1α,25-Dihydroxyvitamin D3-3β-bromoacetate, a potential cancer therapeutic agent: Synthesis and molecular mechanism of action

Article abstract of DOI:10.1016/j.bmcl.2011.02.025

Synthesis of 1α,25-dihydroxyvitamin D₃-3β- bromoacetate (1,25(OH)₂D₃-3-BE), a potential anti-cancer agent is presented. We also report that mechanism of action of 1,25(OH) ₂D₃-3-BE may involve reducti

Full text of DOI:10.1016/j.bmcl.2011.02.025

Bioorganic & Medicinal Chemistry Letters 21 (2011) 2537–2540
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
1a
,25-Dihydroxyvitamin D3-3b-bromoacetate, a potential cancer
therapeutic agent: Synthesis and molecular mechanism of action
Rahul Ray a, , James R. Lambert b
⇑
a Department of Medicine, Boston University School of Medicine, Boston, MA 02118, USA
b Department of Pathology, University of Colorado, Denver, Aurora, CO 80045, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Synthesis of 1
a,25-dihydroxyvitamin D3-3b-bromoacetate (1,25(OH)2D3-3-BE), a potential anti-cancer
Received 3 January 2011
Revised 3 February 2011
Accepted 8 February 2011
Available online 18 February 2011
agent is presented. We also report that mechanism of action of 1,25(OH)2D3-3-BE may involve reduction
of its catabolism, as evidenced by the reduced and delayed expression of 1
D3-24-hydroxylase (CYP24) gene in cellular assays.
a,25-dihydroxyvitamin
Ó 2011 Elsevier Ltd. All rights reserved.
Key Words:
1
1
a
a
,25-Dihydroxyvitamin D3 (1,25(OH)2D3)
,25-Dihydroxyvitamin D3-3b-
bromoacetate (1,25(OH)2D3-3-BE)
Vitamin D receptor (VDR)
Vitamin D receptor-ligand binding domain
(VDR-LBD)
Synthesis of 1,25(OH)2D3-3-BE
Anti-cancer agent
Mechanism of action of 1,25(OH)2D3-3-BE
1a,25-Dihydroxyvitamin D3-24-
hydroxylase (CYP24)
The awareness of the potential beneficial effect of vitamin D in a
number of diseases, and general good health has increasingly be-
come a public health issue.1 In addition, the therapeutic potential
analogs (of 1,25(OH)2D3) have been synthesized based on this con-
cept with the hope of achieving a favorable antiproliferative/toxic-
ity index, yet clinical results are largely disappointing.5,6
of 1
a
,25-dihydroxyvitamin D3 (1,25(OH)2D3), the dihydroxylated
On the other hand, recent clinical trials of 1,25(OH)2D3 either
alone or in combination with common cancer therapeutic drugs
have produced strongly encouraging results, and brought back
the realization that 1,25(OH)2D3 is probably the most promising
‘drug’ if its therapeutic dose can be escalated without causing
toxicity.7–10 High doses of 1,25(OH)2D3 are required to counter
its non-tumor specific spread, and high catabolic rate. Realizing
metabolite of vitamin D3, on established tumors was realized when
the potent antiproliferative property of 1,25(OH)2D3 was discov-
ered in the early 1980’s in leukemic cells.2 Since then the antipro-
liferative and pro-differentiative effects of 1,25(OH)2D3 have been
studied in various cancer cell lines, and as a result, 1,25(OH)2D3
has been investigated as a potential anti-cancer agent.3 However,
clinical trials have demonstrated strong calcemic and calciuric ef-
fects at pharmacological doses (of 1,25(OH)2D3).4 This limitation
provided the search for structural analogs of 1,25(OH)2D3 with
potent antiproliferative yet low calcemic properties.4 This
effort, based solely on a Specific Estrogen Receptor Modulator
(SERM)-like concept adopted the idea that these analogs would
change the conformation of nuclear vitamin D receptor (VDR),
the chief modulator of the biological actions of 1,25(OH)2D3
sufficiently from the parent hormone to separate cell-regulatory
properties from calcemic property. Numerous such ‘non-calcemic’
this unmet potential of 1,25(OH)2D3-therapy, we adopted
a
hypothesis-based alternate approach to increase the half-life of
1,25(OH)2D3, to thus decrease effective dose with reduced toxicity.
