Acetonitrile solubility

[EK8]

I have a sample that is soluble in acetonitrile at 30% acetonitrile/H2O. Any more MeCN results in the separation of water from MeCN, forming two distinct layers. I'm assuming this is due to the high concentration of hydrophilic peptides from my sample. The issue is, I can only get good retention with a HILIC column, where my starting gradient is 97% MeCN.

What can I expect to go wrong when my sample mixes with the starting gradient? I have 200uL sample injected with a 200uL sample loop, with a solvent flow rate of 3.5mL/min. I am assuming that when my sample mixes with the solvent, the water (where my sample is soluble in) and MeCN will separate into layers in the piping prior to entering the column, or in the column? How might this affect my chromatography?

I don't have the option of reducing sample concentration as I need to recover enough for bioactivity assays after fractionation.

[itspip]

With 70% aqueous as the sample solvent and a starting gradient of 3% aqueous I would expect that the chromatography would be poor, ie, the sample will mostly be carried with the injection solvent through the column with a good portion of the sample being unretained followed by a smear of the sample across the column resulting in a large hump of a peak.

What is the concentration of the sample?
Is this a prepatory method? It seems like it with 200 uL injected at 3.5 mL/min.
How do you already know that HILIC is the best method for separation?

[EK8]

With 70% aqueous as the sample solvent and a starting gradient of 3% aqueous I would expect that the chromatography would be poor, ie, the sample will mostly be carried with the injection solvent through the column with a good portion of the sample being unretained followed by a smear of the sample across the column resulting in a large hump of a peak.

What is the concentration of the sample?
Is this a prepatory method? It seems like it with 200 uL injected at 3.5 mL/min.
How do you already know that HILIC is the best method for separation?

Thanks for the reply. I was worried that my sample may be smearing across the column as I have not been able to recover bioactivity to the expected degree from any fraction I collect.

The sample is 200mg/mL, which contains my unknown active compound along with a complex mixture of highly hydrophilic metabolites and are small molecules <500Da. Liquid-liquid extraction even with butanol won't be able to pull out the compounds from aqueous phase.

We ran a C18 column, and everything eluted in the first 2 minutes. We get very good retention and separation with HILIC. We ran analytical before scaling up.

I was thinking further sample cleanup maybe using anion exchange (target compound has a negative charge), but need to figure out how to elute. I can't use salts because I don't know how I'd separate from the rest of the matrix, and I can't have high salt concentration since this has to be testing against cells.

Maybe elute using buffers at different pH? It would have to be something volatile I could remove.

[itspip]

You might want to try dialysis as a sample prep first. That could do a good enough job cleaning up the sample before the prep-LC step. You might also want to determine the max concentration in MeCN and try the prep-LC at that concentration but over multiple injections.

[Multidimensional]

EK8: I agree with "itspip" and had the very same thought about using dialysis. It does sound like you may have a very unusual (and inappropriate) method or do not have the required training to set up this method. I would encourage you to consult with a professional full-time chromatographer, not someone in academia for advice on how to proceed. Using any type of HPLC method where the sample is not fully soluble in the mobile phase will always lead to the creation of a scientifically invalid method of analysis (and invalid results). These types if general questions are not appropriate for a web forum (too much info is missing and you need local assistance for these types of issues).

[EK8]

EK8: I agree with "itspip" and had the very same thought about using dialysis. It does sound like you may have a very unusual (and inappropriate) method or do not have the required training to set up this method. I would encourage you to consult with a professional full-time chromatographer, not someone in academia for advice on how to proceed. Using any type of HPLC method where the sample is not fully soluble in the mobile phase will always lead to the creation of a scientifically invalid method of analysis (and invalid results). These types if general questions are not appropriate for a web forum (too much info is missing and you need local assistance for these types of issues).

Thank you for the comment. Unfortunately, the lab I am in has very little funding, and even lacks experts in the field I am working in, leaving me basically to figure everything out on my own.

I have tried dialysis and the smallest membrane I could find is a 100-500Da rated membrane. The target compound diffuses across the membrane, alone with many of the small molecules in the sample matrix. It cleans up my sample, but not too significantly to improve solubility.

[Multidimensional]

EK8 wrote: " leaving me basically to figure everything out on my own."

Unfortunately, even learning the basics of HPLC is not something that one can ":figure out on their own". Not ever. It takes many years of professional training just to develop a 'basic' level of skills, BASIC, not enough to develop methods. The technique is far more complex than most realize and each instrument (and CDS) requires a great deal of new knowledge to be acquired. If your school/business wants to use HPLC, then they need to understand they will need to invest in hiring a full time professional chromatographer from nidustry OR send the work out to a professional analysis lab. They will waste what little money they have (and of course TIME) leaving someone to "figure it out". They clearly do not understand what they are asking you to do (and you do not have the experience to help them). We see many businesses attempt this poor route thinking they are saving money. In all cases, they loose money and time, with no scientifically correct results. It is a lesson best learned early. Delay, and continue to waste time and money, not learning from others.

