Identification of two integration sites in favor of transgene expression in Trichoderma reesei
- 1.2k Downloads
The ascomycete fungus Trichoderma reesei was widely used as a biotechnological workhorse for production of cellulases and recombinant proteins due to its large capacity of protein secretion. Transgenesis by random integration of a gene of interest (GOI) into the genome of T. reesei can generate series of strains that express different levels of the indicated transgene. The insertion site of the GOI plays an important role in the ultimate production of the targeted proteins. However, so far no systematic studies have been made to identify transgene integration loci for optimal expression of the GOI in T. reesei. Currently, only the locus of exocellobiohydrolases I encoding gene (cbh1) is widely used as a promising integration site to lead to high expression level of the GOI. No additional sites associated with efficient gene expression have been characterized.
To search for gene integration sites that benefit for the secreted expression of GOI, the food-and-mouth disease virus 2A protein was applied for co-expression of an Aspergillus niger lipA gene and Discosoma sp. DsRed1 gene in T. reesei, by random integration of the expression cassette into the genome. We demonstrated that the fluorescent intensity of RFP (red fluorescent protein) inside of the cell was well correlated with the secreted lipase yields, based on which, we successfully developed a high-throughput screening method to screen strains with relatively higher secreted expression of the GOI (in this study, lipase). The copy number and the insertion sites of the transgene were investigated among the selected highly expressed strains. Eventually, in addition to cbh1 gene locus, two other genome insertion loci that efficiently facilitate gene expression in T. reesei were identified.
We have successfully developed a high-throughput screening method to screen strains with optimal expression of the indicated secreted proteins in T. reesei. Moreover, we identified two optimal genome loci for transgene expression, which could provide new approach to modulate gene expression levels while retaining the indicated promoter and culture conditions.
Keywords2A peptide Flow cytometry Trichoderma reesei Integration site Lipase Gene expression
gene of interest
red fluorescent protein
enhanced green fluorescent protein
cellobiohydrolase I gene
fluorescence-activated cell sorting
the food-and-mouth disease virus
autonomously replicating sequences
American type culture collection
The filamentous fungus Trichoderma reesei is industrially important fungi used for the production of cellulases and hemicellulases due to its large protein secretion capacity. The secreted cellulase quantities can exceed 100 g/L culture in several industrial strains after rounds of random mutagenesis . With this property, in addition to the production of cellulases, T. reesei has also been developed into a host platform to express a variety of recombinant proteins, including both native and heterologous proteins [2, 3]. It is well known that gene expression is strongly affected by its copy number in the genome and the local DNA features close to the integration sites [4, 5]. Indeed, we observed this phenomenon by successful expression of a heterologous lipase gene in T. reesei via random insertion of the expression cassette, from which we noticed that the expression levels of lipase varied in different transformants with different insertion site of the lipA gene . In T. reesei, targeting recombinant genes to the cbh1 locus is a typical approach to lead to high expression levels and cbh1 locus is currently the only integration site reported to have positive influence on the expression of recombinant genes . Little is known about whether there are other integration sites in T. reesei genome that would be benefit for gene expression, although T. reesei genome has been well sequenced and annotated [7, 8].
To search for gene integration sites benefit for gene expression, one of the foremost considerations is to develop a high-throughput, low-cost screen method for screening strains with high expression levels of the recombinant protein from a large population of recombinant strains. Fluorescence-activated cell sorting (FACS) is a specialized application of flow cytometry which is a sensitive and quantitative platform for the measurement of particle fluorescence, and is capable of sorting cells with a special characterization at the single cell level. It has been demonstrated that flow cytometric analysis and fluorescence-activated cell sorting of fungal cells is practicable and this technique has yielded valuable results in a number of different fields of research [9, 10, 11]. In T. reesei, germinating spores could be used for FACS sorting , however, most of the secreted proteins might not be well expressed during the stage of germinating spores. Therefore, using FACS based method to screen T. reesei strains bearing relatively high expression of GOI, especially when the GOI is a secreted protein encoding gene, still has lots of challenges. One of the problems is how to manipulate the branched hyphal mats in FACS instruments and how to make the connection between the production yield of extracellular protein and the cell phenotype that can be detected by flow cytometry.
