RELATED APPLICATIONS

This application is a continuation-in-part of application Serial No. 216,379 filed July 8, 1988; application Serial No. 191,986 filed May 9, 1988 and application Serial No. 135,073 filed December 18, 1987.

Each of these three related applications is incorporated herein by reference.

BACKGROUND

Excessive accumulation of lipids in fat cells is a major feature of human and animal obesity. Fat cell lipogenesis is stimulated by insulin, phorbol esters, and in certain conditions by growth hormone.

The insulin receptor tyrosyl kinase has been shown to be important in the actions of insulin. In addition, it has recently been shown that kinase C may be involved in the stimulatory effects of insulin and growth hormone on rat adipocytes.i/ The phorbol ester 12-myristate 13-acetate, an activator of kinase C, stimulates lipogenesis. Its effect is not additive to the effects of insulin or growth hormone. Downregulation of kinase C markedly inhibits the effects of both hormones. -W

Accordingly an investigation was undertaken to evaluate the effect of protein kinase antagonists on glucose uptake which is the first step in the complex lipogenesis metabolic pathway.

__/ J. Smal and P. DeMeyts, Biochem. Biophys. Res. Commun. 147:1232 (1987).

2/ Ibid.

SUMMARY

This invention involves the discovery that protein kinase antagonists, including antagonists of protein kinase C and of the insulin receptor tyrosine kinase, block or completely suppress normal and basal animal fat cell lipogenesis. The degree of lipogenesis inhibition is a function of the effective amount of antagonist available to the fat cells. Hormone-stimulated lipogenesis is reversibly blocked at relatively lower concentrations whereas at higher concentrations basal lipogenesis is also completely suppressed.

Another aspect of the invention entails the administration of a protein kinase antagonist in an amount therapeutically effective to inhibit lipogenesis in a mammal, e.g., a human.

Preferred protein kinase antagonists include lysophingolipids, staurosporine and aminoacridines.

Formula I includes some of the lysosphingolipids useful in the invention:

I. CH ₃ (CH ₂ )12 CH=CH—-CH—CH—CH ₂ 0—

OH NH ₂ wherein X may be, e.g., H (sp ingσsine) , galactose (psychosine) , sulfogalactose,

Gal-Gal NAc

Gal-Glc- (lyso GM _] Sia Gal-NAc

Gal-Glc- (lyso GM ₂ ) Sia Sia-Gal-Glc- (lyso GM3) and derivatives thereof.

For ula II illustrates the molecular structure of sphingosine, the preferred compound for use in this embodiment of the invention:

Staurosporine, available from Kyowa Hakko Kogyo Co., Tokyo, Japan, is a microbial alkaloid having the formula III:

Aminoacridines constitute an additional group of kinase C inhibitors useful in the practice of this invention. This embodiment of the invention specifically includes, among other aminoacridines, the compounds shown by Hannun and Bell, J. Biol. Chem. 263:5124-5131 (1988), to be protein kinase C inhibitors. Acridine orange is preferred and has been shown to block both basal and insulin-stimulated lipogenesis in rat adipocytes. Inhibition is maximum at 10 _f M.

EXEMPLIFICATION OF THE INVENTION This invention involves the discovery that protein kinase antagonists, preferably sphingolipids and (lyso)sphingoli ids, staurosporine and aminoacridines inhibit or completely suppress lipogenesis in the fat cells of mammals. The degree of inhibition is a function of the effective amount of protein kinase antagonist, e.g., sphingolipid or (lyso)sphingolipid or staurosporine or aminoacridine in or available to the fat cells.

More particularly, it has been discovered that the stimulation of fat cell lipogenesis by insulin, growth hormone and phorbol esters is blocked reversibly by the action of such protein kinase antagonists at an effective but relatively lower concentration, whereas at higher concentrations basal lipogenesis (lipogenesis in the absence of hormones) is also completely suppressed.

EXAMPLE I Inhibition of Lipogenesis By Sphinqosine Figures 1A and IB illustrate the effect of sphingosine on the maximal stimulation of lipogenesis in rat adipocytes by phorbol ester (PMA) , human growth hormone (hGH) and insulin.

The incorporation of the tracer D-3[ ³ H]-glucose into lipids is expressed in counts per minute of radioactivity (= cpm) per tube. It is the mean of a triplicate measurement, plus or minus one standard deviation (1 S.D.). The values obtained have been divided by 1,000 to simplify the graph. Thus, a value of 3 on the Figure 1A graph means that 3,000 cpm were actually measured.

