coli C ΔagaS and not because this deletion Daporinad order was exerting a polar effect on downstream genes, namely, kbaY, agaB, agaC, agaD, and agaI (Figures 1 and 8E). Among these genes, kbaY is involved in the last step of the Aga and Gam pathway, while agaBCD, are involved

in Gam uptake and agaI is not needed for the utilization of Aga and Gam as we have shown above. Thus, if the Aga- phenotype in the ΔagaS mutants is due to a polar effect on a downstream gene it would be kbaY. As expected, the EDL933/pJF118HE and E. coli C/pJF118HE grew on Aga whereas the ∆agaS mutants with pJF118HE did not grow (Figure 8A). Importantly, E. coli C and EDL933 ∆agaS mutants with either pJFagaSED or pJFagaSYED grew on Aga (Figures 8A and 8E). Complementation of the Aga- phenotype by pJFagaSED showed that deletion of agaS caused the Aga- phenotype and not because the deletion of agaS had a polar effect on kbaY expression. Although both pJFagaSED and pJFagaSYED complemented the Aga- phenotype they failed to complement the Gam- phenotype in E. coli C ∆agaS (Figures 8B and 8E). It is likely that the deletion in agaS was causing a polar effect on agaBCD. This was tested by using pJFagaBDC to complement the Gam- phenotype. E. coli C ∆agaS/pJFagaBDC did not grow on Gam plates (Figures 8B and 8E). The plasmid, pJFagaBDC, is functional because we have shown that EDL933 which is Gam-

manifests a Gam+ phenotype when it harbors this plasmid (unpublished data). Since neither pJFagaSYED nor pJFagaBDC could complement the Gam- phenotype, the most likely explanation is that the deletion of agaS not only affects MK-1775 manufacturer the Aga/Gam pathway but also exerts polarity on the expression of agaB, agaC, and agaD. If this is the case, then the plasmid, pJFagaSDC, should complement the Gam- phenotype and it does because E. coli C ∆agaS/ pJFagaSDC grew on Gam plates (Figures 8B and 8E). Identical results were obtained when complementation was done on Aga and Gam plates see more without any added nitrogen (data not shown). These experiments raise the question why the partial deletion of agaS in ∆agaS mutants does not exert polarity on kbaY but is polar on further downstream agaBCD genes.

The most likely explanation 5-FU cost is that the strength of the polarity is a function of distance from the mutation [20, 21]. These complementation experiments were done at 30°C because it was observed that at lower temperatures complementation of ∆agaS mutants with these plasmids was better. In addition, complementation by these plasmids was not observed when IPTG was added at a concentration as low as 10 μM (data not shown) suggesting that over-expression of the AgaS protein, unlike over-expression of AgaA and NagA, is detrimental to the cell. These experiments clearly demonstrate that the agaS gene is involved in Aga and Gam utilization. Figure 8 Complementation of ∆ agaS mutants of EDL933 and E. coli C on Aga and Gam plates. EDL933 and E.