We have analyzed 5-HT (Serotonin) in the mice plasma using two different analytical techniques ( ELISA and LC-MS). Surprisingly, the two different analytical techniques report different observations as below.
ELISA: Average of 270 ng/ml
LC-MS: Average of 6000 ng/ml
Does anyone have an idea for the possible cause for the variation?
I appreciate your feedback in advance.
Thanks
Dhaval
Hi,
there are multiple factors that may influence your outcome - the sample preparation, the controls, how the method of analysis was performed and how the data was analysed and normalised. without overviewing what you've exactly done, it's impossible to tell.
D.
These two techniques are completely different in nature. Proper method analysis techniques are needed in each case to provide useful data. I would not report on any comparisons between them as no information regarding the actual methods used were provided. A fundamental understanding of the techniques used and also, as David mentions, a detailed description of the samples, all sample preparation steps, the EXACT methods used, the training of the operator, etc are needed. ELISA is rather simplistic, but there are still many possible variables which effect the results obtained. Compare that technique to something 1000x more complex such as LC-MS and we are not even talking about the same data anymore. A scientifically useful LC-MS requires a great deal of training and knowledge to setup and run (to provide meaningful results). Any LC-MS system will output data (junk, noise, data that looks like sample info, all kinds of info). The data output may or may not have anything to do with the sample under analysis (I see it all the time as a professional consultant). To properly use LC-MS requires that first, proper sample prep is carried out, then a proper HPLC method which follows good chromatography fundamentals are followed. Next, the MS detector must be optimized for the method, the sample, and calibrated. Unfortunately, you should never assume that these steps are followed. The people who ran your sample may have done a great job, but we have no way of knowing based on your question. Serotonin results by either method can not be viewed by just the technique used (we are not talking about placing a solution in a test tube, then passing light through it to make a measurement).
- The answer to your question is: The details in how they were obtained really matter. You provided no details (and can not over the web. Such info would best be obtained on-site) so the differences you observed in the results are related to HOW the two techniques were used. These are not absolute and are subject to wide variation depending on the training of the operator(s) and the specific methods used.
A few months ago I posted the second of two rants* about the damaging consequences of spurious claims, repeated during three decades, that the toxic plant metabolite ouabain is a mammalian hormone. The discussion of analytical selectivity and specificity should be relevant to your situation.
I think the main underlying problem is lack of awareness of the strict critera that apply to the identification and quantitative analysis of small molecules.This might be due in large part to the failure of editors of biomedical journals to require validation according to standard criteria, which have been evolving steadily since the 18th century, if not earlier. It is most important to ensure that initial findings are comprehensively validated because, in biomedical science in particular, they may give rise to a chain of citations by researchers who, quite naturally, are not specialists in analytical science.
If anyone is interested in exploring another example that could be of concern, I'd propose that long-standing claims for extra-pineal sources of melatonin should be subject to a critical review focussed on analytical methodology and its validation.
Finally, returning to the question, I recall that reliably measuring plasma serotonin was at one time considered difficult, due to release from platelets.
* 'Endogenous ouabain (EO)' in mammals: absence of valid experimental evidence
December 2019
DOI: 10.13140/RG.2.2.10144.53765