The food-and-mouth disease virus (FMDV) 2A protein is a very small protein that only contains 16–20 amino acids and it is originally responsible for the cleavage of the FMDV poly-protein at its own carboxyl-terminus [13, 14]. When this 2A sequence was inserted between two or more independent genes to form a single ORF transcription unit, upon translation, the constituent proteins could be cleaved apart at the C-terminus of 2A sequence to generate two or more separated gene products [15, 16, 17]. With this property, the FMDV 2A peptide has been widely applied in co-expression of two or more genes in a variety of eukaryotic systems [18, 19, 20]. Especially the most recently, Subramanian and colleagues reported that heterologous co-expression of a secreted cellobiohydrolase enzyme (Cel7A from Penicillium funiculosum) and an intracellular enhanced green fluorescent protein (eGFP) linked by 2A peptide in T. reesei resulted in a equal expression ratio of eGFP and Cel7A, indicating that the FMDV 2A peptide is also practicable to co-expression multiple genes in T. reesei .
In this study, to search for the integration sites that will benefit for gene expression, we employed 2A peptide to co-express a RFP gene and a secreted heterologous lipase gene lipA in T. reesei and based on which, we established a FACS based high-throughput screening method to screen strains bearing relatively high expression levels of recombinant proteins. Several strains with highly expressed recombinant gene including a strain R5 that the recombinant gene was integrated into cbh1 locus were successfully screened. Furthermore, we surprisingly found two additional strains R3 and R11 that exhibited comparable expression levels of the recombinant gene as that in R5, while cbh1 gene was normally expressed. We subsequently investigated the effects of the integration sites in R3 and R11 on the transgene expression. Our results indicated that integration of recombinant genes into the loci identified in R3 and R11 could result in an optimal expression of the targeted genes. Our study herein provided a feasible and advantageous method to efficiently pick out hyper-secretion strains from a large population of strains that bearing random integration of transgene in T. reesei. Additionally, the two gene integration sites identified here provided new clues for strain engineering to improve the production of recombinant proteins and other bio-products in T. reesei.
Construction and expression of 2A self-cleavage peptide linked poly-protein gene in T. reesei
Three biologically independent positive transformants were inoculated into MM medium with 2% lactose (w/v) as the carbon source to analyze the expression of the recombinant RFP and lipase. The parental strain Tu6, which is a uridine auxotrophic strain, was used as the negative control. We firstly analyzed the effect of uridine on the protein secretion of strains harboring pyr4 gene by growing these three transformants with or without the addition of uridine. Our results indicated that uridine did not affect the protein secretion of strains with functional pyr4 gene (Additional file 1: Figure S1). Therefore, to make the growth conditions consistent, 5 mM uridine was added into the media regardless of whether the strains were uridine auxotrophic strains. After 48 h of induction in lactose, the expressions of RFP in all of these three transformants were observed as red fluorescent mycelium under the fluorescence microscope (Fig. 1b). No red fluorescent was observed in the parental strain Tu6. The extracellular lipase activities in Tu6 and these three transformants were determined in the supernatants of 96 h lactose culture. No lipase activity was detected in Tu6 and the three transformants exhibited obvious lipase activity varied from 10 to 25 IU/g biomass (Fig. 1c). SDS-PAGE analysis showed that compared to Tu6, all of these three transformants had extra lipase bands, further confirming the successful expression of the recombinant lipase (Fig. 1d).
To evaluate the cleavage efficiency of the 2A self-cleavage peptide, western blot analysis was performed using anti-6× His tag antibody to detect the recombinant protein in the supernatant of these three transformants. The data shown in Fig. 1e demonstrated that only three bands were detected in all of the three transformants. Previously, we demonstrated that expression of this lipA gene in T. reesei generated three separated peptides and MALDI-TOF-TOF mass spectrometry analysis indicated all these peptides bands were A. niger lipase , which was consistent with the western blot data here, suggesting that no uncleaved protein were detected in the supernatant. To determine whether the uncleaved LipA-2A-DsRed poly-protein failed to secrete and retained in the cell, western blot using the same anti-6× His tag antibody was performed to detect the intracellular proteins from the mycelia lysate. No significant band signal with the right size was detected in all the analyzed transformants (data not shown). These data indicated that the LipA and RFP from the single transcript frame were completely separated after translation using FMDV 2A self-cleavage peptide. Given that the co-expressed proteins via 2A peptide can be theoretically expressed at equal molar ratios, the fluorescent intensity of RFP in each transformant will be a good reporter to estimate the extracellular lipase production yields, which made it possible to use FACS to screen T. reesei strains with highly expression of secreted protein of interest.
