As a late follower. let me start by asking if you have succeeded in the meanwhile? If not I'm happy to help, in that case I would like to see the sequences of the 4 primers you have used, then I can comment on the procedure with more accuracy.

Himanshu recommends 25-30 bp overlap between the two genes, I agree but usually I am generous and go for 30 or slightly more. I usually do 10 cycles of gel-purified fragment self-assembly without any primers (in a standard PCR mix), and I allow very long annealing and elongation times because the large fragments are not as mobile as short primers. Also, if you can just about see the two gel-purified bands on a gel before you mix them for self-assembly, they will be present in much lower concentration than the standard PCR primers. So you need to give the single stranded fragments time to find each other and assemble and elongate, and you always have the complementary strands as strong competitorswhich is why this step is very inefficient. You cannot really see if it has worked until you do the second step, but all you need is a few full length molecules. The second step is to dilute the assembly reaction and create a new PCR mix, this time with the two outside primers (sense primer for gene 1, antisense primer for gene 2). This is a proper PCR reaction, and it is here where you should see the proper band (sum of the two separate bands minus the small overlap). The programme is standard, shorter annealing times, and elongation time optimised for the length of the larger desired band, and sufficient cycles for amplification. I usually do 25 cycles and test them for water, and a few dilutions of the assembly PCR (10-fold, 100-fold, 1000-fold). If I don't get anything, I do another 10 cycles and check again.

Gibson assembly is an alternative, but you need high quantities of purified overlapping DNA fragments and target vectors for the reaction to work, most Gibson assemblies fail because the purity and concentration of the individual fragments is not high enough.

Good luck :)

What is the extension time? what taq are you using? what is GC content? How much MgCl? What is volume of PCR?

I m using KOD neo Plus tag, extension time is 5min and temperature is 68.

Primers are quite long and their GC content is almost 50% and Mgcl2 is 1mM working in total 50ul reaction mixture

The tag you are using should be sufficient, GC content is in great shape,

Problems that you might want to try out with most likely:

1) extension time: 5 mins may be two short for long genes. never had problem but have heard of it.

2) Try this, 1ul of BSA to tube. It comes at some standard dilution everywhere, dilute 1/10. I dont know why this works, but this has always done the trick

3) MgCl (maybe 1.5mM so you have more MgCl for the clamp to use ), also very unlikely.

4) DNTPs are bad (least likely based on what you said)

Thanks a lot for your kind reply. I will try out the options you suggested.

Sorry,but I could not get the 4th point about DNTPs could you please elaborate.

Thanks

DNTPs go bad overtime from freezing and thawing. You should always have tubes with small batches to avoid this. Lastly I didn't mention this, but check pH of water.

when I did fusion PCR I found that using a lower annealing temperature of about 50 for four cycles and then increasing it to 55-57 for 25 cycles worked very well for me. I also used an extension temperature of about 72.

Thank you so much for the kind reply. is it ok to use higher extension temperature.

i will try this one

thanks

Thanks Chad,

so what pH of water do u recommend.

An important point before designing overlap PCR is the overlap region between the two genes(25-30 bases performs reasonably well).

Apart from all the general PCR precautions of having a good performing PCR mix and enzyme it might be advisable to add equi molar concentrations of both genes. Another trick that you could apply is to allow the partners to anneal and extend for some cycles(usually 15) and then add the FP for gene 1 and RP for gene 2 for another couple of cycles(usually 20).

Thanks Himanshu for your reply...

but without primers how could it be done ... will you please explain, I could not understand...

Given the scenario that you have enough region for overlap between the two genes,these regions should mediate the initial annealing and extensions like a normal PCR (but not exponential in nature)

Please refer the image..it should help.

I would also suggest considering Gibson assembly if you are going to do overlap PCR regularly

now i get it, thanks

so, concentration of dNTPs used should be the same as used in normal PCR mixtures..

thanks

i usually use the same PCR mix for normal as well as overlap PCR.(enzyme,dNTPs primers if needed everything same)

Thanks a lot

Hi Pratima

Try this paper

http://www.biotechniques.com/multimedia/archive/00190/BTN_A_000114017_O_190204a.pdf

Jesper

Hi Jesper,

Thanks for the help. I will surely try this one and Let you know

Thanks

Thanks everyone for your kind Suggestions.

I will try them out and let you know which one works for me...

