Appl Biochem Biotechnol DOI 10.1007/s12010-015-1662-7
A NaBH4 Coupled Ninhydrin-Based Assay for the Quantification of Protein/Enzymes During the Enzymatic Hydrolysis of Pretreated Lignocellulosic Biomass Yiu Ki Mok 1 & Valdeir Arantes 2 & Jack N. Saddler 1
Received: 13 February 2015 / Accepted: 7 May 2015 # Springer Science+Business Media New York 2015
Abstract Accurate protein quantification is necessary in many of the steps during the enzymatic hydrolysis of pretreated lignocellulosic biomass, from the fundamental determination of enzyme kinetics to techno-economic assessments, such as the use of enzyme recycling strategies, evaluation of enzyme costs, and the optimization of various process steps. In the work described here, a modified, more rapid ninhydrinbased protein quantification assay was developed to better quantify enzyme levels in the presence of lignocellulosic biomass derived compounds. The addition of sodium borohydride followed by acid hydrolysis at 130 °C greatly reduced interference from monosaccharides and oligosaccharides and decreased the assay time 6-fold. The modified ninhydrin assay was shown to be more accurate as compared to various traditional colorimetric protein assays when commercial cellulase enzyme mixtures were quantified under typical pretreated lignocellulosic biomass enzymatic hydrolysis conditions. The relatively short assay time and microplate-reading capability of the modified assay indicated that the method could likely be used for high-throughput protein determination. Keywords Lignocellulose . Ninhydrin . Protein quantification . Cellulase enzymes . Enzymatic hydrolysis
* Jack N. Saddler [email protected] 1
Forest Products Biotechnology/Bioenergy Group, Faculty of Forestry, University of British Columbia, 2424 Main Mall, Vancouver, BC V6T 1Z4, Canada
2
Department of Biotechnology, Lorena School of Engineering, University of São Paulo, CP 116, Rod. Itajubá—Lorena, km 74,5 Área I, Estrada Municipal do Campinho S/N 2602-810, Lorena, SP, Brazil
Appl Biochem Biotechnol
Introduction The cost-effective production of sugars from lignocellulosic biomass continues to be a challenge for the bioconversion/biorefining industries. This is primarily due to the costs associated with the large amount of enzyme/proteins needed to achieve effective cellulose hydrolysis to monomeric sugars [1]. Thus, a considerable amount of effort has been invested in decreasing enzyme production costs, increasing enzyme performance, and developing enzyme recycling strategies. The high enzyme/protein loading required to achieve effective hydrolysis has also been shown to be a major limitation in a number of techno-economic analyses of biomass-to-ethanol processes [2, 3]. Therefore, to try to better assess the efficacy of these approaches from both a fundamental enzyme kinetic and techno-economic perspective, an accurate method of assessing cellulase activity/performance is needed. Traditionally, the filter paper activity (FPA) assay, as recommended by the International Union of Pure and Applied Chemistry (IUPAC), is often employed to evaluate cellulase activity/performance [4]. The filter paper activity assay measures cellulase activity on a volumetric basis by determining the amount of cellulase enzymes required to solubilize 4 % of a 50 mg Whatman #1 filter paper strip in 60 min. Although widely used, it is well known that in addition to the composition of the enzyme preparation, substrate characteristics also play a significant role in influencing enzymatic performance [4–6]. As a result, the filter paper activity assay is often not reflective of the hydrolytic performance of a particular enzyme mixture on different pretreated lignocellulosic substrate. Due to the inherent limitations of the filter paper assay, many researchers have supplemented enzymatic activity-based assays with total protein quantification assays as a preferred method of assessing enzyme performance [5, 7]. Unfortunately, the selection of a total protein quantification assay for the accurate determination of total protein concentration in a Bcellulase mixture^ remains challenging, in part due to the complexity of a cellulase mixture, which contain both cellulases and other accessory enzymes and the extent of glycosylation of each of the enzymes [5, 8]. In a previous work by Adney et al. (1995), the protein concentration of multiple commercial cellulase preparations were determined using the modified Lowry, Bradford, and Bicinchoninic acid (BCA) assays as well as Kjeldahl Nitrogen Analysis and UV absorbance at 280 nm [9]. These workers found that, depending on the protein quantification method used, the total protein concentration varied by as much as 90 %. Similar results were reported by McMillan et al. (2011) where the protein concentration of four cellulase preparations varied by 2.6–4.8 times when measured by either the Bradford or BCA assays [7]. This large variability in accurately determining protein concentrations when utilizing different quantification methods is of significant concern, particularly when enzyme activities and kinetics are being evaluated. The enzymatic hydrolysis of lignocellulosic biomass also results in the production of monomeric and oligomeric sugars, which are known to interfere with many protein assays [10–12]. As past work was often carried out at low substrate concentrations and high enzyme loadings, the interfering effects of any released sugars could simply be resolved by diluting the samples below interfering sugar concentrations. However, with the increased hydrolytic efficiency of newer cellulase preparation and the need to move to higher substrate concentrations, higher sugar-to-protein concentration ratios can be expected. As a result, sample dilution is often no longer a viable way of reducing the interfering effects of sugars. To further complicate matters, other compounds that are present or released during biomass pretreatment
Appl Biochem Biotechnol
and hydrolysis, such as phenolics, surfactants, and sugar degradation products (furfural, 5hydroxymethylfurfural (HMF), levulinic acid), are also know to interfere with many protein assays [10, 11, 13–19]. As mentioned earlier, the modified Lowry, Bradford, and Bicinchoninic acid (BCA) assays as well as Kjeldahl Nitrogen Analysis and UV absorbance have all been previously used, with varying degrees of success, to try to accurately quantify cellulase mixtures. In relative contrast, the ninhydrin assay has shown promise due to its specificity for amino acids and decreased susceptibility to interference [20–24]. However, high concentrations of sugars have been shown to be problematic, as were the long hydrolysis times typically required to attain complete protein hydrolysis [10, 24]. Other workers have also reported wide-ranging standard deviations [25]. In the work described below, a modified, accurate, and rapid ninhydrin assay was developed and used to quantify the enzyme/protein present during enzymatic hydrolysis of biomass at relatively low protein and high substrate/sugar concentrations. A 6-fold reduction in total assay time could be achieved by increasing hydrolysis temperatures from the traditional 100 to 130 °C while the interfering influence of released sugars could be greatly reduced by sodium borohydride treatment.
