Jump to content

  • Log in with Facebook Log in with Twitter Log in with Windows Live Log In with Google      Sign In   
  • Create Account

Submit your paper to J Biol Methods today!
Photo
- - - - -

Western Blots on Membrane Proteins


  • Please log in to reply
9 replies to this topic

#1 jmafs

jmafs

    member

  • Active Members
  • Pip
  • 7 posts
0
Neutral

Posted 11 November 2010 - 01:15 PM

I have been running westerns on prostate cancer lysates searching for a 130kDa transmembrane protein, Integrin beta 1. Despite my continuous efforts, I have been unable to find my band of interest. For the most part I see many non-specific bands at low intensity but nothing at 130kDa, where I expect my protein.

I have been able to visualize GAPDH loading control perfectly, which makes me think that my procedure is not completely wrong. I have introduced protease inhibitors to my samples, and keep them on ice at all times to avoid protein degradation. I have beta-mercaptoethanol and SDS in my loading buffer. I have tried varying boiling times for about 5-10 min at 95C. I have tried incubating the primary antibody in blocking buffer, without blocking buffer. I have changed blocking buffers as well from BSA to Normal Goat serum, also filtering this solution. Also, since it is a rather large protein, I have only 10% methanol in my transfer buffer.

I have tried using multiple cell lines, and different lysates to ensure that my samples were not bad. I have also tried different primary antibodies, with the same results.

I am really at a loss of where to go next. All of my exposures end up looking like weak signal in a laddering sort of fashion with non-specific binding, and nothing where I expect the band. Do any of you think this has to do with the protein being a transmembrane protein? Is there any way you treat them differently?

If you have any ideas or suggestions I would VERY MUCH appreciate them!

#2 bob1

bob1

    Thelymitra pulchella

  • Global Moderators
  • PipPipPipPipPipPipPipPipPipPip
  • 5,830 posts
414
Excellent

Posted 11 November 2010 - 04:41 PM

What are your transfer conditions (type (e.g. wet, semidry), Volts/Amps, time, buffer)?

One other transmembrane protein (CFTR) that I can think of doesn't do well with boiling, I think it is denatured at 70 deg C for 15 minutes in denaturing buffer.

How is your antibody stored, and how old is it? Do you have a positive control?

#3 jmafs

jmafs

    member

  • Active Members
  • Pip
  • 7 posts
0
Neutral

Posted 11 November 2010 - 04:57 PM

I have been doing a wet transfer at 90V for 1 hour. I use a standard transfer buffer consisting of 48 mM Tris, 39 mM glycine, 0.1% SDS, 10% methanol. I upped the SDS and lowered the methanol content to cater to a protein >100kDa.

I have tried not boiling the samples as well but that gave similar results, but I could possibly try a less aggressive temperature like 70C. By denaturing buffer do you mean the electrophoresis loading buffer?

My antibody is only about 3 mos old, stored at 4C, per manufacturer instructions. I also attempted a different antibody that was brand new, which gave the same pattern on my blots. I think the next step would be the positive control lysate that the manufacturer (Santa Cruz) recommends.

What about playing with the reducing agent in my loading buffer. Any thoughts on the exclusion of that and if it could help?

Thanks so much!

#4 ksturm

ksturm

    member

  • Active Members
  • Pip
  • 13 posts
1
Neutral

Posted 11 November 2010 - 11:52 PM

Boiling in 95deg is maybe to harsh. Try to put your sample with loading buffer in 60deg water bath for 30min. Cool down in ice water, centrifuge and load supernatant for electrophoresis. Boiling in 95deg is usual method, but often cause protein fragmentation. Your unspecific bands due probably to protein fragments. Another possible explanation for these bands could be unspecific binding of secondary ab. You can test for it by leaving out the primary. However I recommend to use the positive control as you mentioned.

