Albuminated Glycoenzymes: Enzyme Stabilization through Orthogonal

Subscriber access provided by TULANE UNIVERSITY

Article

Albuminated Glycoenzymes: Enzyme Stabilization through Orthogonal Attachment of a Single-Layered Protein Shell around a Central Glycoenzyme Core Dustin W. Ritter, Jared M. Newton, Jason Richard Roberts, and Michael J. McShane Bioconjugate Chem., Just Accepted Manuscript • DOI: 10.1021/acs.bioconjchem.6b00103 • Publication Date (Web): 25 Apr 2016 Downloaded from http://pubs.acs.org on May 3, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Bioconjugate Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Bioconjugate Chemistry

Albuminated Glycoenzymes: Enzyme Stabilization through Orthogonal Attachment of a Single-Layered Protein Shell around a Central Glycoenzyme Core Dustin W. Ritter†, Jared M. Newton†, Jason R. Roberts†, and Michael J. McShane*†‡ † Department of Biomedical Engineering, Texas A&M University, College Station, Texas 77843-3120, United States ‡ Department of Materials Science & Engineering, Texas A&M University, College Station, Texas 77843-3003, United States *Email: [email protected]

ABSTRACT: Here we demonstrate an approach to stabilize enzymes through the orthogonal covalent attachment of albumin on the single-enzyme level. Albuminated glycoenzymes (AGs) based upon glucose oxidase and catalase from Aspergillus niger were prepared in this manner. Gel filtration chromatography and dynamic light scattering support modification, with an increase in hydrodynamic radius of ca. 60% upon albumination. Both AGs demonstrate a marked resistance to aggregation during heating to 90 °C, but this effect is more profound in albuminated catalase. The functional characteristics of albuminated glucose oxidase vary considerably with exposure type. The AG’s thermal inactivation is reduced more than 25 times

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

compared to native glucose oxidase, and moderate stabilization is observed with one-month storage at 37 °C. However, albumination has no effect on operational stability of glucose oxidase.

INTRODUCTION Enzymes are employed across a wide variety of fields, ranging from industrial applications like food processing to more research-focused applications such as molecular biology. In biomedical engineering, enzymes play a particularly significant role in certain types of biosensors. One wellknown example is the glucose biosensor,1 the most common of which relies upon glucose oxidase (GOx). Regardless of the application, enzymes are valued for their ability to convert substrate to product with high selectivity. This is only possible while the enzyme maintains its catalytic activity. At minimum, loss of enzymatic activity translates to financial losses— replacement cost of deactivated enzyme or decreased revenue from reduced product yield. However, in applications such as percutaneous or fully implantable glucose biosensors,1-6 where replacement of deactivated enzyme is nontrivial, the implications are far worse. Loss of enzymatic activity can drastically affect sensor response, leading to a need for frequent recalibration and ultimately sensor replacement. If glucose prediction errors exceed 20% (e.g., resulting from improper recalibration or failure to recalibrate), incorrect therapeutic decisions can be made, which can have severe consequences for patients relying on the sensor.7 Given the ubiquity of enzymes, numerous enzyme stabilization approaches have emerged over time. Many of these techniques are quite effective and include crosslinking,8-13 chemical modification,14-17 and immobilization on a solid support.18-24 However, they often create enzyme constructs orders of magnitude larger than the single enzyme, significantly limiting diffusion. Additionally, these constructs may no longer exist in the solution phase, and initial activity can


Page 2 of 24

Page 3 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


be greatly reduced upon stabilization due to modification of groups involved in catalysis. These properties can preclude use of the stabilized enzyme in applications requiring solution-phase enzyme of near-native size. It is worth noting that when evaluating the stability of an enzyme, loss of activity over time and throughout normal operating conditions are certainly important, but should not be the only factors considered or assessed. In an enzyme-containing device, there are numerous “opportunities”

for

accelerated

enzyme

deactivation—during

fabrication,

processing,

distribution, and operation under extreme or abnormal conditions—to which an enzyme should ideally be stabilized. Depending upon the application, the ability to withstand harsh sterilization procedures, extreme temperatures (high and/or low), multiple freeze-thaw cycles, and exposure to various additives or solvents might be desirable. The novel stabilization approach presented herein involves the orthogonal covalent attachment of bovine serum albumin (BSA) to the glycosylation sites of an enzyme resulting in encasement of the glycoenzyme within an albumin shell (Figure 1). This approach seeks to improve upon an uncontrolled enzyme stabilization strategy commonly used in current glucose biosensors25 through introduction of the BSA in a controlled manner and on the single-enzyme level. The vast literature on BSA supported the hypothesis that thermostability might be improved via albumination. Serum albumin is known to be quite thermostable,26 and many groups have reported that addition of BSA to an enzyme solution improves the enzyme’s thermostability.27-30 Chang and Mahoney showed that addition of BSA to a solution of β-galactosidase from Streptococcus thermophilus resulted in a nine-fold thermostabilization. They also proposed that hydrophobic interactions played a critical role in the stabilizing action imparted by BSA.27 Building upon that work, Marini and co-workers recognized that BSA has molecular chaperone-



