Purification of the Mannitol-specific Enzyme II of the Escherichia coli Phosphoenolpyruvate:Sugar Phosphotransferase System*

SUMMARY The inducible, mannitol-specific Enzyme II of the phosphoenolpyruvate:sugar phosphotransferase sys- tem has been purified approximately 230-fold from Escherichia coli membranes. The enzyme, initially solubilized with deoxycholate, was first subjected to hy- drophobic chromatography on hexyl agarose and then purified by several ion exchange steps in the presence of the nonionic detergent, Lubrol PX. The purified pro- tein appears homogeneous by several criteria and probably consists of a single hind of polypeptide chain with a molecular weight of 60,000 (25%). talyzing

Enzyme II complexes are tightly associated with the membrane and function as the sugar-specific recognition components of the PTS (7). In addition, at least some substrates of the PTS also require another sugar-specific protein for their transport, Enzyme III, which may be either membrane-associated or soluble, and is itself phosphorylated by phospho-HPr during the reaction (2,7,8 In whole cells and in membrane vesicles, this reaction can be shown to occur vectorially such that external sugar is taken up and phosphorylated at the expense of internal sugar-P (11, 12). Thus, an Enzyme II complex can apparently catalyze both active group translocation and exchange group translocation (4).
In order to investigate the mechanisms of these transport reactions in detail, it is necessary to have a highly purified, stable Enzyme II. In this communication, we report the purification to apparent homogeneity of the n-mannitol-specific Enzyme II from E. coli. The purified enzyme has a polypeptide chain molecular weight of 60,000 and catalyzes both phosphoenolpyruvate-and mannitol-l-P-dependent phosphorylation of n-mannitol under appropriate conditions. Our preliminary characterization of Enzyme IImtl shows that it should be suitable for detailed catalytic and physicochemical studies. apparatus. After drying for 10 min on a hotplate, the filters were counted in 5 ml of a toluene-based scintillation fluid. Mannitol-I-P: mannitol transphosphorylation was measured as previously described (10) by the Dowex l-X2 resin column procedure (5). When purified Enzyme IImt' was assayed for either activity, Lubrol PX was present after a dilution of the stock enzyme solution (0.025 to 0.05%, final concentration).
Analytical Procedures-Protein was estimated according to Lowry et al. (15) with bovine serum albumin as the standard. Solutions containing Lubrol PX gave a precipitate in this procedure. Therefore, samples were centrifuged before reading, and appropriate blanks were run. Polyacrylamide gel electrophoresis in the presence of sodium dodecyl sulfate was performed according to Laemmli (16) in l-mmthick slabs. Some samples were concentrated by lyophilization after dialysis against 5 mM NH4HC03 containing 1 mM P-mercaptoethanol before being loaded on the gel. Electrophoresis in gels containing Lubrol PX was at 4°C in 6% polyacrylamide slabs using the Tris/ glycinate system of Jovin et al. (17) with the following modifications.
Lubrol PX (0.125%) was included in the gel and in the upper and lower buffers.
brane in the presence of 0.2 M NaCl. These experiments also revealed that the nonionic detergent, Lubrol PX, although inefficient in solubilization, activated the deoxycholate-extracted enzyme approximately 1.8fold at concentrations greater than 0.1%. A purification strategy was therefore developed using a hydrophobic resin which bound Enzyme IImt' in the presence of deoxycholate and released it upon the addition of Lubrol PX. Of the series of alkyl agaroses available commercially (Miles), we found that the hexyl, octyl, and decyl derivatives all retained Enzyme IImt'. Chromatography on hexyl agarose in extraction buffer, followed by elution with Lubrol PX, afforded a 20-to 25-fold purification as shown in Fig. 1A. After extensive dialysis of the eluted protein against buffer containing Lubrol PX to remove most of the deoxycholate, further purification by ion exchange chromatography on DEAE-cellulose ( Fig. 1B) and w-aminohexyl agarose ( Fig.  1C) resulted in an additional 3-to 4-fold purification. After rechromatography on a small w-aminohexyl agarose column, the enzyme had been purified approximately 235-fold relative to crude E. cob membranes.  (19) supplemented with thiamin, uracil, and arginine (100 pg/ml each) and with 0.5% mannitol as the carbon source. Cells were harvested in the midexponential growth phase (A550 = 1.0 to 1.2) after quickly being cooled to 4°C in an ice bath and were washed twice with cold Medium 63. The bacteria (3 to 3.5 g from 2 liters of culture) were resuspended in 40 ml of 20 mM Tris-HCl buffer, pH 7.5, containing 1 mM dithiothreitol, and were broken by passage through a French Press (10,000 p.s.i.) at 4°C. Cell debris was removed by centrifugation at 10,000 X g for 5 min, and membranes were collected by centrifugation at 100,000 X g for 90 min. They were washed once with 40 ml of the cell resuspension buffer and stored at -70°C until used.
