Professional Documents
Culture Documents
Klara Valko
Physicochemical Characterisation
Analytical Chemistry
Molecular Discovery Research
GlaxoSmithKline
Stevenage, UK
Physchem Forum 2010
Klara Valko
Elimination
Toxicity
Metabolism
Clearance
Free
Concentration
Dose
Target
Solubility
Permeability
logK HSA
logK AGP
logK IAM
fibrinogen
non-proteins (fatty acids, urea, creatinine,
ammonium salts, amino acids, hormones, etc)
Physchem Forum 2010
Klara Valko
K plasma =
%PPB
--------------100 - %PPB
Vr = volume ratio (physiological variation !)
K(protein) = partition coefficient
Physchem Forum 2010
Klara Valko
% HSA binding
%
bound
80
% Binding 0.4mM HSA
60
40
20
0
1
1.5
2.5
3.5
4.5
K (HSA)
K
Copyright GlaxoSmithKline Group of
Companies
Day 1
Day 2
Day 3
KKlara Valko
Rovinj 2009
1.6
3.3
4.5 6.8
Copyright GlaxoSmithKline Group of
Companies
75%
95%
98% 99.8%
8.0
99.9999 %
S
O
HN
K=6.3
O
S
HN
K=5.1
Cl
O
S
HN
Cl
K=7.5
Chromatographic Hydrophobicity
Index obtained on IAM
Red: acids
Blue:bases
Green: neutrals
Black: zwitterions
calcKPPB
% PPB
Empty circle:
training set
Red: acids
Blue:bases
Green: neutrals
compounds
Black: zwitterionic
Physchem Forum 2010
Klara Valko
HPLC
1.5
1.5
logK HSA
logK HSA
Scatter Plot
0.5
-0.5
0.5
-0.5
-1
-1
-4
-2
-4
logd_pH74_acd.value
ACD_logD
-2
PCv5.clogp_481
clogP
Copyright GlaxoSmithKline Group of
Companies
2.5
2.5
logK IAM
logK IAM
Scatter Plot
1.5
1.5
0.5
0.5
-4
-2
logd_pH74_acd.value
-4
-2
PCv5.clogp_481
ACD_logD
clogP
Klara Valko
Rovinj 2009
Vdss =
Dose
litre; or litre/kg
Why VD is important?
t = 0.693
Vdss
Cl
K (plasma
proteins/free)
K (HSA) +
K(AGP) +
K(globulins)
HSA
Globulins,
Lipoproteins,
HDL, LDL etc.
AGP
KHSA
c1 free drug
KAGP
aqueous phase
c2 free drug
KLIPOPROTEINS
Lipoprotein membranes
KGLOB
c2 free drug
KAGP
AGP
Vdss= K(plasma/free)/K(tissue/free)
Copyright GlaxoSmithKline Group of
Companies
KPLASMA
c1 free drug
Plasma
KSA
SA
Tissue
KTISSUE
K (IAM)
K (tissue /free)
1.5
-0.5
-1
-1.5
-1
-0.5
0.5
1.5
predlogVd
logVd= 0.0789clogP +
0.588*posChg - 0.731negChg0.147
n=118 r=0.80 s=0.41
The application of the
HPLC based bio mimetic
binding data can improve
the prediction of in vivo
compound distribution in
comparison to purely in
silico calculations.
insilicologVd
3.5
2.5
1.5
0.5
-0.5
0.5
1.5
2.5
3.5
ModelLogVdu
Different
colours are
from different
projects. Red
compounds
showed strong
specific AGP
binding.
3.5
2.5
1.5
0.5
0
-0.5
0.5
1.5
2.5
3.5
estlogkbtb
logkBTB=1.29(+/-0.10)*logk(IAM)+1.03(+/-0.10)*logk(HSA) 2.37
n=135
r2=0.76
s=0.35 F=212
logkblood=0.14*CHIlogD7.4+0.46*logK(HSA)
+0.43*logkAGP 0.72
Klara Valko
Physchem Forum 8 2010
n=133
r2=0.79
s=0.29 F=164
1.5
0.5
-0.5
-1
-0.5
0.5
1.5
0.5
-0.5
maybe
no
yes
CNS
Conclusions
HPLC method offers binding measurements to individual proteins.
The HPLC based method can reproducibly differentiate between very strong
binders and between some enantiomeric compounds.
Structure -binding relationships can be set up for very high binders to help
modifying structure to reduce binding. The structure binding relationships for
HSA and AGP are very different.
The individual protein binding data can be used to simulate physiological
(pathological) variation of total plasma protein binding.
A model has been established and validated that allows estimation of the total
plasma protein binding from HSA and AGP binding.
The protein binding data together with the membrane partition data (K IAM) can
be used to predict the in vivo volume of distribution (free tissue and plasma
concentration) of compounds.
The bio-mimetic protein and phospholipid binding data can be used for estimating
brain tissue binding, and CNS penetration of the compounds too.
Physchem Forum 2010
Klara Valko
Acknowledgements
Alan Hill
Shenaz Nunhuck
Pat McDonough
Iain Reid
Elisabetta Chiarparin
Silvia Bardoni
UP Group