1a,25-Dihydroxyvitamin D3-3b-bromoacetate (1,25(OH)2D3-3-
BE) was initially developed in our laboratory as an affinity labeling
reagent to map the ligand-binding domain (LBD) of VDR.11 Subse-
quently, we realized that since 1,25(OH)2D3-3-BE covalently at-
taches to VDR-LBD,12–15 it might be physically protected from
interacting with catabolic enzymes, thereby increasing its half-life.
Furthermore, once attached to the VDR-LBD, 1,25(OH)2D3-3-BE
becomes a simple 3-acetate derivative of 1,25(OH)2D3, activating
gene transcription similar to 1,25(OH)2D3. In other words,
1,25(OH)2D3-3-BE becomes de facto 1,25(OH)2D3 with an increased
⇑
Corresponding author. Tel.: +1 617 638 8199; fax: +1 617 638 8194.
E-mail address: bapi@bu.edu (R. Ray).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.02.025

2538
R. Ray, J. R. Lambert / Bioorg. Med. Chem. Lett. 21 (2011) 2537–2540
half-life by this process. We observed that 1,25(OH)2D3-3-BE has a
significantly greater antiproliferative effect than 1,25(OH)2D3 in
normal human keratinocytes, supporting our hypothesis.16
Subsequently several publications from our group and others have
demonstrated significantly stronger antiproliferative effect of
1,25(OH)2D3-3-BE and its mono-hydroxylated analog, 25-hydrox-
yvitamin D3-3b-bromoacetate than 1,25(OH)2D3 in various cancer
cells.17–22 We also observed strong anti-tumor effect of
1,25(OH)2D3-3-BE in mouse model of renal cancer,22 and andro-
gen-insensitive prostate cancer (manuscript in preparation).
The above mentioned results suggest strong therapeutic poten-
tial of 1,25(OH)2D3-3-BE, and warrant extensive in vitro and in vivo
studies, requiring substantial quantity of 1,25(OH)2D3-3-BE. Previ-
ously, we reported synthesis of 1,25(OH)2D3-3-BE with a starting
material that is no longer available23 mandating an alternative
synthetic procedure for this compound, which is delineated in this
Letter. Furthermore, our hypothesis about increased stability of
1,25(OH)2D3-3-BE by making it less available to catabolic enzymes
remains unproven to date. In this Letter, we report results of cellu-
lar studies to support this hypothesis on the catabolism of this
analog.
in a stereospecific manner. In an earlier publication,25 we reported
that N-methylmorpholine-N-oxide/SeO2allylic oxidation of 25-hy-
droxy-5E-[6,19,190-H3]vitamin D3-3b-tert-butyldimethylsilyl ether
to
methylsilyl ether produced an approximately 6:1 ratio of
-OH:1b-OH stereoisomers. However, in the present case, HPLC
1a
,25-dihydroxy-5E-[6,19,190-H3]vitamin D3-3b-tert-butyldi-
1a
analysis of the crude reaction mixture showed the product to be
homogeneous, indicating the presence of a single stereoisomer
(D) in the reaction mixture, which, after purification, was identified
to have the 1a-OH (desired) stereochemistry by comparison with
the NMR spectrum of a known sample.25 We speculate that the
lack of three deuterium atoms in the triene system adjoining the
A-ring in C may have changed the conformation of the A-ring sub-
stantially to have the oncoming oxygen atom introduced exclu-
sively from one side. We also found that stereochemistry at the
1-position depended on the structure of the 1-silyl ether. For
example, when 1-OH was derivatized with a triethylsilyl group,
the 1a-OH:1b-OH ratio was approximately 4:1 with an overall
yield of 45%.
In the next step 1a,25-dihydroxy-5E-vitamin D3-3b-tert-butyl-
dimethylsilyl ether (D) was converted to its 5Z counterpart (E)
by photolysis. Attempts to protect the 1-hydroxyl group in E as tet-
rahydropyranyl (THP) ether resulted in the derivatization of the
Synthesis of 1,25(OH)2D3-3-BE was initiated by converting
25-hydroxyvitamin D3 (A, Duphar Chemicals, Netherlands) to its
trans-variety (B) by the sulfur-dioxide method24 and converting
the crude into the tert-butyldimethylsilylether (TBDMS) derivative
(C) (Scheme 1) in good yield.
