Fungal Metabolites as Potent Protein Kinase Inhibitors - Ingenta Connect

14 downloads 122 Views 137KB Size Report
Fungal Metabolites as Potent Protein Kinase Inhibitors: Identification of a. Novel Metabolite and Novel Activities of Known Metabolites. Masayoshi Oyama† ...
Letters in Drug Design & Discovery, 2004, 1, 24-29

24

Fungal Metabolites as Potent Protein Kinase Inhibitors: Identification of a Novel Metabolite and Novel Activities of Known Metabolites Masayoshi Oyama†, Zhihong Xu†$, Kuo-Hsiung Lee†, Timothy D. Spitzer‡, Peter Kitrinos ‡ , Octerloney B. McDonald ¶ , Rosie R.J. Jones # and Edward P. Garvey*ƒ Received July 30, 2003: Accepted October 7, 2003

†Natural

Products Laboratory, School of Pharmacy, University of North Carolina, Chapel Hill, North Carolina 27599,USA ‡ Computational, Analytical, and Structural Sciences, ¶ Systems Research; and ƒ Biochemical and Analytical Pharmacology, GlaxoSmithKline, Research Triangle Park, North Carolina 27709, USA #GlaxoSmithKline Medicines Research Centre, Gunnels Wood Road, Stevenage Abstract: A novel undecylresorcinol dimer (1) was isolated from Coleophoma sp. and inhibited cFMS receptor tyrosine kinase (IC50 of 0.4 µM), with greater than 10-fold selectivity versus nine other protein kinases. The known fungal metabolites balanol and altenusin inhibited cFMS kinase and pp60c-Src kinase, respectively, even more potently and selectively. Altenusin inhibited pp60c-Src with an IC50 of 20 nM and a selectivity of at least 400-fold versus nine other protein kinases. Balanol inhibited cFMS receptor kinase with an IC50 of 1 nM and selectivities of 14-75-fold versus pp60c-Src and VEGF receptor kinases and greater than 10,000-fold versus seven other kinases.

Keywords: fungal metabolite, protein kinase inhibitor, cFMS receptor tyrosine kinase, pp60c-Src kinase, balanol, altenusin, alternariol. INTRODUCTION Natural products are a rich source for inhibitors of protein kinases (ex, erbstatin, coumarins, staurosporines, lavendustin, etc, [1]). Although many of these compounds are polyhydroxylated aromatics and not considered good chemical templates for pharmaceutical development, there are indeed several natural products (or derivatives) that are currently in clinical trials (ex, flavopiridol, bryostatin-1, PKC412, UCN 01, [2]). In addition, natural products are often used as tool compounds because of their potency or selectivity. For example, wortmannin is routinely used as a potent and selective inhibitor of phosphatidylinositol-3kinase both in vitro and in cells [3,4]. Finally, natural products have been used in structural studies to help elucidate the specific interactions between kinase and inhibitor that lead to potency and selectivity [5-7]. We report here the results of screening fungal extracts against the cFMS receptor tyrosine kinase (also known as CSF-1 or m-CSF receptor tyrosine kinase). This kinase has been implicated in both neoplastic [8] and bone diseases [9] and very few inhibitors have been described to date in the literature. We were interested in identifying possible leads against cFMS receptor kinase itself, and also in using this kinase as a surrogate to identify novel kinase inhibitor chemotypes. Although no new template was identified for drug discovery purposes, a novel fungal metabolite 1 and potent and selective inhibitors were identified. These *Address correspondence to this author at the Biochemical and Analytical Pharmacology, GlaxoSmithKline, Research Triangle Park, North Carolina 27709, USA; Tel.: 919-483-4260; Fax: 919-483-4320; E-mail: [email protected] $Present addresses: Duke University, Durham, North Carolina 27710 1570-1808/04 $45.00+.00