Thanks a Lot

Regards,

Pratima

Thanks for your kind reply

Hi Himanshu,

I am getting little confused with the PCR program. It should be like-

1. mix without primers-- 94(5m),{98(10s);55 or Tm of my primers which is 60(30s);72(1m/kb} *15cycles ,72(7m),20(hold)--- then I will add primers and

now what would be the program?

do we need very first step in which double DNA separates, I think we already have separated one in our mix. So is it ok after adding primer just use the following program

98(10s),Tm of my primers(30s),72(1m/KB) *20 cycles ; 72(7m), 20(hold)

please suggest

hello Sharda,

I tried your method too, but no results. I am getting the band but not of desired size. Its size is around bZIP gene size not the fused one. :(

As a late follower. let me start by asking if you have succeeded in the meanwhile? If not I'm happy to help, in that case I would like to see the sequences of the 4 primers you have used, then I can comment on the procedure with more accuracy.

Himanshu recommends 25-30 bp overlap between the two genes, I agree but usually I am generous and go for 30 or slightly more. I usually do 10 cycles of gel-purified fragment self-assembly without any primers (in a standard PCR mix), and I allow very long annealing and elongation times because the large fragments are not as mobile as short primers. Also, if you can just about see the two gel-purified bands on a gel before you mix them for self-assembly, they will be present in much lower concentration than the standard PCR primers. So you need to give the single stranded fragments time to find each other and assemble and elongate, and you always have the complementary strands as strong competitorswhich is why this step is very inefficient. You cannot really see if it has worked until you do the second step, but all you need is a few full length molecules. The second step is to dilute the assembly reaction and create a new PCR mix, this time with the two outside primers (sense primer for gene 1, antisense primer for gene 2). This is a proper PCR reaction, and it is here where you should see the proper band (sum of the two separate bands minus the small overlap). The programme is standard, shorter annealing times, and elongation time optimised for the length of the larger desired band, and sufficient cycles for amplification. I usually do 25 cycles and test them for water, and a few dilutions of the assembly PCR (10-fold, 100-fold, 1000-fold). If I don't get anything, I do another 10 cycles and check again.

Gibson assembly is an alternative, but you need high quantities of purified overlapping DNA fragments and target vectors for the reaction to work, most Gibson assemblies fail because the purity and concentration of the individual fragments is not high enough.

Good luck :)

Thanks for your kind Reply,

I am still struggling for the fusion PCR product.

Your suggestion is important for me

my overlapping region is 45bp. Tm of overlaping region is 68.2 and outside primers Tm is F=61.6 and R=71.6. I am using Ex-taq.

I got both the genes separately in almost equal amounts. Now, I just want to confirm did I get ur suggestion in right way:

Now I have to use the reaction mix without primers like

Buffer=5ul

dntps=5ul

mgcl2=1.5-2mM

Both genes templete=100ng

taq=1ul

H2O= up to 50

then divide the mix into 5(10ul each) --> then do the reaction with higher annealing and extension time like

94=5m(denaturation)

94=30s

68=Tm of overlap region for 1m(usaually use =30s)

72=extension 3m (final produt is around 3Kb)

For 10 cycles

Now, for 2nd PCR, use 

10uM Primers=1.5ul each

CDNA=from 1st PCR product without purification (dilute 1/10 or 1/100)

rest reaction mix is same

and PCR program is 

same as above except the annealing  time and Tm (here Tm should be used as of Primers) do it for 15-20 cycles. 

waiting for  your Reply,

Thanks,

regards,

Pratima

Hi again, I do not understand why you divide the mix into 5, do you try 5 different conditions? 3'end hybridization of long DNA fragments is very inefficient, trying out different annealing temperatures can be a good idea. I would try 50, 55, 60, and 65 degrees, and I would anneal for 2 minutes. 10 cycles is fine. Most of the DNA molecules will simply become double stranded again and they do nothing in the reaction. Only a very small number of molecules assemble in the way you expect and become double stranded only at the 45bp overlap, which becomes slowly extended during annealing, and also later during elongation. This is impossible to see because only a very small fraction of those molecules will do this.

For the second reaction use just 1 microlitre PCR mix as template, or 1 microlitre of 10-times and 100-times diluted samples, and do a new PCR in 50 microlitres (new reagents, and this time adding the outside primers) with a normal annealing time of 60 degrees for 30 seconds (close to the lowest annealing temperature of the two primers), then 3 minutes elongation at 72 degrees, 15-20 cycles may not be enough, it could take 25 to 30 cycles to get your product.

You must remember that the vast majority of DNA fragments will simply anneal back to double stranded. The higher the annealing temperature of the first PCR, the more you favour the double strand recovery which competes with the assembly. This makes sense because 45bp overlap is very little compared to 1-2 kb overlap.

So the bottom line is, I guess you have used too high temperature during the assembly PCR, and you may need more cycles during the second PCR (25-30 cycles). Try it out and let me know what you get.

Hello,

Thank you so much for your kind reply.

I am just dividing the master mix without any different condition, so that less volume of PCR mix may work more efficiently (only reason)

Sure, I will try the suggestion you gave and let you know as I will have results.

Thanks a lot