Materials and Methods Traditional Acidic Ninhydrin Assay The total protein concentration was measured using the traditional ninhydrin assay following acid hydrolysis calibrated using bovine serum albumin (BSA, Sigma) as a protein standard. One hundred microliters of each protein sample (protein concentration between 0 and 800 μg/ g) was mixed with 50 μL of ultra-pure water in a 0.5 mL screw-cap microcentrifuge tube (Fisherbrand). Samples were diluted on a weight basis to minimize potential errors resulting from large dilution factors as some enzyme preparations contained high protein concentrations. This was followed by the addition of 300 μL of 9 M hydrochloric acid (HCl) to a final concentration of 6 M and subsequently heated in a dry heating bath (MBI Lab Equipment, Kirkland, PQ) at 100 °C for 24 h. After cooling to room temperature, 100 μL of the hydrolysate was transferred into a 1.5-mL microcentrifuge tube (Fisherbrand) and neutralized with 100 μL of 5 M sodium hydroxide (NaOH) (Fisher Scientific). Upon neutralization, 200 μL of the 2 % ninhydrin reagent (Sigma) was added and heated at 100 °C for 10 min. Samples were then cooled to room temperature prior to the addition of 500 μL of 50 % v/v ethanol. Subsequently, 200 μL of the colored solution was transferred to a 96-well microplate (Corning) and the absorbance was read at 560 nm by a microplate reader (PerkinElmer VICTOR3, Woodbridge, ON). All samples were performed in triplicate and experiments were repeated at least twice.
Evaluation of Glucose Interference on the Traditional Ninhydrin Assay To evaluate the possible interfering influence of glucose on the traditional ninhydrin assay, which is the predominant sugar produced during enzymatic hydrolysis of lignocellulosic biomass, BSA (100, 800, and 1600 μg protein/g) and glucose (0, 5, 10, 20, and 40 g/kg) (Sigma) were dissolved in 50 mM sodium acetate buffer (pH=4.8). Fifty microliters of the
Appl Biochem Biotechnol
protein solution and 50 μL of the sugar solution were mixed together, resulting in a final glucose concentration of 2.5, 5, 10, and 20 g/kg and BSA concentrations of 50, 400, and 800 μg/g, respectively. Fifty microliters of ultra-pure water was then added to the protein-sugar mixture. Alternatively, 50 mM sodium acetate buffer (pH=4.8) was added in place of the sugar solution for the 0 g/kg glucose condition. All samples were then quantified using the traditional ninhydrin assay.
Elimination of Glucose Interference with Sodium Borohydride As sodium borohydride (NaBH4) had been previously used to reduce monomeric sugars to sugar alcohols, we investigated the potential use of NaBH4 to remove the interference of sugars on the traditional ninhydrin assay [26]. One hundred microliters of a 20 g/kg glucose solution dissolved in 50 mM sodium acetate buffer (pH=4.8) was incubated with 50 μL of NaBH4 solution prepared by dissolving NaBH4 pellets (Sigma-Aldrich) in ultra-pure water followed by the addition of 0.02 %v/v Antifoam O-30 (Sigma-Aldrich). Three different NaBH 4 concentrations of 4, 6, and 13.3 g/kg were prepared, representing ratios of 1:10, 1:6.67, and 1:3 NaBH4/glucose w/w, respectively. Samples were incubated at room temperature and the reaction was stopped at 5, 15, 30, 60, and 120 min by the addition of 300 μL of 9 M HCl (final concentration of 6 M HCl). Following NaBH4 treatment, the samples were assayed by the traditional ninhydrin assay. Once the required incubation time and concentration of NaBH4 needed to remove glucose interference was determined, the same conditions were then used to assess the possible influence of hemicellulose derived sugars (arabinose, galactose, mannose, and xylose) at concentrations of 20 g/kg. As described in more detail in the BResults and Discussion^ section, the required ratio of NaBH4/glucose and incubation time required to completely eliminate sugar interference was determined to be 1:3 NaBH4/glucose w/w and 60 min. These conditions were subsequently used to develop the modified ninhydrin assay.
Influence of Sodium Borohydride on Protein Measurements To further investigate the possible influence of NaBH4 on the ninhydrin assay, BSA (0– 1600 μg/g) and glucose (40 g/kg) were dissolved in 50 mM sodium acetate buffer (pH= 4.8). Fifty microliters of the protein solution and 50 μL of the sugar solution were mixed together, resulting in a final glucose concentration of 20 g/kg and maximum BSA concentration of 800 μg/g. This protein-sugar mixture was then subjected to NaBH4 treatment as described above. As a comparison without the addition of NaBH4, 50 μL of the protein solution was mixed with 50 μL of 50 mM sodium acetate buffer followed by the addition of 50 μL of ultra-pure water. All samples were then subjected to acid hydrolysis at 100 °C for 24 h followed by quantification with the traditional ninhydrin assay. Upon establishing the effects of NaBH4 on BSA, the protein concentration of commercial cellulase preparations (Celluclast 1.5L, Cellic CTec2, Cellic CTec3) and commercial hemicellulase preparations (HTec, Multifect Xylanase) were also compared with and without the addition of NaBH4. The concentrations of sugars present in the enzyme preparations were measured using high-performance liquid chromatography (HPLC) (Dionex DX-3000, Sunnyvale, CA) as described elsewhere [6].
Appl Biochem Biotechnol
Reduction of Time Required for Complete Protein Hydrolysis As previous work had shown that the time required for complete protein hydrolysis can be reduced by increasing the hydrolysis temperature, we also investigated if the total hydrolysis time can be reduced by increasing the hydrolysis temperature from 100 to 130 °C [27]. Previous work has hydrolyzed protein at temperatures as high as 160 °C to reduce the hydrolysis time. However, these conditions often required specially designed equipment [28]. A temperature of 130 °C was thus selected as a compromise condition between hydrolysis time and the need for specialized equipment. BSA, Cellic CTec2, and HTec were incubated using the NaBH4 conditions described above and hydrolyzed with 9 M HCl (final concentration of 6 M HCl) at 100 and 130 °C. Samples were hydrolyzed in triplicates for 0, 2, 4, 6, 12, 18, and 24 h at 100 °C for 0, 15, 30, 60, 120, and 240 min at 130 °C. The extent of protein hydrolysis was quantified using the ninhydrin assay. Upon determining the required hydrolysis time to attain complete protein hydrolysis at 130 °C, the protein concentration of commercial cellulase preparations (Celluclast 1.5L, Cellic CTec2, Cellic CTec3) and commercial hemicellulases preparations (HTec, Multifect Xylanase) were compared between the traditional hydrolysis conditions of 24 h at 100 °C and the hydrolysis conditions at 130 °C. As described in more detail in the BResults and Discussion^ section, complete protein hydrolysis was attained with a reduced time of 2 h at 130 °C across all of the enzyme preparations. These conditions were subsequently used for the modified ninhydrin assay.
