Synthesis and PTP1B inhibition of novel 4-aryl-1-oxa-9-thiacyclopenta[b]fluorenes

Article abstract of DOI:10.1016/S0960-894X(00)00278-X

Novel 4-aryl-1-oxa-9-thiacyclopenta[b]fluorenes were designed, synthesized, and evaluated as inhibitors of the protein tyrosine phosphatase, PTP1B. Compounds 3 (IC₅₀ = 284 nM) and 4 (IC₅₀ = 74 nM), showed nanomolar potency against PTP1B (TRDI(P)YETD(P)YRK as substrate). Compound 4 also lowered insulin in the diabetic ob/ob mouse at a dose of 10 mg/kg/day, po. (C) 2000 Elsevier Science Ltd. All rights reserved.

Full text of DOI:10.1016/S0960-894X(00)00278-X

Bioorganic & Medicinal Chemistry Letters 10 (2000) 1535±1538
Synthesis and PTP1B Inhibition of Novel 4-Aryl-1-Oxa-9-
Thiacyclopenta[b]¯uorenes
Jay Wrobel,* Zenan Li, Janet Sredy,y Diane R. Sawicki,y Laura Seestaller
and Donald Sullivan
Wyeth-Ayerst Research, Inc. CN 8000, Princeton, NJ 08543±8000, USA
Received 6 January 2000; accepted 26 April 2000
AbstractÐNovel 4-aryl-1-oxa-9-thiacyclopenta[b]¯uorenes were designed, synthesized, and evaluated as inhibitors of the protein
tyrosine phosphatase, PTP1B. Compounds 3 (IC50=284 nM) and 4 (IC50=74 nM), showed nanomolar potency against PTP1B
(TRDI(P)YETD(P)Y(P)YRK as substrate). Compound 4 also lowered insulin in the diabetic ob/ob mouse at a dose of 10 mg/kg/
day, po. # 2000 Elsevier Science Ltd. All rights reserved.
Protein tyrosine phosphatase 1B (PTP1B) appears to
play a major role in insulin sensitivity and the depho-
sphorylation of the insulin receptor (IR) on the basis of
many biochemical and cellular studies1 and according to
a pivotal study with PTP1B knockout mice.2 Depho-
sphylation of the IR correlates with insulin resistance3
and this, in turn, is a prime factor leading to Type II
diabetes.4 Thus, an orally active and selective PTP1B inhi-
bitor could potentially ameliorate insulin resistance and
normalize plasma glucose and insulin without inducing
hypoglycemia, and could, therefore, be a major advance
in the treatment of Type II diabetes.5 We previously
disclosed the PTP1B inhibition and oral antidiabetic
activities of a series 11-aryl benzo[b]naphtho[2,3-d]thio-
phenes, such as 16 (Fig. 1). In an e€ort to de®ne the role
and improve upon the properties of the tetracyclic ring
portion of 1 and 2, we designed, synthesized and tested 4-
aryl-1-oxa-9-thiacyclopenta[b]¯uorenes 3 and 4 (Fig. 1)
and report the results herein.
benzo[b]thiophene with n-butyl lithium, was treated with
this aldehyde (method b) and the secondary alcohol of
the resulting product was reductively eliminated using
conditions of Nutaitis8 to provide methylene congener 6
(method c). Treatment of this benzothiophene-furan 6
with p-anisoyl chloride under standard Friedel-Crafts
conditions (method d) resulted in acylation of 6, con-
comitant intramolecular cyclization and aromatization
to give the 1-oxa-9-thiacyclopenta[b]¯uorene heterocycle
7, which was the ®rst representative of this novel het-
erocyclic ring system. Starting material 6 (42%) was
recovered as well.
The methyl ether of 7 was demethylated with BBr3
(method e) and the resulting phenol was bis-iodinated
with molecular iodine and base to provide di-iodo phe-
nol 8 (method f). Treatment of 8 with (S)-lactic acid,
methyl ester under Mitsunobu conditions, followed by
methyl ester hydrolysis with aqueous potassium hydroxide
gave (R)-2-methyloxyacetic acid derivative 3 (method g).
Likewise, compound 8, when reacted with (S)-2-hydroxy-
3-phenylpropionic acid, methyl ester under similar con-
ditions, followed by basic hydrolysis gave (R)-2-benzyl-
oxyacetic acid derivative 4 (method h).
