Exploration of new geometries in cellulosome-like chimeras

Abstract

In this study, novel cellulosome chimeras exhibiting atypical geometries and binding modes, wherein the targeting and proximity functions were directly incorporated as integral parts of the enzyme components, were designed. Two pivotal cellulosomal enzymes (family 48 and 9 cellulases) were thus appended with an efficient cellulose-binding module (CBM) and an optional cohesin and/or dockerin. Compared to the parental enzymes, the chimeric cellulases exhibited improved activity on crystalline cellulose as opposed to their reduced activity on amorphous cellulose. Nevertheless, the various complexes assembled using these engineered enzymes were somewhat less active on crystalline cellulose than the conventional designer cellulosomes containing the parental enzymes. The diminished activity appeared to reflect the number of protein-protein interactions within a given complex, which presumably impeded the mobility of their catalytic modules. The presence of numerous CBMs in a given complex, however, also reduced their performance. Furthermore, a "covalent cellulosome" that combines in a single polypeptide chain a CBM, together with family 48 and family 9 catalytic modules, also exhibited reduced activity. This study also revealed that the cohesin-dockerin interaction may be reversible under specific conditions. Taken together, the data demonstrate that cellulosome components can be used to generate higher-order functional composites and suggest that enzyme mobility is a critical parameter for cellulosome efficiency.

FIG. 1.

Schematic representations of the recombinant proteins and complexes used in this study. The source of the respective module (see the symbol key) is indicated as follows: gray (C. cellulolyticum), white (C. thermocellum), and black (R. flavefaciens). In the shorthand notation for the engineered enzymes, the numbers 9 and 48 refer to the corresponding GH family (GH9 and GH48) of the catalytic module; M designates the tandem CBM3a and X2 modules of the C. cellulolyticum scaffoldin CipC; uppercase letters C, F, and T indicate the source of the cohesin module, C. cellulolyticum, R. flavefaciens, and C. thermocellum, respectively; lowercase letters c, f, and t indicate the source of the dockerin module, C. cellulolyticum, R. flavefaciens, and C. thermocellum, respectively. The names symmetrical, asymmetrical, cyclic, and polymeric are intended both as descriptive terms of the artificial cellulosome-like geometries and to facilitate subsequent narratives in the text. The component compositions are shown schematically and in the shorthand notation are given below the individual elements of the figure. The hybrid and mixed cellulosomes both contain scaffoldins and are thus based on the native complexes. Novel proteins constructed in this study are indicated by an asterisk.

FIG. 2.

Avicel hydrolysis by the recombinant enzymes alone. The amounts of reducing sugars were determined after 0, 1, 6 and 24 h of incubation at 37°C. (A) GH48 cellulases. The curves are labeled as follows: •, 48t; ○, M-48c; ▾, M-C-48; ▵, M-C-48t; ▪, M-F-48t; and □, Scaf3:48t complex. (B) GH9 cellulases. The curves are labeled as follows: •, 9c; ○, M-T-9c; ▾, M-T-9f; and ▵, the Scaf3:9c complex. The enzyme concentration was 0.1 μM in all kinetic experiments. The data are the means of three independent experiments (variation within 5% of the mean).

FIG. 3.

Preference of parental versus engineered enzymes for Avicel over PAS-cellulose. The enzymes are indicated at the bottom of the graph. The data represent the ratio of the specific activity on Avicel/specific activity on PAS-cellulose. The specific activity on Avicel was estimated after 24 h of hydrolysis (data from Fig. 2). The standard deviations are indicated by the error bars.

FIG. 4.

Electrophoretic mobility of components and assembled complexes on nondenaturing gels. Lane 1, M-C-48 alone; lane 2, M-T-9c alone; lane 3, asymmetrical cellulosome containing M-C-48 and M-T-9c; lane 4, Scaf6 alone; lane 5, 48t alone; lane 6, 9f alone; lane 7, M-48-9c alone; lane 8, mixed cellulosome containing Scaf6, 48t, 9f, and M-48-9c. In each lane, the concentrations of the indicated proteins were 10 μM, except in lane 8 where 5 μM of each component was used. Similar quality gels were obtained for all complexes used in this study.