Cells were incubated without hormone (basal— hatched bars) or with respectively 100 ng/ml PMA, 1000 ng/ml hGH or 100 ng/ml insulin in the absence (bars 0) or in the presence (bars 50) of 50 μM of sphingosine (combined hatched and white bars) .

At 100 μ sphingosine both basal and hormone stimulated lipogenesis are reduced to zero.

The lipogenesis assay has been conducted accord¬ ing to Smal, et al. J. Biol. Chem. 262:11071-11079 (1987) . Briefly, dissected epididymal and retro- peritoneal fat pads were digested under vigorous shaking at 37°C for 30 min with collagenase (l.o mg/ml) in Krebs-Ringer-Hepes (KRH) buffer, pH 7.4, 35 mg/ml dialyzed bovine serum albu ine (BSA), 0.27 mM glucose. After filtration on cheesecloth and 4 washes in KRH with 10 mg/ml BSA, the adipocytes were preincu- bated for 4 hours at 37°C in the same buffer. Sphingosine (100 mM stock solution in ethanol) was added to the cells (final concentration:50 μM in Fig. 1A; 0 to 100 μM in Fig. IB) 10 min before the lipogenesis assay. The same concentration of ethanol, without sphingosine, was added to the control cells.

The lipogenesis assay was performed in triplicate in 6 ml polyethylene vials by adding successively 400 μl of adipocyte suspension (8 X 10 ⁴ cells/ml) , 50 μl of KRH buffer pH 7.4 (1% BSA, 0.27 mM glucose) without hormone (basal lipogenesis) or with human growth hormone (1000 ng/ml) , insulin (100 ng/ml) or phorbol ester PMA (100 ng/ml) and 50 μl D-[3- ³ H]-glucose in a total volume of 0.5 ml. The vials were incubated 2 hours at 37°C under gentle shaking. The incubation was interrupted by adding 5 ml/tube of toluene scintillator (1 liter toluene + 0.3 g of l,4bis[2-

(4-methyl-5 phenyloxazolyl) benzene and 5 g of 2,5-diphenyloxazole under vigorous shaking (30s to break the cells) followed by a rest of at least 1 hour to allow extraction of lipids into the toluene phase before counting. The samples were counted in a Beckman LS 1880 beta counter. The counting efficiencies for the different samples were measured by internal standardization with quenched tritiated standards. The incorporation of D-[3- ³ H]-glucose into lipids is expressed in cpm X 10~ ³ /tube + 1 standard deviation.

Figure 1A shows that 50 μM sphingosine completely blocks the enhancement of lipogenesis by PMA, and markedly decreases the effect of hGH (by 65%) and insulin (by 89%) .

Figure IB shows that the dose-response curve of the effects of sphingosine is in the range described elsewhere for the inhibition of kinase C ^n vitro as well as in vivo on different cell types. ³ / It also shows that at 100 μM sphingosine also completely abolishes basal lipogenesis as well as the effect of insulin, PMA and hGH.

The viability of the adipocytes, assessed by the trypan blue exclusion method remained unchanged in the presence of 200 μM sphingosine. The effect of sphingosine on basal lipogenesis was totally reversible and that on stimulated lipogenesis partially reversible after washing the cells.

3/ Y.A. Hannun, C.R. Loo xs, A.H. Merril Jr. and R.M. Bell, J. Biol. Chem. 261:12504 (1986); E. Wilson, M.C. Olcott, R.M. Bell, A.H. Merril Jr. and J.D. Lambeth, J. Biol. Chem. 261:12616 (1986); Y.A. Hannun, CS. Greenberg and R.M. Bell, Ibid. 262:13620 (1987) .

EXAMPLE II Specificity of Lipogenesis Inhibition By Sphingosine The specificity of the inhibition of lipogenesis by sphingosine was tested by comparison with the effects of sphingomyelin. The critical structural features of lysosphingolipids required for specific inhibition of protein kinase C are a primary a ine group and the hydrophobiσ character of the long chain as shown by Formula I. As Figure 2 shows, sphingomyelin, which has an amide linked fatty acid at the 2-amino position of the sphingoid base, see Formula IV,

does not inhibit either basal lipogenesis or hormone- stimulated lipogenesis at concentrations up to 250 μM. Thus the inhibition of lipogenesis has the same specificity as kinase C inhibition.

To obtain the data reflected by Figure 2, lipogenesis was determined as in Figure IB, as explained in detail supra, not only in the presence of increasing concentrations of sphingosine (closed symbols) but also in the presence of increasing concentrations of sphingomyelin (open symbols) .

