Urate oxidase purification by salting-in crystallization: towards an alternative to chromatography

Abstract

Background: Rasburicase (Fasturtec® or Elitek®, Sanofi-Aventis), the recombinant form of urate oxidase from Aspergillus flavus, is a therapeutic enzyme used to prevent or decrease the high levels of uric acid in blood that can occur as a result of chemotherapy. It is produced by Sanofi-Aventis and currently purified via several standard steps of chromatography. This work explores the feasibility of replacing one or more chromatography steps in the downstream process by a crystallization step. It compares the efficacy of two crystallization techniques that have proven successful on pure urate oxidase, testing them on impure urate oxidase solutions.

Methodology/principal findings: Here we investigate the possibility of purifying urate oxidase directly by crystallization from the fermentation broth. Based on attractive interaction potentials which are known to drive urate oxidase crystallization, two crystallization routes are compared: a) by increased polymer concentration, which induces a depletion attraction and b) by decreased salt concentration, which induces attractive interactions via a salting-in effect. We observe that adding polymer, a very efficient way to crystallize pure urate oxidase through the depletion effect, is not an efficient way to grow crystals from impure solution. On the other hand, we show that dialysis, which decreases salt concentration through its strong salting-in effect, makes purification of urate oxidase from the fermentation broth possible.

Conclusions: The aim of this study is to compare purification efficacy of two crystallization methods. Our findings show that crystallization of urate oxidase from the fermentation broth provides purity comparable to what can be achieved with one chromatography step. This suggests that, in the case of urate oxidase, crystallization could be implemented not only for polishing or concentration during the last steps of purification, but also as an initial capture step, with minimal changes to the current process.

Figure 1. Solubility variations of the recombinant urate oxidase.

A) Solubility of urate oxidase as a function of pH in 50 mM Tris buffer at 20°C without added salt. B) Solubility of urate oxidase as a function of temperature in 50 mM Tris buffer pH 8, taking into account variation of pH as a function of temperature with tris buffer (ΔpH/ΔT = −0.03°C⁻¹). C) Solubility of urate oxidase as a function of salt type and ionic strength in 50 mM Tris buffer pH 8 at 20°C. D) Solubility of urate oxidase as a function of polymer addition (PEG 8000 or Poloxamer P188) with and without salt at 20°C.

Figure 2. Size exclusion chromatography analysis of the 5 urate oxidase pools from the downstream process.

Size exclusion chromatography analysis of the 5 urate oxidase pools from the downstream process. Each pool is analyzed on a Superdex 200 GL column eluted in 50 mM sodium phosphate buffer pH 8 with UV-Vis detection at 280 nm.

Figure 3. IEF analysis of the 5 urate oxidase pools from the downstream process.

IEF analysis of the 5 urate oxidase pools from the downstream process. Pool 2/3 lane identical to that of pool 7, suggesting that the 230 kDa peak corresponds to octomers of urate oxidase.

Figure 4. Crystallization trials of urate oxidase from pool 1 and pool 3 either via addition of poloxamer or via reverse dialysis.

Crystallization trials of urate oxidase from pool 1 and 3. Top) Crystallization conditions of pool 1. A) c_UOx≈35 mg/mL, 5% poloxamer 188, NH₄Cl 20 mM, Tris 5 mM pH 8.5. B) c_UOx≈40 mg/mL, Tris 5 mM pH 7.5, 20 mM NH₄Cl vs. Tris 5 mM pH 7.5. Bottom) Crystallization of pool 3. C) c_UOx≈11 mg/mL, 2.5% poloxamer 188, NH₄Cl 45 mM, Tris 50 mM pH 7.5. D) c_UOx≈68 mg/mL 100 mM NH₄Cl, Tris 50 mM pH 7.5 vs. Tris 50 mM pH 7.5.

Figure 5. Size exclusion chromatography analysis of urate oxidase crystal content versus chromatography steps 1 and 2.

(top) SEC analysis of the crystal content of urate oxidase from pool 1 (unfilled cross) compared to pool 1 (filled circle) and pool 2 (unfilled square): crystallization conditions were (left) 35 mg/mL urate oxidase with 5% poloxamer 188 in 5 mM Tris pH 8.5, 50 mM NH₄Cl; (right) 40 mg/mL urate oxidase in 5 mM Tris pH 8.5, 50 mM NH₄Cl, dialyzed against Tris 5 mM pH 8; (bottom) SEC analysis of the crystal content of urate oxidase from pool 3 (unfilled cross) compared to pool 2 (unfilled square) and pool 4 (crossed square): crystallization conditions were (left) 11 mg/mL urate oxidase with 2.5% poloxamer 188 in 5 mM Tris pH 8.5, 50 mM NH₄Cl; (right) 68 mg/mL urate oxidase in 5 mM Tris pH 8.5, 50 mM NH₄Cl, dialyzed against 5 mM Tris pH 8.

Figure 6. IEF analysis of the urate oxidase after crystallization.

IEF analysis of the urate oxidase. From top to bottom: a) pool 1 ; b) pool 1 after 50 mM NH₄Cl addition and ultrafiltration, c) crystal content from 35 mg/mL urate oxidase pool 1 with 5% poloxamer 188 in 5 mM Tris pH 8.5, 50 mM NH₄Cl; d) supernatant from 35 mg/mL urate oxidase pool 1 with 5% poloxamer 188 in 5 mM Tris pH 8.5, 50 mM NH₄Cl; d) pool 7 ; e) crystal content from 40 mg/mL urate oxidase pool 1 in 5 mM Tris pH 8.5, 50 mM NH₄Cl, dialyzed against 5 mM Tris pH 8; f) supernatant from 40 mg/mL urate oxidase pool 1 in 5 mM Tris pH 8.5, NH₄Cl 50 mM, dialyzed against 5 mM Tris pH 8, g) pool 7.

Figure 7. Spontaneaous urate oxidase crystals from pool 1 at 5°C (top) and SEC analysis (bottom).

Top: Urate oxidase crystals grown spontaneously in pool 1 after one month's storage at 5°C. Bottom: SEC analysis of this urate oxidase crystal content grown in pool 1 after one month storage at 5°C (filled square) and comparison with crystals obtained after reverse dialysis from pool 1 (unfilled cross) and chromatography from pool 2 (unfilled square).