was unaffected by the status of AMPK activity.

Figure 4B: Specific kinase assays performed on NDPK-A immunoprecipitations from 6 the various cell treatments, AMPK activity (closed bars) and CK2 activity (open bars).

7 This approach independently confirms our Western blotting findings.

9 Figure 4C: CK2 kinase assays using as target substrate purified, recombinant wild-type NDPK-A or NDPK-A S122A, NDPK-A S122D and NDPK-A S12OA mutants.

11 CK2a was unable to phosphorylate (left histogram) or bind (right panel) NDPK-A 12 when S120 was mutated to an alanine (S12OA). CK2a was able to bind and 13 phosphorylate NDPK-A S122A, but in contrast is unable to bind / phosphorylate 14 NDPK-A S122D, the AMPK phospho-mimetic mutant. Thus, we further confirm that the phospho-status of S122 determines whether CK2a binds (Figure 4A) and....

16 therefore, phosphorylates (Figure 4B) NDPK-A.

18 FIGURE 5: In vivo phosphorylation of ACC1 by the NDPK-A I AMPK ci complex is 19 independent of CK2 activity

S I S

21 Figure 5A: Extracted cytosol from the HepG2 cells treated as in the previous S..

22 experiments and probed for total ACC1 protein levels (lower blot), stripped and re-: 23 probed for phospho-ACC1 (upper blot). As expected, activation of AMPK resulted in 24 an increase in ACC phosphorylation (lanes 5 and 6). We observe that neither activation nor inhibition of CK2 had any effect on the ability of AMPK to phosphorylate 26 ACC1.

28 Figure 5B: The histogram quantitates the data from Figure 5A as the ratio of 29 phosphoserine to total protein band intensity as measured by densitornetry from four independent experiments (n=4). In viva phosphoserine content of NDPK-A and 31 NDPK-B after treatment reveals a novel relationship between NDPK-A and NDPK-B.

33 Figure 5C: NDPK-A immunoprecipitates from extracted cytosol of treated Hep G2 34 cells probed with a 1/1000 dilution of anti-phosphoserine antibody. We observe that the phosphoserine status of NDPK-A increased when AMPK was activated, but not I when CK2 was activated (compare lanes 5 and 6 to lane 7). Further, this 2 phosphoserine increment was insensitive to CK2 inhibition with TBB (5C, last lane).

4 Figure 5D: NDPK-B immunoprecipitates from extracted cytosol of treated Hep G2 cells probed with a 1/1000 dilution of anti-phosphoserine antibody. The 6 phosphoserine status of NDPKB dramatically increased upon activation of AMPK 7 despite our consistent finding that NDPK-B and AMPK do not interact (6) and this 8 phosphoserine increment was observed to be TBB sensitive, in contrast to that of 9 NDPK-A. These data suggest that under conditions where AMPK was active, CK2 is capable of phosphorylating NDPK-B.

12 FIGURE 6: Activation of AMPK al results in translocation of CK2a from the NDPK-A / 13 AMPK al complex to NDPK-B in the cell cytosol Figures 6A and 6B: Immunoprecipitations of NDPK-A, CK2a and NDPK-B from 16 treated Hep G2 cells. Immunoprecipitates were probed reciprocally as labelled S...

17 underneath the blots. We observe that CK2f3 is associated with both NDPK-A and 18 NDPK-B under all conditions tested (6A and 6B, middle blots), but CK2a is not 19 associated with NDPKA under conditions where AMPK is activated (6A top and bottom blots, lanes 5 and 6). Figure 6B shows that only under conditions where 21 AMPK is activated, does CK2a associate with NDPK-B (6B, top and bottom blots).

23 Figure 6C: Coimmunoprecipitation study using as input, extracted cytosol from either ** 24 NDPK-A wild-type littermate (Wt), NDPK-A null (A -I-), AMPK wild-type littermate (Ct), AMPK al null (al -I-) or AMPK a2 null (a2 -I-) mouse liver. Only in the total absence 26 of NDPK-A, does CK2a associated with NDPK-B under basal conditions. In all other 27 instances, CK2a is associated with NDPK-A.

29 FIGURE 7: Modelling the interaction 31 Working model of the interactions between NDPK-A (left), NDPK-B (right) and 32 CK2a/AMPK al. Panel A (left and right), shows the interaction in the absence of 33 AMPK phosphorylation at S122, where CK2 phosphorylates and inhibits NDPK-A at 34 S120. After AMPK stimulation, the interaction changes and AMPK phosphorylates NDPK-A at S122, resulting in a translocation of CK2a from NDPK-A to NDPK-B, 36 whilst simultaneously permitting substrate channelling of NDPK-A generated ATP to I AMPK al (6). Panel B incorporates our new findings into our recently described 2 working hypothesis and builds upon our previously reported NDPK-A / AMPK al 3 association (6) by including CK2a and the translocation mechanism.

Experimental 7 NDPK / AMPK cDNA constructs and purification 9 pcDNA3.1 constructs containing cDNA encoding for human AMPK al, AMPK a2, AMPK 131, AMPK 132, AMPK yl, AMPK y2 (gift of D. Caning), NDPK-A, NDPK-B (gift 11 of M. L. Lacombe), CK2a and CK213 (gifts of 0. Meek and D. Litchfield respectively) 12 were grown in bacterial culture using an ampicillin resistance selection and affinity 13 purified using a poly-His tag as previously reported (15).

Antibodies used in this study 17 Anti-NDPK-H1 (NM3O1) mouse monoclonal, anti-NDPK-H2 (L-15) goat 18 polyclonal, anti-CK2a mouse monoclonal and CK213 goat polyclonal antibodies 19 were supplied by Santa Cruz (USA). AMPK al, AMPK a2, AMPK 13 and AMPK y: *. S...

specific antibodies and other reagents were kindly provided by D. G. Hardie, S...

21 University of Dundee and D. Carling, Imperial College London. * .*.

22 *** : 23 Cytosolic protein extraction Fractionation of tissue/cells was carried out as previously reported (16). Briefly, 26 homogenisation of tissue or cell extracts was achieved by using mechanical lysis 27 into 3 volumes of ice-cold buffer I (10 mM Iris, 20 mM NaH2PO4, 1 mM EDTA, 28 pH 7.8 containing a fresh protease inhibitor cocktail). Osmolarity was restored by 29 adding 1/20 volume of 2.4 M KCI, 1/40 volume of 1.2 M NaCl and 1/5 volume of 1.25 M sucrose. Mixtures were centrifuged for 5 minutes at 500 g. The pellets 31 contain nuclei. Supernatants were diluted in buffer 11(30 mM imidazole, 120 mM 32 KCI, 30 mM NaH2PO4 and 250 mM sucrose, pH 6.8 with protease inhibitor as 33 above) to a protein concentration of 0.7 mg/mI. Samples were then spun down 34 at 7000 g for 15 minutes to remove mitochondria. The supernatant is then subjected to a 30000 g spin for 30 minutes. The resultant pellets contain the I remaining membrane proteins and the supernatants contain cytosolic extracted 2 proteins.

4 ImmunoDreciDitation 6 Prior to precipitation, dimethyl pimelimidate (DMP) linking was performed as 7 previously described (6). We used a 7 p1 slurry of either DMP-antibody linked 8 protein A or DMP-antibody linked protein G beads as required and added 9 cytosolic protein as detailed in each figure legend. Samples were then shake-incubated for 60 minutes at 4 C. The samples were pelleted in a desktop 11 centrifuge, followed by 3 x 1 ml washes with standard assay buffer containing 1 M 12 NaCI and 3 x 1 ml washes into standard assay buffer. Precipitation pellets were 13 re-suspended in 20 p1 of assay buffer and assayed/probed as described.

Western blot analysis 17 Protein concentrations were determined using the method of Bradford. SDS-S..

18 PAGE was performed using the Novex (Invitrogen) system on 4-12% Bis-Tris 19 polyacrylamide gels and MES buffer allowing for good separation of proteins between 17 -180 kDa. Western blotting analysis was performed as detailed in...., *S..

21 (17) and (6). Briefly, nitrocellulose membranes were blocked in TBS-Tween (0.5% 22 Tween-20) plus 5% milk powder for 30 minutes followed by 4 x 15mm washes in 23 TBS-Tween. Primary antibodies were incubated for 90 minutes followed by 24 further 4 x 15 mm washes. Species-specific secondary antibodies were used according to the manufacturer's instructions and incubated for 45 minutes, 26 followed by 4 x 15 mm washes. Enhanced Chemiluminescent (ECL) reagent 27 (Amersham) was used for visualisation.