The results of polyacrylamide gel electrophoresis in the presence of dodecyl sulfate of samples taken after each purification step are shown in Fig. 2A. After the final step, only a single band, apparent M, = 60,000 (+5%), was visible on the gel. In contrast, if membranes from stationary phase cells were used as starting material, two additional bands (Mr = 29,000 and 28,000) co-purified with the larger component ( Fig.  2A, Panel 7). This result suggests that proteases present in stationary phase cells may partially cleave the native protein during purification.
For this reason, only membranes from midexponential phase cells were used to purify Enzyme IImt'.
Purification of Enzyme ZZ"" -Membranes from 3.5 g of midexponential phase cells were extracted with shaking at 20°C for 30 min in a total volume of 50 ml of 20 mM Tris-HCl, pH 8.4, containing 0.2 M NaCl, 1 mM dithiothreitol, and 0.5% sodium deoxycholate (extraction buffer). The solution was chilled on ice, and all subsequent operations were performed at 4°C. The suspension was centrifuged at 100,000 x g for 90 min, and the supernatant was loaded onto a column of hexyl agarose (3 x 9 cm) equilibrated with extraction buffer. After three column volume washes with extraction buffer, the Enzyme IImt' activity was eluted with three column volumes of the same buffer containing 0.5% Lubrol PX, and 5-ml fractions were collected (see Fig. 1 for profiles of this and the following columns). The peak fractions were pooled and dialyzed against 20 volumes of 20 mM Tris-HCl, pH 8.4, containing 1 mM dithiothreitol and 0.5% Lubrol PX (TDL buffer). Dialysis was continued for 40 h with two changes of the buffer. The solution was then applied to a DEAE-cellulose column (1.5 X 5 cm) equilibrated with TDL buffer. After the column was washed with two column volumes of buffer, it was developed with a 70-ml linear gradient, 0 to 0.3 M NaCl, in TDL buffer (2.5-ml fractions). Enzyme IImt' eluted at about 0.1 M NaCl, and the peak fractions were pooled and dialyzed overnight against 50 volumes of TDL buffer. Next, the solution was loaded onto a column of w-aminohexyl agarose (1.5 x 2 cm). Two column volume washes with TDL buffer were followed by a 30-ml linear gradient, 0 to 0.3 M NaCl in TDL buffer (lml fractions). Activity eluted at about 0.16 M NaCl, and the peak fractions were dialyzed overnight against 100 volumes of TDL buffer. At this point, the enzyme appeared highly purified (CL Fig. 2), and minor impurities were removed by rechromatography on w-aminohexyl agarose (1.5 X 1 cm column) in TDL buffer. After a 4-ml wash, this column was eluted stepwise with 4 ml each of TDL buffer containing 0.05, 0.10, 0.15, and 0.2 M NaCl. Most of the activity came off with the 0.15 M NaCl wash, and this fraction was again dialyzed overnight against 500 ml of TDL buffer and stored in aliquots at -70°C until needed.