inhibitors could possibly be turned into tools or used in structural studies to aid in development of inhibitors of cFMS receptor kinase or related kinases. EXPERIMENTAL PROCEDURES Isolation of Compound 1 The producing microorganism, a mitosporic fungus belonging to the Coelomycete class, was isolated from a plant sample from Malaysia. Growth was maintained on malt extract agar at 25°C. Two 5 mm plugs, taken from the growth edge of the colony, were used to inoculate a 250 ml Erlenmeyer flask containing 50 ml of seed medium. The seed medium contained 1% peptone (Oxoid L34), 4% glycerol, 2% malt extract, 0.1% Junlon (Honeywill and Stein); and the pH was adjusted to 6.3. The inoculated seed flask was incubated for 8 days at 25°C on a rotary shaker (250 rpm). A 1.5 ml aliquot of seed culture was transferred into a 250 ml Erlenmeyer flask containing 50 ml of the production media [4% maltose, 12% Glucidex (Roquette), 1% peptone (Oxoid L37), 1.5% cottonseed flour (Sigma), 1.5% beet molasses, 0.5% MgSO4 •7H 2 O, 0.5% CaCO3 , 0.2% FeSO4 •7H 2 O, 0.002 %; ZnSO4 •2H 2 O, 0.02% LTryptophan, and 12% Vermiculite exfoliated (V3 Silvaperl) 12.0%; pH was not adjusted]. The solid state fermentation was incubated for 23 days at 25°C. Methanol (100 ml) was added to the solid state fermentation (50 ml). After 2 hr, the solid substrate was broken into small pieces and left to extract overnight. The mixture was then filtered through Whatman filter paper. The extract was taken to dryness by rotary evaporation, and then 5 ml of dry methanol was added. After mixing for 30 min, this methanol extract was filtered through filter paper and © 2004 Bentham Science Publishers Ltd.

Fungal Metabolites as Potent Protein Kinase Inhibitors

Letters in Drug Design & Discovery, 2004, Vol. 1, No. 1

25

dried under a N2 stream. The material obtained (440 mg) was dissolved in 4.4 ml of DMSO, and 2 ml was purified by preparative HPLC on a Hypersil HyPURITY Elite C18 column (5 mm particle size, 15 cm x 21.2 mm i.d.) with the following acetonitrile gradient in 20mM NH4OAc buffer: 030 min, a linear gradient of 0-100%; 30-40 min, 100%. Thirty sec (7.5 ml) fractions were collected and samples of each fractions were tested for inhibitory activity against cFMS receptor kinase. The fractions with retention times of 26-33 min showed inhibitory activity and were further purified on the same C18 column. By eluting the material with 40-100% gradient of acetonitrile in 20mM NH4OAc buffer, an active compound (1) was isolated from the fraction with retention times of 12.5-14.5 min. All HPLC procedures were carried out at room temperature, and the flow rate was 15 ml/min.

µl resulted in a final nominal enzyme concentration of 20 nM. [It is noted that enzyme titration experiments indicated that the amount of active kinase was 90% purity using glutothione agarose resin. Protein concentration was approximated with Coomassie Plus reagent (Pierce). The specific activity was 2 nmol/min/mg under the assay conditions described below. Condition for activation (ie, autophosphorylation) of cFMS receptor kinase was as follows: 10 µM enzyme, 100 µM ATP (Sigma), 5 mM MgCl2 (Sigma) in 50 mM Tris-HCL for 90 minutes at room temperature. Enzyme reactions were performed in a volume of 45 µl, using round-bottom polystyrene 96-well plates (Corning Costar) on a Beckman-Coulter Biomek 2000. One µl compound or 1 µl DMSO alone (positive controls) were added to assay plates. Enzyme reactions were performed as follows: 30 µl of a 1.5x substrate reaction mix containing 50 mM MOPS, pH 7.5, 15 mM MgCl2 , 6 µM biotinEAIYAPFAKKK-NH 2 (Quality Controlled Biochemicals, Inc.), 7.5 mM DTT (Sigma), 75 mM NaCl (Sigma), 10 µM ATP (Sigma), and 0.5 µCi [γ- 33 P] ATP (Amersham Pharmacia). The reaction was initiated by the addition of enzyme. Under standard conditions, reaction generated approximately 15,000 cpm compared to a background of approximately 800 cpm. Activated enzyme was diluted 500fold with 50 mM MOPS, pH 7.5 so that the addition of 15