Modification of the Ninhydrin Assay to Quantify Protein During the Enzymatic Hydrolysis of High Concentrations of Biomass The modified ninhydrin method was developed to improve the overall accuracy, compatibility, and speed of the assay. One hundred microliters of protein containing samples (protein concentration between 0 and 800 μg/g) was first incubated with 50 μL of NaBH4 for 60 min at a ratio of 1:3 NaBH4/total sugar (w/w) in a 0.5-mL screw-cap microcentrifuge tube with BSA as a protein standard unless otherwise specified. This was followed by the addition of 300 μL of 9 M HCl (final concentration of 6 M HCl) and subsequent heating in a dry heating bath at 130 °C for 2 h. After cooling to room temperature, 100 μL of the hydrolysate was transferred to a 1.5-mL microcentrifuge tube and neutralized with 100 μL of 5 M NaOH. Upon neutralization, 200 μL of 2 % ninhydrin reagent was added and heated at 100 °C for 10 min. Samples were then cooled to room temperature prior to the addition of 500 μL of 50 %v/v ethanol. Subsequently, 200 μL of the colored solution was transferred to a 96-well microplate and the absorbance was read at 560 nm. All samples were performed in triplicate and experiments were repeated at least twice.
Compatibility of Modified Ninhydrin Assay with Compounds Encountered During Lignocellulosic Biomass Hydrolysis In addition to monosaccharides, lignocellulosic hydrolysates can also contain other compounds such as oligosaccharides and phenolic compounds. To determine the influence of these compounds, a commercial cellulase preparation, Cellic CTec3, was quantified using the modified ninhydrin assay described above after dilution in 50 mM sodium acetate buffer (pH= 4.8) or the water soluble fractions (WSFs) obtained from the SO2 catalyzed steam pretreatment of corn stover (SPCS), poplar (SPP), and lodgepole pine (SPLPP). Each substrate was
Appl Biochem Biotechnol
pretreated under near optimal conditions to maximize carbohydrate recovery while providing a cellulose-rich substrate that was more susceptible to enzymatic hydrolysis [6]. Compositional analyses of the water soluble fractions were carried out in duplicate following the National Renewable Energy Laboratory (NREL) protocols [29]. Additionally, the total amount of phenolics was measured in triplicate using the Lowry-Folin method with phloroglucinol as a standard [30]. These conditions were selected to mimic potential enzymatic hydrolysis conditions where high concentrations of phenolic derivatives and sugars would be present. After dilution, the final total sugar concentration in each enzyme/WSF mixture was always below 20 g/kg. Samples without NaBH4 treatment were also quantified to assess the degree of interference by the WSFs.
Elucidation of Interfering Effects of Individual Lignocellulosic Components To elucidate the individual interfering effects of different lignocellulosic components present in the WSFs, Cellic CTec3 was diluted in the different WSFs and quantified using the modified ninhydrin assay after three different treatments. NaBH4 treatment alone was used to isolate the effects of oligosaccharides as only monosaccharides and the reducing end subunit of oligomeric sugars were reduced [26, 31]. Alternatively, samples were autoclaved at 121 °C for 60 min with 4 % w/w sulfuric acid following the NREL procedure, without subsequent NaBH4 treatment, to determine the effects of oligosaccharides without prior NaBH4 treatment [29]. A combination of both treatments, acid hydrolysis with NaBH4 treatment, was used to determine the effects of lignin derived compounds from steam pretreatment. The pHs of the autoclaved samples were all readjusted to 4.8 using 50 % w/w NaOH. All samples were then hydrolyzed and quantified according to the modified ninhydrin assay described above.
Comparison of Current Protein Assays and the Modified Ninhydrin Assay In order to assess the compatibility of traditional protein assays with lignocellulosic hydrolysates, Cellic CTec3 was measured in triplicate in the presence of 50 mM acetate buffer or pretreatment WSFs as described above using the BCA, Bradford, and Lowry assays (Thermo Fisher Scientific Inc., Rockford, IL) with BSA as the protein standard. Acetone precipitation was conducted prior to protein quantification by the assays according to the manufacturer’s specifications to remove any interfering compounds present in the WSFs. The determined protein values were then compared against those obtained by the modified ninhydrin assay.
Results and Discussion Prior to protein quantification by the ninhydrin assay, the complete hydrolysis of protein to amino acids is typically required. This is traditionally accomplished by heating the protein samples at 100 °C for 24 h with 6 M HCl [24]. As other works have also recommended the use of alkaline conditions where the proteins are hydrolyzed using a strong base, such as 13.5 M NaOH, we first compared the relative merits of the acid or alkaline approach to protein breakdown [25, 32]. Although the acidic and alkaline ninhydrin assays showed a similar linear protein quantification range of 0–800 μg/g, the alkaline ninhydrin assay appeared to be influenced by various biomass derived components [25]. Previous work which used the alkaline ninhydrin
Appl Biochem Biotechnol
method reported that protein concentrations were underestimated by up to 15 % in the presence of the pretreatment liquid obtained from hydrothermally pretreated wheat straw [25]. Another potential drawback of the alkaline method is the potential loss of serine and threonine which can constitute as much as 25 % of the amino acids present in Trichoderma reesei derived cellulases [27, 33]. Thus, because of these potential drawbacks, acidic hydrolysis conditions were used for all subsequent work. Typically, sugars from the fermentation broth as well as downstream processes are present in cellulase preparations, and they are also produced during the hydrolysis of lignocellulosic substrates [12, 34]. As expected, the traditional acid-based ninhydrin assay was influenced by glucose, the predominant sugar released during cellulose hydrolysis (Fig. 1). This interference was partly due to the dehydration of hexose sugars during the traditional protein conditions of 100 °C with 6 M HCl for 24 h to generate sugar degradation products, such as levulinic acid. Previous work has shown that levulinic acid reacts with the ninhydrin reagent to form a cyclic intermediate with the five-carbon ring of ninhydrin, leading to the formation of a red condensation product that absorbs at 560 nm [10]. When the samples were spiked with glucose, an abnormal red color was observed instead of the expected Ruhemann’s purple chromophore. It was apparent that no significant interference (40 %) was observed even at the lowest concentration of glucose tested (5 g/kg). As mentioned earlier, as enzyme costs are currently too high, it can be anticipated that, in the future, there will be increasing pressure to both lower the enzyme loading while achieving even greater hydrolysis yields. As a result, for the acidic ninhydrin assay to be more effective in quantifying proteins under these conditions, it is essential that the problem of sugar interference be resolved.
