Cellulase Mixture Research Articles

Rational approach to optimize cellulase mixtures for hydrolysis of regenerated cellulose containing residual ionic liquid

For the efficient production of glucose for platform chemicals or biofuels, cellulosic biomass is pretreated and subsequently hydrolyzed with cellulases. Although ionic liquids (IL) are known to effectively pretreat cellulosic biomass, the hydrolysis of IL pretreated biomass has not been optimized so far. Here, we present a semi-empirical model to rationally optimize the hydrolysis of pretreated α-cellulose – regenerated from IL and containing residual IL from the pretreatment. First, the influence of the IL MMIM DMP on the individual cellulases endoglucanase I, cellobiohydrolase I and β-glucosidase was investigated. Second, an enzyme loading-dependent model was developed to describe kinetics for the individual cellulases and cellulase mixtures. Third, this model was used to optimize the cellulase mixture for the efficient hydrolysis of regenerated cellulose containing residual IL. Finally, we could significantly increase the initial hydrolysis rate in 10% (v/v) MMIM DMP by 49% and the sugar yield by 10% points.

Ensiling and in vitro digestibility characteristics of Ceratoides arborescens treated with lactic acid bacteria inoculants and cellulase

The effects of different additives, including lactic acid bacteria inoculants, acremonium cellulase and mixtures of lactic acid bacteria inoculants and acremonium cellulase on changes of nutritive components and in vitro digestibility of Ceratoides arborescens during ensiling were investigated. The silages of each treatment were prepared in plastic film bags, kept in an incubator at 30°C and bags were opened after 1, 3, 5, 10, 30 and 45 days at room temperature. The results show that addition of additives and ensiling time did not affect acid detergent lignin content during ensiling (P>0.05). With ensiling time, pH value decreased ( P<0.05), whereas the concentrations of lactic acids, acetic acid, propionic acid, butyric acid and ammonia nitrogen increased (P<0.05). At 45 days of fermentation, the silages treated with the mixture of 50 mg lactic acid bacteria and 200 mg acremonium cellulase showed the lowest ( P<0.05) pH value, neutral detergent fiber and acid detergent fiber contents, and the highest (P<0.05) in vitro digestibility of neutral detergent fiber and acid detergent fiber, water soluble carbohydrate and lactic acids concentrations compared to other treatments. Lactic acid bacteria can improve C. arborescens silage fermentation quality; cellulase had a positive effect on in vitro digestible of neutral detergent fiber. There was favorable interaction between the addition of lactic acid bacteria and cellulase on the silage fermentation quality and chemical composition of C. arborescens .

Cellobiose Dehydrogenase and a Copper-Dependent Polysaccharide Monooxygenase Potentiate Cellulose Degradation by Neurospora crassa

The high cost of enzymes for saccharification of lignocellulosic biomass is a major barrier to the production of second generation biofuels. Using a combination of genetic and biochemical techniques, we report that filamentous fungi use oxidative enzymes to cleave glycosidic bonds in cellulose. Deletion of cdh-1, the gene encoding the major cellobiose dehydrogenase of Neurospora crassa, reduced cellulase activity substantially, and addition of purified cellobiose dehydrogenases from M. thermophila to the Δcdh-1 strain resulted in a 1.6- to 2.0-fold stimulation in cellulase activity. Addition of cellobiose dehydrogenase to a mixture of purified cellulases showed no stimulatory effect. We show that cellobiose dehydrogenase enhances cellulose degradation by coupling the oxidation of cellobiose to the reductive activation of copper-dependent polysaccharide monooxygenases (PMOs) that catalyze the insertion of oxygen into C-H bonds adjacent to the glycosidic linkage. Three of these PMOs were characterized and shown to have different regiospecifities resulting in oxidized products modified at either the reducing or nonreducing end of a glucan chain. In contrast to previous models where oxidative enzymes were thought to produce reactive oxygen species that randomly attacked the substrate, the data here support a direct, enzyme-catalyzed oxidation of cellulose. Cellobiose dehydrogenases and proteins related to the polysaccharide monooxygenases described here are found throughout both ascomycete and basidiomycete fungi, suggesting that this model for oxidative cellulose degradation may be widespread throughout the fungal kingdom. When added to mixtures of cellulases, these proteins enhance cellulose saccharification, suggesting that they could be used to reduce the cost of biofuel production.