FIG. 5.

Analysis of the reversibility of the cohesin-dockerin interaction. (A) Electrophoretic mobility of components and assembled complexes in C. cellulolyticum. Lane 1, M-C-48 alone; lane 2, M-C-48t alone; lane 3, M-T-9c alone; lane 4, complex M-C-48/M-T-9c (asymmetrical cellulosome); lane 5, complex M-C-48t/M-T-9c (symmetrical cellulosome); lane 6, asymmetrical cellulosome mixed with M-C-48t resulting in symmetrical cellulosome plus free M-C-48. In each lane, equimolar concentrations (10 μM) of the indicated proteins were used. (B) Electrophoretic mobility of components and assembled complexes in C. thermocellum. Lane 1, M-C-48t alone; lane 2, M-T-9f alone; lane 3, M-T-9c alone; lane 4, complex M-C-48t/M-T-9f; lane 5, complex M-C-48t/M-T-9f mixed with M-T-9c resulting in complex M-C-48t/M-T-9c (symmetrical cellulosome) plus free M-T-9f; lane 6, M-T-9f alone; lane 7, M-T-9c alone; lane 8, complex M-C-48t/M-T-9c (symmetrical cellulosome). In each lane, equimolar concentrations (10 μM) of the indicated proteins were used. Schematic representations of the assembly and dissociation of the indicated complexes are shown below each gel. See the legend to Fig. 1 for details of subunits, modular components. and shorthand notation. The numbers in the top left-hand corner indicate the corresponding lanes in the nondenaturing gel.

FIG. 6.

Assembly of the cyclic cellulosome. (A) Electrophoretic mobility of components and assembled complexes. Lane 1, M-C-48t alone; lane 2, M-T-9f alone; lane 3, M-F-48t alone; lane 4, M-T-9c alone; lane 5, complex 1 (M-C-48t plus M-T-9f); lane 6, complex 2 (M-F-48t plus M-T-9C); lane 7, complex 1 plus complex 2 (cyclic cellulosome). In lanes 1 to 6, equimolar concentrations (10 μM) of the indicated proteins were used. In lane 7, complexes 1 and 2 were mixed at a final concentration of 5 μM. (B) Gel filtration analysis of the assembled complexes. Injected proteins (100 μl) and the corresponding lanes in panel A are indicated on each chromatogram. mAU refers to milli absorbance units at 280 nm. Vertical lines indicate the positions of molecular mass markers: blue dextran or void volume (>2 MDa), ferritin (440 kDa), aldolase (158 kDa), and bovine serum albumin (67 kDa). For each chromatogram, the indicated proteins were at a concentration of 10 μM, except for cyclic cellulosome, where 5 μM concentrations of complexes 1 and 2 were used. The peak observed at 20 ml in all chromatograms corresponds to the maleic acid contained in the sample buffer. Note that the formation of complex 2 generates a major and minor band on nondenaturing PAGE (panel A, lane 6), but a single peak at the expected mass on gel filtration (panel B). Since the mobilities of the bands differ from those observed for free components (panel A, lanes 3 and 4), it is assumed that the two bands correspond to two different conformations of complex 2.

FIG. 7.

Stability of the asymmetrical and mixed cellulosomes. Samples containing 10 μM asymmetrical complex (A and C) or 5 μM mixed cellulosomes (B and D) were incubated at 37°C. At 0, 1, 6, and 24 h (time of incubation indicated above the lanes), the aliquots were subjected to nondenaturing PAGE (A and B) or to SDS-PAGE (C and D).

FIG. 8.

Kinetic studies of Avicel hydrolysis by the various engineered cellulosomes. The amounts of reducing sugars were determined after 0, 1, 6, and 24 h of incubation at 37°C. The data show the means of four independent experiments (variation within 5% of the mean).

FIG. 9.

Comparative solubilization of microcrystalline cellulose Avicel by the various complexes and free enzyme systems. The composition of the complexes and free enzyme systems is indicated at the bottom of the graph. The data represent the amount of released soluble sugars after 24 h of reaction by the given enzyme system and are summarized from Fig. 8. Error bars indicate the standard deviations.