Transcriptomic and proteomic profiling of peptidase expression in Fasciola hepatica eggs developing at host’s body temperature

From the bovine liver, we isolated 97 live F. hepatica adults. After overnight cultivation, we recovered approx. 228,000 laid eggs, which we divided in three groups. The first group (T0) was immediately frozen at − 80 °C, while the other two groups (T5 and T10) were incubated for 5 and 10 days at 37 °C, respectively. Before incubation and/or freezing, the eggs from each time point were subdivided according to the type of subsequent analysis as follows: 21,000 eggs for RNA sequencing (RNA-seq), 1800 eggs for mass spectrometry analysis, and 50,000 eggs for a biochemical analysis of proteolytic activity. Material for RNA-seq and LC–MS/MS analyses from each time point was divided in three replicates.

Egg viability assay

The effect of incubation at 37 °C on egg development and viability was evaluated using a light-stimulated egg hatch test. Freshly laid eggs were cultivated for 5 and 10 days at 37 °C (T5 and T10 groups, respectively), then the temperature was lowered, and egg developed at 25 °C. Eggs maintained at 25 °C served as a control group. The hatching tests were performed on day 14, 16, 19, 23, 26, 30 and 35 post laying. Eggs were checked regularly until a release of new miracidia was not recorded. In addition to the hatching tests, eggs were regularly inspected with light microscope (Fig. 1c) for visible signs of embryonation characterised by the cell division, which were observed on day 5 in all groups. However, hatching of eggs from groups incubated at 37 °C was substantially delayed comparing to the control group maintained at 25 °C throughout the whole experiment. Approximately 78% of eggs from control group embryonating all time at 25 °C successfully released miracidia at days 14, 16 and 19 (Fig. 1a). Releasing miracidia in group T5 begun on day 16 (7% of eggs hatched) and ended on day 26 with 36% hatched eggs in total. Even more pronounced retardation of development was observed in eggs from group T10, where the hatching started on day 26 (2% of eggs hatched) and ended on day 35 with only 15% of eggs hatched. The egg death rate increased with the length of incubation at 37 °C reaching up to 14, 18 and 42% of total amount of eggs in control, T5 and T10 group, respectively (Fig. 1b). Nevertheless, signs of partial or complete embryonation (eye spot stage, cell division stage, hatched egg) was reported in 82 and 58% of eggs in T5 and T10 groups, respectively.

Figure 1
figure 1

Egg viability assay. (a) Hatch rate dynamics during the egg development. Numbers of hatched eggs in groups the control, T5 and T10 group are presented as percentage of total number of eggs (mean and standard deviation are given). (b) Proportion of four egg development stages. Proportion of eggs from each stage on the day when last newly hatched eggs were observed in the particular group, that is on day 19, 26 and 35 in control, T5 and T10 groups, respectively. (c) Light micrographs illustrating the different stages of egg development. Freshly laid egg, embryonating egg at cell division stage, eye spot stage, hatched egg with an open operculum after the release of miracidium, dead egg (from left to right).

Sequencing, annotation, and peptidases identification

Single-end sequencing generated about 23 million short reads for each sample (Supplementary file 2). A total of 9708 F. hepatica proteins derived from the reference genome PRJEB2528336 were reannotated using UniProt/UniRef100 and MEROPS protein database; 8159 homologues were found in the UniProt database (Supplementary file 3). Moreover, the GO term analysis revealed 1879 unique GO terms across all F. hepatica reference proteins, whereby 73 GO terms were related to proteolytic activity (Supplementary file 4) and we used them to identify a total of 193 peptidases.

Transcripts and proteins present at different time points of egg development

Protein-coding gene transcripts and translated proteins were considered transcribed/translated in a particular time-point group when the median of relative quantification values in TPM (transcripts per million) given by RNA-seq and PPM (parts per million) given by LC–MS/MS of all three replicates was non-zero (for all data, see Supplementary file 5).

Among the 9708 reference proteins, 7475 (77.0%) protein-coding transcripts of F. hepatica eggs were quantified by RNA-seq in at least one time point. Regarding a comparison between the time-point groups, the numbers of transcribed sequences identified in T0, T5, and T10 samples were 7005, 6705, and 6740, respectively (Fig. 2a). Transcription of a vast majority of identified transcripts (6189, that is approx. 83% of all identified transcripts) was detected in eggs throughout the monitored egg development, i.e., in all three time-point groups. The proportion of 6189 shared transcripts from the total transcripts quantified at time points T0, T5, T10 was 88%, 92%, and 91%, respectively.

Figure 2
figure 2

UpSet plots depicting the number of transcripts and proteins quantified in F. hepatica eggs at different points of development. The numbers represent a distribution of protein-coding gene transcripts (a) and translated proteins (b) at three different points of egg development (T0, T5, and T10). Solitary dots represent transcripts or proteins unique to each time point; two or three dots connected by a line represent the number of transcripts or proteins shared between two or three different time points, respectively. The numbers of quantified peptidases are given in parentheses and highlighted in orange.

About 200 transcripts were shared only between pairs of groups (201 transcripts in T0 vs T5, 183 transcripts in T0 vs T10 and 213 transcripts in T5 vs T10 comparisons). Interestingly, 432 protein-coding genes were transcribed only in T0, that is, during early egg formation. Out of all 7475 transcribed genes, we determined 145 transcripts coding proteins with proteolytic activity (based on GO term analysis): this accounts for approx. 1.9% of all expressed transcripts. Majority of identified peptidase transcripts (117) were shared by eggs at all three time points (T0, T5, T10), similarly as for all detected protein-coding gene transcripts.

