Which protein do you detect as a loading control? Maybe the antibody for this protein is not working anymore... If possible use a fresh antibody aliquot.

Cheers, Christian

Have you looked at the gel for total proteins to confirm that you are loading equal amounts of protein in each lane or close to it? 

Given that the answer is yes, besides what Christian pointed out, you have to remember that completion of transfer and binding to membrane (neutral or charged) is a function of size of protein and time. Since your target protein is small, it would transfer quickly but if you are using a larger protein as the "house-keeping" (invariant) you may run into the problem of making a reasonable compromise between partial transfer of larger protein or near full transfer of larger protein by increasing the transfer time and running the risk of loosing some of your smaller target.

An alternative is to use a smaller house-keeping protein.

Best... 

Thanks for your replies Christian and Tausif. 

Tausif, I measured the amount of protein before running the gel and the concentrations were around 10ug.

I tried loading control of size 60kd and 37 kd though, and blot for 1 hour at 90 volt, do you think the transfer time is not sufficient?

By the way, I tried different aliquotes of antibody and ensured that there is no problem with antibody. 

No bands found for loading control but found bands for the other two expression constructs. Are you sure you have  followed protocol rightly? It is mostly likely to be a mistake because there is no resonable explanation accounting for this phenomenon. Maybe you have ignored some important things.

Tanmoyita, you can stain the gel for protein with Ponceau Red prior to transfer and empirically establish complete or near complete transfer. Once you have found the correct conditions (pay attention to smaller markers that they are not getting away as you increase the time or voltage), you can optimize antibody-mediated detection for your house-keeping protein(s).

If you do not have sufficient experience in this technique, in the beginning there are some fiddly bits you have to master. Once you have done that, life will be grand!  ;)

"It is mostly likely to be a mistake because there is no resonable explanation accounting for this phenomenon."

That is a matter of how you define "reasonable"!

Let me give you a few illustrative examples:

1) let's say that Tanmoyita has succeeded in overexposing her target protein and there is a quantitative difference sufficient enough in magnitude that her target is well above the threshold of routine detection limits and this allows her an easy detection;

2) her target antibody has high Ka and low Kd;

3) the house-keeping antibody happens to be not high in affinity or avidity

And that is without counting the differential rates of movement of various size proteins during the electrotransfer.

This assumes that detecting antibody is not non-functional. Optimum conditions (fiddly bits) need to be established.  Of course, it doesn't mean that there are no procedural mistakes.  ;)

Tanmoyita, I would like to elaborate on the good suggestions by Christian and Tausif and focus on the possibility that some basic system checks and controls may be needed. If this is your private supply, you can begin by asking the vendor technical support (or supplier if not commercial) about status and any known issues with those lots of control protein and antibody. The supplier(s) will have proper handling, storage and troubleshooting information. However, common use reagents entail greater risks. Are these control reagents themselves controlled if available and used as a common supply item? That is, does somebody track & log temperature/alarm, reagent access/use and product expiry? Is there universal training or awareness for its storage, handling, restocking and applicability? If not, these may have been unknowingly mishandled (e.g., at RT or even on ice too too long, F/T too many times) By the way, its not worth chasing down this hypothetical in an R&D lab; just purchase new reagents and maintain control systems. Therefore, re-establish the integrity and performance of the controls, and confirm their suitability for this setup as mentioned above.

Even I am having the same problem. My protein of interest (32KD) could be successfully blotted and detected, but when I tried to strip the membrane and reprobe with actin (42KD), I could not detect the band

I will rule out that the actin antibody is bad because someone else tried it and she got a perfect detection with the same actin antibody

Just try to blot for a different loading control like actinin or GAPDH

Anukana

Your answer doesn't seem to solve the question. I just mentioned that I used actin

"Even I am having the same problem."

I take it that you weren't supposed to!  ;)

Stripping and reporting is not without its own issues, especially if the purpose of detecting a housekeeping protein is accurate quantitative representation of protein load/lane.

There can be many reasons for "observed" differences in signal, many of them have nothing to do with procedural errors. The dynamic range of detection in this method is far from 3-4 orders of magnitude.

Rather than stripping blots and re-probing them, one can always devise a simultaneous system of detection if chromophores or fluorescent detection is employed. For such a method to work in a 2-step detection, one would need to rely on the primary antibodies that are developed in different species. 

Thanks to all of you for your replies. I appreciate that and will get back again if able to solve the problem.

Tausif Sir, Could you suggest me any immediate technical change that I should do while running the next blot ( I would do the ponceau  staining anyway) ? 

Hi Tanmoyita, Could you share the blot image and protocols for gel and immunoblot (gel crosslinkage, time run, sample loading id./lane, antibody i.d.s, dilutions, detection chemistry and duration, etc.). There is another possibility of non-specific binding. For example and solution, see: Lasne 2015 Methods Mol Biol.1312:277-83. "Double-blotting: a solution to the problem of nonspecific binding of secondary antibodies in immunoblotting procedures."

Hi Bradford,

I ran 10-20% gradient tris-tricine gel from Biorad. I ran for 45 mins to 1 hour at 120 volt, transfer it for 1 hour at 90 volt.

I attached the image of the blot when it was incubated with GAPDH (Rabbit), but as my protein specific antibody is also rabbit origin and I incubated the blot first with that ab, u can see my protein band in this blot also although there is no GAPDH band found in that lane.