WO1994016712A1

WO1994016712A1 - Creatine phosphate, creatine phosphate analogs, and uses therefor

Info

Publication number: WO1994016712A1
Application number: PCT/US1994/001005
Authority: WO
Inventors: Rima Kaddurah-Daouk; James W. Lillie
Original assignee: Amira, Inc.
Priority date: 1993-01-27
Filing date: 1994-01-27
Publication date: 1994-08-04
Also published as: AU6097994A

Abstract

The present invention relates to a method of inhibiting tumor growth in an individual by administering creatine phosphate (N-phosphorylcreatine) or an analog thereof. The present invention further relates to compositions useful in the present method, including phosphorylcreatine and analogs of phosphorylcreatine. In one embodiment, phosphorylcyclocreatine is provided as the disodium salt, trihydrate (C5H8N3O5P.Na2.3H2O), which has improved solubility in comparison with the dilithium dihydrate of phosphorylcyclocreatine or cyclocreatine itself. As antitumor agents, N-phosphorylcreatine or an analog thereof can be administered to a mammal (e.g., a human) to inhibit the growth of a tumor.

Description

CREATINE PHOSPHATE, CREATINE PHOSPHATE ANALOGS.

AND USES THEREFOR

Description

Background of the Invention Cancer is a leading cause of deaths worldwide.

Numerous compounds are used as anti-tumor agents; many of these agents are non-specific in their activity and kill rapidly proliferating normal cells, as well as tumor cells.

Summary of the Invention The present invention is based on the surprising discovery that cyclocreatine phophate inhibits tumor growth. The present invention relates to creatine phosphate, creatine phosphate analogs, and methods of inhibiting tumor growth in an individual by administering creatine phosphate (phosphorylcreatine) or an analog of creatine phosphate. In one embodiment, phosphorylcyclocreatine is provided as the disodium salt, trihydrate (C₅H₈N₃0₅P* a₂ ^#3H₂θ) , which has improved solubility in comparison with the dilithium dihydrate of phosphorylcyclocreatine or cyclocreatine itself.

As antitumor agents, creatine phosphate or an analog of creatine phosphate can be administered to a mammal (e.g., a human) to inhibit the growth of a tumor. Administration of these compositions according to the present method can result in the inhibition of tumor growth. Inhibition of tumor growth may involve a reduction in the extent of tumor growth, tumor regression (a decrease in size of a tumor) , disappearance of a tumor, as well as reductions in the extent of growth of metastases, regression of metastases, or disappearance of metastases. Brief Description of the Drawings

Figure 1 is an illustration of 3-dimensional models of creatine, N-phosphorylcreatine, cyclocratine, and N- phosphorylcyclocreatine. The models were generated from published X-ray diffraction data for creatine, N- phosphorylcreatine and N-phosphorylcyclocreatine (Phillips, G.N. et al.. J. Am. Chem. Soc. 101: 7120-7121, 1979; Herriott, J.R. and Love, W.E., Acta Cryst. B24: 1014-1027, 1968; Mendel, H. and Hodgkin, D.C.k, Acta Crvst. 2: 443-446, 1954). The coordinates of N-phosphoryl¬ cyclocreatine were used to generate a model of cyclocreatine.

Figure 2 is a histogram illustrating the mM concentration of cyclocreatine (CCr) required to reduce colony formation in soft agar by 50% (IC50) for eight

"high" CK expressing tumor cell lines (High CK Lines) and for four low CK expressing tumor cell lines (Low CK) . The cells were exposed to drug continuously.

Figure 3 is a histogram illustrating the intracellular accumulation of N-phosphorylated CCr in four "high" CK expressing tumor cell lines (High CK) and in two low CK expressing tumor cell lines (Low CK) is shown. Cells were exposed to 14 mM CCr for 48 hours. Cyclocreatine is efficiently phosphorylated in vitro in high CK expressing cell lines but not in low Ck expressing cell lines.

Figure 4 is a graph illustrating the effect of CCr"P administered intraperitoneally (37.5 mg/ml) on human prostate adenocarcinoma cell line-derived tumors in a nude mouse xenograph model (N = 5) . The mean tumor volume for the 5 animals in the control (o) and in the treated (•) group is plotted until day 30, when the dose was doubled. Figures 5A and 5B are graphs illustrating the effect of CCr^"P on human prostate adenocarcinoma cell line- derived tumors in a nude mouse xenograph model (N=5) . Treatment was initiated on day six when all animals had palpable tumors and doubled on day 30 to 50mg/ml CCr'P (injected intraperitoneally) . Figure 5A shows the change in tumor volume over time in each of the 5 animals in the treated group until day 64, where each symbol represents one animal. Figure 5B shows the change in tumor volume over time in each of the 5 animals in the control group (treated with saline injected intraperitoneally) where each symbol represents one animal. Note the different scales in 5A and 5B.

Figure 6 is a graph illustrating the effect of CCr'P administered intraperitoneally (50 mg/ml) on human prostate adenocarcinoma cell line-derived tumors in a nude mouse xenograph model (N = 8) . The mean tumor volume for the 8 animals in the control (•) and in 8 animals in the treated (o) group is plotted until day 39. The dose started at 1 g/kg/day and changed to 2 g/kg/day on day 26.

Figure 7 is a graph illustrating that established human prostate adenocarcinoma cell line-derived tumors in a nude mouse xenograph model (N=4) are responsive to CCr'P administered intraperitoneally, 50 mg/ml (1 g/kg/day) . The means for the 4 animals in the control (•) and the 4 animals in the treated (o) group are plotted until day 25. Figure 8 is a graph illustrating that human cervical carcinoma cell line-derived tumors in a nude mouse xenograph model (N=4) are responsive to CCr'P administered intraperitoneally, 50 mg/ml. The means for the 4 animals in the control (o) and the 4 animals in the CCr'P-treated (•) group are plotted until day 28. Figure 9 is a graph illustrating that human cervical carcinoma cell line-derived tumors in a nude mouse xenograft model (n = 4) are responsive to CCr'P administered intravenously (75 mg/ml) . The means for the 4 animals in the control (•) and the 4 animals in the CCr'P-treated (o) group are plotted. Figure 10 is a Η NMR of the disodium salt of N- phosphorylcyclocreatine (trihydrate) in D₂0, 300 MHz. The chemical shifts are reported in ppm relative to TMS (assinged to zero) . Figure 11 is an HPLC profile of the disodium salt of N-phosphorylcyclocreatine (trihydrate) .

Figure 12 is an ultraviolet (UV) spectrum for the disodium salt of N-phosphorylcyclocreatine (trihydrate) , λ-_^ = 207 nm.

Detailed Description of the Invention

The present invention relates to creatine phosphate, creatine phosphate analogs, and methods of inhibiting tumor growth in an individual by administering N- phosphorylcreatine (creatine phosphate) or an analog of of N-phosphorylcreatine. As shown herein, when cyclocreatine (CCr) , an analog of creatine having antitumor activity, is administered to tumor cells, the compound is phosphorylated by creatine kinase and accumulates in the cells as a synthetic phosphagen, N-phosphorylcyclocreatine (CCr'P) . As is further described herein, accumulation of this phosphagen (due to creatine kinase activity) is detrimental to tumor cell growth (see Example 1) .

Further studies described herein demonstrate that administration of an analog of N-phosphorylcreatine can inhibit tumor growth. In particular, N- phosphorylcyclocreatine was shown to inhibit the growth of human tumor cell xenografts in mice (Examples 2 and 3) . It is reasonable to expect, based on the work described herein in established animal models, that there is a similar effect in humans.