EXAMPLE III

Comparative Sphingosine Inhibition of Lipogenesis and Glucose Uptake

This example involves experiments conducted simultaneously on the same batch of rat adipocytes. Figure 3 shows that lipogenesis and glucose uptake

are inhibited to the same extent. The data imply that the observed sphingosine effect on lipogenesis impacts the glucose uptake step of the lipogenesis metabolic pathway.

To obtain the data reflected by Figure 3, lipo¬ genesis and glucose uptake were measured in parallel on the same batch of adipocytes in each experiment. The lipogenesis protocol is described in reference to Figure 1. For glucose uptake measurements, isolated adipocytes (16 x 10 ⁴ cells/ml) were incubated following the lipogenesis procedure with increasing concentration of sphingosine (0-100 μM) during 10 min. at 37°C prior to the addition of a saturating dose of insulin (100 ng/ml) , hGH (1 μg/ml) , or phorbol ester PMA (100 ng/ml) and 9 nM D-[3- ³ HJ-glucose for two hours at 37 ^β C. The incubation was interrupted by adding successively 3 ml of ice-cold 9 °/00 Na Cl and 1 ml of dinonylphtalate per tube. The tubes were immediately centrifuged at 4°C, 3000 RPM for five minutes. The floating cells were recovered with a 500 μl pipette fitted with a truncated tip, transferred in a scintillation vial containing 10 ml of universal scintillating fluid (Aqua-sol II, NEN Research Product, Boston, MA.) and thoroughly shaken to break the cells before counting. The results, expressed in % of the effect without sphingosine (control) , are plotted as a function of the sphingosine concentration used (number indicated under the bar graphs) . Black bars represent lipogenesis, white bars represent glucose uptake.

EXAMPLE IV

Sphingosine Inhibition Not Conseguent From Inhibition of Hormone-Receptor Binding

The binding of insulin and growth hormone to their receptors was studied as described in J. Smal, J. Closset, G. Hennen and P. DeMeyts, J. Biol. Chem. 62:11071 (1987). ¹²⁵ I-A14 y ^r -insulin and ¹²⁵ I-hGH were prepared as described respectively in J. Markussen and V.D. Larsen, in Insulin: Chemistry, Structure and Function of Insulin and Related Hormones, (D. Brandenburg and A. Wollmer, eds) W. DeGruyter, Berlin, p. 161 (1980) and J. Closset, J. S al, F. Gomez and G. Hennen, Biochem. J. 214:885 (1983) . Isolated adipocytes (8 x 10 ⁵ cells/ml for insulin binding; 16 x 10 ⁵ cells/ml for hGH binding) were preincubated in duplicate (Final volume: 0.5 ml/tube) in binding buffer (Krebs Ringer Hepes pH 7.4, 0.27 mM, 5% BSA (W/W) + 0.5 mg/ml bacitracin and 0.5 μg/ml aprotinin) with increasing concentration of sphingosine (0-200 μM) for ten minutes at 37°C before the addition of ^{1 5} I-insulin (30,000 cpm/tube corre¬ sponding to 30 pM final) or ¹²⁵ I-hGH (50,000 cpm/tube corresponding to 45 PM final) . The incubation time was 45 minutes at 37°C for insulin and 75 minutes at 37"C for hGH. Incubations were interrupted by adding successively 3 ml of ice-cold 9 °/00 NaCl and 1 ml dinonylphtalate per tube. The tubes were immediately centrifuged at 4°C, 3000 RPM for five minutes. The floating cells were recovered with a 500 μl pipette fitted with a truncated tip and the cell associated radioactivity counted (total binding) . Binding of ¹²⁵ I-insulin and ^{1 5} I-hGH in the presence of respectively 10 μg/ml of unlabeled insulin and hGH

(non-specific binding) was measured following the same procedure.

In Figures 4A and 4B the results, expressed as cpm x 10~ ³ /tube of tracer bound in the absence (total binding) or in the presence of 10 μg/ml of unlabeled hormone (N.S. binding) , are expressed as a function of increasing concentration of sphingosine (0-200 μM) .

As Figures 4A and 4B show, sphingosine had no significant effect on insulin binding at 200 μM (Fig. 4A) . In contrast, growth hormone binding to its receptor progressively decreased with increasing concentrations of sphingosine (Fig. 4A) . However, growth hormone specific binding did not decrease below 43% of its control value at a sphingosine concentration (100 μM) which inhibited lipogenesis completel .

