Introduction

The basidiomycete Tremella fuciformis Berk, or ‘white jelly mushroom’ favored by Chinese people as a gourmet and medicinal mushroom (Huang 1993, 1998), is a very interesting organism since it produces single yeast-like conidia (YLCs) derived from mycelium (Huang 2000). YLCs are monocytes that reproduce offspring by budding. Single YLC can form colony by vegetative propagation (Huang 2000). There have been many reports about the ingredients of YLCs, which show health aiding effects (Lin et al. 1982; Kiho et al. 1994; Cheng et al.2002), especially the polysaccharide fraction composed of acidic and neutral heteroglycans which has several pharmacological activities (Gao et al. 1996, 1997), such as enhancing host immune functions (Wang et al. 1983; Xia and Lin 1989; Jing et al. 2002; Yang and Xu 2002) and antitumor activity (Ukai et al. 1972, 1992; Dong et al. 2004; Dong and Qu 2004). Massive production of these active ingredients necessitates the obtainment of YLCs at a large scale. In general, submerged fermentation was adopted because of its expedient industrialization and short fermentation period of 4 days after inoculation, which were conferred by the high propagation rate of YLCs. However, one of the most general problems encountered in submerged fermentation is the deficiency of dissolved oxygen (Huang 2000; Wu et al. 2002). So far as was known, YLCs of T. fuciformis were highly sensitive to oxygen supply for its aerobiosis. With higher cell density and viscosity achieved, oxygen limitation became more severe during late stage of fermentation (Zheng and Lu 2003).

Vitreoscilla hemoglobin (VHb) was first recognized by Wakabayashi et al. (1986). The putative function of VHb was to trap oxygen and feed it to the membrane terminal oxidases, which was reflected by the kinetic constants for oxygen binding to VHb. Practically, Khosla and co-workers first demonstrated that the expression of VHb could enhance growth properties (Khosla and Bailey 1988) and protein synthesis (Khosla et al. 1990) in recombinant Escherichia coli. This finding was supported by the claims of many specialists and scholars that the VHb gene often improved growth as well as protein and metabolite synthesis in bacteria and fungi (Khosravi et al. 1990; Kallio and Bailey 1996; Wei et al. 1998; Wu et al. 2003; Bhave and Chattoo 2003). Therefore, it was concluded that the expression of VHb gene conduced to enhancing growth and metabolism under oxygen-limited conditions.

Previously, an efficient procedure for isolation and regeneration of protoplasts (Xie and Zhu 2003) from T. fuciformis and a modified restriction enzyme-mediated DNA integration (REMI) (Rogers et al. 2004; Guerin and Larochelle 2002) transformation system had been established in our researches (Zhu 2004; Xie et al. 2005). Therefore, in this study, we incorporated VHb gene in YLCs from T. fuciformis to enhance oxygen uptake in submerged fermentation under low dissolved oxygen conditions. This research would pave the way for modifying T. fuciformis genetically and solving the chemical engineering problems with genetic strategies in higher fungi fermentation. Meanwhile, this research is also of great importance to genetic engineering of other edible and medical mushroom.

Materials and methods

Strains and media

T. fuciformis ACCC50546 was purchased from Agricultural Culture Collection of China (ACCC, Beijing) and grown in CM agar medium (1% maltose, 1% glucose, 0.4% tryptone, 0.1% MgSO4·7H2O, K2HPO4, 0.046% KH2PO4, 2% agar) at 25°C. After 10 days, YLCs were isolated from mycelia and maintained on CM agar slants at 8°C. For transformation of T. fuciformis, CM liquid medium was inoculated with YLCs and incubated for 3 days at 25°C. The same medium was used throughout the fermentation experiments. Potassium chloride (0.5 M) was used as an osmotic stabilizer during preparation and regeneration of the protoplasts. Transformants were selected and stored on CM agar medium containing hygromycin at 50 μg/ml (Xie et al. 2004). E. coli DH5α (Stratagene, USA) was used for DNA manipulations, which was cultured at 37°C in LB medium composed of 0.5% yeast extract, 1% tryptone, and 1% NaCl. The lysis buffer was used in preparation of cell lysates from YLCs using glass beads vortexing (50 mM Tris–HCl, pH 8.0, 1% DMSO, 200 mM NaCl, 1 mM EDTA, 1 mM PMSF, 1 μg/mL leupeptin, 1 μg/mL pepstatin A). Glass beads (diameter 500 μm) were soaked in concentrated HCl (w/v, 37%) for 16 h and rinsed thoroughly in distilled water. After that, these glass beads were baked for 16 h at above 150°C, then air-cooled. Before use, the heat-treated beads were chilled at 4°C or on ice for at least 2 h.