The present invention further relates to compositions useful in the present method, including phosphorylcreatine and analogs of phosphorylcreatine and pharmaceutically acceptable salts (e.g., lithium, sodium) thereof. In one embodiment, N-phosphoryl-cyclocreatine is provided as the disodium salt, trihydrate

, which has improved solubility in comparison with cyclocreatine or the dilithiu dihydrate of phosphorylcyclocreatine. The dilithiu salt (dihydrate) can be prepared according to Annesley, T.M. and J.B. Walker, Biomed. Biophvs. Res. Commun.. 74 (1) : 185-190 (1977) and Struve G.E. et al.. J. Org. Chem.. 42.25^ : 4035-4040 (1977). In addition to compounds which act as phosphagens (e.g., N-phosphorylcyclocreatine, N- phosphorylho ocyclocreatine, N-phosphoryl-3- guanidinopropionate) , analogs of N-phosphorylcreatine include compounds such as non-hydrolyzable analogs and/or irreversible inhibitors of creatine kinase. Analogs of N- phosphorocreatine can be obtained by chemical phosphorylation (see e.g., Example 4) or enzymatic phosphorylation (see e.g., Rowley et al.. J. Am. Chem. Soc. , £3_.: 5542-5551 (1971)) of the corresponding creatine analog. Creatine analogs have been previously described in U.S. Serial No. 07/610,418 and U.S. Serial No. 07/812,561, the contents of both of which are incorporated herein by reference. For instance, cyclocreatine (Griffiths, G.R. and J.B. Walker, J. Biol. Chem. , 251: 2049-2054 (1976)), homocyclocreatine (1-carboxyethy1-2- iminoimidazolidine; Roberts, J.J. and J.B. Walker, Arch. Biochem. Biophvs.. 220: 563-571 (1983)), 1-carboxymethyl- 2-iminohexahydropyrimidine (Griffiths, G.R. and J.B. Walker, J. Biol. Chem.. 251: 2049-2054 (1976)), β- guanidinopropionate (Sigma Chemical Co., St. Louis, MO) , guanidinoacetate (Sigma Chemical Co., St. Louis, MO), carbocreatine (Nguyen, Ann Cae Khue, Ph.D. Thesis in Pharmaceutical Chemistry, (University of California, San Francisco, 1983) , l-carboxymethyl-2-iminoimidazolidine-4- one (Jd.), N-methyl-N-amidino-/S-alanine (McLaughlin, A.C. et al.. J. Biol. Chem.. 247: 4382-4388 (1972), N-ethyl-N- amidinoglycine (.Id.), N-methyl-N-amidinoaminomethyl- phosphinic acid (Xd.), (R) -N-amidinoproline (.Id.), (R) -N- amidinoazetidine-2-carboxylic acid (Dietrich, R.F. et al.. Biochemistry. 1£:3180-3186 (1980)), (R)-N-methyl-N- a idinoalanine (Id.) can be converted to the corresponding N-phosphorylated form for use in the present invention. A general formula for phosphorylcreatine and phosphorylcreatine analogs is provided below.

and pharmaceutically acceptable salts thereof, wherein:

(a) Y is selected from the group consisting of: -C0₂H, -NHOH, -N0₂, -S0₃H, -P(=0) (OH) (OH) , and -P(=0) (OH) (H);

(b) A is selected from the group consisting of: C,- C₃ alkyl, C,-C₃ alkenyl, C₂ alkynyl, and C^C_j alkoyl, each having 0-2 bulky and/or hydrophobic substitutents which are selected independently from the group consisting of:

(1) K, where K is selected from the group consisting of: C,-C₃ straight alkyl, C₂-C₃ straight alkenyl, and C,-C₃ straight alkoyl; and

(2) -NH-M, wherein M is selected from the group consisting of: hydrogen and C_j-C₃ alkyl; (c) X is selected from the group consisting of: NR,, CHR-, C_j alkenyl with an R- substituent, O and S, wherein R_x is selected from the group consisting of: (1) hydrogen; and

(2) K where K is selected from the group consisting of: C,-C₃ straight alkyl, C_J-C_J straight alkenyl, C₁-C₃ straight alkoyl, C₃- C₄ branched alkyl, and C₃-C₄ branched alkenyl;

(d) If A is chosen as a C, alkenyl group, then X must also be chosen as an alkenyl group, and if X is chosen as a C_ alkenyl group, then A must be an alkenyl group, wherein A and X are connected by a double bond;

(e) Zj and Z₂ are chosen independently from the group consisting of: =0, -NR₂R₃, -CHR₂R₃, and -NR₂0R₃; wherein, Z_ and Z₂ may not both be =0 and wherein for a selected Z group, R₂ and R₃ are independently selected from the group consisting of:

(1) hydrogen;

(2) K, where K is selected from the group consisting of:

straight alkyl, ^C ₂~C₃ straight alkenyl, C--C₃ straight alkoyl, C₃-C₄ branched alkyl, and C₃-C₄ branched alkenyl; and

(3) B, wherein B is selected from the group consisting of: -C0₂H, -C(=0) (NHOH) , -NH0H, -S0₃H, -N0₂, -OP(=0) (OH) (OH) ,

-0P(=S) (OH) (OH) , -0P(=0) (H) (OH) , -0P(=S) (H) (OH) , -P(=0) (OH) (OH) , -P(=S) (OH) (OH) , -P(=0) (H) (OH) and -P(=S) (H) (OH) , wherein B is optionally connected via a linker selected from the group consisting of: C,-C₂ alkyl, C^ alkenyl, and C,-C₂ alkoyl;

(f) If R. and at least one R₂ or R₃ group are present, Ri may be connected by a single or double bond to an R₂ or R₃ group to form a cycle of 5 to 7 members;

(g) If one or more R₂ or R₃ groups is present, any two may be connected by a single or double bond to form a cycle of 5 to 7 members; and

(h) At least one of R₂ or R₃ must be selected from B.

Further phosphorylcreatine analogs are encompassed by the expanded formula provided below. Such compounds include, for example, bisubstrate analogs and irreversible inhibitors of creatine kinase.

and pharmaceutically acceptable salts thereof.

(a) Y is selected from the group consisting of:

-C0₂H, -NHOH, -N0₂, -S0₃H, -P(=0) (OH) (OJ) , and -P(=0) (OJ) (H) , wherein J is hydrogen, C,-C₆ straight or branched chain alkyl, C₂-C₆ straight or branched alkenyl, or aryl group. (b) A is selected from the group consisting of: c.- C₅ alkyl, C_j-Cj alkenyl, C₂-C₅ alkynyl, and C_-c₅ alkoyl. The chain may have 0-2 bulky and/or hydrophobic substitutents that are intended to enhance the effectiveness of the compound or to prevent its decomposition by pathways such as that in which creatine is cyclized to creatinine. The substituents are selected independently from the following: (1) K, where K is selected from the group consisting of: C.-C₆ straight alkyl, C₂-C₆ straight alkenyl, Cι-C₆ straight alkoyl, C₃- C₆ branched alkyl, C₃-C₆ branched alkenyl, and C₄-C₆ branched alkoyl. K may also have 0-2 substituents, selected to be reactive with an enzymic nucleophile. A halogen, such as Br or Cl, or an epoxy or an acetoxy group are examples of such reactive groups.

(2) An aryl group containing a 1-2 carbocyclic (such as phenyl or naphthyl) or heterocyclic ring (such as indolyl or adeninyl) . The aryl group may have 0-2 substitutents that could react with an enzymic nucleophile such as -CH₂L or - C0CH₂L, in which L is a leaving group such as bromo, chloro, epoxy and acetoxy..

(3) -NH-M, wherein M is hydrogen, C--C₄ straight or branched alkyl or alkoyl, or C₂-C₄ straight or branched alkenyl.

(c) X is selected from the group consisting of:

NR_j, CHR,, C_ alkenyl with an R, substituent, O and S, wherein Rj is selected from the group consisting of: (1) Hydrogen;

(2) K as described above in subsection (b) ;

(3) An aryl group as described in subsection

(b) ; (4 ) C_j-C, α-amino-fo-methyl-fc-adenosylcarboxylic acid attached via the fo-methyl carbon ; (5) C₅-C₉ α-amino-fo-aza-to-methyl-fo- adenosylcarboxylic acid attached via the fo-methyl carbon; and (6) C_j-C₉ α-amino-t>-thia-J.-methyl-t>- adenosylcarboxylic acid attached via the t)-methyl carbon.

(d) If A is chosen as a C- alkenyl group, then X must also be chosen as an alkenyl group, and if X is chosen as a C- alkenyl group, then A must be an alkenyl group, wherein A and X are connected by a double bond.

(e) Zi and Z₂ are chosen independently from the group consisting of: =0, -NR₂R₃, -CHR₂R₃, and -NR₂0R₃; wherein, Z, and Z₂ may not both be =0 and wherein, for a selected Z group, R₂ and R₃ are independently selected from the group consisting of:

(1) Hydrogen; (2) K as described above in subsection (b) ;

(3) An aryl group as described above in subsection (b) ;

(4) C₄-C₈ α-amino-carboxylic acid attached via the t)-carbon; (5) B, wherein B is selected from the group consisting of: -C0₂H, -C(=0) (NH0H) , -NH0H, -S0₃H, -N0₂, -0P(=0) (OH) (0J) , -OP(=S) (OH) (OJ) , -0P(=0) (H) (OJ) , -OP(=S) (H) (OJ), -P(=0) (OH) (OJ), -P(=S) (OH) (OJ) , -P(=0) (H) (OJ) and -P(=S) (H) (OJ) . As above, J is either 5 hydrogen, C--C₆ straight or branched chain alkyl, Cj-Cβ straight or branched alkenyl, or aryl group. B is optionally connected via a linker such as a C.-C^ alkyl, C^ alkenyl, or

alkoyl;

10 (6) -D-E, wherein D is selected from the group consisting of: Cι~C₃ straight alkyl, C₃ branched alkyl, C₂-C₃ straight alkenyl, C₃ branched alkenyl, C--C₃ straight alkoyl, aryl, and aroyl; and