EXAMPLE V

Fig. 5A illustrates the effect of staurosporine on the maximal stimulation of lipogenesis in isolated rat adipocytes by insulin. The incorporation of the tracer D-3[ ³ H]-glucose into lipids is expressed in counts per minute of radioactivity (= cpm) per tube. It is the mean of a triplicate measurement, plus or minus one standard derivation (1 S.D.). Cells were incubated without (basal curve) or with 100 ng/ml insulin (insulin curve) in the presence of increasing doses (0-21.4 μM) of staurosporine. Lipogenesis assays were conducted according to Smal, supra. In Fig. 5B, the results of the experiment shown on Fig. 5A are normalized: insulin- timulated lipogenesis and basal lipogenesis in the presence of increasing doses of staurosporine are expressed in % of their initial values without staurosporine. Stimulated lipogenesis in rat adipocytes are progressively

inhibited by staurosporine in a dose-dependent manner. Basal and insulin-stimulated lipogenesis are completely inhibited by staurosporine at about 2.5 micromolar and 25 micromolar respectively, whereas a 100 micromolar concentration of sphingosine is necessary to yield a comparable inhibitory effect.

EXAMPLE VI

Figures 6A and B illustrate the effect of staurosporine on the maximal stimulation of lipogenesis in rat adipocytes by insulin, phorbol ester, PMA and human growth hormone. Figures 6A and B illustrate the effect of staurosporine on the maximal stimulation of lipogenesis in rat adipocytes by insulin, phorbol ester, PMA and human growth hormone.

The experiments in Figures 6A and B are conducted in exactly the same way as in Figures 1A and B, as explained in detail supra, except that staurosporine, at the concentrations indicated under the horizontal axis (2.14 μM in Figure 6A, 0-21.4 μM in Figure 6B) was used instead of sphingosine. Figure 6A shows that at 2.14 μM, staurosporine almost completely abolishes the effects of PMA and hGH on lipogenesis, and markedly reduces the effect of insulin, without altering basal lipogenesis. Figure 6B shows the dose-response curve for the effect of staurosporine on basal as well as hGH-, PMA-, and insulin- stimulated lipogenesis. At 21.4 μM, staurosporine completely abolishes both basal and stimulated lipogenesis.

EXAMPLE VII

Figures 7A and 7B show, respectively, the percentage of inhibition of lipogenesis and the percentage of inhibition of phosphorylation resulting

from the effects of staurosporine. Figure 7A represents the same data as in Figure 6B, normalized; each curve of Figure 7B is reinforced as the percent of the inhibition of lipogenesis (100% being the lipogenesis measured in the absence of staurosporine, after subtraction of basal lipogenesis from the PMA-, hGH- or insulin-stimulated lipogenesis) , as a function of staurosporine concentration (0-21.4 μM) . Figure 7B shows the percent of the inhibition of kinase activity of purified kinase C and purified insulin receptor tyrosine kinase, as a function of staurosporine concentration (0-21.4 μM) . The kinase activity was measured by the ability of the purified kinases to phosphorylate a synthetic peptide, Arg-Arg-Leu-Ile-Glu-Asp-Ala-Glu-Tyr-Ala-Ala-Gly, as- explained in detail in Y. Fujita-Yamaguchi and S. Kathuria, Biochem. Biophys. Res. Commun. 157:955-962 (1988) .

Purification of insulin receptor kinase: Insulin receptor was purified 2400-fold from Triton x-100-solubilized human placental membranes by sequential affinity chromatography on WGA- and insulin-Sepharose as described in Y. Fujita-Yamaguchi, S. Chou, Y. Sakamoto, and K. Itakura, J. Biol. Chem. 258:5045-5049 (1983).

Tyrosine-specific protein kinase assay: Phosphorylation assays were carried out at 25 ^β C for 40 minutes in 30 μl of 50 mM Tris-Hcl buffer, pH 7.4, containing ImM of the src-related peptide, 2 mM MnCl ₂ , 15 mM MgCl ₂ , 0.1% Triton X-100, and 40 μM [ ₇ - ³ P]ATP (-12,000 cpm/pmole) and the purified insulin receptor (-0.1 μg) , which had been pre- incubated with varying concentrations of one of the inhibitors at 25 " C for 1 hour in the presence or

