Symp05/5_Hong Wan_R.pdf - PhysChem Forum

High throughput screening of pKa by capillary electrophoresis and mass spectrometry (CE/MS)

and long-term validation

Hong Wan

Lead Generation, DMPK and Physical ChemistryAstraZeneca R&D Mölndal

SE-431 83 Mölndal. SWEDENe-mail: [email protected]

5th Physchem Forum, Stevenage, UK, June 19th

• Why screening pKa• Advantages of CE/MS over established methods• High throughput pKa screening method by CE/MS

• Effects of buffer type and ionic strength on pKa• Comparison of CE/UV and CE/MS• Effects of pressure on migration times/sensitivity/mobility

• Comparisons of measured pKa and ref. and predicted values• Reproducibility and accuracy and long-term validation• Summary

Outline

Screening for pKa, why?

• Info about change in charge of molecule……

pKa(dissociation

constant)

Solubility

Lipophilicity

Protein binding

QSAR descriptor Formulation

Bioavailability

AbsorptionPermeabilityHydrogen

bondingpotency

StabilityMetabolism

Is CE/MS better technique for pKa than others?

• Sirius GLpKa – well established and widely used method !

• pION's Gemini for routine pKa ($500/cdp, $760/cpd/co-solvent)

• CE/UV (CombiSep), commercially available

• CE/MS (new technology), AZ, Mölndal, 2003 published.

H. Wan, A. Holmén, M. Någård, W. Lindberg, J. Chromatogr. A. 979 (2002) 369.

H. Wan, A. Holmén, Y. Wang, W. Lindberg, M. Englund, M. Någård, R. Thompson,

Rapid Commun. Mass Spectrom. 17 (2003) 2639.

Advantages of CE/MS over established methods

Specification CE/MS Titration/UV (D-PAS)

Amount of cpd 1 μg, or 1 μL of 10 mM solution

1-2 mg (titration)3 μL of 10 mM DMSO stock (Sirius 3T)

Concentration 1-10 μM (100 μL)

No

> 20 μM (UV)

Purity required

Co-solvent No Yes, for poorly soluble cpds

Accuracy < ± 0.2 ± 0.02-0.2?

Throughputcapacity

>150 cpds/seq.(6h) 4 min/cpd (Sirius T3)

Limitation Poor ionization (ESI) Impurity? poor solubility (precipitation),Ionization group close to chromophore

Yes

Principle of pKa determination by CE

pKa ↔ 50% ionization

}HA{}A}{H{K th

a

−+

=

I0.3281I0.5085z

mmmlog pHpK

2

effm

efftha α+

+⎥⎦

⎤⎢⎣

⎡−

−=

Equations used for pKa calculation

Graphs and table from Miller, Electrophoresis 2002, 23, 2833

High-throughput pKa screening by CE/MS

10-14 different pH buffers Constant ionic strengths

0.2 0.4 0.6 0.8 1 (min)

mAU

0

5

10

15

20

25

30

35

12

3

N

⎟⎟⎠

⎞⎜⎜⎝

⎛−⋅

⋅=−=

eofobs

efftoteofobseff ttV

LLmmm 11

(fit-pKa)

CE separation

Mobility calculation

pKa evaluation

Effective mobility

0 0.5

1 1.5

2 2.5

3 3.5

4 4.5

2.00 4.00 6.00 8.00 10.00 pH

UV non-volatile buffers (Single compound)

20-100 μM

MS Volatile buffers

(Pooled compounds)

pKa2

pKa1

Pressure

Sample pooling and data analysis flow scheme

Reference stand.

