PhysChem Measurements-more than just a number

Emily Freeman Physical and Analytical Sciences UCB

PhysChem Forum June 2007

Overview

PhysChem Report

Overview of the Assays run at UCB

More to learn from AKAS and HTLogD

Summary

Questions?

PhysChem Report

ND Poor Modest GoodAKAS ND <30mM 30-150mM 150-350mM >350mMPSA >113 83-113 <83

AKAS Automated Kinetic Aqueous Solubility

Final DMSO content 5%

HT MultiPROBE II

Aqueous Buffer

44 x compounds in 4hrs Run twice a week

10mM DMSO Stock Solution

Fastest form to dissolve not necessarily the most stable

Kinetic

All NCE’s

AKAS at 4 pH’s

Stomach pH 1 - 3 Duodenum pH 5 – 7Blood pH7.4Colon pH 7 – 8

Run at 4 pH’s to mimic the varying pH range of

the GI tract:

pH3, pH5, pH7.4, pH9.

Picture modified from: Alex Avdeef. Absorption and Drug Development Solubility, Permeability and Charge State. pION, Inc. John Wiley & Sons 2003

Solubility Assays

2nd Tier Solubility

Kinetic - from DMSO Stock

1st Tier Solubility

Pseoudothermodynamic - from solid

QuickSolAKAS Passes at>150uM

Solubility Assays - QuickSol

Analysis by HPLC

Solid Sample

Buffer

shaken for 90 mins

Pseudo-thermodynamic

Starts from solid but not in excess.

Only a single time point not necessarily at the equilibrium.

pH7.4 (pH5 on request)

Dissolution may not be complete in 90 mins.

Why is QSol Important?

Crystal forms can have different solubility

Batches can have different crystal forms

Batch to Batch Variation

Depending on the dissolution energy of the crystal lattice

Do not Average Batches

No DMSO present so no co-solvent effects - AKAS has 5% DMSO

Use the Right Assay for your Question

Assays start from DMSO Stock

Using Solid to dose animals

A Biologist DMPK

= QSol= AKAS

Lipophilicity

aqaq

octoct

[neutral][ionised][neutral][ionised]D(o/w)

++

=Octanol

WaterPartition

Layers are separated and analysed by HPLC

Distribution Around The Body BloodAqueous

TissuesLipophilic

Sample added to octanol and water and shaken for 90 minutes

HTLogD7.4 assay

Why we measure LogD7.4 not LogP

Ratio

D = Distribution coefficient

Octanol Neutral and Ionised species

Water Neutral and Ionised species

LogD7.4 can be run at the same pH for all compounds

It’s run at a pH of interest eg pH7.4 for blood

LogD7.4

P= Partition coefficient

Octanol Neutral species

Water Neutral species

LogP can only be run at a pH where all of the compound is neutral

Would have to be run at a different pH for each compound

depending on it’s pKa

LogP

Shake flask LogD7.4 vs HTLogD7.4

-2

-1

0

1

2

3

4

5

-2.5 -1.5 -0.5 0.5 1.5 2.5 3.5 4.5

Shake Flask LogD7.4

HTL

ogD 7

.4

R2 = 0.94

There’s more to learn from AKAS than just solubility

Stability

pKa Approximation

If a compound is ionisable

Spectral ChangeFamotidine calibration spectra

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

240 260 280 300 320 340 360 380 400wavelength / nm

0.5mM

0.2mM

0.05mM

0.01mM

pH3

pH9

Famotidine solubility samples spectra

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

240 260 280 300 320 340 360 380 400wavelength / nm

pH3

pH5

pH7.4

pH9

S

NNNH2

NH2

SNH2

N SO

O

NH2

Famotidine

Calibration Spectra

Solubility Spectra

Ionisable compoundsAbs

AbspH3pH5

pH7.4pH9

Four point calibration

Spectral change in the solubility

Acidic or Basic Ionisation?