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 24

like properties; that is, when added to an enzyme solution, the enzyme experienced greatly reduced thermal inactivation and aggregation. In some cases, BSA even outperformed αcrystallin, a recognized intracellular molecular chaperone.28 Finally, Myung and Zhang also observed enhanced enzyme stability with the addition of BSA,30 which they explained in the context of the macromolecular crowding effect.31

Figure 1. Cartoon illustrating the orthogonal attachment of multiple BSA macromolecules to a native glycoenzyme to form an AG. In this work, we demonstrate that this technique of preparing albuminated glycoenzymes (AGs), which is applied to both GOx and catalase (Cat), results in dramatic thermostabilization with a moderate increase in long-term storage stability for BSA-modified GOx (BSA-GOx). It is important to point out that while both of the glycoenzymes employed in this study are multimeric, the conjugation strategy employed herein is not expected to prevent subunit dissociation. In the case of GOx, dissociation of the two subunits is accompanied by loss of each flavin adenine dinucleotide (FAD) cofactor,32,

33

rendering the enzyme deactivated.

For

strategies to prevent subunit dissociation, the reader is directed to an excellent review by Fernández-Lafuente.34 While the direct goal of this work was to stabilize GOx and Cat in a manner which permits their incorporation into optical biosensing hydrogels, this approach could be extended or adapted to stabilize other proteins for different applications.

Furthermore,

albumination of non-human-derived proteins using human serum albumin could impart other beneficial effects beyond stabilization, such as reduction of immunogenicity when employed in


Page 5 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


vivo. At this point, the AGs have not been tested for an immunogenic response, and this is currently only hypothetical; feasibility testing for this would need to be investigated in future work. RESULTS AND DISCUSSION Strategy for Glycoenzyme Albumination. GOx from A. niger is a homodimer (ca. 160 kDa) with approximately 18–20% glycosylation,

35, 36

and Cat from A. niger is tetrameric (ca. 385

kDa) with a carbohydrate content of 10.2%.37 In both cases, the enzymes’ glycosylation sites were targeted for the attachment of BSA. Briefly, the single free cysteine of BSA was modified using a thiol-reactive heterobifunctional crosslinker to yield hydrazide-functionalized BSA. Separately, glycosylation sites on GOx or Cat were oxidized using sodium periodate to generate reactive aldehyde groups. Following purification of modified BSA and glycoenzyme, an excess of mono-hydrazide-functionalized BSA was combined with oxidized glycoenzyme to prepare the desired bioconjugate. Resulting hydrazone linkages were reduced for stability, and the AGs were purified by diafiltration, characterized by a variety of methods, and subjected to a host of stability tests. Liquid Chromatography. Figure 2 contains the overlaid chromatograms of the native glycoenzymes (gray lines) and the AGs (maroon lines) for GOx (Figure 2A) and Cat (Figure 2B). In the case of both native glycoenzymes, elution of only one primary peak is detected, which has overlapping absorbance in both the UV (dashed lines) and the visible regions (solid lines) as expected. In contrast, the UV chromatograms for the crude AG samples contain three primary peaks, but only the one at the lowest elution volume overlaps with the single peak detected at the visible wavelengths; this peak corresponds to the AG, and it is sufficiently resolved from the other peaks such that purification from the other components in the crude AG



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

sample is possible. The component which eluted at approximately 76 mL corresponds to BSA, and the peak at approximately 66 mL elution volume is believed to be dimers of BSA.38

Figure 2. Chromatograms of BSA-GOx and native GOx (A), as well as BSA-Cat and native Cat (B). Traces corresponding to AGs are shown in maroon, while those corresponding to native enzymes are shown in gray. Absorbance at 280 nm is indicated by a dashed line, while absorbance at 450 nm (for GOx and BSA-GOx) or 405 nm (for Cat and BSA-Cat) is indicated by a solid line. For visualization purposes, the 450 nm traces for native GOx and BSA-GOx have been multiplied by a factor of ten. Regarding the peaks corresponding to the AGs, presence of absorbance in the visible region supports retention of the enzymes’ respective cofactors, as they contribute the absorbance in this region. Conjugation of the BSA to the glycoenzymes is supported by 1) the large shift in elution


