| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
|
FJ
EXPRESS SUMMARY ARTICLE The Full-length version of this article is also available, published online September 5, 2002 as doi:10.1096/fj.02-0026fje. |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Oncology and Molecular Endocrinology Research Center, Laval University Medical Center (CHUL) and Laval University, Quebec, G1V 4G2, Canada
4Correspondence: Oncology and Molecular Endocrinology Research Center, CHUL, 2705 boul. Laurier, Quebec, QC, G1V 4G2, Canada. E-mail: sxlin{at}crchul.ulaval.ca
SPECIFIC AIM
The goal of this study is to design high-affinity inhibitors for human estrogenic type 1 17ß-hydroxysteroid dehydrogenase (17ß-HSD1) to block the synthesis of estrogen, which plays an important role in breast cancer proliferation, and to test the efficacy of a combined study of molecular modeling, chemical synthesis, inhibition kinetics, and crystallography in the rational design of inhibitors for targeted enzymes.
PRINCIPAL FINDINGS
1. Molecular modeling
The estradiol (E2) structure in the 17ß-HSD1-E2 complex at 2.3 Å and the adenosyl portion of the NADH modeled in the structure of the apo-enzyme were originally used to create the preliminary hybrid inhibitor. The nicotinamide ribose diphosphate portion of NADH was deleted and the 5'-carbon of the adenosyl ribose was linked by an alkyl carbonyl spacer to the C16ß position of E2. Based on the initial model with a linker of 10 methylene groups, a series of hybrid compounds with 5, 6, 7, 8, 9, and 12-methylene groups was generated. Results from molecular modeling suggest that the ideal linker length for these hybrid inhibitors appears to be in the range of 8–9 methylenes, with which the basic interactions between the steroid and the enzyme as well as those between the adenosine core and the enzyme can be maintained.
2. Chemical synthesis
To verify the correctness of our modeling and obtain the best inhibitor in this series, we synthesized the hybrid inhibitors with 7–9 methylenes using E1 (estrogen) and adenosine as starting material. After final purification by silica gel chromatograph and full characterization by infrared spectroscopy, 1H and 13C nuclear magnetic resonance, and mass spectrometry, the hybrid compounds were tested as inhibitors of 17ß-HSD1.
3. Study of hybrid inhibition
Inhibition of the 17ß-HSD activity by the hybrid compounds was preliminarily assayed by measuring the IC50 for the formation of E2 from E1 following an established protocol. These E2 adenosine hybrids strongly inhibit 17ß-HSD1 with IC50 values of 93, 52, and 140 nM for linkers with 7, 8, and 9 CH2, respectively. The optimized linker length was evaluated to be 8 methylenes (compound 8; EM-1745, corresponding to n=9 in the modeling). This spacer length allows for optimal interactions of both reactive moieties—namely, E2 and adenosine—with the substrate and cofactor binding sites of the enzyme. EM-1745 shows a high capacity of inhibition (IC50=52 nM), significantly improved over all previously available inhibitors and natural substrates.
The inhibition kinetics of the most efficient hybrid was studied for estradiol oxidation using the fluorescence signal of NADPH formed from the reaction. The Lineweaver-Burk plots for the kinetics, in the absence and presence of varying EM1745 concentrations, intersect on the same point in the 1/v axis (Fig. 1
A), demonstrating a typical reversible competitive inhibition, coincident with the structural composition of the inhibitor. The respective apparent Km values were plotted against inhibitor concentrations, the slope of which yielded Km/Ki. A Ki value of 3.0 ± 0.8 nM was calculated (Fig. 1A
). As a control, we synthesized and studied the inhibition kinetics of compound 10 (an alkyl-adenosine, abbreviation: MB-329–131 A2, Fig. 1B
) and compound 11 (a 16ß-alkyl-E2). For compound 10, competitive inhibition was observed. This inhibitor has a > 80,000 fold lower inhibition (Ki=250 µM) than that of EM-1745. For compound 11, even weaker inhibition was estimated, which did not permit measurement of the Ki within the solubility range of the compound. The results confirm that both the steroid and adenosine cores in EM-1745 are necessary for the function of the hybrid inhibitor.