29 Overlay bindinQ analysis 31 Purified bacterial recombinant NDPK, AMPK and CK2 subunits were immobilised 32 on nitrocellulose membrane by spotting. The membranes were then blocked and 33 washed as detailed above and a TBS-tween solution containing either 10 or 15 pg 34 of purified recombinant protein (as labelled in figure legends) overlaid on the membrane and incubated for 1 hour with shaking followed by 4 x 15mm washes 1 in TBS-tween (as above) and a 1/1000 dilution of isoform specific probing 2 antibody against the overlaid protein was incubated, as for the Western blotting 3 protocol, followed by detection using secondary antibody and ECL visualisation.

AMPK (SAMS) / CK2 (CK2-tide) activity assay 7 AMPK activity was assayed by measuring the incorporation of 32P from 500 nM 8 [32P]ATP into 1 mM of the synthetic peptide substrate USAMSfl 9 [HMRSAMSGLHLVKRR] exactly as detailed by Sullivan et al in 1994 (18). CK2 activity was measured in a similar manner using the specific synthetic peptide 11 substrate, "CK2-tide" [RRRADDSDDDDD]. In detail, samples were suspended to 12 a final volume of 25 p1 in standard assay buffer (50 mM HEPES, pH 7.0, 50 mM 13 NaF, 1 mM EGTA, 1 mM EDTA, 0.2 % Tween-20, 10 % glycerol). Assays were 14 incubated at 30 C for 10 minutes (for AMPK) and 30 minutes (for CK2) and terminated by spotting a 15 lii aliquot onto a 1 cm2 piece of P-81 16 phosphocellulose paper and washing 3 x 5 minutes in 1% phosphoric acid.

17 Samples were then air-dried and incorporation of labelled phosphate was 18 quantified using a Packard Instant Imager (19). Slight variations of this assay * 19 were used throughout the experiments, substituting the SAMS / CK2-tide peptides with recombinant NDPK, as labelled. * ***

22 Results.. : 24 The NDPK-A / AMPK ci complex interacts with protein kinase CK2 26 In relevant blot figures, each bead-linked precipitating antibody is shown at the 27 top of the blots and the relevant isoform-specific probing antibody at the bottom.

28 Figure 1, left panel shows SOS-PAGE and Western blotting of CK2, NDPK-A and 29 AMPK ci immunoprecipitates (IP) from an equal mixture (5 pg each) of four purified recombinant proteins: CK2a, NDPK-A, AMPK al and Kir4.2 (non- 31 specificity control) probed with specific antibodies against the remaining potential 32 complex components. Figure 1, right panel shows the same experiment repeated 33 using precipitates from a cell extract of human derived liver cell line, Hep G2 as 34 input. Again, Kir4.2 was used as a non-specificity control (+CT, last lane).

Previous work from this laboratory identified a novel interaction between NPDK-A 1 and AMPK al (6) but not involving NDPK-B. Under basal (i.e., unstressed non- 2 AMPK activated) conditions, in both a recombinant reconstituted system and a 3 native liver cell cytosol extract, the catalytic subunit of protein kinase CK2 (CK2a) 4 is present in the cytosolic NDPK-A / AMPK al complex that survives stringent washing with 1 M NaCI.

7 Which subunits of NDPK I AMPK interact with Drotein kinase CK2? 9 Figure 2, upper left panel shows a screen of 6 different peptides used as substrate for either AMPK (closed bars) or protein kinase, CK2 (open bars). The 11 first four peptides (NDPKpep 1, 2, 3 and 4) correspond to exposed regions of 12 NDPK-A (corresponding to amino acids: 49-62, 92-105, 111-125 and 139-152 13 respectively) as previously described ((7)) alongside relevant kinase specificity 14 controls; SAMS peptide (for AMPK) and CK2-tide (for CK2) (20). As previously reported (7), purified active rat liver AMPK al phosphorylates SAMS peptide, * 16 NDPKpep 3, at serine 122 and NDPKpep 4 at serine 144, whilst active * 17 recombinant CK2 is only able to phosphorylate CK2-tide and NDPKpep 3 at S120 18 (first reported by (5) and confirmed later in this study). Figure 2, lower left panel, 19 shows dot blots of each of the four NDPK peptides spotted onto nitrocellulose and overlaid with pure, recombinant CK2a, CK213, AMPK al and a2 catalytic subunits. *S*S

21 As observed previously, AMPK is able to phosphorylate both NDPKpep 3 and 22 NDPKpep 4 and the AMPK al, and this catalytic subunit selectively binds to the 23 NDPKpep 3 region of NDPK-A. We find that CK2a is also able to phosphorylate 24 and bind NDPKpep 3 but CK213 binds a region of NDPK-A not encompassed by the four NDPK peptides used in this study (compare right and left overlay panels 26 and data not shown). These data indicate that both AMPK and CK2a are capable 27 of binding and phosphorylating the NDPKpep 3 region of NDPK-A with previous 28 work identifying the target residues as S122 and S120 respectively. Figure 2, right 29 panel shows an overlay binding matrix of each of the NDPK and AMPK recombinant subunits spotted onto nitrocellulose and overlaid with either 31 recombinant CK2a or CK213. Unlike the peptide data above, we observe that 32 CK213 is capable of binding both isoforms of NDPK (A and B) and both isoforms of 33 AMPK 3 (1 and 2). CK2a was also observed to bind both NDPK isoforms (A and 34 B) and interestingly, only the AMPK y2 isoform (see later).

I InvestiQating the S122-deDendent interaction of CK2 with NDPK-A /AMPK ci 3 In Figure 3, left panel, we present a series of kinase assays (+ AMPK or + CK2 4 specific protein kinases as indicated) each using pure recombinant NDPK-A as target substrate. Also, in columns 2 and 4, we first phosphorylated NDPK-A for 10 6 minutes with the other kinase prior to conducting the kinase assays. In this 7 manner, we sought to determine whether NDPK-A prior phosphorylation by 8 AMPK or CK2 was independent or mutually exclusive for the other kinase. We 9 observe that prior phosphorylation of NDPK-A with CK2 (at S120) has no effect on the ability of AMPK to phosphorylate NDPK-A (second column, left panel) 11 whereas, prior phosphorylation of NDPK-A with AMPK (at S122) abolishes the 12 ability of CK2 to phosphorylate NDPK-A (S120). We previously reported a 13 mechanism, referred to as "substrate channelling", whereby NDPK-generated 14 AlP is specifically channelled to AMPK ci (irrespective of the surrounding ATP concentration) for use as phosphate donor in downstream AMPK phosphorylation 16 events, for example, phosphorylation of acetyl CoA carboxylase (ACC) (6). *.

17 Further, we also recently demonstrated substrate channelling between NDPK-A 18 and AMPK ci to be an S122-dependent process, such that AMPK: **.

19 phosphorylation of S122 is required for substrate channelling to proceed (7). To test this idea with respect to CK2, AMPK and NDPK, we manipulated the S..

21 nucleotide source such that the available AlP for AMPK could only arise from 22 GTP + ADP via NDPK-A. Figure 3, right panel shows this substrate channelling 23 assay using recombinant NDPK-A and CK2a added to purified, active rat liver 24 AMPK in vitro and assayed in the presence of GTP, ADP or GTP + ADP (from which NDPK-A makes ATP) as reported in (6). Alternatively, we applied ATP 26 GDP, to "steal" ATP by making GTP as previously described (6). Aliquots were 27 taken for Western analysis, immunoprecipitated using an NDPK-A antibody and 28 probed for the presence of AMPK ci and CK2a (Figure 3, upper and lower right 29 blot panel, respectively). Our data shows that whenever AMPK is activated either conventionally (lane 4 of blots) or under "substrate channelling" conditions (i.e., 31 when NDPK-A S122 is phosphorylated by AMPK ci), CK2 a is no longer in 32 complex with NDPK-A / AMPK ci (lane 3, lower blot) whereas under "substrate 33 steal" conditions (lane 6), CK2c is present (as it is under basal conditions and 34 with GTP, GDP or ADP alone). Next, we determined the details of the mechanism and then the in vivo fate of CK2a when AMPK is active.