Page 6 of 24

Page 7 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


volume, which is indicative of a drastic increase in the Stokes radius (lower elution volume indicates higher Stokes radius); and 2) the additional UV absorbance as compared to the native glycoenzymes. Concerning the latter point, the UV:visible absorbance ratio doubled upon albumination, increasing from 10:1 for GOx and 1:1 for Cat to 20:1 for BSA-GOx and 2:1 for BSA-modified Cat (BSA-Cat). As BSA absorbs at 280 nm, but does not contribute any appreciable absorbance at 405 nm or 450 nm, this increase in the UV:visible absorbance ratio supports co-elution of BSA and GOx. Particle Sizing. Dynamic light scattering (DLS) was used to determine the hydrodynamic sizes of BSA, the native glycoenzymes, and the AGs (Figure 3). Particle sizing reveals that BSA has a hydrodynamic diameter of 7.959 nm. DLS reveals that native GOx has a hydrodynamic diameter of 12.36 nm, which increases by 7.76 nm to 20.12 nm upon modification with BSA. Likewise, the hydrodynamic diameter of native Cat increases by 9.27 nm following albumination (from 15.38 nm to 24.65 nm). The relative increase (i.e., percent change) in hydrodynamic size is similar for both AGs—62.8% increase for BSA-GOx and 60.3% increase for BSA-Cat.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. Size distributions of BSA, native GOx, and BSA-GOx (A), as well as BSA, native Cat, and BSA-Cat (B). BSA is shown as black bars, native enzymes as gray bars, and AGs are shown as maroon bars. Temperature Ramp. The albuminated and native glycoenzymes were characterized by DLS to investigate the effect of heating on their size distributions (n = 3). In the absence of detergents, it is expected that heat-induced denaturation of the enzymes would lead to aggregation with a corresponding shift in the size distributions toward larger sizes. The hydrodynamic diameter of each form of enzyme was monitored while the temperature of the sample was incrementally increased from 25 °C to 90 °C. The size distributions for the native glycoenzymes and the AGs as a function of temperature are shown in Figure 4. All samples exhibit one primary peak at all tested temperatures; however, the native glycoenzymes (especially Cat) show evidence of higher-molecular-weight species at some lower temperatures. Remarkably, at all tested temperatures, the size distributions of the AGs remain relatively constant as compared to the native glycoenzymes.

Figure 4. Size distributions for native GOx and BSA-GOx, as well as native Cat and BSA-Cat, as a function of temperature (n = 3).


Page 8 of 24

Page 9 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 5 shows the percent change in the size of the primary peak (from the size distributions shown in Figure 4) from its initial size as a function of temperature. At 60 °C, native GOx begins to drastically increase in size, stabilizing near 150% change at 75 °C. This shift of the enzyme’s size distribution toward larger size is indicative of thermal denaturation and subsequent aggregation, which is consistent with the reported melting temperature of GOx (ca. 56–58 °C)3941

as well as reports that GOx forms trimers and tetramers upon thermal denaturation.39 Like

native GOx, the size distribution for BSA-GOx begins to increase in size at approximately 60 °C; however, the rate of change is approximately 5–6 times slower than that of native GOx, and the total percent change after ramping to 90 °C is reduced by a factor of two. The results obtained for native Cat and BSA-Cat are similar to those for native GOx and BSA-GOx, but even more dramatic. Both native Cat and BSA-Cat begin to experience a shift in their size distributions toward larger size at approximately 65 °C; however, after ramping to 90 °C, native Cat increases by 440% of its initial size, whereas BSA-Cat only increases by 40% (an 11-fold difference).



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 24

Figure 5. Effect of heating on the size of BSA-GOx and native GOx (A) and BSA-Cat and native Cat (B). BSA-GOx and native GOx are indicated by maroon squares and gray crosses, respectively, while BSA-Cat and native Cat are indicated by maroon diamonds and grey plus signs, respectively. Error bars represent 95% confidence intervals (n = 3), and dotted lines are intended to be a guide for the eyes. These findings are consistent with reports in the literature that indicate BSA has molecular chaperone-like properties.28,

29

Marini and co-workers first recognized that BSA can act as a

molecular chaperone, reporting that addition of BSA to an enzyme solution significantly reduced thermally induced aggregation.28 Finn and others also noted that BSA inhibits aggregation and even amyloid formation,29 consistent with chaperone-like behavior. Therefore, similar interactions between enzyme and BSA might be conserved in the AGs.


Page 11 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


The inhibition of aggregation shown in Figure 4 and Figure 5 could indicate that thermal denaturation (and thus, inactivation) is reduced upon albumination; alternatively, BSA encasement may be preventing the aggregation of thermally denatured enzyme. Further testing of functional characteristics after exposure to elevated temperatures is necessary to elucidate these findings and is the subject of the Thermostability section to follow. Long-Term Storage Stability. Samples of BSA-GOx and native GOx were stored in glucosefree phosphate-buffered saline (PBS) for four weeks at 37 °C, and enzymatic activity was assayed at various time points to study the effect of albumination on spontaneous deactivation during long-term storage. As shown in Figure 6A, BSA-GOx has 78.4% of the specific activity of native GOx at the initial time point. This could be due to inactivation of a fraction of the enzyme during attachment of BSA or diffusional limitations imparted by the BSA shell. Despite having a lower initial specific activity, the data show that after four weeks, BSA-GOx is approximately three times as active as native GOx (74.4 U/mg vs. 26.0 U/mg, respectively). The activity data were normalized and fit with exponential decays (Figure 6B). For native GOx, a first-order exponential decay fits quite well (R2 > 0.99) with a decay constant of 0.1206/day, corresponding to a half-life of 5.75 days. BSA-GOx is more than 2.5 times as stable, with a decay constant of 0.0453/day (R2 ~ 0.97) and a half-life of 15.3 days.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 24