High-performance hydrolysis of wheat straw using cellulase and thermomechanical pretreatment

Wheat straw was ground and pretreated with a thermomechanical process before exposure to a mixture of cellulases (Celluclast). The thermomechanical process known as DIC (Détente Instantanée Contrôlée), which is similar to the steam explosion process, was used to pretreat the lignocellulosic biomass through a flash expansion. The results demonstrate that crushing the wheat straw increased the hydrolysis yields. The initial hydrolysis rates and the final yields of hydrolysis increased relative to those of the untreated straw as a result of the DIC pretreatment. After optimization, the hydrolysis process resulted in a glucose yield of 31g/100g of straw, which corresponds to approximately 90% of the theoretical cellulose value. A fermentation procedure was performed using the microorganism Saccharomyces cerevisiae, and the results confirmed the production of bioethanol (11.5g/100g of pretreated straw). The adsorption isotherms of cellulases on wheat straw and the enzyme kinetics of hydrolysis of the wheat straw were also studied using different concentrations of Celluclast. The kinetic parameters (maximal velocity, Vemax, and half-saturation constant, Ke) were determined from the initial velocities. Kad (adsorption coefficient) results show that the DIC pretreatment of the straw decreased Kad six-fold compared to that of the untreated straw (15.54 vs 2.52g of enzymes/g of biomass, respectively), confirming that the DIC pretreatment produced a substrate that is more accessible to cellulases.

Scaled-up separation of cellobiohydrolase1 from a cellulase mixture by ion-exchange chromatography

Enzymatic hydrolysis of cellulose often involves cellulases produced by Trichoderma reesei, of which cellobiohydrolase1 (CBH1) is the most abundant (about 60% of total cellulases) and plays an important role in the hydrolysis of crystalline cellulose. A method for separating sufficient quantities from the bulk cellulase cocktail is highly desirable for many studies, such as those that aim to characterize binding and hydrolysis kinetics of CBH1. In this work, CBH1 was separated from other Spezyme CP cellulases by ion-exchange chromatography using an efficient modification of a smaller scale process. The ion-exchange column was connected to a vacuum manifold system to provide a steady flow through parallel columns and thus achieve scale-up for enzyme separation. With five 5-mL columns running in parallel, about 55 mg of CBH1 was separated from 145 mg of Spezyme CP in a single separation. Step elution was used to replace the continuous gradient used at smaller scale. The purified CBH1 was collected in the fraction eluted with a buffer containing 0.33 M salt and showed comparable purity and activity as the enzyme purified by a fast protein liquid chromatography system. The stability of separated CBH1 was studied for up to 2 days and good thermal stability was observed. Separated CBH1 also showed both high adsorption to bacterial microcrystalline cellulose with ~4 μmol/g maximum adsorption and a K(a) of 5.55 ± 2.34 μM(-1) , and good hydrolytic activity based on atomic force microscopy observations that show a reduction in fiber height.

Pilot scale enzymatic deinking of mixed office wastepaper and old newspaper

The performance of a newly developed pilot scale continuous enzymatic deinking system has been evaluated using a mixture of cellulase and xylanase enzymes in the deinking of mixed office wastepaper (MOW) and old newspaper (ONP). Optimizations of the enzymatic deinking processes were carried out, and the optimum conditions obtained for MOW and ONP were different. The highest brightness obtained from enzymatic deinking of MOW and ONP under their respective optimum conditions were about 83.6% and 41.9%, respectively. The deinking efficiency of 6.0% and 6.3% were obtained by enzymatic deinking process using MOW and ONP, respectively. On the other hand, the deinking efficiency of 2.9% and 3.5% were obtained by a chemical deinking process using MOW and ONP, respectively. The findings obtained from present work indicated that enzymes have potential to be used in deinking of MOW, which is difficult to deink by conventional chemical deinking methods.

