PEGylation but not Fc-fusion improves in vivo residence time of a thermostable mutant of bacterial cocaine esterase

Abstract

It is very popular to fuse a protein drug or drug candidate to the Fc domain of immunoglobulin G (IgG) in order to prolong the in vivo half-life. In this study, we have designed, prepared, and tested an Fc-fused thermostable cocaine esterase (CocE) mutant (known as E196–301, with the T172R/G173Q/L196C/I301C substitutions on CocE) expressed in E. Coli. As expected, Fc-fusion does not affect the in vitro enzyme activity and thermal stability of the enzyme and that Fc-E196–301 can favorably bind FcRn with K_d = 386±35 nM. However, Fc-fusion does not prolong the in vivo half-life of E196–301 at all; Fc-E196–301 and E196–301 have essentially the same PK profile (t_1/2 = 0.4±0.1 h) in rats. This is the first time demonstrating that Fc-fusion does not prolong in vivo half-life of a protein. This finding is consistent with the mechanistic understanding that E196–301 and Fc-E196–301 are all degraded primarily through rapid proteolysis in the body. The Fc fusion cannot protect E196–301 from the proteolysis in the body. Nevertheless, it has been demonstrated that PEGylation can effectively protect E196–301, as the PEGylated E196–301, i.e. PEG-E196–301, has a significantly prolonged in vivo half-life. It has also been demonstrated that both E196–301 and PEG-E196–301 have dose-dependent in vivo half-lives (e.g. 19.9±6.4 h for the elimination t_1/2 of 30 mg/kg PEG-E196–301), as the endogenous proteolytic enzymes responsible for proteolysis of E196–301 (PEGylated or not) are nearly saturated by the high plasma concentration produced by a high dose of E196–301 or PEG-E196–301.

Graphical Abstract

Introduction

Cocaine dependence and overdose are major medical and public problems that continue to defy treatment^1–3. Cocaine exerts its psychotropic effects by binding dopamine transporter to block dopamine reuptake^4–5. Despite decades of efforts, classical pharmacodynamic approach to develop a clinically effective dopamine receptor/transporter antagonist has not generated a clinically useful anti-cocaine medication^6–8. Inactivating cocaine in the circulation system to interfere the delivery of cocaine into the central nervous system is an alternative pharmacokinetic approach to address this problem, which can be done by binding cocaine with another molecule such as antibody or accelerating cocaine metabolism into biologically inactive metabolites^9–12. The pharmacokinetic approach using a highly active cocaine-metabolizing enzyme should be more efficient, depending on the catalytic rate constant (k_cat) and Michaelis-Menten constant (K_M). Unlike the antibody/vaccine approach, one enzyme molecule can repeatedly degrade drug molecules. Hence, it would be an ideal medication to develop a highly active exogenous enzyme which can efficiently metabolize cocaine into physiologically inactive metabolites.

Bacterial cocaine esterase (CocE), isolated from the Rhodococcus sp. strain MB1 found in the rhizosphere soil surrounding the coca plant¹³, is known as the most efficient natural enzyme for hydrolyzing cocaine into biologically inactive metabolites. CocE has been shown highly effective against the toxicity of a lethal dose of cocaine in rodents¹⁴. However, a major barrier to clinically utilize CocE is its poor thermostability, with an in vitro half-life of about 12 minutes at physiological temperature (37 °C)¹³. Fortunately, one of our previously reported thermostable mutants, CocE-T172R/G173Q mutant (which is also known as RBP-8000 or TNX-1300) designed through our structure-and-mechanism-based protein design platform¹³, has been shown to be safe and efficacious in a randomized, double-blind, placebo-controlled clinical trial for cocaine toxicity treatment¹⁵. Further, for cocaine dependence treatment, one would like to have a cocaine-metabolizing enzyme with both high cocaine-eliminating capability and a residence time (in vivo half-life) as long as possible to achieve the long-time abstinence from cocaine. Clinical trial results revealed that RBP-8000 had an in vivo half-life ~1.8 h at the dose of 100 mg in humans, good enough for cocaine overdose treatment. However, the in vivo half-life of ~1.8 h is clearly too short to be effective for cocaine dependence treatment.

