Relationships between Lipophilicity and Solubility
Karl J. Box Physical Chemistry Symposium Nov 29th 2006 Sirius Analytical Instruments Limited
Solubility versus calculated lipophilicity LogS vs CLogP -1 -2
LogS
-3 -4 -5 -6 -7 -1
0
1
2
3
4
5
CLogP Presented by James Blake, Array BioPharma – “Finding Drugs within Chemistry Space” 2 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
Measured Solubility versus Measured Lipophilicity Log(Intrinsic Solubility) vs LogP 2 0
Log S
-2 -4 -6 y = -1.0342x - 0.5338 2 R = 0.8976
-8 -10 -2
0
40 out of 78 compounds within ± 1 log unit of best fit line = 51% 3 / 34
2
4
6
8
10
Log P Copyright Sirius Analytical Instruments Ltd. 2006
Measured Solubility versus Measured Lipophilicity Log(Intrinsic Solubility) vs LogP 2 0
LogS = -1.03LogP - 0.53
Log S
-2 -4 -6 y = -1.0342x - 0.5338
-8
2
R = 0.8976
-10 -2
0
2
59 out of 78 compounds within ± 1.5 log units of best fit line = 76% 4 / 34
4
6
8
10
Log P Copyright Sirius Analytical Instruments Ltd. 2006
High and Low Solubility Regions Log(Intrinsic Solubility ) vs LogP 2 0
LogS
-2 -4 -6 y = -1.0342x - 0.5338
-8
R
2
= 0.8976
-10 -2
0
2
4
6
8
10
Log P
High solubility region for a given lipophilicity What often happens ! Poor solubility region for a given lipophilicity
5 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
Compounds in High Solubility Region Log(Intrinsic Solubility) vs LogP OH
2
4-lodophenol
0
H3C
I
OH
Thymol CH3
Log S
-2 CH3
-4 -6 y = -1.0342x - 0.5338
-8
2
R = 0.8976
-10 -2
0
2
4
6
8
10
Log P
Poor solubility region for a given lipophilicity
6 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
Low Solubility Region Log(Intrinsic Solubility) vs LogP glipizide
2
O
NH
O
O
HN
N
0
NH
N
H3C
-2
Log S
O
S
loperamide Cl
-4
N
O CH3
N
OH
H3C
-6 tetracycline and floxacins
-8
folic acid
-10 -2
0
piroxicam, flumequine, nitrofurantoin and many sulphonamides loperamide and terfenadine
2
4
6
8
10
Log P H3C CH3
HO
tetracycline
N
O
CH3 N
HN
H OH
H2N
N
N
NH2
ciprofloxacin
OH OH
7 / 34
O
OH
O
O
N N H
O N
NH NH
O F
folic acid
O
O
OH
O HO
HO
Copyright Sirius Analytical Instruments Ltd. 2006
Low Solubility Compounds with High Melting Points
2 0
Log S
-2 -4 -6 -8 -10 -2
0
2
4
6
8
10
Log P Compounds with MPts > 200oC
8 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
All Compounds with High Melting Points
2 0
Log S
-2 -4 -6 -8 -10 -2
0
2
4
6
8
10
Log P
All Compounds with MPts > 200oC 9 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
Relationship with Melting Points
2 0
Log S
-2 -4 -6 -8 -10 -2
0
2
4
6
8
10
Log P Compounds with MPts > 145oC
10 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
Compounds in High Solubility Region with low Melting Points
OH
2
4-lodophenol MPt = 93oC
0
H3C
I
OH
Thymol MPt = 51oC CH3
Log S
-2
CH3
-4 -6 -8 -10 -2
0
2
4
6
8
10
Log P
11 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
All Compounds with low Melting Points
2 0
Log S
-2 -4 -6 -8 -10 -2
0
2
4
6
8
10
Log P
Compounds with MPts < 100oC 12 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
Relationship with Melting Points Red circles have melting points > 145oC
2
Light Blue circles have melting points < 135oC
0
Log S
-2 -4 -6 -8 -10 -2
0
2
4
6
8
10
Log P Compounds with MPts < 135oC But what is left? Do compounds shown as 13 / 34
have MPts between 135 – 145oC? Copyright Sirius Analytical Instruments Ltd. 2006
Biopharmaceutics Classification System (BCS) *Drug considered HIGHLY SOLUBLE if the highest dose remains in aqueous solution between pH 1 - 8
HIGH SOLUBILITY CLASS 1 a
CLASS 2 b
HIGH PERMEABILITY *Drug considered HIGHLY PERMEABLE if human absorption >90%
CLASS 3 c
LOW PERMEABILITY
a b
* www.fda.gov/cder/guidance/3618fnl.pdf
LOW SOLUBILITY
c
1 3
CLASS 4
2 4
RATE OF DISSOLUTION limits in vivo absorption SOLUBILITY limits absorption flux PERMEABILITY is rate determining
Amidon, G L. Lennernas, H. Shah, V P. Crison, J R. Pharm. Res. 1995, (12(3)) pp 413-420