25-hydroxy group as well, and produced 1
nyloxy, 3b-tert-butyldimethysilyl-5Z-vitamin D3 (F), which
was de-silylated to produce 1 ,25-di-tetrahydropyranyloxy-5Z-
vitamin D3 (G). Finally, was DCC-coupled to bromoacetic
acid, and the resulting product was de-protected to produce
a,25-di-tetrahydropyra-
a
Success of our synthetic scheme depended critically on the
introduction of the 1-hydroxyl group into (C) by allylic oxidation
G
OH
OH
/
OH
H
H
H
O2S
NMO / SeO2
CH2Cl2 / Reflux
/
NaHCO3
1. Liq. SO2 / Reflux
EtOH / Reflux
2. TBDMSCl /
Imidazole / DMF
HO
OTBDMS
OTBDMS
(A)
(D)
(G)
(C)
(B)
OH
OTHP
OH
H
H
H
TBAF / THF
UV / Toluene - Et3N
DHP / PPTS /
CH2Cl2
TBDMSO
OTHP
OH
TBDMSO
OTHP
HO
OTBDMS
(F)
(E)
OH
OTHP
H
H
H
BrCH2COOH /
DCC / DMAP /
CH2Cl2
HO
OTHP
BrCH2COO
OH
BrCH2COO
OTHP
1,25(OH)2D3-3-BE (I)
(H)
Scheme 1. Scheme for the synthesis of 1,25(OH)2D3-3-BE.

R. Ray, J. R. Lambert / Bioorg. Med. Chem. Lett. 21 (2011) 2537–2540
2539
Figure 1. Cartoon depicting interaction between 1,25(OH)2D3, 1,25(OH)2D3-3-BE and VDR. Points to note: (i) reversible and irreversible nature of interaction of 1,25(OH)2D3
and 1,25(OH)2D3-3-BE, respectively, with VDR, leading to different catabolic outcomes, (ii) covalent attachment of 1,25(OH)2D3-3-BE via Cys288 in VDR-LBD, (iii) differential
conformational changes of VDR-LBD induced by 1,25(OH)2D3 and 1,25(OH)2D3-3-BE.
1
a
,25-dihydroxy-5Z-vitamin D3-3b-bromoacetate (I, 1,25(OH)2D3-
of 1,25(OH)2D3 or 1,25(OH)2D3-3-BE or ethanol vehicle (control).
As shown in Figure 2, treatment of A498 cells with 10ꢀ6 M of
1,25(OH)2D3 strongly induced CYP24 mRNA expression. 1,25(OH)2-
D3-3-BE treatment also led to induction of CYP24 mRNA, but at a
significantly higher concentration (approximately 10-fold) to
achieve induction comparable to 1,25(OH)2D3, suggesting
1,25(OH)2D3-3-BE is less efficacious in inducing CYP24 than
1,25(OH)2D3, as predicted by our hypothesis.
3-BE) in nearly quantitative yield.
Signal transduction via 1,25(OH)2D3 is a stepwise process that is
initiated by the highly specific binding between 1,25(OH)2D3 and
VDR. Ligand binding allosterically promotes formation of a hetero-
meric complex with RXR, which allows the complex to bind to
vitamin D response element(s) in chromatin and recruitment of
co-activators, leading to transcription of vitamin D target genes.
According to this dogma, VDR-binding by 1,25(OH)2D3 as well as
its analogs/derivatives such as 1,25(OH)2D3-3-BE is crucial for all
these processes.
