Towards Predictable Transmembrane Transport: QSAR Analysis of the Anion Binding and Anion Transport Properties of Thioureas

Towards Predictable Transmembrane Transport: QSAR Analysis of the Anion Binding and Anion Transport Properties of Thioureas

Philip A. Gale

N. Busschaert, S.J. Bradberry, M. Wenzel, C.J.E. Haynes, J.R. Hiscock, I.L. Kirby, L.E. Karagiannidis, S.J. Moore, N.J. Wells, J. Herniman, G.J. Langley, P.N. Horton, M.E. Light, V. Félix, J.G. Frey, P. A. Gale, Chemical Science 2013, 4, 3036-3045 .

Tools from medicinal chemistry• Hansch analysis: used in medicinal chemistry to examine

quantitative structure-activity relationships across a series of molecules. The log(1/IC50) value is correlated to a linear combination of predictable molecular properties, usually logP (water:octanol partition coefficient) and/ or Hammett constants using multiple regression techniques:

log(1/IC50) = k1π + k2σ + k3

where IC50 = molar concentration required for a standard response

π is related to lipophilicity (logP)

σ = Hammett coefficient (related to binding strength)

C. Hansch, A. Leo and R. W. Taft, Chem. Rev., 1991, 91, 165.

variation of hydrogen bond donor strength (σ) and lipophilicity

thiourea anion binding site

Transporter designR

NH

NH

S

1 R = Br2 R = CF33 R = Cl4 R = CN5 R = COCF36 R = COMe7 R = COOMe8 R = F9 R = H10 R = I11 R = NO2

12 R = O(CO)Me13 R = OCF314 R = OEt15 R = OMe16 R = SMe17 R = SO2Me18 R = CH319 R = CH2CH320 R = (CH2)2CH321 R = (CH2)3CH322 R = (CH2)4CH3

Chloride efflux promoted by a selection of compounds 1-22 (2 mol% thiourea to lipid) from unilamellar POPC vesicles loaded with 489 mM NaCl buffered to pH 7.2 with 5 mM sodium phosphate salts. The vesicles were dispersed in 489 mM NaNO3 buffered to pH 7.2 with 5 mM sodium phosphate salts. At the end of the experiment, detergent was added to lyse the vesicles and calibrate the ISE to 100 % chloride efflux. Each point represents the average of at least 9 trials. DMSO was used as control.

Cl-NO3-

Anion transport at 2% loading

Anion transport trendsHill analysis !E = EmaxCn/(EC50

n+Cn) !E is the magnitude of an observed effect Emax is the maximum value of this effect (in this case Emax = 100 % chloride efflux) C is the concentration of carrier n is the Hill coefficient of sigmoidality EC50 is the effective concentration of carrier required to mediate 50 % of the maximum response

Br

N

S

N

H H

EC50 = 0.80 mol% w.r.t. lipid

A. V. Hill, Biochem. J., 1913, 7, 471.

variation of hydrogen bond donor strength (σ) and lipophilicity

thiourea anion binding site

Transporter designR

NH

NH

S

1 R = Br2 R = CF33 R = Cl4 R = CN5 R = COCF36 R = COMe7 R = COOMe8 R = F9 R = H10 R = I11 R = NO2

12 R = O(CO)Me13 R = OCF314 R = OEt15 R = OMe16 R = SMe17 R = SO2Me18 R = CH319 R = CH2CH320 R = (CH2)2CH321 R = (CH2)3CH322 R = (CH2)4CH3

variation of hydrogen bond donor strength (σ) and lipophilicity

thiourea anion binding site

Transporter designR

NH

NH

S

2 R = CF33 R = Cl4 R = CN5 R = COCF3

7 R = COOMe8 R = F9 R = H10 R = I11 R = NO2

12 R = O(CO)Me13 R = OCF3

15 R = OMe16 R = SMe17 R = SO2Me18 R = CH319 R = CH2CH3

21 R = (CH2)3CH322 R = (CH2)4CH3

Hansch analysis

!

log(1/EC50) = k1A + k2B + k3C + k4

N. Busschaert, S.J. Bradberry, M. Wenzel, C.J.E. Haynes, J.R. Hiscock, I.L. Kirby, L.E. Karagiannidis, S.J. Moore, N.J. Wells, J. Herniman, G.J. Langley, P.N. Horton, M.E. Light, V. Félix, J.G. Frey, P. A. Gale,

Chemical Science 2013, 4, 3036-3045 .

