low recovery / protein loss in the lower range..

[alicee!]

hi al..
this is a general observation for most chromatographic methods i have worked with.. i wish to know if this is usuallly observed.
the expected response/recovery is not obtained in low range solutions when compared to the response for 100% solution..

say, from a 100% solution response.. if i go down to 10%, 1% and further 0.1%.. the response shows a loss in dilutional recovery..
meaning if 100% gives an area value of 100.. the 99% shows 99.. the 2% shows 1.8 (and not 2), 1% shows 0.6 (and not 1) and further 0.1% shows 0.05 (and not 0.1). This loss is observed to increase when the amounts go to the lower range.. starting say 4%.

is this usually observed for chromatographic methods.. why? and are there any solutions to this?
further.. is this considered to be impacting the accuracy?

is this usual for say any specific method like for size exclusion, ion exchange or reverse phase hplc.. or is it unusual?

[tom jupille]

It's not unusual for proteins at all, for reasons that may have nothing to do with the chromatography. Proteins like to stick to silica surfaces. I remember having to silanize glassware (vials, etc.) in order to minimize adsorption. If you are losing a small, constant amount of protein to adsorption onto glassware (or the injection loop, or the stationary phase support, or anywhere else), the results will be just what you are seeing (a larger percentage loss at lower levels).

The reason I'm focusing on the glassware as the culprit is that I would expect any active sites in the system or column to get "passivated" with protein fairly quickly, whereas you use a fresh vial for every sample.

[alicee!]

so.. the major loss would be at the glass ware u mean..? we quite a few times have to use used glassware.. wud it reach saturation at any point?

would there be immediate adsorption.. cos my first few samples also show this loss.. injected in say 5 mins?
would this be constant then?

also.. i have always thought this is an on column loss..

and how does this impact accuracy please?

[tom jupille]

so.. the major loss would be at the glass ware u mean..? we quite a few times have to use used glassware.. wud it reach saturation at any point?

Obviously, I don't know for sure what's going on with *your* samples. I do know from past experience that some proteins can adsorb to glassware. And if you wash the glassware, the adsorbed protein comes off. And if you wash the glassware with hot water and detergent, you can actually etch the glass surface so that it becomes even more adsorptive!

would there be immediate adsorption.. cos my first few samples also show this loss.. injected in say 5 mins?

Yes, it's fast.

would this be constant then?

My experience has been that it's a small, constant loss. In essence, it's an offset in your calibration.

also.. i have always thought this is an on column loss..

That can (and does) also occur. The point I was trying to make was that it's not necessarily a column problem.

and how does this impact accuracy please?

As indicated above, it's an offset (all of your areas are lower than they should be by a constant amount). The absolute error is the same for all samples, so the percentage error is larger for lower levels.

[alicee!]

does anyone have a similar experience..
waht are the usual sample load amounts for say an RP column.. what are the lowest levels that you go for impurity detection.
do you get the expected area response at the lower amounts?

i'd like to correct or compensate for this..

n would like to know how to validate accuracy in such cases.. for impurities..

thanks.

[grzesiek]

i haven't experienced loss fo recovery, at least not in the range that is useful
so say with RP-HPLC i have no problem with the range 0.05% to 100% and lower

i don't have too much experience with the kind of problem you have but if your response funcion has systematic error (this loss of recovery) you only need to explain why it is so - that's what i would do
your response funcion then is not forced zero type, as you would like it to be from what i see

[HW Mueller]

If one knows why one has bad recovery the solution to this problem is usually not far away.

[alicee!]

i haven't experienced loss fo recovery, at least not in the range that is useful
so say with RP-HPLC i have no problem with the range 0.05% to 100% and lower

i don't have too much experience with the kind of problem you have but if your response funcion has systematic error (this loss of recovery) you only need to explain why it is so - that's what i would do
your response funcion then is not forced zero type, as you would like it to be from what i see

could you please tell me what are used as the 100% loading amounts generally for an RP method..
i have a 150*2.1 mm column, 3 micron, 200 Angstrom.. column..
i do a 15 mcg (50 mcl injection of a 0.3mg/ml solution). For the 1% solution, i use a diluted solution and inject 50 mcl)

you think these are appropriate to my conditions?

yes my y intercept is a high negative value, not forced zero. I m not correcting it by forcing through zero.
thanks.

