Primary high-throughput screening of engineered phytases by online monitoring of the oxygen transfer rate of Komagataella phaffii

Abstract

Background: Recombinant phytase production has recently gained increased recognition in phosphate recycling from phytate contained in plant-based side and waste streams. Until now, new phytase variants are evaluated at the end of the expression by standard offline screening procedures, where promising candidates with high activities and protein titers are identified. However, for large mutant libraries, this implies extensive laboratory work for a first screening of hundreds of clones. In this study, for the first time, two synergistic concepts for the primary screening of phytases were investigated.

Results: The aim was to predict high recombinant protein producer strains as well as high volumetric activity phytase variants, based on the development of the respiratory activity over time of the host cell, in this case, Komagataella phaffii (Pichia pastoris). In a first step, the metabolic burden was investigated by cultivating a clone library in YPD medium in a µTOM device. It was found that strains expressing medium or high protein concentrations show clear characteristics of an elevated level of metabolic burden during constitutive expression. However, a high protein concentration does not imply a high enzymatic activity. Therefore, in a second approach, the screening was adapted to screen for phytase variants with high volumetric activity. To do so, a modified Syn6 MES medium was developed, where phytic acid was used as the only phosphate source. Thereby, only clones secreting active phytase and generating free phosphate were able to grow, which was monitored via the oxygen transfer rate. A correlation between the offline measured volumetric phytase activity and µ_max was found. The clones were then ranked according to their online and offline performance and the results matched in 83% of the cases.

Conclusion: Online monitoring of the oxygen transfer rates in 96-well plates allowed for the evaluation of the total protein concentration and the volumetric phytase activity already during the expression. Using these results, also the specific activity can be calculated. In the future, primary screening experiments of large enzyme mutant libraries can be conducted without offline activity assays, to identify promising candidates.

Keywords: Komagataella phaffii; Pichia pastoris; Phytase screening; Recombinant protein production; Volumetric and specific enzyme activity.

Fig. 1

Cultivation of a clone library of a phytase-secreting K. phaffii strain. The cultivation was performed in YPD medium with 10 g/L glucose in a 96-square well MTP with 600 µL filling volume per well, operated at 350 rpm, 50 mm shaking diameter, and 30 °C. The oxygen transfer rate was monitored online in a µTOM device [35]. The data was obtained in duplicates and error bars represent minimal and maximal values. For clear data representation, four selected model clones are presented. Only every second measurement point over time is shown as a symbol. Data for the complete strain library can be found in Figure S2

Fig. 2

Correlation of calculated maximal growth rate (µ_max) with offline measured protein concentration. The online measured oxygen transfer rates (OTR) in Fig. 1 and Figure S2 were used to calculate the maximal growth rate until the first OTR peak with an exponential fit, as shown in Figure S3. The mean total protein concentration was measured at the end of the cultivation after 26 h for each clone with a BCA assay. The mean of duplicates is shown for the offline protein concentrations and growth rates with error bars as minimal and maximal values. The cultivation was performed in YPD medium with 10 g/L glucose in a 96-square well MTP with 600 µL filling volume per well, operated at 350 rpm, 50 mm shaking diameter, and 30 °C in a µTOM device [35]

Fig. 3

Schematic illustration of the phytase screening principle in modified Syn6 MES medium. Phytic acid is used as the only source of phosphate in the modified Syn6 MES medium. If the expression host, K. phaffii, secretes an active variant of the phytase enzyme, phosphate units from the phytic acid molecule are cleaved to generate myo-inositol phosphate and free phosphate. The full conversion to myo-inositol phosphate is hypothetical

Fig. 4

Cultivation of selected model clones of phytase-secreting K. phaffii strains. The cultivation was carried out in the modified Syn6 MES medium with 140 mM MES buffer (pH = 6.0), 10 g/L glucose, and (A) 0.81 g/L, (B) 0.55 g/L and (C) 0.11 g/L phytic acid as the only phosphate source. A 96-square well MTP was used with 600 µL filling volume per well, operated at 350 rpm, 50 mm shaking diameter, and 30 °C. The oxygen transfer rate was monitored online in a µTOM device [35]. The black curve in (A) represents a reference cultivation of the JE11 (1) clone in the standard Syn6 MES medium with 1 g/L KH₂PO₄, for comparison. The data was obtained in triplicates and error bars represent the standard deviations. For clear data representation, only every third or fifth measurement point over time is shown as a symbol

Fig. 5

Cultivation of a clone library (JE10) of a phytase-secreting K. phaffii strain. The cultivation was performed in the modified Syn6 MES medium with 140 mM MES buffer (pH = 6.0), 10 g/L glucose, and 0.55 g/L phytic acid in a 96-square well MTP with 600 µL filling volume per well, operated at 350 rpm, 50 mm shaking diameter, and 30 °C. The oxygen transfer rate was monitored online in a µTOM device [35]. A The mean OTR data for 22 of the 48 tested clones is shown without error bars for clarity. B The mean OTR data for another 22 of the 48 tested clones is shown without error bars for clarity. C The OTR data of the four selected model clones is displayed. The data was obtained in duplicates and error bars represent minimal and maximal values. For clear data representation, only every third measurement point over time is shown as a symbol

Fig. 6

Correlation of calculated maximal growth rate (µ_max) with offline measured phytase activity. The online measured oxygen transfer rates (OTR) from 96 clones (Fig. 5 and Figure S7) were used to calculate the maximal growth rate until the first OTR peak with an exponential fit, as shown in Figure S3. The mean phytase activity was measured at the end of the cultivation after 26 h for each clone with a 4-MUP assay. The mean of duplicates is shown for the offline phytase activities and growth rates with error bars as minimal and maximal values. The cultivation was performed in the modified Syn6 MES medium with 140 mM MES buffer (pH = 6.0), 10 g/L glucose, and 0.55 g/L phytic acid in a 96-square well MTP with 600 µL filling volume per well, operated at 350 rpm, 50 mm shaking diameter, and 30 °C in a µTOM device [35]. A linear fit was performed, to visualize the correlation between the online-measured maximal growth rate and the offline-measured volumetric phytase activity. The empty vector clones are marked as filled red circles and the wild type clones, which contain the non-modified phytase gene, are marked as filled blue diamonds

Fig. 7

Ranking of clone performance by the maximal growth rate and offline measured phytase activity. First, all 96 tested clones were ranked according to their maximal growth rate until the first oxygen transfer rate (OTR) peak and second, according to their offline measured phytase activity (data shown in Fig. 6). The clone with the highest maximal growth rate would be ranked as number 1 on the y-axis and the clone with the highest phytase activity as number 1 on the x-axis. The orange box and the star symbols mark the top 25% of clones, evaluated by the growth rate from the online measured OTR (24 clones). The empty vector clones are marked as filled red circles and the wild type clones as filled blue stars. Clones, which would be lost in this screening round, are marked as filled grey circles