15 wherein E is selected from the group^' consisting of:

-(P0₃)_nNMP, wherein n is 0-2, NMP is a ribonucleotide monophosphate connected via the 5'-phosphate, 3'-phosphate or the

20 aromatic ring of the base, and (=S) may be substituted for one or more (=0) in E. When the latter type of substitution is made in NMP, the NMP group is converted to a ribonucleotide monothiophosphate;

25 -[P(=0) (0CH₃) (0) ]_m-Q, where m is 0-3 and Q is a ribonucleoside connected via the ribose or the aromatic ring of the base and (=S) may be substituted for one or more (=0) in E;

30 -[P(=0) (OH) (CH₂) ]_m-Q, wherein m is 0-

3, Q is a ribonucleoside connected via the ribose or the aromatic ring of the base, and (=S) may be substituted for one or more (=0) in E; -[P(=0) (H) (CH₂) ]_m-Q, wherein m is 0-3, Q is a ribonucleoside connected via the ribose or the aromatic ring of the base and (=S) may be substituted for one or more (=0) in E; and an aryl group containing 0-3 substituents chosen independently from the group consisting of: Cl, Br, epoxy, acetoxy, -0G, -C(=0)G, and -C0₂G, where G is independently selected from the group consisting of: C--C₆ straight alkyl, C₂-C₆ straight alkenyl, C--C₆ straight alkoyl, C₃-C₆ branched alkyl, C₃-C₆ branched alkenyl, and C₄-C₆ branched alkoyl. E may be attached at any point to D.

If D is an alkyl or alkenyl group, D may be connected at either or both ends by an amide linkage. An amide linkage can be in either orientation (-CONH- or -NHCO-) , and if two amide linkages are present their orientations may be the same or different. (7) -E, wherein E is as described above in subsection (6) , and, if E is aryl, E may be connected by an amide linkage, which can be in either orientation.

(f) If R| and at least one R₂ or R₃ group are present, R_! may be connected by a single or double bond to an R₂ or R₃ group to form a cycle of 5 to 7 members. (g) If one or more R₂ or R₃ groups is present, any two may be connected by a single or double bond to form a cycle of 5 to 7 members. Thus, when present, an R₂ group may be joined to another R₂ group or to an R₃ group, and conversely, an R₃ group may be joined to another R₃ group or to an R₂ group, in the manner described.

(h) At least one of R₂ or R₃ must be selected from the group consisting of -B, -D-E, and -E.

As antitumor agents, phosphorylcreatine or an analog of phosphorylcreatine, such as N-phosphorylcyclocreatine, can be administered to an individual (e.g., a mammal, such as a human) alone or in combination with another drug (e.g., another analog of phosphorylcreatine or other antitumor agent), to inhibit the growth of a tumor (i.e., as agents in antitumor therapy) . As antitumor agents for cancer chemotherapy, N-phosphorylcreatine and suitable N- phosphorylcreatine analogs can prevent, reduce, or eliminate neoplastic disease. Administration of these compositions according to the present method can result in the inhibition of tumor growth in an individual having a tumor (e.g., sarcoma, carcinoma, neuroblastoma, retinoblastoma, melanoma, glioma, teratoma) . Inhibition of tumor growth may involve a reduction in the extent of tumor growth, tumor regression (a decrease in size of a tumor) or disappearance of a tumor as illustrated in Examples 2 and 3. Inhibition of growth of metastases may also result. Due to their low toxicity, compositions of the present invention may be useful for prolonged therapy to prevent recurrence of disease.

The results reported in the Examples suggest that the active form of cyclocreatine in vitro is CCr^'P. The mechaniεm by which CCr inhibits cell growth in vivo is unknown. CCr'P may regulate cellular processes, independent of ATP regeneration, and CK may only be necessary to convert an inactive prodrug (free CCr) to its active, phosphorylated form (CCr'P) . In such a case, CCr'P may interact with a cellular protein or lipid that normally binds creatine phosphate, because the structure of the two molecules is similar (Figure 1) . It has been suggested, for instance, that creatine phosphate regulates allosterically the activity of several enzymes involved in carbohydrate metabolism (Fu, J.Y. and R.G. Kemp, J. Biol. Chem. 248: 1124-1125, 1973). Tumor cells may undergo changes in membrane permeability; preliminary data indicates that, when added exogenously to cells in vitro. N-phosphorylcyclocreatine accumulates intracellularly. In addition, or alternatively, N-phosphorylcyclocreatine could function as an antitumor agent by modifying extracellular processes, without entering cells. These extracellular functions could be ATP dependent or mediated through regulation of ion transport, for example.

Thus, N-phosphorylcreatine and analogs thereof, such as N-phosphorylcyclocreatine, may exert their function intracellularly or extracellularly. In any event, by use of the N-phosphorylated form of creatine analogs according to the method of the present invention, it may be possible to bypass any need for CK activity. Whether N- phosphorylcreatine or an analog of N-phosphorylcreatine will be effective against a selected tumor type can be determined empirically by combining the cells with the compound and determining the effect on cells. When practical, a biopsy of a tumor can be obtained, and liye cells can be tested for sensitivity to drug in vitro.

Mode of Administration For in vivo applications, N-phophorylcreatine or an analog thereof can be administered by a variety of routes, including, but not necessarily limited to, oral (e.g., dietary) , topical, transdermal, inhalation, or parenteral (e.g., subcutaneous, intramuscular, intravenous injection, intraperitoneal) routes of administration, for example. A therapeutically effective amount (i.e., one that is sufficient to produce the desired effect of inhibition of tumor growth in an individual) of a composition comprising N-phosphorylcreatine or an analog thereof is administered to the individual. Those skilled in the art will determine the actual amount of drug to be administered is based on factors such as the route of administration, size and age of the individual, the severity of symptoms, other medical conditions and the desired aim of treatment (desired effect) .

N-phosphorylcreatine or N-phosphorylcreatine analogs will be formulated in a manner suitable for the selected route of administration (e.g. , powder, tablet, capsule, cream or ointment, solution, emulsion) . An appropriate composition comprising a creatine analog can be prepared in a physiologically acceptable vehicle or carrier. For example, a composition in tablet form can include one or more additives such as a filler (e.g., lactose), a binder (e.g., gelatin, carboxymethyl-cellulose, gum arabic) , a flavoring agent, a coloring agent, or coating material as desired. For cream or ointment preparations, crystalline powder can be mixed with an appropriate base, such as a polyethylene glycol (PEG) cream base. For solutions or emulsions in general, carriers include aqueous or alcoholic/aqueous solutions, emulsions or suspensions, including saline and buffered media. Parenteral vehicles can include sodium chloride solution, Ringer's dextrose, dextrose and sodium chloride, lactated Ringer's or fixed oils. In addition, intravenous vehicles can include fluid and nutrient replenisherε, saline and phosphate buffered saline (PBS) and electrolyte replenishers, such as those based on Ringer's dextrose. Preservatives and other additives can also be present. For example, antimicrobial, antioxidant, chelating agents, and inert gases can be added. (See, generally. Remington's Pharmaceutical Sciences. 16th Edition, Mack, Ed., 1980) .

The present invention will be further illustrated by the following examples, which are not intended to be limiting in any way.

EXAMPLE 1

Cyclocreatine (CCr) is the most active substrate analog of creatine to date. The compound was first synthesized in 1971 (Rowley, G.L. et al.. J. Am. Chem. Soc.. 93: 5542-5551, 1971) and has been studied extensively by Walker and colleagues in studies of energy metabolism in ischemia (reviewed in Walker, J.B., Adv. Enzvmol.. 50: 177-241, 1979) . CCr is phosphorylated efficiently in vitro and in vivo by creatine kinase to yield l-carboxymethyl-3-phosphono-2-iminoimidazolidine or cyclocreatine phosphate (CCr~P; See Figure 1) . Cyclocreatine and creatine are structurally related; cyclocreatine phosphate and creatine phosphate are also structurally related (Figure 1; Phillips, G.N. et al. , J^_. Am. Chem. Soc. 101: 7120-7121, 1979).

Methods and Materials Compounds

Cyclocreatine (CCr) was prepared according to published protocols (Griffiths, G.R. and J.B. Walker, ____ Biol. Chem. 251: 2049-2054, 1976). Determination of free and N-phosphorylated CCr in cultured tumor cells was by the colorimetric assay using the chromogenic reagent Na₃ [Fe(CN)₅NH₃] (Griffiths, G.R. and J.B. Walker, J. Biol. Chem. 251: 2049-2054, 1976) . Confluent dishes of each cell line were incubated with CCr for the time indicated, washed 3x with PBS, and fixed with 1 ml of 0.2 M perchloric acid. The perchloric acid wash was assayed spectrophotometrically for free and N-phosphorylated CCr as described (Griffiths, G.R. and J.B. Walker, J. Biol. Chem. 251: 2049-2054, 1976).