2003 manual data analysis



Current pKa screening - integrated & automated assay

Automated sample pooling (mass)

CE/MS

Automated migration time conversion

pKa evaluation

pKa reported to database

Generate sample run list (Abase)

AZ10181832-001 NA

pH meff res Fitting Option:2.477 287.909 x 287.806016 -0.102984 287.909 2.4773.012 243.266 x -243.266 243.266 3.0124.046 178.427 x -178.427 178.427 4.046 Value: Ionization: Ion. Func. Gr.4.565 163.527 x -163.527 163.527 4.565 pKa1:5.140 163.972 x -163.972 163.972 5.14 pKa2:5.633 163.325 x -163.325 163.325 5.633 pKa3:6.060 164.855 x -164.855 164.855 6.06 r2:7.086 151.090 x -151.09 151.09 7.0868.020 115.490 x -115.49 115.49 8.028.380 72.070 x -72.07 72.07 8.389.080 26.064 x -26.064 26.064 9.089.470 9.778 x -9.778 9.778 9.47

10.010 5.132 x -5.132 5.132 10.0110.530 2.184 x -2.184 2.184 10.53

0 00 00 00 00 00 0

Comment:

8.33

0.999

3.01

-400

-200

0

0

50

100

150

200

250

300

350

1 2 3 4 5 6 7 8 9 10 11 12 13 14pH

mef

f

B1B1

B2B2

Di-BaseDi-Base

Approved

NVNV

N

N

N

N

B1

B2

CE/MS instrumentation

HPCE3D , 1100 series LC/MSD trap SL (Agilent)

• Untreated fused silica capillary (50-60 cm/50 SEK) can be used as long as it works. • Buffers without filtration (stock solutions can be used up to >3 years).• Sheath liquid (1 L) can be used for more than 6 months (99% recycling).

• Higher initial investment

• Long-term savings

On-line CE/UV/MS experimental setup

UV @ 21.6 cmMS @ > 75 cmMS w/o UV @ 55 cm

Air coolingUV Detector interface

Capillary cassette

Capillary inlet

To MS

(Agilent 1100 series LC/MSD Trap)

Pressure (40-50 mbar)

Courtesy from Agilent (with small modifications)

Sensitivity from volatile vs non-volatile buffers in CE/MS

• Non-volatile buffer decreased signals at higher concentration (ion suppression)

• Ion source contamination by non-volatile buffer

020406080

100120140160

25 m

M NH4A

C50

mM N

H4AC

75 m

M NH4A

C10

0 mM N

H4AC

25 m

M sodiu

m/phos

phate

50 m

M sodiu

m/phos

phate

75 m

M sodiu

m/phos

phate

100 m

M sodiu

m/phos

phate

LidocainePapaverine

S/N

Not detectable

10 μM

Effects of buffer type and ionic strength on pKa

• Titration method uses physiological buffer with ionic strength at 0.15 M.• Ionic strengths (0.025 - 0.15 M) have a small effect on pKa (Δ pKa=0.064)

0

2

4

6

8

10

12

0 2 4 6 8 10 12

CE/UV (non-volatile buffer, I=0.05)CE/MS (volatile buffer, I=0.025)

pKa by CE/UV

pKa by CE/MS

Nicotine

PapaverinePilocarpine

NicotineLidocaine

Codeine

PropranololAtenolol

I0.3281I0.5085z

mmmlog pHpK

2

effm

efftha α+

+⎥⎦

⎤⎢⎣

⎡−

−=

0.00

1.00

2.00

3.00

4.00

5.00

6.00

7.00

8.00

9.00

10.00

11.00

12.00

pKa1(B) pKa2(HA)

I=0.01I=0.1I=0.15I=0.50I=1.0I=3.0

NH

ONH2

OHTryptophan (ACD)9.55/ 9.34/ 9.01/ 9.32/ 9.3/ 9.92

9.51/2.34 (CE/MS, I=0.025)

9.43/2.30 (CE/UV, I=0.01)

2.37 /2.37/ 2.23/ 2.39/ 2.75

Comparison of UV and MS (Ion trap) sensitivity

UV working concentration: 20-100 μM

0 5 10 15 20 25 30 35 40 Time [min]

0.0

0.2

0.4

0.6

0.8

1.07x10

Intens.[mAU]

Propranolol

AtenololCodeine

LidocainePapverine

Pilocarpine

UV

MS(EIC)

EOF

Sample: 0.5 µMAmmonium acetate,(pH=7) I=0.025Pressure: 40 mbar

Sensitivity and reproducibility of pressure-assisted-CE/MS

.