Recently, we designed a promising new mutant of E172–173 with extra L196C/I301C substitutions (denoted as enzyme E196–301), which introduces a pair of cross-subunit disulfide bonds on the dimer interface. Compared to E172–173, E196–301 has not only a considerably prolonged in vitro half-life (>100 days) at 37 °C, but also a significantly improved catalytic efficiency (k_cat/K_M) against cocaine (about 150% increase)¹⁶. Our current effort to further develop E196–301 for cocaine dependence treatment is focused on extending its residence time in vivo. A well-known approach for in vivo half-life extension is genetic fusion of the functional protein to the Fc domain of immunoglobulin G (IgG). Generally speaking, Fc fusion enables interaction with the neonatal Fc receptor (FcRn) at acidic pH (~6), which protects bound molecule from intracellular lysosomal degradation by recycling them back to the circulation¹⁷. Fc-FcRn interaction contributes to the extraordinarily long in vivo half-life of IgG proteins in humans. Therefore, engineering protein to an Fc fusion protein form has the potential to significantly increase the in vivo half-life^18–20. Using this technology, we successfully extended another protein-based therapeutic entity to ~107 hours in rats²¹.

In this study, we first tried to test Fc fusion to E196–301 with the goal to extend the in vivo half-life. The C-terminus of E196–301 was fused to the N-terminus of the hinge region of Fc with a Gly₆Ser linker between them. Notably, given that E196–301 has been proven as a dimer with a pair of cross-unit disulfide bonds, a previously used monomeric Fc variant, known as Fc(M1) (the A1Q/C6S/C12S/C15S/P24S mutant of human IgG1 Fc), for fusion with another protein²¹, was used for fusing to E196–301. The purpose of using a monomeric Fc(M1) variant, instead of wild-type Fc, is to avoid the disulfide-bonding mediated Fc-Fc dimerization within the hinge domain such that the E196–301 region of the fusion protein will still retain a dimer structure, as shown in Figure 1A. The designed Fc-E196–301 fusion protein was prepared and tested for its in vitro activities and in vivo profiles. Based on the in vitro data obtained, E196–301 fusion with Fc(M1) did not change the high catalytic activity and thermal stability of E196–301, and the fusion protein Fc-E196–301 revealed significant binding ability to FcRn, as expected. However, the protein fusion with Fc(M1) did not significantly prolong the in vivo half-life of E196–301 according to pharmacokinetic data in rats. Then, we aimed to understand why the Fc fusion did not prolong the in vivo half-life of E196–301 at all through further analysis. Based on the mechanistic understanding, PEGylation was then performed to extend the in vivo half-life of E196–301. Indeed, a PEGylated E196–301 had a significantly prolonged in vivo half-life compared to E196–301 and Fc-E196–301.

Figure 1.

Molecular structure and in vitro characterization of Fc-E196–301 protein. (A) The dimer structure of Fc-E196–301; (B) SDS-PAGE gel for E196–301 (lane 1) and Fc-E196–301 (lane 2); (C) Plot of the remaining enzyme activity (measured in triplicate) vs time (day) of incubation at 37 °C. (D) Plot of the Fc-E196–301 (or E196–301) binding to FcRn (measured in triplicate) vs the Fc-E196–301 concentration at pH 6.

Results and Discussion

In vitro characterization of the designed Fc-E196–301.