Fig. 1 Relative increase in the measured protein concentration of BSA by the traditional ninhydrin assay with increasing concentrations of glucose up to 40 g/kg
Appl Biochem Biotechnol
Other researchers have also tried to reduce the interference encountered when carrying out various protein assays, for example, using methods such as protein precipitation with trichloroacetic acid or acetone [35]. However, with the increase in the hydrolytic potential of more recent Bcellulase cocktails^, lower overall protein loadings can be expected to be used when hydrolyzing lignocellulosic biomass, with protein precipitation methods resulting in some protein loss. Alternatively, other workers have used sodium borohydride (NaBH4) to reduce aldehydes and ketones to alcohols and reduce monomeric sugars to their respective alcohols [26, 31]. Thus, we next investigated whether NaBH4 treatment could minimize sugar interference with the ninhydrin-based protein assay. As mentioned earlier, higher substrate concentrations (>20 % w/w solids) will likely be required to achieve economically feasible lignocellulose hydrolysis processes [2, 3]. Under these types of processing conditions, the hydrolysate is expected to contain higher sugar and protein concentrations [2, 3, 25]. Thus, although dilution of the samples will be required to provide the accurate quantification of protein concentration within the linear protein standard range of 0 to 800 μg/g when using the ninhydrin assay, relatively high sugar concentrations (~20 g/kg) will likely remain [2, 3]. To try to simulate these types of conditions, various potential biomass derived sugars at a 20 g/kg concentration were incubated with different ratios of NaBH4 (w/w), for up to 120 min, followed by protein quantification using the traditional ninhydrin assay. Regardless of the concentration of NaBH4 added, it was apparent that glucose reduction occurred rapidly and that no significant changes in absorbance occurred after 60 min (Fig. 2a). At a NaBH4/glucose (w/w) ratio of 1:10 and 1:6.67, the glucose continued to interfere with the protein assay. However, as the ratio of NaBH4 to glucose was increased to 1:3 (w/w), little or no absorbance was detected at 560 nm after 60 min, indicating that the glucose interference of the ninhydrin assay had been minimized. To assess the possible reactivity of NaBH4 with the other types of hemicellulose derived sugars that might be encountered during enzymatic hydrolysis, arabinose, galactose, mannose, and xylose, at a concentration of 20 g/kg, were each also incubated at a ratio of 1:3 NaBH4/ sugar (w/w) for 60 min and quantified by the modified ninhydrin assay (Fig. 2b). As expected, without the addition of NaBH4, the galactose and mannose interfered with the assay due to their decomposition to levulinic acid. Although levulinic acid is not produced during the degradation of pentose sugars, both the arabinose and xylose also interfered with the assay. The observed increase in absorbance was likely due to the yellow color resulting from furfural production during the acid hydrolysis of the pentose sugars. However, with the addition of NaBH4, the interfering influence of the hemicellulosic sugars was minimized, as indicated by the lack of absorbance at 560 nm. Therefore, a ratio of 1:3 NaBH4/total sugar (w/w) and incubation time of 60 min were used for all subsequent work. Having established that the interference of up to 20 g/kg of sugars could be minimized with the addition of NaBH4, its compatibility with various proteins was next evaluated starting with BSA in the absence or presence of glucose. The samples were hydrolyzed using traditional acidic hydrolysis conditions with NaBH4 only applied to the glucose containing samples. The absorbance at 560 nm was then plotted against the protein concentration of BSA, resulting in a linear protein standard curve (Fig. 3). A comparison of the slope between the samples with (y= 2.0579x, R2 =0.9995) and without NaBH4 treatment (y=2.1132x, R2 =0.9996) indicated that NaBH4 addition had no significant effect on protein quantification. To further evaluate the compatibility of the NaBH4 treatment, commercial enzyme preparations were also assayed as they may contain residual sugars from their fermentation broths and surfactants that act as
Appl Biochem Biotechnol
Fig. 2 a Influence of different NaBH4 treatment conditions on the absorbance response of ninhydrin in the presence of 20 g/kg glucose. b Interference of various monomeric sugars at a concentration of 20 g/kg with and without NaBH4 treatment at a ratio of 1:3 NaBH4/sugar (w/w) for 60 min
protein stabilizers [34]. It was apparent that some of the commercial cellulase and xylanase preparations contained relatively high levels of sugar, with Cellic CTec2 containing the highest concentration at 265.5 g/kg glucose (Table 1). Although we had anticipated that the protein values of the enzyme preparations determined by the traditional ninhydrin assay would be overestimated, due to sugar degradation products resulting from the acid hydrolysis step, the addition of NaBH4 had little effect on the final
Appl Biochem Biotechnol
Fig. 3 Comparison of the protein concentration determined after BSA (20 g/kg glucose) was diluted in 50 mM sodium acetate buffer, using either the traditional ninhydrin assay or the traditional ninhydrin assay after NaBH4 treatment
protein concentrations determined (Fig. 4). This lack of sugar interference was likely due to the significant dilution (>100×) that was used prior to quantification due to the high initial concentration of protein and sugars. These results indicated that NaBH4 treatment is not needed when protein dilution is sufficient to reduce sugar concentration while maintaining a high protein/sugar ratio. However, under typical conditions, the dilution of samples alone is unlikely to eliminate the interference of sugars as high protein/sugar ratios are unlikely to be typically encountered. Although the addition of NaBH4 can greatly improve the compatibility of the ninhydrin assay with lignocellulosic biomass hydrolysis conditions, long hydrolysis times continue to be required for the complete hydrolysis of protein to amino acids. As previous work had shown that the complete hydrolysis of chicken egg white lysozyme could be achieved after 45 min at 160 °C, we next assessed whether complete cellulase hydrolysis could also be achieved within Table 1 Concentration of monomeric sugars present in commercial enzyme preparations Enzyme preparations
Concentration (g/kg) Ara
Gal
Glu
Xyl
Man
Celluclast 1.5L
n.d.
0.3
5.8
0.1
0.9
Cellic CTec2
n.d.
3.3
265.5
0.5
3.2
Cellic CTec3
n.d.
2.2
17.3
0.5
2.6
HTec
n.d.
0.5
0.9
0.01
0.9
Multifect Xylanase
n.d.
