Limits...
A Comparison of transgenic and wild type soybean seeds: analysis of transcriptome profiles using RNA-Seq.

Lambirth KC, Whaley AM, Blakley IC, Schlueter JA, Bost KL, Loraine AE, Piller KJ - BMC Biotechnol. (2015)

Bottom Line: However, the effects of expressing recombinant protein at high levels on bean physiology are not well understood.ST77 (hTG) and ST111 (hMBP) had significantly less differences with 52 and 307 differentially expressed genes respectively.Gene ontology enrichment analysis found that the upregulated genes in the 764 line were annotated with functions related to endopeptidase inhibitors and protein synthesis, but suppressed expression of genes annotated to the nuclear pore and to protein transport.

View Article: PubMed Central - PubMed

Affiliation: Department of Biological Sciences, University of North Carolina at Charlotte, Charlotte, NC, 28223, USA. kclambirth@uncc.edu.

ABSTRACT

Background: Soybean (Glycine max) has been bred for thousands of years to produce seeds rich in protein for human and animal consumption, making them an appealing bioreactor for producing valuable recombinant proteins at high levels. However, the effects of expressing recombinant protein at high levels on bean physiology are not well understood. To address this, we investigated whether gene expression within transgenic soybean seed tissue is altered when large amounts of recombinant proteins are being produced and stored exclusively in the seeds. We used RNA-Seq to survey gene expression in three transgenic soybean lines expressing recombinant protein at levels representing up to 1.61 % of total protein in seed tissues. The three lines included: ST77, expressing human thyroglobulin protein (hTG), ST111, expressing human myelin basic protein (hMBP), and 764, expressing a mutant, nontoxic form of a staphylococcal subunit vaccine protein (mSEB). All lines selected for analysis were homozygous and contained a single copy of the transgene.

Methods: Each transgenic soybean seed was screened for transgene presence and recombinant protein expression via PCR and western blotting.  Whole seed mRNA was extracted and cDNA libraries constructed for Illumina sequencing.  Following alignment to the soybean reference genome, differential gene expression analysis was conducted using edgeR and cufflinks.  Functional analysis of differentially expressed genes was carried out using the gene ontology analysis tool AgriGO.

Results: The transcriptomes of nine seeds from each transgenic line were sequenced and compared with wild type seeds. Native soybean gene expression was significantly altered in line 764 (mSEB) with more than 3000 genes being upregulated or downregulated. ST77 (hTG) and ST111 (hMBP) had significantly less differences with 52 and 307 differentially expressed genes respectively. Gene ontology enrichment analysis found that the upregulated genes in the 764 line were annotated with functions related to endopeptidase inhibitors and protein synthesis, but suppressed expression of genes annotated to the nuclear pore and to protein transport. No significant gene ontology terms were detected in ST77, and only a few genes involved in photosynthesis and thylakoid functions were downregulated in ST111. Despite these differences, transgenic plants and seeds appeared phenotypically similar to non-transgenic controls. There was no correlation between recombinant protein expression level and the quantity of differentially expressed genes detected.

Conclusions: Measurable unscripted gene expression changes were detected in the seed transcriptomes of all three transgenic soybean lines analyzed, with line 764 being substantially altered. Differences detected at the transcript level may be due to T-DNA insert locations, random mutations following transformation or direct effects of the recombinant protein itself, or a combination of these. The physiological consequences of such changes remain unknown.

No MeSH data available.


Related in: MedlinePlus

GOseq results from the edgeR and cufflinks merged DE gene list. The numbers over each bar indicate the total number of differentially expressed (DE) genes in each category. The length and direction of the arrow in each bar indicate how many of the genes were up (arrows pointing up) or down (arrows pointing down) regulated in that category. The red dashed line indicates the percentage of all GO-annotated genes that are DE from edgeR and cufflinks in this line (~4 %). All terms shown have an FDR of 0.05 or smaller
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
getmorefigures.php?uid=PMC4591623&req=5

Fig7: GOseq results from the edgeR and cufflinks merged DE gene list. The numbers over each bar indicate the total number of differentially expressed (DE) genes in each category. The length and direction of the arrow in each bar indicate how many of the genes were up (arrows pointing up) or down (arrows pointing down) regulated in that category. The red dashed line indicates the percentage of all GO-annotated genes that are DE from edgeR and cufflinks in this line (~4 %). All terms shown have an FDR of 0.05 or smaller

