Growth of Euglena gracilis and Euglena mutabilis in pre-treatment media
With a goal of assessing the impact of elevated S and N on Cd tolerance, the impact of these pre-treatments on Euglena growth was assessed. Cell counts of E. gracilis were reported every 4 days while those of E. mutabilis were recorded every 7 days (Fig. 1). Both species were grown in MAM (Modified Acid Medium) [16], and MAM with elevated S, or N. The different sampling intervals were selected to be consistent with previous research on these organisms [10, 17, 18, 19]. Both E. gracilis and E. mutabilisexhibited the greatest increase in cell counts in cultures pre-treated with elevated concentrations of S, while the lowest increase in cell counts was in cultures pre-treated with elevated concentrations of N.
|
Figure 1: Growth of E. gracilis (a) and E. mutabilis (b) is unaffected by increased S but decreased with increased N. MAM with elevated S, and MAM with elevated N. 20,000 cells of E. gracilis and E. mutabilis were inoculated into their treatments and grown for 44 and 42 days, respectively, with E. gracilis cells were counted and transferred into fresh media every 4 days while E. mutabilis cells were counted and transferred every 7 days. A one-way ANOVA was used to calculate statistical difference in cell counts of the S (MAM + MgSO4) or N (MAM + NH3NO3) treatment compared to control growth (MAM). Differences between control cells and cells in the S treatment are denoted by an asterixis (*), while differences between control cells and cells in the N treatment are denoted by a dagger (☨). Error bars represent standard deviation in viable cell counts between biological replicates.
|
Euglena gracilis and Euglena mutabilis cell viability after exposure to 25\(\varvec{\mu }\varvec{M}\) CdCl2
Pretreating E. gracilis cultures with elevated concentrations of S and N prior to CdCl2 exposure resulted in a higher cell counts, after 8 days, relative to non-pretreated cells (Fig. 2a, b). In contrast, pretreated E. mutabilis cultures did not show a consistent or statistically significant difference in viable cells relative to non-pretreated cells after 21 days of CdCl2 exposure (Fig. 2b). The impact of pre-treatment on E. gracilis prompted an investigation, using RNA-sequencing, to determine which genes were responding to cadmium exposure, and if there are differences in gene transcript levels in cells that have been pre-treated with S or N relative to those that were not pre-treated. The time course was repeated for a shortened 8 days and at this point there was a significant difference in cells counts between the pretreated and non-pretreated cultures and there was enough biomass for reliable RNA isolation (Fig. 2b). Cultures from an 8 day time course were used for RNA isolation.
While there was an increase in E. gracilis tolerance to CdCl2 with pre-treatment, E. mutabilis cultures exposed to CdCl2 with or without pre-treatment in elevated S or N had a higher cell count than the E. gracilis cultures at 14 and 21 days. With no pre-treatment or following elevated S or N pre-treatment the E. mutabilis cultures contained 3.5 times, 1.9 times, and 1.8 times the number of viable cells as the 20 day E. gracilis cultures (Fig. 2a, c). This indicated that only E. gracilis responded to pre-treatment but that E. mutabilis was the more tolerant culture.
Trypan Blue Solution (Sigma-Aldrich, Mississauga) was used to distinguish between living and dead cells in E. gracilis and E. mutabilis cultures to assess cell viability (Table 1). While there were substantially fewer E. gracilis cells present in the cultures exposed to CdCl2 the fraction of dead cells was higher in the E. mutabilis cultures. Assessing the differences in cell growth and cell viability between species indicated that only E. gracilis had increased CdCl2 tolerance following pre-treatment with S or N. Therefore, to investigate the impact of S and N pre-treatment on gene expression change E. gracilis cultures were chosen and the 8 day time point selected to ensure a statistical difference in growth of pre-treated cultures and sufficient biomass for RNA isolation.
RNA-seq, Differential Gene Expression, and Gene Ontology Term Analysis
RNA-sequencing was carried out on six different E. gracilis samples: non-pretreated cultures, S-pretreated cultures, and N-pretreated cultures both with and without exposure to 25μM CdCl2. Each sample type had six biological replicates sequenced for a total of 36 samples with an average of 23.3 million raw paired end (PE) reads per library (Table S1). Following the removal of adapter and poor-quality sequences there was an average of 19.5 million trimmed PE reads per library. A de novo transcriptome was assembled using a combined total of 117.3 million trimmed PE reads. Once completed, the transfrag assembly represented 838,537 transcripts with an N50 value of 892 and these transcripts were from 459,533 Euglena gracilis genes. BUSCO analysis comparing E. gracilis transfrag assembly to the eukaryota Orthodb v10 orthologs showed 82.8% of orthologs within the database were found in the E. gracilis de novo assembly. An average of 23.2 million trimmed paired-end reads were aligned to the de novo transcriptome assembly across all libraries (Table S2).