Mr (g/mol) PSA solvent acc surface area

volume Surface Area Molecular Volume Vsmax Vsmin PI Molar refractivity Molecular volume

Parachor Index of refraction TPSA Bp

logD (pH1,7) logD (pH4,6) logD (pH6,5) logD (pH7,4) logD (pH8)

pKa (arom NH) pKa (alkyl NH) pKa2 (NH) logS (pH1,7) logS (pH4,6) logS (pH6,5) logS (pH7,4) logS (pH8)

Vx logPS Vd polarizability MW AMW Sv Se Sp Ss Mv Me Mp Ms

SCBO ARR RBN RBF ZM1 ZM1V ZM2 ZM2V Qindex SNar HNar GNar Xt Dz Ram Pol LPRS VDA MSD SMTI SMTIV GMTI

GMTIV Xu SPI W WA Har Har2 QW TI2 HyDp RHyDp w ww Rww D/D Wap WhetZ Whetm

Whetv Whete Whetp J JhetZ Jhetm Jhetv Jhete Jhetp MAXDN MAXDP DELS TIE S0K S1K S2K S3K PHI BLI PW2 PW3

PW4 PW5 PJI2 CSI ECC AECC DECC MDDD UNIP CENT VAR BAC Lop ICR D/Dr06

GGI1 GGI2 GGI3 GGI4 GGI5 GGI6 GGI7 GGI8 GGI9 GGI10

JGI1 JGI2 JGI3 JGI4 JGI5 JGI6 JGI7 JGI8 JGI9 JGI10 JGT

W3D J3D H3D AGDD DDI ADDD

G1 G2 RGyr SPAM SPH ASP FDI PJI3 L/Bw SEig HOMA RCI AROM HOMT

DISPm QXXm QYYm QZZm DISPv

QXXv QYYv QZZv DISPe QXXe QYYe QZZe DISPp QXXp QYYp QZZp

L1u L2u L3u P1u P2u G1u G2u G3u E1u E2u E3u L1m L2m L3m P1m P2m G1m G2m G3m E1m

E2m E3m L1v L2v L3v P1v P2v G1v G2v G3v E1v E2v

E3v L1e L2e L3e P1e P2e G1e G2e G3e E1e E2e E3e L1p L2p L3p P1p P2p

G1p G2p G3p E1p E2p E3p

L1s L2s L3s P1s P2s

G1s G2s G3s E1s E2s E3s

Tu Tm Tv Te Tp Ts

Au Am Av Ae Ap As

Gu Gm Gs

Ku Km Kv Ke Kp Ks

Du Dm Dv De Dp

Ds Vu Vm Vv Ve Vp Vs nHAcc Ui Hy AMR TPSA(NO) TPSA(Tot)

BLTF96 BLTD48 BLTA96

http://www.vcclab.org/lab/indexhlp/dragon_descr.html

Molecular parameters

M. J. Hynes, J. Chem. Soc. Dalton Trans., 1993, 311; Hammett constants for substituents in para-position taken from: A. Hansch et al., Chem. Rev., 1991,91, 165.

Graphical representation of the correlation between anion binding (logKa) and the Hammett constant σp for compounds 1-22 (excluding 1, 6, 14 and 20). Linear fits are represented by a blue line. (a). interaction with Cl- vs. Hammett constant; (b) interaction with H2PO4 vs. Hammett constant-; (c) interaction with HCO3- vs. Hammett constant.

logKa(Cl-) = 0.55(±0.03)*σp + 1.17(±0.01)

N = 18, R² = 0.96, R²adj = 0.96, RMSE = 0.04, F = 424

logKa(H2PO4-) = 0.85(±0.06)*σp + 2.38(±0.02)

N = 17, R² = 0.92, R²adj = 0.91, RMSE = 0.09, F = 167

logKa(HCO3-) = 0.88(±0.10)*σp + 2.40(±0.04)

N = 16, R² = 0.84, R²adj = 0.83, RMSE = 0.13, F = 73

QSAR analysis of anion binding

LipophilicityThe logP of a compound can be correlated to the retention time in a reverse phase HPLC experiment (elution with water:methanol gradient). The retention times were correlated with calculated logP values to identify the best model.

C. Giaginis et al., J. Liq. Chromatogr. Relat. Technol., 2008, 31, 79; (b) R. Quesada and co-workers, Chem. Commun., 2012, 48, 5274. www.vcclabs.org; I. Moriguchi et al., Chem. Pharm. Bull., 1992, 40, 127.