2 Activation of AMPK al results in exclusion of CK2a from the cytosolic NDPK-A / 3 AMPK al complex in an intact cell system The Hep G2 cell line was treated for 20 minutes with a variety of agents as 6 follows: 4,5,6,7-tetrabromobenzotriazole (TBB), a potent selective inhibitor of 7 CK2; PKAi and PKC1, cell permeant pseudosubstrate inhibitors of PKA and PKC 8 respectively; spermidine (Sp), an activator of CK2 (21); metformin (Met) and 9 phenformin (Phen), both activators of AMPK or a combination of these as indicated at the top of Figure 4A. Cell cytosol was extracted and 11 immunoprecipitations (with either antibody against NDPK-A or CK2a) were 12 performed. Each precipitation was probed for the presence of either the fat 13 synthesis regulatory enzyme acetyl CoA carboxylase 1 (ACC1), AMPK al, CK2a, 14 CK2f3 or NDPK-A as labelled in the left panels. We find that treatment with three different kinase inhibitors (TBB, PKAi, PKCi) had no effect of association of CK2a * **.

16 within the complex whereas, when exposed to AMPK activators (metformin / * * : 17 phenformin), CK2a was lost from the complex, as shown using reciprocal S..

18 immunoprecipitation / Western blotting in Figure 4A (lanes 5 and 6). Activation of 19 CK2 with spermidine had no effect on the association but activation of AMPK always resulted in exclusion of CK2a from the complex (n=3). Interestingly, CK213 **.S 21 remained associated under all conditions tested and its association with the 22 NDPK-A / AMPK al complex was unaffected by the status of AMPK activity. ** 23 Thus, we find that activation of AMPK fractures the 2a, 213 CK2 heterodimer such 24 that only the $3 subunit remains with the complex. Figure 4B shows specific kinase assays performed on NDPK-A immunoprecipitations from the various cell-applied 26 treatments (AMPK activity, closed bars and CK2 activity, open bars). This 27 approach independently confirmed our Western blotting findings. Using 28 precipitated kinase activity as readout, CK2 activity was only absent when 29 metformin or phenformin had been added to the cells. Figure 4C demonstrates that CK2a phosphorylates S120 on NDPK-A and is unable to phosphorylate or 31 bind (see blot, right) NDPK-A when S120 is mutated to an alanine (S12OA). In a 32 related experiment, when we mutated NDPKA at the adjacent AMPK target site 33 S122 (A -dephospho-mimetic and D -phosphomimetic mutations), we observe 34 that CK2a is able to bind and phosphorylate NDPKA S122A, but in contrast is unable to bind I phosphorylate NDPK-A S122D, the AMPK phospho-mimetic 1 mutant. Thus, we further confirm that the phospho-status of S122 determines 2 whether CK2a binds (Figure 4A) and therefore, phosphorylates (Figure 4B) 3 NDPK-A.

In vivo phosphorylation of ACC1 by the NDPK-A I AMPK al complex is 6 independent of CK2 activity 8 Previously we reported that the cytosolic NDPK-A I AMPK al complex associated 9 with and regulated ACC1 in vivo ((6) and (7)). In Figure 5A, we probed extracted cytosol from the HepG2 cells (treated as in the previous experiments) for total 11 ACC1 protein levels (lower blot) and reprobed for phospho-ACC1 (upper blot).

12 We find that treatment of the Hep G2 cells with three kinase inhibitors (TBB, PKAi 13 and PKCi) and interestingly, the CK2 activator spermidine (column 7, Figure 4B 14 open bar), had no effect on phosphostatus of ACCI. As expected, activation of AMPK resulted in an increase in ACC phosphorylation (lanes 5 and 6). Thus, we * **.

16 observe that neither activation nor inhibition of CK2 (with spermidine and TBB * . : 17 respectively -see Figure 4B), had any effect on the ability of AMPK to 18 phosphorylate ACC1. The histogram in Figure 5B quantitates and confirms these * S..

19 data as the ratio of phosphoserine to total protein band intensity as measured by densitometry from four independent experiments. *5** * . S

22 In vivo phosphoserine content of NDPK-A and NDPK-B after treatment reveals a 23 novel relationship between NDPK-A and NDPK-B Next, we investigated the effects of the Hep G2 cell treatments shown above on 26 the in v/va phosphoserine status of NDPK-A using NDPK-B as a unoninteractingn 27 control using a phosphoserine specific antibody to probe precipitated NDPK-A 28 (Figure 5C) or NDPK-B (Figure 50) as indicated. Figure 5C shows that the 29 phosphoserine content of NDPK-A increased when AMPK was activated, but not when CK2 was activated (compare lanes 5 and 6 to lane 7). Further, this 31 phosphoserine increment wasinsensitive to CK2 inhibition with TBB (5C, last 32 lane). Originally intended as a control, we used an NDPK-B immunoprecipitate 33 probed with the phosphoserine antibody and as shown in Figure 5D, we observed 34 an interesting finding. The phosphoserine status of NDPK-B dramatically increased upon activation of AMPK despite our consistent finding that NDPK-B I and AMPK do not interact (6) and (7). Surprisingly, this phosphoserine increment 2 was observed to be TBB sensitive (Figure 5D), in contrast to that of NDPK-A 3 (Figure 5C, last lane). This unexpected CK2 sensitivity suggested that under 4 conditions where AMPK was active, CK2 was now phosphorylating NDPK-B.

Since NDPK-B is capable of binding CK2a and CK213 (Figure 2, right hand 6 overlays), we hypothesised that CK2a might be translocating from the NDPK-A / 7 AMPK al complex to NDPK-B upon activation of AMPK within the NDPK-A I 8 AMPK al / CK2 complex.

Activation of AMPK al results in translocation of CK2a from the NDPK-A I AMPK 11 al comDlex to NDPK-B in the cell cvtosol 13 Figures 6A and 6B show immunoprecipitations of NDPK-A, CK2a and NDPK-B 14 from Hep G2 cells manipulated for kinase activities as described previously.

These immunoprecipitates were probed as labelled underneath the blots.

16 Whereas we observe that CK213 is associated with both NDPK-A and NDPK-B 17 precipitates under all conditions tested (6A and 6B, middle blots), we confirm our 18 previous observation that CK2a is not associated with NDPK-A under conditions 19 where AMPK is activated (6A top and bottom blots, lanes 5 and 6). Figure 6B shows that only under conditions where AMPK is activated, does CK2a associate S...

21 with NDPK-B (6B, top and bottom blots). This AMPK-dependent association of 22 CK2a with NDPK-B explains the CK2-dependent phosphoserine increment upon 23 AMPK activation (Figure 5D). Figure 6C shows co-immunoprecipitation data 24 using as input cytosol from either NDPK-A wild type litter mate (Wt), NDPK-A null (A -I-), AMPK wild-type littermate (Ct), AMPK al null (al -I-) or AMPK a2 null (a2 26 -I-) mouse liver as used previously in (7). These data demonstrate that only in the 27 null state for NDPK-A, is CK2a associated with NDPK-B under baseline 28 conditions. In all other instances, CK2a is associated with NDPK-A, as might be 29 expected since activated AMPK is required for release of CK2a from NDPK-A.

Use of these NDPK and AMPK null tissue samples provides further proof that this 31 interaction is physiologically relevant and validates the specificity of our reported 32 interactions in vitro and in cell lines. The data show that CK2a associates with 33 NDPK-B in the NDPK-A null mouse liver cytosol and future work will determine 34 the relationship to the phenotypes of the NDPK-A mouse consequent upon the basal association of CK2a with NDPK-B (22).

2 Modeling the interaction 4 Figure 7 presents our working model of the interactions between NDPK-A (left), NDPK-B (right) and CK2a / AMPK cii. Panel A (left and right), shows the basal 6 interaction, i.e., in the absence of AMPK phosphorylation at S122, where CK2 7 phosphorylates and inhibits NDPK-A at S120 ((5) and confirmed by us in Figure 8 4C). Then, the lower left part of Panel A shows how that interaction changes 9 when AMPK is activated and phosphorylates NDPK-A at S122, resulting in a translocation of CK2cx from NDPK-A to NDPK-B, whilst simultaneously permitting 11 substrate channelling of NDPK-A generated ATP to AMPK cii (6), which in turn 12 controls AMPK in the absence of AMP, provided S122 remains phosphorylated.

13 Panel B incorporates our new findings into our recently described working 14 hypothesis and builds upon our previously reported NDPK-A I AMPK cii association (6) by including CK2ci and the translocation mechanism. a a * S.. *

17 Discussion *.* 19 The principal finding of the current study is that the constitutively active pleiotropic. * a kinase, CK2 forms part of the cytosolic NDPK-A / AMPK cii complex under a..

21 resting, but not under AMPK-activated conditions. Once the NDPK-A I AMPK cii 22 complex starts to substrate channel, i.e., under conditions where NDPK-A S122 is 23 phosphorylated by AMPK cii (as shown in (7)), CK2ci is released from the 24 complex and translocates to the closely related, NDPK-B isoform. The potential cellular implications of this translocation mechanism linking the two most widely 26 expressed NDPK family members are discussed below.