Mentions: We next performed a gene ontology (GO) enrichment analysis using GOseq [44] which accounts for selection biases in RNA-Seq data in which larger, more highly expressed transcripts are preferentially detected as differentially expressed. In line 764 we detected at least one read sequence from ~42,000 of the 56,000 annotated soybean genes, and of these ~42,000 expressed genes, approximately 3800 (9 %) were differentially expressed. The input list consisted of approximately 1500 genes that were considered differentially expressed after combining the lists from both edgeR and cufflinks. Thus, on average, we expected that approximately 3.5 % of genes in any random sample of expressed genes would be differentially expressed. However, there were several GO categories that exceeded this 3.5 % threshold and are grouped according to their parent terms in Fig. 7. A more detailed flowchart of all GO terms can be found in Additional file 1: Figure S3. Of 16 genes annotated to the term “nuclear pore”, nine were differentially expressed, and all were downregulated. Of the 490 genes annotated to the term “structural constituent of ribosome”, 47 were differentially expressed, and of these 94 % were upregulated. All DE genes annotated as protease inhibitors were upregulated, including 8 of 19 genes encoding serine-type endopeptidase inhibitors, and 10 of 60 genes encoding peptidase inhibitor and regulator activity. Intracellular transport also appeared affected in the 764 samples, as 8 % annotated to non-membrane intracellular organelles were differentially expressed and most (82 %) were upregulated. All 10 DE genes encoding mitochondrial function were also upregulated. In addition, several genes (5 of 8) annotated with the biological process term “response to wounding” were upregulated. Taken together, these results suggested that protein synthesis was more active in the 764 seeds as compared to the nontransgenic controls. These results also suggest that aspects of intracellular transport and nuclear pore structures may be altered. The annotation of upregulated genes involved in wounding responses and peptidase inhibitors suggests that some aspects of a physical stress response may have been activated.Fig. 7


A Comparison of transgenic and wild type soybean seeds: analysis of transcriptome profiles using RNA-Seq.

Lambirth KC, Whaley AM, Blakley IC, Schlueter JA, Bost KL, Loraine AE, Piller KJ - BMC Biotechnol. (2015)

GOseq results from the edgeR and cufflinks merged DE gene list. The numbers over each bar indicate the total number of differentially expressed (DE) genes in each category. The length and direction of the arrow in each bar indicate how many of the genes were up (arrows pointing up) or down (arrows pointing down) regulated in that category. The red dashed line indicates the percentage of all GO-annotated genes that are DE from edgeR and cufflinks in this line (~4 %). All terms shown have an FDR of 0.05 or smaller
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

License 1 - License 2
Show All Figures
getmorefigures.php?uid=PMC4591623&req=5

Fig7: GOseq results from the edgeR and cufflinks merged DE gene list. The numbers over each bar indicate the total number of differentially expressed (DE) genes in each category. The length and direction of the arrow in each bar indicate how many of the genes were up (arrows pointing up) or down (arrows pointing down) regulated in that category. The red dashed line indicates the percentage of all GO-annotated genes that are DE from edgeR and cufflinks in this line (~4 %). All terms shown have an FDR of 0.05 or smaller
Mentions: We next performed a gene ontology (GO) enrichment analysis using GOseq [44] which accounts for selection biases in RNA-Seq data in which larger, more highly expressed transcripts are preferentially detected as differentially expressed. In line 764 we detected at least one read sequence from ~42,000 of the 56,000 annotated soybean genes, and of these ~42,000 expressed genes, approximately 3800 (9 %) were differentially expressed. The input list consisted of approximately 1500 genes that were considered differentially expressed after combining the lists from both edgeR and cufflinks. Thus, on average, we expected that approximately 3.5 % of genes in any random sample of expressed genes would be differentially expressed. However, there were several GO categories that exceeded this 3.5 % threshold and are grouped according to their parent terms in Fig. 7. A more detailed flowchart of all GO terms can be found in Additional file 1: Figure S3. Of 16 genes annotated to the term “nuclear pore”, nine were differentially expressed, and all were downregulated. Of the 490 genes annotated to the term “structural constituent of ribosome”, 47 were differentially expressed, and of these 94 % were upregulated. All DE genes annotated as protease inhibitors were upregulated, including 8 of 19 genes encoding serine-type endopeptidase inhibitors, and 10 of 60 genes encoding peptidase inhibitor and regulator activity. Intracellular transport also appeared affected in the 764 samples, as 8 % annotated to non-membrane intracellular organelles were differentially expressed and most (82 %) were upregulated. All 10 DE genes encoding mitochondrial function were also upregulated. In addition, several genes (5 of 8) annotated with the biological process term “response to wounding” were upregulated. Taken together, these results suggested that protein synthesis was more active in the 764 seeds as compared to the nontransgenic controls. These results also suggest that aspects of intracellular transport and nuclear pore structures may be altered. The annotation of upregulated genes involved in wounding responses and peptidase inhibitors suggests that some aspects of a physical stress response may have been activated.Fig. 7