Figure 6. Specific activity (A) and normalized activity (B) of BSA-GOx (maroon squares) and native GOx (gray crosses) in the absence of glucose over four weeks at 37 °C. Error bars represent 95% confidence intervals (n = 3). The data were fit with first-order exponential decays; the maroon lines corresponds to the BSA-GOx data and the gray lines corresponds to the native GOx data. Operational Stability. As GOx reacts with glucose and oxygen, hydrogen peroxide is produced, which has been shown to deactivate the enzyme at a faster rate than spontaneous deactivation.42 Our group has previously reported peroxide-mediated deactivation for PEGylated GOx,43 as well as glutaraldehyde-modified PEGylated GOx.44 Immobilization of GOx has been shown to largely mitigate deactivation by hydrogen peroxide, presumably due to increased rigidity.45 To test for any protective effect imparted by albumination, native GOx and BSA-GOx were continuously exposed to glucose-containing PBS at room temperature for up to 24 h. As


Page 13 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


shown in Figure 7, after 6 h of continuous glucose exposure, BSA-GOx retained less of its initial activity than native GOx (72.4% vs. 90.7%). Likewise, at the 12 h time point, BSA-GOx retained less initial activity (44.8% vs. 55.7%). However, after 24 h, both native GOx and BSA-GOx lost approximately 75% of their initial specific activity. As albumination does not appear to stabilize GOx against deactivation by hydrogen peroxide, it follows that albumination does not significantly rigidify the native enzyme.

Figure 7. Normalized activity of BSA-GOx (maroon squares) and native GOx (gray crosses) exposed to glucose for 24 h at room temperature. Error bars represent 95% confidence intervals (n = 3), and dotted lines are intended to be a guide for the eyes. Thermostability. For many proteins, exposure to elevated temperatures induces thermal denaturation and loss of tertiary and/or quaternary structure, which is often accompanied by a loss of functional properties (e.g., enzymes lose catalytic abilities). Thermal denaturation of GOx results in an irreversible transition to a compact denatured state with the enzyme’s dimeric structure preserved but dissociation of the FAD cofactor.39 To determine whether albumination imparts stability at elevated temperatures, enzymatic activity was assayed following exposure to 60 °C for up to 1 h. Figure 8 shows that BSA-GOx is remarkably more stable than native GOx. After 1 h, native GOx only retains 1.42% of its initial activity, while BSA-GOx retains 37.7% of its initial activity—a 26.5-fold improvement in stability.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 24

Figure 8. Normalized activity of BSA-GOx (maroon squares) and native GOx (gray crosses) with exposure to extreme temperature (60 °C). Error bars represent 95% confidence intervals (n = 3). Native GOx data were fit with a first-order exponential decay (gray line), while BSA-GOx data were fit with both a first-order (maroon solid line) and second-order exponential decay (maroon dashed line). The data were fit with first-order exponential decays (solid lines). For native GOx, a first-order exponential decay fit well (R2 > 0.99) with a decay constant of 0.1157/min, corresponding to a half-life of 5.99 min. The data for BSA-GOx were also fit with a first-order exponential decay, which yielded a decay constant and half-life of 0.0144/min and 48.1 min, respectively; however, with a coefficient of determination of approximately 0.91, it was determined that the behavior might be better described using a multi-exponential decay. As such, the activity data for BSAGOx were fit with a second-order exponential decay (dashed line), which fit quite well as expected (R2 = 1). Two-thirds of the enzyme population had a long half-life of 71.7 min, while the remaining one-third of the enzyme population had a short half-life of 3.53 min. Therefore, it appears that albumination imparts a protective effect upon a majority of the enzyme population, translating to an overall increase in the thermostability of the AG compared to the native GOx.


Page 15 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


These results agree well with others’ findings, which indicate that even non-covalent interactions with BSA can improve the thermal stability of enzymes.27-30 In summary, we have demonstrated that two glycoenzymes, GOx and Cat, can be bioconjugated with BSA to form AGs. Liquid chromatographs exhibit a large shift in the elution volume accompanied by a doubling of the UV:visible absorbance ratio following albumination, which demonstrates successful conjugation of the enzymes with BSA. DLS indicates that the AGs are approximately 60% larger than the respective native glycoenzymes, providing secondary confirmation that modification has occurred. Both BSA-GOx and BSA-Cat resist aggregation during heating to 90 °C, but this effect is more profound in the latter AG. This seems to suggest that the previously reported molecular chaperone-like properties of BSA might be conserved in the AGs. The response of BSA-GOx to the test conditions, in terms of its functional characteristics, vary considerably. BSA-GOx thermostability is excellent compared to the native glycoenzyme, and moderate stabilization is observed with storage for one month at physiological temperature in the absence of glucose, but operational stability of GOx appears to be unaffected by albumination. Consequently, careful consideration of the expected conditions is recommended before BSAGOx is utilized in a particular application. EXPERIMENTAL PROCEDURES Materials. GOx from Aspergillus niger was obtained from BBI Enzymes (GO3A, 270 U/mg solid, 75% protein by Lowry). Cat from A. niger was obtained from EMD Millipore (219261, 6450 U/mg solid, 37.6% protein by UV absorbance) and purified by gel filtration chromatography to remove contaminants. BSA (A7906) and peroxidase from Amoracia



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 24

rusticana (type II, 188 U/mg solid) were obtained from Sigma. N-(β-maleimidopropionic acid) hydrazide (BMPH) was obtained from Thermo Scientific. Preparation of AGs. The procedure used to synthesize the AGs is depicted schematically in Figure 9. BSA (74.25 mg, 1) was dissolved in 1.85 mL of 100 mM sodium phosphate containing 154 mM NaCl (pH 7.2). BMPH was dissolved in 150 µL of dimethyl sulfoxide at a concentration of 50 mM and added to the BSA solution, yielding a 6.67 molar excess of BMPH to BSA. The reaction solution was mixed for 2 h at room temperature, followed by removal of excess BMPH using a desalting column equilibrated with 100 mM sodium phosphate containing 154 mM NaCl (pH 7.2) to yield hydrazide-functionalized BSA (2).