0.9
5.9
0.1
1.0
Ara arabinose, Gal galactose, Glu glucose, Xyl xylose, Man mannose, n.d. not detected
Appl Biochem Biotechnol
Fig. 4 Comparison of total protein concentrations of different commercial enzyme preparations quantified using the traditional ninhydrin assay with and without NaBH4 treatment
a shorter period of time [28]. To evaluate the effects of elevated hydrolysis temperatures, commercial cellulase (Cellic CTec2) and xylanase preparations (HTec) as well as BSA were hydrolyzed at 100 and 130 °C. Complete hydrolysis was defined as the lack of any increase in absorbance at 560 nm upon quantification by the ninhydrin assay. For all of the protein samples, it was evident that no further hydrolysis occurs after 18 h at 100 °C. In contrast, only 2 h was required at 130 °C to reach the same level of hydrolysis, suggesting that the overall speed of the ninhydrin assay can be significantly increased by using temperatures above 100 °C (Fig. 5a–c). Since different proteins hydrolyze at different rates depending on their amino acid composition, the applicability of the reduced hydrolysis time was next assessed on various enzyme mixtures [27]. When the protein concentrations of three different commercial cellulase and two hemicellulase preparations hydrolyzed for 24 h at 100 °C and 2 h at 130 °C were compared (Fig. 6), it was apparent that both hydrolysis conditions resulted in identical protein concentrations across all five enzyme preparations. Additionally, the standard errors obtained at 130 °C (
A NaBH4 Coupled Ninhydrin-Based Assay for the Quantification of Protein/Enzymes During the Enzymatic Hydrolysis of Pretreated Lignocellulosic Biomass Yiu Ki Mok 1 & Valdeir Arantes 2 & Jack N. Saddler 1
Received: 13 February 2015 / Accepted: 7 May 2015 # Springer Science+Business Media New York 2015
Abstract Accurate protein quantification is necessary in many of the steps during the enzymatic hydrolysis of pretreated lignocellulosic biomass, from the fundamental determination of enzyme kinetics to techno-economic assessments, such as the use of enzyme recycling strategies, evaluation of enzyme costs, and the optimization of various process steps. In the work described here, a modified, more rapid ninhydrinbased protein quantification assay was developed to better quantify enzyme levels in the presence of lignocellulosic biomass derived compounds. The addition of sodium borohydride followed by acid hydrolysis at 130 °C greatly reduced interference from monosaccharides and oligosaccharides and decreased the assay time 6-fold. The modified ninhydrin assay was shown to be more accurate as compared to various traditional colorimetric protein assays when commercial cellulase enzyme mixtures were quantified under typical pretreated lignocellulosic biomass enzymatic hydrolysis conditions. The relatively short assay time and microplate-reading capability of the modified assay indicated that the method could likely be used for high-throughput protein determination. Keywords Lignocellulose . Ninhydrin . Protein quantification . Cellulase enzymes . Enzymatic hydrolysis
* Jack N. Saddler [email protected] 1
Forest Products Biotechnology/Bioenergy Group, Faculty of Forestry, University of British Columbia, 2424 Main Mall, Vancouver, BC V6T 1Z4, Canada
2
Department of Biotechnology, Lorena School of Engineering, University of São Paulo, CP 116, Rod. Itajubá—Lorena, km 74,5 Área I, Estrada Municipal do Campinho S/N 2602-810, Lorena, SP, Brazil
Appl Biochem Biotechnol
Introduction The cost-effective production of sugars from lignocellulosic biomass continues to be a challenge for the bioconversion/biorefining industries. This is primarily due to the costs associated with the large amount of enzyme/proteins needed to achieve effective cellulose hydrolysis to monomeric sugars [1]. Thus, a considerable amount of effort has been invested in decreasing enzyme production costs, increasing enzyme performance, and developing enzyme recycling strategies. The high enzyme/protein loading required to achieve effective hydrolysis has also been shown to be a major limitation in a number of techno-economic analyses of biomass-to-ethanol processes [2, 3]. Therefore, to try to better assess the efficacy of these approaches from both a fundamental enzyme kinetic and techno-economic perspective, an accurate method of assessing cellulase activity/performance is needed. Traditionally, the filter paper activity (FPA) assay, as recommended by the International Union of Pure and Applied Chemistry (IUPAC), is often employed to evaluate cellulase activity/performance [4]. The filter paper activity assay measures cellulase activity on a volumetric basis by determining the amount of cellulase enzymes required to solubilize 4 % of a 50 mg Whatman #1 filter paper strip in 60 min. Although widely used, it is well known that in addition to the composition of the enzyme preparation, substrate characteristics also play a significant role in influencing enzymatic performance [4–6]. As a result, the filter paper activity assay is often not reflective of the hydrolytic performance of a particular enzyme mixture on different pretreated lignocellulosic substrate. Due to the inherent limitations of the filter paper assay, many researchers have supplemented enzymatic activity-based assays with total protein quantification assays as a preferred method of assessing enzyme performance [5, 7]. Unfortunately, the selection of a total protein quantification assay for the accurate determination of total protein concentration in a Bcellulase mixture^ remains challenging, in part due to the complexity of a cellulase mixture, which contain both cellulases and other accessory enzymes and the extent of glycosylation of each of the enzymes [5, 8]. In a previous work by Adney et al. (1995), the protein concentration of multiple commercial cellulase preparations were determined using the modified Lowry, Bradford, and Bicinchoninic acid (BCA) assays as well as Kjeldahl Nitrogen Analysis and UV absorbance at 280 nm [9]. These workers found that, depending on the protein quantification method used, the total protein concentration varied by as much as 90 %. Similar results were reported by McMillan et al. (2011) where the protein concentration of four cellulase preparations varied by 2.6–4.8 times when measured by either the Bradford or BCA assays [7]. This large variability in accurately determining protein concentrations when utilizing different quantification methods is of significant concern, particularly when enzyme activities and kinetics are being evaluated. The enzymatic hydrolysis of lignocellulosic biomass also results in the production of monomeric and oligomeric sugars, which are known to interfere with many protein assays [10–12]. As past work was often carried out at low substrate concentrations and high enzyme loadings, the interfering effects of any released sugars could simply be resolved by diluting the samples below interfering sugar concentrations. However, with the increased hydrolytic efficiency of newer cellulase preparation and the need to move to higher substrate concentrations, higher sugar-to-protein concentration ratios can be expected. As a result, sample dilution is often no longer a viable way of reducing the interfering effects of sugars. To further complicate matters, other compounds that are present or released during biomass pretreatment
Appl Biochem Biotechnol
and hydrolysis, such as phenolics, surfactants, and sugar degradation products (furfural, 5hydroxymethylfurfural (HMF), levulinic acid), are also know to interfere with many protein assays [10, 11, 13–19]. As mentioned earlier, the modified Lowry, Bradford, and Bicinchoninic acid (BCA) assays as well as Kjeldahl Nitrogen Analysis and UV absorbance have all been previously used, with varying degrees of success, to try to accurately quantify cellulase mixtures. In relative contrast, the ninhydrin assay has shown promise due to its specificity for amino acids and decreased susceptibility to interference [20–24]. However, high concentrations of sugars have been shown to be problematic, as were the long hydrolysis times typically required to attain complete protein hydrolysis [10, 24]. Other workers have also reported wide-ranging standard deviations [25]. In the work described below, a modified, accurate, and rapid ninhydrin assay was developed and used to quantify the enzyme/protein present during enzymatic hydrolysis of biomass at relatively low protein and high substrate/sugar concentrations. A 6-fold reduction in total assay time could be achieved by increasing hydrolysis temperatures from the traditional 100 to 130 °C while the interfering influence of released sugars could be greatly reduced by sodium borohydride treatment.