Chemistry
According to Scheme 1, commercially available 2,3-
dimethylfuran 5 was subjected to Vilsmeier conditions to
a€ord 4,5-dimethylfuran 2-carboxaldehyde (method a).7
2-Lithio-benzo[b]thiophene, produced via reaction of
For the preparation of 2, compound 96 was subjected to
molecular iodine and base (method i) and the resulting
di-iodophenol was further treated with (S)-2-hydroxy-3-
phenylpropionic acid, methyl ester under Mitsunobu
conditions. Methyl ester hydrolysis with aqueous
potassium hydroxide (method j) a€orded compound 2.
Analytical data for the target compounds and inter-
mediates are provided in footnote 17.
*Corresponding author at Wyeth-Ayerst Research, Inc., 145 King of
Prussia Road, Radnor, PA 19087, USA. Tel.: +1-610-341-2480; fax:
+1-610-989-4588; e-mail: wrobelj@war.wyeth.com
yPresent Address: Institute for Diabetes Discovery, Business Park
Drive, Branford, CT 06405, USA.
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00278-X

1536
J. Wrobel et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1535±1538
Figure 1.
Results and Discussion
The inhibitory activity of our compounds against
recombinant PTP1B9 was assessed using, as substrate,
the phosphotyrosyl dodecapeptide, TRDI(P)YETD(P)Y
(P)YRK, corresponding to the 1142±1153 insulin receptor
kinase regulatory domain, phosphorylated on the 1146,
1150, and 1151 tyrosine residues as described previously.6,10
Figure 2. Cation-p interactions of lysines 116 and 120 of PTP1B with
the tetracyclic-rings of 1 found via docking experiments.
obtainedthroughour synthesis e€orts, we prepared instead
the diiodide-containing analogue 4. Thus, the obvious
analogue from the benzo[b]naphtho[2,3-d]thiophene
series to compare to compound 4 was the diodide 2. This
latter compound contained the same ring system as 1
with the minor changes of substitution of two phenolic
ring bromines by iodine atoms and the tetracyclic ring
bromine by a hydrogen atom.
We previously reported that compound 1 had an IC50
against hPTP1B of 61 nM, making it among the most
potent PTP1B inhibitors reported to date.6 In our dock-
ing studies of 1 with the X-ray crystal structure PTP1B,6
we noticed that the tetracyclic ring portion interacted with
lysine-120 and lysine-116 via cation-p type interactions
described by Dougherty11 (see Fig. 2). There was a par-
ticularly strong interaction (ꢀ3 A) between lysine-120
side-chain nitrogen atom and the naphthalene ring of
the tetracycle. In an e€ort to capitalize on this possible
interaction, we designed the 4-aryl-1-oxa-9-thiacyclopenta
[b]¯uorene ring system (as shown in analogues 3 and 4)
because the terminal furan ring of this system would
provide a more electron rich aromatic environment that
could potentially have stronger interactions with the
ammonium side chain of lysine-120.
Consistent with our hypothesis gleaned from our docking
studies, compound 4, with an IC50 of 74 nM, was more
potent than 2 (IC50 of 179 nM). Structural studies are
necessary to con®rm our docking results and therefore
other reasons for this potency enhancement are possible.
For instance, the length of the tetracyclic portion
increased by 1 A when going from 2 to 4. This increase
in length results in an extension of lipophilic moieties that
are available for binding via hydrophobic interactions.
Other di€erences that could play a role in the increased
potency of 4 over 2 include di€erences in dipole moments
and the potential H-bonding site of the furan oxygen
atom of 4 that is not present in 2.
