The role of chromatography in physico-chemical characterisation

1
The role of chromatography in physico-chemical
characterisation

• Shenaz Nunhuck
• CASS, GSK

2
Why do we need physchem measurements?

• Physicochemical properties of drugs influence
  their absorption and distribution in vivo
• Systemic absorption of drug involves a number of
  rate processes
• Distribution of the drug in the body
• Dissolution of the drug in the body fluids
• Permeation across the cell membranes to reach the
  site of action.
• Key physicochemical parameters influencing these
  processes are lipophilicity, solubility, pKa,
  permeability

3

• PHYSCHEM
• ASSAYS

LIPOPHILICITY LogD (oct), CHI
IONISATION CONSTANT
PLASMA PROTEIN BINDING
AQUEOUS SOLUBILITY
MEMBRANE PERMEABILITY
4
Sample flow process
5
Lipophilicity measurements

• Chromatographic Hydrophobicity Index (CHI)
• Immobilised artificial membrane (IAM) partition
• Protein binding (human serum albumin,
  alpha-1-acid-glycoprotein)
• Octanol/water partition coefficient (LogP/D(oct)

6
Theoretical basis of using chromatography for
measuring lipophilicity

• Different compounds travel at different speeds in
  the chromatographic system.
• The differential migration depends on the
  interaction of compounds between the mobile and
  stationary phase.
• Retention factor is directly related to the
  chromatographic partition coefficient.

k number of mol in the stationary phase/number
of mol in the mobile phase k (tR - t0 )/ t0
log k log K log (Vs/Vm) k is retention
factor log K is the log of the chromatographic
partition coefficient Vs/Vm is constant column
parameter (the ratio of the mobile and
stationary phase volumes)
7
Chromatographic Hydrophobicity Index, CHI

• Fast gradient methods the gradient retention
  time is proportional to the compound
  lipophilicity.
• Fast gradient retention time obtained on
  commercially available C-18 stationary phase
  converted to Chromatographic Hydrophobicity Index
  (CHI) this is the chromatographic lipophilicity.
• The CHI Indices at three different pHs are
  determined from the gradient retention times
  obtained by injecting the compound into a HPLC
  system.
• Dynamic range extended by the gradient method.
• Can be expressed on a logP/D scale.
• (CHIlogD 0.054CHI -1.467)

8
Generic Gradient HPLC ( Four minute CHI method)

• CHI is derived directly from a reversed phase
  chromatographic gradient retention time.
• Luna C-18 column, bufferacetonitrile gradient,
  pH2
• Luna C-18 column, bufferacetonitrile gradient,
  pH7.4
• Luna C-18 column, bufferacetonitrile gradient,
  pH 10.5
• Each run time is 4 minutes.
• Retention times are converted to CHI
  lipophilicity values after calibration.
• Column Luna C18(2), 50 x 3.0 mm id, 5 ?m
• Flow 1.00 ml/min
• Mobile phase
• A 50 mM ammonium acetate pH 7.4/10.5,
• 0.1M H3PO4
• B Acetonitrile
• Gradient 0-100 B in 2.5 minutes,
• hold at 100 B for 0.5 minute,
• return to 0 B in 0.2 minute,
• equilibrate at 0 B for 1.8 minutes

9
Derivation of CHI

• A set of calibration compounds of known CHI
  values (determined isocratically) is run.
• A plot of Rt v/s CHI gives the calibration curve.
• Research compounds are run. The Rt is converted
  to CHI from the coefficients of the calibration
  curve.
• CHI (slope x Rt) intercept

10
Immobilised Artificial Membrane (IAM)

• Immobilised Artificial Membrane-
  phosphatidylcholine (PC) head group with an ester
  linkage between two acyl chains and the glycerol
  backbone of the PC molecule.
• Phosphatidylcholine (PC) is the major
  phospholipid found in cell membranes.
• IAM stationary phases prepared from PC analogs
  closely mimic the surface of a biological cell
  membrane.
• CHI IAM are extensively used in GSK for various
  purposes
• Brain penetration models
• Hepatoxicity models

Schematic diagram of the IAM.PC (CH2)12
stationary phase surface
11
Fast HPLC method to measure interaction with
Immobilised Artificial Membrane (IAM), CHI IAM
tR 7.4 CHI IAM Compound
3.284 49.4 Octanophenone
3.167 45.7 Heptanophenone
3.033 41.8 Hexanophenone
2.866 37.3 Valerophenone
2.658 32 Butirophenone
2.415 25.9 Propiophenone
2.093 17.2 Acetophenone
1.893 11.5 Acetanilide
1.648 2.9 Paracetamol

• A set of calibration compounds of known CHI IAM
  values (previously determined isocratically) is
  run.
• A plot of Rt v/s CHI gives the calibration curve.
• Research compounds are run. The Rt is converted
  to CHI from the coefficients of the calibration
  curve.
• CHI (slope x Rt) intercept

