Phorboxazole analogues induce association of cdk4 with extranuclear cytokeratin intermediate filaments

Article abstract of DOI:10.1021/ja057087e

The cellular localization profile and molecular association of the phorboxazoles were examined with a streamlined target elucidation system using synthetic fluorescent probes. Cellular image analyses identified the binding of phorboxazole analogues to cyt

Full text of DOI:10.1021/ja057087e

Published on Web 03/03/2006
Phorboxazole Analogues Induce Association of cdk4 with Extranuclear
Cytokeratin Intermediate Filaments
Craig J. Forsyth,*,† Lu Ying, Jiehao Chen, and James J. La Clair*,‡
†
†,§
Department of Chemistry, UniVersity of Minnesota, Minneapolis, Minnesota 55455, and
the Xenobe Research Institute, P.O. Box 4073, San Diego, California 92164-4073
Received October 18, 2005; E-mail: forsyth@chem.umn.edu; i@xenobe.org
Phorboxazoles A and B (1 and 2, Figure 1) were isolated from
1
2
sponges Phorbas sp. and Raspailia sp. collected off Western
Australia in 1993 and 1996, respectively. These scarce natural
products arrest cancer cell growth in S phase at low to subnanomolar
1,3
concentrations, making them promising candidates for therapeutic
development.1 Although substantial efforts have yielded several
a,4
5
total syntheses, the cellular biology of 1 and 2 has remained
6
undefined. Summarized here is a fluorescent label-based study
reliant upon total synthesis5a,7 that elucidates the cellular uptake,
localization, and biomolecular association of the phorboxazoles.
This work provides the first detailed insight into a potential mode
of the phorboxazoles’ unique cytostatic activity.
An abiotic N,N-dialkyl-7-aminocoumarin derivative was selected
as the label because this class of blue fluorescent dye does not
interfere with protein association or inhibit cellular uptake. It was
Figure 1. Structures of phorboxazoles and synthetic variants.
6
previously determined that modification of the C33 hemiketal (3,
Figure 1) or the C45,46-vinyl bromide (4) of phorboxazole A did
not substantially diminish cytostatic activity against a panel of
human cancer cells.3 Therefore, fluorescent analogues 5 and 6
,8
(Figure 1) were prepared via Sonogashira couplings between the
corresponding vinyl iodides and C46 terminal alkynes (see Sup-
porting Information).9
The cellular uptake of 5 and 6 was examined in HeLa cells by
LED fluorescence microscopy.10 The resulting images were com-
11
pared with those obtained using control fluorophores 7-9 and
6
established probes 10-12 (Figure 2). The uptake of 5 and 6 (Figure
2) was comparable to that of protein phosphatase ligand 10,
membrane intercalator 11, and DNA binder 12. Probes 5 and 6
localized on intermediate-sized filaments (IF) and led to a dramatic
restructuring of the IF to form a large aggregate (IFC) adjacent to
the nucleus (Figure 2).
Uptake and localization studies were accompanied by micro-
scopic stop-flow cell cycle staging using a fluorescent staging
cocktail to determine the phase of HeLa cells.12 Probes 5 and 6
induced complete HeLa cell cycle arrest at S phase at 10.2 nM 5
and 8.2 nM 6. Next, HeLa cells were separately treated with 5 and
6
for 4 h, then fractionated into nuclear, cytosolic, endoplasmic
10
reticulum, Golgi, and membrane partitions. Fractions that dis-
played visual fluorescence ([label] > ∼1 µM) were obtained from
the cytosolic partitions derived from each probe. Native PAGE
analyses of the pooled fluorescent fractions yielded fluorescent
bands at ∼32 and ∼54 kDa (Figure 3). However, only the band
near 54 kDa persisted after an affinity purification step using an
amine-reactive agarose gel, Affigel 10, displaying an anti-N,N-
dimethyl-7-aminocoumarin-4-acetamide antibody.
