Mobility of salts on unbonded bare silica HILIC column

Question

I have two chemical compounds that I put on cells, to see if they can diffuse into the cytosol. The compounds are similar in structure, but contain a key modification that alter the polarity completely. After the cells are incubated, they are washed with PBS several times to remove excess compound. PBS is used not to rupture the cells prematurely. The goal here is just to see compound that actually went into the cell, not something that was there from the beginning in the stimulation phase.

The two compounds I test are different in polarity. The first one has good retention on RP18 and therefore residual PBS salt is not a problem to sperate off and divert into waste before the analysis of the actual compound.

The second compound is so polar that it requires a hilic phase for proper retention. The manufacturer describes the hilic phase as bare unbounded silica.

I analyse in LCMS the cell lysates after protein precipitation and centrifugation. The composition of this sample is 2 parts methanol and 1 part PBS. When I inject the sample that contains the lysate which was incubated with the polar compound, I am wondering if the salts from the PBS and the lysate are mobile on the silica if enough water is present? In my gradient I go up from 10 percent water in ACN to 100 percent water both with 10 mM NH4Fa in 15 min, staying for 4 min at 100 % (NH4Fa itself is recommended by the manufacturer which itself is a salt) I am asking this question, because from normal phase separation from purification in chemical synthesis we know, that salts stick very strongly to silica and maybe never come off it. However, in an analytical context maybe the bare silica has properties, that change this behavior?

I am also wondering if anyone would have experience with the retention time behaviour of PBS on hilic phases (so, that I can divert it as well). My goal is obviously to not let not volatile salts clog the ESI source of the instrument and make sure that the column is not loading up immobile salt deposition, while I can circumvent ion suppression of my compound by PBS. Any experience on this is highly appreciated.

Answer 1

The OP asks about phosphate-buffered saline solution and its potential retention on bare silica column under the gradient conditions provided as follows:

10 percent water in ACN to 100 percent water both with 10 mM NH4Fa in 15 min, staying at 4 min at 100 %

Yes, your concern LC-MS is valid in the sense that even inorganic ions can retain on bare silica in hydrophilic interaction liquid chromatography mode. And we do not want phosphate ions to reach the mass spectrometer. Now the term bare silica is deceptive, even if the manufacturer calls it bare silica. All "silicas" are different, in the HPLC sense due to their physicochemical properties.

See this paper for example: Takayama, Nobuyuki, Lee Wah Lim, and Toyohide Takeuchi. "Retention behavior of inorganic anions in hydrophilic interaction chromatography." Analytical Sciences 33.5 (2017): 619-623.

Even here most inorganic ions showed retention at higher ACN concentration using bonded phases. For example, in this study they show the effect of typical UV-active inorganic ions (phosphate excluded because it does not absorb UV).

Another issue is you start with 90% ACN in the beginning. This can virtually precipitate phosphate ion (this is a close range where phosphate becomes insoluble in ACN). In short, if possible switch to a simpler UV-method for HILIC in order to use your analyte X with phosphonate groups. Alternatively, find by alternative detection methods, where your analyte elutes, and use that window that goes the MS, the rest should can go to the waste.

Your gradient is too wide. I will not recommend going to 100% water. HILIC mode is notoriously slow to equilibrate and your retention times will vary here and there. Try 90-to-60 ACN at the most.