28 We initially observed the association of the NDPK-A / AMPK ci complex with 29 protein kinase CK2 when using it as a control kinase in our study of the mechanism of interaction between NDPK-A and AMPK cii (7). Biondi et al, first 31 showed that CK2 phosphorylated NDPK at serine 120 in vivo (5). Our previous 32 work demonstrated that in viva, AMPK cii phosphorylated NDPK-A at the 33 adjacent S122 site and, that phosphorylation was both necessary and sufficient to 34 initiate the substrate channelling mechanism (7). We have proved that substrate channelling permits AMPK to remain active even when ATP is very low because 1 the AlP is synthesised locally within the complex via NDPK-A. Initially, we 2 expressed and purified recombinant CK2a, NDPK-A, AMPK al and Kir4.2, 3 reconstituting the system in vitro.

Parallel to this, we immunoprecipitated using specific antibodies to each 6 component from a human derived liver cell line, Hep G2 and observed that the 7 NDPK-A I AMPK al complex interacted specifically with protein kinase CK2a. We 8 then performed a binding matrix experiment, as previously employed (6, 15) and 9 determined that CK2a is capable of binding both to NDPK-A and NDPK-B in the S120 region. We found that CK213 is also capable of binding NDPK-A and NDPK- 11 B in another, as yet unidentified, region and that the subunit of AMPK binds 12 CK2p. Interestingly, we found that only the y2 subunit isoform of AMPK could 13 bind CK2a. It should be noted from our recent study, that the composition of 14 NDPK-A / AMPK al complexes associated with the cystic fibrosis transmembrane conductance regulator, CFTR, was specifically the AMPK al, AMPK y2 isozyme 16 (15). Further work will be needed to investigate the significance of this finding and 17 the role of AMPK y2 in these systems, but we predict that CK213 might be bound 18 to CFTR, since the NDPK-A I AMPK al complex also binds that protein (15). : .*. *0*

In order to investigate kinase hierarchy, we used each kinase to pre-21 phosphorylate NDPK-A and then investigated the ability of the other to 22 subsequently phosphorylate its specific target residue in the presence of 32P- 23 ATP (Figure 3, left panel). We found that prior phosphorylation of NDPK-A by 24 CK2 made no difference to the ability of AMPK alto phosphorylate NDPK-A.

However, prior phosphorylation of NDPK-A with AMPK abolished CK2a- 26 dependent phosphorylation of NDPK-A.

28 In order to test this mechanism in relation to our previously identified substrate 29 channelling/substrate steal mechanism (6), we initiated both channel' and steal' conditions in a reconstituted recombinant system and probed for changes in 31 AMPK al and CK2a coprecipitation with NDPK-A (Figure 3, left panel). We found 32 that under conditions where substrate channelling occurred, CK2a was not 33 present in the NDPK-A I AMPK al complex and conversely, during substrate 34 steal (NDPK-A sequestration of ATP with GDP to generate GTP, thus shuthng off AMPK), CK2a remains a part of the complex (as it is in baseline conditions).

2 We next tested whether the mechanism of exclusion of CK2a from the cytosolic 3 NDPK-A / AMPK al complex was also present in a Hep G2 cell system, under a 4 variety of protein kinase manipulations (Figure 4). We observed that treatment with specific CK2, PKA and PKC inhibitors had no effect of association of CK2a 6 within the complex whereas when the cells were treated with AMPK activators 7 (metformin I phenformin), CK2a was expelled. Interestingly, CK213 was found to 8 be present under all conditions tested and its association with NDPK-A / AMPK 9 al complex was unaffected by AMPK activity. This finding is consistent with the known CK2 a and 13 "interaction dynamic" and unexplained differences in CK2a 11 function alone verses differential roles of the intact heterodimer (23).

13 We suggest that CK213 remains as an anchor for CK2a within the NDPK-A / 14 AMPK al complex, to permit re-association when AMPK is no longer active.

Further investigation confirmed a report from Biondi et al, that CK2a * r* * S..

16 phosphorylates S120 on NDPK-A (5). We find that CK2a is unable to * . : 17 phosphorylate or bind NDPKA when S120 is mutated to an alanine (S12OA) *00 18 (Figure 4C). Further, when we mutated NDPK-A at the AMPK target site (S122) : 19 introducing either dephosphomimetic (A) or a phospho-mimetic (D) mutations, we observed that CK2a was only able to bind and phosphorylate NDPK-A S122A. In S...

21 contrast, CK2a was unable to bind the NDPK-A S122D, the AMPK phospho- 22 mimetic mutant. Thus, we confirm that the phospho-status of S122 determines 23 whether CK2a binds (Figure 4A) and therefore, phosphorylates (Figure 4B) 24 NDPK-A and conclude that negative charge at S122 prevents CK2a from interacting with NDPK-A. This type of interaction mechanism is not without 26 precedent, as Daval et al recently revealed that prior AMPK phosphorylation of 27 hormone sensitive lipase (HSL) at residue S565 precluded PKA phosphorylation I 28 interaction at S563 (24).

Previously, we reported that the cytosolic NDPK-A / AMPK al complex 31 associated with and regulated a key fat controller' enzyme ACC1 in vivo by 32 substrate channelling as an AMP-independent process ((6) and (7)). Therefore, 33 we tested whether the in vivo phosphorylation status of ACC1 by the NDPK-A I 34 AMPK ci complex involved CK2a, given our previous observations of a lack of CK2a in the complex when substrate channelling occurs. Using the extracted 1 cytosol from the treated HepG2 cells, we probed for total and phospho-ACC 1 2 protein levels. We found that the neither the inhibitors; TBB, PKAi and PKCi nor 3 the CK2 activator, spermidine had any effect on phosphostatus of ACC1. As 4 expected, activation of AMPK resulted in an increase in ACC phosphorylation.

Thus, neither activation nor inhibition of CK2 (with spermidine and TBB 6 respectively) had any direct effect on ACC1 phosphostatus or on the ability of 7 AMPK to phosphorylate ACC1.

9 With this negative result for the relationship between CK2 and fat metabolism at hand, we checked the in vivo phosphoserine status of NDPK-A using the closely 11 related but non-interacting, NDPK-B as a control. Strikingly, the phosphoserine 12 status of NDPKA was CK2-independent whereas the phosphostatus of NDPK-B 13 was CK2-dependent but, crucially, only upon AMPK activation. This was 14 surprising because our previous work demonstrated repeatedly that NDPK-B and AMPK ci do not interact (6, 15) but from these conflicting data, we hypothesised 16 that CK2a might translocate from the NDPK-A I AMPK ci complex to NDPK-B., 17 upon AMPK activation. Figure 6A and 6B reveal that CK2a only associates with 18 NDPK-B when AMPK is activated by metformin and phenformin i.e., CK2u 19 translocates from NDPK-A to NDPK-B when substrate channelling occurs. This hypothesis was substantiated in Figure 6C by using NDPK-A, AMPK ci and 21 AMPK a2 null mouse liver cytosol extracts as previously reported (22, 25). These 22 data showed that CK2a associated with NDPK-B in baseline conditions only when 23 NDPK-A was genomically deleted, otherwise, the interactions were as observed 24 in the Hep G2 cells.

26 Why might NDPK-B need CK2a under substrate channelling conditions? 28 Our new finding is that under cell stress conditions (i.e., activation of AMPK), our 29 previously identified substrate channelling" mechanism results in disassociation of CK2a from the cytosolic NDPK-A I AMPK ci complex. Our idea is that when 31 AlP concentrations fall, substrate channelling will redress the shortage of ATP 32 (6) by permitting NDPK-A to activate AMPK cl-dependent phosphorylation of 33 ACC1, thus switching off energy consuming fat synthesis (6).

1 Now, our latest data suggest that the concomitantly released CK2a translocates 2 to NDPK-B without affecting the fat synthetic pathways (Figure 6A and 6B). We 3 hypothesise that the functional significance of this translocation might be to aid 4 the NDPK-A / AMPK al complex in conserving cellular ATP by an independent route as follows. Since CK2a translocation to NDPK-B is inhibitory to the ability of 6 NDPK-B to engage in phosphotransfer and NDPK-B normally synthesizes GTP 7 for heterotrimeric G proteins, the net result could be to modulate I inhibit 8 transmembrane signalling and conserve ATP by inhibiting receptor signalling and 9 NDPK-B utilisation of ATP to make GTP. Thus, induction of regulation of NDPK-B by CK2a under these conditions could have a significant impact on multiple 11 heterotrimeric G-protein pathways as described by others in (26, 27).