Column IAM PC2 (CH2)12 150 x 4.6 Flow rate 2
ml/min Gradient 0 to 2.5 min 0 to 70
acetonitrile 2.5 to 3.3 min 70
acetonitrile 3.3 to 3.5 min 0
acetonitrile Buffer 50mM NH4AC, pH 7.4 Cycle
time 4 min
Ref Valko et al, Rapid gradient HPLC method for
measuring drug interactions with immobilised
artificial membrane Comparison with other
lipophilicity measures. J Pharm Sci 891085-1096
12
4-way parallel HPLC system
C-18 pH 10.5
C-18 pH 7.4
IAM
C-18 pH 2
13
4 chromatograms of one compound by the 4-way HPLC

• Typical 4-way chromatograms of a base

14
CHI values and the acid/base character
CHIs measured at 3 pHs provide an automatic way
of grouping molecules according to acid/base
character without a need for structural
information.
15
Automated Data Processing

• All data from the 4 systems are stored in a
  single data file
• Complex data processing.major rate-limiting step
  as this requires visual inspection of each
  chromatogram for identifying the major peak.
• Datect LC-validator software is run after data is
  collected on Chemstation
• Chromatograms are automatically inspected and
  validated by the software.

16
Automated review of Chromatograms

• Peaks of interest are identified, their shape
  examined and retention time automatically
  transferred into excel spreadsheets for further
  data processing.
• Software highlights any anomalies and generates
  explanatory error messages prompting expert
  visual inspection.
• Customised alerts are set up by the user.
• Most of our validation failures occur when the
• Major components peak area is less than 80 of
  total, indicating that the compound is probably
  impure.
• Major peak retention times are not identical at
  various wavelengths.
• Major peak absorbance is weak indicating lack of
  chromophore or absence of compound.

Datect LC-validator (www.datect.com).
17
Commercially available 4-way HPLC-MS instrument
from Waters
18
Biomimetic hplc stationary phases (HSA, RSA, AGP)

• Used to measure the binding affinity of compounds
  to proteins.
• Plasma protein binding affects the unbound (free)
  drug concentration available to diffuse from the
  blood and reach the target tissue.
• Commercially available human and rat serum
  albumin and a-acid glycoprotein hplc stationary
  phases (available from ChromTech Ltd)

19
Plasma Protein Binding

• Fast generic gradient hplc method based on
  propan-2-ol gradient and chemically bonded HSA or
  RSA column.
• Warfarin site is the major binding site on HSA.
• By injecting a racemic mixture of warfarin on the
  column, the R and S enantiomer are separated
  indicating the warfarin site is intact and the
  column is suitable for use.
• Only 6 minutes analysis time

Column Chromtech HSA 50 x 3 mm Flow rate 1.8
ml/min at 300C Mobile phase 50 mM ammonium
acetate pH7.4/Propan-2-ol Gradient 0 - 3 min 0
to 30 propan-2-ol 3 to 5 min 30 propan-2-ol 5
to 5.1 min 0 propan-2-ol Cycle time 6 min
Ref Valko et al, 2003.Fast gradient HPLC method
to determine compounds binding to human serum
albumin. Relationships with octanol/water and
immobilised artificial membrane lipophilicity. J
Pharm Sci 922236-2248
20
Calibration and Results
System is calibrated using literature plasma
protein binding data.
Calibration compounds Literature binding
Nizatidine 35
Bromazepam 60
Carbamazepine 75
Piroxicam 94.5
Nicardipine 95
Warfarin 98
Ketoprofen 98.7
Indomethacin 99
Diclofenac 99.8

• Calculate Binding
• logK slope log(tR) intercept
• K B / (101-B)

• Results are reported as bound or logK

21
Plasma Protein Binding

• Reproducibility on the column is very good the
  gradient retention time is within 0.1 min from
  day to day
• The hplc method is fast, simple and is easily
  automated.
• The use of calibration compensates for any
  changes in the column properties and hence
  increases the accuracy of the determination.
• The hplc procedure can discriminate easily in the
  high binding region (better than the traditional
  ultrafiltration or equilibrium dialysis methods)
  as the percentage of drug bound to the protein is
  measured and not the free drug.
• Approximately 400 injections per column

22
Chromatographic methods for quantitative assays

• HPLC is a powerful technique for separation and
  quantification
• Suitable approach for determination of compound
  concentration
• Applied as end-point for logP(octanol) and
  solubility determination

23
LogP(octanol) shake-flask determination

• Equilibration of the compound between n-octanol
  and water in 96-well plate
• Determination of concentration of the compound in
  each phase by fast gradient hplc method.
• The syringe in the autosampler is set to sample
  first at the depth of the octanol phase in the
  well and then at the depth of the aqueous phase
  without any cross contamination
• Ratio of the peak areas obtained from the aqueous
  and octanol phases directly provides partition
  coefficients

LogD(oct) Log( Peak area of sample in
octanol phase x Injn vol. (aqu) Peak
area of sample in aqueous phase) Injn vol. (oct)
24
Why is aqueous solubility important in early drug
discovery?

