Back to the Top
The following message was posted to: PharmPK
Dear Group
We have spent an extensive amount of time measuring protein-binding of
one of our drugs. It seems (even by dialysis) that the drug is 100%
bound (the dialysate contains <<1 ng/ml even though the plasma cell is
spiked 250, 2500 or 25000 ng/ml). This is a result we have seen
consistently on a number of occasions and by a number of techniques. The
big question is:
Can a drug that is effectively 100% protein-bound show any efficacy and
does anyone know of other drugs that have this same strange
characteristic??
Best wishes
David Jones
NeuroSearch A/S
Denmark
Back to the Top
The following message was posted to: PharmPK
David
Drugs tightly bound to albumin do exist. Indeed albumin has been
used as a carrier to bring drugs to the target
eg
J Immunol. 2003 May 1;170(9):4793-801
Albumin-based drug delivery as novel therapeutic approach for
rheumatoid arthritis.
Wunder A, Muller-Ladner U, Stelzer EH, Funk J, Neumann E, Stehle G,
Pap T, Sinn H, Gay S, Fiehn C.
Regards
Colin
Professor Colin Garner BPharm PhD DSc FRCPath
Chief Executive Officer
--
Xceleron Ltd
York Biocentre
Innovation Way
Heslington
York YO10 5NY
United Kingdom
Switchboard: +44(0)1904 561561
Direct line: +44(0)1904 561568
FAX: +44(0)1904 561560
Email: colin.garner.-at-.xceleron.com
Back to the Top
The following message was posted to: PharmPK
According to the law of mass action you must have equilibrium between
the
free fraction and the bound fraction, except if you have a covalent
binding.
Only the free fraction is active because it diffuses freely towards
tissues.
The bound fraction is temporarily retained in the plasma compartment
Francoise BREE
BIOPREDIC/ DRUGGABLIS
Parc d'affaire de la Breteche
Batiment B1
35760 Saint Gregoire
France
Back to the Top
The following message was posted to: PharmPK
David,
A couple of questions that might shed light on the phenomenon you're
observing: How long are you dialyzing for (we routinely dialyze for 20
hours, but many highly bound compounds, including warfarin, do not reach
equilibrium)? Do you know your compound is stable in plasma? Do you
know its aqueous solubility (on the receiver side)? Have you tested for
non-specific binding (to the receiver side)? These issues could result
in a near-zero concentration on the receiver side.
Adrian Sheldon
In Vitro ADMET
Charles River Labs
Back to the Top
Hi David,
IMHO, the fup (fraction unbound in plasma) alone cannot be used to
try to
predict what will happen in vivo. It's an "end point" measurement, not a
dynamic one. A very low fraction unbound is certainly a sign that the
compound has high affinity for the plasma proteins, but what is
important to
know is whether the target has a higher affinity than serum proteins
or not.
If your compound has a very high affinity for its target (greater than
affinity for other proteins, i assume that your compound's target is a
protein too), a very low plasmatic fu will not prevent the compound
to be
active in vivo, because the equilibrium will be displaced toward the
target.
Of course, you must have a way to estimate affinity for the target in
vitro
to know that, and i can easily imagine that is is not practically
doable for
all sorts of compounds.
Best regards
Cedric
Back to the Top
David,
I am assuming that the low concentration in the dialysate is not due
to poor recovery (precipitation or binding to the device). One trick
to look at highly bound drugs is to dilute the plasma - a 10-fold
dilution will expand the 90 - 100% dynamic range by 10-fold and allow
you to discriminate between highly bound drugs.
Yes, highly bound drugs can be efficaceous. Interactions with
transporters or other carriers can strip plasma proteins of drug and
allow entry to cells. What is more important than the extent of
binding is the off-rate for equilibriation.
Note that, if a drug is bound predominantly to a high
capacity/concentration plasma component, such as albumin, even 99.9%
binding does not indicate high affinity. The theoretical values are;
90% bound 60.73 micromolar
99% bound 6.07 micromolar
99.9% bound 0.61 micromolar
99.99% bound 0.06 micromolar
So, to achieve tight binding single digit nanomolar Ks values the
binding
would have to be e.g. 99.999%. Note that this is not the case for
carrier proteins with lower concentrations, including AAG.
I hope that this helps.
All the very best,
Bernard
--
Bernard Murray, Ph.D.
Research Investigator/Scientific Leader
Drug Metabolism, PCS, PPD, GPRD, Abbott Laboratories, Chicago, USA
Bernard.Murray.at.abbott.com
Back to the Top
Hi David,
Given 100% binding, I am interested as to the clearance of the
compound. If a compound is truly 100% bound then there will be no
free drug for uptake into hepatocytes, glomerular filtration etc. I
suspect it is more likely that the free fraction is below your LOQ or
it is subject to non-specific binding.