Figure 2. Comparative analysis of cellular uptake, localization, and
response of phorboxazole probes 5 and 6, controls 7, 8, or 9, and organelle
labels 10-12. Images depict the localization of each probe 10 min after
exposure of a 0.34 cm well containing ∼10 HeLa cells/cm with 350 µL
of a 10 nM solution of probe, followed by washing the cells twice with
media. Bars denote the relative intensity of fluorescence obtained per cell.
Localization on the IF was confirmed by immunofluorescent staining with
a green fluorescent anti-cytokeratin mAb (see Supporting Information).
2
6
Samples of the 54 kDa band in the blot obtained using 6 (Figure
the strongest correlations were to human cytokeratins KRT1, KRT9,
and KRT10. This proteomic identification of cytokeratin targeting
a) were submitted to sequence analysis.1 Among three replicates,
3
3
14
†
University of Minnesota.
was consistent with visualized localization of 5 and 6 on IF (Figure
‡
Xenobe Research Institute.
15
2
). KRT10 and the cancer biomarker KRT18 were then cloned
§
Present address: Memorial Sloan-Kettering Cancer Center, New York, NY
1
4
10021, USA.
from a HeLa cDNA library, expressed, and gel purified. The
3858
9
J. AM. CHEM. SOC. 2006, 128, 3858-3859
10.1021/ja057087e CCC: $33.50 © 2006 American Chemical Society

C O M M U N I C A T I O N S
The small-molecule-mediated anchoring of cdk4 upon IF de-
scribed here represents an unexploited paradigm for altering the
activity of cellular proteins. Studies to elucidate the details of the
molecular associations of phorboxazole derivatives and their cellular
receptors and the effects of these interactions upon cell cycle
progression are continuing.
Acknowledgment. Financial support from the NIH (R01
CA099950) is gratefully acknowledged. We thank B. Wang for
helpful discussions, and T. Sch u¨ tz for assistance with protein gel
analysis, cloning, and expression.
Supporting Information Available: Synthetic procedures and
characterization data, and protocols for cellular studies and protein
analyses. This material is available free of charge via the Internet at
http://pubs.acs.org.
Figure 3. Affinity analyses of 5, 6, and 9 in HeLa cell cytosolic fractions.
a) Fluorescent gel analysis depicting fluorescence from 5 and 6 bound to
protein targets, including a native blot analysis using anti-DYE antibody
denoting bound complex. (b) Coomassie-stained SDS-PAGE gel depicting
the pull down of a protein with mass of 32 kDa with an anti-KRT10 affinity
resin in the presence of 5 and 6. (c) Repetition of the pull-down assay in
(
References
(b) using purified cdk4. (d) Blot analysis depicting cdk4 isolated using a
(
1) (a) Searle, P. A.; Molinski, T. F. J. Am. Chem. Soc. 1995, 117, 8126. (b)
Searle, P. A.; Molinski, T. F.; Brzezinski, L. J.; Leahy, J. W. J. Am. Chem.
Soc. 1996, 118, 9422. (c) Molinski, T. F. Tetrahedron Lett. 1996, 37,
KRT10 affinity resin.
7879.
affinities of 5 and 6 were determined by equilibrium dialysis,
(2) Capon, R. J.; Skene, C.; Liu, E. H.; Lacey, E.; Gill, J. H.; Heiland, K.;
yielding association constants (K
and 0.2 and 0.3 nM to KRT18, respectively. Probe 6 was also
assayed against cloned and expressed KRT1 (K ) 32 nM) and
KRT9 (K ) 2.3 nM). Control experiments using 7, 8, and 9 (Figure
) showed insignificant affinity (K > 100 µM) to either KRT10
or KRT18.
a
) of 1.2 and 3.2 nM to KRT10,
Friedel, T. Nat. Prod. Res. 2004, 18, 305.