• Solubility is a key property for gastrointestinal
  absorption of orally administered drugs.
• Affects bioavailability
• Helpful in drug formulation stages for optimal
  drug delivery route and optimization
• Insoluble compounds may compromise screening
  results.
• Various solubilities
• DMSO precipitative solubility
• Solubility from solids
• Solubility in simulated intestinal fluid (SIF)

25
HPLC-based Precipitative Aqueous Solubility
incubation filtration
Sample 500 uM in pH 7.4 aqueous buffer
data in GSK database
compounds dissolved in DMSO at 10 mM
fast gradient generic HPLC method
Tecan
50 uM standard in DMSO
26
Quantification by HPLC

• Fast automated sample preparation
• Gradient HPLC method same as the CHI method
• Sample and standard solutions injected next to
  each other (single point calibration)
• Data collected at two wavelengths
• Impurities separated
• Automated data processing using in-house macro
• Macro identifies the peak of interest in the
  standard solution and matches it with that in the
  sample solution
• Peak area and retention time data exported to
  excel

Solubility of sample Peak area of sample
X Conc. of standard Peak area of standard
27
Artificial Membrane Permeability

• High throughput assessment of compound intestinal
  permeability
• Cultured cell monolayer with reconstituted lipid
  membrane

• Lipid is egg phosphatidyl choline and cholesterol
  dissolved in n-decane.
• Permeation experiment is initiated by adding the
  compound to the bottom well and stopped at a
  pre-determined elapsed time.
• Samples are analysed by HPLC/UV or/MS

P (Vd/a t) ln (RX)/R(1-X) 1/(1R) t is the equilibration time in s Vd is the volume of donor solution in cm3 Vr is the volume of acceptor solution in cm3 a is the membarne surface area in cm2 R is the Vd/Vr ratio A is the acceptor side peak area D is the donor side peak area X is the A/D peak area ratio
28
Advantages of using HPLC technique

• The compounds retention time can be directly
  related to the distribution between the
  stationary and the mobile phase, there is no need
  for concentration determination.
• By changing the stationary phases and the mobile
  phase composition various types of lipophilic
  interactions can be investigated.
• Impurities do not affect results as they are
  separated from the main peak and the compound of
  interest can be identified.

29
Advantages of using HPLC technique

• Only small amount of material is needed.
• Parallel systems can be used to lower cost and
  increase throughput.
• With generic gradient hplc method, one method can
  be used with a variety of compounds there is no
  need for individual customised method
  development.
• Provides an excellent platform for computer
  controlled automated measurements with
  computerised data acquisition.

30
CONCLUSIONS

• HPLC provides an excellent generic platform for
  measuring lipophilicity, acid/base character and
  bio-mimetic partition properties.
• With the application of gradient methods and
  system calibration with known compounds, large
  amounts of reproducible data are obtained
  covering a wide dynamic range of the property.
• The extensive application of automated platforms
  and parallelised chromatography has enabled
  hundreds of thousands of determinations to be
  made per annum with a minimum of labour.
• The data are suitable to build local and general
  models to predict developability properties in
  early stages of drug discovery.

31
Acknowledgements

• Klara Valko
• Chris Bevan
• Alan Hill
• Pat McDonough

32
GlaxoSmithKline is currently recruiting for a
Physicochemical Scientist in Harlow - Essex,
Southeast England.At GlaxoSmithKline (GSK), one
of the world's leading healthcare businesses, we
discover, develop and produce products that help
people live longer, do more and feel
better.Minimum Requirements You will have a
BSc or equivalent experience in Chemistry,
Analytical Chemistry or related discipline and
have experience within an analytical laboratory
environment.
33
Difference between CHIlogP and octanol/water logP

• Chromatographic lipophilicity is not the same as
  the octanol/water lipophilicity
• H-bond donor compounds (Series 1) partition more
  into octanol because the octanol OH groups can
  interact with the solute H-bond donor group
  resulting in higher logP values (i.e they look
  more lipophilic).
• The two scales of lipophilicity can be aligned by
  introducing a H-bond acidity (A) term or a count
  of H-bond acid groups on the molecules (HBDC)
• LogPoct 0.05CHIlogP 0.41HBDC 1.41
• N86 r0.94 s0.40
• where HBC Hydrogen bond donor count
• LogPoct 0.054CHIlogP 1.319A 1.877
• N86 r0.97 s0.29
• where A Calculated Hydrogen Bond Acidity