A minimal set of bacterial cellulases for consolidated bioprocessing of lignocellulose

Cost-effective release of fermentable sugars from non-food biomass through biomass pretreatment/enzymatic hydrolysis is still the largest obstacle to second-generation biorefineries. Therefore, the hydrolysis performance of 21 bacterial cellulase mixtures containing the glycoside hydrolase family 5 Bacillus subtilis endoglucanase (BsCel5), family 9 Clostridium phytofermentans processive endoglucanase (CpCel9), and family 48 C. phytofermentans cellobiohydrolase (CpCel48) was studied on partially ordered low-accessibility microcrystalline cellulose (Avicel) and disordered high-accessibility regenerated amorphous cellulose (RAC). Faster hydrolysis rates and higher digestibilities were obtained on RAC than on Avicel. The optimal ratios for maximum cellulose digestibility were dynamic for Avicel but nearly fixed for RAC. Processive endoglucanase CpCel9 was the most important for high cellulose digestibility regardless of substrate type. This study provides important information for the construction of a minimal set of bacterial cellulases for the consolidated bioprocessing bacteria, such as Bacillus subtilis, for converting lignocellulose to biocommodities in a single step.

Quantitative Proteomic Approach for Cellulose Degradation by Neurospora crassa

Conversion of plant biomass to soluble sugars is the primary bottleneck associated with production of economically viable cellulosic fuels and chemicals. To better understand the biochemical route that filamentous fungi use to degrade plant biomass, we have taken a quantitative proteomics approach to characterizing the secretome of Neurospora crassa during growth on microcrystalline cellulose. Thirteen proteins were quantified in the N. crassa secretome using a combination of Absolute Quantification (AQUA) and Absolute SILAC to verify protein concentrations. Four of these enzymes including 2 cellobiohydrolases (CBH-1 and GH6-2), an endoglucanase (GH5-1), and a β-glucosidase (GH3-4) were then chosen to reconstitute a defined cellulase mixture in vitro. These enzymes were assayed alone and in mixtures and the activity of the reconstituted set was then compared to the crude mixture of N. crassa secretome proteins. Results show that while these 4 proteins represent 63-65% of the total secretome by weight, they account for just 43% of the total activity on microcrystalline cellulose after 24 h of hydrolysis. This result and quantitative proteomic data on other less abundant proteins secreted by Neurospora suggest that proteins other than canonical fungal cellulases may play an important role in cellulose degradation by fungi.

A mechanistic model for rational design of optimal cellulase mixtures

A model-based framework is described that permits the optimal composition of cellulase enzyme mixtures to be found for lignocellulose hydrolysis. The rates of hydrolysis are shown to be dependent on the nature of the substrate. For bacterial microcrystalline cellulose (BMCC) hydrolyzed by a ternary cellulase mixture of EG2, CBHI, and CBHII, the optimal predicted mixture was 1:0:1 EG2:CBHI:CBHII at 24 h and 1:1:0 at 72 h, at loadings of 10 mg enzyme per g substrate. The model was validated with measurements of soluble cello-oligosaccharide production from BMCC during both single enzyme and mixed enzyme hydrolysis. Three-dimensional diagrams illustrating cellulose conversion were developed for mixtures of EG2, CBHI, CBHII acting on BMCC and predicted for other substrates with a range of substrate properties. Model predictions agreed well with experimental values of conversion after 24 h for a variety of enzyme mixtures. The predicted mixture performances for substrates with varying properties demonstrated the effects of initial degree of polymerization (DP) and surface area on the performance of cellulase mixtures. For substrates with a higher initial DP, endoglucanase enzymes accounted for a larger fraction of the optimal mixture. Substrates with low surface areas showed significantly reduced hydrolysis rates regardless of mixture composition. These insights, along with the quantitative predictions, demonstrate the utility of this model-based framework for optimizing cellulase mixtures.