Materials and Methods Traditional Acidic Ninhydrin Assay The total protein concentration was measured using the traditional ninhydrin assay following acid hydrolysis calibrated using bovine serum albumin (BSA, Sigma) as a protein standard. One hundred microliters of each protein sample (protein concentration between 0 and 800 μg/ g) was mixed with 50 μL of ultra-pure water in a 0.5 mL screw-cap microcentrifuge tube (Fisherbrand). Samples were diluted on a weight basis to minimize potential errors resulting from large dilution factors as some enzyme preparations contained high protein concentrations. This was followed by the addition of 300 μL of 9 M hydrochloric acid (HCl) to a final concentration of 6 M and subsequently heated in a dry heating bath (MBI Lab Equipment, Kirkland, PQ) at 100 °C for 24 h. After cooling to room temperature, 100 μL of the hydrolysate was transferred into a 1.5-mL microcentrifuge tube (Fisherbrand) and neutralized with 100 μL of 5 M sodium hydroxide (NaOH) (Fisher Scientific). Upon neutralization, 200 μL of the 2 % ninhydrin reagent (Sigma) was added and heated at 100 °C for 10 min. Samples were then cooled to room temperature prior to the addition of 500 μL of 50 % v/v ethanol. Subsequently, 200 μL of the colored solution was transferred to a 96-well microplate (Corning) and the absorbance was read at 560 nm by a microplate reader (PerkinElmer VICTOR3, Woodbridge, ON). All samples were performed in triplicate and experiments were repeated at least twice.
Evaluation of Glucose Interference on the Traditional Ninhydrin Assay To evaluate the possible interfering influence of glucose on the traditional ninhydrin assay, which is the predominant sugar produced during enzymatic hydrolysis of lignocellulosic biomass, BSA (100, 800, and 1600 μg protein/g) and glucose (0, 5, 10, 20, and 40 g/kg) (Sigma) were dissolved in 50 mM sodium acetate buffer (pH=4.8). Fifty microliters of the
Appl Biochem Biotechnol
protein solution and 50 μL of the sugar solution were mixed together, resulting in a final glucose concentration of 2.5, 5, 10, and 20 g/kg and BSA concentrations of 50, 400, and 800 μg/g, respectively. Fifty microliters of ultra-pure water was then added to the protein-sugar mixture. Alternatively, 50 mM sodium acetate buffer (pH=4.8) was added in place of the sugar solution for the 0 g/kg glucose condition. All samples were then quantified using the traditional ninhydrin assay.
Elimination of Glucose Interference with Sodium Borohydride As sodium borohydride (NaBH4) had been previously used to reduce monomeric sugars to sugar alcohols, we investigated the potential use of NaBH4 to remove the interference of sugars on the traditional ninhydrin assay [26]. One hundred microliters of a 20 g/kg glucose solution dissolved in 50 mM sodium acetate buffer (pH=4.8) was incubated with 50 μL of NaBH4 solution prepared by dissolving NaBH4 pellets (Sigma-Aldrich) in ultra-pure water followed by the addition of 0.02 %v/v Antifoam O-30 (Sigma-Aldrich). Three different NaBH 4 concentrations of 4, 6, and 13.3 g/kg were prepared, representing ratios of 1:10, 1:6.67, and 1:3 NaBH4/glucose w/w, respectively. Samples were incubated at room temperature and the reaction was stopped at 5, 15, 30, 60, and 120 min by the addition of 300 μL of 9 M HCl (final concentration of 6 M HCl). Following NaBH4 treatment, the samples were assayed by the traditional ninhydrin assay. Once the required incubation time and concentration of NaBH4 needed to remove glucose interference was determined, the same conditions were then used to assess the possible influence of hemicellulose derived sugars (arabinose, galactose, mannose, and xylose) at concentrations of 20 g/kg. As described in more detail in the BResults and Discussion^ section, the required ratio of NaBH4/glucose and incubation time required to completely eliminate sugar interference was determined to be 1:3 NaBH4/glucose w/w and 60 min. These conditions were subsequently used to develop the modified ninhydrin assay.
Influence of Sodium Borohydride on Protein Measurements To further investigate the possible influence of NaBH4 on the ninhydrin assay, BSA (0– 1600 μg/g) and glucose (40 g/kg) were dissolved in 50 mM sodium acetate buffer (pH= 4.8). Fifty microliters of the protein solution and 50 μL of the sugar solution were mixed together, resulting in a final glucose concentration of 20 g/kg and maximum BSA concentration of 800 μg/g. This protein-sugar mixture was then subjected to NaBH4 treatment as described above. As a comparison without the addition of NaBH4, 50 μL of the protein solution was mixed with 50 μL of 50 mM sodium acetate buffer followed by the addition of 50 μL of ultra-pure water. All samples were then subjected to acid hydrolysis at 100 °C for 24 h followed by quantification with the traditional ninhydrin assay. Upon establishing the effects of NaBH4 on BSA, the protein concentration of commercial cellulase preparations (Celluclast 1.5L, Cellic CTec2, Cellic CTec3) and commercial hemicellulase preparations (HTec, Multifect Xylanase) were also compared with and without the addition of NaBH4. The concentrations of sugars present in the enzyme preparations were measured using high-performance liquid chromatography (HPLC) (Dionex DX-3000, Sunnyvale, CA) as described elsewhere [6].
Appl Biochem Biotechnol
Reduction of Time Required for Complete Protein Hydrolysis As previous work had shown that the time required for complete protein hydrolysis can be reduced by increasing the hydrolysis temperature, we also investigated if the total hydrolysis time can be reduced by increasing the hydrolysis temperature from 100 to 130 °C [27]. Previous work has hydrolyzed protein at temperatures as high as 160 °C to reduce the hydrolysis time. However, these conditions often required specially designed equipment [28]. A temperature of 130 °C was thus selected as a compromise condition between hydrolysis time and the need for specialized equipment. BSA, Cellic CTec2, and HTec were incubated using the NaBH4 conditions described above and hydrolyzed with 9 M HCl (final concentration of 6 M HCl) at 100 and 130 °C. Samples were hydrolyzed in triplicates for 0, 2, 4, 6, 12, 18, and 24 h at 100 °C for 0, 15, 30, 60, 120, and 240 min at 130 °C. The extent of protein hydrolysis was quantified using the ninhydrin assay. Upon determining the required hydrolysis time to attain complete protein hydrolysis at 130 °C, the protein concentration of commercial cellulase preparations (Celluclast 1.5L, Cellic CTec2, Cellic CTec3) and commercial hemicellulases preparations (HTec, Multifect Xylanase) were compared between the traditional hydrolysis conditions of 24 h at 100 °C and the hydrolysis conditions at 130 °C. As described in more detail in the BResults and Discussion^ section, complete protein hydrolysis was attained with a reduced time of 2 h at 130 °C across all of the enzyme preparations. These conditions were subsequently used for the modified ninhydrin assay.