Plasmid construction

To construct the plasmid expression vector, designated as pVHb (Fig. 1e), intermediate plasmid pBI-VHb (Fig. 1c) was generated by incorporating VHb gene from plasmid pOK12-VHb (Fig. 1a) digested with BamH I and Sac I into plasmid pBI121 (Fig. 1b) instead of beta-glucuronidase (GUS) gene. Then, plasmid pBI-VHb was cut using Hind III and EcoR I. The reading frame fragment comprising the CaMV35S promoter, Nos terminator, and VHb gene sequence was recovered. Finally, pVHb was constructed by inserting this fragment excised from pBI-VHb using cohesive-end ligation at the multiple cloning site of pCAMBIA1300 (Fig. 1d) digested by restriction enzymes Hind III and EcoR I simultaneously. Thus, the resultant plasmid, pVHb, consisted of a pCAMBIA1300 backbone containing the hph and VHb genes, each of which was put under the control of CaMV35S promoter and followed by different terminators.

Fig. 1
figure 1

Scheme of the construction of expression vector pVHb. Enzyme sites and reading frames are shown in full for each plasmid. Two promoters are both PCaMV35S. Vitreoscilla hemoglobin (VHb) is the target gene. Hph is a hygromycin resistance gene. Termination sequences of VHb and hph genes are different, one of which is Tnos following VHb gene, and the other is CaMV35S polyA. RB and LB right and left T-border, respectively. It should be noted that the replication origins of all plasmids are not identical

Full size image

Protoplast generation and transformation procedure by REMI

YLCs were cultivated for 3 days with shaking (150 rpm) at 25°C. These conidia were harvested by centrifugation at 4,000×g for 5 min and rinsed with osmotic stabilizer (0.5 M KCl). Then, the YLCs were incubated for 3 h at 35°C in 1 mL 20 mg/mL lywallzyme (Guang Dong Institute of Microbiology, China) containing 0.5 M KCl (Xie and Zhu 2003). After incubation, these protoplasts were washed free of enzyme and transferred to TPB solution (0.6 M KCl, 25 mM CaCl2, H2O). The whole mixture (300 μL) of plasmid and Hind III (150 U) was added to 100 μL of protoplast suspension (109 cells/mL) in an Eppendorf tube and chilled on ice for 5 min. After that, 20 μL of PEG4000/S was added and mixed briefly, and then the mixture was kept for 5 min on ice. Subsequently, the mixture was incubated at 25°C for an hour after the addition of 1 mL of PEG4000/S and subjected to heat shock for 20 min at 35°C (Schiestl and Petes 1991; Zhu 2004). Transformed protoplast suspension was centrifugated and mixed with 1 mL of CM liquid medium containing 0.5 M potassium chloride. After resuspension, 100 μL of the whole mixture was poured onto CM agar medium plates plus 50 μg/mL hygromycin. Seven days later, the transformants were selected according to the resistance to hygromycin and transferred to new medium.

