Dear all,
I am wondering why i am getting only less proteins identified from a total protein lysate separated on a 1D gel. Did the in-gel digestion by cutting individual lanes in to 24 bands, followed by LCMS analysis using Agilent Q-TOF and Mascot search. We have even used Agilent SpectrumMill for the protein identification. But still getting only few proteins identified (Less than 1000 proteins). From the literature what I came across to see is that using the proteomic strategies several groups have identified and reported more than 2000 proteins. So where am I missing the rest of protein information? Can anyone suggest me on how to improve the protein identification? Does it has to do anything with the sample preparation/LCMS parameters/Mascot-SpectrumMIll search parameters. Please help me in resolving this problem.
Thanks in advance,
Murali
Hi,
It is quite difficult to troubleshoot your findings (even though I don't feel around 1000 proteins identified is a problem..), but there are a few things that you could evaluate:
- What are your expectations regarding the number of proteins based on? Are you analyzing the same sample (prepared by the same sample prep) as used in the article you're referring to?
- Does your gel suggest that there are 2000 proteins on it? How did you stain it, Coomassie or silver? Did you fix the gel before staining, did you destain it? These things can influence the success of in gel digestion and thus of protein identification;
- What do your base peak chromatograms look like? Did you use a steep gradient for the LC-MS analysis? What are your MS/MS settings: how many fragmentation spectra per cycle were you acquiring, did you use dynamic exclusion?
- Did you reduce and alkylate your proteins prior to digestion? What protease did you use for digestion? Are your search settings correct, for example with regard to mass accuracy, fixed and variable modifications?
- When searching your results, did you combine all 24 files into one search file, or did you consider all 24 files separately?
Hope this helps getting you in the right direction, good luck!
Hi,
besides what Eef asked, It would be very helpful to state what is the proteome of interest and what organism it is. btw - 1000 is not a few. sounds right for a qtof...
Hi Eef and David,
Thank you very much for your suggestions.
Its a mammalian cell line that I am using and extracted only the intracellular proteins. Didn't use any reagents that will aid in extracting the membrane proteins also.
My gel didn't' suggest that there should be 2000 proteins, since I have seen lot of articles that have used different mammalian cell lines for global proteome profiling. Did the staining of the gel using coomassie and prior to trypsin digestion we have detained the gel. But gel fixing as such we haven't done.
And regarding the LCMS part we have used a step gradient- 45%B in 90 minutes(we standardized this using standard BSA digest from Agilent and got good coverage too with 500fmol (71% sequence coverage using mascot). All the steps prior to trypsin digestion what you are mentioning is performed and the error tolerance was set as 10 and 20 ppm for MS and MS/MS during mascot search. And finally did combined all the 24 files to single file and analyzed using Proteome Scaffold.
Was curious to know whether 600 is a good number for a global proteome profiling !!!
well, if you're looking at soluble proteins only, then that's most probably the reason you're not getting beyond the 1000 protein mark. obviously, accommodating for modifications and glycosylations may change that number a little bit, as well as allowing for experimental artifacts (such as acrylamide binding to lysines).
Hello Muralidharan,
I have encountered the samle problem with my samples from some company people where they reported just some 600 proteins and some people have reported over 2000 proteins in the same sample. Even I got around 2000 proteins in similar samples before so I requested them and they told to re-run the sample as something was wrong with spectra obtained. I think for getting whole proteins you need to run your sample 2-3 times and then combine the results to get whole protein profile. I am waiting for my results, will update you further if I come to knw about any other important reason behind this. But m damn sure that there are more number of proteins which youstill need to detect.
All the best, hope it helps !
Hi Ajay,
thank you soo much for sharing the experience.
Regards,
M
- Badges
- Science topic
- Similar topics
- Biological Science
- Molecular Biology
- Biomolecules
- Proteins