14 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
Biopharmaceutics Classification System (BCS) LOW PERMEABILITY CLASS 3
HIGH PERMEABILITY CLASS 1
HIGH SOLUBILITY
CLASS 4
CLASS 2
LOW SOLUBILITY
15 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
LogP vs LogS and the BCS
2 0
Log S
-2 -4 -6 -8 -10 -2
0
2
4
6
8
10
Log P Class 1 – High Solubility, High fraction absorbed Class 2 – Low Solubility, High fraction absorbed Class 3 – High Solubility, Low fraction absorbed Class 4 – Low Solubility, Low fraction absorbed 16 / 34
Not known (includes topical anesthetics)
Copyright Sirius Analytical Instruments Ltd. 2006
LogP vs LogS and the BCS
2 0
3
1
0
2
3 1 4 2
Log S
-2 -4 -6 -8 -10 -2
2
4
6
8
10
Log P Class 1 – High Solubility, High fraction absorbed Class 2 – Low Solubility, High fraction absorbed Class 3 – High Solubility, Low fraction absorbed Class 4 – Low Solubility, Low fraction absorbed 17 / 34
4? Copyright Sirius Analytical Instruments Ltd. 2006
Actively transported and paracellularly transported compounds
2 0
Log S
-2 -4 -6 Active transport or paracellular transport
-8
Sulfasalazine Fa < 20%
-10 -2
0
2
4
6
8
10
Log P Class 1 – High Solubility, High fraction absorbed Class 2 – Low Solubility, High fraction absorbed Class 3 – High Solubility, Low fraction absorbed Class 4 – Low Solubility, Low fraction absorbed 18 / 34
Marketed drugs. All compounds >20% fraction absorbed except one. Copyright Sirius Analytical Instruments Ltd. 2006
Problems with reported literature values
2 0
Log S
-2 -4 -6 Loperamide measured at Sirius
-8 -10 -2
0
2
4
6
8
10
Log P Class 1 – High Solubility, High fraction absorbed Class 2 – Low Solubility, High fraction absorbed Class 3 – High Solubility, Low fraction absorbed Class 4 – Low Solubility, Low fraction absorbed 19 / 34
Not known (includes topical anesthetics)
Copyright Sirius Analytical Instruments Ltd. 2006
Problems with reported literature values
2 Loperamide becomes Class 3 – three orders of magnitude difference between measured and literature solubility values
0
Log S
-2 -4 -6 -8 -10 -2
0
2
4
6
8
10
Log P Class 1 – High Solubility, High fraction absorbed Class 2 – Low Solubility, High fraction absorbed Class 3 – High Solubility, Low fraction absorbed Class 4 – Low Solubility, Low fraction absorbed 20 / 34
Not known (includes topical anesthetics)
Copyright Sirius Analytical Instruments Ltd. 2006
Now let’s return to the compounds with missing melting points Same slide as slide 13. Red circles have melting points > 145oC
2
Light Blue circles have melting points < 135oC
0
Log S
-2 -4 -6 -8 -10 -2
0
2
4
6
8
10
Log P But what is left? Do compounds shown as
21 / 34
have MPts between 135 – 145oC?
Copyright Sirius Analytical Instruments Ltd. 2006
The missing compounds can not form supersaturated solutions All other compounds not marked in green have the ability to form supersaturated solutions. We call these compounds chasers.
2 0
Chasers have kinetic solubilities greater than their intrinsic solubility.
Log S
-2 -4 -6 -8 -10 -2
0
2
4
6
8
10
Log P These compounds cannot form supersaturated solutions. When the pH is right, they fall out of solution immediately the solubility limit is exceeded. We call these compounds Non-Chasers. The kinetic solubility and Intrinsic solubility of non-chasers is equal. 22 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
Can we predict whether a sample is a non-chaser? Chlorpromazine
Imipramine
S
Amitryptyline N
Non-chaser
N
Cl
Non-chaser CH3 N
Desipramine
CH 3
Chlorprothixene
Non-chaser
CH3
Non-chaser
CH 3
Cl
Non-chaser converts to chaser
N
Nortriptyline NH
S
CH 3
N
Trimipramine
CH 3
N
N
CH3
CH3
CH3
Maprotiline
N
Non-chaser
Non-chaser
CH3
N CH3
CH3
NH
CH3
Chaser NH
CH 3
Secondary and tertiary amines with logP > 4.
23 / 34 Similar structures, but maprotiline contains a -CH2-CH2- bridge.