Figure 9. Synthesis of BSA-modified glycoenzymes—AGs. Oxidation of multiple glycosylation sites on the glycoenzyme yields aldehydes, which are then reacted with hydrazide-functionalized BSA. The single hydrazide group of each BSA is covalently attached at the oxidized glycosylation sites, and the resulting hydrazone linkages are reduced with NaBH3CN for stability. The glycoenzyme (3) was dissolved in 100 mM sodium phosphate containing 154 mM NaCl (pH 7.2) at a concentration of 3.33 mg/mL for GOx and 5.22 mg/mL for Cat. Separately, 4.3 mg of NaIO4 was dissolved in 100 µL of deionized water, then immediately added to 900 µL of the glycoenzyme solution and slowly agitated in the dark for 1 h at room temperature to yield 4. It is important to note that proteins exposed to oxidants such as NaIO4 have been reported to form higher-order oligomers in certain cases due to intermolecular crosslinking. However, Nakamura


Page 17 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


and others exposed GOx from A. niger to a five-fold higher concentration of NaIO4 for 5 h and found that the size and shape of the oxidized GOx was essentially the same as the native GOx.46 Nevertheless, a small amount of aggregated glycoenzyme might be observed following oxidation, but it will be removed during subsequent desalting or purification. The reaction was quenched by the addition of excess glycerol (1.25 µL). The oxidized glycoenzyme was diluted to 1.75 mL by the addition of 750 µL of 100 mM sodium phosphate containing 154 mM NaCl (pH 7.2), then purified using a desalting column equilibrated with the same phosphate buffer to remove excess NaIO4, glycerol, and degradation products from glycerol quenching (i.e., formaldehyde and formic acid). The solutions of oxidized glycoenzyme and hydrazide-functionalized BSA were combined, yielding a ca. 60-fold molar excess of BSA to glycoenzyme. The mixture was reacted in the dark for 2 h at room temperature under gentle agitation to yield 5. In a fume hood, 40 µL of 5 M NaBH3CN in 1 N NaOH was added. Caution: NaBH3CN is extremely toxic; as such, all operations should be performed with care in a fume hood. The NaBH3CN was reacted with the bioconjugate for 30 min at room temperature under gentle agitation to yield the AG (6). Finally, the conjugate was purified from low-molecular-weight contaminants by diafiltration and transferred into 10 mM sodium phosphate containing 154 mM NaCl (pH 7.2). Liquid Chromatography. Gel filtration chromatography was performed to purify the AGs and approximate their size. The crude sample was injected into a liquid chromatography system (GE Healthcare Life Sciences model ÄKTAexplorer 10) equipped with a gel-filtration column (GE Healthcare Life Sciences model HiLoad Superdex 200 PG) equilibrated with phosphate buffer (10 mM sodium phosphate, 154 mM NaCl, pH 7.2). Absorbance was monitored at 280 nm and 450 nm (for GOx) or 405 nm (for Cat), and 1 mL fractions were collected.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 24

Particle Sizing. A photon correlation spectrometer (Malvern model Zetasizer Nano ZS) was used to acquire size distributions of the AGs, as well as the native glycoenzymes and BSA, to determine the change in size following modification. Also, it is important to determine whether oligomerization occurred during oxidation or subsequent conjugation. Disposable 3.5 mL cuvettes were filled with 1 mL enzyme in 10 mM sodium phosphate containing 154 mM NaCl (pH 7.2). Temperature Ramp. A photon correlation spectrometer (Malvern model Zetasizer Nano ZS) was used to determine the effect of heating on the size of AGs and native glycoenzymes (n = 3). Standard 3.5 mL glass cuvettes were filled with enzyme (1 mL, ca. 0.735 mg/mL) in 10 mM sodium phosphate containing 154 mM NaCl (pH 7.2). The temperature of the cuvette holder was incrementally increased from 25 °C to 90 °C in steps of 5 °C, and the enzyme solution was allowed to equilibrate to each temperature for 2 min prior to collection of sizing data. Enzymatic Activity Assays. Enzymatic assays of BSA-GOx and native GOx samples were performed to determine activity following exposure to various challenges.47 In all cases, enzymatic activity measurements were performed in triplicate at pH 5.1 and 35 °C using a UV/Vis spectrophotometer (Agilent model Cary 300) equipped with a water-thermostatted multicell holder to maintain the appropriate temperature. Absorbance data were collected at approximately 30 Hz for approximately 2.5 min. Data were fit via linear regression using the maximum number of points to obtain a norm of residuals less than 0.025. This procedure was adopted to minimize the effects of nonlinearities present at later time points in absorbance data collected for samples of high catalytic activity. Long-Term Storage Stability. To observe the effects of albumination on the spontaneous denaturation of GOx, enzymatic activity was assayed over a period of four weeks with the


