Back to the Top
Dear List.
Does anyone have any suggestions on how best to obtain accurate
plasma protein binding measures for compounds which exhibit very
rapid metabolism/decomposition in plasma?
Even if samples are placed on ice immediately after they are drawn
they continue to decompose whilst they cool. Pre-loading collection
tubes with ascorbic acid stabilises the samples to the extent that we
can take measurements of the parent fraction, but this obviously
affects the level of plasma protein binding.
Ultracentrifugation and equilibrium dialysis are seemingly too slow,
we hope ultrafiltration may provide better results; suggestions would
be gratefully received?
Robert A. Comley
Imaging Scientist
Clinical Imaging Applications Group
Translational Medicine and Genetics
GlaxoSmithKline
e-mail: Robert.A.Comley.-a-.gsk.com
[Ultrafiltration will take 10-15 min centrifuge time. Temperature
control might help but could effect the protein binding equilbrium - db]
Back to the Top
The following message was posted to: PharmPK
Dear Robert:
There are several layers to your question that need to be peeled away
from each other.
First, you cite both metabolism and decomposition (better still
chemical instability, either innate or sample collection/processing-
induced). These are quite different and need to be understood
separately.
If the primary concern is metabolic change to the parent compound,
this has several points of measurement, all of which may be different.
1. What is occuring in the sample (i.e., whole blood) in the instant
before it is drawn from the subject?
2. If one is drawing into any of the commerically available
collection tubes, there is a choice to be made between tubes with
zero additives:
--- changes in metabolism can be attributed to changes in sample
temperature, pH, precipitation, deactivation of blood-based metabolic
enzymes
or one of a number of anticoagulants, antioxidants, etc
--- metabolism might be stopped by cell lysing, precipitation,
enzymatic deactivation, denaturation, temperature, pH
3. In any event, if the metabolic half-life of the parent compound is
extremely short, it is likely that the binding constant can be taken
as equivalent to zero because there is no parent drug lasting through
the processing steps cited above.
4. That raises the issue of whether a metabolite can be used as a
surrogate marker for the binding constant - this is likely dependent
on whether the metabolite is sufficiently similar to the parent to
serve the role and the type binding available to either parent or
metabolite in the constantly changing sample you are processing.
5. So, if you can prove that your newly ex vivo sample is metabolized
much more rapidly than will allow for direct measurement of parent/
protein binding, it may be that you can eliminate the constant from
consideration.
6. If the half-life isn't that short, there are a variety of stop-
flow spectrophotometric and NMR methods that may assist you, in
addition to the methods you cite.
Second, if you are contending with chemical instability, you must
separate instabilities due to in vivo stability (which may go beyond
metabolism) and sample processing (#3 above). For instance, some
compounds are chemically stable but deteriorate on ex vivo exposure
to light, while others are chemically instable in vivo, but are
"immediately" stabilized in the collection tube.
Again, if the half-life of chemical deterioration is so rapid that
there is effectively no parent left within the time you have taken
during sample processing, the constant is estimated at zero.
So, a clear assessment of the moving parts is your initial task. You
may get a free pass at k=0, or a rough time scheduling the proper
cutting edge stop-flow or NMR time.
Best of luck,
Ian Davis
Director, Operations
Strategic Consutling Services
Pharsight Corporation
Back to the Top
The following message was posted to: PharmPK
Dear Dr Comley,
Let me suggest the use of a technology known as "surface plasmon
resonance biosensoring".
Provided you have access to this fairly expensive equipment,
including purified proteins of interest immobilized on biosensors,
measurements are extremely quick. This makes the method extremely
attractive for investigations on unstable compounds.
If you are not quite familiar with the technology, the principle
consists in immobilizing a protein/receptor on a biosensor, and
(indirectly) measure ligand-induced changes in its molecular weight.
The measures essentially give access to parameters of binding
kinetics (affinity constants, number of binding sites). The fraction
bound is a secondary parameter, derived from the above.
The beauty of the technique is that you do not need an assay method
for the ligand. Not even a radiotracer.
See for instance: Frostell-Karlsson A, Remaeus A, Roos H, Andersson
K, Borg P, Hamalainen M, Karlsson R. Biosensor analysis of the
interaction between immobilized human serum albumin and drug
compounds for prediction of human serum albumin binding levels. J Med
Chem. 2000 May 18;43(10):1986-1992.
Hope this helps,
Henri MERDJAN
Head of Drug Metabolism and Pharmacokinetics
NOVEXEL S.A., Parc Biocitech, 102 Route de Noisy
F-93230 Romainville, France
Tel +33 (0)1 57 14 07 45
Back to the Top
The following message was posted to: PharmPK
Dear Robert:
Consider changing your approach to cooling and sample collection, if
possible. The normal routine for blood sampling in the clinic is to
first collect the blood (e.g into a vacutainer) and then transfer the
tube into wet ice. As you've pointed out, your sample is decomposing
while it cools. You've got warm blood, inside a vial with a warm
inner wall, and it's made of glass which is not the ideal material
for heat transfer -- thus, it takes many minutes to get cold. It's
even worse if you collect in a syringe, transfer to a tube, and then
cool the tube because you are also adding an aeration step during the
transfer.
The cooling process will be accelerated considerably by collecting
blood into a vial which is already cold (<4*C) and by increasing the
surface area of blood exposed to the cold surface. This means
collections into smaller vials (so that the cylindrical shape of the
blood in the vial is both higher and smaller in diameter and thus in
contact with more cold surfaces). We use thermoelectric cooling and
automation, but you could try this manually with a few whole blood
samples to determine whether it helps? You can observe the
improvement in cooling by placing a thermocouple into your standard
vial, monitoring it with a voltmeter, and emulating your current
process. Then compare to a vial that is prechilled, taller and
narrower. Note the time it takes to reach a refrigerated setpoint,
such as 3*C.
There is an alternate method of ultrafiltration, using membrane
probes immersed in the sample itself. There is a comparison of this
method with centrifugal ultrafiltration in H. Lam, M. Davies and C.E.
Lunte, "Vacuum ultrafiltration sampling for determination of plasma
protein binding of drugs" J. Pharmaceutical and Biomedical Analysis:
14 (1996) 1753-1757. This can be done using a cold sample.
Candice Kissinger
Senior V.P. Research
BASi
West Lafayette, Indiana
Back to the Top
Hi Robert,
In the usual sense, db is right that ultrafiltration takes 10s of
minutes, but it is not a linear process. Instead of putting 0.4mL
into a filter and spinning for 30 minutes to get 50uL, spread the
plasma out over several filters - say 10 at 100uL each - and spin
them fast for 1 minute. (If some burst, discard them.) You will get
5-10uL filtrate in each tube, pool them and analyse the pooled
filtrate. If you are right that it is plasma proteins that are
responsible for the decomposition, then once filtered the compound
should be stable. Check this against whole plasma kept at the same
temperature for 1 minute.
Some of the other replies about the blood sampling should help.
Regards.
Ted
Back to the Top
The following message was posted to: PharmPK
Hi all,
I have seen several comments and suggestions regarding plasma protein
binding methods, but none have mentioned high throughput methods such
as the "BD Gentest Serum Binding System". The company claims
excellent results for a number of drugs, but I wonder if anyone is
using this on a regular basis and can comment on the pros and cons
(limitations) of this or other commercially available systems.
Thanks,
William Jones
Research Associate
College of Pharmacy
The Ohio State University
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)