1,25(OH)2D3-3-BE is a VDR affinity alkylating agent. In an ear-
lier study, we described that 1,25(OH)2D3-3-BE is capable of specif-
ically labeling native VDR in ROS 17/2.8 bone cells and calf thymus
nuclear extract,12 and rapidly titrating recombinant VDR with 1:1
stoichiometry.16 We also reported that 1,25(OH)2D3-3-BE cova-
lently labels a single cysteine residue (Cys288) in the VDR-LBD (as
shown in the cartoon in Fig. 1).14 Furthermore, we demonstrated
that VDR is essential for the growth inhibitory activity of
25-hydroxyvitamin D3-3-BE, a prototype of 1,25(OH)2D3-3-BE
without the 1-hydroxyl group in ALVA-31 prostate cancer cells.19
The other mechanistic aspect of our hypothesis delineates that
1,25(OH)2D3-3-BE, after interacting with VDR, is protected from
catabolism, thus increasing its half-life, and enhancing its
anti-growth property as evidenced in several studies.17–22
In the kinetic experiment cells were treated with 10ꢀ8 M of
either 1,25(OH)2D3 or 1,25(OH)2D3-3-BE and CYP24 message was
detected at various time-points. As shown in Figure 3, induction
of CYP24 mRNA was detectable by 4 h in 1,25(OH)2D3-treated cells
and continued to increase over 16 h. In contrast, 1,25(OH)2D3-3-BE
showed a much delayed kinetics of induction of CYP24 mRNA, with
CYP24 mRNA first being detectable at 16 h.
1a,25-Dihydroxyvitamin D3-24-hydroxylase (CYP24) is a cyto-
chrome P450 enzyme that introduces a hydroxyl group at the
24-position in the side chain of 1,25(OH)2D3 that initiates the mul-
ti-step catabolic degradation.26 Interaction between VDR and
1,25(OH)2D3 is an equilibrium process so that the steady state
would always contain ‘free’ 1,25(OH)2D3 (not bound to VDR) which
will be rapidly catabolized by CYP24 (Fig. 1).
In contrast, covalent attachment of 1,25(OH)2D3-3-BE inside the
ligand-binding pocket of VDR14 would eliminate/reduce ‘free’
1,25(OH)2D3 in the steady state, thereby potentially requiring re-
duced and delayed expression of CYP24 (Fig. 1).
Figure 2. Comparison of the dose dependent increase of CYP24 mRNA by
1,25(OH)2D3 and 1,25(OH)2D3-3-BE. A498 cells were treated with ethanol control
or 10ꢀ6 M of 1,25(OH)2D3 and 1,25(OH)2D3-3-BE for 24 h, total RNA was prepared
and CYP24 mRNA levels assessed by RT-PCR. b-Actin mRNA levels were determined
for each sample as control. The results are representative of two independent
experiments.
We chose A498 kidney cancer cells to test our hypothesis. In the
dose–response experiment A498 cells were treated with 10ꢀ6
M

2540
R. Ray, J. R. Lambert / Bioorg. Med. Chem. Lett. 21 (2011) 2537–2540
References and notes
1. Holick, M. F. Nutr. Rev. 2008, 66, S182.
2. Abe, E.; Miyaura, C.; Sakagami, H.; Takeda, M.; Konno, K.; Yamazaki, T.; Yoshiki,
S.; Suda, T. Proc. Natl. Acad. Sci. U.S.A. 1981, 78, 4990.
3. Deeb, K. K.; Trump, D. L.; Johnson, C. S. Nat. Rev. Cancer 2007, 7, 684.
4. Matsuda, S.; Jones, G. Mol. Cancer Ther. 2006, 5, 797.
5. Dalhoff, K.; Dancey, J.; Astrup, L.; Skovsgaard, T.; Hamberg, K. J.; Lofts, F. J.;
Rosmorduc, O.; Erlinger, S.; BachHansen, J.; Steward, W. P.; Skov, T.; Burcharth,
F.; Evans, T. R. Br. J. Cancer 2003, 89, 252.
6. Evans, T. R.; Colston, K. W.; Lofts, F. J.; Cunningham, D.; Anthoney, D. A.; Gogas,
H.; de Bono, J. S.; Hamberg, K. J.; Skov, T.; Mansi, J. L. Br. J. Cancer 2002, 86, 680.
7. Muindi, J. R.; Johnson, C. S.; Trump, D. L.; Christy, R.; Engler, K. L.; Fakih, M. G.
Cancer Chemother. Pharmacol. 2009, 65, 33.
8. Fakih, M. G.; Trump, D. L.; Muindi, J. R.; Black, J. D.; Bernardi, R. J.; Creaven, P. J.;
Schwartz, J.; Brattain, M. G.; Hutson, A.; French, R.; Johnson, C. S. Clin. Cancer
Res. 2007, 13, 1216.