I have often observed colleagues attenuating an assay EC50 using an
equilibrium, fraction unbound measurement in order to determine
concentrations that need to be achieved in vivo. However, when we
have investigated the phenomena of the effect of free fraction on
efficacy, the rightwards shift is less pronounced - due to the
affinity of the compound for the target receptor.
The NSAIDs are an example of compounds with very high affinities to
albumin with equilibrium binding constants for ibuprofen of 2.73 x
10E6 for one site and 1.95 x 10E4 for other site (see reviews by
Kragh-Hansen:
Molecular aspects of ligands binding to serum albumin. [Review] [297
refs]1981
Practical aspects of the binding-binding and enzymatic properties of
human serum albumin. [Review] [139 refs]2002)
Determination of compound affinity to protein (plasma protein or
receptor) is important and can be done via fluorescent methods with
fluorescent probes or using isothermal titration calorimetry (ITC).
Through knowledge of your equilibrium binding Ka or Kd for the
various protein binding sites, you can calculate the theoretical free
fraction when using albumin or AAG.
Binding to any protein is a dynamic process and can be observed using
Biacore technology or even the 'old fashioned' stop flow techniques.
Unless it is covalently bound to the albumin, your compound will
dissociate given the right conditions. We see this when our hepatic
extraction ratio is larger than our free fraction - free compound is
cleared AND so is that which has dissociated from the plasma protein.
There is also scope for compounds to be displaced from albumin (see
literature by Amitava Dasgupta). The off rate is a controlling factor
in these observations and has been observed to be rate limiting in
the uptake of fatty acids by some investigators.
I hope this helps,
Mark
Mark Baker
UCB
United Kingdom
Back to the Top
The following message was posted to: PharmPK
Warfarin is a classic example of extensively bound drug 99.99 % bound to
plasma proteins mostly on to serum albumin. A word of caution is you
have
to ensure that you are not losing any drug due to adsorption onto the
devices and when using dialysis technique you have to correct for volume
loss due to long dialysis times.
Hope this helps.
Prasad Tata, Ph.D.
Mallinckrodt, Inc.
Back to the Top
The following message was posted to: PharmPK
Charles River Labs,
I have done a lot of protein binding experiments using
ultracentrifugation and ultrafiltration. Regarding
non-specific binding, ultrafiltration is the worst,
ultracentrifugation is better. It has been well known
that Dialysis supposes to give negligible non-specific
binding (NSB), since NSB happens on the both sides,
then NSB can be eliminated.
Hope this can help.
Xiaodong
Back to the Top
The following message was posted to: PharmPK
Dear Cedric,
You say that if the "compound has a very high affinity for its
target.... the equilibrium will be displaced toward the target". Others
have replied along the same line. I think this depends. It depends on
where the target presides and the number of targets (number of binding
sites) and, obviously, the rate constants (on & off). If the target is
an intracellular receptor (e.g., intranuclear) and present at low
abundance (e.g., 1000 sites per cell, i.e., no sink conditions) this
will hardly affect the equilibrium because in the first instance this
will be between the free drug outside and inside the cell (cytoplasm).
In other words, a drug that enters cells through passive diffusion, and
has an intracellular target, could be very dependent on plasma protein
binding for ist action. However, when there is active transport
("transporters and other carriers", see Bernard Murray's reply) involved
this situation changes, obviously. One way of looking at it is that this
is Nature's way of preventing xenobiotics of diffusing into cells by
mopping them up through plasma protein binding (and then delivering them
to liver & kidneys for elimination).
I'd like to hear what you think.
Kind regards,
Frederik Pruijn
Back to the Top
The following message was posted to: PharmPK
Dear Frederik,
I completely agree with your point of view : a lot of physiological
parameters should be taken into account when trying to make
predictions from
in vitro data. I just wanted to point out that fu alone cannot
predict in
vivo behaviour, the main reason for me being that it is not a "dynamic"
measure (contrary to the affinity constants mentionned by Bernard, or
other
constants -Ka, Kd...- that define equilibriums more dynamically).
