Hi Rohail, It might be the primary antibodies. Typically WBs denature proteins meaning that for a positive signal you need a primary Ab that is able to recognise the denatured form of your protein. Typically in ELISAs the capture and detection Abs recognise fully formed protein (an may not be able to bind denatured protein in a WB scenario). 

I would try a primary Ab specifically designed for WB against your protein of interest. 

Hi Rohail, It might be the primary antibodies. Typically WBs denature proteins meaning that for a positive signal you need a primary Ab that is able to recognise the denatured form of your protein. Typically in ELISAs the capture and detection Abs recognise fully formed protein (an may not be able to bind denatured protein in a WB scenario). 

I would try a primary Ab specifically designed for WB against your protein of interest. 

Dear Rohail,

All your intend mitochondrial proteins are single chain? if they have more than one subunit, during WB procedure they may be dissociated and you can not see them in WB but in ELISA they will find and make signal. But if all the four proteins are single chain, I think it may be related to your capture Ab in WB. Because in WB you encounter denatured proteins but in ELISA you will assay more native form of the protein.

On the other hand ELISA has more false positive results. So if you have not false positive in the ELISA method, I suggest you changing your capture Ab in WB.

Best

Mehdi

Agree with above answers. I think, first of all you should find out where is the error? I have few suggestions.

1. Are you sure the positivity of ELISA? You can repeat the ELISA with known positive and known negative with number of time to check it.

2. Normally we run a SDS Gel parallel to the WB as a control. That means, run two gels with same samples at same time. One can be used for WB and other can be stained with SDS staining (eg: Coomassie brilliant blue) solution. After WB has been done, we can compare the two gels, whether the bands are presence in WB or not.

In your case, first of all you should confirm the presence of 4 enzymes in SDS gel.

3. If the 1 and 2 are correct (that means, if the samples are positive and they give strong enough bands in SDS) the problem is almost with the WB.

4. As James and Mehdi answers, it may be due to the problem of binding primary antibody with your interest protein.  In WB, enzyme is denaturing which can or cannot be recognized by primary antibody. You can use different specific primary Ab and compare the results. 

5. Or you can use other techniques for identifying your protein.

I think point 2 of Bhagya will provide you the necessary information to determine your next move.

Tip: Make sure you do not remove the stacking gel when you do the coomassie staining. It might be that the boiling step / denaturing by SDS, causes your protein to self-aggregate and so it does not move into your gel. If this is the case, you will find a blue mark in your stacking.

Thanks for your helpful answers ,Hi,James D Porter,Mehdi Hedayati, Bhagya Deepachandi,Wesley Jongbloed.... yes i know that on western blotting protein lost its conformation but why other two proteins are positive on western blotting?yes sir all of my protein in single polypeptide chain and its original conformation(But every protein has several domains).I do not think that my result is false positive because i repeat ELISA 3 times and compared my results with positive and negative control.yes Deepachandi i have checked every parameter that you mentioned.

Occasionally, if you could obtain one protein in a reasonable amount and other one in low level you get + signal on elisa but encounter to recognize on less sensitive system. see this provision

I agree with the above comments and would add:

Have you considered running a non denaturing gel? If you detect all 4 proteins by this method, it would confirm that denaturing was the issue.

I agree with all the comments above, but it could be as simple as being one of the very few proteins that just don't blot.

I think there is no problem about sensitivity and protein amount , because ELISA can detect all 4 proteins, but the main problem is related to 1) protein epitope changing in WB in comparison to ELISA and 2) Antibody specificity, so it is better repeat the WB with another capture antibodies.

Also, who is the antibody manufacturer? I've had WB's that wouldn't show bands for months... until I bought a new primary antibody from a different company, even though they had the same epitope. 

Yes, I agree with Terra, and I told as a problem solving, changing the capture antibody, which is apply with changing producer company.

Of course. do not forget, washing step if not suitable in WB, may be elute week bond capture antibody, So check it with Ab producer or use lower ionic strength, lower concentration of detergent too.

best

Check the pore size of the membrane u use for blotting and try blotting it at slightly lower voltage. There are chances that the protein is small enough to pass through the membrane and so u never get them on blot.  For eg, for a 16 kDa protein I use 0.2 µm pore size of the membrane with 80 V for one hour while others in my lab use 0.45 µm pore size with 100 V for 1 h for proteins with higher molecular weight.

Are you using the same clone of anitbody for those two missing proteins in both WB and ELISA? If you they are the same, u need to make sure they are suitable for WB detection and that they are polyclonal. 

Thanks for your helpful answers,Thanks terra and mehdi,okay i am going to check all of my proteins on native gel,my primary antibody is the patient serum.i will tryi to use fresh serum.yes i am using same clone of antibody for those two missing proteins in both WB and ELISA

It is probably due to:

1- the antibodies used in the ELISA system are conformational antibody. In normal SDS-PAGE, the conformational epitope will be lost hence explaining no signal on western. Try Native PAGE instead.

2- if you still get nothing, the reason might be that the protein concentration is too low for Western Blot. In such a case, you can do a pull-down experiment by attaching your antibody to dynabeads and incubate your serum/plasma overnight followed by a centrifugation.

good luck!

Thanks for your answers