Modification of the Ninhydrin Assay to Quantify Protein During the Enzymatic Hydrolysis of High Concentrations of Biomass The modified ninhydrin method was developed to improve the overall accuracy, compatibility, and speed of the assay. One hundred microliters of protein containing samples (protein concentration between 0 and 800 μg/g) was first incubated with 50 μL of NaBH4 for 60 min at a ratio of 1:3 NaBH4/total sugar (w/w) in a 0.5-mL screw-cap microcentrifuge tube with BSA as a protein standard unless otherwise specified. This was followed by the addition of 300 μL of 9 M HCl (final concentration of 6 M HCl) and subsequent heating in a dry heating bath at 130 °C for 2 h. After cooling to room temperature, 100 μL of the hydrolysate was transferred to a 1.5-mL microcentrifuge tube and neutralized with 100 μL of 5 M NaOH. Upon neutralization, 200 μL of 2 % ninhydrin reagent was added and heated at 100 °C for 10 min. Samples were then cooled to room temperature prior to the addition of 500 μL of 50 %v/v ethanol. Subsequently, 200 μL of the colored solution was transferred to a 96-well microplate and the absorbance was read at 560 nm. All samples were performed in triplicate and experiments were repeated at least twice.
Compatibility of Modified Ninhydrin Assay with Compounds Encountered During Lignocellulosic Biomass Hydrolysis In addition to monosaccharides, lignocellulosic hydrolysates can also contain other compounds such as oligosaccharides and phenolic compounds. To determine the influence of these compounds, a commercial cellulase preparation, Cellic CTec3, was quantified using the modified ninhydrin assay described above after dilution in 50 mM sodium acetate buffer (pH= 4.8) or the water soluble fractions (WSFs) obtained from the SO2 catalyzed steam pretreatment of corn stover (SPCS), poplar (SPP), and lodgepole pine (SPLPP). Each substrate was
Appl Biochem Biotechnol
pretreated under near optimal conditions to maximize carbohydrate recovery while providing a cellulose-rich substrate that was more susceptible to enzymatic hydrolysis [6]. Compositional analyses of the water soluble fractions were carried out in duplicate following the National Renewable Energy Laboratory (NREL) protocols [29]. Additionally, the total amount of phenolics was measured in triplicate using the Lowry-Folin method with phloroglucinol as a standard [30]. These conditions were selected to mimic potential enzymatic hydrolysis conditions where high concentrations of phenolic derivatives and sugars would be present. After dilution, the final total sugar concentration in each enzyme/WSF mixture was always below 20 g/kg. Samples without NaBH4 treatment were also quantified to assess the degree of interference by the WSFs.
Elucidation of Interfering Effects of Individual Lignocellulosic Components To elucidate the individual interfering effects of different lignocellulosic components present in the WSFs, Cellic CTec3 was diluted in the different WSFs and quantified using the modified ninhydrin assay after three different treatments. NaBH4 treatment alone was used to isolate the effects of oligosaccharides as only monosaccharides and the reducing end subunit of oligomeric sugars were reduced [26, 31]. Alternatively, samples were autoclaved at 121 °C for 60 min with 4 % w/w sulfuric acid following the NREL procedure, without subsequent NaBH4 treatment, to determine the effects of oligosaccharides without prior NaBH4 treatment [29]. A combination of both treatments, acid hydrolysis with NaBH4 treatment, was used to determine the effects of lignin derived compounds from steam pretreatment. The pHs of the autoclaved samples were all readjusted to 4.8 using 50 % w/w NaOH. All samples were then hydrolyzed and quantified according to the modified ninhydrin assay described above.
Comparison of Current Protein Assays and the Modified Ninhydrin Assay In order to assess the compatibility of traditional protein assays with lignocellulosic hydrolysates, Cellic CTec3 was measured in triplicate in the presence of 50 mM acetate buffer or pretreatment WSFs as described above using the BCA, Bradford, and Lowry assays (Thermo Fisher Scientific Inc., Rockford, IL) with BSA as the protein standard. Acetone precipitation was conducted prior to protein quantification by the assays according to the manufacturer’s specifications to remove any interfering compounds present in the WSFs. The determined protein values were then compared against those obtained by the modified ninhydrin assay.
Results and Discussion Prior to protein quantification by the ninhydrin assay, the complete hydrolysis of protein to amino acids is typically required. This is traditionally accomplished by heating the protein samples at 100 °C for 24 h with 6 M HCl [24]. As other works have also recommended the use of alkaline conditions where the proteins are hydrolyzed using a strong base, such as 13.5 M NaOH, we first compared the relative merits of the acid or alkaline approach to protein breakdown [25, 32]. Although the acidic and alkaline ninhydrin assays showed a similar linear protein quantification range of 0–800 μg/g, the alkaline ninhydrin assay appeared to be influenced by various biomass derived components [25]. Previous work which used the alkaline ninhydrin
Appl Biochem Biotechnol
method reported that protein concentrations were underestimated by up to 15 % in the presence of the pretreatment liquid obtained from hydrothermally pretreated wheat straw [25]. Another potential drawback of the alkaline method is the potential loss of serine and threonine which can constitute as much as 25 % of the amino acids present in Trichoderma reesei derived cellulases [27, 33]. Thus, because of these potential drawbacks, acidic hydrolysis conditions were used for all subsequent work. Typically, sugars from the fermentation broth as well as downstream processes are present in cellulase preparations, and they are also produced during the hydrolysis of lignocellulosic substrates [12, 34]. As expected, the traditional acid-based ninhydrin assay was influenced by glucose, the predominant sugar released during cellulose hydrolysis (Fig. 1). This interference was partly due to the dehydration of hexose sugars during the traditional protein conditions of 100 °C with 6 M HCl for 24 h to generate sugar degradation products, such as levulinic acid. Previous work has shown that levulinic acid reacts with the ninhydrin reagent to form a cyclic intermediate with the five-carbon ring of ninhydrin, leading to the formation of a red condensation product that absorbs at 560 nm [10]. When the samples were spiked with glucose, an abnormal red color was observed instead of the expected Ruhemann’s purple chromophore. It was apparent that no significant interference (40 %) was observed even at the lowest concentration of glucose tested (5 g/kg). As mentioned earlier, as enzyme costs are currently too high, it can be anticipated that, in the future, there will be increasing pressure to both lower the enzyme loading while achieving even greater hydrolysis yields. As a result, for the acidic ninhydrin assay to be more effective in quantifying proteins under these conditions, it is essential that the problem of sugar interference be resolved.
