All restriction enzymes were purchased from New England Biolabs (Beverly, MA, USA) and DNA polymerase was obtained from Takara (Dalian, Liaoning, China). Yeast extracts and peptone were purchased from Oxoid (Basingstoke, UK). Luria Bertani (LB) medium and yeast extract peptone glucose (YPD) components were purchased from Beijing Chemical Industry (Beijing, China). Escherichia coli TOP10 cells were used for plasmid maintenance. Zeocin, P. pastoris X33, and pGAPzαA were purchased from Invitrogen (Carlsbad, CA, USA) and used for the expression process.

YPD broth (1% yeast extract, 2% peptone, and 2% glucose) containing Zeocin (100, 200, 300, and 400 µg/ml) was employed for selection of P. pastoris transformants at 30°C. Low salt LB (1% tryptone, 0.5% yeast extract, and 0.5% NaCl, pH 7.0) with Zeocin (25 µg/ml) were employed to culture E. coli TOP10 transformants at 37°C.

A 560-base pair (bp) Pre-Onc-DV3 fusion gene fragment was synthesized by GenScript Corporation (Nanjing, China) after optimization. Both pGAPzαA and the Pre-Onc-DV3 gene were double digested by BstBI and EcoRI, and the digested gene was inserted into the open reading frame under control of the GAP promoter, resulting in the pGAPZαA-Pre-onconase-PFV-DV3 expression vector (Fig. 1). The recombinant plasmid was transferred into E. coli Top10 cells, and screened on low salt LB containing 25 µg/ml Zeocin. The extracted recombinant plasmid was identified by digesting with BglII and EcoRI, linearized with BlnI, and transformed into P. pastoris X33 cells by electroporation. The electrocompetent P. pastoris X33 cells were prepared according to standard methods, and using an electroporator (Scientz, Ningbo, China) following the manufacturer's instructions. The electroporation conditions were 1.5 kV, 25 µF, and 200 Ω.

The transformed X33 clones were screened on YPD agar plates containing 100 µg/ml zeocin for 3 days. The chosen transformants were screened for Zeocin-level clones from the YPD agar plates containing a higher Zeocin concentration (200, 300, and 400 µg/ml). All of these high-level secretion clones were confirmed by PCR using X33 transformant DNA, and selected according to the Onc-DV3 recombinant protein concentration on SDS-PAGE. The promising X33 transformants were grown in 10 ml YPD medium in a 100-ml flask for 24 h. Then, the culture was transferred to 100 ml YPD medium for 24 h in a 1,000-ml flask. The inoculums were cultivated to OD₆₀₀ of 0.5 as starting seed cells to investigate the effects of culture parameters: i) carbon source cells were cultured in 2% (w/v) of different carbon sources (glucose, glycerol, methanol, sucrose, sorbitol); ii) nitrogen source cells were cultured in 2% (w/v) of different nitrogen sources [tryptone, peptone, yeast extract, beef extract, (NH₄)₂SO₄]; iii) the cells were cultured under different temperatures (21, 24, 27, 30, and 33°C); and (iv) the cells were cultured under a different pH (4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5, 8.0).

The concentration of the Onc-DV3 recombinant protein was determined by SDS-PAGE and RNA hydrolytic activity. All of the data are expressed as the mean ± standard deviation (SD) of independent triplicate determinations.

Western blot analysis was performed to determine the concentration of Onc-DV3. Briefly, samples were collected at various time-points (35, 45, 55, 65, and 75 h), and resolved by SDS-PAGE. Then, the proteins were electrophoretically transferred to polyvinylidene fluoride (PVDF) membranes (Millipore Co., Ltd., Boston, MA, USA) using a DYCZ-40G transfer blotter (Beijing Liuyi Co., Ltd., Beijing, China). The membranes were incubated in blocking buffer (10 mM Tris-HCl, 150 mM NaCl, 5% skim milk, pH 7.5) at room temperature for 2 h, after which they were incubated with a polyclonal antibody of DV3 at room temperature for 4 h. Next, the membranes were washed three times with TBS containing 0.05% Tween-20, and incubated with goat anti-rabbit IgG peroxidase conjugate (diluted 1:100; Boster Co., Ltd., Wuhan, Hubei, China) at room temperature for 4 h. The membrane was incubated with a substrate solution containing 0.06% (w/v) chloronaphthol and 0.01% H₂O₂ (both from Beijing Chemicals) after washing.

A 50-l fermentor was applied to evaluate the ability for larger scale constitutive expression of Onc-DV3 immunotoxin using the GAP promoter in P. pastoris, with the aforementioned method. The time-course for biomass and production of Onc-DV3 in the 50-l fermentor was determined. The production value of the recombinant immunotoxin increased with biomass until 72 h. The biomass (OD₆₀₀) was elevated to ~300 during fed-batch fermentation in a 50-l fermentor. The production of Onc-DV3 in the 50-l fermentor reached about 2×10⁶ U/l after 72 h of fermentation, compared to 6×10⁵ U/l in a shake flask.