FACS screen and quantification analysis of the correlation between red fluorescence intensity and lipase activity
A lipase hyper-secretion strain with cbh1 gene deletion was obtained from FACS screening
Two genome insertion sites benefit for gene expression were identified via plasmid rescue method
From the data shown in Fig. 3a, two strains R3 and R11 exhibited comparative production levels of lipase as that in R5, while cbh1 gene was normally expressed in these two strains. To figure out the factors associated with the higher expression levels of recombinant lipase in strain R3 and R11, gene copy number and the insertion sites of the transgene were analyzed. Quantitative PCR (qPCR) was employed to identify the copy number of lipase gene in strain R3 and R11 using N10, a strain only contains one copy of lipase gene, as the reference strain . The data shown in Additional file 3: Figure S3a indicated that both R3 and R11 contained only one copy of the expression cassette of Lipase-2A-DsRed. To determine the chromosomal location of the inserted transgene in strain R3 and R11, plasmid rescue strategy was used to capture the fragment including the flanking chromosomal region of the transgene. Total genomic DNA of R3 and R11 were extracted and digested with restriction enzyme Sal I, which is the only restriction site of plasmid pSKLR (Fig. 1a), and then religated and transformed into E. coli cells. Twenty four colonies for strain R3 and 33 colonies for strain R11 were recovered under selection against ampicillin. Considering that both the transformed plasmids pSKpyr4 and pSKLR contain ampicillin resistance gene, colony PCR using primers targeting in lipase gene was first performed to get rid of colonies containing the flanking genomic sequence at the pSKpyr4 insertion site. Eleven out of 24 colonies and 9 out of 33 colonies were verified to contain lipase gene in strain R3 and R11, respectively (Additional file 4: Figure S4a, b).
To determine whether the integration sites identified in this study were conserved in other Trichoderma species, we performed sequence alignments using 4 kb sequence surrounding the R3 and R11 insertion sites as queries to against 9 sequenced Trichoderma species including T. reesei Rut C-30, T. asperellum CBS 433.97, T. asperellum TR356, T. citrinoviride TUCIM 6016, T. gamsii T6085, T. harzianum CBS 226.95, T. harzianum TR274, T. longibrachiatum ATCC 18648 and T. virens. Gv29-8 (Additional file 6: Table S1). All of these genome sequences are available in JGI Genome Portal (https://genome.jgi.doe.gov/portal/). The results showed that T. reesei species RutC30 and QM6a shared 100% identity and 100% coverage with both 4 kb R3 and R11 loci surrounding sequences, while T. citrinoviride shared 98% coverage and 81.59% identity with R3 locus sequence, 87% coverage and 89.97% identity with R11 sequences, and T. longibrachiatum shared 75% coverage and 85.57% identity with R3 locus sequence, 88% coverage and 89.23% identity with R11 sequences. However, other Trichoderma species had a significantly lower coverage ratio with both R3 and R11 locus sequences (Additional file 6: Table S2), implying that the identified locus may be only applied for a few Trichoderma species.
Functional analysis of the above two transgene insertion sites
In general, the copy number and the integration site of an introduced transgene are highly related to the ultimate expression level of the target gene [4, 5, 23]. Since in both R3 and R11 strain, there was only one copy number of the Lipase-2A-DsRed cassette in the genome, the integration sites identified in R3 and R11 may play a key role in their higher expression level of the recombinant gene. In T. reesei, cbh1 gene was one of the highly expressed endogenous genes and targeting transgenes into cbh1 site was one of the efficient strategies to improve the expression levels of the indicated genes . Our data demonstrated that the production levels of lipase in R3 and R11 were comparable to that in strain R5, implying that the integration sites in R3 (R3 locus) and R11 (R11 locus) might have the same or even better effect on gene expression compared to cbh1 locus. To verify this hypothesis, we constructed strains, in which cbh1 gene was integrated into R3 locus, R11 locus, or other random sites in the genome, respectively. Meanwhile, cbh1 gene in its native locus was replaced by lipA gene. These generated strains were, respectively, named as R3cbh1, R11cbh1 and Rcbh1 accordingly. To improve the homologous recombination ratio, the strain Tu6∆ku70  was used as parental strain to generate R3cbh1 and R11cbh1 strains and Tu6 was used to generate Rcbh1 series strains. The copy number of cbh1 gene in these transformants was first determined using qPCR. The data shown in Additional file 3: Figure S3b only listed the resulted strains, which contains single copy number of cbh1gene.
