Back to the Top
Dear Group,
I have a problem in reproducibility of our assay to estimate the drug
in liver tissue. We are using artemisinin as internal standard. The
response of drug as well as internal standard is not consistent ,
even in calibration and QC samples. We have checked and ensured that
this is not a problem due to instrument, chromatography or negligent
analysis.
Brief description of method for your information:
Liver tissue is homoginised in methanol. Sample, standard and QC were
subjected to LLE. (many different solvents, and pH conditions tried).
Dried residue reconstituted in 50% Acetonitrile water and injected to
synergy fusion RP column and eluted with ammonium acetate and
acetonitrile gradient mobile phase. Appropriate MRMs used for MSMS
detection.
Method works very well with neat samples and plasma samples.
My question is:
Why liver tissue extraction is not robust?
Why liver tissue behaves so differently than plasma?
Is it possible that drug and IS are binding to liver tissue and not
getting extracted easily?
Are there any general tips to be followed for tissue sample preparation?
Has anyone estimated artemisini in liver tissue?
my sincere thanks in advance for the input.
Regards,
vinayak
Back to the Top
The following message was posted to: PharmPK
Vinayak,
Why are you using methanol instead of a buffer for homogenizing the
liver? There may be some metabolism of the analyte and / internal
standard upon incubation in liver homogenate, but you have to mimimize
it using different approaches and validate the extraction procedure
with respect to metabolic loss of both the analytes.
Back to the Top
The following message was posted to: PharmPK
Hi Vinayak,
We did a study long time ago showing that artemisinin binds to
hemoglobin in red blood cells. The binding is irreversible and probably
leads to the destruction of the molecule. I would think that the same
happens when liver tissues are involved. US Svensson (later on US
Simonsson) has done several studies involving liver microsomes and
tissues and a search on PubMed will give you some general information,
although I don't think they extracted the artemisinin from liver
cells/tissues.
Toufigh Gordi
Back to the Top
Dear Satya,
We used methanol for homoginisation in our earlier studies and was
working well. Due to the presence of methanol, metabolism should not
be happening as all enzymes will be inactivated. However we have
monitored metabolite also. I feel there is no metabolism. In any
case, this wont explain inconsistency in response.
I will be pleased to know if you have any other input.
Thanks
Back to the Top
Dear Vinayak,
I believe that the source of your variability is most likely in the
extraction step - are you performing the extraction long enough - for
example 1 hour on a shaker ? What extraction solvents do you use ? I
think homogenizing in methanol isn't such a good idea, because
methanol is very likely miscible with your organic extraction
solvent. Instead, for tissue homogenizing I'd rather suggest using
simple TRIS buffer with optional enzyme inhibitor cocktail or
addition of salts (ammonium or zinc sulphate).
I also think that you are getting a lot of heavy matrix effects in
your MSMS because of your methanol use in the homogenization
(solubilization of membrane phospholipids...which you're not going to
loose during the extraction). Try setting another MS chromatogram
trace in positive mode m/z 184-->184 or 522->184. Matrix effects
often cause pronounced variability in response. Try multiple
injections of a same sample.
Have you tried injecting a small volume of the methanolic homogenate
(after centrifugation to eliminate the precipitated protein) ? Doing
so, you might identify whether your variability leads from
homogenization or extraction procedure.
All the best,
Jurij Trontelj
Faculty of pharmacy
University in Ljubljana
Slovenia
Back to the Top
The following message was posted to: PharmPK
Back up a little. Try adding all materials to mobile phase, then to a
surrogate extract(never exposed to tissue). If these agree within
10-15%,
extract the tissue and add materials to the actual extract (previously
extracted). If there is still agreement, add material to the
extraction mix
with tissue and perform the extraction. You can add a duration of
incubation to push the system. This should demonstrate-not give you a
"feel" as to whether there is metabolic activity toward methanol, your
analyte or internal standard.
Ed O'Connor, Ph.D.
Laboratory Director
Matrix BioAnalytical Laboratories
25 Science Park at Yale
New Haven, CT 06511
Web: www.matrixbioanalytical.com
Email: eoconnor.aaa.matrixbioanalytical.com
Back to the Top
[Our campus network is being 'upgraded'. As part of that process the
boomer.org server IP number will change in the next day or so (only,
I hope) but there may be some disruption of access during this time -
db]
The following message was posted to: PharmPK
Hi Vinayak,
As Toufigh said, it is known that artemisinin binds irreversibly to
RBC and something similar may be happening in case of liver. But i
dont know any mechanism (apart from metabolism) for its so called
destruction.
But if this is the case, then binding of a drug to liver tissue and
its destruction in you case is questionable, as you are using methanol
homogenised liver tissue ( I suppose you are still in method
validation process and homogenising the liver with methanol prior to
spiking of drug). Are you using the homogenate supernatant or the
suspension for preparation of calibration standards and QC?
The only point in my previous thread was that by using methanol
homogenised liver tissue for validation, you may loose some
information on stability of analyte in liver matrix (this information
may be critical when you analyse unknown samples).
Is this problem only in case of internal standard (artemisinin) or the
analyte also? Which analyte you are measuring in liver?
Back to the Top
Dea Satya and Toufigh,
Thank you for your discussion. Yes I homoginise in methanol prior to
spiking of analytes. This instability issue is there with anlytes
too. I am monitoring the expected metabolite also along with analyte.
Since I have metabolite standard, I spike that too in the matrix as
one of the analyte. I tried to homoginize the liver in PBS and
spiked the analytes and internal standard. I still have the same
problem. Is there any possiblity of some enzymes being active even on
homoginisation in methanol? (That may explain formation of a
different metabolite, other than the expected one, which I am already
monitoring)
Irreversible binding is one way to look at it. But how do we prove it?
Thanks again for this discussion.
Regards,
Vinayak
Back to the Top
The following message was posted to: PharmPK
Vinayak,
As Ed O'Connor said, it would be important for you first to
demonstrate (and not give you a "feel") as to whether there is
metabolic activity toward methanol, your
analyte or internal standard.
Also if you dont mind it would be interesting to try other solvents
like acetonitrile, TCA in place of methanol and do repeated
extractions (just a ray of hope for better results!!!!!).
As long as irreversible binding issue is concerned, you may want to
digest the extract using proteases and amino-peptidases. Following
reference may be useful.
Journal of Pharmacology and Experimental therapeutics Vol 219 (1)
page 207-212.
Back to the Top
The following message was posted to: PharmPK
I have seen notes about using acetonitrile instead of methanol for
homogenization step for metabolism ID work. Maybe try acetonitrile?
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Artemisinin in liver tissue" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)