(167a) A Quantitative Determination of the Specific and Nonspecific Interaction Forces Between Carbohydrate-Binding Module of CBH1/Cel7A and Lignocellulosic Biomass

Introduction

Enzymatic hydrolysis of lignocellulosic biomass has been
considered as one of the key steps in the biochemical conversion of cellulosic
biomass to sustainable biofuels and biochemicals. One of the major inhibitory
mechanisms preventing efficient enzymatic hydrolysis is the nonproductive
binding of cellulase to lignin. It has been suggested that cellulases adsorb to
lignin and cellulose via hydrophobic, hydrogen bonding and ionic bond
interactions. However, the exact mechanism of cellulase binding to cellulose
and lignin remain unknown. Thus, elucidating the mechanism behind cellulase
adsorption to both cellulose and lignin is required for the development of
effective enzymes or pretreatment technologies for biofuel production. To
understand the specific effect of lignin on cellulase adsorption, a set of reference
substrates derived from kraft, sulfite and organosolv pulping was utilized to
identify the types of forces involved in enzymatic interactions with cellulosic
materials. The reference substrates used have known surface composition of
cellulose and lignin.

Most of fungal cellulases have two modules namely a
catalytic domain (CD) module and a carbohydrate-binding module (CBM). It is
believed that the CBM plays a role in increasing the concentration of CD on the
substrate surface and thus increasing the enzymatic action on the insoluble
substrate surface. Therefore, studying direct interaction mechanism between the
lignocellulose surface and CBM allows us to elucidate the nonproductive binding
component of cellulases to lignin.

Atomic force microscopy (AFM) has the unique ability to
quantify interactions between two surfaces at the molecular level and in native
environments. We utilized AFM to measure the total forces acting between CBM
and reference substrates in buffered medium. By performing adhesion force
measurements between CBM and substrate surfaces, the binding affinity of CBM to
lignin and cellulose was quantified. In addition, the effect of lignin type on
CBM binding affinity was discussed. Poisson statistical analysis model was then
applied to the data to decouple the adhesion force into specific and
nonspecific components. Based on the quantitative comparison of specific and
non-specific forces between CBM and substrate surfaces, the dominant type of
forces effective in enzymatic binding to cellulose and lignin were determined.

Materials and Methods

CBM of Trichoderma reesei Cellobiohydrolase I (Cel7A)
was expressed in Escherichia coli (E. coli) BL21 (DE3) cells with
ketosteroid isomerase (KSI) tag using the pET31b(+) expression vector. The histidine-tagged
fusion protein was purified on Ni-NTA agarose column with 10-250 mM imidazole
gradient. CBM peptide was cleaved from KSI using cyanogen bromide. CBM was
further purified from KSI after cleavage using reversed-phase chromatography
with 0-60% acetonitrile gradient in 0.1% trifluoroacetic acid (TFA). The presence of CBM in
the eluent was confirmed by mass spectroscopy analysis. The fraction containing
CBM was lyophilized and resuspended in Tris-HCl (10mM, pH 7.5, 150 mM NaCl) buffer.

Modified organosolv, sulfite, and kraft pulping were used to
produce reference substrates from poplar. The chemical composition of the
reference substrates was analyzed via Technical Association of the Pulp
and Paper Industry (TAPPI) test methods (T236, T204, T222, T211, and T249).
Surface lignin coverage was characterized by X-ray photoelectron spectroscopy
(XPS). Prior to force measurements, fibers of reference lignocellulosic
substrates were attached to silicon wafers coated with poly-L-lysine (PLL). AFM
cantilevers were functionalized with CBM and interaction forces between
substrate surfaces and CBM were collected in 50 mM sodium acetate buffer at pH
4.8 under
ambient temperature and humidity. A 5 x 5 μm² area of substrate
surface was scanned at a 1 Hz rate. Spatially
collected force-distance retraction curves were processed using the Nanoscope
Analysis 1.5 software to determine the number of adhesion events and the
magnitudes of adhesion forces measured between CBM and substrates'
surfaces. Poisson statistical model was applied to adhesion force data to
decouple the total interaction force into specific (hydrogen bonding) and
nonspecific (Lifshitz-van der Waals and hydrophobic forces) components.

Results and Discussion

Our results showed that both hydrogen bonding and
hydrophobic interactions contribute to enzymatic binding to lignocellulosic
substrate surfaces. As the surface lignin coverage increases on the substrates,
the overall adhesion forces of CBM to substrate surface increases in a linear manner
(Figure 1). The slope of the linear line for kraft pulped (KP) substrates is 1.5
folds higher than that for the (sulfite pulped) SP substrates. Although we
don't have a third point for the organosolv pulped (OP) substrates, based on
the behavior of two data points obtained for the OP substrates, the trend
observed for the KP and SP substrates is most likely valid for the OP
substrates (Figure 1) . This suggests that the effect of kraft lignin on
increasing the binding forces of CBM is higher than the effect of
lignosulfonates and organosolv lignin on such interactions. Lignin-rich
substrates showed higher binding strength to CBM in comparison to lignin-free
substrates. This suggests that the affinity of CBM to lignin is higher than
cellulose.

In addition to overall adhesion forces, the decoupled
specific and nonspecific forces were compared. As the surface lignin coverage
increases, the nonspecific forces increased for all types of pulped substrates.
However, substrates with kraft lignin had 1.3 and 1.4 folds higher nonspecific
forces than the substrates with lignosulfonates and organosolv lignin,
respectively. This suggests that in the presence of hydrophobic lignin (kraft
lignin) on substrates' surfaces, hydrophobic interactions contribute
significantly to the overall interactions with CBM in comparison to other types
of lignins investigated. When the specific and nonspecific forces were compared,
it was found that the specific forces are higher than the nonspecific forces. The
specific forces of lignin-rich substrates derived from kraft pulping, sulfite
pulping and organosolv pulping were 1.3, 1.6 and 1.4 folds higher than the
nonspecific forces of the same substrates, respectively. Having 1.6 fold greater
specific forces for the lignosulfonates suggest that hydrogen bonding
interactions are more dominant in comparison to kraft lignin and organosolv
lignin. Additionally, as the surface lignin coverage increases, the specific
forces increase. This suggests that hydrogen bonding interactions involved in
nonproductive binding of CBM to lignin. In summary, both hydrogen bonding and
hydrophobic interactions contribute to enzymatic binding to lignocellulosic substrate
surfaces. However, the chemical structure of the
lignin available on the substrate surface determines the type of interactions
that dominate how lignocellulosic surfaces interact with other surfaces of
interest.

Conclusion

Understanding the types of interaction forces that dominate
the binding of enzymes to the biomass surface or specific chemical groups will
allow protein engineers to design new enzymes that can specifically target
cellulose groups in the biomass. In addition, by applying specific pretreatment
technologies capable of reducing the hydrophobicity of lignin and reducing the
number of chemical groups in lignin structure that are involved in hydrogen
bonding, nonspecific binding of cellulase enzyme to lignin can be minimized.

Acknowledgment

This work is supported by National Science Foundation (NSF),
Grant number: 1067012. We gratefully acknowledge Dr. Dmitri Tolkatchev and
Prof. Alla Kostyukova from the Gene and Linda Voiland School of Chemical
Engineering and Bioengineering at Washington State University, for their
valuable help in CBM expression and purification.

Figure 1. Adhesion forces of
CBM to reference substrates as a function of surface lignin coverage. R²
of linear regression lines are 0.99 and 0.97 for KP and SP substrates,
respectively. Errors bars indicate the standard error of the mean.