N

NH+

O

R OH

O

R O

Neutral at pH9

Ionised at pH3

Neutral at pH3

Ionised at pH9

Basic Shift Acidic Shift

Spectral ChangeFamotidine calibration spectra

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0.5

240 260 280 300 320 340 360 380 400wavelength / nm

0.5mM

0.2mM

0.05mM

0.01mM

pH3

pH9

Famotidine solubility samples spectra

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

240 260 280 300 320 340 360 380 400wavelength / nm

pH3

pH5

pH7.4

pH9

S

NNNH2

NH2

SNH2

N SO

O

NH2

Famotidine

Isosbestic point

Neutral and charged curves cross: absorbance is the same

Calibration Spectra

Solubility Spectra

Ionisable compoundsAbs

AbspH3pH5

pH7.4pH9

Basic ionisation event between pH5 and pH7.4

pH3 pH5 pH7.4 pH9244 162 83 75

AKAS results

pKa approximation from AKAS

pKa Estimation

Soluble at Lower pHs

Basic ionisation event between pH5 and pH7.4

pH3 pH5 pH7.4 pH9244 162 83 75

% Ionised

Ionisable Compounds

pH3 pH5 pH7.4 pH982 89 163 184

pKa Estimation

Soluble at Higher pHs Soluble at all pH’s ionised or not

pH3 pH5 pH7.4 pH9>350 >350 >350 >350

Ibuprofen

Acidic ionisation event between pH5 and pH7.4

Examples of AKAS approximating pKa

pH3 (uM) pH5 (uM) pH7.4 (uM) pH9 (uM)2 >350 >350 >350

AKASIsoxicam

pH3 (uM) pH5 (uM) pH7.4 (uM) pH9 (uM)144 >350 >350 >350

AKASSulfasalazine

pH3 (uM) pH5 (uM) pH7.4 (uM) pH9 (uM)>350 >350 240 0

AKASTerfenadine

NS

O

NH

OH

OO

ON

N NH

SO O

NN

OH

O

OH

NOH

OH

pKa

pKa

pKa

pH Stability

pH3 pH5 pH7.4 pH9>350 >350 >350 ND

O

O

Compound X

No spectral change in the calibration

Spectral change in the solubility

Unstable ester at pH9?

90min shakePossible

stability issues

Calibration Spectra

Sample Spectra

pH3pH5

pH7.4pH9

pH Stability issues – from AKAS UV spectra

Compound Y

pH3pH5

pH7.4pH9

Calibration Spectra

Sample SpectrapH3 pH5 pH7.4 pH9ND >350 >350 >350

Unstable at pH3?

There’s more to learn from AKAS than just solubility

Stability

pKa Approximation

If a compound is ionisable

So in 6hrs we have discovered:

Why we run a purity check before HTLogD7.4

Accept over 90% purity

Indicates which HPLC pH method to use

Using QC as a time zero we can identify if the compound has

degraded over time

QC Purity Check by HPLC

HTLogD7.4 Assay

QC and HTLogD7.4

Octanol 254nm

Octanol 280nm

Aqueous 254nm

Aqueous 280nm

QC 254nm

PhysChem Report

ND Poor Modest GoodAKAS ND <30mM 30-150mM 150-350mM >350mMPSA >113 83-113 <83

PhysChem Report

ND Poor Modest GoodAKAS ND <30mM 30-150mM 150-350mM >350mMPSA >113 83-113 <83

Calculated values

ACD Labs predicted pKa

Polar Surface Area Absorption

Calculate % ionised

Number of H bond Donors and Acceptors Physical Properties

Confirms pKa in AKAS and suggests other pKa’s that may not be visible in AKAS

+ or – Sulphur?

Benefit of Our Report Over Databases

Averaged and concatenated data

Comments not looked at – stability issues highlighted

Difficult to see PhysChem data as a package

ACD labs pKa

Which AKAS results do I use?

Use concatenated as batches may differ

Average?

or Concatenated?

PhysChem Report

ND Poor Modest GoodAKAS ND <30mM 30-150mM 150-350mM >350mMPSA >113 83-113 <83

Summary

The PhysChem Report

Measured Values, AKAS, QSol, HTLogD.

