Putative irreversible inhibitors of the human sodium-dependent bile acid transporter (hASBT; SLC10A2) supportthe role of transmembrane domain 7 in substrate binding/translocation

Supplementary Material

Pharmaceutical Research

Pablo M. González*, Naissan Hussainzada, Peter W. Swaan, Alexander D. MacKerell Jr., and James E. Polli

*Departamento de Farmacia, Facultad de Química, Pontificia Universidad Católica de Chile, Santiago, Chile

Table of Contents

1. Synthesis of Electrophilic CDCA Derivatives[including Scheme 1 (Synthetic route to obtain 3β-Cl-CDCA) and Scheme 2 (Synthetic route for the preparation of 7α-Ms-CDCA] page 3

Fig. S1.A. 13C-NMR 3-Cl-CDCApage 5

Fig. S1.B. 1H-NMR 3-Cl-CDCApage 5

Fig. S2.A. 13C-NMR 7-Ms-CDCApage 6

Fig. S2.B. 1H-NMR 7-Ms-CDCApage 6

Fig S3.hASBT-inhibition profile of 7-Ms-CDCApage 7

Fig S4.hASBT-inhibition profile of 3-Cl-CDCApage 7

Fig. S5.A. Remaining hASBT activity after pre-incubation with 3-Cl-CDCApage 8

Fig. S5.B. Kitz and Wilson plot for 3-Cl-CDCApage 8

Fig. S6. Remaining hOCTN2 activity after pre-incubation with 3-Cl-CDCApage 9

Table S1.

Appendix 1 page 10

REFERENCES page 12

1. Synthesis of Electrophilic CDCA Derivatives.

The putatively irreversible hASBT inhibitors 3-chloro-7-hydroxy-5-cholan-24-oic acid (5) and 3-hydroxy-7-mesyloxy-5-cholan-24-oic acid (9) (3-Cl-CDCA, and 7-Ms-CDCA, respectively) were synthesized from 3,7-dihydroxy-5-cholan-24-oic acid (chenodeoxycholic acid, 1) in several steps. First, the benzyl ester of 1 was prepared using DMF as solvent. Chenodeoxycholic acid benzyl ester (2) was obtained in almost quantitative yield. Ester 2 was then reacted with p-toluenesulfonyl chloride (TsCl) in pyridine (1), obtaining benzyl 3-(tosyloxy)-7-hydroxy-5-cholanoate (3) in 82% yield. Chlorination of the 3-position was achieved by reacting 3 with an excess of pyridine hydrochloride in pyridine at 90oC as described previously (2). The substitution product 4 was purified by column chromatography in mixture ethyl acetate/hexanes and obtained in 57% yield. Catalytic hydrogenation of benzyl ester 4 produced 3-Cl-CDCA (5) as a white fluffy solid (Scheme S1).

Scheme 1. Synthetic route to obtain 3-Cl-CDCA

7-Ms-CDCA (9) was obtained from chenodeoxycholate benzyl ester 2 after several steps (Scheme 2). First, 3-oxo-7-hydroxy-5-cholanoate benzyl ester (6) was obtained by refluxing 2 and 50% Ag2CO3/celite in toluene as described previously (3). Compound 6 was obtained in 89% yield after chromatographic purification. 3-oxo-7-mesyloxy-5-cholanoate benzyl ester (7) was prepared as described for 3 using an excess of mesyl chloride (MsCl) and refluxing in pyridine (68% yield). The 3-oxo-precursor 7 was reduced to the 3-hydroxy derivative using NaBH4 in methanol (4) (75% yield after chromatography). 7-Ms-CDCA (9) was obtained as an off-white solid after catalytic hydrogenolysis.

1D 13C-NMR and 1H-NMR spectra of final products 5 and 9are shown in Figs S1 and S2 panes A and B, respectively.

Scheme 2. Synthetic route for the preparation of 7-Ms-CDCA.

Fig. S1.A. 13C-NMR 3-Cl-CDCA

Fig. S1.B. 1H-NMR 3-Cl-CDCA

Fig. S2.A.13C-NMR 7-Ms-CDCA

Fig. S2.B.1H-NMR 7-Ms-CDCA

Fig S3. hASBT-inhibition profile of 7-Ms-CDCA

Fig. S4. hASBT-inhibition profile of 3-Cl-CDCA

Fig. S5.A. Remaining hASBT activity after pre-incubation with 3-Cl-CDCA

Fig. S5.B. Kitz and Wilson plot for 3-Cl-CDCA

Fig. S6. Remaining hOCTN2 activity after pre-incubation with 3-Cl-CDCA

Table S1.

hASBT inactivation rate and inhibitory affinity of 3-Cl-CDCA and 7-Ms-CDCA obtained from Eqn2.

Parameter / 3-Cl-CDCA / 7-Ms-CDCA
k3 (min-1) / 0.399 / 0.136
Ki (M) / 2.17 / 3.59

Appendix 1

The objective of this appendix is to show the derivation of the irreversible inhibition model. Kitz and Wilson provided the model in the following form:

(A1)

where J is remaining TCA flux after pre-incubation (i.e. remaining hASBT activity), J0 is TCA flux without pre-incubation, k3 is inactivation rate, Ki is binding affinity, I is the irreversible inhibitor concentration, and t is pre-incubation time (i.e. duration of pre-incubation). Eqn A1 is the same as eqn 1. However, Kitz and Wilson did not show the derivation in great detail, perhaps since eqn A1 was not directly used in regression analysis. A detailed derivation is provided here, since the present manuscript directly employs a form of the equation for regression analysis.

In the model for irreversible inhibition, the amount of active transporter is:

(A2)

where E is the amount of transporter that is not irreversibly inactivated, Eu is the amount of unbound transporter, and EI is the amount of transporter-inhibitor complex prior to inactivation.

(A3)

Substituting eqn A2 into eqn A3 and solving for EI yields

(A4)

The rate of loss of amount of transporter due to inactivation is

(A5)

Substituting eqn A4 into eqn A5 yields

(A6)

Solving eqn A6 yields

(A7)

where E0 is the initial amount of transporter.

The amount of transporter is assessed functionally by measuring active TCA flux. Hence, eqn A7 yields

(A8)

Eqn A8 is Eqn 1 in the main text.

REFERENCES

1.Iida T, Chang F. Potential bile acid metabolites. 7. 3,7,12-trihydroxy-5beta-cholanic acids and related compounds. Journal of Organic Chemistry. 1982 January, 1982;47:2972-8.

2.Chang F, Blickenstaff R, Feldstein A, Gray J, McCaleb G, Sprunt D. Seroflocculating Steroids. III. Chloro and other bile acid derivatives. Journal of the American Chemical Society. 1957;79:2164-7.

3.Jones S, Selinsky B. Efficient route to 7alpha-(benzyloxy)-3-(dioxolane-Cholestane-24(R)-ol, a key intermediate in the synthesis of squalamine. Journal of Organic Chemistry. 1998;63:3786-9.

4.Uekawa T, Ishigami K, Kithara T. Short-step synthesis of chenodiol from sigmasterol. Bioscience, Biothechnology and Biochemistry. 2004;68(6):1332-7.

1