Western blot mystery - possibly a sample preparation problem? (View forum version)


Posted 28 March 2012 - 07:33 PM


I've been trying for a few weeks to run some western blots in my lab, with limited success. I really need to get it working (my PI is getting very annoyed with me), but I'm having a hell of a time figuring out why it isn't working. I'm the only one in my current lab running westerns, so I'd really appreciate some help!

The samples I'm using are from total cell lysates of rat spleens, mouse thymus, and MCF10A human cells. I've been able to detect actin and tubulin, though the detection isn't as high as I would like. That may be due to an old antibody, though. The bigger problem is that I haven't been able to detect Bcl-2 in my samples. It's a smaller protein (~25kDa), which may be part of the problem. Using a positive control mouse spleen sample from Santa Cruz Biotech (where my antibody is from), I've gotten a Bcl-2 signal, so that suggests it has something to do with how my samples are prepared.

I'm preparing my samples according to this: (http://www.biobanks....on 20030807.pdf), briefly: cells are lysed in SDS (basically laemmli buffer without bromophenol) by pipetting and vigorous vortexing, then incubating at 70 degrees for 20 minutes. Quantification of my samples by BioRad RC/DC puts them at about 2mg/mL. I've been loading 20ug per lane (15uL well sizes, but I don't like overfilling them).

An observation that may be related is that the blue line in the running gel runs very quickly through to the bottom, and gets very dispersed. The bottom few markers on my protein ladder are also smearing. I've tried using gels from 8% to 12%, and varying the voltage from 125 to 200, and nothing seems to have an effect. I've ordered some pre-cast 4-20% gels to see if that helps.

Does anyone have any ideas what may be going wrong? I've been frustrated with this for far too long, this *should* be a real simple, quick assay and I should have been done with it weeks ago. I think my PI is going to be very annoyed if I don't have any data for him before he leaves for AACR this weekend!



Posted 28 March 2012 - 07:45 PM

Add some protease inhibitors to your lysis buffer - a lot will be needed for the spleens, they are full of proteases.

Try boiling your samples for 5 min instead of 70 deg C for 20 min.


Posted 28 March 2012 - 08:16 PM

Perhaps you can try loading more than 20 ug of protein? If you could lyse your cells with smaller lysis buffer volume, that should concentrate your protein and you can keep using low loading volume. You also didn't mention what %of gel did you use the first time around? If you use higher concentration gel (may be 12 to 15%), you could slow down the run, and then may be you could also try stopping the run before your loading marker ran out of gel.


Posted 28 March 2012 - 10:03 PM

Thanks for the suggestions bob. I forgot to mention that I do boil my samples for 10 minutes just before loading them. Also, I have used a protease inhibitor cocktail to see if it would make a difference, and no luck. Staining with coomassie doesn't look any difference with or without inhibitors.

Lysing in 2% SDS at 70 degrees *should* kill proteases (of course, nothing is guaranteed).

Zienpiggie, in lieu of better ideas, I was going to try to get a higher concentration like you mention. In my most recent try, I used 12% acrylamide, ran at 130V, and stopped it before the ladder was halfway across the gel. The bromophenol blue still dispersed and was barely visible within 20 minutes, but a 25kDa protein couldn't have migrated off the gel.


Posted 29 March 2012 - 06:39 AM

25 kDa is a small protein and could have migrated off. your ladder can tell you if it did.

more likely, however, is that the protein passed through the membrane during transfer.

what are your transfer conditions (membrane, buffer, time, current and/or voltage, etc)?


Posted 29 March 2012 - 08:20 AM

My ladder's 10kDa band is still well within the gel. I've been using BioRad's wet transfer system, 100V for 1hour, chilled with an ice pack. I use nitrocellulose with 0.2um pore size. Transfer buffer contains 20% methanol.


Posted 29 March 2012 - 09:48 AM

When you commassie your gel, can you see distinct protein bands on your samples around the 25 kda? Also, may be you could ponceau your membrane to make sure you have a decent transfer. I also don't fully understand what you mean by the bromophenol blue becomes barely visible within 20 minutes?

SCBT Tech Support

Posted 29 March 2012 - 10:31 AM


Greetings from Santa Cruz.

I see you are having trouble using your Bcl-2 antibody for WB. Did you already contact Technical Support here at Santa Cruz for suggestions? We would be glad to help you try to help you get this Western Blot to work for you. If you want to contact us, we would be glad to offer suggestions to your protocol for this antibody!

Please feel free to give us a call or send us an email and we will be glad to help you. You can reach us in several ways:
Toll Free: 1.800.457.3801 ext. 2
Live chat: www.scbt.com
Email: scbt@scbt.com

Good luck in your research!


Santa Cruz Technical Support Team

Santa Cruz Biotechnology, www.scbt.com
Email: scbt@scbt.com
Phone: +1-800-457-3801 ext. 2
Fax: +1-831-457-6013


Posted 29 March 2012 - 10:53 AM

Zienpiggie, I see distinct bands on the coomassie stain gel, but almost exclusively at 50kDa or above. There's one band at around 30kDa that's smudgy and has variable migration in each lane. The bromophenol blue disperses rapidly as it migrates through the separating gel, so in the positive control sample it's hard to see it. In my samples it's easier to see, but still dispersed and generally lower than it should be.

SCBT, I have contacted tech support, and I was sent a replacement antibody, which didn't solve the problem. I may give you another call.


Posted 30 March 2012 - 08:36 AM

the bromphenol blue band becoming diffuse (then sharpens by the end of the run, if you run far enough) may indicate that your sds is old and decomposing. you may want to try to run the gel with a fresh lot of sds.

also, you may want to use a gradient gel (to at least 15%).


Posted 30 March 2012 - 02:48 PM

Also by any chance, is your gel assembly tight? or if they are made fresh? If they are dried out, the gels shrink and may become detached from the plate then your samples could be seeping out of the gel instead of running down. Nevertheless, it sounds like you are not getting much protein in the 25 kda to begin with. I think increasing the protein amount to load with fresh SDS probably the next thing to try.


Posted 06 April 2012 - 01:38 PM

Okay, I found the problem. It was simply that the gels I was making weren't working with my samples. Pre-cast gels work fine, and I get nice, clear, bright bands where Bcl-2 is supposed to be.

Odd thing is, my ladder worked even with the gels I was making. And using fresh SDS didn't help (ALL my reagents for casting gels were purchased within the last 2 months, except for the tris). I'm really not sure why my gels weren't working, since I've made hundreds of gels before in other labs and they've always worked fine. But I also don't really give a damn, I found something that works. Posted Image

Now I've got a contamination issue in my organ extracts... but hell, that's a way better problem to have than bands not showing up when they should, so I'm still happy.

Thanks for the help everyone!