Proteomics-based Compositional Analysis of Complex Cellulase–Hemicellulase Mixtures

Efficient deconstruction of cellulosic biomass to fermentable sugars for fuel and chemical production is accomplished by a complex mixture of cellulases, hemicellulases, and accessory enzymes (e.g., >50 extracellular proteins). Cellulolytic enzyme mixtures, produced industrially mostly using fungi like Trichoderma reesei, are poorly characterized in terms of their protein composition and its correlation to hydrolytic activity on cellulosic biomass. The secretomes of commercial glycosyl hydrolase-producing microbes was explored using a proteomics approach with high-throughput quantification using liquid chromatography-tandem mass spectrometry (LC-MS/MS). Here, we show that proteomics-based spectral counting approach is a reasonably accurate and rapid analytical technique that can be used to determine protein composition of complex glycosyl hydrolase mixtures that also correlates with the specific activity of individual enzymes present within the mixture. For example, a strong linear correlation was seen between Avicelase activity and total cellobiohydrolase content. Reliable, quantitative and cheaper analytical methods that provide insight into the cellulosic biomass degrading fungal and bacterial secretomes would lead to further improvements toward commercialization of plant biomass-derived fuels and chemicals.

Inhibition of enzymatic hydrolysis by residual lignins from softwood—study of enzyme binding and inactivation on lignin‐rich surface

Lignin-derived inhibition is a major obstacle restricting the enzymatic hydrolysis of cell wall polysaccharides especially with softwood lignocellulosics. Enzyme adsorption on lignin is suggested to contribute to the inhibitory effect of lignin. The interaction of cellulases with softwood lignin was studied in the present work with commercial Trichoderma reesei cellulases (Celluclast) and lignin-rich residues isolated from steam pretreated softwood (SPS) by enzymatic and acid hydrolysis. Both lignin preparations inhibited the hydrolysis of microcrystalline cellulose (Avicel) and adsorbed the major cellulases present in the commercial cellulase mixture. The adsorption phenomenon was studied at low temperature (4°C) and at the typical hydrolysis temperature (45°C) by following activities of free and lignin-bound enzymes. Severe inactivation of the lignin-bound enzymes was observed at 45°C, however at 4°C the enzymes retained well their activity. Furthermore, SDS-PAGE analysis of the lignin-bound enzymes indicated that very strong interactions form between the residue and the enzymes at 45°C, because the enzymes were not released from the residue in the electrophoresis. These results suggest that heat-induced denaturation may take place on the surface of softwood lignin at the hydrolysis temperature.

Equilibrium Water Contents of Cellulose Films Determined via Solvent Exchange and Quartz Crystal Microbalance with Dissipation Monitoring

Model cellulose surfaces have attracted increasing attention for studying interactions with cell wall matrix polymers and as substrates for enzymatic degradation studies. Quartz crystal microbalance with dissipation monitoring (QCM-D) solvent exchange studies showed that the water content of regenerated cellulose (RC) films was proportional to the film thickness (d) and was consistent with about five water molecules per anhydroglucose unit. Sulfated nanocrystalline cellulose (SNC) and desulfated nanocrystalline cellulose (DNC) films had comparable water contents and contained about five times more water than RC films. A cellulase mixture served as a probe for studies of substrate accessibility and degradation. Cellulase adsorption onto RC films was independent of d, whereas degradation times increased with d. However, adsorption onto SNC and DNC films increased with d, whereas cellulase degradation times for DNC films were independent of studied d. Enhanced access to guest molecules for SNC and DNC films revealed they are more porous than RC films.