Fig. 1 Relative increase in the measured protein concentration of BSA by the traditional ninhydrin assay with increasing concentrations of glucose up to 40 g/kg
Appl Biochem Biotechnol
Other researchers have also tried to reduce the interference encountered when carrying out various protein assays, for example, using methods such as protein precipitation with trichloroacetic acid or acetone [35]. However, with the increase in the hydrolytic potential of more recent Bcellulase cocktails^, lower overall protein loadings can be expected to be used when hydrolyzing lignocellulosic biomass, with protein precipitation methods resulting in some protein loss. Alternatively, other workers have used sodium borohydride (NaBH4) to reduce aldehydes and ketones to alcohols and reduce monomeric sugars to their respective alcohols [26, 31]. Thus, we next investigated whether NaBH4 treatment could minimize sugar interference with the ninhydrin-based protein assay. As mentioned earlier, higher substrate concentrations (>20 % w/w solids) will likely be required to achieve economically feasible lignocellulose hydrolysis processes [2, 3]. Under these types of processing conditions, the hydrolysate is expected to contain higher sugar and protein concentrations [2, 3, 25]. Thus, although dilution of the samples will be required to provide the accurate quantification of protein concentration within the linear protein standard range of 0 to 800 μg/g when using the ninhydrin assay, relatively high sugar concentrations (~20 g/kg) will likely remain [2, 3]. To try to simulate these types of conditions, various potential biomass derived sugars at a 20 g/kg concentration were incubated with different ratios of NaBH4 (w/w), for up to 120 min, followed by protein quantification using the traditional ninhydrin assay. Regardless of the concentration of NaBH4 added, it was apparent that glucose reduction occurred rapidly and that no significant changes in absorbance occurred after 60 min (Fig. 2a). At a NaBH4/glucose (w/w) ratio of 1:10 and 1:6.67, the glucose continued to interfere with the protein assay. However, as the ratio of NaBH4 to glucose was increased to 1:3 (w/w), little or no absorbance was detected at 560 nm after 60 min, indicating that the glucose interference of the ninhydrin assay had been minimized. To assess the possible reactivity of NaBH4 with the other types of hemicellulose derived sugars that might be encountered during enzymatic hydrolysis, arabinose, galactose, mannose, and xylose, at a concentration of 20 g/kg, were each also incubated at a ratio of 1:3 NaBH4/ sugar (w/w) for 60 min and quantified by the modified ninhydrin assay (Fig. 2b). As expected, without the addition of NaBH4, the galactose and mannose interfered with the assay due to their decomposition to levulinic acid. Although levulinic acid is not produced during the degradation of pentose sugars, both the arabinose and xylose also interfered with the assay. The observed increase in absorbance was likely due to the yellow color resulting from furfural production during the acid hydrolysis of the pentose sugars. However, with the addition of NaBH4, the interfering influence of the hemicellulosic sugars was minimized, as indicated by the lack of absorbance at 560 nm. Therefore, a ratio of 1:3 NaBH4/total sugar (w/w) and incubation time of 60 min were used for all subsequent work. Having established that the interference of up to 20 g/kg of sugars could be minimized with the addition of NaBH4, its compatibility with various proteins was next evaluated starting with BSA in the absence or presence of glucose. The samples were hydrolyzed using traditional acidic hydrolysis conditions with NaBH4 only applied to the glucose containing samples. The absorbance at 560 nm was then plotted against the protein concentration of BSA, resulting in a linear protein standard curve (Fig. 3). A comparison of the slope between the samples with (y= 2.0579x, R2 =0.9995) and without NaBH4 treatment (y=2.1132x, R2 =0.9996) indicated that NaBH4 addition had no significant effect on protein quantification. To further evaluate the compatibility of the NaBH4 treatment, commercial enzyme preparations were also assayed as they may contain residual sugars from their fermentation broths and surfactants that act as
Appl Biochem Biotechnol
Fig. 2 a Influence of different NaBH4 treatment conditions on the absorbance response of ninhydrin in the presence of 20 g/kg glucose. b Interference of various monomeric sugars at a concentration of 20 g/kg with and without NaBH4 treatment at a ratio of 1:3 NaBH4/sugar (w/w) for 60 min
protein stabilizers [34]. It was apparent that some of the commercial cellulase and xylanase preparations contained relatively high levels of sugar, with Cellic CTec2 containing the highest concentration at 265.5 g/kg glucose (Table 1). Although we had anticipated that the protein values of the enzyme preparations determined by the traditional ninhydrin assay would be overestimated, due to sugar degradation products resulting from the acid hydrolysis step, the addition of NaBH4 had little effect on the final
Appl Biochem Biotechnol
Fig. 3 Comparison of the protein concentration determined after BSA (20 g/kg glucose) was diluted in 50 mM sodium acetate buffer, using either the traditional ninhydrin assay or the traditional ninhydrin assay after NaBH4 treatment
protein concentrations determined (Fig. 4). This lack of sugar interference was likely due to the significant dilution (>100×) that was used prior to quantification due to the high initial concentration of protein and sugars. These results indicated that NaBH4 treatment is not needed when protein dilution is sufficient to reduce sugar concentration while maintaining a high protein/sugar ratio. However, under typical conditions, the dilution of samples alone is unlikely to eliminate the interference of sugars as high protein/sugar ratios are unlikely to be typically encountered. Although the addition of NaBH4 can greatly improve the compatibility of the ninhydrin assay with lignocellulosic biomass hydrolysis conditions, long hydrolysis times continue to be required for the complete hydrolysis of protein to amino acids. As previous work had shown that the complete hydrolysis of chicken egg white lysozyme could be achieved after 45 min at 160 °C, we next assessed whether complete cellulase hydrolysis could also be achieved within Table 1 Concentration of monomeric sugars present in commercial enzyme preparations Enzyme preparations
Concentration (g/kg) Ara
Gal
Glu
Xyl
Man
Celluclast 1.5L
n.d.
0.3
5.8
0.1
0.9
Cellic CTec2
n.d.
3.3
265.5
0.5
3.2
Cellic CTec3
n.d.
2.2
17.3
0.5
2.6
HTec
n.d.
0.5
0.9
0.01
0.9
Multifect Xylanase
n.d.
0.9
5.9
0.1
1.0
Ara arabinose, Gal galactose, Glu glucose, Xyl xylose, Man mannose, n.d. not detected
Appl Biochem Biotechnol
Fig. 4 Comparison of total protein concentrations of different commercial enzyme preparations quantified using the traditional ninhydrin assay with and without NaBH4 treatment
a shorter period of time [28]. To evaluate the effects of elevated hydrolysis temperatures, commercial cellulase (Cellic CTec2) and xylanase preparations (HTec) as well as BSA were hydrolyzed at 100 and 130 °C. Complete hydrolysis was defined as the lack of any increase in absorbance at 560 nm upon quantification by the ninhydrin assay. For all of the protein samples, it was evident that no further hydrolysis occurs after 18 h at 100 °C. In contrast, only 2 h was required at 130 °C to reach the same level of hydrolysis, suggesting that the overall speed of the ninhydrin assay can be significantly increased by using temperatures above 100 °C (Fig. 5a–c). Since different proteins hydrolyze at different rates depending on their amino acid composition, the applicability of the reduced hydrolysis time was next assessed on various enzyme mixtures [27]. When the protein concentrations of three different commercial cellulase and two hemicellulase preparations hydrolyzed for 24 h at 100 °C and 2 h at 130 °C were compared (Fig. 6), it was apparent that both hydrolysis conditions resulted in identical protein concentrations across all five enzyme preparations. Additionally, the standard errors obtained at 130 °C (