(3) Uckun, F. M.; Forsyth, C. J. Bioorg. Med. Chem. Lett. 2001, 11, 1181.
(
4) Uckun, F. M.; Narla, R. M.; Forsyth, C. J.; Lee, C. S.; Ahmed, F.; Cink,
R. D. U.S. Patent 6797721, 2004; WO 01/36394, 2001; Priority: US 99-
165452 19991115.
a
a
(
5) (a) Forsyth, C. J.; Ahmed, F.; Cink, R. D.; Lee, C. S. J. Am. Chem. Soc.
2
a
1998, 120, 5597. (b) Evans, D. A.; Fitch, D. M.; Smith, T. E.; Cee, V. J.
J. Am. Chem. Soc. 2000, 122, 10033. (c) Smith, A. B.; Minbiole, K. P.;
Verhoest, P. R.; Schelhaas, M. J. Am. Chem. Soc. 2001, 123, 10942. (d)
Pattenden, G.; Gonz a´ lez, M. A.; Little, P. B.; Millan, D. S.; Plowright,
A. T.; Tornos, J. A.; Ye, T. Org. Biomol. Chem. 2003, 1, 4173. (e)
Williams, D. R.; Kiryanov, A. A.; Emde, U.; Clark, M. P.; Berliner, M.
A.; Reeves, J. T. Angew. Chem., Int. Ed. 2003, 42, 1258. (f) Li, D.-R.; et
al. Chem.sEur. J. 2006, 12, 1185.
We then screened for proteins whose association with KRT10
was affected by 5 and 6. First, resins coated with an anti-KRT10
antibody were used to isolate protein complexes from HeLa cell
lysates exposed to 5 and 6. SDS-PAGE analysis of the proteins
returned from these experiments revealed a second major protein
(6) (a) Sandler, J. S.; Fenical, W.; Gulledge, B. M.; Chamberlin, A. R.; La
Clair, J. J. J. Am. Chem. Soc. 2005, 127, 9320. (b) La Clair, J. J.; et al.
ChemBioChem 2006, in press.
(ca. 32 kDa) that was associated with KRT10 in the presence of
probes 5 and 6 (Figure 3b). LC-MS/MS analysis indicated that
(7) Synthesis and characterization of 3 and 4 are documented in: Lee, C. S.
Ph.D. Thesis, University of Minnesota, Minneapolis, MN, 1999.
this protein was cyclin-dependent kinase 4 (cdk4), an essential
(8) Smith, A. B., III; Razler, T. M.; Pettit, G. R.; Chapuis, J.-C. Org. Lett.
16
component of G1-S phase cell cycle progression and a validated
anticancer drug target.17 This assignment was confirmed by
2005, 7, 4403.
(9) Chen, J.; Ying, L.; Hansen, T. M.; Engler, M. M.; Lee, C. S.; La Clair,
J. J.; Forsyth, C. J. Bioorg. Med. Chem. Lett. 2006, 16, 901.
10) (a) Huber, L. A.; Pfaller, K.; Vietor, I. Circ. Res. 2003, 92, 962. (b) Chen,
X.; Murphy, R. F. J. Biomed. Biotechnol. 2005, 4, 87.
repetition of the binding experiments in vitro using purified cdk4
(
(
(Figure 3c). Next, HeLa cell lysates were screened using a KRT10
11) Images were collected at 100 ms/image at 30 min intervals over 24 h for
each compound. (a) Herman, P.; Maliwal, B. P.; Lin, H. J.; Lakowicz, J.
R. J. Microsc. 2001, 203, 176. (b) Repetto, L.; Piano, E.; Pontiggia, C.
Opt. Lett. 2004, 29, 1132.
12) B o¨ cker, W.; Gantenberg, H.-W.; M u¨ ller, W.-U.; Streffer, C. Phys. Med.