It has been reported that the chromosome location of a trangene could affect its expression at the level of transcription [23, 25]. To investigate whether the effect of insertion sites on gene expression occurred at the transcriptional level or post-transcriptional level, qRT-PCR was performed in strain Tu6, Rcbh1-1, Rcbh1-2, R3, R3cbh1-1, R3cbh1-2, R11, R11cbh1-1 and R11cbh1-2 to detect the transcriptional levels of cbh1 gene. The data exhibited in Fig. 6d showed that the expression pattern of cbh1 gene in these strains at the transcriptional level was highly consistent with their protein expression pattern, suggesting that the position effects on gene expression in these concerned strains from the present study occurred at the transcriptional level.
Although the market share of T. reesei based cellulases and hemi-cellulases were diminishing in recent years, the application of using T. reesei as a cell factory to produce other valuable protein products still has great potential due to its large capacity of protein secretion. For construction of host strain to express recombinant proteins, gene copy number and the selected promoter are usually considered as the major factors to affect the transgene expression. However, the expression levels of the transgenes are quite different when placed in different chromosomal locations. This phenomenon has been reported in several systems such as Escherichia coli , Drosophila melanogaster , human cells , S. cerevisiae  as well as T. reesei . Considering that chromosomal integration was required for the construction of recombinant T. reesei strains to express homologous or heterologous genes, in addition to the gene copy number and the promoter, epigenetic effects in surrounding of the chromosomal integration site should also be an important consideration when constructing recombinant T. reesei strains.
Compared to the random integration, site-specific integration has lots of advantages such as the ability to direct transgenes to a neutral location to avoid insertion mutagenesis. However, since no systematic studies have been made to identify transgene integration loci that enable an optimal gene expression in T. reesei, site-specific integration could only be a downstream step of the random integration, due to the uncertainties about which loci could lead to the optimal transgene expression. Currently, the most commonly used site for targeted integration has been cbh1, a locus that harbors the expression of the main endogenous secreted protein CBHI in T. reesei. Integration of transgene into cbh1 locus to replace cbh1 gene has been proved to be an efficient strategy to result in sufficiently high levels of transgene expression . However, the strong cbh1 promoter was usually used in this strategy and suppression of cbh1 gene also can contribute to improve the expression level of target gene , which made it hard to determine whether the high expression levels were caused by the integration position effect. Furthermore, deletion of cbh1 gene can cause growth defect on cellulose based carbon sources, because of the decreased efficiency of releasing cellobiose from cellulose. Therefore, searching for additional favorable integration sites in T. reesei for secreted expression of transgenes will be especially important.
For the above purpose, we employed the FMDV 2A peptide to co-expression of a secreted heterologous lipase gene lipA and RFP protein in T. reesei. In this system, the fluorescent intensity of the intracellular RFP can be used as a good reporter to indicate the production levels of the secreted lipase. We then used this system to successfully develop a FACS based high throughput screening method to screen strains with optimal transgene expression, from a pool of recombinant strains bearing random integration. Among the 46 screened strains, strain R5, R3, and R11 caught our attention to perform further investigation. Our results showed that R5 was a strain that the transgene was inserted into cbh1 site and the native cbh1 gene was replaced, while R3 and R11 displayed similar expression levels of transgene without a deletion of cbh1 gene (Fig. 3a). The single copy number of transgene in strain R3 and R11 indicated that the integration site in these two strains might play a major role in the high transgene expression levels. The results of the recombinant expression of endogenous cbh1 gene in the newly identified R3 and R11 locus confirmed this hypothesis (Fig. 6). Our study here provided a promising screening method to screen strains with higher secreted expression of transgene and with this approach, we further revealed two previously unrecognized loci that enable transgenes be reliably expressed at high levels by integration into a single locus. However, these two genome loci were only conserved in a few Trichoderma species according to the sequences alignment data shown in Additional file 6: Table S2. In addition, one of the limitations of the screening method we developed here is the low transformation efficiency of T. reesei. The efficiency of the commonly used PEG mediated protoplast transformation was only 200–300 colonies per microgram plasmid DNA when using pyr4 as the selected marker , which is far from to cover all of the genome insertion loci. Hence, establishment of a more efficient transformation system will contribute to identify more integration sites that benefit for the transgene expression.
