SUPPLEMENTARY MATERIAL
Effect of the Int-arm binding sites of phage HK022 on the reaction directionality
The role of each of the five tight Int-binding sites on the P and P’ arm of att (P1-2, P'1-3, Online Fig. 1) in determining the directionality of the recombination reaction has been investigated by testing how triple-base mutations (TCA to GTC marked as ten mutations) affect their integrative and excisive activities. P1 and P’3 sites have been shown to be essential for integration but not for excision, P2 and P’1 for excision but not integration and P’2 for both reactions (Numrych et al. 1990). Since the Int-binding sites and the P and P’ arms of and HK022 are almost identical (Yagil et al. 1989) we subjected the HK022 sites to a similar ten mutational analysis using a 2.7 kb vector (pUC18) encoding either an attP site (carrying the P and P’ arms), an attL site (carrying the P’ arm) or an attR site (carrying the P arm). The effect of each ten mutation on integrative and excisive recombination was assayed as follows. The DNA substrates in each integrative (attP x attB) reaction included an attP mutant on the circular pUC18 plasmid [Online Fig. 1A(a)] and a linear radioactive attB fragment [Online Fig. 1A(c)]. As a positive control, the reaction also included a wild type attP cloned on a larger circular 4 kb vector [pEMBL19, Online Fig. 1A(b)]. attB x attP reactions could yield two radioactive linear products [Online Fig. 1A(d)], the longer representing the product of the wild type control and the shorter reaction product of the mutated substrate. The excisive attL x attR reactions were done similarly, a ten mutation on an attL circular plasmid was reacted with a linear radioactive wild type attR fragment and each attR mutant circular plasmid was reacted with a linear radioactive wild type attL fragment. The reactions likewise included wild type controls on a larger plasmid (scheme not shown). Both integrative (attP x attB) and excisive (attL x attR) reactions included purified Int and IHF and the latter also included purified Xis. Radiograms of sample gels for the integrative and excisive reactions are shown in Online Figs. 1B and 1C, respectively. L represents the presence (+) of the larger control substrate whose recombination product is marked on the radiogram by the upper arrows. S represents the presence of the smaller mutated substrate and its product is marked by the arrows below. The bottom arrows on the radiograms present the unreacted att substrates. The content of each lane is specified in the legend of Online Fig. 1. Online Fig. 1D represents the calculated integrative and excisive activities of each ten mutation relative to the wild type activity. The results clearly show that the P’3 and P1 sites direct integrative reaction because their mutations abolished integration but not excision. Likewise, P’1 and P2 sites are important for excision and P’2 is important for both reactions. In Online Fig. 1C the internal larger wild type control (top band) showed a weaker product that was consistent along both experiments. However this did not interfere with the analysis of the results. These results agree exactly with those of the system (Numrych et al. 1990).
Legend to Supplementary Figure
Online Figure 1. Effect of ten mutations on the directionality of the site-specific recombination reaction in vitro. A. Scheme of the integrative experiment (see text). B. Autoradiogram of the integrative reactions. Lanes a-c in are control reactions. Lane a shows the product of wild type attP on the smaller plasmid alone Lane b shows the reaction with the wild type attP alone on the larger plasmid. Lane c shows both wild type attP substrates mixed in the same reaction. Lanes d-f show the results of the three different ten mutations on the P’ arm and lanes g, h show the results with the two mutations on the P arm. C. Both types of excisive reactions. In lanes a-e the circular plasmid carried attL sites reacting with the wild type labeled linear attR site. Lanes a-b are the wild type controls and lanes c-e show the relevant P’ mutations on the attL plasmid. In lanes f-h the circular plasmid carried attR sites reacting with the wild type labeled linear attL, lanes f being the wild type control. The intensity of each mutated product relative to the intensity of the wild type product. D. Calculated results of 5 independent experiments relative to the wild type controls.Asterisks indicate radioactive label.
1