[EK8]

EK8 wrote: " leaving me basically to figure everything out on my own."

Unfortunately, even learning the basics of HPLC is not something that one can ":figure out on their own". Not ever. It takes many years of professional training just to develop a 'basic' level of skills, BASIC, not enough to develop methods. The technique is far more complex than most realize and each instrument (and CDS) requires a great deal of new knowledge to be acquired. If your school/business wants to use HPLC, then they need to understand they will need to invest in hiring a full time professional chromatographer from nidustry OR send the work out to a professional analysis lab. They will waste what little money they have (and of course TIME) leaving someone to "figure it out". They clearly do not understand what they are asking you to do (and you do not have the experience to help them). We see many businesses attempt this poor route thinking they are saving money. In all cases, they loose money and time, with no scientifically correct results. It is a lesson best learned early. Delay, and continue to waste time and money, not learning from others.

Believe me I have tried to get them to collaborate or contract out this work. They just refuse and are being very unrealistic.

I know the target analyte is soluble in methanol. Would it be worth trying to use Methanol/H2O r as solvents as opposed to MeCn/H2O?

[itspip]

I know the target analyte is soluble in methanol. Would it be worth trying to use Methanol/H2O r as solvents as opposed to MeCn/H2O?

The issue here is that methanol is a protic solvent. Since you are trying HILIC any protic solvent will act as the strong solvent.

I agree with Multidimensional that you're most likely going to need to work with a collaborator/consultant for this method development. Hound the company; either spend the money now and save time, or waste time on troubleshooting and developing a method that may or may not work for what is needed.

[H.Thomas]

You could try to use at-column dilution: https://www.waters.com/content/dam/wate ... 008462.pdf

[Gaetan Glauser]

Hi,

We were in a very similar situation a few years ago. A bioactive unknown compound (for which stucture AND concentration were unknown) which we could only track by bioactivity. It would elute in the solvent peak in reverse-phase and could only be satisfactorily retained with HILIC. It was poorly soluble in ACN so we tested different things and even published a (quite cool in my opinion) paper on at-column dilution for HILIC-based prep HPLC (https://doi.org/10.1016/j.chroma.2017.11.071). We did all kinds of tests on solubility, molecular cut-off filters, SPE sorbents, buffers, and used different HILIC and ion-exchange columns. We purified several fractions and fractions of fractions, performed NMR on active fractions, identified certain compounds, retested these compounds which proved inactive, restarted the entire process with different chemistries etc. But despite years of efforts we never found what the compound actually was.

I agree with previous comments that this is not a job for beginners nor for labs which have few resources. I even believe that you may need several experts from different fields (e.g. analytical chemist, natural product chemist, organic chemist, biologist). Have a clear discussion with your supervisors so they really realize what it takes to identify an unknown bioactive susbtance from a complex biological extract. Sometimes you can be very lucky and get your molecule almost by serendipity, but in most cases you'll struggle a lot, especially if only HILIC works since there are lots of highly concentrated primary metabolites able to interfere.

Good luck!

[EK8]

Hi,

We were in a very similar situation a few years ago. A bioactive unknown compound (for which stucture AND concentration were unknown) which we could only track by bioactivity. It would elute in the solvent peak in reverse-phase and could only be satisfactorily retained with HILIC. It was poorly soluble in ACN so we tested different things and even published a (quite cool in my opinion) paper on at-column dilution for HILIC-based prep HPLC (https://doi.org/10.1016/j.chroma.2017.11.071). We did all kinds of tests on solubility, molecular cut-off filters, SPE sorbents, buffers, and used different HILIC and ion-exchange columns. We purified several fractions and fractions of fractions, performed NMR on active fractions, identified certain compounds, retested these compounds which proved inactive, restarted the entire process with different chemistries etc. But despite years of efforts we never found what the compound actually was.

I agree with previous comments that this is not a job for beginners nor for labs which have few resources. I even believe that you may need several experts from different fields (e.g. analytical chemist, natural product chemist, organic chemist, biologist). Have a clear discussion with your supervisors so they really realize what it takes to identify an unknown bioactive susbtance from a complex biological extract. Sometimes you can be very lucky and get your molecule almost by serendipity, but in most cases you'll struggle a lot, especially if only HILIC works since there are lots of highly concentrated primary metabolites able to interfere.

Good luck!

Hello. Thank you for sharing your experience. I am curious as to what kind of bioactivity were you tracking? I am looking at antibacterial activity. It's disheartening to hear that it was never identified.

Another issue I seem to be having is non-specific inhibition in fractions. I even speedvac my collections multiple times, but for some reason I see activity across multiple fractions that adds up more than the expected total. Did you ever experience this?

[Gaetan Glauser]

I am not an expert in the bioassay part. That was a bioassay based on nematode dormancy. In our case, this was rather an on/off situation (active or inactive) where a small dilution could already shut off activity so we had to carefully evaluate our dilutions before testing.