Synergistic action of xylanase and mannanase improves the total hydrolysis of softwood

The impact of xylan and glucomannan hydrolysis on cellulose hydrolysis was studied on five pretreated softwood substrates with different xylan and glucomannan contents, both varying from 0.2% to 6.9%, using mixtures of purified enzymes.The supplementation of pure cellulase mixture with non-specific endoglucanase TrCel7B and xylanase TrXyn11 enhanced the hydrolysis of all substrates, except the steam pretreated spruce, by more than 50%. The addition of endo-β-mannanase increased the overall hydrolysis yield by 20–25%, liberating significantly more glucose than theoretically present in glucomannan.When supplemented together, xylanolytic and mannanolytic enzymes acted synergistically with cellulases. Moreover, a linear correlation was observed between the hydrolysis of polysaccharides, irrespective of the composition, indicating that glucomannan and xylan form a complex network of polysaccharides around the cellulosic fibres extending throughout the lignocellulosic matrix. Both hemicellulolytic enzymes are crucial as accessory enzymes when designing efficient mixtures for the total hydrolysis of lignocellulosic substrates containing both hemicelluloses.

Influence of particle size and multi-enzyme supplementation of fibrous diets on total tract digestibility and performance of weaning (8–20 kg) and growing (20–40 kg) pigs

The aim of the study was to evaluate the effects of fibre source (cassava residue, CR and sweet potato vines, SPV), particle size (1 mm and 3 mm screen size) and multi-enzyme supplementation (mixture of α-amylase, β-glucanase, cellulase and protease) on growth performance and the coefficient of total tract apparent digestibility (CTTAD) in weaning and growing pigs (Landrace × Yorkshire). The experimental diets were based on maize, fish meal, full fat soybean, soybean meal, soybean oil and CR or SPV. The dietary levels of CR and SPV were 200 and 350 g/kg (as fed) in the post-weaning and growing period, respectively. The experiment was arranged according to a 2 × 2 × 2 factorial completely randomized design with five replications, and lasted 60 days with two periods: from 30 to 60 days of age (post-weaning) and 60 to 90 days of age (growing). In each period, pigs fed the CR diet had higher CTTAD of dietary components (except for ether extract), average daily gain and lower feed conversion ratio than those fed the SPV diet (P<0.05). The reduction in particle size increased the CTTAD of dietary components and average daily gain in the post-weaning period (P<0.05), but not in the growing period (P>0.05). Addition of the enzyme mixture improved the CTTAD and growth performance in the post-weaning period (P<0.05). There was an interaction between particle size and multi-enzyme supplementation on the CTTAD of crude protein and neutral detergent fibre (P<0.05) in the post-weaning period, such that multi-enzyme supplementation increased the CTTAD of CP and NDF in the larger particle size diet (0.80 versus 0.75, and 0.58 versus 0.51, respectively), while there was no changes in the small particle size diet (0.80 versus 0.78, and 0.59 versus 0.55, respectively). However, in the growing period multi-enzyme supplementation had no positive effect on the performance and the CTTAD of dietary components, with the exception of the CTTAD of crude protein and neutral detergent fibre. In conclusion, a reduction in particle size and multi-enzyme supplementation improved the CTTAD of dietary components and the growth performance in the post-weaning period, but not in the growing period.

Compatible Ionic liquid‐cellulases system for hydrolysis of lignocellulosic biomass