PCR and Southern hybridization

Hundreds of transgenic YLCs had appeared after transformation. PCR and Southern hybridization (Sambrook et al. 1989; Kim et al. 1999) were performed to confirm whether or not the conidia grown on CM agar medium were transformants. Chromosomal DNAs of nontransgenic and transgenic YLCs were isolated employing the modified DNAzol protocols (Marco and Roberta 2003). The chromosomal DNAs of the transformants and the nontransgenic strain as the control were used as the templates of the PCR to confirm the integration of the pVHb using the hph gene-specific primers (forward primer: 5′-CGGATGATTCCTACGCGAGC-3′, reverse primer: 5′-TTCCTCCGGATCGGTGAAGC-3′). PCRs were performed in a final volume of 20 μL with the following reagents: 1.0 μg of chromosomal DNA, 100 pmol of each primer, 200 μM dNTPs, 1.5 mM magnesium chloride, 2.5 U Taq polymerase, 2 μL PCR buffer, and double distilled water. PCR cycling was implemented as follows: initial denaturation at 94°C for 5 min and then 30 cycles of 94°C (45 s), 52°C (45 s), and 72°C (90 s). After the completion of the 30 cycles, the reaction mixture was incubated for 10 min at 72°C. The products were analyzed by the ethidium bromide staining after 1.0% agarose gel electrophoresis (Sambrook et al. 1989). At the same time, Southern hybridization using the DIG system was carried out. Approximately 5 μg of genomic DNA from the transformant was digested with the restriction enzyme Hind III, size-fractionated by electrophoresis on a 1% agarose gel. The DNA fragments in the agarose gel were transferred onto a Hybond N+ Nylon membrane (Amersham, Hong Kong). A probe for VHb was prepared with the fragment amplified by PCR with two primers specific to VHb gene using pVHb as template. Labeling of the DNA probe, hybridization, and signal detection were conducted using the method recommended by the manufacturer (Amersham).

Protein extraction and Western blot analysis

The expression of VHb was analyzed by Western blot in transformants (Sambrook et al. 1989; Wilfred et al. 1994; Sooyoung et al. 2005). Control and transgenic YLCs were cultured in 100 ml CM liquid medium and harvested after 3 days. YLCs were centrifuged at 4,000×g at 4°C for 10 min, then resuspended in an equal volume of chilled lysis buffer. The suspension was placed in a sturdy tube, and PMSF (10 μL of 100 mM PMSF per milliliter of cell suspension) and 2 g of chilled glass beads per gram of cell wet weight were added at this point. Subsequently, the cells were vortexed for 1 min and kept on ice for 1 min. This process was repeated for five times. Protein was extracted from disrupted cells in the lysis buffer by ultracentrifugation at 4°C for 30 min at 30,000×g after removing the glass beads (Wilfred et al. 1994). Protein concentration was estimated by the Bradford method (Suh et al. 2000). The proteins were separated on a 12.6% SDS-PAGE gel (Shen et al. 2003) and blotted onto an Immuno-Blot polyvinylidene fluoride (PVDF) membrane (Sigma, USA). Western blotting and hybridization were performed according to the manufacturer’s instructions. The anti-VHb antibodies (rabbit) were used at a dilution of 1:5,000, and secondary anti-rabbit IgG-alkaline phosphatase conjugate was used at a dilution of 1:5,000 (Sigma). The hemoglobin was produced in recombinant E. coli and purified as the standard.

Batch fermentation at shake flask and bioreactor scale

Submerged fermentation experiments were carried out in shake flasks and bioreactor to investigate the effect on the growth of YLCs by the expression of VHb gene. The wild-type YLCs and transformant cultures at the late exponential growth phase were inoculated into three 250-mL shake flasks, each containing 100 mL medium with tailor-made bungs, and incubated at 25°C and 150 rpm in an orbital shaker for 5 days. Meanwhile, bioreactor cultivations were performed in a 4-L in situ sterilizable bioreactor (Set 4V, Setric Genic Industriel, Toulouse, France) with an effective working volume of 3 L. The process was monitored and controlled using bioprocess automation software (Process Control Systems AG, Wetzikon, Switzerland). Mid-log-phase CM-grown YLCs were transferred to the bioreactor containing CM liquid medium. The cultivation conditions were at 25°C, pH 7.0, and 150-rpm agitation. To ensure anaerobic conditions, the bioreactor was sparged with nitrogen at a flow rate of 0.5 volume per volume per minute. In these two experiments, samples were taken at 12-h intervals. Cell growth was followed by measuring the optical density at 600 nm with spectrophotometer (Milton Roy).

