Effect of starvation and the viable-but-nonculturable state on green fluorescent protein (GFP) fluorescence in GFP-tagged Pseudomonas fluorescens A506.

The green fluorescent protein (GFP) gene, gfp, of the jellyfish Aequorea victoria is being used as a reporter system for gene expression and as a marker for tracking prokaryotes and eukaryotes. Cells that have been genetically altered with the gfp gene produce a protein that fluoresces when it is excited by UV light. This unique phenotype allows gfp-tagged cells to be specifically monitored by nondestructive means. In this study we determined whether a gfp-tagged strain of Pseudomonas fluorescens continued to fluoresce under conditions under which the cells were starved, viable but nonculturable (VBNC), or dead. Epifluorescent microscopy, flow cytometry, and spectrofluorometry were used to measure fluorescence intensity in starved, VBNC, and dead or dying cells. Results obtained by using flow cytometry indicated that microcosms containing VBNC cells, which were obtained by incubation under stress conditions (starvation at 37.5 degrees C), fluoresced at an intensity that was at least 80% of the intensity of nonstressed cultures. Similarly, microcosms containing starved cells incubated at 5 and 30 degrees C had fluorescence intensities that were 90 to 110% of the intensity of nonstressed cells. VBNC cells remained fluorescent during the entire 6-month incubation period. In addition, cells starved at 5 or 30 degrees C remained fluorescent for at least 11 months. Treatment of the cells with UV light or incubation at 39 or 50 degrees C resulted in a loss of GFP from the cells. There was a strong correlation between cell death and leakage of GFP from the cells, although the extent of leakage varied depending on the treatment. Most dead cells were not GFP fluorescent, but a small proportion of the dead cells retained some GFP at a lower concentration than the concentration in live cells. Our results suggest that gfp-tagged cells remain fluorescent following starvation and entry into the VBNC state but that fluorescence is lost when the cells die, presumably because membrane integrity is lost.