To confirm that the major electrophoretic band in our purified preparation was indeed Enzyme IImtl, a nondenaturing gel in the presence of Lubrol PX was run. Again, a single band was obtained, and elution of the enzyme from a parallel, unstained gel showed that Enzyme II"" activity co-migrated with this band (Fig. 2B)  Membranes from 3.5 g of midexponential phase cells were used as starting material. Enzyme II"" was purified as described under "Experimental Procedures" and in the legend to Fig. 1. Activities listed refer to mannitol-1-P formed in the phosphoenolpyruvate-dependent reaction.
We have also examined several catalytic properties of the purified protein. Phosphoenolpyruvate-dependent sugar phosphorylation was not detected with fructose, mannose, or methyl-cu-glucoside as substrates. Sorbitol  did serve as a substrate with a velocity at 0.1 mM equal to about 3% of that observed with mannitol.3 Purified Enzyme IImtl, like the activity in crude membranes (lo), was very sensitive to sulfhydryl reagents. Phosphoenolpyruvate-dependent activity was rapidly lost during exposure of the enzyme to Nethylmaleimide.
This result indicated that at least one free sulfhydryl group was necessary for catalytic function.  (2) "Yields greater than 100% are probably due to activation of Enzyme II"" by Lubrol PX, which was present in all the assays except those of crude membranes. bands close to it on the acidic side were evident (Fig. 2C). The major component was slightly acidic with a p1 of 6.2. The minor bands possibly represented modified Enzyme IIm molecules arising, for example, by oxidation or deamidation during purification or electrofocusing. It seems unlikely that they were contaminants consisting of unrelated proteins since single bands were obtained in the two other gel systems. Judging from these experiments, we estimate that the protein was >95% homogeneous and that it consisted of a single kind of polypeptide chain? The latter conclusion, however, must Biochemical and genetic investigations have provided evidence suggesting that Enzyme II complexes of the PTS are responsible for the sugar-P:sugar transphosphorylation reactions detected in intact bacterial cells and isolated membranes (10). We have confiied this conclusion with purified Enzyme II"*' which we have found to catalyze phosphoryl transfer from mannitol-1-P to ['4C]mannitol in the absence of the soluble PTS enzymes. The pH-activity profiles of this reaction for both unfractionated membranes and the purified protein gave maxima near pH 7. In contrast, purified Enzyme IImt' exhibited two maxima, near pH 7 and pH 9, in the phosphoenolpyruvate-dependent phosphorylation of mannitol (not shown). This was probably due to contributions to the shape of the curve by either or both of the soluble PTS enzymes. We have also conducted preliminary initial rate kinetic studies of the transphosphorylation reaction. However, strong substrate inhibition was exhibited by both mannitol and mannitol-1-P which complicated detailed kinetic anaIyses of the data obtained.
Purification procedures have been reported previously for the membrane-bound Enzyme II specific for glucose and mannose ("Enzyme IIB") from E. coli (20) and the lactosespecific Enzyme II from Staphylococcus aureus (21). However, evidence for homogeneity other than a single band on a dodecyl sulfate gel was not presented in either case. Furthermore, purification relative to crude membranes was only about lo-fold for the E. coli enzyme and 20-fold for the one from S. aureus, and both preparations were unstable (20,21). In contrast, our purified Enzyme II"' is stable for several days at 4°C and for at least 2 months at -70°C.
In the present study, we have used hydrophobic chromatography as the most effective step in the purification of Enzyme IImt'. After subsequent ion exchange chromatographic steps, however, the overah yield was only about 25%, with the greatest loss occurring during DEAE-cellulose chromatography. Nevertheless, we have shown that our preparation of Enzyme IImtl is apparently homogeneous with a subunit molecular weight of 60,000. Furthermore, we have demonstrated the ability of the purified enzyme to catalyze mannitol-1-P: mannitol transphosphorylation in the absence of the soluble PTS enzymes and the apparent independence of the phosphoenolpyruvate reaction from an Enzyme III requirement. It may be necessary to scale up the purification procedure and to increase the yield in order to obtain enough material for more detailed physical and chemical analyses of the enzyme. However, the method described in this communication yields a sufficient quantity of stable Enzyme IImt' for a variety of studies which should be important in determining the mechanism of PTS-mediated transport in E. coli.