Results

Transformation of protoplasts with REMI

In general, there were four methods available in genetic transformation of basidiomycetes: electroporation (Kuo et al. 2004), Agrobacterium-mediated transformation (Chen et al. 2000), particle bombardment (Sunagawa and Magae 2002), and REMI (Kim et al. 2004; Lu et al. 1994). Compared to other three methods, REMI was simple and economical. It required no special instruments. When the REMI was carried out using the restriction enzyme Hind III which digested pVHb at the unique site, genetic transformation of protoplasts generated from YLCs of T. fuciformis was successfully implemented with the expression plasmid pVHb, and the transformation yield was estimated as 2.17×104 transformants per microgram of plasmid DNA. The transformation frequency was higher than that of other mushroom, such as Ganoderma Lucidum (Kim et al. 2004), Trametes versicolor (Kim et al. 2002), Lentinula edodes (Sato et al. 1998), and Coprinus cinereus (Granado et al. 1997).

Identification of T-DNA insertional transformants with PCR

Among the resistant colonies, three candidates randomly selected were further confirmed the integration of T-DNA into the host chromosomes with PCR. The chromosomal DNAs of the three transformants and the nontransgenic strain as the control were used as the templates for the PCR using the hph-specific primers. When the products were analyzed by the agarose gel electrophoresis, the expected 1.0-kb amplified bands appeared (Fig. 2). No false positives were detected by PCR amplification among the three antibiotic resistant cultures. This band was not amplified from the control strain. So, the amplified bands made it clear that exogenous gene was integrated into the genome of YLCs.

Fig. 2
figure 2

PCR identification of three transformants and the control with hph-specific primers. Lane 1: a negative control of PCR; lanes 2–4: transformants T1, T2, T3; lane CK: the wild-type yeast-like conidia (YLCs). M λ-EcoT14I digest DNA marker. PCR products are about 1.0 kb

Full size image

Copy of VHb gene in transformants from YLCs of T. fuciformis

To confirm the insertion of VHb gene in genome DNA and determine its copy number, the three transformants detected by PCR were analyzed for the VHb gene integration by Southern blotting using VHb gene as the probe. Genomic DNA from transformants was digested with Hind III and hybridized with the DIG-labeled VHb gene probe. There was only one Hind III cutting site within the plasmid pVHb. Therefore, hybridization bands could be used to determine the copy number of VHb in genome. As shown in Fig. 3, the different Southern hybridization profiles of the three picked transformants indicated that the DNA was inserted into different sites in the genome. Two hybridizing bands were detected in transformant T2, indicating two insertions within the T2 genome. Transformants T1 and T3 showed only one hybridizing band, which implied that there was only one copy of the target gene. Hybridizing signal did not appear in the control strain (Fig. 3, lane CK). Bands of different sizes were visualized as imaged in Fig. 3, lanes 1–3. From the result of Southern hybridization, we concluded that the plasmid pVHb DNA was successfully integrated into the genomic DNAs of YLCs from T. fuciformis by REMI.

Fig. 3
figure 3

Southern blotting of genomic DNA from transgenic YLCs of Tremella fuciformis using the DIG system. Lane CK: the wild-type YLCs, lanes 1–3: transformants T1, T2, and T3

Full size image

Expression of VHb gene in transformed YLCs

Western blot analysis was performed to demonstrate the expression of VHb in YLCs from T. fuciformis transformed with the plasmid pVHb. Whole cell extracts were electrophoresed on a 12.6% SDS-PAGE gel. The proteins were then transferred onto a nitrocellulose membrane and incubated with VHb antiserum (Promega, USA). As shown in Fig. 4, two bands that comigrated with the VHb standard were present in transformants (Fig. 4, lanes 3 and 4). This band, however, did not appear in the control without harboring VHb gene (Fig. 4, lane CK). These results indicated that VHb protein of the correct molecular weight had been produced in T. fuciformis intracellularly.