Integration position effects on transgene expression levels have been reported to associate with chromatin structure . For example, transcriptional silencing could be caused if the transgenes located in telomeric region , while higher transcriptional level could be obtained when transgenes were close to the DNA replication origins . Our results showed that the high expression of transgenes occurred at the transcriptional level in both R3 and R11 integration (Fig. 6d). Further analysis of the sequence around R11 locus revealed that there were about 5 kb AT rich sequences in the downstream of insertion site (Fig. 5c and Additional file 5), which is a part of the intron region of a predicted fungal specific transcription factor (protein ID 68425) encoding gene. It has been well known that the high AT content sequence is a common character for DNA replication initiation in bacterial, archaeal, and eukaryotic replicons . In yeast, AT-rich DNA sequence can contribute to remove the nucleosomes by the RSC chromatin remodeling complex to form the nucleosome-free regions (NFRs) . Considering that NFRs were associated with the initiation of transcription of most genes, the high transgene expression levels in R11 locus might be benefited from the 5 kb AT-rich sequence in the downstream of the insertion site, although little studies associated the function of AT-rich DNA sequence was performed in T. reesei to date.
For R3 locus, the recombinant gene was integrated into the 5′UTR region of the cel3c gene. No specific characterized sequence was found around the R3 locus. However, integration occurred in R3 locus might disrupt the expression of cel3c gene, which predicted to be a β-glucosidase encoding gene . Recently, it has been reported that dysfunction of a β-glucosidase encoding gene cel3d in T. reesei resulted in higher secretion of cellulases . In addition, in N. crassa, deletion of β-glucosidases could efficiently decrease the carbon catabolite repression (CCR) effect, thereby allowing the induction of cellulases under cellobiose, cellotriose and cellotetraose . Considering that decreased CCR could result in higher expression of cellulose-induced proteins , we presumed that the high expression level in R3 locus might result from the disruption of cel3c gene. To verify this hypothesis, we made a cel3c deletion strain and compared the total secreted protein levels and cellulase activity between ∆cel3c and its parental strain. However, unexpectedly, our data demonstrated that the deletion of cel3c gene had no significant influence on the expression of secreted proteins (data not shown). We then checked the upstream region of the cel3c gene and found the promoter region of cel3c has been predicted to have putative XYR1-binding site . It is well known that the transcriptional factor xyr1 is the major activator for most cellulases genes, which might contribute to the high expression level of the targeted gene. However, the mechanism behind the R3 and R11 position effect on the transgene expression still needs to be further investigated.
The chromosomal location of transgenes plays an important role in the ultimate recombinant protein yield. To search and identify loci for optimal transgene expression in T. reesei, we employed 2A mediated multiple proteins co-expression system to simultaneously express a secreted lipase gene and the RFP gene, by randomly integration of the expression cassette into T. reesei genome. Our data demonstrated that in this system, the production levels of the extracellular lipase were well correlated with the intracellular RFP fluorescent intensity, thereby allowing us to perform flow cytometry sorting for screening strains with better secretion yields of the transgene (in this case, lipA gene). We subsequently investigated the copy number and the integrated sites among the screened strains and eventually identified two optimal loci for transgene expression in T. reesei. In the short term our study here provided a promising strategy to construct optimal T. reesei host strains for the production of recombinant proteins, and in the longer term, it raised the question how these two loci eventually affected the gene expression. Further investigation of the behind mechanism will contribute to a biological understanding of the epigenetic effects on transgene expression.
Microbial strains and growth conditions
Trichoderma reesei Tu6 strain (ATCC MYA-256)  was obtained from American Type Culture Collection (ATCC). T. reesei Tu6∆ku70  was kindly provided by Prof. Dr. Monika Schmoll (AIT Austrian Institute of Technology, Austria). T. reesei strain R3, R5, R11 were constructed by co-transforming strain Tu6 with plasmid pSK-pyr4  and pSKLR. The plasmid of pSKLR was derived from plasmid pBluescript SK (+) by inserting a fused DNA fragment containing the dsRed1 gene, 2A sequence and a heterologous lipase gene lipA under the control of the cbh1 promoter (Pcbh1) and cbh1 terminator (Tcbh1) from T. reesei (Fig. 1a). Transformants were selected on MM media  without adding uridine and verified by diagnostic PCR.