Ionic liquids (ILs) have been increasingly recognized as novel solvents for dissolution and pretreatment of cellulose. However, cellulases are inactivated in the presence of ILs, even when present at low concentrations. To more fully exploit the benefits of ILs it is critical to develop a compatible IL-cellulases system in which the IL is able to effectively solubilize and activate the lignocellulosic biomass, and the cellulases possess high stability and activity. In this study, we investigated the stability and activity of a commercially available cellulases mixture in the presence of different concentrations of 1-ethyl-3-methylimidazolium acetate ([Emim][OAc]). A mixture of cellulases and β-glucosidase (Celluclast1.5L, from Trichoderma reesei, and Novozyme188, from Aspergillus niger, respectively) retained 77% and 65% of its original activity after being pre-incubated in 15% and 20% (w/v) IL solutions, respectively, at 50°C for 3 h. The cellulases mixture also retained high activity in 15% [Emim][OAc] to hydrolyze Avicel, a model substrate for cellulose analysis, with conversion efficiency of approximately 91%. Notably, the presence of different amounts of yellow poplar lignin did not interfere significantly with the enzymatic hydrolysis of Avicel. Using this IL-cellulase system (15% [Emim][OAc]), the saccharification of yellow poplar biomass was also significantly improved (33%) compared to the untreated control (3%) during the first hour of enzymatic hydrolysis. Together, these findings provide compelling evidence that [Emim][OAc] was compatible with the cellulase mixture, and this compatible IL-cellulases system is promising for efficient activation and hydrolysis of native biomass to produce biofuels and co-products from the individual biomass components.

Enzymatic Treatment as a Pre‐Step to Remove Cellulose Films from Sensors

AbstractIn this work an enzymatic treatment is proposed as a preparative, cleaning protocol to remove cellulose films from resonators and sensors. Quartz crystal and surface plasmon gold sensors, coated with ultrathin films of cellulose are used in studies of molecular (for example, polymer and surfactant) adsorption. The sensors are usually recycled after removal of the film, with limited success, after one of two treatments, either hot acid or ammoniac solutions. In the proposed, improved protocol a mixture of cellulases from Aspergillus species, are used as a pre‐treatment to facilitate the release of the cellulose film from the surfaces of the sensors. Two concentrations of NaCl solutions were considered in the enzymatic treatment, 1 and 10 mM, at given enzyme solution concentration, temperature and pH. It was found that after 80 min, the water contact angle after treatment with both salt concentration conditions reached a plateau. The average water contact angle after integration of the enzymatic and ammoniac treatments was found to be low enough, between 6.4 and 7.1 deg to allow reuse the sensors. It is concluded that the use of the ammoniac cleaning solution after the enzymatic treatment is a very convenient, safe and less time consuming way to remove the cellulose films from the sensors to be recycled.

Screening of cellulases for biofuel production: Online monitoring of the enzymatic hydrolysis of insoluble cellulose using high‐throughput scattered light detection

A new prospective cellulase assay simultaneously combining high-throughput, online analysis and insoluble cellulosic substrates is described. The hydrolysis of three different insoluble cellulosic substrates, catalysed by a commercial cellulase preparation from Trichoderma reesei (Celluclast), was monitored using the BioLector - allowing online monitoring of scattered light intensities in a continuously shaken microtiter plate. Cellulase activities could be quantitatively assayed using the BioLector. At low cellulase/cellulose ratios, the Michaelis-Menten parameters of the cellulase mixture were mainly affected by the crystallinity index of the cellulose. Here, the apparent maximum cellulase activities inversely correlated with the crystallinity index of the cellulose. At high cellulase/cellulose ratios the particle size of the cellulose, defining the external surface area accessible to the cellulases, was the key determining factor for cellulase activity. The developed technique was also successfully applied to evaluate the pH optimum of cellulases. Moreover, the non-hydrolytic deagglomeration of cellulose particles was investigated, for the first time, using high-throughput scattered light detection. In conclusion, this cellulase assay ideally links high-throughput, online analysis and realistic insoluble cellulosic substrates in one simple system. It will considerably simplify and accelerate fundamental research on cellulase screening.