13 NDPK-A, NDPK-B and CK2a -possible interactions NDPKs were the first described metastatic suppressors in human tumors such as ** 16 breast carcinoma (28), melanoma (30) and oral squamous cell carcinomas (31).

17 Paradoxically, NDPK-A is also reported to act as a metastasis promotor in other 18 human tumors, including neuroblastoma (29). The reasons are obscure but the 19 presence of an S12OG mutation in NDPK-A has been detected in 14 -30% of *...

patients with advanced stages of neuroblastoma (29). NDPK-A S12OG is reported 21 to reduce cell adhesion and increase cell migration and was observed to be more *.S.

22 effective than wild-type NDPK-A at promoting neuroblastoma metastasis (29). An.. : 23 S12OA mutation, which our current data shows results in CK2a exclusion from the 24 NDPK-A / AMPK al complex and translocation to NDPK-B, induced tumour cell motility inhibition in a human breast carcinoma cell line (32). S12OA is also 26 reported to exhibit reduced serine autophosphorylation levels in NDPKs (33).

27 NDPK-B induces activation of the oncogene, c-myc, by an unknown mechanism 28 and leads to metastasis (35). Further work will be needed to see whether CK2 29 translocation mediates this interaction. Also, CK2 inhibition has been shown to reverse the cancer phenotype (36), again by an unknown mechanism, presumed 31 to involve its reported phosphorylation of p53 (37, 38).

33 Thus, the site of CK2 phosphorylation on NDPK at S120 shows important biology.

34 The interactions presented reveal a molecular mechanism connecting these apparently disparate metastatic effectors. Thus, overall, the molecular I mechanism of interaction presented in this study might illuminate a hitherto 2 unknown link between NDPK-A and CK2a in common cancers, whereby loss of 3 the CK2ci target on NDPK-A (S120) results in firstly, a decrease in tumour cell 4 motility and secondly, a potential increase in CK2a interaction with NDPK-B, a known oncogenic protein.

7 In Conclusion

9 AMPK al phosphorylates NDPK-A at S122 making CK2a dissociate from the NDPK-A I AMPK al cytosolic complex and translocate to NDPK-B. Thus, we 11 postulate the existence of a multi-functional cytosolic protein kinase complex, at 12 the crossroads of cancer, energy and fat metabolism.

14 Accordingly, we anticipate that pharmaceutically acceptable agents which have the property of disrupting the binding of CK2ci to NDPK-B, or promoting the binding of **I 16 CK2a to NDPK-A could be useful in the prevention and treatment of cell * 17 proliferative conditions, such as cancers, which result from an increased level of 18 binding of CK2a to NDPK-B or a reduced level of binding of CK2a to NDPK-A, * 19 whether due to loss of the CK2a target on NDPK-A or for any other reason. * *** ****

21 Similarly, pharmaceutically acceptable agents which have the property of disrupting * * ** * )1* 22 the binding of CK2a to NDPK-A, or promoting the binding of CK2a to NDPK-B..

23 could be useful in the prevention and treatment of cell proliferative conditions, 24 such as cancers which result from an increased level of binding of CK2a to NDPK-A, or a reduced level of binding of CK2a to NDPK-B, whether due to loss 26 of the CK2cz target on NDPK-B or for any other reason.

28 Screening Assay We also propose a high throughput, non-radioactive assay to screen test agents 31 which may have the property of modulating the binding of CK2a to an isoform of 32 NDPK (e.g. NDPK-B) or a specific mutant thereof, e.g. S12OA or S12OG mutants 33 of NDPK-A. Suitable test agents include peptides which comprise the binding site 34 of an isoform of NDPK (e.g. NDPK-B), for CK2a, or variants thereof. Such test I agents could therefore be expected to compete with isoforms of NDPK to bind 2 free CK2o.

4 In order to screen for test agent which may have the property of disrupting the binding of CK2a. to NDPK-B, a plurality of test agents are introduced to either 0) 6 samples of Hep G2 cells or (ii) lysates of Hep G2 cells, in wells of a multi-well 7 plate and incubated. After a period of time, complexes of CK2a and NDPK-B are 8 immunoprecipitated using an antibody to NDPK-B immobilised on beads. NDPK- 9 B can then be adsorbed onto the surface of a multi-well plate or may remain bound to the beads during the next stage of the assay.

12 The presence of CK2a can then be assayed either by using an ELISA which 13 employs either an anti-phosphoserine antibody that binds phosphorylated serine 14 residues on NDPK-B or an anti-CK2a antibody. The anti-phosphoserine antibody or anti-CK2o antibody may be enzyme labelled, or a further enzyme labelled * 16 antibody which binds to the anti-phosphoserine or anti-CK2a antibody may be.

17 added. Substrates for a reaction which is catalysed by the label enzyme are then 18 added. The label enzyme catalyses a detectable and quantifiable reaction, such * 19 as a reaction which generates a coloured substance, thereby enabling quantification of the amount of CK2a which bound to NDPK-B. ***

22 Where a test agent has disrupted the binding of CK2a to NDPK-B, the colour 23 generated by the label enzyme will be lower, enabling test agents which have the 24 property of disrupting the binding of CK2a to NDPK-B to be determined and selected as lead compounds for further development and testing. Modifications 26 to and variations of this ELISA will be apparent to one skilled in the art. The same 27 test can be used to assess whether the test agent has the property of promoting 28 the binding of CK2a to NDPK-B. A corresponding test can be used to assess 29 whether the test agent has the property of modulating the binding of CK2a to NDPK-A. A cell line in which an isoform of NDPK (such as NDPK-A) has been 31 mutated to a specific mutant of the isoform may be used to assess whether a test 32 agent has the property of modulating the binding of CK2o to a specific mutant of 33 the isoform of NDPK.

I Further modifications and variations may be made within the scope of the invention 2 herein disclosed.

4 References 6 1. Heidbuchel, H., Callewaert, G., Vereecke, J., and Carmeliet, E. (1992) 7 Membrane-bound nucleoside diphosphate kinase activity in atrial cells of frog, 8 guinea pig, and human. Circ Res 71, 808-820 9 2. Engel, M., Veron, M., Theisinger, B., Lacombe, M. L., Seib, T., Dooley, S., and Welter, C. (1995) A novel serine/threonirie-specific protein 11 phosphotransferase activity of Nm23/nucleoside-diphosphate kinase. Eur J 12 Biochem 234, 200-207 13 3. Roymans, D., Willems, R., Van Blockstaele, D. R., and Slegers, H. (2002) 14 Nucleoside diphosphate kinase (NDPK/NM23) and the waltz with multiple partners: possible consequences in tumor metastasis. Clin Exp Metastasis 19, 16 465-476 S.....

17 4. Bosnar, M. H., De Gunzburg, J., Bago, R., Brecevic, L., Weber, I., and 18 Pavelic, J. (2004) Subcellular localization of A and B Nm23/NDPK subunits. S..

