The a-amylase from the thermoacidophilic eubacterium Alicyclobacillus (Bacillus) acidocaldarius strain ATCC 27009 was studied as an example of an acidophilic protein. The enzyme was purified from the culture fluid. On an SDS/polyacrylamide gel, the protein exhibited an apparent molecular mass of 160 kDa, which is approximately 15% higher than that predicted from the nucleotide sequence. The difference is due to the enzyme being a glycoprotein. Deglycosylation or synthesis of the enzyme in Escherichia coli gave a product with the mass expected for the mature protein.The amylase hydrolyzed starch at random and from the inside, and its main hydrolysis products were maltotriose and maltose. It also formed glucose from starch (by hydrolysing the intermediate product maltotetraose to glucose and maltotriose) and exhibited some pullulanase activity. The pH and temperature optima were pH 3 and 75 "C, respectively, characterizing the enzyme as being thermoacidophilic. Alignment of the sequence of the enzyme with that of its closests neutrophilic relatives and with that of a-1,4 or a-1,6 glycosidic-bond hydrolyzing enzymes of known threedimensional structure showed that the acidophilic a-amylase contains approximately 30% less charged residues than do its closests relatives, that these residues are replaced by neutral polar residues, and that hot spots for these exchanges are likely to be located at the surface of the protein.Literature data show that similar effects are observed in three other acidophilic proteins. It is proposed that these proteins have adapted to the acidic environment by reducing the density of both positive and negative charges at their surface, that this effect circumvents electrostatic repulsion of charged groups at low pH, and thereby contributes to the acidostability of these proteins.Enzymes optimally active under extreme conditions are both of biotechnological importance and of scientific interest (for reviews see [I-31). Much work has been devoted to the mechanism of thermostability. However, its outcome is complex. Theoretical considerations suggest that the exchange of a single amino acid residue in an enzyme can increase its temperature optimum by more than 10°C [3]. Studies on forced molecular evolution have provided support for this notion. In such thermo-drill experiments, a gene encoding a mesophilic enzyme is introduced into a thermophilic organism, and by gradually increasing the growth tem-Correspondence to E. P. Bakker, Abteilung Mikrobiologie,