Cells

Transformed and non-transformed cells were obtained from the American Type Culture Collection (ATCC) , Rockville, MD. Cell lines obtained from the ATCC were grown as suggested by the ATCC Catalogue of Cell Lines and Hybridomas, Hay et al. (eds.)

Creatine Kinase Assays

Cell lines were harvested at confluency after being washed 2 or 3 times with phosphate buffered saline (PBS) and concentrated by centrifugation. The resulting pellets were resuspended in 50 to 200 μl of 0.2M Tris/O.IM NaCl were lysed by three cycles of freeze-thawing. Particulate matter was removed by centrifugation and the supernatant was used for creatine kinase enzyme assays, creatine kinase isozyme electrophoresis assays, or to measure total protein levels of cell extracts. Protein content of cellular extracts was determined using the Bio-Rad Protein Assay (Bio-Rad, #500-0006) .

Total creatine kinase activity was determined in a coupled reaction using a commercially available kit (Sigma Chemical Co. , CK Kit #10) in which the production of ATP by creatine kinase is coupled to the reduction of nicotina ide adenine dinucleotide phosphate (NADP) . 3H-thvmidine Assays

The rate of incorporation of ³H-thymidine into DNA was adopted as a measure of cell proliferation. The cells were plated in 96 well plates and drugs were added the next day. Tritiated thymidine (³H-thymidine; Amersham) was added to a final concentration of 2 μCi/ml at 1 to 24 hours before harvest, depending upon the length of the experiment. To harvest, media was removed by aspiration or inversion and the cells were incubated for 5 minutes at 37°C in 100 μl of trypsin (2.5%, Sigma Chemical Corp., St. Louis, MO) to detach the cells. Cells were harvested with a Skatron or Tom-Tec cell harvester, collected on 102 x 258 mm glass fiber filter paper (Pharmacia) , sealed in a bag containing liquid scintillant (LKB) and were counted in a 1205 Betaplate liquid scintillation counter (Wallac) to determine cell-associated radioactivity. All samples for each time point were taken in duplicate or triplicate. To obtain values for growth as a percentage of control for drug treated cultures, the incorporation of 3H-thymidine in the presence of drug was divided by the incorporation of 3H-thymidine in the absence of drug and was then multiplied by a factor of 100. The drug dose required to reduce incorporation of 3H-thymidine 50% (IC50) was determined by linear regression.

Stem Cell Assay

Cells were incubated in IMDM Enriched Medium, consisting of 77% Iscove's Modified Dulbecco's Medium (IMDM) , 2 mM L-glutamine, 4 mM CaCl₂, 2300 mg/1 NaCl₂, 3 U/ml insulin, 0.5 mg/ml DEAE Dextran, 1.5% BSA, 10% fetal bovine serum, 10% horse serum, 2 Mm Na pyruvate, 100 U/ml Pen/Strep. The soft agar was prepared as a standard assay consisting of two layers: a base feeder layer of 0.5% agar and a less solid top layer (0.3% agar), which holds the tumor cells. After 21 days, the colonies were strained by dropwise addition of a vital stain (p-iodonitrotetralium violet; 0.5 ml or 0.5 mg/ml solution in water) to determine viability and to aid visualization. After 24 hours at 37°C, viable colonies consisting of greater than 50 cells were identified at 40X magnification with a phase contrast microscope.

Selective Anti-Tumor Cell Activity of CCr In Vitro: Inhibition of Colony Formation of Established Tumor Cell Lines in Soft Agar

The creatine analog CCr was assayed for its ability to inhibit the colony formation of established lines of tumor cells in vitro. Tumor cell lines expressing a high level (CK > 0.1 U/mg as determined as described above) of creatine kinase, NCI-H69, ME-180, A2058-055, Y79, 293, OVCAR-3, SK-N-MC and SW48, were growth inhibited in soft agar 50% by less than 10 mM CCr (Figure 2) . In contrast, two low (CK < 0.1 U/mg as determined as described above) CK expressing lines — A2058, A2058-032 — were insensitive to CCr at the highest concentration tested (56 mM) , and two other low CK expressing lines — CaSki, and U87-MG — were inhibited 50% at 23 mM and 40 mM, respectively. These results revealed that tumor cell lines expressing a high level of creatine kinase (CK > 0.1 U/mg) can be efficiently growth inhibited by CCr and that tumor cells having a low level of creatine kinase (approximately 0.01 U/mg) are less sensitive to CCr in vitro.

Tumor cell lines growth-inhibited by CCr accumulate the synthetic phosphagen CCr-P

To determine whether the specificity of CCr against high CK expressing tumor lines in vitro is related to the accumulation of the synthetic phosphagen CCr^'P, the accumulation of CCr'P in high and low CK expressing tumor lines was determined. Figure 3 shows that four high CK expressing lines (ME-180, DU145, MCF-7 and OVCAR-3) that are growth-sensitive to CCr, as determined by incorporation of ³H-thymidine, accumulated a substantial level of CCr'P (between 0.33 and 0.63 μmoles CCr"P/mg cellular protein) . In contrast, the two low CK expressing lines (U87 and A2058) , which are growth-resistant to CCr, accumulated no detectable CCr'P. All six cell lines accumulated between 5 and 27 μmoles free CCr/mg cellular protein (data not shown) . The fact that cell lines most sensitive to CCr in tissue culture are those able to accumulate large quantities of CCr'P suggests that accumulation of this synthetic phosphagen may be necessary for growth inhibition.

EXAMPLE 2

Cyclocreatine Phosphate is Correlated

With Tumor Regression in Vivo

EXPERIMENT I Animals

Athymic (CDl-Nu/Nu) male mice were obtained from Charles River Breeding Labs at an age of 4-6 weeks and delivered to the University of Massachusetts Animal Medicine Facility. Mice were allowed to grow and acclimate for 5 months before injection of tumor cells. To identify each mouse, a small hole was punched at a specific position of the outermost part of the right or left ear. The average weight of the mice at the beginning of the experiment was 27.0 grams. Mice were housed as described below. For this experiment 10 animals were used, 5 treated and 5 untreated. Mice were housed at the II Biotech Park Animal facility operated by the University of Massachusetts Medical Center, 55 Lake Avenue North, Worcester, MA. The animal room maintains a temperature of 30°C, and a relative humidity of 40-70%. The animals are housed in polycarbonate microisolator cages (Lab Products, Inc., Maywood, NJ) . The cages are lined with autoclaved bedding (Sani-Chip, P.J. Murphy, Inc., Montville, NJ) , and provided with autoclaved feed (Purina Diet 5010, Purina, Inc.) and sterilized water. Animals are manipulated for husbandry and experimental treatment under a laminar flow workstation (Nuaire, Inc.), and their cages/bedding changed twice weekly. Once each week terramycin was added to the water, and the water changed twice weekly. Care and handling procedures closely adhered to the guidelines set up by the University of Massachusetts Animal Care and Use Committee (ACUC) .

Cells

The DU-145 human prostate adenocarcinoma was obtained from the American Type Culture collection (ATCC) Rockville, MD and grown according to the protocol suggested by the ATCC of Cell Lines and Hybridomas (6th edition, Hay et al. eds., 1988). The DU145 cells were derived from a primary tissue obtained from a 60 year old Caucasian male with metastatic carcinoma of the prostate which metaεtasized to the brain, liver and bone marrow. The cells were passaged twice weekly for 10 months before injection at a subcultivation ratio of 1:10, 1:15 by trypsinization. Medium consisted of 90% Minimum Essential Medium (Eagle's with Earle's Balanced Salts) without L-glutamine, J.R.H. Biosciences 51-412, EMEM) , 10% fetal bovine serum, 100 U/ml, Penicillin/Streptomycin, 2 mM L-glutamine, 1 mM Na pyruvate (J.R.H. Biosciences). Cells were grown in 75 cm² flasks at 5% C0₂, 37°C.

To harvest sufficient cells for the experiment, 16 150 cm² about 80% confluent flasks of DU-145 cells were treated with 6 mis of 0.25% trypsin until the cells detached from the flask. Trypsin was inactivated with 10 mis of MEM medium + 10% FBS. Cells were collected by centrifugation for 5 minutes at 1500 rpm. The supernatant was removed by aspiration and the pellet was resuspended in 20 ml of plain (no serum) media. The cell suspension was centrifuged again and the pellet was resuspended in 20 mis of serum free MEM media. A cell count was performed to determine cell number and viability, then 2.5 x 10⁶ cells in 0.2 mis phosphate-buffered saline were injected subcutaneously over the proximal right femur of each mouse. Injections were performed with a 1 cc syringe and a 23 g needle.

