Immunofluorescence using polyclonal antibody - (Apr/21/2007 )
I'm tring to localize a predicted cytosolic protein in human fibroblasts by using anti-rabbit polyclonal antibody. This is the protocol I followed:
1. Fixation: either 4% Formalin or 100% methanol tried, seems no difference
2. Washing with PBS 3 times
3. Permalization: 0.1% Triton in PBS
4. Washing with PBS 3 times
5. Blocking: 5% BSA in PBS for 30min
6. Primary antibody dilution in 1% BSA: 1:10 for 1 and a half hours
7. Washing with PBS 3 times
8. Secondary antibody dilution in 1% BSA: donkey anti-rabbit FITC 1:100 for 1 hour
9. Washing: PBS
10.Antifade solution and microscopy
However, I detected a high background of green fluorescence almost all around the cell body which might overshadow my target protein. There is no difference between using 8 weeks polyclonal antibody and pre-bleed serum (serum before giving immunogen, supposingly no antibody) of the same rabbit. On the other hand, westernblot could actually produce potent bands of my target protein especially after agonist stimulation (over 10 times darker band) while there was almost nothing on the blot if replacing primary antibody with pre-bleed.
I think the fixation and permalization shouldn't have much problem as the FITC can go into the cells. Blocking might have problem. Should I use human sera for blocking Fc receptors of the cells? (Supposingly no expression of Fc receptors on fibroblasts!) Or I should use donkey sera for blocking? (same specise as 2nd antibody) Or the antibody dilution too low to make a clean background? Should I try 1:100 or more diluted?
Based on the phenomena that antibody and pre-bleed showed the same pattern of fluorescence, I believe non-specific bindings are dominant, overshadowing my target. It seems the primary antibodies are not mediated in this assay or the common factors between 8 week antibody and pre-bleed are dominantly effective. Any experienced fellow can give some comments and suggestions? Thanks a lot!
1. Fixation: either 4% Formalin or 100% methanol tried, seems no difference
2. Washing with PBS 3 times
3. Permalization: 0.1% Triton in PBS
4. Washing with PBS 3 times
5. Blocking: 5% BSA in PBS for 30min
6. Primary antibody dilution in 1% BSA: 1:10 for 1 and a half hours
7. Washing with PBS 3 times
8. Secondary antibody dilution in 1% BSA: donkey anti-rabbit FITC 1:100 for 1 hour
9. Washing: PBS
10.Antifade solution and microscopy
However, I detected a high background of green fluorescence almost all around the cell body which might overshadow my target protein. There is no difference between using 8 weeks polyclonal antibody and pre-bleed serum (serum before giving immunogen, supposingly no antibody) of the same rabbit. On the other hand, westernblot could actually produce potent bands of my target protein especially after agonist stimulation (over 10 times darker band) while there was almost nothing on the blot if replacing primary antibody with pre-bleed.
I think the fixation and permalization shouldn't have much problem as the FITC can go into the cells. Blocking might have problem. Should I use human sera for blocking Fc receptors of the cells? (Supposingly no expression of Fc receptors on fibroblasts!) Or I should use donkey sera for blocking? (same specise as 2nd antibody) Or the antibody dilution too low to make a clean background? Should I try 1:100 or more diluted?
Based on the phenomena that antibody and pre-bleed showed the same pattern of fluorescence, I believe non-specific bindings are dominant, overshadowing my target. It seems the primary antibodies are not mediated in this assay or the common factors between 8 week antibody and pre-bleed are dominantly effective. Any experienced fellow can give some comments and suggestions? Thanks a lot!
Hi,
In IF I have never used blocking (as in don't have your step 4 and 5, but I use a slightly different method) and usually don't have problem with background, but here go:
- It sounds like you are testing your Ab. If that is the case, better do a serial dilution. Since with 1:10 you got much background, then dilute 1:100, 1:200, 1:500, 1:1000 and more if possible. (The dilution factor could be deduced from your WB. How good is your WB?)
- If you want to do blocking: according to Jackson, for blocking, normal serum (5%v/v) from the host species of the 2nd Ab is recommended (for yours, donkey serum)
- Have you done other IF before this? How are those? The washing step (step 2) may not be clean enough. For methanol, you have to wash until buffer is no longer got pushed out of the coverslip (till you can wet it easily). For formalin also, wash as much as you can.
1. Fixation: either 4% Formalin or 100% methanol tried, seems no difference
2. Washing with PBS 3 times
3. Permalization: 0.1% Triton in PBS
4. Washing with PBS 3 times
5. Blocking: 5% BSA in PBS for 30min
6. Primary antibody dilution in 1% BSA: 1:10 for 1 and a half hours
7. Washing with PBS 3 times
8. Secondary antibody dilution in 1% BSA: donkey anti-rabbit FITC 1:100 for 1 hour
9. Washing: PBS
10.Antifade solution and microscopy
However, I detected a high background of green fluorescence almost all around the cell body which might overshadow my target protein. There is no difference between using 8 weeks polyclonal antibody and pre-bleed serum (serum before giving immunogen, supposingly no antibody) of the same rabbit. On the other hand, westernblot could actually produce potent bands of my target protein especially after agonist stimulation (over 10 times darker band) while there was almost nothing on the blot if replacing primary antibody with pre-bleed.