Biological pretreatment of cellulose: Enhancing enzymatic hydrolysis rate using cellulose-binding domains from cellulases

In this study, cellulose-binding domains (CBDs) of cellulases from Trichoderma reesei were used in a pretreatment step and were found to effectively reduce the crystallinity of cellulose (both Avicel and fibrous cellulose). This, in turn, led to higher glucose concentrations (up to 25% increase) in subsequent hydrolysis of cellulose using a mixture of cellulases and without the need for any intermediate purification step. CBDs were shown to be active in a range of temperatures (up to 50°C), while cellulase hydrolytic activity was greatly reduced after incubation at 50°C. This was explained by retention of full binding capacity after incubation at 50°C for 15 h. Our findings suggest that CBDs may be a valuable tool in pretreating cellulose and eventually afford faster enzymatic conversion of cellulose to glucose, thus contributing to more affordable processes in the production of biofuels.

Enzyme Treatments to Enhance Oil Recovery from Condensed Corn Distillers Solubles

AbstractThe objective of this study was to determine the effect of enzyme hydrolysis of various corn components on oil recovery from condensed corn distillers solubles (CCDS). Hydrolysis with a commercial protease significantly increased oil recovery as the enzyme concentration increased, with the greatest oil recovery being 70% at 10% v/w (dry weight basis) enzyme concentration. Increasing centrifugal force from 8,500 to 12,240×g was only slightly effective for the non‐enzyme treated samples. Reducing CCDS particle size by grinding with a mortar and pestle increased oil recovery to 83% when an enzyme combination of a commercial cellulase mixture and a protease was used. Particle size reduction of CCDS by high‐speed blending resulted in low oil recovery, but the oil recovery was significantly improved after enzyme treatment. Zein‐lipid interaction was very strong when tested in a model system, with only 10% of the oil being freed by centrifugation alone. Following enzyme hydrolysis of the zein‐oil complex with a protease, oil recovery was increased to 97%. Overall, enzyme hydrolysis and further particle size reduction showed a small, albeit statistically significant, effect in increasing oil recovery from CCDS. These small increases may not justify the use of enzymes or processing modifications to reduce particle size in the ethanol industry, nonetheless, these data may provide a reference or insight to design more effective treatments for oil recovery.

Rhamnolipids as affinity foaming agent for selective collection of β-glucosidase from cellulase enzyme mixture

Selective and effective separation can potentially be achieved with affinity foam fractionation using simple foaming setup and operation. In this study the use of affinity foam fractionation for selective collection and enrichment of β-glucosidase from a cellulase enzyme mixture was evaluated. Rhamnolipids, a group of glycolipids produced most commonly by Pseudomonas aeruginosa, were used as the affinity foaming agent, because of their foaming property and the presence of dirhamnose moiety (a potential substrate analog for β-glucosidase) in some rhamnolipids. The effects of aeration rate, medium pH, cellulase concentration and rhamnolipid concentration on the foam fractionation performance were examined. Among the pH studied (3.1, 5.0, 7.0 and 9.0), pH 5 was clearly the optimal for selective enrichment of β-glucosidase, presumably corresponding to the high binding affinity between the enzyme and the substrate analog (dirhamnose). With adequate rhamnolipid concentrations (≥0.1 g/L), the aeration rate of 0.1 L/min (i.e., 2 VVM) for 50 ml test samples was found to give the highest enrichment; higher aeration rates produced wetter foam and, thus, lower (diluted) enzyme activity in the foamate. The enrichment increased with the increasing rhamnolipids-to-cellulase ratio, in the range of 0–2 (w/w) investigated in this study. The finding indicated that rhamnolipids were the limiting compounds in these systems so that the amount of surfactant–enzyme complexes formed and removed into the foam phase would increase when more rhamnolipids were added. At the rhamnolipids-to-cellulase ratio of 2, the β-glucosidase activity in the foamate was about 9 times as high as the activity in the original sample of cellulase mixture and about 17 times the activity in the remaining solution (after foaming). The overall FPU (filter paper unit, a measurement of total cellulase activity) and the activities of endo- and exo-glucanases were only enriched 70–150%. The feasibility of affinity foam fractionation was demonstrated.