Total ion chromatogram (TIC)

SENS-C10.D: EIC 260 ±All MS

0

2 4 x10

Intens.

1 2 3 4 5 6 7 8 Time [min]

1 SENS-C20.D: EIC 260 ±All MS

0.0 0.5 1.0

6 x10

1 2 3 4 5 6 7 8 Time [min]

1SENS-C30.D: EIC 260 ±All MS

0.0 0.5

7 x10

1 2 3 4 5 6 7 8 Time [min]

0.1 μM S/N=1

1 μM (propranolol) S/N=235

10 μM S/N=864

1

SENS-C10.D: TIC ±All

0

1 6 x10

Intens.

1 2 3 4 5 6 7 8 Time [min]

0 2

6 x10

1 2 3 4 5 6 7 8 Time [min]


0

2 7 x10

1 2 3 4 5 6 7 8 Time [min]

0.1 μM

1 μM (11 mixed compounds)

10 μM


1

1

1

(EIC) and migration times after 30 runs

Ammonium acetate(pH=7) I=0.025;

pressure, 40 mbar

PKA-C036.D: EIC 79; 248 ±All,

2

4

66x10

MS

0 2 4 6 8 10 12 14 16 18Time [min]

0

5

UV-PKAC036.D: UV Chromatogram, 210.4 nmAR-H073379

AR-H073379

DMSO (2.5%)

pH=3

DMSO

?25 μΜ

On-line tandem UV and MS detection

Sensitive and selective MS detection over UV

79.3 125.0 167.0199.0

248.0 (M+1)

317.9

391.0

+MS, 7.3min (#974)

0.00

0.25

0.50

0.75

1.00

1.25

6x10Intens.

100 150 200 250 300 350 400 450 m/z

AR-H073379

Effects of pressure on migration times/sensitivity

+++++++++

+++++

++++

+++

++

+

++ +++++ +

+

+

++

++

+

+

+

++ ++ +

+

+ ++

+

+ ++

+

+ ++

+

+++

+

+

+

+

++++

CE capillary

sheath liquid

nebulizing gas

+++ ++ ++

CE sprayer tip

• Pressure reduced migration times

• Negligible signal suppression

• Improved RSD of effective mobility

• High throughput capacity

160

170

180

190

200

210

220

230

0 10 20 30 40 50 mbar

AtenololCodeineLidocaineQuinidinePilocarpinePropranololNicotine Papaverine

Pilocarpine+Papaverine

Codeine+Propranolol

0

500000

1000000

1500000

2000000

Intensity

0 10 20 30 40 50 mbar

AtenololCodeineLidocaineQuinidinePilocarpinePropranololNicotine Papaverine

Mass scan 70-660 m/zTime (s)

• Effective mobility shift around pKa (an interesting observation)• Mobility shift caused by pressure doesn’t affect pKa values.

H. Wan, A. Holmén, M. Någård, W. Lindberg, J. Chromatogr. A. 979 (2002) 369.

2 3 4 5 6 7 8 9 10 11

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5 Ibuprofen

50 mbar 25 mbar 0 mbar

Mef

f(10-4

cm2 V-1

s-1)

pH

pKa=4.5

Effects of pressure on effective mobility

min0 5 10 15 20 25 30 35

mAU

0

20

40

60

80

100

(c) P=50 mbar

(b) P=25 mbar

(a) P=0 mbar

1 2+3+N

1N

3

1

2

OH

O

OH

O

3 Benzoic acid

2 Ibuprofen

1 Lidocaine

(A) pH=3

40 min

NH

O

N

CH3

CH3

CH3

CH3

How many compounds can be pooled by CE/MS?