19 Exp Cell Res 298, 275-284 5. Biondi, R. M., Engel, M., Sauane, M., Welter, C., Issinger, 0. G., Jimenez de 21 Asua, L., and Passeron, S. (1996) Inhibition of nucleoside diphosphate kinase 22 activity by in vitro phosphorylation by protein kinase CK2. Differential 23 phosphorylation of NDP kinases in HeLa cells in culture. FEBS Left 399, 183-24 187 6. Crawford, R. M., Treharne, K. J., Best, 0. G., Muimo, R., Riemen, C. E., and 26 Mehta, A. (2005) A novel physical and functional association between 27 nucleoside diphosphate kinase A and AMP-activated protein kinase alphal in 28 liver and lung. Biochem J 392, 201 -209 29 7. Crawford, R. M., Treharne, K. J., Arnaud-Dabernat, S., Daniel, J. Y., Foretz, M., Viollet, B., and Mehta, A. (2006) Understanding the Molecular Basis of the 31 Interaction between NDPK-A and AMPK {alpha}1. Mol Cell Biol 26, 5921-32 5931 33 8. Stapleton, D., Mitcheihill, K. I., Gao, G., Widmer, J., Michell, B. J., Teh, T., 34 House, C. M., Fernandez, C. S., Cox, T., Witters, L. A., and Kemp, B. E. (1996) Mammalian AMP-activated protein kinase subfamily. J Biol Chem 271, 36 611-614 1 9. Nielsen, J. N., Mustard, K. J., Graham, D. A., Vu, H., MacDonald, C. S., 2 Pilegaard, H., Goodyear, L. J., Hardie, D. G., Richter, E. A., and 3 Wojtaszewski, J. F. (2003) 5'-AMP-activated protein kinase activity and 4 subunit expression in exercise-trained human skeletal muscle. J AppI Physiol 94, 631-641 6 10. Hardie, D. G., and Carling, D. (1997) The AMP-activated protein kinase-fuel 7 gauge of the mammalian cell? Eur J Biochem 246, 259-273 8 11. Hardie, D. G., and Hawley, S. A. (2001) AMP-activated protein kinase: the 9 energy charge hypothesis revisited. Bioessays 23, 1112-1119 12. Pinna, L. A. (2003) The raison d'etre of constitutively active protein kinases: 11 the lesson of CK2. Acc Chem Res 36, 378-384 12 13. Meggio, F., and Pinna, L. A. (2003) One-thousand-and-one substrates of 13 protein kinase CK2? Faseb J 17, 349-368 14 14. Meggio, F., Mann, 0., and Pinna, L. A. (1994) Substrate specificity of protein kinase CK2. Cell Mol Biol Res 40, 401-409 16 15. Crawford, R. M., Trehame, K. J., Best, 0. G., Riemen, C. E., Muimo, R., 17 Gruenert, D. C., Arnaud-Dabernat, S., Daniel, J. V., and Mehta, A. (2006) 18 NDPK-A (but not NDPK-B) and AMPK alphal (but not AMPK alpha2) bind the * S. 19 cystic fibrosis transmembrane conductance regulator in epithelial cell membranes. Cell Signal Feb 6 (epub ahead of print). Ac. No: 16466905 21 16. Ranki, H. J., Crawford, R. M., Budas, G. R., and Jovanovic, A. (2002) Ageing 22 is associated with a decrease in the number of sarcolemmal ATP-sensitive K+ 23 channels in a gender-dependent manner. Mech Ageing Dev 123, 695-705 24 17. Crawford, R. M., Ranki, H. J., Boning, C. H., Budas, G. R., and Jovanovic, A. (2002) Creatine kinase is physically associated with the cardiac ATP-sensitive 26 K+ channel in vivo. Faseb J 16, 102-1 04 27 18. Sullivan, J. E., Carey, F., Caning, D., and Ben, R. K. (1994) Characterisation 28 of 5'-AMP-activated protein kinase in human liver using specific peptide 29 substrates and the effects of 5-AMP analogues on enzyme activity. Biochem Biophys Res Commun 200, 1551-1 556 31 19. Muimo, R., Hornickova, Z., Riemen, C. E., Gerke, V., Matthews, H., and 32 Mehta, A. (2000) Histidine phosphorylation of annexin I in airway epithelia. J 33 Biol Chem 275, 36632-36636 34 20. Mann, 0., Meggio, F., and Pinria, L. A. (1994) Design and synthesis of two new peptide substrates for the specific and sensitive monitoring of casein 36 kinases-1 and -2. Biochem Biophys Res Commun 198, 898-905 1 21. Leroy, D., Filhol, 0., Delcros, J. G., Pares, S., Chambaz, E. M., and Cochet, 2 C. (1997) Chemical features of the protein kinase CK2 polyamine binding site.

3 Biochemistry 36, 1242-1250 4 22. Arnaud-Dabernat, S., Bourbon, P. M., Dierich, A., Le Meur, M., and Daniel, J. Y. (2003) Knockout mice as model systems for studying nm23/NDP kinase 6 gene functions. Application to the nm23-M1 gene. J Bioenerg Biomembr 35, 7 19-30 8 23. Litchfield, D. W., Bosc, D. G., Canton, D. A., Saulnier, R. B., Vilk, G., and 9 Zhang, C. (2001) Functional specialization of CK2 isoforms and characterization of isoform-specific binding partners. Mol Cell Biochem 227, 11 21-29 12 24. Daval, M., Foufelle, F., and Ferre, P. (2006) Functions of AMP-activated 13 protein kinase in adipose tissue. J Physiol M 25. Viollet, B., Andreelli, F., Jorgensen, S. B., Perrin, C., Geloen, A., Flamez, D., Mu, J., Lenzner, C., Baud, 0., Bennoun, M., Gomas, E., Nicolas, G., 16 Wojtaszewski, J. F., Kahn, A., Carling, 0., Schuit, F. C., Bimbaum, M. J., 17 Richter, E. A., Burcelin, R., and Vaulont, S. (2003) The AMP-activated protein.

18 kinase alpha2 catalytic subunit controls whole-body insulin sensitivity. J Clin 19 lnvestlll,91-98: :.

26. Lukov, G. L., Baker, C. M., Ludtke, P. J., Hu, 1., Carter, M. D., Hackett, R. A., 21 Thulin, C. D., and WUlardson, B. M. (2006) Mechanism of assembly of G S* 22 protein beta gamma subunits by CK2-phosphorylated phosducin-like protein 23 and the cytosolic chaperonin complex. J Biol Chem.. : 24 27. Lukov, G. L., Myung, C. S., Mclntire, W. E., Shao, J., Zimmerman, S. S., Garrison, J. C., and Willardson, B. M. (2004) Role of the isoprenyl pocket of 26 the G protein beta gamma subunit complex in the binding of phosducin and 27 phosducin-like protein. Biochemistry 43, 5651-5660 28 28. Salerno, M., Palmieri, D., Bouadis, A., Halverson, D., and Steeg, P. S. (2005) 29 Nm23-H1 metastasis suppressor expression level influences the binding properties, stability, and function of the kinase suppressor of Rasi (KSR1) Erk 31 scaffold in breast carcinoma cells. Mol Cell Biol 25, 1379-1388 32 29. Almgren, M. A., Hennksson, K. C., Fujimoto, J., and Chang, C. L. (2004) 33 Nucleoside diphosphate kinase Nnm23-H1 promotes metastasis of NB69- 34 derived human neuroblastoma. Mol Cancer Res 2, 387-394 30. Cervoni, L., Egistelli, L., Eufemi, M., d'Abusco, A. S., Altieri, F., Lascu, I., 36 Turano, C., and Giartosio, A. (2006) DNA sequences acting as binding sites for NM23/NDPK proteins in melanoma M14 cells. J Cell Biochem 98, 421-2 428 3 31. Yang, Z., Wen, V., and Pu, P. (2003) [The regulation of nm23-H1/NDPK-A in 4 different processes of regional lymph node metastases of oral squamous cell carcinomas]. Hua Xi Kou Qiang Vi Xue Za Zhi 21, 263-266 6 32. Otero, A. S., Doyle, M. B., Hartsough, M. T., and Steeg, P. S. (1999) Wildtype 7 NM23-H1, but not its S120 mutants, suppresses desensitization of muscarinic 8 potassium current. Biochim Biophys Acta 1449, 157-1 68 9 33. Freije, J. M., Blay, P., MacDonald, N. J., Manrow, R. E., and Steeg, P. S. (1997) Site-directed mutation of Nm23-H1. Mutations lacking motility 11 suppressive capacity upon transfection are deficient in histidine-dependent 12 protein phosphotransferase pathways in vitro. J Biol Chem 272, 5525-5532 13 34. MacDonald, N. J., De Ia Rosa, A., Benedict, M. A., Freije, J. M., Krutsch, H., 14 and Steeg, P. S. (1993) A serine phosphorylation of Nm23, and not its nucleoside diphosphate kinase activity, correlates with suppression of tumor a.

16 metastatic potential. J Biol Chem 268, 25780-25789 17 35. Postel, E. H., Berberich, S. J., Rooney, J. W., and Kaetzel, 0. M. (2000) 18 Human NM23/nucleoside diphosphate kinase regulates gene expression 19 through DNA binding to nuclease-hypersensitive transcriptional elements. J ioenerg Biomembr 32, 277-284 21 36. Ahmad, K. A., Wang, G., Slaton, J., Unger, G., and Ahmed, K. (2005) 22 Targeting CK2 for cancer therapy. Anticancer Drugs 16, 1037-1 043 23 37. Isten, M. E., Weber, J. E., and Litchfield, D. W. (2005) CK2 interacting 24 proteins: emerging paradigms for CK2 regulation? Mol Cell Biochem 274, 115-124 26 38. McKendrick, L., Milne, D., and Meek, D. (1999) Protein kinase CK2- 27 dependent regulation of p53 function: evidence that the phosphorylation 28 status of the serine 386 (CK2) site of p53 is constitutive and stable. Mol Cell 29 Biochem 191,187-199 39. Landesman-Bollag, E., Song, D. H., Romieu-Mourez, R., Sussman, D. J., 31 Cardiff, R. 0., Sonenshein, G. E., and Seldin, D. C. (2001) Protein kinase 32 CK2: signaling and tumorigenesis in the mammary gland. Mol Cell Biochem 33 227, 153-165 34 40. Landesman-Bollag, E., Romieu-Mourez, R., Song, 0. H., Sonenshein, G. E., Cardiff, R. D., and Seldin, 0. C. (2001) Protein kinase CK2 in mammary gland 36 tumorigenesis. Oncogene 20, 3247-3257 1 41. Romieu-Mourez, R., Landesman-Bollag, E., Seldin, D. C., and Sonenshein, 2 G. E. (2002) Protein kinase CK2 promotes aberrant activation of nuclear 3 factorkappaB, transformed phenotype, and survival of breast cancer cells.