absence of 0.1 μM insulin or IGF-I. The reaction was terminated by the addition of 50 μl of 5% trichloroacetic acid and 20 μl of bovine serum albumin (19 mg/nl) . After incubating this solution at 0°C for 30 minutes, the proteins were precipitated by centrifugation. Duplicate 35 μl aliquots of the supernatant were spotted on pieces of phospho- cellulose paper (Whatman, P-81) . The papers were extensively washed in 75 mM phosphoric acid. Incorporation of ³² P into the peptide was quantitated by counting in a liquid scintillation counter. The kinase activity of purified protein kinase C was examined using histone HIS (Sigma) as a substrate under the conditions described in T. Akiyama, E. Nishida, J. Ishida, N. Soyi, H. Ogarawa, M. Hoshi, Y. Miyata, and S. Sakai, J. Biol. Chem. 261:15648-15651 (1986) . Comparison of Figures 7A and 7B shows that the dose-response curve for the inhibition of insulin-stimulated lipogenesis by staurosporine parallels the inhibition of insulin receptor tyrosine kinase by the drug, while the inhibition o f hGH and PMA stimulation parallels the inhibition of protein kinase C.

EXAMPLE VIII Figures 8A and 8B illustrate the effects of staurosporine on insulin and hGH binding to their receptors in isolated rat adipocytes. The experiments were conducted as in Figures 4A and 4B, except that staurosporine (0-21.4 μM) was used instead of sphingosine. The data show that staurosporine has little or no effect on hormone binding at concentrations which almost completely inhibit hormone-stimulated lipogenesis.

EXAMPLE IX

Figure 9 illustrates the effects of acridine orange (an active inhibitor of protein kinase C) and the lack of effect of 9-acridine carboxylic acid (inactive on protein kinase C) on basal lipogenesis as well as on lipogenesis stimulated by hGH, PMA, and insulin. The assay is explained in detail in Smal, J. , S. Kathuria and P. DeMeyts, FEBS Lett. 244:465-468 (1989). The experiments were conducted as explained in reference to Figure IB, except that acridine orange (closed symbols) or 9-acridine- carboxylic acid (open symbols) were used at 0-100 μM instead of sphingosine. The data shows that acridine orange completely inhibits the lipogenesis stimulated by insulin, PMA and hGH as well as basal lipogenesis.

EXAMPLE X

Figures 10A and 10B illustrate the effects of acridine orange on insulin and hGH binding to their receptors in isolated rat adipocytes. The experiments are conducted as in Figures 4A and 4B except that acridine orange (0-100 μM) was used instead of sphingosine. The data show that at 10 μM, a dose at which inhibition of lipogenesis is almost maximal, the drug has no effect on hormone binding, while 100 μM inhibits the binding of both hormones by 50%.

The free fat cell assay procedures which yielded the data demonstrating lipogenesis inhibition reported in the foregoing examples was developed within the pharmaceutical industry and is now an industry standard. This data is cogent evidence, which would be accepted by a person of ordinary skill in the art, of the utility of the invention to inhibit lipogenesis and thus to control obesity in mammals, including humans.

Administration and Dosage The selection of a mode of administration and of dosage level is within the competence and discretion of persons skilled in the art. The invention in its broader aspects includes a pharmaceutical composition containing a protein kinase antagonist, e.g., staurosporine or a sphingolipid or a lysosphingolipid or an a inoacridine useful to block lipid metabolism in mammals, including man. The antagonist may be administered per se or in combination with therapeutically acceptable adjuvants, e.g., liposomes.

Parenteral administration is preferred. Percutaneous administration using gels or creams containing a permeabilizing agent such as dimethylsulfoxide may be used. The protein kinase antagonist, i.e. , staurosporine or a sphingolipid or lysosphingolipid or an aminoacridine is administered in an amount therapeutically effective to partially or completely suppress the hormonal or basal stimulation of lipogenesis in mammalian fat cells.

The degree of suppression is a function of dosage level. Preferably about 15 to about 100 mg/kg of patient weight of protein kinase antagonist is appropriate for sphingolipids. Administration of at least about 15 mg/kg is appropriate to partially suppress lipogenesis. Administration of at least about 60 mg/kg is indicated for the complete suppression of basal lipogenesis. For staurosporine at least about 15 mg/kg is indicated for complete suppression of basal lipogenesis. Administration may appropriately continue over an appropriate time period until the desired therapeutic effect is observed.

A number of drugs are known to interfere with lipogenesis, e.g., hormone (catecholamines, glucagon, ACTH) clofibrate, halogenate, fenfluramine, amphetamine, cinchocaine, chlorpromazine, and some derivatives thereof. None of these drugs suppress lipogenesis as effectively as the protein kinase antagonists of this invention.