[alicee!]

so.. the major loss would be at the glass ware u mean..? we quite a few times have to use used glassware.. wud it reach saturation at any point?

Obviously, I don't know for sure what's going on with *your* samples. I do know from past experience that some proteins can adsorb to glassware. And if you wash the glassware, the adsorbed protein comes off. And if you wash the glassware with hot water and detergent, you can actually etch the glass surface so that it becomes even more adsorptive!

would there be immediate adsorption.. cos my first few samples also show this loss.. injected in say 5 mins?

Yes, it's fast.

would this be constant then?

My experience has been that it's a small, constant loss. In essence, it's an offset in your calibration.

also.. i have always thought this is an on column loss..

That can (and does) also occur. The point I was trying to make was that it's not necessarily a column problem.

and how does this impact accuracy please?

As indicated above, it's an offset (all of your areas are lower than they should be by a constant amount). The absolute error is the same for all samples, so the percentage error is larger for lower levels.

i have a few answers to the above..
i m using silinized glassware and hence there would be least protein adsorption.
im mostly using standard vials and systems.. which are considered appropriate for suchanalysis. i see similar behavior on two different brand systems..

[alicee!]

If one knows why one has bad recovery the solution to this problem is usually not far away.

i suspect this is an on-column loss..
low recovery/elution..
but i donot observe (the remaining protein) elution in the next blnak run i.e. a blank aafter a sample injection which shows a low recovery/response.

this i observe with my RP, SEC and IEC methods..
the area response at lower amounts is lower than expected at 100% region.
but the response at these low amounts is linear when a curve of 0.2-2% is evaluated.
thanks.

[grzesiek]

could you please tell me what are used as the 100% loading amounts generally for an RP method..
i have a 150*2.1 mm column, 3 micron, 200 Angstrom.. column..
i do a 15 mcg (50 mcl injection of a 0.3mg/ml solution). For the 1% solution, i use a diluted solution and inject 50 mcl)

you think these are appropriate to my conditions?

yes my y intercept is a high negative value, not forced zero. I m not correcting it by forcing through zero.
thanks.

well I ususally do inject sample at concentration that gives me absorbance of the main peak at around 1.0 -1.5, after that(1.5 or 2) the detector response is not linear (or at least it is not expected to be)

with the real data seen I think that people will have more ideas to help you, please give as the numbers - conc/area and maybe one chromatogram of 100% sample and 1%

[HW Mueller]

You have a problem which you should fix chromatographically/chemically. If you can not solve the problem then you should at least prove consistency, but I have no experience with this, maybe others do.
I remember that I once had a recovery problem across all important concentrations of the analyte. I was in the position to say that the analysis was an estimate within a certain range, and just worked with that.

[alicee!]

could you please tell me what are used as the 100% loading amounts generally for an RP method..
i have a 150*2.1 mm column, 3 micron, 200 Angstrom.. column..
i do a 15 mcg (50 mcl injection of a 0.3mg/ml solution). For the 1% solution, i use a diluted solution and inject 50 mcl)

you think these are appropriate to my conditions?

yes my y intercept is a high negative value, not forced zero. I m not correcting it by forcing through zero.
thanks.

well I ususally do inject sample at concentration that gives me absorbance of the main peak at around 1.0 -1.5, after that(1.5 or 2) the detector response is not linear (or at least it is not expected to be)

with the real data seen I think that people will have more ideas to help you, please give as the numbers - conc/area and maybe one chromatogram of 100% sample and 1%

i would like to know what your 100% sample amount and area response is.. and what is the lowest % (0.05%) that you can reliably detect and what is the area value for this point.
thanks.
PS: will try to send the chromatograms

[alicee!]

You have a problem which you should fix chromatographically/chemically. If you can not solve the problem then you should at least prove consistency, but I have no experience with this, maybe others do.
I remember that I once had a recovery problem across all important concentrations of the analyte. I was in the position to say that the analysis was an estimate within a certain range, and just worked with that.

when you have no recovery problems, what are the protein amounts and ranges you are talking about please?
thanks.

[grzesiek]

amount depends on chromophores (I don't know if spelling's good) area is usually few millions for 100%
for smaller concentration few hundreds is usually good