Measurement of Tumors

Tumors were measured from the exterior with mm calipers held horizontally to the body of the mouse to determine the width, and perpendicularly to determine the height of the tumor. These measurements were multiplied by the tumor's thickness as determined by the calipers to the nearest 0.1 mm to estimate the volume of each tumor.

Drug Administration

N-phosphorylcyclocreatine was prepared as a disodium salt (trihydrate) as described in Example 4. Lot #AM 709 was used. Six days after the cells were injected, all animals had palpable tumors and the treatment was initiated. One group of 5 animals received 0.2 mis twice daily of CCr^'P at 37.5 mg/ml in a saline vehicle pH 7.0, while the other group of 5 animals received only the 0.9% εaline vehicle. All injections were performed intraperitoneally (i.p.), once in the morning at 9:00, and once in the evening at 5:00, six days per week. On the seventh day the animals were not manipulated. On day 30 the dose was doubled by increasing the volume to 0.3 is per injection and the concentration to 50mgs/ml. Cyclocreatine-Phosphate at 50mg/ml was prepared fresh daily in deionized water pH 7.0. Saline was not used as the drug vehicle because the compound is prepared as a disodium salt, therefore a 50 mg/ml solution contains nearly 0.6% sodium. The control animals were injected intravenously with 0.9% buffered saline (pH 7.0) . Just prior to injecting animals, the solution was filtered in a laminar flow hood using a Corning 115 ml micropore filter system (0.2μ) to assure sterility.

Results

When CCr^'P was administered i.p. to animals bearing DU145 human prostate tumors, all of the treated tumors stopped growing by day 11 and began regressing (N = 5) . The mean starting tumor volume was 9 mm³, and by day 20, the mean control tumors reached nearly 70 mm³ while the treated tumors were less than 20 mm³ (see Figure 4). From day 6, when the treatment was initiated, until day 30 when the dose was doubled, the treated tumors showed significant growth delay (p = 0.003; ANOVA) . By day 27 the treated tumors had began growing again and had reached a mean of about 60 mm³, when one of the treated tumors began to regress, followed by its disappearance several days later (see Figure 5A to compare the progress of individual tumors) . On day 27 a second tumor began to regress from a volume of 85.3 mm³, while the other three tumors continued to grow at the same rate as the controls. By day 32 the second tumor had disappeared, and by day 42 a third tumor was regressing from a volume of 334.4 mm³. On day 60 it also disappeared, leaving the five treated animals with 2 tumors. The control tumors continued to grow slowly but none regressed (Figure 5B) . On day 66, the mean control tumor volume was 830 mm³ and the mean treated volume was 290 mm³ (N = 5) . The difference between the two groups for the entire period was significant (p = 0.008; ANOVA) .

None of the animals appeared sickly nor did they display any abnormal behavior throughout the experiment, and all mice gained weight steadily. The control animals gained 1.6 g on average and the cCr-P-treated mice gained 3.5 g on average. Thus, it may be concluded from this experiment that CCr^'P acts as an anti-tumor agent, without any notable signs of toxicity, in a mouse xenograft model for human prostate cancer.

Table 1

Mean weight change in athymic mice bearing DU-145 cell line-derived tumors and treated with either CCr^'P or saline. The final weight measurement was taken on day 64, the first on day 5.

EXPERIMENT II Animals

Athymic (CDl-Nu/Nu) male mice were obtained from Charles River Breeding Labs at an age of 4-6 weeks and delivered to the University of Massachusetts Animal Medicine Facility. Mice were allowed to grow and acclimate for 2 weeks before injection of tumor cells. To identify each mouse, a small hole was punched at a specific position of the outermost part of the right or left ear. The average weight of the mice at the beginning of the experiment was 27.6 grams. Mice were housed as described above. For this experiment 16 animals were used, 8 treated and 8 placebo treated.

Cells DU 145 cells were cultured as described above and

2.5 x 10⁶ cells in 0.2 mis phosphate-buffered saline were injected subcutaneously over the proximal right femur of each mouse. Injections were performed with a 1 cc syringe and a 23 g needle. Tumors were measured as described above.

Drug Administration

N-phosphorylcyclocreatine was prepared as a disodium salt (trihydrate) as described in Example 4. Lot #AM 709 was used. Five days after the cells were injected, all animals had palpable tumors and the treatment was initiated. One group of 8 animals received 0.3 is twice daily of CCr'P at 50 mg/ml (1.0 g/kg/day) in a water vehicle pH 7.0, while the other group of 8 animals received saline pH 7.0. All injections were performed intraperitoneally, once in the morning at 9:00, and once in the evening at 5:00, six days per week. On the seventh day the animals were not manipulated. On day 26 the dose was doubled to 2.0 g/kg/day by increasing the volume to

0.6 mis per injection and maintaining the concentration at 50 mgs/ml.

Results

In this experiment, when CCr^'P was administered i.p. to 8 animals bearing DU145 human prostate tumors, the treated tumors grew more slowly. The mean starting tumor volume was 15 mm³, and by day 25, the day before the dose was doubled, the mean control tumors had reached nearly 400 mm³ while the treated tumors were just over 200 mm³ (see Figure 6) . By day 25 one of the treated tumors had begun to regress and it disappeared several days later. None of the control animals' tumors regressed and all continued to grow rapidly. Although only one of eight tumors regressed in this experiment compared to three out of five in the previous, it should be noted that these tumor grew faster, and the animals were much younger. Despite the variability between the two groups, the difference in tumor volumes from days 5-39 was nearly significant (p = 0.055; ANOVA) .

Although the animals used in this experiment were fully grown, both groups gained weight nearly equally during the 39 day experiment. The control animals gained 4.0 g on average while the experimentals gained 3.6 g. Both groups of animals appeared very healthy and active throughout the experiment and none lost any weight. In this experiment, phosphorylcyclocreatine behaved as a tumor growth inhibitor while causing no apparent signs of toxicity.

Table 2

Mean weight change in athymic mice bearing DU-145 cell line-derived tumors and treated with either CCr^'P or saline. The final weight measurement was taken on day 39

EXPERIMENT III

Animals

Athymic (CD-Nu/Nu) male mice were obtained from Charles River Breeding Labs at an age of 4-6 weeks and delivered to the University of Massachusetts Animal Medicine Facility. Mice were allowed to grow and acclimate for 2 weeks before injection of tumor cells. To identify each mouse, a small hole was punched at a specific position of the outermost part of the right or left ear. The average weight of the mice at the beginning of the experiment was 27.6 grams. Mice were housed as described above. For this experiment, 8 animals were used, 4 treated and 4 placebo treated.

Cells

DU 145 cells were cultured as described above and 2.5 x 10⁶ cells in 0.2 is phosphate-buffered saline were injected subcutaneously over the proximal right femur of each mouse. Injections were performed with a 1 cc syringe and a 23 g needle. Tumors were measured as described above.

Drug Administration N-phosphorylcyclocreatine was prepared as a disodium salt (trihydrate) as described in Example 4. Lot #AM 709 was used. Eight days after the cells were injected, all animals had established palpable tumors (over 50 mm³) and the treatment was initiated. One group of 4 animals received 0.3 mis twice daily of CCr^'P at 50 mg/ml (1.0 g/kg/day) in a water vehicle pH 7.0, while the other group of 4 animals received saline pH 7.0. All injections were performed intraperitoneally, once in the morning at 9:00, and once in the evening at 5:00, six days per weeks. On the seventh day the animals were not manipulated.

Results

Human prostate tumors were initiated in 8 athymic male mice, and allowed to grow for 8 days until well established (> 50 mm³) . When CCr^'P was administered i.p. to 4 animals bearing DU145 dell-derived human prostate tumors, the treated tumors grew more slowly. The mean starting tumor volume was 67.0 mm³ for the CCr^'P-treated animals, 55.6 mm³ for the controls. On day 25, the mean control tumors had reached an average size of 343.8 mm³ while the treated tumors were 186.0 mm³ (Figure 7). By day 13 one of the treated tumors had begun to regress and it disappeared on day 19. None of the control animals' tumors regressed in this experiment, and all continued to grow rapidly. Only one of four tumors regressed in this experiment, but it is consistent with the previous two experiments in that for only treated animals do tumors actually disappear. Although the difference in tumor growth between the two groups was not quite significant (p = 0.08; ANOVA) , note how the treated tumors started larger and ended up considerably smaller. These young animals gained more weight when treated with i.p. test drug than with i.p. saline. While the controls gained 3.1 grams on average, the treated animals gained 6.7 grams over 25 days (see Table 3). None of the animals displayed any abnormal behavior. All mice in each group were active and appeared healthy during the 25 day experiment.

TABLE 3

Mean weight change in athymic mice bearing established DU- 145 cell line derived tumors and treated with either CCr'P or saline. The final weight measurement was taken on day 25, the first on day 0.