We first wanted to know whether the fusion protein Fc-E196–301 can be expressed in E. Coli. as a soluble recombinant protein or not. Previously, we reported a series of CocE mutants^{13, 16} with in vitro and in vivo characterization, demonstrating that all of the previously tested CocE mutants including E196–301¹⁶ can be expressed as soluble proteins. To minimize the possible systematic errors, we simultaneously prepared E196–301 and Fc-E196–301 under the same experimental conditions. After centrifugation of the cell-lysates, the soluble supernatant parts were applied to a cobalt-chelating resin for purification via the His₆-tag on the N-terminus. As seen in Figure 1B, Fc-E196–301 was expressed as a soluble recombinant protein, and catalytic activity of purified Fc-E196–301 against cocaine is the same as that of the unfused E196–301 (data not shown here)¹⁶. From a 1 L culture, we were able to routinely obtain approximately 50 mg of soluble and purified Fc-E196–301 protein.

Based on the encouraging thermal stability data of E196–301 reported previously¹⁶, we also tested the stability of Fc-E196–301 at 37 °C. Not surprisingly, as shown in Figure 1C, Fc-E196–301 still retained ~50% of the enzyme activity after incubation at 37 °C for 100 days, which is comparable to that of the unfused protein E196–301. The rationale of the Fc fusion technology is that the Fc domain of the Fc-fused protein will bind to FcRn at an acidic pH (~6) and, thus, the fusion protein will be salvaged from intracellular lysosomal degradation in the same manner as IgG. To confirm the binding ability of Fc-E196–301 to FcRn, we performed the EIA-based in vitro FcRn binding assay at pH 6. Fc-E196–301 showed the capability of binding rat FcRn in a dose-dependent manner, with K_d (dissociation constant) = 386 ± 35 nM. The K_d value of human IgG1 to rat FcRn was reported to be ~1000 nM²², which is close to 386 ± 35 nM of Fc-E196–301 determined in this study, suggesting that our expressed and purified Fc-E196–301 protein was able to bind rat FcRn in a similar way that human IgG1 does.

Biological half-life of Fc-E196–301 in comparison with E196–301.

The Fc-E196–301 protein was tested for its pharmacokinetic (PK) profile in rats along with E196–301. Rats (n = 3 for each group) were administered intravenously (IV) with the purified Fc-E196–301 or E196–301 at a dose of 1 mg/kg of body weight. The blood samples were collected at 2, 5, 10, 15, 30, and 60 min after the enzyme injection. Depicted in Figure 2A are the time courses of the active enzyme concentrations remained in the circulation after the IV administration. As shown in Figure 2A, we did not observe any extension of the in vivo half-life when comparing the time course of Fc-E196–301 to that of E196–301. There was no significant difference (p = 0.7041) between the time courses of Fc-E196–301 and E196–301, according to the two-way (protein entity × time) analysis of variance (ANOVA) using the GraphPad Prism 7 software. As seen in Figure 2A, the measured residual enzymatic activity over time followed a single-exponential equation (one-phase decay), rather than the usually observed double-exponential equation for the two-phase decay including the fast distribution phase and the slow elimination phase. In other words, the elimination phase is as fast as the distribution phase for both Fc-E196–301 and E196–301.

Figure 2.

Time courses of E196–301 and Fc-E196–301 proteins in rats. (A) Plots of the remaining active enzyme concentrations vs time (min) after IV administration of 1 mg/kg E196–301 or Fc-E196–301 (n=7 per group); (B) Western blots of the serum samples collected from rats at various time points; (C) Plots of the remaining active enzyme concentrations vs time (h) after IV administration of 1, 10, or 30 mg/kg E196–301 (n=3–7 per group).