In order to directly compare the e€ect on activity of the 1-
oxa-9-thiacyclopenta[b]¯uorene ring system with the benzo
[b]naphtho[2,3-d]thiophene ring system of compound 1,
the substituents of both compounds would have to be
identical. Since the tribromo derivative of the 1-oxa-9-thia-
cyclopenta[b]¯uorene analogous to 1 was not readily
The (R)-benzyl moiety emanating from the acetic acid
a-carbon of 1 was incorporated into 1 on the basis of
Scheme 1. Reagents and conditions: (a) DMF, POCl3, rt, 3.5 h, 76%; (b) 2-lithio-benzo[b]thiophene, THF, 78 ꢁC, 1 h, 99%; (c) NaBH4, TFA, THF,
rt, 3.5 h, 42%; (d) p-anisoyl chloride, SnCl4, CS2, 78 to 0 ꢁC, 9 h, 25%; (e) BBr3, CH2Cl2, 78 ꢁC to rt, 2.5 h, 94%; (f) I2, NaOH, MeOH, 0 ꢁC to rt, 15
h, 54%; (g) 8 to 3 (S)-lactic acid, methyl ester, diethyl azodicarboxylate, PPh3, benzene, re¯ux, 3 h, 52%/KOH, H2O, dioxane, rt, 32 h, 95%; (h) 8 to 4
(S)-HOCH2(CH2Ph)CO2Me, diethyl azodicarboxylate, PPh3, benzene, 60 ꢁC, 3 h, 54%/KOH, H2O, dioxane, rt, 24 h, 93%; (i) I2, NaOH, MeOH, 0 ꢁC
to rt, 15 h, 54%; (j) (S)-HOCH2(CH2Ph)CO2Me, diethyl azodicarboxylate, PPh3, benzene, 60 ꢁC, 3 h, 86%/KOH, H2O, dioxane, rt, 24 h, 99%.

J. Wrobel et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1535±1538
1537
the aforementioned docking studies of 1, and earlier
analogues.6 In these studies, the benzyl moiety of 1 ®lled an
empty pocket within the enzyme active site. According to
these docking results, a substituent smaller than a benzyl
group would not ®ll this pocket as well, and was therefore
proposed to be less active. To examine this hypothesis, the
1-oxa-9-thiacyclopenta[b]¯uorene analogue 3 that con-
tained an (R)-methyl moiety on the acetic acid a-carbon
was prepared. This compound, with an IC50 of 284 nM,
was indeed more than 4-fold less potent than 4.
11. Ma, J. C.; Dougherty, D. A. Chem. Rev. 1997, 97, 1303.
12. Wrobel, J.; Li, Z.; Dietrich, A.; McCaleb, M.; Mihan, B.;
Sredy, J.; Sullivan, D. J. Med. Chem. 1998, 41, 1084.
13. Plasma glucose and insulin values of 200±350 mg/dL and
300±775 mIU/mL, respectively, in untreated mice.
14. Sredy, J.; Sawicki, D. R.; Flam, B. R.; Sullivan, D. Meta-
bolism 1995, 44, 1074.
15. Sredy, J. unpublished data.
16. Ciglitazone reference standard at 100 mg/kg dose: 43%
decrease in glucose and 39% decrease in insulin insulin.
17. Analytical data. Melting points were determined on an
electrothermal capillary melting point apparatus and are not
corrected. Proton magnetic resonance (1H NMR) spectra were
recorded at 200 MHz (Varian XL-200), 300 MHz (VXR-300
or Bruker DPX300), or at 400 MHz (Bruker AM-400 or VXR-
400). Electron Impact (EI, IE=70 eV) and chemical ionization
(CI, isobutane reagent gas) mass spectra were recorded on a
Finnigan model 8230 spectrometer. Fast atom bombardment
(FAB) were recorded on a Kratos MS50. High Resolution MS
were recorded on a Bruker 9.4 T FTMS. Analyses (C, H, N)
were carried out on a modi®ed Perkin±Elmer model 240 CHN
analyzer. Analytical results for elements were within Æ0.4% of
the theoretical values.
The compounds were also evaluated as antidiabetic
agents in the ob/ob mouse model.6,12,13 Insulin resistance
in this model, has been associated with a reduction in
insulin-induced protein tyrosine phosphorylation in tissues
such as liver, and markedly elevated PTPase activities in
these tissues have been observed leading to the conclusion
that PTPases may cause or contribute to the reduced
phosphorylation of the IR.14 Increased abundance of
PTP1B in the livers of ob/ob mice has also been noted.15
Although neither 2 nor 4 signi®cantly lowered plasma
glucose values in these studies, the 1-oxa-9-thiacyclopenta
[b]¯uorene analogue 4 demonstrated a statistically sig-
ni®cant (p < 0.05) insulin lowering e€ect of 43% at a dose
of 10 mg/kg dose.16 The benzo[b]naphtho[2,3-d]thiophene
analogue 2 also demonstrated a statistically signi®cant
(p<0.05) insulin lowering e€ect (68%) although it was
at a considerably higher dose (25 mg/kg) than that for 4.