Biomass in the culture broth was determined by cell density expressed as optical absorbance (OD₆₀₀). The protein concentration was determined using the Bradford assay kit (Sangon, Shanghai). Cell growth was monitored using a model UV-4802H spectrophotometer (Unico, Shanghai, China) at a wavelength of 490 nm, and converted to dry cell weight using a predetermined correlation factor.

The culture broth was collected after 72 h, and yeast cells were removed by centrifugation at 8,000 rpm for 30 min at 4°C. The centrifuged liquid was ultrafiltrated and concentrated by ultrafiltration using the Amfore ultrafiltration membrane 10000. The concentrated recombinant immunotoxin was loaded onto a SP Sepharose Fast Flow column (5.0×20 cm) that was pre-equilibrated with buffer A (20 mM PB, pH 7.0). The bound recombinant immunotoxin was eluted with buffer B (1 mol/l NaCl). The eluted protein was mixed with buffer C (20 mM PB, pH 7.0) and further loaded onto the Sephadex G-75 column (1.6×100 cm) that was pre-equilibrated with the same buffer. The purity of the recombinant immunotoxin from the final eluate was analyzed by SDS-PAGE.

In order to determine the RNA degradation activity of onconase in vitro, 10 µg Pichia RNA was added in a 100-µl reaction buffer (100 mmol/l Tris-HCl, pH 8.0), and then purified onconase solutions at different concentrations (100, 10, 1, and 0.1 µg/ml) were added. The tubes were incubated in a water bath at 37°C for 30 min. The control group was incubated under the same conditions, but standard RNase A was added to replace the onconase. RNA degradation was assessed by agarose gel electrophoresis.

With 1 mg/ml solution of yeast transfer RNA (tRNA) as a substrate, various agents were successively added to assay the enzymatic activity of the purified Onc-DV3 by drawing a portrait of change values of OD₂₆₀ along the RNase activity. The experiment was repeated at least three times.

The paired highly metastatic and non-metastatic cell lines used were as follows: i) PC-3M-1E8, a human prostate cell line (highly metastatic), ii) PC-3M-2B4, a human prostate cell line, iii) MDA-MB-231, a human breast carcinoma cell line (highly metastatic), iv) MCF-7, a human breast carcinoma cell line, v) PG-BE1, a human lung cancer cell line (highly metastatic), vi) PG-LH7, a human lung cancer cell line, vii) L-02, a human normal liver cell, and viii) HEK293, a human embryonic kidney cell line.

All of the cell lines were kept in our laboratory in RPMI-1640 medium (Invitrogen) with 20% fetal bovine serum and 10% DMSO at −80°C. Cells were cultured in RPMI-1640 medium with 10% fetal bovine serum, 100 U/ml penicillin, and 100 µg/ml streptomycin at 37°C in 5% CO₂. The cells (PC-3M-1E8, PC-3M-2B4, MDA-MB-231, MCF-7, PG-BE1, PG-LH7, L-02, HEK293) were seeded on 96-well flat-bottomed plates at 3,000 cells/well, and incubated at 37°C in 5% CO₂ atmosphere for 24 h. Then, the cells were treated with Onc-DV3 after removing the medium at serial dilutions (0.05, 0.10, 0.20, 0.40, 0.80, 1.60, and 3.20 µM). There were six individual wells for each dose. The blank control was the dilution medium without added sample. The cells were incubated in 96-well flat-bottomed plates at 37°C in 5% CO₂ atmosphere for 72 h, after which the supernatant was removed, and the cells were washed with PBS (pH 7.4). The cells were incubated in medium containing 20 µl MTT solution (5 mg/ml) at 37°C for 4 h. In order to dissolve the formed formazan crystals, 100 µl dimethyl sulfoxide (DMSO) was added after removing the supernatant. The OD was measured using the ELISA microplate reader at 490 nm. Cytotoxicity to the cancer cells was determined by survival rate (%). The experiment was repeated at least three times. Data are presented as mean ± SD.

A synthesized 560 bp Pre-Onc-DV3 gene was inserted into the pGAPzαA vector under control of the GAP promoter, by integrating the coding gene in frame with the S. cerevisiae α-factor pre-secretion signal sequence. The constitutive expression of pGAPzα-Pre-Onc-DV3 is shown (Fig. 1). The extracted recombinant plasmid was identified by digesting with BglII and EcoRI, producing a DNA fragment ~1,050 bp (Fig. 2). To confirm validity of the open reading frame, the full-length promoter and Pre-Onc-DV3 was sequenced (data not shown). Zeocin was used to screen the E. coli TOP10 transformants.

The reconstructed vector was extracted and digested with BlnI to obtain a linear recombinant vector. Then, the linear vector of pGAPZαA-Pre-Onc-DV3 was employed to transform P. pastoris X33 by electroporation. The X33 transformants harboring the recombinant vector were screened depending on their resistance to Zeocin on YPD plates containing different concentrations of Zeocin. The 21 transformants were selected on YPD plates containing 400 µg/ml Zeocin, and were confirmed by genomic DNA PCR analysis. The expression ability of these transformants was determined by measuring the medium supernatant for RNA degradation efficiency.