Concerning why Fc-E196–301 did not have an improved biological half-life, compared to E196–301, despite its binding with FcRn, there might be a couple of possibilities. First, through binding with FcRn and the FcRn-mediated recycling mechanism discussed above, the Fc-E196–301 protein might have a longer in vivo half-life in rats, but in an inactive form such that we were unable to detect its enzyme activity in rat plasma. The other possibility is that Fc-E196–301 was degraded rapidly in the body. To explore the first possibility, we further analyzed the Fc-E196–301 protein concentrations in blood samples using western blotting, allowing us to semi-quantitatively determine the relative amounts of Fc fusion protein in the blood samples no matter whether the fusion protein is enzymatically active or not. Depicted in Figure 2B are the results of western blotting analysis on the blood samples from rats. As shown in Figure 2B, full-length Fc-E196–301 protein was detected with an expected molecular weight of ~110 kDa at different time points of the PK study, and the full-length Fc-E196–301 protein band rapidly became weaker and disappeared. The change of the protein band from western blotting (Figure 2B) is consistent with the decrease of the measured enzyme activity (Figure 2A). These data exclude the possibility of Fc-E196–301 circulating as an enzymatically inactive form. So, it is more likely that both E196–301 and Fc-E196–301 were degraded rapidly in a same rate in vivo. Thus, the Fc fusion was able to generate the desirable binding affinity of the fusion protein Fc-E196–301 with FcRn, but unable to slow down the rapid degradation of E196–301 in vivo.

E196–301 was degraded in a dose-dependent manner in vivo.

It has been known that another CocE mutant, i.e. CocE-T172R/G173Q, was eliminated primarily through rapid proteolysis in the body, as revealed by the results of radio-labeled CocE-T172R/G173Q eliminating assay and immunohistochemistry data in rats²³. In comparison with CocE-T172R/G173Q, E196–301 only has extra L196C/I301C substitutions designed to form a pair of cross-subunit disulfide bonds in order to thermally stabilize the dimer structure without changing any residues on the dimer surface¹⁶. Our finding that the Fc-E196–301 and E196–301 proteins had the similarly short in vivo half-lives suggests that both Fc-E196–301 and E196–301 were also degraded primarily through rapid proteolysis in the body. The Fc fusion did not protect E196–301 from proteolysis in the body. To examine this mechanistic understanding/assumption, we further examined the PK profiles of E196–301 at different doses in rats. As shown in Figure 2C and Table 1, the in vivo half-life (t_1/2) of E196–301 was dependent on the dose: t_1/2 = 0.4 ± 0.1 h at the dose of 1 mg/kg, 2.1 ± 0.5 h at the dose of 10 mg/kg dosing, and 2.4 ± 0.3 h at the dose of 30 mg/kg (see Table 1). The dose-dependence of the observed in vivo half-life is consistent with the aforementioned mechanism that the enzyme was eliminated primarily through rapid proteolysis in the body, suggesting that the endogenous proteolytic enzymes responsible for proteolysis of E196–301 were nearly saturated by the high plasma concentration of E196–301 provided by a high dose (10 or 30 mg/kg).

Table 1.

In vivo half-lives of E196–301 and PEG-E196–301 and their dose dependence

Enzyme	Dose (mg/kg)	In vivo t1/2 (h)a
E196–301	1	0.4±0.1
E196–301	10	2.1±0.5
E196–301	30	2.4±0.3
PEG-E196–301	1	0.10±0.05 (distribution); 3.7±0.6 (elimination)
PEG-E196–301	10	1.2±0.7 (distribution); 16.4±6.9 (elimination)
PEG-E196–301	30	1.1±0.5 (distribution); 19.9±6.4 (elimination)

^aDetermined by fitting to the single-exponential equation (one-phase decay) for E196–301 or the double-exponential equation (two-phase decay) for PEG-E196–301.

Effects of PEGylation on the in vivo half-life of E196–301.

Now that the popularly used Fc fusion technology failed to prolong the in vivo half-life of E196–301, we wanted to know how effective PEGylation can protect E196–301 from the rapid proteolysis in the body. According to the previous cocaine detoxification tests in rats¹⁶, a PEGylated E196–301 protein at a single dose of 30 mg/kg (IV) fully protected mice from a daily lethal dose of cocaine (180 mg/kg, LD₁₀₀) for at least three days. However, the PK profile of E196–301 (PEGylated or not) has not been characterized in any animal species. In this study, we determined the PK profiles of both the un-PEGylated and PEGylated E196–301 proteins for the first time.