Bottom Line: However, the effects of expressing recombinant protein at high levels on bean physiology are not well understood.ST77 (hTG) and ST111 (hMBP) had significantly less differences with 52 and 307 differentially expressed genes respectively.Gene ontology enrichment analysis found that the upregulated genes in the 764 line were annotated with functions related to endopeptidase inhibitors and protein synthesis, but suppressed expression of genes annotated to the nuclear pore and to protein transport.

View Article: PubMed Central - PubMed

Affiliation: Department of Biological Sciences, University of North Carolina at Charlotte, Charlotte, NC, 28223, USA. kclambirth@uncc.edu.

ABSTRACT

Background: Soybean (Glycine max) has been bred for thousands of years to produce seeds rich in protein for human and animal consumption, making them an appealing bioreactor for producing valuable recombinant proteins at high levels. However, the effects of expressing recombinant protein at high levels on bean physiology are not well understood. To address this, we investigated whether gene expression within transgenic soybean seed tissue is altered when large amounts of recombinant proteins are being produced and stored exclusively in the seeds. We used RNA-Seq to survey gene expression in three transgenic soybean lines expressing recombinant protein at levels representing up to 1.61 % of total protein in seed tissues. The three lines included: ST77, expressing human thyroglobulin protein (hTG), ST111, expressing human myelin basic protein (hMBP), and 764, expressing a mutant, nontoxic form of a staphylococcal subunit vaccine protein (mSEB). All lines selected for analysis were homozygous and contained a single copy of the transgene.

Methods: Each transgenic soybean seed was screened for transgene presence and recombinant protein expression via PCR and western blotting.  Whole seed mRNA was extracted and cDNA libraries constructed for Illumina sequencing.  Following alignment to the soybean reference genome, differential gene expression analysis was conducted using edgeR and cufflinks.  Functional analysis of differentially expressed genes was carried out using the gene ontology analysis tool AgriGO.

Results: The transcriptomes of nine seeds from each transgenic line were sequenced and compared with wild type seeds. Native soybean gene expression was significantly altered in line 764 (mSEB) with more than 3000 genes being upregulated or downregulated. ST77 (hTG) and ST111 (hMBP) had significantly less differences with 52 and 307 differentially expressed genes respectively. Gene ontology enrichment analysis found that the upregulated genes in the 764 line were annotated with functions related to endopeptidase inhibitors and protein synthesis, but suppressed expression of genes annotated to the nuclear pore and to protein transport. No significant gene ontology terms were detected in ST77, and only a few genes involved in photosynthesis and thylakoid functions were downregulated in ST111. Despite these differences, transgenic plants and seeds appeared phenotypically similar to non-transgenic controls. There was no correlation between recombinant protein expression level and the quantity of differentially expressed genes detected.

Conclusions: Measurable unscripted gene expression changes were detected in the seed transcriptomes of all three transgenic soybean lines analyzed, with line 764 being substantially altered. Differences detected at the transcript level may be due to T-DNA insert locations, random mutations following transformation or direct effects of the recombinant protein itself, or a combination of these. The physiological consequences of such changes remain unknown.

No MeSH data available.


Related in: MedlinePlus