Best regards,
Cedric
Back to the Top
The following message was posted to: PharmPK
100% binding is no guarantee of no biological activity. Pyrethroid
insecticides (log P=5-10) are only drugs in the sense of controlling
invertebrate parasites. Nevertheless, they are fantastically
strongly bound
in all situations, eg in the mammalian circulation, and yet they are
also
very biologically active (eg neurotoxic). The key is that their
binding is
very readily reversible - they merely hitchhike from one lipophilic
substrate to another. I'm sure some more conventional drugs will also
behave in this way. If you can work out what distinguishes a
strongly-bound
but readily mobile compound from one that is strongly bound but stuck
on one
place, I'd be interested to know how,
Regards,
Kim Travis
Syngenta
Back to the Top
Hi Kim
Could you provide some references on protein binidng of pyrethroids
in rodents? We are working on type II pyrethroids and this info will
help us in modeling
satheesh
Back to the Top
The following message was posted to: PharmPK
I find the discussion on this topic very fascinating. To my opinion the
discussion is now mainly focussing on mechanisms of action. I want to
bring
in some other aspects.
In the assessment of (preclinical) pharmacokinetics some other issues
regarding very high protein-binding play also a role:
* what is the margin of safety, i.e. how essential is it to know the
exact
protein binding in the various species. If the MOS is low, then you
have to
be very sure of the figures you give for the protein binding
* what happens in the clinical situation in case e.g. albumin levels
drop
or in case of interactions with other drugs with a high
protein-binding/high affinity for albumin.
Adrienne
Back to the Top
The following message was posted to: PharmPK
Dear Kim,
As somebody else has already pointed out the Biacore technology is one
way of distinguishing between the different types of interaction you
mention.
Go and check out their Website, particularly this one (but there's heaps
more):
http://biacore.com/lifesciences/technology/introduction/data_interaction
/index.html
Briefly, it comes down to the on and off rates of binding, as others
have explained too.
HTH
Frederik Pruijn
Frederik B. Pruijn PhD MSc (Senior Research Fellow)
Experimental Oncology Group
Auckland Cancer Society Research Centre
Faculty of Medical and Health Sciences
The University of Auckland
Private Bag 92019
Auckland
New Zealand
Phone: +64-9-3737 599 x86939 or x86090
Fax: +64-9-3737 571
E-mail: f.pruijn.-at-.auckland.ac.nz
Back to the Top
The following message was posted to: PharmPK
Dear Adrienne,
Your question asks about the role of protein binding in
pharmacodynamics and the role of protein binding in pharmacokinetics.
In my view protein binding is very important in pharmacodynamics,
because it is only the unbound drug concentrations that can leave the
blood and bind to the target in the relevant tissue(s). At steady
state unbound plasma concentrations are in equilibrium with the free
concentrations at the site of action. Therefore, a fixed ratio exists
between total plasma concentration and the free concentration at the
site of action. The activity of many compounds is determined in vitro
and expressed as IC50's or EC50s. These are free concentrations, thus
correspond to the free concentrations at the site of action at steady-
state. In order to know the corresponding total plasma concentration
at steady state in the blood in the different species, one only has
to correct for protein binding.
Assumptions are that the IC50's in the different species are the
same, the compound binds reversibly to the target and there is no
active transport to the site of action.
In pharmacokinetics protein binding is less important. If albumine
levels are decreased for instance, the free drug that is come
available is almost instantly sucked away by the tissues. As a
result, total concentrations and free concentration in the blood are
hardly affected. For more info, see a nice review of Benet and
Hoener, Clin Pharmacol Ther. 2002 Mar;71(3):115-21, Changes in plasma
protein binding have little clinical relevance.
Best regards (en groeten),
Kees
Kees Bol
Director Clinical Research
Kinesis Holding B.V.
Consultants in Drug Development
Lage Mosten 29
4822 NK Breda
The Netherlands
Back to the Top
The following message was posted to: PharmPK
Dear Kees and Adrienne,
With respect to your comments on the role of protein
binding in pharmacodynamics and pharmacokinetics (actually
not directly related to the topic of the original thread):
Kees wrote:
>In order to know the corresponding total plasma
>concentration at steady state in the blood in the
>different species, one only has to correct for protein
>binding.
One of the discussion points in the thread was that this
correction is not so simple as suggested here, because
binding to the target affects the unbound concentration in
the vicinity of the target.
>Assumptions are that the IC50's in the different species
>are the same
Indeed, but this assumption needs some justification!
There are many exceptions on this 'rule'.
> In pharmacokinetics protein binding is less important.
I don't agree with this general statement. Plasma protein
binding is equally important for PK and PD. The fact that
plasma protein binding has little clinical relevance (see
below) does not imply that plasma protein binding is not
important!
>If albumine levels are decreased for instance, the free
>drug that is come available is almost instantly sucked
>away by the tissues. As a result, total concentrations
>and free concentration in the blood are hardly affected.