4,5-Dimethyl-2-furaldehyde: yellow oil: MS (ESI): [M+H]+,
125.
Benzo[b]thiophen-2-yl-(2,3-dimethyl-furan-5-yl)-methanol. Yellow
oil: 1H NMR (DMSO-d6) d 7.89 (d, J=8 Hz, 1H), 7.76 (d, J=8
Hz, 1H), 7.26±7.356 (m, 2H), 7.25 (s, 1H), 6.36 (s, 1H), 6.08 (s,
1H), 5.92 (s, 1H), 2.13 (s, 3H), 1.86 (s, 3H); MS (EI): [M+], 258.
Benzo[b]thiophen-2-yl-(2,3-dimethyl-furan-5-yl)-methane (6). Oil:
1H NMR (CDCl3) d 7.74 (d, J=8 Hz, 1H), 7.67 (d, J=8 Hz,
1H), 7.22±7.35 (m, 2H), 7.08 (d, J=1 Hz, 1H), 5.93 (s, 1H),
4.15 (s, 2H), 2.17 (s, 3H), 1.90 (s, 3H); MS (EI): [M+], 242.
4-(2,3-Dimethyl-1-oxa-9-thia-cyclopenta[b]¯uoren-4-yl)-phenyl
Acknowledgements
1
methyl ether (7). Light-yellow solid: H NMR (CDCl3) d 7.81
(s, 1H), 7.76 (d, J=8 Hz, 1H), 7.24±7.36 (m, 3H), 7.00±7.11 (m,
3H), 6.85 (d, J=8 Hz, 1H), 3.96 (s, 3H), 2.37 (s, 3H), 1.55 (s,
3H); MS (EI): [M+], 358.
We are grateful to the Discovery Analytical Chemistry
Department of Wyeth-Ayerst for elemental analyses,
400-MHz 1H NMR, mass spectroscopy and HPLC data.
4-(2,3-Dimethyl-1-oxa-9-thia-cyclopenta[b]¯uoren-4-yl)-phenol.
Brown: mp 174±175 ꢁC: H NMR (CDCl3) d 7.81 (s, 1H), 7.76
1
(d, J=8 Hz, 1H), 7.27±7.31 (m, 3H), 7.00±7.08 (m, 3H), 6.87 (d,
J=8 Hz, 1H), 5.00 (s, 1H), 2.37 (s, 3H), 1.58 (s, 3H); MS (EI):
[M+], 344.
References and Notes
4-(2,3-Dimethyl-1-oxa-9-thia-cyclopenta[b]¯uoren-4-yl)-2,6-
diiodo-phenol (8). White solid: 1H NMR (CDCl3) d 7.83 (s,
1H), 7.81 (s, 2H), 7.80 (d, J=8 Hz, 1H), 7.33 (dd, J=8, 7 Hz,
1H), 7.15 (dd, J=8, 7 Hz, 1H), 6.98 (d, J=8 Hz, 1H), 5.99 (s,
1H), 2.37 (s, 3H), 1.61 (s, 3H); MS (EI): [M+], 596.
(R)-2-[4-(2,3-Dimethyl-1-oxa-9-thia-cyclopenta[b]¯uoren-4-yl)-2,6-
diiodo-phenoxy]-propionic acid (3). O€-white solid: mp 218±
1. Byon, J. C. H.; Kusari, A. B.; Kusari, J. Mol. Cell. Bio-
chem. 1998, 182, 101.
2. Elchebly, M.; Payette, P.; Michaliszyn, E.; Cromlish, W.;
Collins, S.; Loy, A. L.; Normandin, D.; Cheng, A.; Himms-
Hagen, J.; Chan, C.-C.; Ramachandran, C.; Gresser, M. J.;
Tremblay, M. L.; Kennedy, B. P. Science 1999, 283, 1544.
3. Goldstein, B. J.; Ahmad, F.; Ding, W.; Li, P.-M.; Zhang,
W.-R. Mol. Cell. Biochem. 1998, 182, 91.