• DMSO < 5% in sample

• Constant mobility

• High MS resolution

• More than 150 cpds/sequence

0.00

0.50

1.00

1.50

2.00

2.50

3.00

3.50

0 1 2 3 4 5

PilocarpineLidocainePropranololAtenololPapaverine

012345678

0 1 2 3 4 5

DMSO (Peak height)

DMSO (Peak area)

DMSO %

DMSO %

μeff

0

1

2

3

4

5

6

7

8

9

10

11

0 1 2 3 4 5 6 7 8 9 10 11

N

O

O

OH

F

CH3

NNO

O

O

O

OH

CH3

NN

OO

CH3

pKa (Cal.) ACD6

pKa (CE/MS)

0

1

2

3

4

5

6

7

8

9

10

11

0 1 2 3 4 5 6 7 8 9 10 11

R2=0.992

pKa (literature)

pKa (CE/MS) (a) (b)

Cinoxacin

Flumequine

Phenylbutazone

R2=0.880

Comparison of measured pKa and lit./predicted values



Poor prediction for conjugated structures

Long-term validation of CE/MS pKa (2005-2008)

Scatter Plot

pKa(2005)

2

3

4

5

6

7

8

9

10

11

2 3 4 5 6 7 8 9 10 11

Antipyrine

Benzocaine

Diazepam

Furosemide

Ethionamide

Phenylbutazone

WarfarinMethotrexate

Glyburide

Enalapril

Sulfacetamide

Meclizine

Papaverine HCL

Amiodarone

Flumequine

Cimetidine

Tamoxifen

Clozapine

lidocaine

clonidine

Phenytoin

Quinine

Terbutaline

Haloperidol

ClofazimineTerfenadine

propranolol

Imipramine

Paracetmol

O

N

R2 =0.996

pKa: 8.71 (2005)

7.88/7.66 (2008)

Tamoxifen

10 buffers

14 buffers

An example of poorly soluble compound

-0.2

0

0.2

0.4

0.6

0.8

1

1.2

2 3 4 5 6 7 8 9 10 11

1. Ionization not close to chromophore2. Low solubility 0.45 μM (pH 7.4), LogD=6.5pKa (DPAS): nopKa (GlpKa): co-solvent, precipitation ?pKa (CE/MS): 8.71 (2005), 7.88/7.76 (2008)

(ref: 7.6)

Ref. (Karl Box et al., Anal. Chem., 2003, 75, 883-892).

HN+O

CH3

CH3

CH3

NO

CH3

CH3

CH3

H+

pKa = 8.7

Tamoxifen

pKa=8.71

AZ10009015-095 NA

pH meff res Fitting Option:2.362 87.540 x 85.8015775 -1.738423 87.54 2.3622.810 87.797 x 85.8010667 -1.995933 87.797 2.814.120 84.083 x 85.7856568 1.7026568 84.083 4.12 Value: Ionization: Ion. Func. Gr.4.720 89.378 x 85.7373889 -3.640611 89.378 4.72 pKa1:5.331 89.326 x 85.5392192 -3.786781 89.326 5.331 pKa2:5.714 80.971 x 85.1701977 4.1991977 80.971 5.714 pKa3:6.281 78.424 x 83.5164221 5.0924221 78.424 6.281 r2:7.127 71.436 x 71.9841923 0.5481923 71.436 7.1278.454 12.957 x 16.9045823 3.9475823 12.957 8.4548.826 14.763 x 8.09575778 -6.667242 14.763 8.8269.470 11.515 x 1.98220358 -9.532796 11.515 9.479.848 5.279 x 0.84143209 -4.437568 5.279 9.848

10.260 4.675 x 0.32782088 -4.347179 4.675 10.2610.776 2.467 x 0.10018246 -2.366818 2.467 10.776

0 00 00 00 00 00 0

7.76

0.991

Comment:

-20-10

010

0

10

20

30

40

50

60

70

80

90

100

1 2 3 4 5 6 7 8 9 10 11 12 13 14

pH

mef

f

NVNV

B1B1

Mono-BaseMono-Base

Approved

NVNV

O

N

pKa7.8

• Seven compounds as on-line QC for pKa screening (single measurement)

• Reproducibility <0.2 pKa units

Long-term reproducibility of QC (2005-2008)