9. Beer, T. M.; Javle, M. M.; Ryan, C. W.; Garzotto, M.; Lam, G. N.; Wong, A.;
Henner, W. D.; Johnson, C. S.; Trump, D. L. Cancer Chemother. Pharmacol. 2007,
59, 581.
10. Trump, D. L.; Potter, D. M.; Muindi, J.; Brufsky, A.; Johnson, C. S. Cancer 2006,
106, 2136.
11. Ray, R.; Ray, S.; Holick, M. F. Bioorg. Chem. 1994, 22, 276.
12. Ray, R.; Swamy, N.; MacDonald, P. N.; Ray, S.; Haussler, M. R.; Holick, M. F. J.
Biol. Chem. 1996, 271, 2012.
Figure 3. Kinetics of CYP24 message induction by 1,25(OH)2D3 and 1,25(OH)2D3-3-
BE. A498 cells were treated with 10ꢀ8 M of 1,25(OH)2D3 and 1,25(OH)2D3-3-BE or
ethanol control for 16 h, total RNA was prepared at specified times and CYP24
mRNA levels assessed by RT-PCR. b-Actin mRNA levels were determined for each
sample as control. The results are representative of two independent experiments.
13. Swamy, N.; Kounine, M.; Ray, R. Arch. Biochem. Biophys. 1997, 348, 91.
14. Swamy, N.; Xu, W.; Paz, N.; Hsieh, J.-C.; Haussler, M. R.; Maalouf, G. J.; Mohr, S.
C.; Ray, R. Biochemistry 2000, 39, 12162.
15. Mohr, S. C.; Swamy, N.; Xu, W.; Ray, R. Steroids 2001, 66, 189.
16. Chen, M. L.; Ray, S.; Swamy, N.; Holick, M. F.; Ray, R. Arch. Biochem. Biophys.
1999, 370, 34.
17. Swamy, N.; Persons, K. S.; Chen, T. C.; Ray, R. J. Cell. Biochem. 2003, 89, 909.
18. Swamy, N.; Chen, T. C.; Peleg, S.; Dhawan, P.; Christakos, S.; Stewart, L. V.;
Weigel, N. L.; Mehta, R. G.; Holick, M. F.; Ray, R. Clin. Cancer Res. 2004, 10, 8018.
19. Lambert, J. L.; Young, C. D.; Persons, K. S.; Ray, R. Biochem. Biophys. Res.
Commun. 2007, 361, 189.
20. Lange, T. S.; Singh, R. K.; Kim, K. K.; Zou, Y.; Kalkunte, S. S.; Sholler, G. L.;
Swamy, N.; Brard, L. Chem. Biol. Drug Des. 2007, 70, 302.
21. Persons, K. S.; Eddy, V. J.; Chadid, S.; Deoliveira, R.; Saha, A. K.; Ray, R.
Anticancer Res. 2010, 30, 1875.
22. Lambert, J. R.; Eddy, V. J.; Young, C. D.; Persons, K. S.; Sarkar, S.; Kelly, J. A.;
Genova, E.; Lucia, M. S.; Faller, D. V.; Faller, R. Cancer Prev. Res. (Phila) 2010, 3,
1596.
23. Ray, R.; Holick, S. A.; Holick, M. F. J. Chem. Soc., Chem. Commun. 1985, 11, 702.
24. Yamada, S.; Suzuki, t.; Takayama, H.; Miyamoto, K.; Matsunaga, I.; Nawata, Y. J.
Org. Chem. 1983, 48, 3483.
Taken together, results of Figures
2 and 3 suggest that
1,25(OH)2D3-3-BE is less effective in inducing CYP24 gene
expression than an equivalent amount of 1,25(OH)2D3. CYP24 is a
catabolic enzyme that initiates the degradation of 1,25(OH)2D3.
Therefore, reduced and delayed expression of CYP24 gene implies
decreased catabolism of 1,25(OH)2D3-3-BE, as predicted by our
hypothesis.