The R3cbh1 and R11cbh1 series strains were generated by co-transforming strain Tu6∆ku70 with three DNA fragments including a linearized pSK-pyr4, a fragment containing the expression cassette of Pcbh1-lipA-Tcbh1 and a fragment including the expression cassette of Pcbh1-cbh1-Tcbh2 flanked with 2 kb upstream and downstream DNA sequence of R3 or R11 locus. ∆cbh1::lipA strains were generated by only co-transforming strain Tu6∆ku70 with two DNA fragments including a linearized pSK-pyr4 and a fragment containing the expression cassette of Pcbh1-lipA-Tcbh1. Transformants were selected on minimal media without adding uridine and tested for genotypes by diagnostic PCR. The Rcbh1 series strains were created by co-transforming strain Tu6 with linearized plasmid pSK-pyr4 and a fragment including the expression cassette Pcbh1-cbh1-Tcbh2. Transformants were selected on minimal media without the addition of uridine and verified by diagnostic PCR. Strain ∆cbh1 was generated by transforming strain Tu6∆ku70 with a DNA fragment that contained pyr4 expression cassette flanked with 2 kb upstream and downstream DNA sequence of cbh1 gene. Transformants were selected on minimal media without uridine and verified by diagnostic PCR. All the primers used in this study were listed in Additional file 7: Table S3.
For conidiation, T. reesei strains were grown for 5–6 days at 28 °C on potato dextrose agar plates (PDA) or PDA supplemented with 5 mM uridine when necessary. For measurement of the secreted proteins, 2 × 106 conidia/mL were inoculated into 50 mL of liquid minimal medium with the indicated carbon source in 250 mL flasks and grown at 28 °C on a rotary shaker (200 rpm) in continuous dark condition for the indicated time. 5 mM uridine was added when using Tu6 or Tu6∆ku70 as a control to compare the protein secretion change in other recombinant strains, which derived from these two strains. 1% glycerol (w/v) was added when using strains with cbh1 gene deletion background.
Sphere-protoplasting and transformation of T. reesei
The indicated T. reesei strains were inoculated into slants and cultivated for 5–7 days for conidiation. The conidia were suspended by adding 3 mL sterile H2O into the slants and filtered through a metal sieve of 200-mesh. Then the filtered conidia were inoculated into 100 mL minimal medium with 2% glucose as carbon source at 2× 106/mL and grown at 28 °C on a rotary shaker (220 rpm) for 13–14 h. The germinated mycelium from this culture was collected by filtering through a 200-mesh sieve and re-suspended into 15 mL of 0.2 M phosphate buffer, pH7.4, which contained 150 mg lysing enzyme (Sigma, L1412) and 15 mg cellulase (Onozuka, R-10), and incubated at 30 °C on a rotary shaker (80 rpm) for 1.5–2 h. The culture was added with equal volume of 0.6 M sorbitol solution including 0.6 M sorbitol and10 mM Tris–HCl, pH7.0 and filtered using 200-mesh sieve to remove the undigested mycelia. The flowthrough including protoplasts were centrifuged at room temperature at 2800 rpm for 6–8 min to precipitate the protoplasts. The protoplasts were washed twice using 1.2 M sorbitol solution containing 1.2 M sorbitol, 50 mM CaCl2 and 10 mM Tris–HCl, pH7.4 and re-suspended in 200 μL of the same solution.
For transformation, the above protoplasts suspension was mixed with 5–10 μg transformed DNA (the total DNA volume should be ≤ 20 μL) and 50 μL PEG solution including 50% PEG4000, 50 mM CaCl2 and 10 mM Tris–HCl, pH7.4. The mixture was incubated on ice for 30 min and then added with 1 mL PEG solution and incubated at room temperature for 20 min. 1 mL sorbitol was added after PEG treatment. For normal transformation, the mixture was added into MM media with 0.8% agar and spread onto selective plates containing 1 M sorbitol and the plates were incubated at 28 °C for 3–5 days . For flow cytometry screening, the protoplasts preparation and transformation were performed as described above, excepting that the last step was changed (see the details below).