The effects of a few carbon sources on the production of Onc-DV3 immunotoxin during the fermentation process were evaluated. The X33 transformant with the highest expression level of recombinant immunotoxin was selected as an initial seed and cultured in modified YPD in which the glucose was 2% (w/v), but was replaced with different carbon sources (glucose, sorbitol, sucrose, methanol, or glycerol). The results demonstrated that glucose was the most favorable carbon source, with a maximum expression level of immunotoxin of ~310 U/l after 60 h of culture. The results are shown in Fig. 3A. Significantly, the biomass achieved the highest level when glycerol was used as a carbon source, and immunotoxin protein production was much lower than that of glucose as a carbon source. Compared to glucose and glycerol, both the biomass and production level of Onc-DV3 immunotoxin were much lower when using methanol, sucrose, or sorbitol as the carbon source. The main nitrogen sources were evaluated for their capacity to produce Onc-DV3 immunotoxin [tryptone, peptone, yeast extract, beef extract, (NH₄)₂SO₄]. The results demonstrated that peptone was the most favorable nitrogen source. The maximum expression level of immunotoxin was ~310 U/l after 60 h of culture (Fig. 3B). The cells were grown in YPD medium (pH 7.0) at 21, 24, 27, 30, and 33°C to evaluate the effects of temperature on biomass and Onc-DV3 immunotoxin production (Fig. 3D). The biomass and production of Onc-DV3 did not show remarkable variance under different temperatures. The selected temperature for Onc-DV3 expression was at 30°C during fermentation, according to the empirical of higher enzyme activity at higher culture temperature (8). The effects of different medium initial pH (4.5, 5.0, 5.5, 6.0, 6.5, 7.0, 7.5, 8.0) at 30°C on biomass and Onc-DV3 production were also checked during culture. The results showed (Fig. 3C) that the initial pH of the YPD medium had little effect on the biomass, but Onc-DV3 expression level was highest at pH 7.0.

The expression levels of recombinant immunotoxin at different times during the culture were analyzed by western blot analysis using anti-onconase polyclonal antibody. The results are shown in Figs. 4 and 5. The production of the recombinant immunotoxin was presented at 35 h and reached the highest production at 65 h in a flask culture.

By using optimized conditions, a 50-l fermentor was applied to evaluate the ability for large-scale constitutive expression of Onc-DV3 immunotoxin using GAP promoter in P. pastoris, with the above selected method. The time-course for biomass and production of Onc-DV3 in the 50-l fermentor are shown in Fig. 6. The production value of the recombinant immunotoxin increased with the biomass until 72 h. The biomass (OD₆₀₀) was elevated to ~300 during fed-batch fermentation in a 50-l fermentor. The production of Onc-DV3 in the 50 l fermentor reached about 2×10³ U/l after 72 h of fermentation compared with that in the shake flask culture of 3.1×10² U/l. The results revealed that the constitutive expression of the Onc-DV3 recombinant immunotoxin was less hazardous to P. pastoris. It is viable to efficiently produce Onc-DV3 recombinant immunotoxin in the 50 l fermentor by adopting the constitutive fermentation strategy established in this study.

The recombinant immunotoxin was isolated from the medium supernatant after ultrafiltration using membranes of 60 and 10 kDa and purified by SP Sepharose ion exchange chromatography and Sephadex G-75 gel chromatography. The purity of the recombinant immunotoxin reached 97.0% with the recovery level of 12.8% (Table I). The purified immunotoxin showed a single band on SDS-PAGE at ~20.0 kDa (Fig. 7), which was near the theoretical value for this recombinant protein.

The biological activity assay of Onc-DV3 was carried out by agrose electrophoresis and increment ultraviolet absorption (Fig. 8). RNA hydrolytic activity of the biological activity of Onc-DV3 was analyzed according to RNase assay from the onconase domain of Onc-DV3 (Table II). The results confirmed that the recombinant immunotoxin Onc-DV3 produced by the GAP promoter-derived expression system maintained high biological activity.

In conclusion, Onc-DV3 recombinant immunotoxin was expressed under the control of the GAP promoter in the P. pastoris expression system with high production and high biological activity.

Every strain of the tumor cells was cultured and treated with immunotoxin Onc-DV3 and incubated for 72 h. The survival rate of the cells as indicated by OD₄₉₀ is shown in Table III. The trials of Onc-DV3 with tumor cells showed high efficient inhibition in highly metastatic tumor cells (PC-3M-1E8, PG-BE1, MDA-MB-231; the survival rate varied from 24.79±7.31 to 27.61±9.16), middle efficiency on poorly metastatic tumor cells (PC-3M-2B4, MCF-7, PG-LH7, the survival rate varied from 53.38±17.37 to 59.64±8.03), and no effect on normal cells (L-02, HEK293).