|
4. Crystallographic structure
The crystal structure of 17ß-HSD1-EM1745 complex was resolved to 1.6 Å. The final model containing 2181 nonhydrogen protein atoms, 1 inhibitor molecule, 3 glycerol molecules, and 231 water molecules, was refined to an R factor of 18.5% and Rfree of 20.5%. The mean B factor of the structure is 21.5 Å2. Both the E2 moiety and the adenosine moiety of the inhibitor demonstrated well-defined electron density at a contour level of 3.0
in the A weighted 2Fo-Fc map. The density of the linker portion can be clearly seen at 2.0 in the same map. After superimposing the structure of 17ß-HSD1-EM-1745 onto that of the 17ß-HSD1-E2-NADP complex, we found that the steroid and adenosine moieties of the inhibitor fit very well into the positions where E2 and NADP had been respectively located in the ternary complex (Fig. 2
). More hydrogen bonds and hydrophobic interactions have been identified between the hybrid inhibitor and the enzyme than those found between ligands and the enzyme in 17ß-HSD1-E2-NADP ternary complex. We suggest that the presence of the steroid and adenosine cores in hybrid inhibitor mutually induces the closed conformation of the enzyme around the binding site.
|
CONCLUSIONS AND SIGNIFICANCE
The affinity of a conjunction of two ligands could be much higher than that of each ligand with respect to the enzyme. Indeed, it was found that the affinity of the reaction intermediate Phe-AMP to yeast phenylalanyl-tRNA synthetase was much higher than that of either Phe or AMP. Moreover, the dissociation constant KPhe·AMP (4x10-9 M) was much closer to the product of KPhe and KAMP than those to either KPhe (30x10-6 M) or KAMP (1x10-3 M) alone. This agrees with theoretical approximations. It was pointed out that if the formation of the covalent bond between Phe and AMP is not followed by essential conformational changes in the enzyme as opposed to the two ligands bound separately (a likely case due to the small size of the substrates), one should expect:
 |
 |
Following a similar hypothesis, synthesizing new types of hybrid inhibitors for 17ß-HSD1 that would combine moieties from NAD(P)H and analogs of E2 was proposed.
Using human estrogenic 17ß-HSD as an example, we have established a rational design of hybrid inhibitors by combining molecular modeling, chemical synthesis, inhibition study, and crystallography. Moreover, the affinity of the hybrid inhibitor has been significantly increased by making use of the binding energy of both moieties. Indeed, the newly synthesized inhibitor EM-1745, with an 8-methylene linker, showed a strong inhibition of the enzyme activity competitive with the substrate with a Ki of 3.0 ± 0.8 nM for the enzyme’s oxidation, which is a significant improvement over previously available inhibitors and natural substrates for the enzyme.
The competitive nature of E2 and the hybrid inhibitor EM-1745 facilitates a comparison of their affinities using Ki and/or KD. As reported earlier, the KD of E2 is 4.7 ± 0.9 µM. In fact, the KD of E2 and the Ki of the compound 10 (250 µM) yield a product of 1.2 nM, which is very close to the experimental Ki of EM-1745 (3.0±0.8 nM). This approximation using E2 and compound 10 as the two cores coincides well with the case of Phe-AMP and phenylalanyl-tRNA synthetase described above. This strongly suggests that our initial idea has worked out as expected.
The crystallographic results have also demonstrated that the new hybrid inhibitor interacts strongly with the enzyme at the steroid and the adenine cores, supporting our initial idea of the hybrid design. Several remarkable phenomena have been observed in the inhibitor binding pocket. 1) The steroid core and the adenine core bind to the enzyme in a manner similar to that observed in the structure of 17ß-HSD1-E2-NADP ternary complex; 2) The electron density is strong in the steroid and the adenine core but somewhat weaker in the linker part, which explains the flexibility in the linker; 3). O17 of steroid core forms strong hydrogen bonds with Tyr155 (2.86 Å) and Ser142 (2.60 Å) as observed in 17ß-HSD1-E2 binary complex structure; 4) a glycerol molecule was defined in the corresponding position of nicotinamide ring of NADP in the structure of 17ß-HSD1-E2-NADP ternary complex; this glycerol contributes to the stability of the hybrid inhibitor by forming a third hydrogen bond to O17 of steroid core; and 5) the strong interactions between the hybrid inhibitor and enzyme induces a close conformation of the substrate entry path compared to free enzyme and several complexes.
Concerning the potential affinity and selectivity from bisubstrate inhibitor, some successful attempts have been reported for enzymes such as insulin receptor tyrosine kinase, serotonin N-acetyltransferase, and acetyl-CoA carboxylase. Further improvement of the hybrids in our study will benefit from the present results with EM-1745. Thus it is shown that rational drug design can provide a strong tool, when a combined approach is used. The present methodology may be applicable to other enzyme systems, especially steroidogenic enzymes.
|
FOOTNOTES
1 To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.02-0026fje; to cite this article, use FASEB J. (September 5, 2002) 10.1096/fj.02-0026fje 
2 These authors contributed equally to the work.
3 Present address: Department of Biophysics and Biophysical Chemistry, School of Medicine, Johns Hopkins University, Baltimore, MD 21205, USA.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