4 Cancer Res 62, 6770-6778 42 Evans, J. M. M., Donnelly, L. A., Emslie-Smith, A. M., Alessi, D. R., and 6 Morris, A. 0. (2005) Metformin and reduced risk of cancer in diabetic patients.

7 BMJ 330, 1304-1305 8 43. Hulin, B., McCarthy, P. A., Gibbs, E. M. (1996) The glitazone family of 9 antidiabetic agents, Current Pharmaceutical Design, 2, 85-10 1 44. Lee, K-S., Park, J-H., Lee, S., Lim, H-J., Jang, Y., Park, H-Y (2006) 11 Troglitazone inhibits endothelial cell proliferation through suppression of 12 casein kinase 2 activity, Biotech. Biophys. Research Comm., 346, 83-88 13 45. Valentijn, L. J., Koster, J., Versteeg, R., (2006) Read-through transcript from 14 NM23-H1 into the neighboring NM23-H2 gene encodes a novel protein, Genomics, 87(4):483-9.

16 46. Braun, S., Mauch, C., Boukamp, P., Werner, S, (2006) Novel roles of NM23 17 proteins in skin homeostasis, repair and disease, Oncogene, 1-11.

S *SS * *. * S * *S** **** * S 5.5 * S * *5 S

Claims

I Claims 3 1. A method of identifying an agent which has the property

of modulating the 4 binding of CK2a to an isoform of NDPK, or a specific mutant thereof, the method comprising selecting a test agent and assessing whether the test 6 agent has the property of modulating the binding of CK2cx to the isoform of 7 NDPK, or specific mutant thereof.

9
2. A method according to claim 1, wherein the modulation is disruption, the method is a method of identifying an agent which has the property of 11 disrupting the binding of CK2a to an isoform of NDPK, or a specific mutant 12 thereof, and the method comprising selecting a test agent and assessing 13 whether the test agent has the property of disrupting the binding of CK2a to 14 the isoform of NDPK, or specific mutant thereof.

16
3. A method according to claim 1 or claim 2, wherein the isoform of NDPK is 17 NDPK-AorNDPK-B. * **

19
4. A method according to any one preceding claim, wherein what is assessed is * the binding of CK2ci, which is not bound to CK2f3, to the isoform of NDPK, 21 which has CK2 bound thereto. * S*

23
5. A method according to any one preceding claim, in which the isoform of 24 NDPK is NDPK-B, the method is a method of identifying an agent which has the property of disrupting the binding of CK2a to NDPK-B, or specific mutant 26 of NDPK-B, and the method comprises assessing whether the test agent has 27 the property of disrupting the binding of CK2ci to NDPK-B, or the specific 28 mutant of NDPK-B.

6. A method according to any one preceding claim, in which the isoform of 31 NDPK is NDPK-A, the method is a method of identifying an agent which has 32 the property of promoting the binding of CK2o to NDPK-A, or a specific 33 mutant of NDPK-A, and the method comprises assessing whether the test 34 agent has the property of modulating the binding of CK2a to NDPK-A, or the specific mutant of NDPK-A.

1 7. A method according to claim 6, wherein the isoform of NDPK-A, or a specific 2 mutant thereof, is an S12OG or S12OA mutant of NDPK-A, the method is a 3 method of identifying an agent which has the property of promoting the 4 binding of CK2ci to the said mutant of NDPK-A, and the method comprises assessing whether the test agent has the property of modulating the binding 6 of CK2a to the said mutant of NDPK-A.

8 8. A method according to any one preceding claim, which is a method of 9 identifying an agent which may be used to modulate the binding of CK2a to the isoform of NDPK, or specific mutant thereof, in a subject in need thereof.

12 9. A method according to any one preceding claim, in which a plurality of test 13 agents are screened simultaneously and/or sequentially to assess whether 14 they have the property of modulating the binding of CK2a to the isoform of NDPK, or specific mutant thereof.

17 10. A method according to any one preceding claim, wherein assessing whether a 18 test agent has the property of modulating the binding of CK2a to the isoform 19 of NDPK, or specific mutant thereof, comprises exposing the test agent to (i) * CK2a, or a variant, fragment and/or fusion thereof, and/or (ii) the isoform of 21 NDPK, or specific mutant thereof, or a variant, fragment and/or fusion of the *...

22 isoform of NDPK, or specific mutant thereof, and determining whether the test 23 agent has modulated the binding of CK2a to the isoform of NDPK, or specific 24 mutant thereof, or variant, fragment and/or fusion of the isoform of NDPK, or specific mutant thereof.

27 11. A method according to claim 10, wherein the test agent is exposed to (i) 28 CK2a, or a variant, fragment and/or fusion thereof, and (ii) the isoform of 29 NDPK, or specific mutant thereof, or a variant, fragment and/or fusion of the isoform of NDPK, or specific mutant thereof, in a cell-based assay.

32 12. A method according to claim 10 or claim 11, wherein whether a test agent has 33 the property of modulating the binding of CK2a to NDPK-B is determined by 34 carrying out a binding assay which provides a quantitative or qualitative measure of the binding of CK2a, or a variant, fragment and/or fusion thereof, 36 to the isoform of NDPK, or specific mutant thereof, or a variant, fragment I and/or fusion of the isoform of NDPK, or specific mutant thereof, or a 2 parameter related to the said binding.

4 13. A method according to claim 12, wherein one of CK2a, or a variant, fragment and/or fusion thereof, or the isoform of NDPK, or specific mutant thereof, or a 6 variant, fragment and/or fusion of the isoform of NDPK, or specific mutant 7 thereof, is concentrated and/or separated using a recognition molecule which 8 specifically binds one of CK2a, or a variant, fragment and/or fusion thereof, or 9 the isoform of NDPK, or specific mutant thereof, or a variant, fragment and/or fusion of the isoform of NDPK, or specific mutant thereof.

12 14. A method according to claim 13, comprising the step of determining whether 13 the other of CK2a, or a variant, fragment and/or fusion thereof, or the isoform 14 of NDPK, or specific mutant thereof, or a variant, fragment and/or fusion of the isoform of N DPK, or specific mutant thereof, has been concentrated and/or 16 separated along with its binding partner by carrying out an assay for an * *** 17 activity of the said other of CK2a, or a variant, fragment and/or fusion thereof, * : 18 or the isoform of NDPK, or specific mutant thereof, or a variant, fragment S..

19 and/or fusion of the isoform of NDPK, or specific mutant thereof. : 21 15. A method according to claim 10 or claim 11 wherein whether a test agent has *. *** * .5* 22 the property of modulating the binding of CK2a to the isoform of NDPK, or 23 specific mutant thereof, is determined by carrying out a binding assay which 24 provides a quantitative or qualitative measure of an activity of CK2a, or a variant, fragment and/or fusion thereof, or the isoform of NDPK, or specific 26 mutant thereof, or a variant, fragment and/or fusion of the isoform of NDPK, or 27 the specific mutant thereof, which activity varies depending on the binding of 28 CK2a, or a variant, fragment and/or fusion thereof, to the isoform of NDPK, or 29 specific mutant thereof, or a variant, fragment and/or fusion of the isoform of NDPK, or specific mutant thereof.

32 16. A method according to claim 10 or claim 11, wherein whether a test agent has 33 the property of modulating the binding of CK2a to the isoform of NDPK, or a 34 specific mutant thereof, is determined by exposing the test agent to intracellular CK2a, or a variant, fragment and/or fusion thereof, and the 36 isoform of NDPK, or specific mutant thereof, or a variant, fragment and/or 1 fusion of the isoform of NDPK, or specific mutant thereof, in a cell-based 2 assay.

4 17. A method according to claim 10 or claim 11, wherein whether a test agent has the property of modulating the binding of CK2a to the isoform of NDPK, or 6 specific mutant thereof, is determined by introducing the test agent to an 7 experimental animal and assessing whether the test agent has the property of 8 modulating the binding of CK2a to the isoform of NDPK, or specific mutant 9 thereof, in the experimental animal.

