In accordance with previous results, the activity of extracellular proteases from Pseudomonas fluorescens MFO is maximal at a growth temperature of 17.5°C, well below the optimal growth temperature. In addition, the activities of three periplasmic phosphatases display the same growth temperature optimum. Chemostat experiments have shown that it is the growth temperature itself and not the value of the growth rate that regulates these activities. In contrast, a foreign periplasmic phosphatase, expressed under the control of its own promoter, displays a different sensitivity toward temperature. We conclude that in the psychrotrophic strain P. fluorescens MFO, growth temperature exerts a specific control upon the activity of certain enzymes. The critical temperature (17.5C) is within the range of normal growth, suggesting that this control is probably different from a cold shock or heat shock response.Pseudomonas fluorescens is well known as a major psychrotrophic contaminant of raw milk stored in refrigerated tanks (16). Many studies of this bacterium have been concerned with the production of deleterious extracellular enzymes, such as thermostable proteases (8,15,27). Among the numerous observations concerning these enzymes, it has been repeatedly shown that most strains maximally produce proteases at a temperature (15 to 20°C) well below the optimal growth temperature (25 to 30°C) (13,19,25). However, no studies have yet dealt with the mechanism of regulation of protease production with regard to temperature.At this stage two main questions can be raised regarding the elucidation of this mechanism. The first one relates to the specificity of this temperature effect, i.e., whether it is restricted to the production of proteases or extended to the production of other enzymes. To this end, the activities of extracellular proteases as a function of growth temperature were compared with those of several periplasmic phosphatases. The exported enzymes all showed the same regulation by temperature even though they are clearly differentially regulated by other growth conditions. Thus, it was important to determine whether this temperature effect might involve protein export through the cytoplasmic membrane. If so, any foreign exported protein should be submitted to the same effect. The expression of a gene from the mesophilic species Escherichia coli, under the control of its own promoter, was studied in P. fluorescens at different growth temperatures. In this case, a temperature effect similar to that observed with the native enzymes was not demonstrated.The second question is whether the temperature itself is the direct cause of the regulation or an indirect effector acting through the growth rate variation; such an indirect effect has indeed been demonstrated for several activities or proteins in mesophilic bacteria (5). To answer this question, the activities of the two acidic phosphatases were assayed in cells grown in a chemostat at two different temperatures and * Corresponding author. several dilution rates. The ...