Page 19 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


enzyme samples stored at 37 °C (elevated temperature to accelerate deactivation and simulate physiological conditions) in the absence of glucose. To limit sample contamination, multiple samples were prepared for each enzyme type and each time point to prevent the need for samples to be repeatedly opened; that is, samples were only opened on the day they were assayed for enzymatic activity. Operational Stability. To test for the effect of glucose on native GOx and BSA-GOx enzymatic activity, samples were continuously exposed to PBS containing glucose at room temperature for up to 24 h. For these experiments, a dynamic dialysis system in a microplate format (10-well MicroDialyzer) was obtained from Spectrum Labs (Note: this product has been discontinued). The MicroDialyzer has ten wells, each of which can be filled with up to 500 µL of sample that is in direct contact with a 12-14 kDa regenerated cellulose dialysis membrane (Spectrum Labs model Spectra/Por 2). Beneath the dialysis membrane, a stirred dialysate solution is pumped through the dialysis chamber. For our purposes, enzyme solutions (250 µL, ca. 0.25 mg/mL) were added to the wells (n = 3), and PBS containing glucose (100 mg/dL) was pumped through the dialysis chamber at 4 mL/min using a peristaltic pump. After the appropriate glucose exposure time (6, 12, or 24 h), glucose-containing PBS was removed from the dialysis chamber, and 1 L of fresh PBS (no glucose) was pumped through the system at 4 mL/min (ca. 4 h) to remove excess glucose and product from the sample wells. Because the samples may change volume due to osmotic and hydrostatic forces—and because there is a possibility for the membrane to be compromised during testing—the samples were transferred to a standard microplate and concentration was determined using a multimode microplate reader (Tecan model Infinite M200 PRO series). Finally, the samples were assayed for enzymatic activity as described above.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 24

Thermostability. To determine if attachment of BSA provided stabilization of the enzyme at elevated temperatures (i.e., where thermally-induced denaturation dominates), enzyme samples (0.73 mg/mL) were immersed into a 60 °C water bath for up to 1 h. Separate replicate samples (n = 3) were prepared for each time point (5 min, 15 min, and 1 h). Upon removal from the water bath, samples were immediately cooled to 4 °C, then assayed for enzymatic activity as described above. ACKNOWLEDGMENTS This work was supported, in part, by the National Institutes of Health (1R43DK093139) and the Texas Engineering Experiment Station. Dustin W. Ritter thanks the National Science Foundation for financial support through the Graduate Research Fellowship (NGE-0750732). The authors also gratefully acknowledge the Protein Chemistry Laboratory at Texas A&M University for performing gel-filtration chromatography and providing technical expertise, comments, and suggestions. Use of the Texas A&M University Materials Characterization Facility is acknowledged. REFERENCES (1) (2) (3) (4) (5)

Heller, A., and Feldman, B. (2008) Electrochemical glucose sensors and their applications in diabetes management. Chem. Rev. 108, 2482-2505. Brown, J. Q., Srivastava, R., and McShane, M. J. (2005) Encapsulation of glucose oxidase and an oxygen-quenched fluorophore in polyelectrolyte-coated calcium alginate microspheres as optical glucose sensor systems. Biosens. Bioelectron. 21, 212-216. Stein, E. W., Grant, P. S., Zhu, H., and McShane, M. J. (2007) Microscale enzymatic optical biosensors using mass transport limiting nanofilms. 1. Fabrication and characterization using glucose as a model analyte. Anal. Chem. 79, 1339-1348. Stein, E. W., Singh, S., and McShane, M. J. (2008) Microscale enzymatic optical biosensors using mass transport limiting nanofilms. 2. Response modulation by varying analyte transport properties. Anal. Chem. 80, 1408-1417. Singh, S., and McShane, M. (2010) Enhancing the longevity of microparticle-based glucose sensors towards 1 month continuous operation. Biosens. Bioelectron. 25, 10751081.


Page 21 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(6)

(7) (8) (9) (10) (11)

(12) (13) (14) (15) (16) (17)

(18) (19) (20)