Extracts in DMSO and samples from HPLC fractionation were assayed directly. For titrations of purified compounds, the stock concentrations were measured by NMR and then serially diluted 1 to 3 and added to the reactions to produce an 11-point dose-response curve ranging from 0.0001 to 10 µM. Primary data were normalized to control values [100*(U1-C2)/(C1-C2), where C1 is CPM in the presence of no compound, C2 is CPM in the presence of excess EDTA, and U is CPM in the presence of compound]. Equation 1 y = Vmax (1-(xN/(IC50N+xN))) + Y2

(Eq. 1)

where Vmax is maximal signal, x is the concentration of inhibitor, N is the hill coefficient, and Y2 is the background signal, fitted to the data to determine apparent IC50 value for the inhibitor. RESULTS AND DISCUSSION Isolation and Structure Identification of Compound 1 Originally, 20,000 fungal extracts, which represented 5,000 organisms grown under four different conditions, were screened against a number of molecular targets. Because of the reproducibility and robustness of the initial activity against cFMS receptor kinase, several of these extracts were prepared in larger amounts to isolate and elucidate the structure of the inhibitory metabolite. HPLC fractionation led to the isolation of an inhibitor against cFMS receptor kinase from a Malaysian fungus, Coleophoma sp. (Class Coelomycete). Compound 1 was isolated as colorless oil and gave an [M+H]+ ion at m/z 571.4034, corresponding to the formula C35H55O6 (calcd. 571.3999) in the HRESIMS. UV absorption bands were present at 263 and 300 nm. The IR spectrum showed absorbance for hydroxyl (3599 cm-1) and ester carbonyl (1644 cm-1 ) groups. The 1 H NMR spectrum showed three aromatic protons in an A2 B spin system [δ 6.71 (2H, d, J = 2.0 Hz) and 6.75 (1H, d, J = 2.0 Hz)], two meta-coupled protons [δ 6.11 and 6.13 (each d, J = 2.2 Hz)], one oxymethine proton (δ 5.04), and two terminal methyl protons (δ 0.84 and 0.87). Nineteen

26

Letters in Drug Design & Discovery, 2004, Vol. 1, No. 1

Garvey et al.

HO HO

OH+

O OH

C+:

OH

m/z 263

OH

OH+

HO

OH m/z 571

OH m/z 309 OH

C

OH+ OH

m/z 291 OH

HO

2' 3'

1'

7'

8'

9'

10' 11'

6' 5' OH 16 14 12 10 9 17 15 13 11

12'

14' 13'

15'

16'

O

4'

8

7

O 2

1

OH 3 4

6 1

17'

5 OH

Fig. (1). Mass fragmentation and structure of 1. MS data generated as described in Experimental Procedures. Numbering system describing atoms of compound 1 is the same as used to describe NMR assignments listed in Table 1.

methylene groups were identified from the HMQC spectra (Table 1).

13 C

NMR and

Table 1.

Chemical Shifts and Assignments for Compound 1 13C Chemical

Shift (ppm)

1H Integral and Multiplicity2

1

-

-

144.0

2

-

-

109.0

3

-

-

159.7

4

6.14

1H, d, J=2.2 Hz

100.4

5

-

-

160.2

6

6.12

1H, d, J=2.2 Hz

108.7

7

2.56

2H, ≈ t, J≈7.8 Hz

34.4

8

1.45

2H, m

31.4

9 - 14

1.24 – 1.29

14H, m

28.6 – 29.2

15

1.24 – 1.29

2H, m

31.2

16

1.24 – 1.29

2H, m

22.0

17

0.85

3H, t, J=7.4 Hz

13.8

1’