11 18. A method according to claim 17, wherein the experimental animal has been 12 genetically altered so that a greater or lesser amount of CK2ci, or a variant, 13 fragment and/or fusion thereof, binds NDPK, or specific mutant thereof, or a 14 variant, fragment and/or fusion of the isoform of NDPK, or specific mutant thereof, in at least some cells of the experimental animal under basal 16 conditions, or in response to a specific stimulus, than would be the case in 17 equivalent circumstances in the equivalent wild type animal. *..

19 19. A method according to any one preceding claim, wherein the test agent is a * fragment of an isoform of NDPK, or a variant or fusion thereof.

22 20. The use of (i) an isofomi of NDPK, or a specific mutant thereof, or a variant, ....

23 fragment and/or fusion of the isoform of NDPK, or specific mutant thereof, and * . 24 (ii) CK2ci, or a variant, fragment and/or fusion of CK2a, in a screening assay to identify agents which have the property of modulating the binding of CK2a 26 to the isoform of NDPK, or specific mutant thereof.

28 21. A kit comprising (i) an isoform of NDPK, or a specific mutant thereof, or a 29 variant, fragment and/or fusion of the isoform of NDPK, or specific mutant thereof, and (ii) CK2a, or a variant, fragment and/or fusion of CK2a, and (iii) 31 one or more test agents suitable for screening to assess whether one or more 32 of the said test agents has the property of modulating the binding of CK2cz to 33 the isoform of NDPK, or specific mutant thereof.

22. A kit according to claim 21 in which CK2a, or a variant, fragment and/or fusion 36 of CK2a is not bound to CK23.

2 23. The use of an experimental animal, which has been genetically altered so that 3 a greater or lower amount of CK2a binds an isoform of NDPK, or a specific 4 mutant thereof, under basal conditions or in response to a specific stimulus than would be the case in equivalent circumstances in the corresponding wild 6 type animal, and a test agent, in an experiment to assess whether the test 7 agent has the property of modulating the binding of CK2a to the isoform of 8 NDPK, or specific mutant thereof.

24. A method of developing a pharmaceutical product, comprising the steps of 11 identifying one or more agents by a method according to any one of claims 1 12 to 19 and modifying the one or more identified agents to optimise their 13 pharmaceutical properties.

25. A pharmaceutical product developed by a method according to claim 24. S..

17 26. An agent identified by a method according to any one of claims 1 to 19, or * * : 18 developed by a method according to claim 24, for use in therapy. * * . **.S

27. A gene therapy vector which is operable, when introduced into a host, to 21 cause cells within the host to express a peptide agent identified by a method *.

22 according to any one of claims 1 to 19, or developed by the method of claim 23 24.

28. A pharmaceutically acceptable agent which has the property of modulating 26 the binding of CK2a to an isoform of NDPK, or a specific mutant thereof, in 27 vivo.

29 29. A pharmaceutically acceptable agent according to claim 28, which has the property of disrupting the binding of CK2o to NDPK-B in vivo.

32 30. A pharmaceutically acceptable agent according to claim 28 or 29, wherein the 33 agent is a fragment of the or an isoform of NDPK, or a variant and/or fusion 34 thereof, which binds to CK2a in vivo and which has the property of disrupting the binding of CK2a to the isoform of NDPK, or variant and/or fusion thereof I by competing with the isoform of NDPK, variant and/or fusion thereof, to bind 2 CK2a.

4 31. A gene therapy vector which is operable, when introduced into a host, to cause cells within the host to express a peptide which has the property of 6 modulating the binding of CK2o to an isoforrn of NDPK, or a specific mutant 7 thereof, in viva 9 32. A gene therapy vector according to claim 31, wherein the peptide is a fragment of the or an isoform of NDPK, or a variant and/or fusion thereof, 11 which binds to CK2a in vivo and which, when expressed within the cells of the 12 host, has the property of disrupting the binding of CK2o to the isoform of 13 NDPK by competing with the isoform of NDPK to bind CK2a.

33. A method of modulating the binding of CK2ci to an isoform of NDPK, or a 16 specific mutant of NDPK found in the patient, in a patient in need thereof, the:.

17 method comprising administering an effective amount of an agent which has 18 the property of modulating the binding of CK2a to the isoform of NDPK, or 19 specific mutant thereof, or an effective amount of a gene therapy vector which s..

is operable, when introduced into a patient, to cause cells within the patient to 21 express an effective amount of a peptide agent which has the property of 22 modulating the binding of CK2a to the isoform of NDPK, or specific mutant 23 thereof.

34. A method according to claim 33, wherein the agent was identified by the 26 method of any one of claims 1 to 19, or developed by the method of claim 24.

28 35. A method according to claim 33 or claim 34, wherein the agent has the 29 property of modulating the binding of CK2a to the isoform of NDPK, or specific mutant thereof, because it has the property of modifying CK2a or the isoform 31 of NDPK, or specific mutant thereof, to reduce or remove the affinity of CK2a 32 for the isoform of NDPK, or specific mutant thereof, or vice versa, and thereby 33 has the property of disrupting the binding of CK2a to the isoform of NDPK, or 34 specific mutant thereof.

1 36. A method according to any one of claims 33 to 35, wherein the agent disrupts 2 the binding of CK2o to the isoform of NDPK, or specific mutant thereof, 3 because it competes with the binding of CK2a to the isoform of NDPK, or 4 specific mutant thereof.

6 37. A method according to any one of claims 33 to 36, which is a method of 7 treating and/or preventing a cancer in which increased binding of CK2a to 8 NDPK-B is a causative factor, or a cancer in which increased binding of CK2a 9 to NDPK-A is a causative factor.

11 38. The use of an agent which modulates the binding of CK2a to an isoform of 12 NDPK, or a specific mutant thereof, in the manufacture of a medicament for 13 treating and/or preventing a condition associated with modulated binding of 14 CK2a to the isoform of NDPK, or a specific mutant thereof.

16 39. A use according to claim 38, wherein the agent was identified by the method * *.* 17 of any one of claims 1 to 19 or developed by a method according to claim 24. * *** 19 40. A use according to claim 37 or claim 38, wherein the medicament is for the: treatment and/or prevention of cancer which results from a mutation in the 21 binding site of an isoform of NDPK for CK2a. *.*** 23 41. A method of assessing whether a patient has an increased amount of CK2a 24 bound to an isoform of NDPK, or a mutant thereof, compared to a reference figure, the method comprising taking a sample of one or more cells from the 26 patient, lysing the one or more cells and measuring the amount of CK2a 27 bound to the isoform of NDPK, or mutant thereof, in the lysed cells.

29 42. A method of assessing the distribution of CK2a, in cells of a patient, the method comprising the steps of taking a sample of cells from a patient, lysing 31 the cells, and assaying two or more of (i) the amount of CK2a, which is bound 32 to a first isoform of NDPK, or a mutant thereof, (ii) the amount of CK2a which 33 is bound to a second isoform of NDPK, or mutant thereof, and (iii) the amount 34 of CK2a which is not bound to either isoform of NDPK, or mutant thereof.

1 43. A method according to claim 42, wherein the first isoform of NDPK is NDPK-A 2 and the second isoform of NDPK is NDPK-B.

4 44. A method of assessing the distribution of CK2a in cells within a patient according to claim 42 or claim 43, comprising assaying the (i) the amount of 6 CK2o which is bound to the first isoform of NDPK, or mutant thereof, (ii) the 7 amount of CK2a which is bound to the second isoform of NDPK, or mutant 8 thereof, and (iii) the amount of CK2a which is not bound to either the first or 9 second isoforms of NDPK, or mutants thereof, in a single lysed sample.

11 45. The use of (i) an inhibitor of CK2a, and (ii) an agent which has the property of 12 modulating the binding of CK2a to an isoform of NDPK, or a specific mutant 13 thereof, in the treatment and/or prevention of a disease associated with 14 increased or decreased binding of CK2a to the isoform of NDPK, or specific mutantthereof.

16. a.

17 46. A use according to claim 45, wherein the disease is a cancer associated with * 18 increased binding of CK2a to NDPK-B and the agent which has the property 19 of modulating the binding of CK2a to NDPK-B has the property of disrupting: ..

the binding of CK2a to NDPK-B, or the property of promoting the binding of 21 CK2atoNDPK-A. p..) I.

PS P

23 47. A use according to claim 45 or claim 46, further comprising the use of an 24 agent which inhibits AMPK.

26 48. A use according to any one of claims 45 to 47, wherein the inhibitor of CK2a 27 is a glitazone.

29 49. A composition comprising (i) an inhibitor of CK2a and (ii) an agent which has the property of modulating the binding of CK2a to an isoform of NDPK, or a 31 mutant thereof.