[wptmdoorn]

Hi folks,

A small part within a project I have been running is about detecting creatinine in urine. We have used a reversed phase HPLC method and succesfully created a method to detect it. The resulting chromatograms are seen below (red is the creatinine stock solution/blue is the extracted urine monster)



The conditions were as follows:

Extraction: centrifugating for 800g / 5min, then dilute the centrifugated urine 1:10 with mobile phase
HPLC Conditions:

• Column: C18
• Mobile phase: 95% phosphate buffer (pH 3,2) / 5% acetonitril)
• Flow: 1,0 ml/min
• Detection: UV (236 nm)

As you can see in the urine there is an additional peak at around the two minutes. Would anyone have an idea what it is? Or maybe a method/way to identify this method?

Thanks in advance.

[Arkcon]

If someone is an expert on RP-HPLC analysis of urine extracts, they may weigh in with an opinion shortly, or eventually.  Your sample prep seems a little atypical to me -- although it can just be a simpler prep that your purposes allows.  Often, there is some sort of selective extraction with a solvent to remove things that aren't of interest and concentrate things that are..  So you can look into those sorts o methods.  Now, you've got a monster peak in the physiological sample -- all you know is that its more non-polar.  If you could extract the spectrum from a PDA detector, we would have a hint.  If you could run it on an LC-MS, we'd have a better hint, or maybe the answer.  But as it stands now, I don't have a clue.

Another thing you can try is spiking some standard into your sample before you begin sample processing , to see if the creatine is changing .  But that seems very unlikely given your simple sample prep.  Still, its good advice to do that when you want to make a comparison.

[wptmdoorn]

If someone is an expert on RP-HPLC analysis of urine extracts, they may weigh in with an opinion shortly, or eventually.  Your sample prep seems a little atypical to me -- although it can just be a simpler prep that your purposes allows.  Often, there is some sort of selective extraction with a solvent to remove things that aren't of interest and concentrate things that are..  So you can look into those sorts o methods.  Now, you've got a monster peak in the physiological sample -- all you know is that its more non-polar.  If you could extract the spectrum from a PDA detector, we would have a hint.  If you could run it on an LC-MS, we'd have a better hint, or maybe the answer.  But as it stands now, I don't have a clue.

Another thing you can try is spiking some standard into your sample before you begin sample processing , to see if the creatine is changing .  But that seems very unlikely given your simple sample prep.  Still, its good advice to do that when you want to make a comparison.

The extraction was a little atypical for me too. However this had been used in the literature and worked fine for us too. I was thinking of some simple liquid-liquid extraction to remove atleast some of the interrupting compounds. I have however no idea how to setup a L/L-extraction where I can deny the other compound from being in the other phase as the creatinine itself (since it is a bit more nonpolar like you stated, but how to seperate those).

We spiked the urine samples, and the first peak actually changed - so I can surely state that the first peak will be creatinine as the retention times with the stock of creatinine are also the same. Any sort of MS will not be possible within the project, there is however a DAD detector available. Apart from this; in the end we are detecting creatinine so it is not the end of the world if it cannot be resolved which compound actually elutes as secondary.

[Arkcon]

Its possible the second peak is a mixture of compounds, which is another thing that the Diode Array Detector can help with.  The only thing that worries me is the the second peak is or contains creatine metabolites, which you may also want to quantitate.  Or maybe not.