-

-

142.6

2’,6’

6.72

2H, d, J=2.0

115.0

Fragment ions at m/z 263.2039 and 309.2075 in the positive ESIMS spectrum were assignable to undecylresorcinol (C 17 H 27 O 2 , calcd. 263.2011) and undecylresorcylic acid (C18 H 29 O 4 , calcd. 309.2066) as shown in Fig. 1. Consequently, the structure of the cFMS inhibitor was determined to be 8-(3,5-dihydroxyphenyl)-1propyloctyl 2,4-dihydroxy-6-undecylbenzoate, a novel undecylresorcinol dimer depicted as 1 (Fig. 1). It is noted that although this structure has not been previously reported to our knowledge, similar compounds appeared recently in a patent application [10] as inhibitors of HIV integrase. No other publications of these compounds have appeared to date.

Assignment1

Likewise, by the same procedure, we isolated three active metabolites from two other extracts. The structures were elucidated using the same analytical techniques described above, and the proposed structures were identical to the previously described fungal metabolites balanol [11], alutenusin, and alternariol [12] (Fig.2). Balanol was from an extract of a Acremonium species isolated from the Quercus roots from an English garden. Alutenusin and alternariol were both isolated from an extract of an Ulocladium species isolated from the dead stem of Rumex acetosa in England.

1H Chemical

Shift (ppm)

Fungal Metabolites as Potent Protein Kinase Inhibitors

Letters in Drug Design & Discovery, 2004, Vol. 1, No. 1

(Table 1)contd..... 13C Chemical

Shift (ppm)

1H Integral and Multiplicity2

3’,5’

-

-

153.5

4’

6.76

1H, t, J=2.0

110.3

7’

2.46

2H, ≈ t, J≈7.8 Hz

35.2

8’

1.50

2H, m

30.9

9’ – 12’

1.24 – 1.29

8H, m

28.6 – 29.2

13’

1.32

2H, m

24.9

Assignment1

1H Chemical

Shift (ppm)

14’

5.05

1H, m

74.2

15’

1.56

2H, m

35.8

16’

1.36

2H, m

18.1

17’

0.88

3H, t, J=7.4 Hz

13.9

C=O

-

-

169.3

compound 1 was noteworthy in that cFMS receptor kinase was the only kinase inhibited with a potency less than 1 µM and was 10-fold more potent compared to inhibiting VEGF receptor kinase, a closely related tyrosine receptor kinase. cSrc and CDK4 were less potently inhibited, and none of the other kinases were not inhibited at the highest concentration tested (10 µM). Thus, compound 1 was a novel metabolite that potently and selectively inhibited cFMS receptor kinase. OH

O

OH

NH

O O

O

N H

HO OH

O

Balanol OH

1 assignment numbering is described in Fig. 1 2 s=singlet, d=doublet, t=triplet, m=multiplet

O HO

O

O

Inhibition of cFMS Receptor Kinase and Nine Other Protein Kinases by Isolated Fungal Metabolites The potency of inhibition by these isolated metabolites were assessed against cFMS receptor kinase and nine other protein kinases listed in Table 2. As was expected based on following the purification of the compounds by the inhibition of cFMS receptor kinase, each metabolite inhibited this kinase (Fig. 3). The potency and selectivity of

27

HO

O

OH HO

OH OH

O Alternariol

Altenusin

Fig. (3). Chemical structures for balanol, alternariol, and altenusin.

Fig. (2). Inhibition of cFMS receptor tyrosine kinase by the four natural products. Concentration of purified natural product [balanol (filled circles), alternariol (open circles), compound 1 (open triangles), and altenusin (filled triangles)] was determined by NMR and serially diluted 1 to 3 in DMSO. cFMS receptor kinase assay is described in Experimental Procedures. Equation 1 was fited to the data to generate the titration curves and to calculate the IC50 values presented in Table 2.

