An analytical approach to the problems of phage recombination and reproduction. III. Cross reactivation.

Abstract The theoretical implications of cross reactivation (CR) of genetic markers from ultraviolet (UV)-irradiated phages are investigated in this paper. In the analysis of CR, a theory (double-crossover theory) based on the assumption that marker rescue takes place by crossovers between the markers and the closest UV hits on both sides has been applied. The fit of this theory to the CR data for UV doses higher than 50 phage-lethal hits (PLH) is very good when high negative interference is taken into account. The parameters involved in the theory are: (1) the mean length L 0 of local genetic damages (damages hitting the marker used for CR); (2) the fraction g of PLH that produces genetic damages; (3) the mean number b of crossovers per target unit (measured in the proximity of the marker) between a phage or a set of partial replicas of irradiated and carrier type; (4) the average number n of phages or sets of partial replicas of irradiated type capable of recombining with the carrier type in the manner specified by the double-crossover theory; and (5) the ratio TU/MU between target units TU and map units MU in the proximity of the marker. Some of these parameters occur also in other phenomena and have been evaluated before. The mean lengths L 0 of local damages for several r II markers are found to be around 0.0030/ g TU. Considering that in T4 phage 0.8 L = 0.0040 TU of UV damages (genetic and functional) for the entire genome as determined on the basis of multiplicity-reactivation data (Barricelli, 1960) . Using CR data for UV doses higher than 50 PLH, n has been estimated ( n = 2.4). The implications of this result for the question whether irradiated phages multiply before CR rescue are discussed. At UV doses lower than 40 PLH, the theory does not fit the data well. The CR frequency is greater than predicted by the theory, and the clone size of cross-reactivated phages is many times greater than at higher doses. The difference cannot be accounted for simply by the increased CR frequency. CR of two closely linked markers has yielded a method of calculating the value of TU/MU between the two markers when g is known. The values vary between a minimum of about TU = 125 g MU (or 51 g SU, where SU = standard unit) (Barricelli, 1960) to a maximum of about TU = 500 g MU (or 208 g SU). These changes indicate a fluctuation of the crossover frequency per TU. All the TU/MU values have been obtained by using markers in the r IIA cistron. Outside this region one may expect to find larger TU/MU values ( Barricelli, 1960 , Fig. 5). CR of “surrounded” markers—one r + surrounded by two r's , as in the cross r 1 + r 2 r 3 + × UVr 1 r 2 + r 3 in which only wild-type yielding complexes are scored—gives information on the relative crossover frequency in the regions near the UV hits as compared to the central region of the undamaged segment between them. The analysis of such data has shown that the crossover frequency is significantly lower in the regions close to the UV damages. This method of analysis is valid only for UV doses higher than 50 PLH, when the mean length of an undamaged area between two UV damages is probably not much larger than the mean length of a heterozygous region. It can, therefore, not be used as an argument against the idea that crossovers are limited to the homologous regions of overlapping replicas. However, the result is contrary to what would be predicted by all theories implying that crossovers should be particularly frequent at the end points of an undamaged region (terminal crossovers) or at the end points of a partial replica. A theoretical model for the interpretation of CR is presented.