You can have effects of forskolin other than stimulation of adenylyl cyclase that can stimulate PKA by an indirerct pathway. Try using deoxiforskolin or dideoxiforskolin and see if you have the same effect .

Dear Erik

Please try searching the Pubchem database and search for forskolin or any of the ones thenRoberto mentions. It will given you a list of targets- levels of regulation, etc. here is an example for forskolin http://pubchem.ncbi.nlm.nih.gov/assay/assay.cgi?cid=47936

Good luck!

Dear Erik,

I am trying to show that adenylyl cyclase activates CREB phosphorylation in HBEC and am having a lot of troubles with WB. I used forskolin as a positive control and got several times an higher amount of p-CREB in cells treated with forskolin compared with controls. On the other hand I intoxicated the cells with adenylyl cyclase but very rarely I observed an increase in p-CREB in the cells treated with this toxin. I also keep getting a weak signal and the presence of the signal in the negative control which is not distinguishable from the treated cells with adenylyl cyclase. My goal is that of showing that cAMP activates phosphorylation of CREB and thus PKA signalling pathway which would regulate a gene I am interested in. I am convinced that my problem is in the lysates, I think that the cells become activated when passing from culture to lyasates and present always a baseline CREB phosphorylation and since the efficiency of my nuclear protein fractionation is very low I cannot get above the threshold and do not see any difference between controls and treated cells. I know that you are experienced with p-CREB western blot and would like to ask if you can suggest a good working protocol for lysing the epithelial cells without activating them. Hope that my experience is of some help and look forward to your suggestion. Thank you.

Dear Anna,

thanks for your post! I may be able to assist in your work if I understand better what you do. What I understand is that HBEC are epithelial cells. Are they primary cells or an immortalized cell line? Do they need serum in their medium or is it serum-free? Regarding forskolin you may use too low or too high amounts. In my current cell lines (both from the brain) there is an IC50 for forskolin around 1 micromolar, if I use more than 10, 30 or 100µM the pCRE level are lower than at 1µM. This is partly due to the DMSO in which I solve it. What works best for me is to treat the cells with the appropriate amount of forskolin (if necessary do a dose-response first) for the appropriate time (may vary from minutes to an hour), discard the medium an put 2x LDS-PAGE sample buffer (I use the NuPage system from Invitrogen) directly upon the cells, lyze them in the sonifier, heat for 10 minutes at 70 degrees Celsius, shock-cool them for a minute on ice (or in a -20o C fridge), centrifuge them at Eppendorf tube max speed 13-15000 rpm and then put the samples on the gel etc.

Another point may be the pCREB antibody. I recommend the CellSignaling 87G3 Rabbit monoclonal at 1:500-1:1000. Secondary antibody depends on what detection system you use.

Hope this helps, if you have further question, go ahead and write me!

Best wishes,

Erik

Dear Erik,

Thank you very much for your helpful answer.

I use Human Bronchial Epithelial cells (immortalised cell lines of various types: Calu 3, A549 and VA10). I am trying to show that the adenylyl cyclase expressed by Bordetella pertussis is activating the PKA signalling pathway which in turn would upregulate MUC5AC expression. I used forskolin as a positive control in a concentration of 20 microg/mL treating the cells for 30 minutes. I also dissolve forskolin in DMSO and I had good results with forskolin, meaning that I got over phosphorylation of p-CREB but I could not see the same effect when treating the cells with bacterial adenylyl cyclase (3, 1 and 0.3 micro/mL). The medium I use for Calu 3 cells is 1:1 DMEM and F12 with 10% Fetal calf serum. Initially, I was detecting a band at 100kDa which I think was caused by the serum in the medium. After that I improved my method and have been able to get rid of that. In order to avoid cell activation, which was evident because of the p-CREB signal in the negative control (TUC, solvent for adenylyl cyclase), I started lysing the cells directly in the plate without scraping or trypsinising them before lysis but I still am getting p-CREB signal in the negative control. I use p-CREB antibody CellSignaling mouse1B6 (D= 1000), I tried with Milk and BSA and with PVDF and Nitrocellulose mebranes, the differences were minimal. I tried lysing the cells using a lysis buffer which is homemade but similar to RIPA, then I changed my protocol and tried nuclear fractionation using an hypotonic buffer with NP40, centrifuging to separate the cytosolic fraction and then lysing the nuclei with a buffer containing 0.1% SDS and 0.5% sodium deoxycholate which should be enough to avoid sonication. In the second case I obtained less background but still signal in my negative control and very weak signal in the samples treated with bacterial adenylyl cyclase. I am going to try lysing the cells with Laemmli buffer directly in the plate and see what happens. I would put 5X Laemmli buffer directly on the cells on ice, then scrape them and boil for 5 minutes. Should I also shock cool them after boiling and centrifuge? Do you think the problem is in the concentration of adenylyl cyclase used? Thank you so much again. I follow you very closely and you have been of great help so many times. Thank you!

Dear Anna,

so there is something that raises pCREB levels in your controls. Can it be that there is naturally a high cAMP/pCREB level in these cell lines? If so, this should be suppressable by treatment with PKA antagonists like Rp-8-Br-cAMPS (see Gjertsen et al., JBC, 1995 and some my old papers). If Rp-8-Br-cAMPS (1mM !) Pretreatment does not lower pCREB levels, CREB is phosphorylated by another protein kinase than PKA. There are some, e.g. Calcium-Calmodulin-dependent kinase, pP90rsk, p38.

All the above, however, does not solve the question why the bacterial adenylate cyclases have no effect. I suggest to contact Prof. Dr. Roland Seifert at the Hannover Medical School for this question. Dr. Seifert works with adenylate cyclases and their pharmacology. He may have an idea.

Sorry, that I have no more immediate ideas. Best wishes to you and all the best for your next experiments!

Erik