The results of the clustering analyses and the results of DGE analysis for pretreated and non-pretreated E. gracilis cultures exposed to 0 M and 25 M CdCl2 are presented in supplementary Figures S1-S2. Overall, DGE analysis of the non-pretreated, S pretreated and N pretreated E.
gracilis cultures identified 185, 136 and 311 unique transcripts respectively (Figure 3). The results of the DGE analysis indicated that the response of E. gracilis to CdCl2 varied depending on the pre-treatment used. While the transcripts that did not overlap are of significance, the overlap of the 10 transcripts differentially expressed in all treatments and the 22 that specifically overlapped between the S and N pretreated cultures represent genes that could be investigated in the future. That investigation would benefit from improved annotation of the E. gracilis genome sequence.
To identify the potential genes encoding proteins that were associated with these transcript level changes, a blastx search was conducted using the NCBI non-redundant protein and SWISS-Prot databases as well as the Ensembl databases for Arabidopsis thaliana, Chlamydomonas reinhardtii, Synechocystis sp., Homo sapiens, and Trypanosoma brucei. Overall, the results of the differential gene expression analysis identified four main classes of genes encoding proteins that showed transcript level changes. These four classes are transmembrane transport, stress response, metabolism, and metal binding (Figure 4). When E. gracilis cultures were pretreated with S and N prior to CdCl2 exposure, genes encoding potential ABC-transporter proteins showed decreased transcript levels when compared to non-pretreated cultures that were exposed to CdCl2. Differential gene expression (DGE) analysis found transcript level changes for genes encoding various stress response proteins including potential chemotaxis proteins and heat shock proteins. When cultures were grown without any form of pre-treatment, there was an increase in transcript levels for genes encoding potential chemotaxis proteins which switched to a decrease in transcript levels when the cultures were pretreated with either S or N. With both the S and N pretreated cultures, an increase in transcript levels was identified for genes encoding potential heat shock proteins. For proteins related to metabolism, there were specific results seen based on the pre-treatment used. With a S pre-treatment there was a decrease in transcript levels for genes encoding cystathionine gamma lyase as well as cystathionine beta synthase. When cultures were pretreated with N, there was a decrease in transcript levels for genes encoding threonine dehydratase and an increase in transcript levels for genes encoding 2-isopropylmalate synthase. In N pretreated cultures, there were also transcript level changes for genes encoding proteins related to metal binding, including calmodulin and serine/threonine kinase which showed increased transcript levels and phospholipid transporting ATPase and L-ascorbate peroxidase which showed decreased transcript levels.
Analysis using the PANTHER classification system revealed enrichment of GO-terms which were consistent with the DGE analysis results. While a complete table of the results of the GO-term analysis are included in the supplementary material, the following results will focus on a few GO-terms of note (Table S53). This included cysteine metabolic process, ABC-type transporter activity, iron-S cluster binding and metal ion binding (Table 2). These GO-terms were enriched in pretreated E. gracilis cultures that were exposed to CdCl2, indicating their potential importance in the overall cadmium tolerance of E. gracilis.
RT-qPCR
RT-qPCR was completed on select transcripts that were identified as differentially expressed during the DGE analysis. Transcripts were selected based on relevant biological functions of predicted proteins or degree of transcript level change. The results indicate that in some instances transcripts, showed similar trends in level changes to the results obtained with RNA-sequencing. These transcripts were, TRINITY_109159, hsp70, and threonine dehydratase. RNA-sequencing results showed that the transcript for TRINITY_109159 was increased in S pretreated cultures that were exposed to CdCl2 but was decreased in S pretreated cultures without CdCl2. We were unable to find sequence similarity to TRINITY_109159 within any of the databases we screened. The RT-qPCR results also supported the detection of increased transcript levels in S pretreated cultures that were exposed to CdCl2 relative to non-pretreated cultures, and S pretreated cultures without CdCl2. Notably RNA sequencing indicated an increase in the transcript for the hsp70 gene in S pretreated cultures that were exposed to CdCl2 when compared to S pretreated cultures that were not exposed to CdCl2. The RT-qPCR results further showed an increase in the level of these transcripts in S pretreated cultures exposed to CdCl2, although the level of change was not at the same level, as seen in the DGE analysis. Finally, our DGE analysis identified a decrease in the level of transcripts for a gene encoding threonine dehydratase in N pretreated cultures that were exposed to CdCl2. The results of the RT-qPCR also showed a trend of decreased transcript levels in N pretreated cultures exposed to CdCl2. These RT-qPCR results were consistent with the DGE results however, not all met the same log2-fold cutoffs used in the RNA-Seq analysis.