EXAMPLE 3

Phosphorylcyclocreatine Inhibits the Growth of

Human Cervical Tumor Cells in a Xenograft Model

INTRAPERITQNEAL ADMINISTRATION Cells

The ME180 human cervical carcinoma was obtained from the American Type Culture Collection (ATCC) Rockville, MD and grown according to the protocol suggested by the ATCC Catalogue of Cell lines and Hybridomas (6th edition, Hay et al. eds., 1988). ME180 cells were derived from a metastasis to omentum of a human epidermoid carcinoma of the cervix that originated from a 66 year old Caucasian female with a primary cervical carcinoma that progressed rapidly (Sykes et al. , 1970) . The cells were passaged twice weekly for 17 months before injection at a subcultivation ratio of 1:5, 1:10 by trypsinization. Medium consisted of 90% Minimum Essential Medium (Eagle's with Earle's Balanced Salts) without L- glutamine, J.R.H. Biosciences 51-412, EMEM) , 10% fetal bovine serum, 100 U/ml, Penicillin/Streptomycin, 2 mM L- glutamine, 1 mM Na pyruvate (J.R.H. Biosciences) . Cells were grown in 75 cm² flasks at 5% C0₂, 37°C.

To harvest sufficient cells for the experiment, 12 225 cm² about 75% confluent flasks of ME-180 cells were treated with 6 is of 0.25% trypsin until the cells detached from the flask. Trypsin was inactivated with 10 mis of MEM medium + 10% FBS. Cells were collected by centrifugation for 5 minutes at 1500 rpm. The supernatant was removed by aspiration and the pellet was resuspended in 20 ml of plain (no serum) media. The cell suspension was centrifuged again and the pellet was resuspended in 20 mis of serum free MEM media. A cell count was performed to determine cell number and viability. 2.0 x 10⁶ cells were injected subcutaneously over the proximal right femur of each mouse. Injections were performed with a 1 cc syringe and a 21 g needle. Tumors were measured as described above.

Animals

Athymic (BalbC-Nu/Nu) female mice were obtained from Jackson Breeding Labs at an age of 5-7 weeks and delivered to the University of Massachusetts Animal Medicine Facility. Mice were allowed to grow and acclimate for 3 weeks before injection of tumor cells. To identify each mouse, a small hole was punched at a specific position of the outermost part of the right or left ear. Altogether 8 mice were used, 4 control and 4 treated. The average weight of the mice at the beginning of the experiment was 21.7 grams. Mice were housed at the II Biotech Park Animal facility as described above. Druq Administration

N-phosphorylcyclocreatine (CCr^'P) was prepared as a disodium salt (trihydrate) as described in Example 4. Lot #AM-709 was used. CCr'P was prepared fresh daily in deionized water pH 7.0. Saline was not used as the drug vehicle because the compound is synthesized as a disodium salt, therefore a 50 mg/ml solution contains nearly 0.6% sodium. The control animals were injected intraperitoneally with 0.9% buffered saline (pH 7.0). Just prior to injecting animals, the solution was filtered under a laminar flow hood using a Corning 115 ml micropore filter system (0.2μ) to assure sterility. Injections were given to each experimental animal twice daily, at 9 a.m. and 5 p.m. , for six days each week. All manipulations were performed under a microisolator hood. A 26 3/4 gauge needle was used with a lcc syringe to administer the solution and injections were made into one of the tail veins. Each experimental animal received 0.2 ml i.v. of drug (75 mg/ml) each injection. The control mice received 0.2 ml of saline i.p. Animals were treated during days 7- 28 following injection of tumor cells.

Results

Human cervical carcinomas were initiated in 8 athymic female mice. Seven days following injection of the ME-180 cells, the mean tumor volume in the treated group (N=4) and in the 4 controls was 9 mm³. Treatment began on the seventh day and continued until the 28th.- As can be seen in Figure 8, tumors in the treated group reached a mean of only 242.6 mm³ by day 28, whereas mean tumor size in the control group increased to 591.6 mm³. The difference in growth rates between the two groups was significant (p = 0.002; ANOVA) . Animal weights in both groups increased steadily during the experiment. The control animals gained an average of 2.0 g, and weights of those on the experimental drug increased by 1.3 g. Table 4 below shows relative weight gain in each group. All of the animals appeared healthy, active and did not display any abnormal behavior throughout the experiment. Thus based upon the results from this experiment, it appears that cCr-P acts as an anti-tumor agent in this model without any notable signs of toxicity.

TABLE 4

Mean weight change in athymic mice bearing ME-180 cell line derived tumors and treated with either CCr'P or saline. The data reflects 22 days of measurements where the last weight measurement was taken on day 28.

INTRAVENOUS ADMINISTRATION

Animals

Athymic (BalbC-Nu/Nu) female mice were obtained from Jackson Breeding Labs at an age of 5-7 weeks and delivered to the University of Massachusetts Animal Medicine Facility. Mice were allowed to grow and acclimate for 2 weeks before injection of tumor cells. To identify each ouse, a small hole was punched at a specific position of the outermost part of the right or left ear. The average weight of the mice at the beginning of the experiment was 21.6 grams. Mice were housed as described above.

Cells

The ME-180 cervical carcinoma cell line was obtained and maintained as described above. In each of the 8 animals, 2.0 x 10⁶ cells were injected subcutaneously over the proximal right femur. Injections were performed with a 1 cc syringe and a 21 g needle. Tumors were measured with calipers as described above.

Drug Administration

N-phosphorylcyclocreatine (CCr'P) was prepared as a disodium salt (trihydrate) as described in Example 4. Lot #AM 709 was used. N-Phosphorylcyclocreatine at 75 mg/ml was prepared fresh daily in deionized water pH 7.0. Saline was not used as the drug vehicle because the compound is prepared as a disodium salt, therefore a 75 mg/ml solution contains nearly 0.9% sodium. The control animals were injected intravenously with 0.9% buffered saline (ph 7.0). Just prior to injecting animals, the solution was filtered in a laminar flow hood using a Corning 115 ml micropore filter system (0.2μ) to assure sterility. Intravenous injections were given to each experimental animal twice daily, at 9 a.m. and 5 p.m. , for five days. All manipulations were performed under a hood. A 30 gauge needle was used with a lcc syringe to administer the solution and injections were made into one of the tail veins. Each experimental animal received 0.2 ml i.v. of drug (75 mg/ml) each injection. The control mice received 0.2 ml of saline i.v. Animals were treated during days 4-8 following injection of tumor cells. Results

Human cervical carcinomas were initiated in 8 athymic female mice. Four days following injection of the ME-180 cells, the mean tumor volume in the 4 mice in the treated group and that of the 4 controls was 7 mm³ (all animals had palpable tumors) . Treatment began on the fourth day and continued until the eighth. Tumor measurements continued to be recorded after the drug administration ended. As shown in Figure 9, tumors in the treated group had barely reached 26 mm³ by day 8, whereas mean tumor size in the control group increased to about 70 mm³.

Animal weights in both groups did not change significantly over the period monitored. The control animals lost an average of 0.5 g, those on the experimental drug increased by 0.2 g. Table 5 (below) shows relative weight gain in each group. The changes are slight probably because of the short time period. All of the animals appeared healthy, active and did not display any abnormal behavior throughout the experiment. Thus based upon the preliminary results from this experiment, it appears that CCr^'P acts as an anti-tumor agent in this model without any notable signs of toxicity.

TABLE 5

Mean weight change in athymic mice bearing ME-180 cell line derived tumors and treated with either CCr'P or saline. The data reflects 5 days of measurements where the last weight measurement was taken on day 8.

EXAMPLE 4 Preparation of Disodium l-Carboxymethyl-2-imino- 3-Phosphonoimidazolidine Trihydrate fNa₂PCC*3H₂0)

The preparative route used was:

too exchange chromatography

( 1) Li₂PCC» 2H₂0 > (NH₄) ₂PCC

(2 ) (NH₄) ₂PCC + 2 NaOH Na,PCC* 3H,0

(I) Preparation of Lot # AM720 (Large scale)

(A) Preparation of dilithium l-carboxymethyl-2-imino-3- phosphonoimidazolidine dihydrate (Li₂PCC«2H₂0)

The dilithium salt of l-carboxymethyl-2-imino-3- phosphonoimidazolidine (Li₂PCC^«2H₂0) was prepared by the method of T.M. Annesley and J.B. Walker (Biochem. Biophvs Res. Commun. 74, 185-190 (1977)). (B) Preparation of diammonium l-carboxymethyl-2-imino-3- phosphonoimidazolidine ( (NH₄)₂PCC)

The dilithium salt of part (A) was converted to the diammonium salt by ion-exchange chromatography. The column (8.3 cm in diameter with a bed height of 35.5 cm) contained 2 L of Dowex AG 1-X8 anion exchange resin. The resin was a mixture of mesh sizes (50-100, 100-200, and 200-400) and had been generated into the bicarbonate form by treatment with 1 M aqueous ammonium bicarbonate (10.5 L in total) followed by treatment with deionized water (ca. 5 L in total) . The resin was then packed and pre-washed with 2 L of 0.025 M NH₄HC0₃ solution.