>For more info, see a nice review of Benet and Hoener,
>Clin Pharmacol Ther. 2002 Mar;71(3):115-21, Changes in
>plasma protein binding have little clinical relevance.
Yes, but there is one mistake. A change in plasma protein
binding (thus a changed fraction unbound fu = Cu / C)
cannot be associated with both unchanged C and Cu. In
fact, the TOTAL concentration (at steady state) in blood
or plasma is strongly affected by plasma protein binding,
and the unbound concentration is hardly affected by fu
(except for intravenous administration of high hepatic
extraction drugs).
Best regards,
Hans Proost
Johannes H. Proost
Dept. of Pharmacokinetics and Drug Delivery
University Centre for Pharmacy
Antonius Deusinglaan 1
9713 AV Groningen, The Netherlands
tel. 31-50 363 3292
fax 31-50 363 3247
Email: j.h.proost.aaa.rug.nl
Back to the Top
Dear Kim,
Biacore is a viable option for measuring on and off rates (a good
introduction: Frostell-Karlsson et al. (2000) J Med Chem 43:
1986-1992). However a problem exists with tightly bound compounds
where it is difficult to detach them from the albumin that is bound
to the plate. A recent alternative is the Akubio system which is
based on resonance acoustic profiling (http://www.akubio.com/). Two
older, other alternative exist in stop flow fluorescence and dilution
experiments. The first using fluorescent detection of dynamic changes
in the tryptophan of albumin, and the second measuring the effect of
diluting an albumin-compound concentration.
The utility of these techniques depends on your compound - solubility
etc.
These rates, combine to give a Ka or Kd (which can also be measured
separately if you have multiple binding sites).
Hope this helps,
Mark
Mark Baker
UCB
United Kingdom
Back to the Top
The following message was posted to: PharmPK
Hi Xiaodong,
Currently I'm doing protein binding experiments and I
found my results are quite different from the
literatures. I'm using ultrafiltration method. How
different will the result be? I mean how worse will it
be? Mine is almost ten times higher than the
literature.
Thank you.
Cheers,
Xin
[Ten times tighter binding? Are you loosing material by binding to
the filter, holder? - db]
Back to the Top
Have you check for non-specific bindings? High protein bound compound
tend to be more stick to all those surfaces.
Best Regard. Ta Kung
Back to the Top
The following message was posted to: PharmPK
All,
For very highly bound compounds or compounds with a non-specific binding
problem, the method we recommend is an ultra-centrifugation-based
approach.
There have been several reports demonstrating the validity of this
method.
Howard
Howard C. Haspel, Ph.D.
Study Director/Scientist II --- In Vitro ADMET
Metabolism and Pharmacokinetics Department
Charles River Laboratories
Preclinical Services
57 Union Street
Worcester, MA 01608 USA
Email: Howard.Haspel.-at-.us.crl.com
VOX: (508)890-0125
FAX: (508)797-4091
Back to the Top
Dear Xin,
ultrafiltration is very reliable in case of compounds that go freely
through the filtration membran, which many compounds don't.
We always perform a controll experimetn in buffer and when the
concentration in the filtrate is >= 75% of the concentatio in the
retentat we use the method (about 50% of the compounds fail). Which
method was used for your compound in the literature?
Regards,
Markus
Back to the Top
The following message was posted to: PharmPK
Hi Xin
For ultafiltration method you need to correct the results of protein
binding or non specific binding to the device.
We have reported the equation that you can use for correction of
this method. Please see following paper.
Drug Metab Dispos, 2004, 32, 11, 1213-7
Thanks
Amir Heydari
Back to the Top
An example of highly protein bound drug could Antimalarial drugs such
as Atovaquone which is highly protein bound > 99%. From my
understanding, the efficacy will also depend upon the affinity of the
drug for the target. If the affinity for the target is higher than
the affinity for the protein binding, then the drug might get
displaced from the protein binding site and attach to the target
site. Of course this is a speculation, but could also be a possibility.
Indranil Bhattacharya
Ph.D candidate
Dept. of Pharmaceutical Sciences
State University of New York at Buffalo
Usa
Back to the Top
An example of highly protein bound drug could Antimalarial drugs such
as Atovaquone which is highly protein bound > 99%. From my
understanding, the efficacy will also depend upon the affinity of the
drug for the target. If the affinity for the target is higher than
the affinity for the protein binding, then the drug might get
displaced from the protein binding site and attach to the target
site. Of course this is a speculation, but could also be a possibility.
Indranil Bhattacharya
Ph.D candidate
Dept. of Pharmaceutical Sciences
State University of New York at Buffalo
Usa
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)