Flow cytometry and FACS screening
The construct pSKLR was co-transformed into the T. reesei strain Tu6 (an auxotrophic strain) with the plasmid pSKpyr4, which contains the pyr4 gene that complements uridine auxotrophic stains. For regeneration, the transformed protoplast suspension was transferred into 50 mL minimal medium as described previously , except that the carbon source 2% glucose (w/v) was substituted with 0.1% glucose (w/v), 0.1% glycerol (w/v) and 2% lactose (w/v), in addition, 1 M sorbitol was included as osmotic stabilizer.
For flow cytometry analysis, after 72 h incubation, all of the regenerated mycelium above was collected for another protoplast preparation. Protoplast suspensions were filtered through a metal sieve of 400-mesh and cytometrically analyzed and sorted with a FACSAria (BD Biosciences) using phosphate buffered saline as a sheath fluid. The sheath pressure was set at 70 psi, and the defection plate voltage was set at 5000 V (default ‘‘low’’ setting). A 488-nm coherent sapphire solid state laser was used for excitation, and emission was measured at 576/26 nm. The photomultiplier tube voltage was set at 330 V for forward scatter, 330 V for side scatter, and 650 V for RFP. The threshold value for event detection was set at 5000 on forward scattering. The drop drive frequency was set to approximately 87 kHz, and the amplitude was set to approximately 33 V; the drop delay value was approximately 44.78.
Protoplasts with the highest fluorescence value (top 0.03%) were directly sorted into 1 M sorbitol included minimal medium with 0.1% glucose (w/v), 0.1% glycerol (w/v) and 2% lactose (w/v) as carbon source and incubated 72 h at 28 °C. Protoplast preparation and sorting procedure were repeated. In the second round of sorting, single protoplasts were sorted into individual wells of 24 well plates containing the same medium as above.
Measurement of in vivo RFP fluorescence in T. reesei mycelium
The sorted protoplasts in three 24-well plates were incubated for 96 h at 600 rpm at 28 °C in a microplate shaker (Multitron II. Infors HT). The mycelia from each well were then filtered through 200 mesh filter (30 µm pore diameter) and added 500 µL Tris–HCl (pH7.5) and lysed using a mini-bead beater (Biospec Products, Bartlesville, Okla.) with 0.5 mm diameter glass beads. The mixture was centrifuged for 5 min at 12,000 rpm and the supernatant was carefully removed for analysis. The RFP fluorescent was measured using a Synergy H4 Hybrid Microplate Reader with 557 nm as the excitation wavelength and 585 nm as the emission wavelength.
Enzyme activity assay of lipase and cellobiohydrolase
Lipase activity was quantitatively determined by an alkali titration method  using olive oil as the substrate when using the supernatant of T. reesei culture from flasks. The reaction was carried out in 50 mM Tris–HCl buffer, pH 7.5 for 10 min at 45 °C. One unit of lipase activity was defined as the amount of lipase necessary to liberate 1 µmol fatty acid from olive oil per min under the standard assay conditions. Lipase activity was assayed by the colorimetric method using 4-nitrophenyl palmitate as substrate when the supernatant of T. reesei culture from 24 well plates was used for analysis. The assay was performed as described by Kumar , except that the reaction temperature was 40 °C. One unit (IU) of lipase activity was defined as the amount of enzyme that liberates 1 µmol 4-nitrophenol per minute under assay conditions. Cellobiohydrolase activity was measured with soluble 4-methylumbelliferyl-ß-d-cellobiose (Sigma) as the substrate as previously described .
Total proteins (about 50–100 µg) from the supernatant of 96 h shake flasks in minimal medium with 2% lactose (w/v) as the carbon source were separated by SDS-PAGE on 12.5% polyacrylamide gel. Proteins were blotted onto PVDF membrane using Trans-Blot Electrophoretic Transfer (BioRad). The membrane was treated with a diluted (1:1000) anti-His antibody (Tiangen, China) and detected with BCIP/NBT detection kit (CWBIO, China) according to the manufacturer’s introduction. For the western blot analysis of RFP and beta-actin, intracellular proteins were extracted by grinding the frozen mycelia to a fine powder and adding with the HEPES lysis buffer including 50 mM HEPES pH 7.5, 5 mM EDTA pH 8.0, 2 mM EGTA pH 8.0, 100 mM NaCl, 1% Triton X (v/v) and 10% glycerol (w/v). The mixture was centrifuged at 12,000 rpm for 10 min at 4 °C. The supernatant was used to perform western blot using anti-RFP antibody and anti-beta-actin antibody (YESEN, China).