Roberts, J. R., Park, J., Helton, K., Wisniewski, N., and McShane, M. J. (2012) Biofouling of polymer hydrogel materials and its effect on diffusion and enzyme-based lumniescent glucose sensor functional characteristics. J. Diabetes Sci. Technol. 6, 12671275. Clarke, W. L., Cox, D., Gonder-Frederick, L. A., Carter, W., and Pohl, S. L. (1987) Evaluating Clinical Accuracy of Systems for Self-Monitoring of Blood Glucose. Diabetes Care 10, 622-628. Quiocho, F. A., and Richards, F. M. (1964) Intermolecular cross linking of a protein in the crystalline state: carboxypeptidase-A. Proc. Natl. Acad. Sci. U. S. A. 52, 833. Cao, L., van Rantwijk, F., and Sheldon, R. A. (2000) Cross-Linked Enzyme Aggregates: A Simple and Effective Method for the Immobilization of Penicillin Acylase. Org. Lett. 2, 1361-1364. Cao, L., Langen, L. v., and Sheldon, R. A. (2003) Immobilised enzymes: carrier-bound or carrier-free? Curr. Opin. Biotechnol. 14, 387-394. Schoevaart, R., Wolbers, M. W., Golubovic, M., Ottens, M., Kieboom, A. P. G., van Rantwijk, F., van der Wielen, L. A. M., and Sheldon, R. A. (2004) Preparation, optimization, and structures of cross-linked enzyme aggregates (CLEAs). Biotechnol. Bioeng. 87, 754-762. Sheldon, R. A., Schoevaart, R., and Van Langen, L. M. (2005) Cross-linked enzyme aggregates (CLEAs): A novel and versatile method for enzyme immobilization (a review). Biocatal. Biotransform. 23, 141-147. Sheldon, R. (2011) Characteristic features and biotechnological applications of crosslinked enzyme aggregates (CLEAs). Appl. Microbiol. Biotechnol. 92, 467-477. Lopez-Cruz, J. I., Viniegra-Gonzalez, G., and Hernandez-Arana, A. (2006) Thermostability of native and pegylated Myceliophthora thermophila laccase in aqueous and mixed solvents. Bioconjugate Chem. 17, 1093-1098. Matsushima, A., Kodera, Y., Hiroto, M., Nishimura, H., and Inada, Y. (1996) Bioconjugates of proteins and polyethylene glycol: potent tools in biotechnological processes. J. Mol. Catal. B: Enzym. 2, 1-17. Rodriguez-Martinez, J. A., Rivera-Rivera, I., Sola, R. J., and Griebenow, K. (2009) Enzymatic activity and thermal stability of PEG-alpha-chymotrypsin conjugates. Biotechnol. Lett. 31, 883-887. Veronese, F. M., Caliceti, P., Schiavon, O., and Sartore, L. (1992) Preparation and properties of monomethoxypoly(ethylene glycol)-modified enzymes for therapeutic applications, in Poly(ethylene Glycol) Chemistry: Biotechnical and Biomedical Applications (Harris, J. M., Ed.) pp 127-137, Plenum Press, New York. Mateo, C., Palomo, J. M., Fernandez-Lorente, G., Guisán, J. M., and Fernández-Lafuente, R. (2007) Improvement of enzyme activity, stability and selectivity via immobilization techniques. Enzyme Microb. Technol. 40, 1451-1463. Rodrigues, R. C., Berenguer-Murcia, Á., and Fernández-Lafuente, R. (2011) Coupling Chemical Modification and Immobilization to Improve the Catalytic Performance of Enzymes. Adv. Synth. Catal. 353, 2216-2238. Rodrigues, R. C., Ortiz, C., Berenguer-Murcia, Á., Torres, R., and Fernández-Lafuente, R. (2013) Modifying enzyme activity and selectivity by immobilization. Chem. Soc. Rev. 42, 6290-6307.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(21)

(22) (23)

(24)

(25) (26) (27) (28) (29) (30) (31) (32) (33) (34) (35)