28

Letters in Drug Design & Discovery, 2004, Vol. 1, No. 1

Table 2.

Garvey et al.

Inhibition of cFMS Receptor Kinase and Other Kinases by Fungal Metabolites1 cFMSR

VEGFR2

Erb2R

Erb4R

EGFR

pp60c-Src

CDK4

PKCζζ

PLK

GSK-3

compound 1

0.4

3

>10

>10

>10

7

6

>10

>10

>10

balanol

0.001

0.07

>10

>10

>10

0.01

>10

10

>10

7

alternariol

0.3

0.4

5

>10

>10

1

>10

>10

>10

0.7

altenusin

8

8

>10

>10

>10

0.02

>10

>10

>10

>10

1 All IC values are averages of two separate determinations, with standard errors less than 15%. The range of Hill coefficients was from 0.8 to 1.1 indicating that one 50 inhibitor molecule bound per enzyme. Units are µM.

Compared to compound 1, the potency and selectivity of balanol (IC50 value of 1 nM) was striking. Inhibition of pp60c-Src and VEGF receptor kinase was potent but at least 1/14- and 1/75-fold less potent than inhibition of cFMS receptor kinase. All other kinases were greater than 1/10,000fold less sensitive. Balanol was first described as a nonselective inhibitor of human protein kinase C (PKC) isozymes [11]. In a subsequent publication, analogs were described with enyzme inhibition again focused on PKC isozymes (with limited data for protein kinase A), and balanol itself had potency against PKCα, βII, and ε of 67, 30, and 38 nM, respectively [13]. In the first report of balanol [11], it was stated that the ζ isozyme was 1/100-fold less sensitive compared to the other isozymes. Using these two reports, the potency of balanol against PKCζ would be approximately 5 µM, in close agreement with the potency reported here (10 µM). Other serine/threonine protein kinases are also inhibited by balanol in the nM potency range [14]. However, there has been only one report on inhibition of tyrosine protein kinases – balanol did not inhibit EGF receptor kinase or Src kinase [15]. The first finding is consistent with our finding here, whereas, the lack of inhibition of Src in the earlier report is not. We do not know the cause of this discrepancy. Finally, we reisolated the known fungal metabolites alternariol and altenusin. Most reports on these and similar natural products have been focused on the mechanism of their toxicity (ex, light induced DNA cross linking [16]). Little data on the inhibition of protein kinases are available in the literature. However, given the polyhydroxylated aromatic nature of these metabolites, it was not surprising that these were kinase inhibitors. Alternariol can be best described as a moderately potent, fairly nonselective kinase inhibitor (Table 2). The more interesting aspect of these data is the emergence of potent and highly selective inhibition of pp60c-Src kinase in going from the restricted three-ring system of alternariol to the flexible two-ring system of altenusin. Altenusin was a 20 nM inhibitor of pp60c-Src kinase with almost 500-fold selectivity against all other kinases tested. Altenusin and alternariol would therefore be excellent tool compounds for structural studies with c-Src kinase to understand how potency and selectivity can be introduced into c-Src inhibitors.

known metabolites with novel potencies and selectivities. Most interestingly, balanol was found to be a low nM inhibitor with good selectivity vs. other tyrosine protein kinases and exquisite selectivity against the other kinases tested. Balanol is the most potent inhibitor described to date against this kinase, and it may represent a starting point for a pharmaceutical program directed at cFMS receptor kinase. Furthermore, through the combination of screening one kinase and having a panel of other kinases for secondary selectivity measurement, we also were able to demonstrate that altenusin is a potent and highly selective inhibitor of pp60c-Src. Structural studies of pp60c-Src and both altenusin and its relatively weak analog alternariol may help elucidate how selectivity can be built into inhibitors of this particular enzyme. ACKNOWLEDGEMENTS We thank the Natural Product group at the Stevenage, UK site at the time of this study for providing all of the fungal extracts and for advice throughout the work. We also thank the following for providing the kinase selectivity data for the nine other kinases studied: Anne Truesdale, John Van Arnold, Brian Reep, Stephanie Schweiker, Laurie Kane, and Frank Preugschat. FOOTNOTES 1 O.B. McDonald, S.D. Chamberlain, J.G. Conway, et al. (manuscript in progress).

