Pseudomonas species, ubiquitous in soil and water [1, 2], are of considerable scientific and technological importance and comprise a taxon of metabolically versatile organisms capable of utilizing a wide range of simple and complex organic compounds . During the last few years, Pseudomonas strains have been increasingly studied with increasing interest because of their importance in the fields of medicine, food technology, environmental microbiology and phytopathology . They are known to be involved in the biodegradation of natural and toxic man-made chemical compounds. In addition, the bacterial genus Pseudomonas is a prolific producer of a number of extracellular enzymes, including lipase and amylase .
α-amylase (E.C 220.127.116.11) catalyses the hydrolysis of α-D-(1, 4) glycosidic linkages in starch and related carbohydrates. It is a key enzyme in the production of starch derivatives and also widely used in food, textile, paper, detergent, clinical, pharmaceutical and other industrial fields [6, 7].
In the last decades, there has been lots of research on the amylases from animals , plants  terrestrial bacteria  and marine bacteria . Interest in bacterial amylases has increased due to their biotechnological applications . Many microorganisms such as bacteria, yeast, and fungi are known to secrete amylases during their growth on starch substrates, which makes starch derivates available to cells.
In conventional multifactor experiments, optimization of plant extraction is usually carried out by varying a single factor while keeping all other factors fixed at a specific set of conditions. It is not only time-consuming, but also usually incapable of reaching the true optimum due to ignoring the interactions among variables . Response surface designs such as central composite design (CCD) and Box Behnken Design (BBD) are commonly selected for performing optimization studies. Compared to the CCD method, the BBD technique is considered the most suitable for evaluating quadratic response surfaces particularly in cases when predicting the response at the extreme level is not the goal of the model. The BBD technique is a three level design based upon the combination of two-level factorial and incomplete block designs [12, 13]. The BBD method employs a spherical design with excellent predictability within the design space and it requires less experiment than the FFD or CCD with the same number of factors . In addition, the BBD technique is rotatable or nearly rotatable regardless of the number of factors under consideration .
In our study, a new strain C2 was isolated from soil regularly contaminated with olive washing wastewater and the diversity of their extracellular hydrolytic enzymes (lipase and amylase) was studied. The results showed that stain C2, which was identified as Pseudomonas luteola, had a high amylase activity. Optimization extraction was carried out using Box Behnken Design (BBD). This amylase showed a high activity, lead to expect a great commercial value and a good prospect for industrial applications.