EFFECTS OF STORAGE ON DIRECT ESTIMATES OF BACTERIAL NUMBERS OF PRESERVED SEAWATER SAMPLES

There is a dramatic decrease, on average 39% during the first 40 days, in bacterial cell numbers with storage time of glutaraldehyde preserved seawater samples using the Acridine Orange fluorochrome method of estimating bacterial numbers commonly used in most laboratories. This decline in cell numbers was modelled accurately so that predictions of the initial population could be calculated. However, these models could not be applied to samples from a different year and/or location. Similar losses were found in formalin fixed samples stored in brown glass bottles and when using the DNA-specific fluorochrome DAPA. This indicates that the apparent cell loss with storage time seen in this investigation was not peculiar to the preservative (glutaraldehyde), the storage bottles ( polystyrene tissue culture flasks) or the fluorochrome (Acridine Orange). Attachment to the inner surfaces of sample bottles containing preserved seawater samples stored for several hundred days did account for 41-48% of the original seawater bacterial concentration before storage. Sanification of stored preserved seawater prior to storage. increased the concentration of unattached bacteria by 5-199%. Such inconsistency of cell release indicates that treatment by sanification cannot ensure full recovery of cells and an accurate estimate of the original concentration of bacteria in the preserved seawater prior to storage. Furthermore, as 29-91% of the original concentration remained unaccounted for in both the attached and unattached population of bacteria, there seems to be a genuine loss of cells within the sample bottle that cannot be explained solely by attachment. The effectiveness of the fixatives to preserve or cross link the nucleic acids, dissolution of the protein and cell lysis are examined as possible explanations. A decrease in the mean cell size and shift in size-frequency distribution also was observed with storage time of preserved seawater samples. Estimates of bacterial numbers, biovolume and biomass from preserved seawater samples stored for weeks and months will substantially underestimate the importance of bacteria in oceanic ecosystems. There is no significant difference between bacterial counts carried out at sea and the same samples (preserved, stained, filtered and mounted on slides), stored frozen and recounted on land and we, therefore, recommend this method of sample preparation and storage.