REFERENCES [1] [2]

[3] [4] [5] [6] [7]

CONCLUSIONS cFMS receptor kinase is a therapeutic target with relatively few inhibitors described to date. In screening against a library of fungal extracts, we isolated a novel metabolite that selectively inhibited this kinase and also

[8] [9] [10]

Lawerence, D.S., Niu, J, Pharmocol. Ther., 1998, 77, 81. Fabbro, D., Ruetz, S., Buchdunger, E., Cowan-Jacob, S.W., Fendrich, G., Liebetanz, J., Mestan, J., O’Reilly, T., Traxler, P., Chaudhuri, B., Fretz, H., Zimmermann, J., Meyer, T., Caravatti, G., Furet, P., Manley, P.W. Pharmacol. Ther., 2002, 93, 79. Arcaro, A., Wymann, M.P., Biochem. J., 1993, 296, 297. Powis, G., Bonjouklian, R., Berggren, M.M., Gallegos, A., Abraham, R., Ashendel, C., Zalkow, L, Matter, W.F., Dodge, J., Grindey, G. Cancer Res., 1994, 54, 2419. Prade, L., Engh, R.A., Girod, A.; Kinzel, V., Huber, R., Bossemeyer, D. Structure, 1997, 5, 1627. Lamers, M.B., Antson, A.A., Hubbard, R.E., Scott, R.K., Williams, D.H., J. Mol. Biol., 1999, 285, 713. Johnson, L.N., De Moliner, E., Brown, N.R.; Song, H., Barford, D., Endicott, J.A., Noble, M.E. Pharmacol. Ther., 2002, 93, 113. Sapi, E., Kacinski, B.M. Proc. Soc. Exp. Biol. Med., 1999, 220, 1. Filvaroff, E., Derynck, R. Curr. Biol., 1998, 8, R679. Silverman, K.C., Lingham, R.B., Singh, S.B., Zink, D.L. Teran, A. 2 0 0 0 , HIV integrase inhibitors. (U.S. Patent application #6,124,327).

Fungal Metabolites as Potent Protein Kinase Inhibitors [11] [12] [13]

Kulanthaivel, P., Hallock, Y.F., Boros, C., Hamilton, S.M., Janzen, W.P., Ballas, L.M., Loomis, C.R., Jiang, J.B., Katz, B., Steiner, J.R., Clardy, J. J. Am. Chem. Soc., 1993, 115, 6452. Karlovsky, P. Nat. Toxins, 1999, 7, 1. Lampe, J.W., Biggers, C.K., Defauw, J.M., Foglesong, R.J., Hall, S.E., Heerding, J.M., Hollinshead, S.P., Hughes, P.F., Jagdmann, G.E. Jr., Johnson, M.G., Lai, Y.-S., Lowden, C.T., Lynch, M.P., Mendoza, J.S., Murphy, M.M., Wilson, J.W., Ballas, L.M., Carter,

Letters in Drug Design & Discovery, 2004, Vol. 1, No. 1

[14] [15] [16]

29

K., Darges, J.W., Davis, J.E., Hubbard, F.R., Stamper, M.L. J. Med. Chem., 2002, 45, 2624. Setyawan, J., Koide, K., Diller, T.C., Bunnage, M.E., Taylor, S.S., Nicolaou, K.C., Brunton, L.L. Mol. Pharmacol., 1999, 56, 370. Koide, K., Bunnage, M.E., Gomez Paloma, L., Kanter, J.R., Taylor, S.S., Brunton, L.L., Nicolau, K.C. Chem. Biol., 1995, 2, 601. DiCosmo, F. Straus, N.A. Experientia, 1985, 41, 1188.