#5 jmafs

jmafs

    member

  • Active Members
  • Pip
  • 7 posts
0
Neutral

Posted 12 November 2010 - 06:29 AM

I have tried not boiling the samples at all (straight from the freezer to the gel after defrosting) and got the same pattern as when I boiled at 95C. I'm not really sure how that adds up...I have not attempted to incubate at 60C though. What kind of centrifugation speed are we looking at here, 10,000xg? Your idea for a run without the primary is a good one, and I tried it with no signal, so I believe that my secondary ab is not the issue here. Hopefully after trying the positive control things will start to make more sense.

#6 madrius1

madrius1

    Veteran

  • Active Members
  • PipPipPipPipPipPipPipPipPipPip
  • 114 posts
0
Neutral

Posted 12 November 2010 - 09:33 AM

Unfortunately, Santa Cruz sell a lot of crappy Abs. Sometimes though, their antibody are very strong and work well for quite a long time, with the advantage of coming in 1ml format. Are all your abs from Santa Cruz? Could you refer to any publication looking at this protein and use the same antibody?

As you explained, it really seems to me like the problem is your primary antibody. I routinely blot 250 kda proteins and boil them for 5 mins at 95 with absolutely no degradation.

#7 jmafs

jmafs

    member

  • Active Members
  • Pip
  • 7 posts
0
Neutral

Posted 12 November 2010 - 10:12 AM

Yes, the Ab is from Santa Cruz. I have a GAPDH antibody from them that works fine, and I also tried buying an antibody against integrin beta 1 from AbCam, with the same results. Most of the publications that use my antibody are for immunoprecipitation, and the ones that use WB don't really provide much detail to their methods. In my independent research I have found that many groups trying to perform a WB on integrin beta 1 use antibodies from independent labs rather than buy them commercially.

I think that the boiling question is less about the size of the protein, but the fact that it is membrane bound. This is becoming a very puzzling dilemma....

Does anyone have an opinion on if my reducing agent (I've tried DTT and BME) is having an effect in all of this?

Thanks for the input!

#8 madrius1

madrius1

    Veteran

  • Active Members
  • PipPipPipPipPipPipPipPipPipPip
  • 114 posts
0
Neutral

Posted 12 November 2010 - 11:56 AM

Since you have no more reducing agents when you probe with the antibody, changing the concentration during migration will not change anything, unless the reduction somehow destroys the epitope targeted by your ab. But it seems very unlikely.

And since you say that a lot of people do their WB with homemade antibodies, I think you have found the problem ^^

The ab from Santa Cruz can probably be optimized to give some results. But you could also try to get a sample antibody from another lab, or do your own. I would prefer the second option.

Good luck!

#9 mdfenko

mdfenko

    an elder

  • Active Members
  • PipPipPipPipPipPipPipPipPipPip
  • 2,811 posts
136
Excellent

Posted 12 November 2010 - 01:30 PM

antibodies that are good for immunoprecipitation aren't necessarily good for immunoblotting (native vs denatured protein). you should look at the data sheet that came with your antibody and see if it is recommended for immunoblotting.

you may need to find another antibody that is recommended for immunoblotting.
talent does what it can
genius does what it must
i do what i get paid to do

#10 NicoleValenz

NicoleValenz

    member

  • Active Members
  • Pip
  • 5 posts
0
Neutral

Posted 04 January 2011 - 06:01 PM

Hi,
I had similar troubles until a labmate told me some membrane-bound proteins are better detected using TCA to extract the lysate and DON'T SPIN DOWN! You'll have to sonicate the lysate to make it clear before you run it, but changing to 10% TCA made all the difference for me.
Boiling, DTT/b-mercaptoethanol, etc. are important only if you're concerned about the state of the protein that the antibody recognizes--native, reduced. If you're spinning down, then you're losing the membrane fraction and this could account for your low signal.

Edited by NicoleValenz, 04 January 2011 - 06:04 PM.





Home - About - Terms of Service - Privacy - Contact Us

©1999-2013 Protocol Online, All rights reserved.