Can approximate pKa from AKAS

Possible to highlight stability issues from AKAS and HTLogD

Calculated Values, pKa, PSA, H-bond Donors/Acceptors

Limitations with Spotfire

Acknowledgements

•PhysChem, Cambridge, UK• John Cooper• Benedicte Fau• Dave Sherwood

•PhysChem, Slough, UK• Richard Taylor• Barbara Mason• Christine Prosser• Justin Staniforth• Femi Kumuyi

•PhysChem, Braine, Belgium• Luc Quere• Liliane Ellens• Benoit Matthew• Geraldine Longfils

•Informatics, Slough• Mike Sanderson• Sarah Archibald• Dave Lee• Phil Ashworth• Steve Guise• Matt Page

Questions

Questions?

References

Li Di and E H Kerns. Profiling drug-like properties in discovery research. Current opinion in Chemical Biology 2003, 7:402-408

Li Di and E H Kerns. Physicochemical profiling: overview of the screens.DDT. Vol.1, No.4 2004

Li Di and E H Kerns. Pharmaceutical profiling in drug discovery. Therapeutic focus. DDT Vol.8 No.7 April 2003

T Hartman, J Schmitt. Lipophilicity-beyond octanol/water: a short comparison of modern technologies. DDT Vol.1 No.4 2004

Alex Avdeef. Absorption and Drug Development Solubility, Permeability and Charge State. pION, Inc. John Wiley & Sons 2003

AKAS solubility of naphthalene/quinoline series(vs time)

0

10

20

30

40

50

60

70

Pre Q2 2002 Q2 2002 Q3 2002 Q4 2002 Q1 2003

Date synthesised

% T

otal

AKAS solubility of P38 compounds(vs time)

0

10

20

30

40

50

60

Pre Q2 2002 Q2 2002 Q3 2002 Q4 2002 Q1 2003

Date synthesised

% T

otal

pKa Estimation With AKASTamoxifen

Basic ionisation event between pH5 and pH7.4

O

O

O

OH

OH

Fluorescein Acidic ionisation event between

pH3 and pH5

pH3 pH5 pH7.4 pH9>350 >350 16 1

pH3 pH5 pH7.4 pH9113 >350 >350 >350

Ionised NeutralpKa

IonisedNeutralpKa

Base Acid

O

N

Soluble at low pH as ionised. Soluble at high pH as ionised

Ionised Neutral Neutral Ionised

Where Discovery PhysChem Fits in UCB

To avoid late stage PhysChem problems screen early in discovery

Discovery PhysChem

Early stage Solubility and Permeability data

Later stage research supported by colleagues in Cambridge and Braine.

PhysChem Profile

Lipophilicity HTLogD7.4

pKaACD LABS prediction

PurityQC

HPLC

StabilityQC

HPLC

PermeabilityPAMPA

PhysChem Profile

SolubilityAKASQSol

PAMPA in development

Currently in development

Donor

AcceptorpH7.4 Buffer

Sample in pH7.4 Buffer

Membrane Phosphatidyl Choline

Single Mechanism

Passive Diffusion

High Through Put

Papp is a rate cm/s-7

PAMPA is like a Kinetic Lipophilicity

The rate in which a compound crosses a membrane

Low values cross Slowly High values Fast

GI PAMPA(all NCEs)

BBB PAMPA(all NCEs)

Caco-2(request)

Permeability

SnapSol(request)

BioSol(all NCEs)

AKAS(all NCEs)

QSol(request)

TSol(request)

Solubility

HT Log D7.4(all NCEs)

BIO-PAMPA(all NCEs)

Lipophilicity

10 mM DMSO Stock

Log P(request)

pKa(request)

Discovery PhysChem (Slh)

PhysChem (Braine + Cam)

DMPK (Slh)QC(all NCEs)

Log D(request)

Cascade

Solubility Assays – SnapSol – BioSol

•SnapSol – kinetic – 1pH-Can run 190 compounds in 2hrs-Data processing is very fast-Compound detected by UV plate reader

•BioSol – Kinetic – at the conditions used in the biological primary assays.

HT MultiPROBE II

SnapSol(request)

BioSol(all NCEs)