QC AZ_1 AZ_2 AZ_3 AZ_4 AZ_5 AZ_6 AZ_7

Compound Atenolol Lidocaine Papaverine Nicotine Propranolol Quinine Codeine

Structure

pKa(B1/B2) Mean (n=50) 9.73 8.16 6.05 8.45(3.12) 9.67 8.88(4.00) 8.45

SD 0.09 0.09 0.11 0.10(0.27) 0.08 0.10(0.19) 0.08

Ref. 9.58 7.90 6.39/5.95 8.12(3.12) 9.508.50(4.1)

8.90 8.54

8.218.30

CE/UV 9.61 7.92 6.38 8.02(3.10) 9.49 8.39(4.14) 7.97

NH

O

N

CH3

CH3

CH3

CH3

N

NC

ONH

OH

CH3

CH3

N

N

CH2

OH

O

CH3H

O

N

OH

O

CH3

CH3

H

NH2

OONHCH3

CH3

OH

NH2

OONH

OH

Examples of determination of poorly soluble compounds

N

NH

S

N

O

O

CH3O

CH3 CH3

CH3

(c) Hexetidine

8.9

4.14

• Possible to measure pKa with solubility < 1 µM

• pKa =50% ionization ↔ (max. mobility)/2

• Mobility ≈ charge/mass

N

O

CH3

CH3

CH3

(b) Tamoxifen

(f) Omeprazole

-2

-1.5

-1

-0.5

0

0.5

1

1.5

2

(a) Imipramine(b) Tamoxifen(c) Hexetidine(d) Ketoconazole(e) Mebendazole(f) Omeprazole (g) Indomethacin(h) Sulfacetamide

μeff

2 4 6 8 10 (pH)

N N

NH2CH3

CH3 CH3

CH3CH3

pKa fitting example pH 2.5-10.5/10 buffers(2.3-10.8/14 buffers)

-1.5

-1

-0.5

0

0.5

1

1.5

2

2.5

2 4 6 8 10 12

Omeprazole

Clozapine

Chlorpromazine

Phenylbutazone

Ampicillin

Mebendazole

-3

-2

-1

0

1

2

3

2 4 6 8 10 12

SulfacetamideParacetmolClonidineCimetidineDebrisoquinEthionamideAntipyrine

N

NH

NH2

13 (ref)?

N

NH

S

N

O

O

CH3O

CH3 CH3

CH3

Omeprazole

Debrisoquin

S NH

O

O

O

NH2

CH3

5.270.08?

Sulfacetamide

8.9

4.14 mono-base

di-base

mono-base/acid

mono-acid

Summary

• Pressure-assisted CE/MS for pKa:• High throughput screening pKa from 2.3 to 10.8

• Reproducibility and accuracy of pKa ± 0.2 units

• Insensitive to compound purity, requiring minute sample

• Concn: (1-10 µM) beneficial for poorly soluble cpds

• Independent on type of buffer, capillary length, DMSO concn.

• Provide charge and structure/stability information (MS)

Acknowledgements• Anders Holmén, Mats Någård, Walter Lindberg, Yudong Wang, Marie

Englund, Richard Thompson, Johan Ulander

H. Wan, A. Holmén, M. Någård, W. Lindberg, J. Chromatogr. A. 979 (2002) 369.H. Wan, A. Holmén, Y. Wang, W. Lindberg, M. Friberg, M. Någård, R. Thompson, Rapid Commun. Mass Spectrom. 17 (2003) 2639.H. Wan, R. Thompson; Drug Discovery Today, Technologies, 2(2005), 171.H. Wan, J. Ulander, Expert Opinion on Drug Metabolism & Toxicology, 2 (2006), 1389.

• Eva Emanuelsson (validation), Fredrik Bergström (new pKa-fit template test)

• Eva Thorin (Activitybase/Dixy group), Michael Wirth Färdigh (automation)

• Zhiping You (new pKa-fit program), AZ Boston

• Lead generation, Physical Chemistry group, AZ Mölndal

Symp05/5_Hong Wan_R.pdf - PhysChem Forum

Documents

Transcript of Symp05/5_Hong Wan_R.pdf - PhysChem Forum