4. Kahn, C. R. Diabetes 1994, 43, 1066.
5. Evans, J. L.; Jallal, B. Exp. Opin. Invest. Drugs 1999, 8,
139.
6. Wrobel, J.; Sredy, J.; Moxham, C.; Dietrich, A.; Li, Z.;
Sawicki, D. R.; Seestaller, L.; Wu, L.; Katz, A.; Sullivan, D.;
Tio, C.; Zhang, Z.-Y. J. Med. Chem. 1999, 42, 3199.
7. Martin, S. F.; Gluchowski, C.; Campbell, C. L.; Chapman,
R. C. J. Org. Chem. 1984, 49.
8. Nutaitis, C. F.; Charles, F.; Patragnoni, R.; Goodkin, G.;
Neighbour, B.; Obaza-Nutaitis, J. Org. Prep. Proced. Int.
1991, 23, 403.
219 ꢁC: H NMR (CDCl3) d 7.97 (s, 2H), 7.85 (s, 1H), 7.81 (d,
1
J=8, Hz, 1H), 7.34 (ddd, J=8, 7, 1 Hz, 1H), 7.12 (ddd, J=8, 7, 1
Hz, 1H), 6.87 (d, J=8 Hz, 1H), 5.46 (q, J=7 Hz, 1H, CH), 2.38
(s, 3H), 1.77 (d, J=7 Hz, 3H), 1.61 (s, 3H); MS (+FAB): [M+],
667.8, [M+H]+, 668.9. Anal. calcd for C25H18I2O4S: C, 44.93;
H, 2.72; N, 0.00. Found: C, 44.77; H, 2.63; N, 0.20.
(R)-2-[4-(2,3-Dimethyl-1-oxa-9-thia-cyclopenta[b]¯uoren-4-yl)-2,6-
diiodo-phenoxy]-3-phenyl-propionic acid (4). White solid: mp
215±217 ꢁC 1H NMR (DMSO-d6) d 8.20 (s, 1H), 7.96 (d, J=8
Hz, 1H), 7.92 (d, J=2 Hz, 2H), 7.45±7.30 (m, 5H), 7.27 (dd,
J=7 Hz, 1H), 7.16 (dd, J=7, 1 Hz, 1H), 6.84 (d, J=8 Hz, 1H),
5.36 (t, J=6 Hz, 1H), 3.43 (dd, J=6 Hz, 2H), 2.37 (s, 3H),
1.57 (s, 3H); MS (+FAB): [M+H]+, 745; High-resolution MS
(ESI) calcd for C31H22I2O4S: [M+H]+ 744.94010. Found:
744.94090.
9. Barford, D.; Keller, J. C.; Flint, A. J.; Tonks, N. K. J. Mol.
Biol. 1994, 239, 726.
10. The reported IC50 values are from direct regression curve
analyses that are signi®cant (p<0.05).
4-(Benzo[b]naphtho[2,3-d]thiophen-11-yl)-2,6-diiodo-phenol.
White solid: mp 213±214 ꢁC: MS (-FAB): [M-H] , 576.8. Anal.

1538
J. Wrobel et al. / Bioorg. Med. Chem. Lett. 10 (2000) 1535±1538
calcd for C22H12I2OS: C, 45.70; H, 2.09; N, 0.00. Found: C,
45.82; H, 2.07; N, 0.30.
(R)-Benzo[b]naphtho[2,3-d]thiophen-11-yl-2,6-diiodo-phenoxy)-3-
phenyl-propionic acid (2). White solid: mp 115±117 ꢁC: NMR
(CDCl3) d 8.36 (s, 1H), 7.95 (dd J=8, 1 Hz, 1H), 7.87 (d, J=2
Hz, 1H), 7.85 (d, J=2 Hz, 1H), 7.78 (dd, J=8, 1 Hz, 1H), 7.27±
7.58 (m, 9 H), 7.12 (ddd, J=8, 7, 1 Hz, 1H), 6.77 (d, J=8, 1 Hz,
1H), 5.56 (t, J=7 Hz, 1H), 3.67±3.55 (m, 2H); MS (EI): [M+],
726. Anal. calcd for C31H20I2O3S: C, 51.26; H, 2.77; N, 0.00.
Found: C, 51.49; H, 2.87; N, 0.13.