Evaluation of options for harvest of a recombinant E. Coli fermentation producing a domain antibody using ultra scale-down techniques and pilot-scale verification

Abstract

Ultra scale-down (USD) methods operating at the millilitre scale were used to characterise full-scale processing of E. coli fermentation broths autolysed to different extents for release of a domain antibody. The focus was on the primary clarification stages involving continuous centrifugation followed by depth filtration. The performance of this sequence was predicted by USD studies to decrease significantly with increased extents of cell lysis. The use of polyethyleneimine reagent was studied to treat the lysed cell broth by precipitation of soluble contaminants such as DNA and flocculation of cell debris material. The USD studies were used to predict the impact of this treatment on the performance and here it was found that the fermentation could be run to maximum productivity using an acceptable clarification process (e.g., a centrifugation stage operating at 0.11 L/m(2) equivalent gravity settling area per hour followed by a resultant required depth filter area of 0.07 m(2) /L supernatant). A range of USD predictions was verified at the pilot scale for centrifugation followed by depth filtration. © 2015 American Institute of Chemical Engineers Biotechnol. Prog., 32:382-392, 2016.

Keywords: E. coli; centrifugation; domain antibody; fermentation; filtration; flocculation.

Figure 1

Process studies undertaken. Flowsheets show (a) the USD study of the broth processability as a function of harvest time; (b) pilot scale study of recoverability of flocculated broth with complementary USD study of centrifugation stage.

Figure 2

Representative fermentation monitoring profiles for 1 L studies. Given are (a) and (b) oxygen uptake rate, carbon dioxide evolution rate, and respiratory quotient; (c) biomass expressed as optical density at 600 nm, wcw, dcw; (d) culture capacitance at 1000 KHz, ratio of measured to predicted K _L a (see Results text) and viscosity expressed as flow consistency index (n ranges from 0.66 to 1.0 see Table 1).

Figure 3

Representative dAb and DNA profiles for 1 L fermentation studies. Given are (a) extracellular and intracellular dAb concentration; (b) specific intracellular and total dAb concentration, (c) %dAb and DNA concentration in culture supernatant; (d) parity plot comparing release into extracellular environment of both dAb product and DNA. The parity line allows for the initial presence of DNA at time of induction (equivalent to 12% release) and the line then shows relationship if dAb and remaining DNA are released to equal extents. To note, 100% dAb release refers to maximum product available at that point in fermentation.

Figure 4

Scanning electron microscopy images of fermentation samples. 20 L scale, Fermentation 2 (Table 1) samples are for postinduction and (total process) times of: (a) 0 (19.3) h, (b) 17.7 (37.0) h, and (c) 42.6 (61.9) h.

Figure 5

Particle size distribution for PEI‐flocculated E. coli cell broth. The ratio of cell broth to PEI solution is 24:1 for final PEI concentrations up to 1% w/v and 7.7:1 for higher concentrations. The PEI concentrations in the feed were varied from 0 to 50% w/v to give the final resultant overall PEI concentrations (w/v of broth as indicated in graphs). The material used for this study was autolysed broth collected at harvest (i.e., 49 h post induction). The concentrations of PEI g/g wcw used were (a) 0, (b) 0.006, (c) 0.018, (d) 0.030, (e) 0.065, (f) 0.195, (g) 0.325, and (h) 0.391.

Figure 6

USD shear stress, centrifugation and supernatant filtration studies. (a) % solids remaining (Eq. 2) after USD centrifugation for untreated and flocculated cell broth at 0 and at 0.53 × 10⁶ W/kg shear stress. Centrifugation conditions used: 6000 rpm for 7 min, max RCF 3,400g, Σ _T = 0.16 m² (see Eq. 1), V _lab /tΣ _T = 3.04 × 10⁻⁸ m/s (0.011 L/m²/h). Lines are best fit by eye. (b) Depth filtration studies of supernatant from untreated sheared broth (n = 2) with V _max evaluated using Eq. 3; supernatant from flocculated sheared broth (low values only – see text), n = 1. The volumetric concentration of the PEI used was 0.5% (w/v of broth) giving PEI concentrations ranging from 0.028 to 0.036 g/g wcw.

Figure 7

USD shear stress and centrifugation studies for clarification of untreated and flocculated broths produced at 20 L fermentation scale–effect of energy dissipation rate. Centrifugation conditions used 8000 rpm for 6 min, max RCF 6,010g, Σ _T = 0.28 m² (see Eq. 1). V _lab /tΣ _T = 2.01 × 10⁻⁸ m/s (equivalent to 0.072 L/m²/h). The PEI concentration of the flocculated broth was 0.5% (w/v of broth), giving PEI concentration of 0.033 and 0.041 g/g wcw for Fermentation 1 and 2, respectively.

Figure 8

Pilot‐scale centrifugation of flocculated fermentation broth. Centrifuges used and operating conditions:(a) Disc stack centrifuge of solids holding capacity 0.25 L, Σ _{( c = 0.4)} = 680 m², operated with a low shear stress hydro‐hermetic feed zone at 50 L/h to give (Q/Σ)_ds = 2.01 × 10⁻⁸ m/s (0.072 L/m²/h) and time to steady state predicted at 3.6 min (b) a tubular bowl centrifuge (CARR Powerfuge) of solids holding capacity 1.0 L, Σ _{( c = 0.9)} = 930 m²) operated at 60 L/h to give (Q/Σ)_tb = 1.78 × 10⁻⁸ m/s (0.064 L/m²/h), and time to steady state predicted at 5.5 min. The dashed lines indicate the USD prediction (from Figure 8) for given energy dissipation rates. Supernatant for filtration studies (Figure 9) is collected between predicted time to steady state and the predicted time to solids breakthrough.

Figure 9

Filtration of supernatant from pilot‐scale centrifugation of flocculated broth. See Figure 8 for details of supernatant preparation using: (▴,▵) solid bowl tubular centrifuge (Powerfuge); (•) disc stack centrifuge. Filtration was by 0.1 m² depth filter, pore size 0.05–0.2 μm, operated at 100 L/m²/h.

Figure 10

Relationship between dAb and DNA release, solids remaining after centrifugation and depth filter area for (a) untreated and (b) flocculated broths. Here 100% dAb recovery refers to total extracellular dAb available after fermentation time of 66 h. Centrifugation conditions for equivalent Q/Σ = 3.04 × 10⁻⁸ m/s (0.011 L/m²/h) for a high shear stress, nonhermetic feed zone.