Problem with TSK Super SW3000 column

[Rocnex1]

Hello Everybody,

Can anybody help me with the problem I had with TSK SuperSW3000 column?

I am using the column (with guard column) for monoclonal antibody samples. The samples were in PBS with 0.01% tween. I usually include injections of p-amino benzoic acid to check the performance of the columns. The mobile phase is PBS at pH 7.4. With a new guard column, the # of theoretical plates went down from 35000 to 32000 after 60 injections. And pressure went up from 100 bars to 110 bars. As per Tosha's tech support, I cleaned the columns by reversing the direction and rising with water, 1 M NaCl, water, 20% ACN and water for 5 column volume each. Then equilibrated with MP. But the pressure did not go back to normal and the # of theoretical plates even went down to 25000.

Has anybody had this kind of problem before? Or it is very normal to see this after 60 injections?

Thanks a lot!

[HW Mueller]

Maybe you can find my comments and questions on this column in past discussions, but in short: Here it was also used for mab (separating it from its Fab!), any organic modifyer changed the performance considerably (usually broadening peaks), but this was reversible via flushing with lots of PBS. Unfortunatly, TFA and higher perfluoro acids permanently changed the column in that fashion (without a change in backpressure). I never found out what´s behind this. I can not accept some of this as "normal", maybe if enough people point out such problems the manufacturer will correspond with us more seriously?

[Rocnex1]

Maybe you can find my comments and questions on this column in past discussions, but in short: Here it was also used for mab (separating it from its Fab!), any organic modifyer changed the performance considerably (usually broadening peaks), but this was reversible via flushing with lots of PBS. Unfortunatly, TFA and higher perfluoro acids permanently changed the column in that fashion (without a change in backpressure). I never found out what´s behind this. I can not accept some of this as "normal", maybe if enough people point out such problems the manufacturer will correspond with us more seriously?

Hello HW,

Thank you for your reply. I am not very sure about your comment "TFA and higher perfluoro acids permanently changed the column in that fashion ". I didn't add any organic modifier. The amino benzoic acid was suggested by Tosha to check the perofrmance of the column, I suppose it does not do any harm to the column.

And my problem was not about reproducibility of the separatioin. It was always the case that after that many injections, performance went down significantly. Did you indicate that Tw may have caused the problem?

[tom jupille]

You may have to change your guard columns more often or else pretreat the samples to get rid of the detergent. It sounds like the crud is breaking through the guard to contaminate the analytical column.

[Rocnex1]

Tom,

Thank you for your thought. Is it unusual that a guard column needs to be replaced after 60 injections? sometimes I have more than 60 injections in one sequence, so I have to break the sequence and replace the guard column?

You think the Tween in the samples are really srewing up the column? do you know any way to remove tween from samples? I often have a lot of samles, a simple pretreating method will help a lot.

thanks!

[HW Mueller]

During and after TFA the peaks had widened drastically, after some days with PBS the peaks narrowed a bit, but never again to the sharpness obtained prior to TFA contact.
You said that you washed the column with 20% ACN, and thereafter it was worse? No surprise to me. I don´t remember whether I used ACN, but MeOH and 2-propanol required long washes with PBS to reverse deleterious effects.
If even these alcohols leave a lasting effect than one could expect a detergent to be even more effective.
This was likely mentioned before: A 5µL injection of the mab in the wrong matrix (low ionic strength or an alcohol/H2O mixture) can ruin the resolution. So it would also not be surprising to me that, because of the Tween, you started out with a lower resolution than what this column is able to provide, and then the detergent built up on the column ....? Maybe you washed the column when you changed the guard and thus washed most of the detergent out? Anyway, if your Tween is the problem I wouldn´t expect a guard column to hold back very much of it.
Also, what´s the stationary phase of your guard?

[Rocnex1]

The guard column I uased has the same stationary phase as the TSK 3000 column.

When I replaced the guard column, I did not use extra wash except normal rinsing and equilibrating.

I talked to Toso ( Jack Tally), he suggested me to clean the columns with 1M NaCl at pH 3 to remove hydrophilic contaminant, after the cleaning with 1 M NaCl pH 7 and 20% ACN failed (actually theoretical plates went down after the first cleaning). But this cleaning procedure also made the performance much worse. Now he suggested me to use 8 M Urea as the last resolve. He mentioned that the Tween in the samples really caused problems because it is strongly retained on the column and very difficult to remove.

Can you suggest something to remove Tween from the column? Or do you know any other column which may tolerate Tween better than TSK?

Thanks a lot!

[Uwe Neue]

Several comments:

I would not have worried about the change in plate count from 35000 to 32000. This is nearly within the measurement error.

If this is real, and is of concern to you, the next best thing would have been to replace the guard column, since most junk builds up on the guard column. This is what they are good for...

If the replacement of the guard column after 60 injections is too expensive for you, you cna disconnect the column and wash THE GUARD COLUMN separately. Then you do not run the risk of damaging your analytical column with unnecessary washes.

How quickly a guard column goes belly-up depends on the amount of gunk that it is supposed to pick up. I once managed to kill a guard in a single injection, but my analytical column was still perfectly fine.

[HW Mueller]

Most of the detergent can be removed easily from samples via ultrafilter centrifuge devices based on micro vials: You spin out most of the liquid add PBS, spin again, add PBS, repeat if needed. You may loose some protein, some mab are quite sticky.
Some people participating in this forum raise eyebrows when NaCl is mentioned. Though I have seen no attack by 0.15M NaCl at pH ~7.2 on test stainless I would not let 1M Cl- at pH 3 contact my stainless columns, etc. The urea would be used to remove proteins off the column, if your Mab is highly purified and didn´t tail very badly (initially) where would protein contamination come from? If you have membrane proteins present or other sticky ones they could certainly cause problems.
The only thing I can think of about removing Tween (if that´s causing the problem) is washing the column for a long time (over night or longer) with your best mobile phase, probabply PBS. Since the "rinse" at the time of changing the guard helped a bit not all is lost. I would be very interested in finding out whether you could ever restore the column to its original performance.