If you are using a nickase based Cas9, then you would need two separate gRNAs whereas if the Cas9 is programmed to make a DSB then you would only need one gRNA.

If you are planning on using a repair template it is highly recommended to use the Cas9 nickase mutant. Introduction of a nick, compared to a double-strand break, is described to be repaired preferentially through HDR which in this case is desired. Double-strand breaks are more likely to be repaired by NHEJ, so using two sgRNAs which cut in close proximity might have an increased probability of inducing NHEJ rather than HDR. However, of course using 2 sgRNAs and the nickase will greatly improve cutting specificity and decrease off-target effects.

For further reading I suggest Ran et al., 2013, Nature Protocols and papers from the Zhang Lab in general.

Thanks for the answers. I read that paper, but still have some confusion. If I use Cas9 nickase mutant, I need to have two gRNAs and a donor template,  can use single strand oligos here?is it must to create a silent mutation on the PAM site of donor template, if not what is the success rate? 

Is it possible to introduce a single point mutation using single stand oligos (donor template) and DSBs using WT Cas9?

thanks again both of you.

I am also new to this field and trying to work out the best way to make my mutation.

Figure 6C in the Ran paper shows greater HDR for wtCas9 than nickase mutant so i thought it should work fine with one gRNA and double strand break?