Depicted in Figure 3A is the PEGylation reaction used in this study. The purified enzymes, including E196–301, Fc-E196–301, and PEGylated E196–301 (denoted as PEG-E196–301 for convenience), were analyzed by SDS-PAGE. As shown in Figure 3B, PEG-E196–301 (lane 3) displayed a much higher overall molecular weight than E196–301 (un-PEGylated enzyme) and Fc-E196–301. The protein bands are reasonably close to the anticipated molecular weights, i.e. ~65 kDa for E196–301, ~110 kDa for Fc-E196–301, and ~210 kDa for PEG-E196–301 in which each subunit of the dimer is covalently linked to a PEG 40 kDa molecule. The active enzyme concentration in the plasma sample collected at each time-point was analyzed by detecting the enzyme activity against cocaine as described above.

Figure 3.

PEGylation reaction to prepare PEG-E196–301 and its characterization. (A) The PEGylation reaction; (B) SDS-PAGE gel for Fc-E196–301 (lane 1), E196–301 (lane 2), PEG-E196–301 (lane 3), and the marker (lane 4); (C) Plots of the remaining active enzyme concentrations vs time (h) after IV administration of 1 mg/kg E196–301 or PEG-E196–301 (n=4 per group); (D) Plots of the remaining active enzyme concentrations vs time (h) after IV administration of 1, 10, or 30 mg/kg PEG-E196–301 (n=3 or 4 per group).

Given in Figure 3C are the time courses of the remaining enzymatic activity after IV administration of E196–301 or PEG-E196–301, showing that PEGylation remarkably slowed down the enzyme elimination in rats. Notably, the time-dependent active enzyme (PEG-E196–301) concentration ([E]_t) followed the well-known double-exponential equation (two-phase decay)²⁴ implemented in the GraphPad Prism 7 software: (${[E]}_t=A{e}^{-k_{1}t}+B{e}^{-k_{2}t}$) which accounts for both the enzyme distribution process (the fast phase, associated with k₁) and elimination process (the slow phase, associated with k₂). The half-life (t_1/2) associated with the enzyme elimination rate constant k₂ is known as the elimination half-life or biological half-life.

Depicted in Figure 3D are the PK profiles of PEG-E196–301 associated with various doses, showing the dose-dependence of the PK profile in rats. As summarized in Table 1, both the distribution and elimination half-lives of PEG-E196–301 are dose-dependent. At a dose of 1 mg/kg, PEG-E196–301 had t_1/2(distribution) = 0.10 ± 0.05 h and t_1/2(elimination) = 3.7 ± 0.6 h. Increasing the dose from 1 mg/kg to 10 mg/kg, t_1/2(distribution) and t_1/2(elimination) considerably improved to 1.2 ± 0.7 h and 16.4 ± 6.9 h, respectively. Further increasing the dose from 10 mg/kg to 30 mg/kg, the changes in t_1/2(distribution) and t_1/2(elimination) were not significant: t_1/2(distribution) = 1.1 ± 0.5 h and t_1/2(elimination) = 19.9 ± 6.4 h when the dose was 30 mg/kg. These PK data suggest that the endogenous proteolytic enzymes responsible for the proteolysis of PEG-E196–301 were nearly saturated by the high plasma concentration of PEG-E196–301 associated with the high dose of 10 or 30 mg/kg.

Finally, we examined whether the enzyme(s) responsible for the rapid degradation of the exogenous proteases exist(s) in plasma by mixing the purified enzyme with the plasma collected from an untreated mouse and then measuring the cocaine hydrolysis activity of the mixture in vitro at different time points thereafter. The experimental data (not shown) revealed no significant changes in the enzyme activity over the time, suggesting that the protease(s) responsible for the rapid degradation of the enzyme do(es) not exist in plasma. As well known, there are numerous endogenous proteolytic enzymes in the body, including proteases in extracellular compartments, intracellular compartments, and the cell membrane on the surface²⁵. So, the specific protease(s) responsible for the rapid degradation of the exogenous enzyme in the body is/are still unknown. Further studies are required in order to identify specific enzyme(s) responsible for the rapid degradation of the exogenous enzyme.