A mixture of Li₂PCC (80.1 g, 0.295 mol), 200 mL of 0.025 M NH₄HC0₃, and approximately 200 mL of deionized water was applied to the ion-exchange column. The column was washed successively with 0.025 M (0.94 L) , 0.10 M (1 L) 0.25 M (1 L) , 0.40 M (1 L) , 0.5 M (0.70 L) , and 0.65 M (2 L) NH₄HC0₃. The product, identified using polyethylenimine (PEI) cellulose thin-layer chromatography (refer to G.L. Rowley and G.L. Kenyon, Anal. Biochem. , 58, 525-533 (1974) ; plates were obtained from J.T. Baker Inc.), was eluted at buffer concentrations of 0.25 M to 0.65 M. The fractions containing the product were pooled and evaporated to dryness at temperatures less than 40°C under reduced pressure. During the final stages of solvent evaporation it was necessary to add several drops of concentrated aqueous ammonia to increase the pH of the product mixture to ca 7.6. Residual ammonium bicarbonate was removed by evaporation of methanol (2 x ca 100 mL) . The white residue was dried over sodium hydroxide in vacuo. 55.7 g (80% yield) of (NH₄)₂PCC were obtained. Product purity determined by HPLC to be 94.6%: A Regis reverse-phase fully endcapped VAL-U-PAK HP column (5 micron o.d.), 0.03M K₂HP0₃ (pH 7.5) eluent with a flow rate of 0.5 mL/ in, and U.V. detection at 210 nM were employed for the HPLC analysis; the retention time of the product was 4.6 to 4.7 min.

(C) Preparation of disodium l-carboxymethyl-2-imino-3- phosphonoimidazolidine trihydrate (Na₂PCC«3H₂0)

Diammonium l-carboxymethyl-2-imino-3- phosphonoimidazolidine (55.3 g, approximately 0.215 mol, water content not determined) , 3 M aqueous sodium hydroxide (148 mL, 0.444 mol) and deionized water (548 mL) were stirred briefly. The resulting solution was chilled on an ice/salt bath to a temperature of 8°C, 6 M aqueous hydrochloric acid (68 mL) was added over 2.5 hour period to the efficiently stirred cold solution, such that a final pH of 7.0 to 7.2 was attained. The pH was then increased to between 7.2 and 7.6 by addition of 3M NaOH and the solution was filtered to remove any particulate matter. The product was crystallized from the filtrate as follows: 2.25 volumes of absolute ethanol (1.8 L) followed by a sufficient amount of 3M NaOH to increase the pH to between 7.2 and 7.6 were added and the mixture cooled in a -40°C freezer. Seed crystals were then added to the cold solution so as to produce a cloudy mixture. The crystallization solution was then transferred .to a -27°C freezer. Over a four day period the solution remained at -27°C and a further 0.6 volumes of absolute ethanol (0.45 L) were added. The mixture was then allowed to stand at room temperature for 1.5 hours and the crystals were filtered, washed with HPLC grade isopropanol (90 mL) and dried in vacuo to constant weight, 42.4 g (61% yield) of Na₂PCC^#3H₂0 were obtained. Purity determined by HPLC (as above) was 99.5%. Phosphate content (w/w % as sodium monobasic phosphate or as sodium dibasic phosphate) was established to be less than 0.5% (but greater than 0.1%) by precipitation with 0.2 M silver nitrate and comparison to similarly treated standard solutions which contained 0.1%, 0.5%, 1.0% and 2.5% phosphate, ^!H NMR (D₂0, 300 MHz) δ 3.75 (s, 2H) , 3.65 (m, 2H) , 3.50 (m, 2H) . The product was designated lot # AM720. The *H NMR, HPLC profile, and UV spectrum for lot ≠ AM 720 are shown in Figures 10-12, respectively. A second and third crop of the product (10.1 g, 15% yield) were obtained by the addition of a total of 3.75 volumes of absolute ethanol (3 L) to the cold crystallization filtrate and sufficient 3 M NaOH to maintain a pH between 7.2 and 7.6. HPLC analysis indicated a purity of 99.0%. No free phosphate or chloride ions were present as determined by a silver nitrate test.

Total yield of Na₂PCC«3H₂0 was approximately 76% (52.5 g) .

(II) Preparation of Lot # AM709 (Small scale)

Preparation of disodium l-carboxymethyl-2-imino-3- phosphonoimidazolidine trihydrate (Na₂PCC*3H₂0)

Batch A

A mixture of diammonium l-carboxymethyl-2-imino-3- phosphonoimidazolidine (12.3 g, approximately 47.8 mmol, water content not determined) , 33 mL of 3 M aqueous sodium hydroxide (99 mmol) and 130 mL of deionized water was chilled on an ice-salt bath. A total of 13 mL of 6 M aqueous hydrochloric acid was added dropwise to the stirred solution such that a pH of ca 7.6 was attained. Addition of 1 L of absolute ethanol and sufficient 3 M NaOH to maintain a pH of 7.6 followed by cooling on an ice/salt bath produced an oil. The liquor was decanted from the oil, seeded with crystals, the flask walls scratched, and then allowed to warm to room temperature. Over this period the cloudy appearance dissipated and crystals formed. The crystals (3.0 g) were then filtered. A portion of the crystallization filtrate was chilled, and then added to the oil which was previously removed. This caused the oil to solidify. The solid was transferred to a filter funnel and washed with HPLC grade isopropanol;

4.0 g were obtained. The crystals and solid were combined and dried in vacuo. yielding 45% Na₂PCC»3H₂0. The sample tested negative for phosphate or chloride ions.

Batch B A mixture of diammonium l-carboxymethyl-2-imino-3- phosphonoimidazolidine (3.0 g, approximately 11.7 mmol, water content not determined) , 3 M aqueous sodium hydroxide (8 mL, 24 mmol) and deionized water (30 mL) was chilled on an ice/salt bath. 6 M aqueous hydrochloric acid was added dropwise to the stirred solution until a pH of ca. 7.6 was attained. After filtration 200 mL of absolute ethanol, and then sufficient 3 M NaOH so as to maintain a pH between 7.2 and 7.6, were added. The mixture was cooled on an ice/salt bath, seed crystals were added, and after 0.5 h an oil was visible. The liquor was decanted from the oil, warmed to room temperature,and the crystals present (1.2 g) were then filtered. A portion of the crystallization filtrate was added to the oil which caused the oil to whiten and then solidify. The solid was transferred to a filter funnel and washed with HPLC grade isopropanol; 2.5 g were obtained. The crystals and solid were combined and dried in vacuo, yielding 81% Na₂PCC*3H₂0. The sample tested negative for phosphate or chloride ions. Lot ≠ AM709 The products from Batch A and B were combined. Purity was determined by HPLC analysis to be 99.6%.

Anal. Calcd. for C_jH_gN₃0jPNa₂»3H₂0: C, 18.70; H, 4.39; N, 13.08. Found: C, 18.62; H, 4.30; N, 12.88.

Eguivalents

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

Claims

1. A method of inhibiting the growth of a tumor in a mammal comprising administering to the mammal a therapeutically effective amount of N-phosphoryl creatine or an analog thereof.

2. A method of inhibiting growth of a tumor in a mammal comprising administering to the mammal a therapeutically effective amount of a compound selected from the group consisting of: N- phosphorylcreatine and analogs of N- phosphorylcreatine, wherein the compound is specified by the general formula:

and pharmaceutically acceptable salts thereof, wherein:

(a) Y is selected from the group consisting of: -C0₂H, -NH0H, -N0₂, -S0₃H, -P(=0) (OH) (OJ) , and -P(=0) (OJ) (H) , wherein J is selected from the group consisting of: hydrogen, C--C₆ straight chain alkyl, C₃-C₆ branched alkyl, C₂-C₆ straight alkenyl, C₃-C₆ branched alkenyl and aryl;