Plasmid rescue method
Genomic DNA was extracted from strain R3 and R11 as follows: 100 mg mycelium was break down in 400 μL lysis buffer (50 mM sodium phosphate at pH7.4, 1 mM EDTA and 5% glycerol) with 0.3 g silica beads (0.5 mm) using Bead Beater (BioSpec, USA). The lysis was incubated at 65 °C for 30 min and thereafter added with 80 μL Tris–HCl, pH 7.5. The supernatant was transferred into another tube and added with an equal volume of phenol:chloroform and centrifuged. DNA pellet was precipitated by 2 volume of 100% ethanol and washed by 75% ethanol (v/v) and dissolved into 20 μL ddH2O.
10 μg of gDNA was digested with Sal I at 37 °C for overnight. The digested DNA was precipitated with 3 M sodium acetate, pH 5.2 and ethanol and dissolved in 50 µL ddH2O. T4 DNA ligase was used for cyclizing the digested DNA and the reactions were incubated overnight at 16 °C and thereafter precipitated with 3 M sodium acetate, pH 5.2 and ethanol. Samples were dissolved in 10 µL of ddH2O and 1 µg of ligated DNA was transformed into Escherichia coli strain DH5α competent cells (TransGene Biotech, China) and colonies were selected for ampicillin resistance. The plasmids included in these colonies were sequenced to capture the flanking genomic sequence at insertion sites.
Gene copy number analysis by qPCR
Genome DNA of the tested strains was extracted as described as above. The DNA concentration and purity was analysed by spectrophotometry (NanoDrop 2000C, Thermo Scientific) and thereafter diluted to the concentration of 20 ng/μL for qPCR reaction. qPCRs were performed in an ABI 7000 real-time detection system using a TransStart Green qPCR SuperMix (TransGene Biotech, China). Total reaction volume for each sample was 20 μL and contained 20 ng of genomic DNA, 10 pmol of each primer, 10 μL of SYBR Green I master mix (TransStart Green qPCR SuperMix, TransGene Biotech, China) and PCR condition was used as the default protocol recommended by the manufacturer. All PCRs were carried out in triplicates within a plate, and two different plates were set up for the same samples. Analysis of the expression level was done using actin gene as a reference. The Ct value of each test gene was calculated by subtraction of reference gene Ct value from each test gene Ct value (Ct = Cttest − Ctactin). A control strain, which was known to only contain one copy number of the test gene, was used as reference strain to quantify the gene copy number of the test strains by the 2−ΔΔCT method .
Statistical significance tests
Statistical significance was determined by t test analysis by the false discovery rate (FDR) approach using the Prism GraphPad software. Asterisks indicate significant differences (* P < 0.05; ** P < 0.01; *** P < 0.001). ns, not significant.
LQ participated in the conception of the study and carried out the majority of the experiments and prepared the manuscript; XJ was involved in FACS screening and western blotting; ZD and JH was involved in the project leadership and participated in technical directions and editing the manuscript; XC was involved in the conception of the study and participated in the guidance with experimental strategies and technical direction. All authors read and approved the final manuscript.
We thank Monika Schmoll (AIT Austrian Institute of Technology, Austria) for providing the ku70 gene deletion strain Tu6∆ku70; We thank Tong Zhao (Institute of Microbiology, Chinese Academy of Sciences) for her assistance with the flow cytometric sorting experiment.
The authors declare that they have no competing interests.
Availability of supporting data
All data generated and analyzed during this study are included in this article and its supplementary information files.
Ethics approval and consent to participate
This work was supported by the National Natural Science Foundation of China (30970073 and 31741002) and by the scientific research innovation team construction program of Fujian Normal University (IRTL1702).
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
- 21.Subramanian V, Schuster LA, Moore KT, Taylor LE 2nd, Baker JO, Vander Wall TA, Linger JG, Himmel ME, Decker SR. A versatile 2A peptide-based bicistronic protein expressing platform for the industrial cellulase producing fungus, Trichoderma reesei. Biotechnol Biofuels. 2017;10:34.CrossRefPubMedPubMedCentralGoogle Scholar
- 26.Sousa C, de Lorenzo V, Cebolla A. Modulation of gene expression through chromosomal positioning in Escherichia coli. Microbiology+. 1997;143:2071–8.Google Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.