Conclusion

In the present study, we have designed, prepared, and tested an Fc-fused thermostable CocE mutant (known as E196–301, with the T172R/G173Q/L196C/I301C substitutions on CocE), i.e. Fc-E196–301, expressed in E. Coli. It has been demonstrated that the Fc fusion does not affect the in vitro enzyme activity and thermal stability of the enzyme and that Fc-E196–301 can favorably bind rat FcRn with K_d = 386 ± 35 nM. However, it turns out that the Fc fusion, despite its binding with FcRn, does not prolong the in vivo half-life of E196–301 at all; Fc-E196–301 and E196–301 have essentially the same PK profile (t_1/2 = 0.4 ± 0.1 h) in rats. This finding is consistent with the mechanistic understanding/assumption that E196–301 and Fc-E196–301 are all degraded primarily through rapid proteolysis in the body. The Fc fusion is unable to protect E196–301 from proteolysis. Nevertheless, it has been demonstrated that PEGylation can effectively protect E196–301, as the PEGylated E196–301, i.e. PEG-E196–301, has a significantly prolonged in vivo half-life. It has also been demonstrated that both E196–301 (un-PEGylated protein) and PEG-E196–301 (PEGylated protein) have dose-dependent in vivo half-lives, as the endogenous proteolytic enzymes responsible for proteolysis of E196–301 (PEGylated or not) are nearly saturated by the high plasma concentration of E196–301 or PEG-E196–301 associated with a high dose (10 or 30 mg/kg).

Materials and Methods

Materials.

The cDNAs for E196–301 (T172R/G173Q/L196C/I301C mutant of bacterial CocE) and for Fc(M1) (A1Q/C6S/C12S/C15S/P24S mutant of human IgG1 Fc) were generated in our previous studies^{16, 21}. The protein expression vector of pET-28a was provided by Dr. Jon Thorson at the University of Kentucky. Phusion DNA polymerase, restriction endonucleases, T4 DNA ligase and HisPur™ Cobalt Resin were purchased from Thermo Fisher Scientific (Waltham, MA). Branched polyethylene glycol (PEG) with molecular weight of 40 kDa was purchased from JenKem. (−)-Cocaine was provided by the National Institute on Drug Abuse Drug Supply Program, and [³H](−)-Cocaine (50 Ci/mmol) was ordered from PerkinElmer. Sprague-Darley rats (200 – 250 g) were ordered from Harlan. All other chemicals were purchased from Thermo Fisher Scientific.

Construction of plasmid.

The expression plasmid pET28a-E196–301-G₆S-Fc(M1) was constructed, containing a sequence encoding E196–301 followed by a Gly₆Ser sequence linked to the N-terminus of Fc(M1). Overlap extension PCR was used to fuse E196–301 to Fc(M1). Primers used are: P1: 5’-CGGCTAGCATGGTGGACGGGAAT-3’; P2: 5’-AGGCTCCTGGGAGCCACCGCCACCGCCACCTCGCTTGATAATCGGCAG-3’; P3:5’=ATCAAGCGAGGTGGCGGTGGCGGTGGCTCCCAGGAGCCTAAGTCCTCC-3’; P4: 5’-CCCAAGCTTCTATTTACCCGGAGACAG-3’. P1-P2 and P3-P4 primer pairs were used to amplify E196–301 and Fc(M1), respectively. The obtained two DNA fragments were fused by another PCR using P1 and P4 for the far ends. The final PCR product was digested with restriction enzymes Nhe I and Hind III and then was ligated into a pET-28a vector that had previously been digested with Nhe I and Hind III to obtain the pET28a-E196–301-G6S-Fc(M1) which encoded an N-terminal fusion of E196–301-G₆S-Fc(M1) with a His₆-tag.

Protein expression and purification.