Biol. 1996, 41, 523.
affinity gel. Cdk4 was repetitively returned after extraction of cell
lysates exposed to 5 and 6, but not 9 (Figure 3d). These experiments
provide conclusive evidence that the phorboxazole probes 5 and 6
induce an association between cdk4 and KRT10. Associations were
confirmed by determination of in vitro binding affinities of 5, 6,
(
(
13) Sequencing analyses via trypsin digestion, nanoLC-MS/MS, and Swiss
protein database search were performed at the Mayo Proteomics Research
Center, Mayo Clinic College of Medicine, Rochester, MN.
14) Franke, W. W.; Schmid, E.; Weber, K.; Osborn M. Exp. Cell. Res. 1979,
and 9 to cdk4 by equilibrium dialysis. Analogues 5 (K
and 6 (K ) 120 nM) bound tightly to cdk4, while control 9 (K
µM) did not.
Perturbation of cdk4 structure and function is well-known to
a
) 350 nM)
(
(
a
a
>
118, 95.
5
15) (a) Linder, S.; Havelka, A. M.; Ueno, T.; Shoshan, M. C. Cancer Lett.
2004, 214, 1. (b) Scheuemann, P.; Hosch, S. B.; Izbicki, J. R. Dis.
Esophagus 2001, 14, 85. (c) Chen, P. H.; Ornelles, D. A.; Shenk, T. J.
Virol. 1993, 67, 3508.
18
inhibit cell cycle progression. The sequestration of cdk4 upon
cytosolic IF proteins19 induced by 1-6 would reduce the amount
of free cellular cdk4 and prevent nuclear translocation of active
cdk4-cyclin D1 complexes required for phosphorylation of the
retinoblastoma (Rb) protein and consequent cell cycle progression.16
In a similar model, the human papillomavirus type 16 protein E1-
(16) (a) Kato, J.; Matsushime, H.; Hiebert, S. W.; Ewen, M. E.; Sherr, C. J.
Genes DeV. 1993, 7, 331. (b) Sherr, C. J. Science 1996, 274, 1672.
(
17) Grillo, M.; Bott, M. J.; Khandke, N.; McGinnis, J. P.; Miranda, M.;
Meyyappan, M.; Rosfjord, E. C.; Rabindran, S. K. Breast Cancer Res.
Treat. 2005, Dec 1, 1.
(
18) (a) Paramio, J. M.; Casanova, M. L.; Segrelles, C.; Mittnacht, S.; Lane,
E. B.; Jorcano, J. L. Mol. Cell Biol. 1999, 19, 3086. (b) Mettus, R. V.;
Rane, S. G. Oncogene 2003, 22, 8413. (c) Lee, H. Z.; Leung, H. W.; Lai,
M. Y.; Wu, C. H. Anticancer Res. 2005, 25, 959. (d) Tao, G. Z.; Rott, L.
S.; Lowe, A. W.; Omary, M. B. J. Biol. Chem. 2002, 277, 19295.
2
E4 is believed to cause G cell cycle arrest by retention of cdk1/
cyclin B1 upon cytoplasmic keratin filaments.20 In contrast,
sequestration of PKB/PKCς upon KRT10 apparently occurs without
(19) Coulombe, P. A.; Wong, P. Nat. Cell Biol. 2004, 6, 699.
(
(
20) Davy, C. E.; et al. J. Virol. 2005, 79, 3998.
21) Paramio, J. M.; Segrelles, C.; Ruiz, S.; Jorcano, J. L. Mol. Cell Biol. 2001,
21, 7449.
21
induction by xenobiotic small molecules. This results in inhibition
of cyclin D1 expression and Rb phosphorylation via the phosphoino-
21
(22) Li, J.; Joo, H. S.; Tsai, M.-D. Mol. Biochem. 2003, 42, 13476.
sitide 3-kinase signal transduction pathway. Several endogenous
proteins are known to inhibit cdk4 activity via direct binding.22
JA057087E
J. AM. CHEM. SOC.
9
VOL. 128, NO. 12, 2006 3859