We reported earlier that 1,25(OH)2D3-3-BE changes the confor-
mation of VDR differently from 1,25(OH)2D3 as reflected in the en-
hanced stabilization of VDR-hOCVDRE (human osteocalcin vitamin
D response element) complex in COS-1 cells and promotion of a
longer stimulation of CYP24 mRNA expression in keratinocytes
compared with 1,25(OH)2D3.16 Furthermore, recently we reported
that compounds similar to 1,25(OH)2D3-3-BE with alkylating bro-
moacetate group at 1- and 11-positions of 1,25(OH)2D3 specifically
labeled VDR-LBD, yet their antiproliferative activity in keratino-
cytes was similar to that of 1,25(OH)2D3.27 Therefore, considering
all the information we ascribe enhanced antiproliferative activity
of 1,25(OH)2D3-3-BE to a combination of its increased half-life
25. Ray, R.; Vicchio, D.; Yergey, A.; Holick, M. F. Steroids 1992, 57, 142.
26. Omdahl, J. L.; Swamy, N.; Serda, R.; Annalora, A.; Berne, M.; Ray, R. J. Steroid
Biochem. Mol. Biol. 2004, 89–90, 159.
27. Kaya, T.; Swamy, N.; Persons, K. S.; Ray, S.; Ray, R. Bioorg. Chem. 2009, 37, 57.
28. Bouillon, R.; Carmeliet, G.; Verlinden, L.; van Etten, E.; Verstuyf, A. Luderer,
H.F.; Lieben, L.; Mathieu, C.; Demay, M. Endocr. Rev. 2008, 6, 726.
Notes
(by covalent labeling of VDR-LBD) and
VDR-conformation upon binding (Fig. 1).
a unique change in
Several non-genomic pathways for the action of 1,25(OH)2D3
and its analogs have also been reported.28 We have observed that
PI3K/Akt non-genomic pathway is involved in the activity of
1,25(OH)2D3-3-BE in kidney cancer cells.22 Therefore, mechanism
of action of 1,25(OH)2D3-3-BE may involve a combination of geno-
mic and non-genomic pathways.
Synthesis: (i)
A mixture of (C) (760 mg) and SeO2 (192 mg) in 15 ml of
anhydrous CH2Cl2 was refluxed under argon for 30 min followed by cooling to
room temperature and addition of a solution of N-methylmorpholine-N-oxide
(850 mg) in 15 ml of anhydrous CH2Cl2. The mixture was refluxed for an
additional 60 min. After usual workup and purification trans-1
dihydroxyvitamin D3-3b-TBDMS ether (D) was obtained in 95% yield.
(ii) A toluene (10 ml) solution of D (80 mg), anthracene (10 mg), Et3N (40
a,25-
l
l) in
a quartz test tube was irradiated from a Hanovia medium pressure mercury arc
In conclusion, we report an efficient synthesis of
1,25(OH)2D3-3-BE, a potential therapeutic agent for cancer. In
addition, we provide data from cellular studies to support our
hypothesis about its intrinsic mechanism of action.
lamp for 75 min. After preparative TLC purification 1a,25-dihydroxyvitamin
D3-3-TBDMS ether (E) was obtained in 67% yield.
Cellular studies: (i) Induction of 1a,25-dihydroxyvitamin D3-24hydroxylase
(CYP24) gene expression by 1,25(OH)2D3 and 1,25(OH)2D3-3-BE (dose–
response): A498 kidney cancer cells (ATCC, Manasas, VA) were treated with
10ꢀ6 M of 1,25(OH)2D3, 1,25(OH)2D3-3-BE or ethanol (vehicle control) for 24 h.
Total RNA was prepared and subjected to reverse transcription using Moloney
murine leukemia virus (MMLV) reverse transcriptase and random hexamers
under standard conditions. Following cDNA synthesis, PCR was performed
using gene specific primers to the vitamin D target gene, CYP24 and b-actin
(control). The products were analyzed on 1% agarose gels.
(ii) Induction of CYP24 gene expression by 1,25(OH)2D3-3-BE (kinetics): In this
experiment, A498 cells were treated with 10ꢀ8 M of 1,25(OH)2D3-3-BE and
1,25(OH)2D3 and RNA prepared over the course of 16 h for the analysis of
CYP24 mRNA levels by RT-PCR.
Acknowledgments
This work was supported by National Cancer Institute Grant CA
127629, Department of Defense Grant PC 051136, National Cancer
Institute CA126317 (sub-contract) and Community Technology
Fund grant, Boston University to R.R., and American Cancer Society
Research Scholar Grant RSG-04-170-01-CNE to J.R.L.