ClogP calculated using Daylight v4.73 ClogP calculated using VCC labs website

R2 = 0.955 R2 = 0.719

M. J. Hynes, J. Chem. Soc. Dalton Trans., 1993, 311; Hammett constants for substituents in para position taken from: A. Hansch et al., Chem. Rev., 1991,91, 165.

Graphical representation of the correlation between anion binding (logKa) and the Hammett constant σp for compounds 1-22 (excluding 1, 6, 14 and 20). Linear fits are represented by a blue line. (a). interaction with Cl- vs. Hammett constant; (b) interaction with H2PO4 vs. Hammett constant-; (c) interaction with HCO3- vs. Hammett constant.

logKa(Cl-) = 0.55(±0.03)*σp + 1.17(±0.01)

N = 18, R² = 0.96, R²adj = 0.96, RMSE = 0.04, F = 424

logKa(H2PO4-) = 0.85(±0.06)*σp + 2.38(±0.02)

N = 17, R² = 0.92, R²adj = 0.91, RMSE = 0.09, F = 167

logKa(HCO3-) = 0.88(±0.10)*σp + 2.40(±0.04)

N = 16, R² = 0.84, R²adj = 0.83, RMSE = 0.13, F = 73

QSAR analysis of anion binding

Mr (g/mol) PSA solvent acc surface area

volume Surface Area Molecular Volume Vsmax Vsmin PI Molar refractivity Molecular volume

Parachor Index of refraction TPSA Bp

logD (pH1,7) logD (pH4,6) logD (pH6,5) logD (pH7,4) logD (pH8)

pKa (arom NH) pKa (alkyl NH) pKa2 (NH) logS (pH1,7) logS (pH4,6) logS (pH6,5) logS (pH7,4) logS (pH8)

Vx logPS Vd polarizability MW AMW Sv Se Sp Ss Mv Me Mp Ms

SCBO ARR RBN RBF ZM1 ZM1V ZM2 ZM2V Qindex SNar HNar GNar Xt Dz Ram Pol LPRS VDA MSD SMTI SMTIV GMTI

GMTIV Xu SPI W WA Har Har2 QW TI2 HyDp RHyDp w ww Rww D/D Wap WhetZ Whetm

Whetv Whete Whetp J JhetZ Jhetm Jhetv Jhete Jhetp MAXDN MAXDP DELS TIE S0K S1K S2K S3K PHI BLI PW2 PW3

PW4 PW5 PJI2 CSI ECC AECC DECC MDDD UNIP CENT VAR BAC Lop ICR D/Dr06

GGI1 GGI2 GGI3 GGI4 GGI5 GGI6 GGI7 GGI8 GGI9 GGI10

JGI1 JGI2 JGI3 JGI4 JGI5 JGI6 JGI7 JGI8 JGI9 JGI10 JGT

W3D J3D H3D AGDD DDI ADDD

G1 G2 RGyr SPAM SPH ASP FDI PJI3 L/Bw SEig HOMA RCI AROM HOMT

DISPm QXXm QYYm QZZm DISPv

QXXv QYYv QZZv DISPe QXXe QYYe QZZe DISPp QXXp QYYp QZZp

L1u L2u L3u P1u P2u G1u G2u G3u E1u E2u E3u L1m L2m L3m P1m P2m G1m G2m G3m E1m

E2m E3m L1v L2v L3v P1v P2v G1v G2v G3v E1v E2v

E3v L1e L2e L3e P1e P2e G1e G2e G3e E1e E2e E3e L1p L2p L3p P1p P2p

G1p G2p G3p E1p E2p E3p

L1s L2s L3s P1s P2s

G1s G2s G3s E1s E2s E3s

Tu Tm Tv Te Tp Ts

Au Am Av Ae Ap As

Gu Gm Gs

Ku Km Kv Ke Kp Ks

Du Dm Dv De Dp

Ds Vu Vm Vv Ve Vp Vs nHAcc Ui Hy AMR TPSA(NO) TPSA(Tot)

BLTF96 BLTD48 BLTA96

http://www.vcclab.org/lab/indexhlp/dragon_descr.html

Molecular parameters

!log(1/EC50) = 0.82(±0.08)*π + 0.66(±0.18)*σp – 0.26(±0.07)*ΔSPAN – 0.43

!!!!!

