Back to the Top
Dear all,
Good morning! We studied the plasma protein binding (PPB) of a test drug using ultrafiltration method (Amicon Ultra-0.5 centrifugal filter devices, 10KD, 0.5mL). We added 0.4mL human, rat, or dog plasma to the device. After a centrifugation (6000g, 5min), the filtrate was about 0.1mL. We determined the concentration of drug in the filter, and the PPB was calcuted according to the following: PPB=1-(concentration of drug in filter / added concentration in plasma).
Surprisely, the calculated PPB in rat plasma was 6.4%-16.9% at three different concentrations, PPB in dog plasma was -24.4%-(-55%), and PPB in human plasma was -33.3%-(-38.9)%.
I don't know why the PPB was caculated to be negative values? In addition, we also added 0.4mL spiked phosphate buffered solution sample to the device. After the centrifugation, the concentration of test drug in the filtrate was also higher than the spiked concentration.
The analytical method was full validated, and the analytical process was qualified.
Best wishes,
Yuanting Zheng
Back to the Top
Dear Yuanting,
I suspect there may be possibility of nonspecific binding with the device as other parameters you have already cross checked. There are certain categories of molecule which binds to the device and thus gives aberrant results. Nonspecific binding can be reduced by using the pretreated device with surfactant. Depending upon the chemical nature of your molecule you can use the types of surfactant( anionic, cationic or nonionic etc). The pretreatment of the device with these surfactant will reduce nonspecific binding. If this also doesn't work then equillibrium dialysis can be opted.
thanks,
with regards,
--
Rashmi Shiju
Back to the Top
The following message was posted to: PharmPK
watch out for a possible chance for nonspecific binding or the
stability of the drug in plasma
Back to the Top
Dear Sir,
Have you used LC-MS/MS for analysis of PPB samples?.
If yes then it may be due to matrix effect as you are comparing donor (spiked plasma samples) with receiver (filtrate which is devoid of plasma protein).
Hope this will help.
Regards,
Rahul Vats
Back to the Top
The following message was posted to: PharmPK
Dear Yuanting Zheng,
We have used the same (Amicon Ultra-0.5 centrifugal filter devices, 10KD, 0.5mL).
My question is , if you're getting same negative values for QC? ,
We are calculating % protein bound ={ 1- (Concentration in filterate / concentration in plasma after centrigugation)}
As if you calculate with intial plasma Concentration , It may show negative values.
Total drug = Concentartion in filterate + Con in plasma after centrifugation
If your calculating as you have written (PPB=1-(concentration of drug in filter / added concentration in plasma)).it may give negative result ..
Dear All,
Correct me if iam wrong..as iam new to this assay.
Regards,
Indira
Back to the Top
The following message was posted to: PharmPK
Dear All,
Iam sorry for my message, May be iam just got confused with the formula for a min..
Dear Yuanting Zheng,
Iam sorry.
Reagrds,
indira
Back to the Top
The following message was posted to: PharmPK
Amicon devices are notorious for non-specific binding. I suggest using Harvard 96 well plates (equilibrium dialysis) which have shown not have similar non-specific binding. You should be doing mass balance calculations. Stanley Cotler
Back to the Top
The following message was posted to: PharmPK
Hi,
I would think non-specific binding (very likely) and poor solubility (possible) caused the problem. Total drug concentration is what you added to plasma. I don't understand the reason to use the concentration in plasma after centrifugation.
XS
The following message was posted to: PharmPK
Hi,
I feel that it is not a non specific binding issue, but may be a stability and/or solubility issue. You information tells me that after centrifugation, your samples'conc. in filtrates increased either the drug dissolevd in buffer or plasma. So you initial samples were sitting there much longer than the filtrate, correct. If yes, it maybe a stability issue. If I were you, I would do the drug incubation in both buffer and palsma for 30 or 60 min to check the stability.