Copyright Sirius Analytical Instruments Ltd. 2006
Other non-chasers….. Cl
Chlorpheniramine
Meclizine
Diphenhydramine
N CH3 N
N
CH3
N O
CH3
CH3
N
Cl CH3
H3C
HO
H
HN
N
CH3
N
N
F
Sertraline
H
Cl
24 / 34
F
N
Cl
F S
Cl
N
Flupenthixol
S
Prochlorperazine Copyright Sirius Analytical Instruments Ltd. 2006
More non-chasers.….. O
CH3
H
H2C
Verapamil
H
O
N
H
CH3
CH3
HO H3C
CH3
H3 C O
N
O
O N
CH H3C
H3C
CH3
N
Deprenyl
Quinine
N CH3
O
CH3
H
O
S
CH3
H
O
O
N O
O
H3C
I
CH3
I H3C
O N
Amiodarone CH3
25 / 34
N CH3
Diltiazem Copyright Sirius Analytical Instruments Ltd. 2006
..….. and some chasers H3C
CH3
CH3
HO H3C
H3C
Terfenadine
NH H3C
N
CH3 O
O
N
CH3
Cl
Cl
OH
HO
Cl
N
H3C
HO
N
HO
HN
Cl
Amodiaquin
NH2
H3C
N
OH
Metoclopramide
N
CH3
26 / 34
NH2
Loperamide
N
O
N
NH2
Pyrimethamine
OH O
Nadolol
NH
CH3 CH3
CH3
Copyright Sirius Analytical Instruments Ltd. 2006
References – chasers and non-chasers First paper, introducing the concept of chasing equilibrium – Stuart, M. Box, K. Chasing equilibrium: measuring the intrinsic solubility of weak acids and bases. Anal. Chem. 2005, 77(4), 983-990
Second paper, collaborative research to validate method and introduce concept of nonchasers – Box, K J. Völgyi, G. Baka, E. Stuart, M. Takács-Novák, K. Comer, J E A. Equilibrium vs. kinetic measurements of aqueous solubility, and the ability of compounds to supersaturate in solution - a validation study. J. Pharm. Sci. 2006, 95, 1298-1307.
27 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
Non-chasers While the similarities between some non-chasing structures are obvious, we don’t yet have strict rules for predicting nonchasing from structure. Since introducing CheqSol in March 2004, we have found only a few non-chasing acids, but about 20% of bases have been non-chasers. Supersaturation impacts on drug bioavailability and must be considered during formulation and manufacturing. Some attempts at predicting non-chasing compounds are shown in the following slides. 28 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
Possible implications of chasing vs. non-chasing The propensity of a compound to supersaturate and remain so for a reasonable time might have implications in drug adsorption For example, a weak base might dissolve fully in the stomach but precipitate on entering the high pH environment of the upper intestinal tract. A better understanding of this would enable better adsorption models to be constructed. Do non-chasers fall out of solution as amorphous material whereas chasers produce crystalline precipitate? Amorphous materials are amenable to solid state dispersion nanoparticle delivery methods. Conversely, do the non-chasers have some kind of structured or ordered solution phase (liquid crystals, micelles, aggregates) that prevents supersaturation? Is it possible to formulate supersaturated solutions that stay in solution long enough such that absorption is enhanced? Conversely, if a compound is administered in supersaturated solution, could it crash out of solution with unpleasant side effects? 29 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
Correlating non-chasing with melting point Melting point – Many non-chasers have low melting points * – According to the Merck Index the free forms of Chlorpheniramine, Verapamil, Imipramine and Chlorpromazine are oils. The free form of nortryptyline is structurally similar to Imipramine and may also be an oil. Do non-chasers precipitate as an oily phase which cannot change into crystals? – Quinine forms crystals with a M.pt of 177ºC, but it forms a trihydrate with a very low M.pt of 57ºC. Does it come out of solution as a fluid droplet which does not further crystallise?
* With thanks to Rod Kittlety, AstraZeneca, Alderley Park, Macclesfield, UK 30 / 34
Copyright Sirius Analytical Instruments Ltd. 2006
Correlating non-chasing with number of H-bond donors
31 / 34
Compound
No. of H-Bond Donors
No. of H-Bond Acceptors
Total Polar Surface Area
No. of rotatable bonds
Amiodarone
0
4
42.7
11
Amitriptyline
0
1
3.2
3
Chlorpheniramine
0
2
16.1
5
Chlorpromazine
0
2
31.8
4
Deprenyl
0
1
3.2
5
Desipramine
1
2
15.3
4
Diltiazem
0
6
84.4
7
Diphenhydramine
0
2
12.5
6
Flupenthixol
1
3
52.0
6
Imipramine
0
2
6.5
4
Meclizine
0
2
6.5
5
Nortriptyline
1
1
12.0
3
Prochlorperazine
0
3
35.0
4
Quinine
1
4
45.6
4
Sertraline
1
1
12.0
2
Trimipramine
0
2
6.5
4
Verapamil
0
6
64.0
14
All weak bases with zero or one H-Bond Donor only
Copyright Sirius Analytical Instruments Ltd. 2006
Correlating non-chasing with SELMA descriptors * Simple rule-based system
Blue = chaser Red = non-chaser
-15
-20
PLS-analysis of Astra (SELMA) set of chemical descriptors shows that polarisation properties of the molecules e.g. that they are amphiphiles (with surfactant properties) causes non-chasing.
-25
-30
-35
-40
Rules:
-45
1.
-50
0.06
0.08
Aver. pos. charge
0.1
0.12
If (Aver. pos. charge