N = 18, R² = 0.89, R²adj = 0.87, RMSE = 0.21, F = 42

R

NH

NH

S

1 R = Br2 R = CF33 R = Cl4 R = CN5 R = COCF36 R = COMe7 R = COOMe8 R = F9 R = H10 R = I11 R = NO2

12 R = O(CO)Me13 R = OCF314 R = OEt15 R = OMe16 R = SMe17 R = SO2Me18 R = CH319 R = CH2CH320 R = (CH2)2CH321 R = (CH2)3CH322 R = (CH2)4CH3

Final model: relative parameters

Cl-NO3

-

Calculated using JMP® 9.0.0

N. Busschaert, S.J. Bradberry, M. Wenzel, C.J.E. Haynes, J.R. Hiscock, I.L. Kirby, L.E. Karagiannidis, S.J. Moore, N.J. Wells, J. Herniman, G.J. Langley, P.N. Horton, M.E. Light, V. Félix, J.G. Frey, P. A. Gale,

Chemical Science 2013, 4, 3036-3045 .

Linear combinations of parameters

Calculated using JMP® 9.0.0

Graphical depiction of the values of the coefficients in the equation when the descriptor values are scaled to have a mean of zero and a range of two using JMP 9.0.0. This shows that lipophilicity (RT or logP) has the greatest effect on anion transport.

!

log(1/EC50) = 0.82(±0.08)*π + 0.66(±0.18)*σp – 0.26(±0.07)*ΔSPAN – 0.43

!!N = 18, R² = 0.89, R²adj = 0.87, RMSE = 0.21, F = 42

N. Busschaert, S.J. Bradberry, M. Wenzel, C.J.E. Haynes, J.R. Hiscock, I.L. Kirby, L.E. Karagiannidis, S.J. Moore, N.J. Wells, J. Herniman, G.J. Langley, P.N. Horton, M.E. Light, V. Félix, J.G. Frey, P. A. Gale,

Chemical Science 2013, 4, 3036-3045 .

Predictions

N. Busschaert, S.J. Bradberry, M. Wenzel, C.J.E. Haynes, J.R. Hiscock, I.L. Kirby, L.E. Karagiannidis, S.J. Moore, N.J. Wells, J. Herniman, G.J. Langley, P.N. Horton, M.E. Light, V. Félix, J.G. Frey, P. A. Gale,

Chemical Science 2013, 4, 3036-3045 .

Conclusions

“From anion receptors to transporters”

P.A. Gale, Acc. Chem. Res. 2011, 44, 216-226.

!!Hansch analysis of anion transport results may reveal the molecular parameters that we should optimise in order to design an efficient anion transporter. For the thioureas studied these are logP, affinity and molecular size. !!!!!

“Small molecule lipid bilayer anion transporters for biological applications” N. Busschaert and

P.A. Gale,

Angew. Chem. Int. Ed., 2013, 52, 1374-1382.

“Anion transporters and biological systems”

P.A. Gale, R. Pérez-Tomás and R. Quesada, Acc. Chem. Res. 2013, DOI: 10.1021/

ar400019p.

2041-6520(2013)4:8;1-N

ISSN 2041-6520

Chemical Sciencewww.rsc.org/chemicalscience Volume 4 | Number 8 | August 2013 | Pages 2979–3348

EDGE ARTICLEPhilip A. Gale et al.Towards predictable transmembrane transport: QSAR analysis of anion binding and transport

!!!!

Research support and collaborations in Southampton: !

Dr. Mark E. Light Dr John Langley

Dr Neil Wells Julie Herniman

Prof. Jonathan Essex Prof. Jeremy Frey

!

Current group: !

Dr Jenny Hiscock Dr Wim van Rossom

Dr Nathalie Busschaert !

Isabelle Kirby Louise Karagiannides

Francesca Piana Stuart Berry

Xin Wu Michael Spooner

Former group members: !

Dr Christine C. Tong Dr Korakot Navakhun

Dr Joachim Garric Dr Claudia Caltagirone

Dr Gareth W. Bates Dr Marco Wenzel Dr Stephen Moore

Sam Bradberry Dr Cally Haynes

Prof. Jeff Davis Dr William Harrell Jr.

!Prof Janez Plavec Dr Damjan Makuc Prof Kate Jolliffe

!Prof Tony Davis

Dr Hennie Valkenier !

Dr Roberto Quesada !

Prof. Ricardo Pérez-Tomás !

Prof. Vítor Félix !!!

CM1005 Supramolecular Chemistry in Water