I think the fixation and permalization shouldn't have much problem as the FITC can go into the cells. Blocking might have problem. Should I use human sera for blocking Fc receptors of the cells? (Supposingly no expression of Fc receptors on fibroblasts!) Or I should use donkey sera for blocking? (same specise as 2nd antibody) Or the antibody dilution too low to make a clean background? Should I try 1:100 or more diluted?
Based on the phenomena that antibody and pre-bleed showed the same pattern of fluorescence, I believe non-specific bindings are dominant, overshadowing my target. It seems the primary antibodies are not mediated in this assay or the common factors between 8 week antibody and pre-bleed are dominantly effective. Any experienced fellow can give some comments and suggestions? Thanks a lot!
definitely try different Ab concentrations but also we use goat or horse serum rather than BSA or FBS for blocking...although I can't remember why and I never had any major problems with background. Hope this helps
Our lab normally don't use BSA. We use FBS and in some more meticulous cases, they add goat serum too. The main reason is because BSA and FDB have bovine IgG, which could cause cross reaction with 2nd antibody and thus highbackground. But for most IF, we don't have problem with background, although we don't do blocking step. I think you most likely got problem due to too much Ab. When you tested by WB, the Ab is very specific? Can be block by your antigen completely? Very strong or very weak?
Thanks for the comments and suggestions from you all. It's the first time I do IF, no experience before. 
Has anyone of you used unpurified polyclonal antibody serum to do IF? I think this might be the major concern. It's not so specific as I could detect quite a few bands in western. My target should have dramatic increase after agonist stimulation while the other non-specific bands don't change in my western blot, however, I never observed any difference in IF between control and treatment. Another thing I don't understand is about my irrelavant antibody control. What I used was the pre-bleed from the same rabbit (we generate this antibody from invitrogen, pre-bleed is the serum of the same rabbit before giving the immunogen, supposingly the best control). It's very clean in my western blot with nothing on the film, neither specific nor non-specific bands shown up. While when I used it in the IF, no difference from antibody one. All are the same between control and treatments, between pre-bleed and antibody. While when I directly gave 2ndary antibody right after blocking, nothing showed up, no fluorescence at all, indicating my blocking was effective, right?
I'm a bit confused about why we block during IF (Already tried my own serum and BSA for blocking 
). Are we worried that there are a lot of Fc receptors present at cell surface which might cross bind the 2ndary antibody? (I used fibroblasts, supposingly no Fc receptors, even have, human and rabbit shouldn't have much in common, I assume)
What's in my mind now is there is something present in both pre-bleed and antibody sera which couldn't be detected by western blot, while it dominates the IF reaction, and overshadows my target. If it is true, nothing more can be done before I generate a monoclonal antibody. But anyway, I will try a few different dilutions and donkey sera for blocking as suggested. Thank you all! 
Has anyone of you used unpurified polyclonal antibody serum to do IF? I think this might be the major concern. It's not so specific as I could detect quite a few bands in western. My target should have dramatic increase after agonist stimulation while the other non-specific bands don't change in my western blot, however, I never observed any difference in IF between control and treatment. Another thing I don't understand is about my irrelavant antibody control. What I used was the pre-bleed from the same rabbit (we generate this antibody from invitrogen, pre-bleed is the serum of the same rabbit before giving the immunogen, supposingly the best control). It's very clean in my western blot with nothing on the film, neither specific nor non-specific bands shown up. While when I used it in the IF, no difference from antibody one. All are the same between control and treatments, between pre-bleed and antibody. While when I directly gave 2ndary antibody right after blocking, nothing showed up, no fluorescence at all, indicating my blocking was effective, right?
I'm a bit confused about why we block during IF (Already tried my own serum and BSA for blocking
). Are we worried that there are a lot of Fc receptors present at cell surface which might cross bind the 2ndary antibody? (I used fibroblasts, supposingly no Fc receptors, even have, human and rabbit shouldn't have much in common, I assume) What's in my mind now is there is something present in both pre-bleed and antibody sera which couldn't be detected by western blot, while it dominates the IF reaction, and overshadows my target. If it is true, nothing more can be done before I generate a monoclonal antibody. But anyway, I will try a few different dilutions and donkey sera for blocking as suggested. Thank you all!
If it is the first time you are doing IF, I think that it may be a good idea to have a positive control, i.e., doing IF using another Ab that the lab has used before and know that it work well for IF, got a nice morphology, e.g. a Golgi marker, endosome marker..., something bright and strong and clear. make sure that the problem is really due to your Ab, you know. Then, if you say that prebleeding serum gave you the same problem, I really think that it is the background. Try washing more. Rinse cells before fixation and wash well after fixing, too. I truly don't think that the problem is due to block in this case, but better be safe than be sorry. Get a bit more used to the technique before trying shortcut, I suppose. Good luck.