I don't understand. How do you use gel extraction? At what step in the process does what you're talking about come in?
I have a limited understanding of the process. In my mind, these are the steps:
-PCR
-Run Gel + Cut out bands
-Purify Bands (Wedgie Prep)
-Clone
-MiniPrep
Sounds like you're already gel extracting (Wedgie Prep). I don't know what a Wedgie Prep is (mostly used Qiagen gel prep kits) but the point is you are running your product on a gel, cutting out only the band you want, and purifying from there. The other option is, if you are certain your PCR reaction is optimal (high yield and no side products), you can directly purify, cut, and clone - it saves you half an hour to run the gel but is really suboptimal.
We did that for another subcloning we've trying to get to work, and it did the trick. We've actually already tried multiple temp ranges. Didn't do a thing.
Are you using a program to predict your ideal annealing temp, or are you mucking about blindly? It's much faster and cheaper to run a simulation
in silico than
in PCR maquinam. We used a program called MacVector that did a great job (anytime I checked the prediction using a temperature gradient, it invariably predicted the best temp for optimizing yield vs side products); but if you don't have MacVector there are some tools available on the Web. I just found
http://www.promega.com/biomath/calc11.htm
by Googling "calculate annealing temperature" but you should check around and see what else is out there. Btw sometimes a degree or two difference in the annealing temp makes a huge difference in the product yield, so if you try a number of temps that are all 2 degrees apart you may miss your optimal point. That's another reason why you need a prediction program. Even if it isn't the world's best prediction program, it will give you a general idea, and then you can try 10 reactions that are all 0.2 degrees apart.
Yes, we PCR out of a cDNA library. And, no, we do not get one sharp band. Our gels are actually really poor.
Whoa there li'l buddy,
you need to optimize your PCR before you do anything else. That's why your subcloning isn't working. The reason you're cloning beta-amyloid and whatever else is because that's what you're amplifying, along with six thousand other genes probably.
Where's your postdoc? What's his deal? Anyways, there is a ton of literature on how to optimize PCRs. Your lab must have a book on basic biochem techniques; check it for ways to optimize PCRs. Frankly though, I've never found the little tricks like adding DMSO or whatever to be very useful; it's really mostly about the specificity of your primers, the purity of your sample, and your annealing temperature (in that order).
Sounds to me like your problem is probably your primers. You need very specific primers to amplify out of cDNA, as you have lots of other very similar sequences in your starting sample. To amplify from cDNA your primers should be at least 18 bp long (not counting add-ons like restriction sites). If there are too many nonspecific products despite temperature optimization, make your primers longer.
Another good way to improve specificity is to use a touchdown PCR. That is, start with a really high annealing temperature, such that you get very few reactions (but the ones you get are very likely to be specific). Then decrease the annealing temp by 0.5 - 1 degree per cycle. On the last cycle you'll have a really low annealing temp (thus low specificity) but it won't really matter, because the vast majority of the templates in your tube will be the right thing already (amplified in prior, low-yield, highly specific cycles). I usually bridge the optimal annealing temp predicted by MacVector (e.g., if optimal is 55 degrees, I will start at 63 and go down to 48 over 30 cycles). The manual for your PCR machine should tell you how to program this.
I am very excited about this!
http://www.promega.com/vectors/t_vectors.htm#b01
Which one do you use? pGem-T or pTarget?
We used pGem; I haven't used pTarget. But your link says it's a mammalian expression vector - you don't need that, you're not going to express it anywhere. All you want is a minimal vector with the beta-gal gene, which is the pGem-T.
But first fix your PCR. If you are amplifying junk, you are ligating junk, regardless of what vector you stick it in. GIGO. No T-vector is going to help you with that.
