Transport of phage P22 DNA across the cytoplasmic membrane

Abstract

Although a great deal is known about the life cycle of bacteriophage P22, the mechanism of phage DNA transport into Salmonella is poorly understood. P22 DNA is initially ejected into the periplasmic space and subsequently transported into the host cytoplasm. Three phage-encoded proteins (gp16, gp20, and gp7) are coejected with the DNA. To test the hypothesis that one or more of these proteins mediate transport of the DNA across the cytoplasmic membrane, we purified gp16, gp20, and gp7 and analyzed their ability to associate with membranes and to facilitate DNA uptake into membrane vesicles in vitro. Membrane association experiments revealed that gp16 partitioned into the membrane fraction, while gp20 and gp7 remained in the soluble fraction. Moreover, the addition of gp16, but not gp7 or gp20, to liposomes preloaded with a fluorescent dye promoted release of the dye. Transport of (32)P-labeled DNA into liposomes occurred only in the presence of gp16 and an artificially created membrane potential. Taken together, these results suggest that gp16 partitions into the cytoplasmic membrane and mediates the active transport of P22 DNA across the cytoplasmic membrane of Salmonella.

FIG. 1.

Membrane partitioning of ejection proteins. Membrane partitioning of individual ejection proteins (A) and the preformed protein complex (B) in crude membrane extract from Salmonella serovar Typhimurium. Proteins were detected by Western blotting with the rabbit anti-six-His antibody after transfer of the proteins to a PVDF membrane. S, soluble fraction; M, membrane fraction; PPC, preformed protein complex. (C) Partitioning of the preformed protein complex in Triton X-114 was detected by Western blotting with the rabbit anti-six-His antibody after transfer of the proteins to a PVDF membrane. Aq, aqueous phase; Det, detergent phase. (D) Controls demonstrating partitioning of BSA into the aqueous phase and the partitioning of the E. coli lactose permease (LacY) into the detergent phase. Proteins were detected by staining with Coomassie brilliant blue. (E) Liposomal partitioning of the ejection proteins. Proteins were detected by probing the PVDF membrane with the rabbit anti-six-His antibody. S, soluble fraction; L, liposomal fraction. (F) Domain of gp16 responsible for membrane association. Truncated constructs of gp16 were cloned and expressed in the E. coli expression host and purified using a nickel column. Each construct was subjected to the liposomal partitioning assay to determine if the construct would partition into the liposomal fraction or remain in the soluble fraction. The different constructs were detected on the PVDF membrane by using rabbit anti-six-His antibody.

FIG. 2.

Membrane-disrupting activity of the ejection proteins. Samples were excited at 490 nm and were read at an emission wavelength of 510 nm. Ethanol was used as the positive control, inducing almost complete lysis of the liposomes, and BSA was used as the negative control. The percent relative fluorescence was determined by calculating the ratio of the relative fluorescence values of the different samples to the relative fluorescence exhibited with 0.2% Triton X-100. (A) The ejection proteins, BSA, and ethanol were added to liposomes with encapsulated fluorescent dye to determine their membrane-disrupting activity. The bar graph depicts the percent relative fluorescence at the 2-min time point. (B) Increasing concentrations of the gp16 protein were added to liposomes with encapsulated calcein dye to determine if dye leakage is concentration dependent. The graph reflects the percent relative fluorescence 2 min after the protein was added. (C) Triton X-100, gp16, and gp7 were added to the liposomes after 2 min of background fluorescence reading. After another 2 min, Triton X-100 was added to the liposomes with gp7 to demonstrate that the fluorescent dye was still inside the liposomes.

FIG. 3.

Transport of DNA inside liposomes is gp16 and membrane potential dependent. Ejection proteins were incubated with radiolabeled DNA prior to adding liposomes to the mixture. After 1 h at room temperature, DNase I was added to digest all untransported DNA. Liposomes were disrupted with Triton X-100 plus proteinase K, and samples were run in a 5% native polyacrylamide gel. The gel was vacuum dried and visualized with a phosphorimager. The artificial membrane potential was created by maintaining 100 mM KCl inside and 20 mM KCl outside the liposomes. The addition of valinomycin equilibrates the [K⁺] inside and outside the liposomes, generating a net negative charge inside the liposomes. Radiolabeled DNA was loaded in the first lane to mark the mobility of the DNA.