Fig. 4
figure 4

Western blot analysis of whole-cell extracts from the control and transformants. Lane CK: control (the wild-type YLCs); lanes 1, 2: VHb standard (14 kDa) from Escherichia coli; lanes 3, 4: transformants

Full size image

Effect of VHb expression on cell growth

In this work, batch fermentations in Erlenmeyer flasks were performed under microaerobic conditions to explore the effect of VHb gene expression on cell growth. The wild-type YLCs and transformant were grown at 25°C in CM liquid medium. The experiment profiles were shown in Fig. 5. When cultivated in Erlenmeyer flasks, the difference in growth rates between wild-type and the VHb-expressing strains during the initial phase of cultivation was negligible. Substantial differences were seen in the late stage of cultivation. The transformant exhibited higher cell viability and grew to a higher cell density than the wild-type YLCs. Furthermore, significant differences in the late exponential-phase specific growth rates were observed. So it was demonstrated that VHb-expressing YLCs had higher specific growth rates and final cell densities compared to a control strain without VHb gene when both were cultivated under the same poorly aerated conditions.

Fig. 5
figure 5

Comparison of the growth properties of the wild YLCs (VHb−; filled symbols) and transformant (VHb+; open symbols) in Erlenmeyer flasks. Cells were grown under the same conditions as described previously. The growth curves as measured in terms of A 600 are shown for these two strains. The data were the mean of three repeats

Full size image

To compare the profiles in large-scale cultivation, bioreactor studies were also carried out in CM liquid medium. The results showed that there were significantly different growth profiles between the control and VHb+ strains under oxygen-limited conditions (Fig. 6). It was undoubted VHb expression resulted in enhanced growth of the transformed strain. Doubling time of the VHb transformant was approximately 12 h shorter than that of the control VHb- strain, and the average specific growth rate was higher by 25%.

Fig. 6
figure 6

Comparison of grown profiles between VHb+ (open symbols) and VHb− (filled symbols) strains under oxygen-limiting conditions in the CM liquid medium. The profiles of optical densities (○, ●) as measured in terms of A 600 are shown. Error bars for the optical density values indicate standard deviation from mean for three batches

Full size image

Discussion

The expression of VHb gene in bacteria was studied extensively (Hikmet et al. 2001; Pauli et al. 1994). There was limited number of literature about the expression of this gene in fungi including Saccharomyces cerevisiae (Wilfred et al. 1994), Penicillium chrysogenum (Sun et al. 2002), Pichia pastoris (Wu et al. 2003), and Aspergillus terreus (Lin et al. 2004). However, to our knowledge, there were scarce reports on genetic transformation of VHb gene in basidiomycetes.

In this study, we sought the way of intracellular expression of Vitreoscilla hemoglobin gene to improve the physiological state of the YLCs from T. fuciformis for increasing cell density in submerged cultivation. For the purpose, T. fuciformis transformation was carried out by the REMI method. Many hph-resistant transformants were generated. PCR and Southern blotting analysis indicated that the transformation was successful, and T-DNA was integrated into the recipient cell chromosome. These transformants were valuable in the cloning of the genes related in the mutant characteristics (Granado et al. 1997; Riggle and Kumamoto 1998). Western blot analysis showed that the molecular weight of expressed VHb protein in T. fuciformis was identical to that of the standard. This result proved that VHb gene was expressed well in some basidiomycetes. The effect of intracellular VHb expression on the growth of YLCs from T. fuciformis was also investigated. Similar to the VHb effects observed in S. cerevisiae (Wilfred et al. 1994), the function of active Vitreoscilla hemoglobin protein in this modified strain could improve cell growth and final cell density at the hypoxic state, as compared to the wild-type strain. This metabolic effect of intracellular VHb was seen more clearly in fed-batch fermentation.

Our data revealed the significance of VHb protein in promoting the growth rate as well as extending the period of stationary phase in T. fuciformis. Probably, it was this particular property of bacterial hemoglobin that might play an important role in prolonging the viability of transformed cells. Also, it may improve their sluggish growth conditions, which was encountered during microaerobic conditions. Meanwhile, it was illustrated that VHb gene driven by CaMV35S promoter could be stably expressed throughout the cultivation and had a positive effect on the growth of cells.

Because the molecular mechanism of the influence of VHb on aerobic metabolism is not known in Vitreoscilla or in any other organism, the presence and function of VHb in enhancing respiration and formation of ATP has been considered to be the reason for the improvement in this system. To explain the VHb mechanism in T. fuciformis, detailed investigations must be performed at the molecular level.