XS
Back to the Top
Hello Yuanting,
The negative protein binding value might be something to do with your donor solution. Have you determined the concentration of the donor solution in the cup before or after filtration? Once I happened to measure caffeine protein binding in human plasma (obtained from a local blood bank) using ultrafiltration technique. The filtrate concentration was higher than the nominal donor concentration we used, and consequently a negative protein binding value was obtained. Later we found out that the blank human plasma already contained significant amount of caffeine, thus the filtrate concentration was higher than the nominal donor concentration since caffeine proteing binding was not that high. Best wishes,
KJ
Back to the Top
Dear all,
Thank you for all your suggestions!
I have determined that there were no test drugs in the blank plasma samples.
The test drug was a derivative of amino acid, so I think its solubility was well in both plasma and PBS.
I think that both the non-specific binding and the poor stability may reduce the drug concentrations in the filtrates, but my promblem is the concentrations in the filtrates was higher than the added concentrations. In addition, I have study the drug stability in rat plasma for 8 hours at room temperature, and no obvious degration was observed.
We determined the drug concentration using a LC/MS/MS method. To avoid of matrix effects, the standard curves was preprared with the filtrates of rat, human, and dog plasmas,respectively. However, because we need large amount of filtrates, we use the Amicon devices-5mL instead of Amicon-0.5mL. I don't know if these two devices can be so different.
I'm going to do this experiment using equillibrium dialysis method, but I really want to know the reason for the aberrant results.
Thank you all!
Yuanting
Back to the Top
Dear Yuanting,
This is an interesting observation. I have seen so many responses to your question, and was wondering if an approach considering all responses is judicious.
First reason that immediately strikes the mind would be non-specific binding, however is it really the one!!!! We first need to understand that the numbers in rat plasma were POSITIVE. So where is non-specific binding hidden wrt rat plasma.
I also saw a response indicating solubility issue. Well, I here believe that the PPB was tested at low, medium and high concentrations, which to some extent would throw some light on the likelihood. However, if you had only tested at one concentration, and if the compound precipiates (I am not sure if this is likely, because you would expect reasonable solubility in plasma, else the compound credibility for development is at question), I would recommend testing at a low concentration. The way we should interpret the solubility issue is, if the drug precipitates in plasma, it would essentially be filtered down and result in higher concentration in the filtrate. This would have resulted in higher value in numerator, resulting in negative protein binding.
What concerns me most is that the value in rat plasma is positive. Is there something to do with the matrix. The only instance you would see positive value in rat plasma and negative in other matrices is when there is a background niose for rat plasma hidden under the compound peak resulting in overestimation of concentration in plasma, however I still have reservation in this interpretation (the only reason I was thinking in these lines is because I am trying to overweigh two matrices to one).
Although my mail is not conclusive, I would consider ultracentrifugation or dialysis method for this compound,
Hope this helps,
Regards, Jagannath
Back to the Top
Dear all
I had also experienced these kind of issues regarding protein binding experiments but from bio analytical perspective.
1. Whether your Internal standard concentration in the batch is consistent throughout or there is any enhancement or reduction in response in LCMSMS.
2. The diluent used for preparing the Calibration curve is also plays a major role. Eg Methanol may cause huge variation while spiking, since there is possibility of two different scientist spiked the methanol containing analyte can give variations too. 3. Did your CC higher concentration prepared from the same stock of solution used for spiking in protein binding experiment.
4. Is there any possible contamination of micropipette tips used for the assay or the micropipette used for spiking both CC and protein binding samples is same or calibrated to show equal volume of dispense. I also had faced problem with calibration failure of pipette used.
5. I also like to ensure while taking the sample are you actually taking the sample from the centre of the solution filtrate.
6. There is also possibility matrix effect which plays a major role. Just ensure during your LCMSMS analysis by employing transition 184.0/184.0 for possible elution of phospholipids interference at this paticluar retention time.
I had listed out possibilities, I hope it helps.
Thank you
Karthikeyan.k
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 "Plasma protein binding was calculated to be negative" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)