(b) A is selected from the group consisting of: C_j- C_s alkyl, C_-C₃ alkenyl, C₂-C₅ alkynyl, and C_:-C₅ alkoyl, each having 0-2 bulky and/or hydrophobic substitutents which are selected independently from the group consisting of: (1) K, where K is selected from the group consisting of: C.-C₆ straight alkyl, C^-Cg straight alkenyl, C_j-C₆ straight alkoyl, C₃- C₆ branched alkyl, C₃-C₆ branched alkenyl, and C₄-C₆ branched alkoyl, K having 0-2 substituents that could react with an enzymic nucleophile independently selected from the group consisting of: bromo, chloro, epoxy and acetoxy; (2) an aryl group selected from the group consisting of: a 1-2 ring carbocycle and a 1-2 ring heterocycle, wherein the aryl group contains 0-2 substitutents that could react with an enzymic nucleophile independently selected from the group consisting of: -CH₂L and -COCH₂L where L is a leaving group independently selected from the group consisting of: bromo, chloro, epoxy and acetoxy; and (3) -NH-M, wherein M is selected from the group consisting of: hydrogen, C_-C₄ alkyl, C₂-C₄ alkenyl, C,-C₄ alkoyl, C₃-C_A branched alkyl, C₃-C₄ branched alkenyl, and C₄ branched alkoyl;

(c) X is selected from the group consisting of:

NR,, CHR-, C, alkenyl with an R, substituent, O and S, wherein R_ is selected from the group consisting of: (1) hydrogen; (2) K where K is selected from the group consisting of: C,-C₆ straight alkyl, C₂-C₆ straight alkenyl, C,-C₆ straight alkoyl, C₃- C₆ branched alkyl, C₃-C₆ branched alkenyl, and C₄-C₆ branched alkoyl, K having 0-2 substitutents that could react with an enzymic nucleophile independently selected from the group consisting of: bromo, chloro, epoxy and acetoxy;

(3) an aryl group selected from the group consisting of: a 1-2 ring carbocycle and a 1-2 ring heterocycle, wherein the aryl group contains 0-2 substitutents that could react with an enzymic nucleophile independently selected from the group consisting of: -CH₂L and -COCH₂L where L is a leaving group independently selected from the group consisting of: bromo, chloro, epoxy and acetoxy;

(4) a C₅-C₉ α-amino-fo-methyl-fo- adenosylcarboxylic acid attached via the Z>- methyl carbon;

(5) a C₅-C₉ α-amino-fo-aza-fo-methyl-fo- adenosylcarboxylic acid attached via the fo-methyl carbon; and

(6) a C₃-C₉ cr-amino-<_>-thia-£)-methyl-S)- adenosylcarboxylic acid attached via the

&>-methyl carbon;

(d) If A is chosen as a C, alkenyl group, then X must also be chosen as an alkenyl group, and if X is chosen as a C, alkenyl group, then A must be an alkenyl group, wherein A and X are connected by a double bond;

(e) Z, and Z₂ are chosen independently from the group consisting of: =0, -NR₂R₃, -CHR₂R₃, and -NR₂OR₃; wherein, Z, and Z₂ may not both be =0 and wherein for a selected Z group, R₂ and R₃ are independently selected from the group consisting of:

(1) hydrogen; 5 (2) K, where K is selected from the group consisting of: C,-C₆ straight alkyl, C₂-C₆ straight alkenyl, C,-^ straight alkoyl, C₃-C₆ branched alkyl, C₃-C₆ branched alkenyl, and C₄-C₆ branched alkoyl, K having 10 0-2 substitutents that could react with an enzymic nucleophile independently selected from the group consisting of: bromo, chloro, epoxy and acetoxy;

(3) an aryl group selected from the group

15 consisting of: a 1-2 ring carbocycle and a

1-2 ring heterocycle, wherein the aryl group contains 0-2 substitutents that could react with an enzymic nucleophile independently selected from the group

20 consisting of: -CH₂L and -COCH₂L where L is a leaving group independently selected from the group consisting of: bromo, chloro, epoxy and acetoxy;

(4) a C₄-C₈ α-amino-carboxylic acid attached via 25 the fo-carbon;

(5) B, wherein B is selected from the group consisting of: -C0₂H, -C(=0) (NHOH) , -NHOH, -S0₃H, -N0₂, -OP(=0) (OH) (OJ) ,

-OP(=S) (OH) (OJ) , -0P(=0) (H) (OJ) , 30 -OP(=S) (H) (OJ) , -P(=0) (OH) (OJ) ,

-P(=S) (OH) (OJ) , -P(=0) (H) (OJ) and -P(=S) (H) (OJ) , wherein J is selected from the group consisting of: hydrogen, C_-C₆ straight alkyl, C₃-C₆ branched alkyl, C₂-C₆ straight alkenyl, C₃-C₆ branched alkenyl and aryl; wherein B is optionally connected via a linker selected from the group consisting of: C--C₂ alkyl, C₂ alkenyl, and Cj-Cz 5 alkoyl;

(6) -D-E, wherein D is selected from the group consisting of: C,-C₃ straight alkyl, C₃ branched alkyl,

straight alkenyl, C₃ branched alkenyl, C--C₃ straight alkoyl, 10 aryl, and aroyl; and wherein E is selected from the group consisting of:

-(P0₃)_nNMP, wherein n is 0-2, NMP is a ribonucleotide monophosphate connected via 15 the 5'-phosphate, 3'-phosphate or the aromatic ring of the base, and (=S) may be substituted for one or more (=0) in E;

-[P(=0) (0CH₃) (0) ]_m-Q, where m is 0-3 and Q is a ribonucleoside connected via the

20 ribose or the aromatic ring of the base and

(=S) may be substituted for one or more (=0) in E;

~[P(=0) (OH) (CH₂) ]_m-Q, wherein m is 0- 3, Q is a ribonucleoside connected via the 25 ribose or the aromatic ring of the base, and (=S) may be substituted for one or more (=0) in E;

-[P(=0) (H) (CH₂) ]_m-Q, wherein m is 0-3, .. Q is a ribonucleoside connected via the

30 ribose or the aromatic ring of the base and

(=S) may be substituted for one or more (=0) in E; and an aryl group containing 0-3 substituents chosen independently from the group consisting of: Cl, Br, epoxy, acetoxy, -0G, -C(=0)G, and -C0₂G, where G is independently selected from the group consisting of: C_-C₆ 5 straight alkyl, C^-C_β straight alkenyl, C,-C₆ straight alkoyl, C₃-C₆ branched alkyl, C₃-C₆ branched alkenyl, C₄-C₆ branched alkoyl; wherein E may be attached at any point to D, and

10 if D is alkyl or alkenyl, D may be connected at either or both ends by an amide linkage; and (7) -E, wherein E is selected from the group consisting of:

15 -(P0₃)_nNMP, wherein n is 0-2, NMP is a ribonucleotide monophosphate connected via the 5'-phosphate, 3'-phosphate or the aromatic ring of the base, and (=S) may be substituted for one or more (=0) in E;

20 _[p(=o) (0CH₃) (0) ]_m-Q, where is 0-3 and Q is a ribonucleoside connected via the ribose or the aromatic ring of the base and (=S) may be substituted for one or more (=0) in E;

25 -[P(=0) (OH) (CH₂) ]_m-Q, wherein m is 0-

3, Q is a ribonucleoside connected _.via the ribose or the aromatic ring of the base, and (=S) may be substituted for one or more. (=0) in E;

30 -[P(=0) (H) (CH₂) ]_m-Q, wherein m is 0-3,

Q is a ribonucleoside connected via the ribose or the aromatic ring of the base and (=S) may be substituted for one or more (=0) in E; and an aryl group containing 0-3 substituents chosen independently from the group consisting of: Cl, Br, epoxy, acetoxy, -0G, -C(=0)G, and -C0₂G, where G is independently selected from the group consisting of: C_j-C₆ straight alkyl, C^-Cg straight alkenyl, C,-C₆ straight alkoyl, C₃- C₆ branched alkyl, C₃-C₆ branched alkenyl, C₄-C₆ branched alkoyl; and if E is aryl, E may be connected by an amide linkage;

(f) If R, and at least one R₂ or R₃ group are present, R, may be connected by a single or double bond to an R₂ or R₃ group to form a cycle of 5 to 7 members;

3. The method of Claim 2 wherein the pharmaceutically acceptable salt is a sodium salt.

4. A method of inhibiting growth of a tumor in a mammal, comprising administering to the mammal a therapeutically effective amount of a composition comprising a compound selected from the group consisting of:

(a) N-phosphorylcreatine;

(b) N-phosphorylcylcocreatine; (c) N-phosphorylhomocyclocreatine;

(d) N-phosphoryl-/S-guanidinoproprionate;

(e) N-phosphoryl-l-carboxymethyl-2- iminohexahydropyrimidine; (f) N-phosphorylguanidinoacetate; (g) N-phosphorylcarbocreatine; and (h) N-phosphoryl-l-carboxymethyl-2- iminoimidazolidine-4-one

5. The method of Claim 4 wherein the composition further comprises a pharmaceutically acceptable carrier.

6. A method of inhibiting growth of a tumor cell comprising contacting said cell with N-phosphoryl creatine or an analog thereof.

7. A composition comprising N-phosphorylcyclocreatine disodium trihydrate.

8. The composition of Claim 7 further comprising a pharmaceutically acceptable carrier.