E196–301 and Fc-E196–301 were expressed in Escherichia coli BL-21 (DE3) cells grown at 37 °C. Protein expression was induced with 0.5 mM isopropyl-β-thiogalactopyranoside for ~15 h at 18 °C. The harvested cells were resuspended in 50 mM Tris-HCl, pH 7.4, 150 mM NaCl buffer with protease inhibitor cocktail and then lysed using a French press. For each protein, the resulting cell debris was removed by centrifugation (15,000 g, 15 min). The supernatant was then purified by HisPur cobalt resin (Thermo Fisher Scientific) following the manufacturer’s instructions. The eluted fractions were concentrated by an Amicon Ultra-50K centrifuge (Millipore, Burlington, MA). The protein concentrations were determined using a Coomassie Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA) with bovine serum albumin as a standard. Purified proteins were analyzed by SDS/PAGE using a 4–12% Tris-Glycine Mini Protein Gel. The protein samples were mixed SDS-loading buffer and heated at 98 °C for 10 min in the presence of a reducing agent DTT. After electrophoresis, the protein gel was stained by a conventional Coomassie blue method.

ELISA assay.

In order to study the interaction between FcRn and Fc-E196–301, soluble and functional rat FcRn protein was prepared by using the method reported previously²⁶. The wells of the EIA clear flat bottom plate (Corning-Costar, NY, USA) were coated with 200 ng of His₆-tagged rat FcRn in coating buffer (0.05 M carbonate buffer, pH 9.6) and incubated overnight at 4 °C. After washing with EIA washing buffer (0.01 M potassium phosphate buffer, pH 6.0, 0.05% (v/w) Tween-20) for three times, the wells were blocked with blocking buffer (1 mg/ml casein in 0.01 M potassium phosphate buffer, pH 6.0). After washing as described above, Fc-E196–301 was serially diluted with the blocking buffer and added to the wells, followed by incubation for one hour at room temperature (RT). After extensive washing, horseradish peroxidase (HRP)-conjugated goat anti-human IgG Fc antibody (Novus Biologicals, Littleton, CO) was added to the wells and incubated for one hour at RT, followed by another extensive washing. Finally, visualization was done by the addition of 50 μL of ELISA developing solution (NEOGEN, Lexington, KY) containing 3,3’,5,5’-Tetramethylbenzidine (TMB) and hydrogen peroxide. The plate was incubated for 2 min with plate shaking and the OD at 610 nm was read in an automated ELISA plate reader (TECAN, Switzerland) for an hour. The steady state affinity constant was obtained using an equilibrium binding model supplied by the GraphPad Prism 7 software (GraphPad Software, La Jolla, CA). Each dot is the representative of the average of triplicates and its values are expressed as mean ± standard deviation.

PEGylation.

Purified E196–301 was conjugated with maleimide linked branched PEG 40-kDa (JenKem Technology, Plano, TX) overnight in the presence of gentle shaking in phosphate-buffered, pH 8.0, at a PEG:E196–301 molar ratio of 20:1. The PEGylated protein was purified and concentrated by the same His₆-tag affinitive chromatography method mentioned above.

Determination of in vivo half-life in rats.

Animal studies were performed in the animal laboratories of the University of Kentucky’s Division of Laboratory Animal Resources (DLAR) facility in accordance with the Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the National Institutes of Health. At the end of the animal studies, animals were euthanized by pentobarbital overdose (250 mg/kg, IP). The experimental protocols were approved by the IACUC (Institutional Animal Care and Use Committee) at the University of Kentucky. Rats were injected intravenously (IV) with different doses (1, 10, or 30 mg/kg for E196–301 and PEG-E196–301; 1 mg/kg for Fc-E196–301) of body weight of the purified enzyme. Blood samples were taken by puncturing saphenous veins with a needle. Approximately 50 μl of blood was collected into a heparin-coated capillary tube at various time points after enzyme injection. Blood samples were centrifuged at 5000 g for 15 min to obtain the serum. The serum was tested for enzyme activity using a sensitive radiometric method as described in our previous studies^{21, 27}. In addition, blood samples from Fc-E196–301 group were further analyzed by western blotting using anti-human Fc antibody (Thermo Fisher Scientific).