Page 22 of 24

Cantone, S., Ferrario, V., Corici, L., Ebert, C., Fattor, D., Spizzo, P., and Gardossi, L. (2013) Efficient immobilisation of industrial biocatalysts: criteria and constraints for the selection of organic polymeric carriers and immobilisation methods. Chem. Soc. Rev. 42, 6262-6276. Bolivar, J. M., Wilson, L., Ferrarotti, S. A., Guisán, J. M., Fernández-Lafuente, R., and Mateo, C. (2006) Improvement of the stability of alcohol dehydrogenase by covalent immobilization on glyoxyl-agarose. J. Biotechnol. 125, 85-94. Pessela, B. C. C., Mateo, C., Fuentes, M., Vian, A., Garcı́a, J. L., Carrascosa, A. V., Guisán, J. M., and Fernández-Lafuente, R. (2003) The immobilization of a thermophilic β-galactosidase on Sepabeads supports decreases product inhibition: Complete hydrolysis of lactose in dairy products. Enzyme Microb. Technol. 33, 199-205. Mateo, C., Monti, R., Pessela, B. C. C., Fuentes, M., Torres, R., Manuel Guisán, J., and Fernández-Lafuente, R. (2004) Immobilization of Lactase from Kluyveromyces lactis Greatly Reduces the Inhibition Promoted by Glucose. Full Hydrolysis of Lactose in Milk. Biotechnol. Progr. 20, 1259-1262. Koudelka-Hep, M., de Rooij, N. F., and Strike, D. J. (1997) Immobilization of Enzymes on Microelectrodes Using Chemical Crosslinking, in Immobilization of Enzymes and Cells (Bickerstaff, G. F., Ed.) pp 83-85, Humana Press. Wetzel, R., Becker, M., Behlke, J., Billwitz, H., Böhm, S., Ebert, B., Hamann, H., Krumbiegel, J., and Lassmann, G. (1980) Temperature Behaviour of Human Serum Albumin. Eur. J. Biochem. 104, 469-478. Chang, B. S., and Mahoney, R. R. (1995) Enzyme thermostabilization by bovine serum albumin and other proteins: evidence for hydrophobic interactions. Biotechnol. Appl. Biochem. 22, 203-214. Marini, I., Moschini, R., Corso, A. D., and Mura, U. (2005) Chaperone-like features of bovine serum albumin: a comparison with α-crystallin. Cell. Mol. Life Sci. 62, 30923099. Finn, T. E., Nunez, A. C., Sunde, M., and Easterbrook-Smith, S. B. (2012) Serum Albumin Prevents Protein Aggregation and Amyloid Formation and Retains Chaperonelike Activity in the Presence of Physiological Ligands. J. Biol. Chem. 287, 21530-21540. Myung, S., Zhang, Y. H. P., and Sanchez Ruiz, J. (2013) Non-Complexed Four Cascade Enzyme Mixture: Simple Purification and Synergetic Co-stabilization. PLoS ONE 8, e61500. Zimmerman, S. B., and Trach, S. O. (1991) Estimation of macromolecule concentrations and excluded volume effects for the cytoplasm of Escherichia coli. J. Mol. Biol. 222, 599-620. Gouda, M. D., Singh, S. A., Rao, A. G. A., Thakur, M. S., and Karanth, N. G. (2003) Thermal inactivation of glucose oxidase - Mechanism and stabilization using additives. J. Biol. Chem. 278, 24324-24333. Jones, M. N., Manley, P., and Wilkinson, A. (1982) The dissociation of glucose oxidase by sodium n-dodecyl sulphate. Biochem. J. 203, 285-291. Fernández-Lafuente, R. (2009) Stabilization of multimeric enzymes: Strategies to prevent subunit dissociation. Enzyme Microb. Technol. 45, 405-418. Tsuge, H., Natsuaki, O., and Ohashi, K. (1975) Purification, properties, and molecular features of glucose oxidase from Aspergillus niger. J. Biochem. 78, 835-843.


Page 23 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(36) (37) (38) (39) (40)

(41) (42) (43) (44) (45)

(46) (47)

Pazur, J. H., Kleppe, K., and Ball, E. M. (1963) The glycoprotein nature of some fungal carbohydrases. Arch. Biochem. Biophys. 103, 515-516. Kikuchi-Torii, K., Hayashi, S., Nakamoto, H., and Nakamura, S. (1982) Properties of Aspergillus niger Catalase. J. Biochem. 92, 1449-1456. Squire, P. G., Moser, P., and O'Konski, C. T. (1968) Hydrodynamic properties of bovine serum albumin monomer and dimer. Biochemistry 7, 4261-4272. Zoldák, G., Zubrik, A., Musatov, A., Stupák, M., and Sedlák, E. (2004) Irreversible Thermal Denaturation of Glucose Oxidase from Aspergillus niger Is the Transition to the Denatured State with Residual Structure. J. Biol. Chem. 279, 47601-47609. Seymour, S. L., and Klinman, J. P. (2002) Comparison of rates and kinetic isotope effects using PEG-modified variants and glycoforms of glucose oxidase: the relationship of modification of the protein envelope to C−H activation and tunneling. Biochemistry 41, 8747-8758. Haouz, A., Glandières, J. M., and Alpert, B. (2001) Involvement of protein dynamics in enzyme stability: The case of glucose oxidase. FEBS Lett. 506, 216-220. Tse, P. H. S., and Gough, D. A. (1987) Time-dependent inactivation of immobilized glucose oxidase and catalase. Biotechnol. Bioeng. 29, 705-713. Ritter, D. W., Roberts, J. R., and McShane, M. J. (2013) Glycosylation site-targeted PEGylation of glucose oxidase retains native enzymatic activity. Enzyme Microb. Technol. 52, 279-285. Ritter, D. W., Newton, J. M., and McShane, M. J. (2014) Modification of PEGylated enzyme with glutaraldehyde can enhance stability while avoiding intermolecular crosslinking. RSC Adv. 4, 28036-28040. Betancor, L., López-Gallego, F., Hidalgo, A., Alonso-Morales, N., Dellamora-Ortiz, G., Guisán, J. M., and Fernández-Lafuente, R. (2006) Preparation of a very stable immobilized biocatalyst of glucose oxidase from Aspergillus niger. J. Biotechnol. 121, 284-289. Nakamura, S., Hayashi, S., and Koga, K. (1976) Effect of periodate oxidation on the structure and properties of glucose oxidase. Biochim. Biophys. Acta, Enzymol. 445, 294308. Bergmeyer, H. U., Gawehn, K., and Grassl, M. (1974) Methods of enzymatic analysis, Vol. I, 2nd ed., Academic Press, New York.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Table of Contents Graphic:

Glucose oxidase, catalase, etc.

Reduced aggregation & thermal inactivation; Improved long-term storage stability


Page 24 of 24

Albuminated Glycoenzymes: Enzyme Stabilization through Orthogonal

Recommend Documents