LC–MS/MS analyses revealed 907 translated proteins in F. hepatica eggs (9.3% of the 9708 reference proteins). As in the case of transcripts, the majority of produced proteins (533, approx. 59% of all identified proteins) was shared across the time points (Fig. 2b), but the proportion of shared proteins from all proteins detected at each point was 76% (T0), 60% (T5) and 98% (T10), which indicates that the proteome is more variable than the transcriptome. The data show that the highest level of unique protein translation tends to occur up to the fifth day of egg development. The number of proteins identified in the T10 group was considerably lower than in samples from the earlier time points. The most abundant proteins in all studied samples were antioxidant and detoxification enzymes (thioredoxin peroxidase, glutathione S-transferase), an iron-storage protein (ferritin), structural proteins (actin, tubulin, myoglobin), proteins important for energetic metabolism (glyceraldehyde-3-phosphate dehydrogenase) and heat-shock proteins (alpha-crystallin domain containing heat-shock protein). Approximately 3.9% (35 in total) of all quantified proteins were characterised as peptidases based on GO term analysis.

Peptidase-coding transcripts and translated peptidases evaluated as the most abundant (based on the TPM/PPM values) in T0, T5, and T10 samples are listed in Table 1. Relative abundances of all proteins with assigned proteolytic activity identified by LC–MS/MS (n = 27) are shown in Fig. 3. The dynamics of peptidase class composition as evidenced by RNA-seq and MS data is not tightly correlated (Fig. 4). A large proportion of peptidase transcripts is comprised of aspartic peptidases, followed by cysteine and threonine peptidases. Finally, the lower expression of metallopeptidases is comparable with that of serine peptidases. On the protein level, threonine peptidases, which form a majority of translated peptidases, are followed by cysteine and aspartic peptidases, which represent about one quarter of the whole peptidase spectrum. The proportion of metallopeptidases and serine peptidases was, according to recorded PPM values, the lowest.

Table 1 Top ten most abundant peptidases identified in F. hepatica eggs at different points of development.
Figure 3
figure 3

Relative abundances of peptidases identified by LC–MS/MS. Relative abundance values (PPM) are shown on the logarithmic scale. Different isoforms of signalase, proteasome subunit α, proteasome subunit β, and proteasome endopeptidase complex are presented as single columns. Peptidases marked by an asterisk (n = 10) were significantly differentially expressed. Cathepsin L isoforms were named according to Cwiklinski et al.50 nomenclature.

Figure 4
figure 4

Representation of individual catalytic classes among the transcribed/translated peptidases. Percentages were counted based on relative abundance values in TPM/PPM.

A comparison of transcriptome and proteome changes during egg development

Statistically significant changes in the gene expression were evaluated by comparing expression levels between pairs of samples of F. hepatica eggs; in the following, we refer to them as log2 fold change (log2FC). Data from an earlier development stage (T0 or T5) served as reference for analysing data from eggs belonging to a later stage of development (T5 or T10, respectively). All statistically significant changes in the expression (log2FC) are listed in Supplementary file 5. A comparative analysis of transcriptomic data resulted in the identification of 3577 unique transcripts significantly changed in at least one compared sample pair. Using > 1 and <  − 1 as thresholds of log2FC, the numbers of transcripts detected as significantly upregulated were 590, 927, and 107, whereas the numbers significantly downregulated were 287, 1112, and 85, when comparing the T0 vs. T5, T0 vs. T10, and T5 vs. T10 group, respectively. The number of proteins with significantly changed levels was notably lower: all in all, in the corresponding stage comparisons we detected 35, 0, and 44 upregulated and 40, 18 and 147 downregulated proteins. This suggests that the lowest number of changed transcripts—and conversely the highest number of changed proteins—was observed in a comparison of T5 vs. T10 time points.

Figure 5 shows these molecular changes in F. hepatica eggs in a graphical form. Box plots (Fig. 5a,b) provide a summarised overview of the dynamics of the transcriptome and proteome at different points of development, where the height of the boxes represents an overall variation in RNA or protein levels. Figure 5a shows a low interquartile range (log2FC < 2.4), which indicates that 50% of significantly differentially expressed transcripts show minor changes in expression throughout the process of egg maturation. Only a few transcripts show major changes. Compared to the data from the other time points, T5–T10 show slightly less pronounced changes in transcript levels.

Figure 5
figure 5

Statistically significant changes in the transcriptome and proteome in F. hepatica eggs at different points of development. The box plots provide an overview of transcriptome (a) and proteome (b) variability in different sample pairs. Transcripts and proteins with the most prominent changes in gene expression levels are located the furthest from zero on the y-axis. The heat maps show a global overview of changes in RNA levels of differentially expressed genes (c) and differentially translated proteins (d) at single gene/protein resolution. The gene/protein ranking is based on relative abundance detected at T0, with the most abundant transcripts appearing at the top of the heat map (red) and those with the lowest abundance appearing at the bottom (green). Differences in the expression/translation are given as Z-score values. Genes which were not expressed at a particular time point are presented in grey.

The same analysis was conducted for the proteome (Fig. 5b). As with the transcriptome, analysis of the proteomic data revealed that the number of changes in protein expression slightly decreases as time progresses, which is most noticeable when comparing changes between T0 and T10. Interestingly, the dispersion of upregulated and downregulated genes was much more symmetric in the transcriptomic data, while the proteome showed a clear tendency towards a higher number of downregulated proteins. For a full description (numerical representation of the box plots), see Supplementary file 6.

Heat maps (Fig. 5c,d) show an overall profile of these changes at individual protein-coding gene/protein resolution. Most striking changes in gene expression are found in a comparison between the T0 and T5 group. The variability of proteome (in terms of translated proteins) increased with age up to time point T5, while in 10-days-old eggs we did not detect a large part of proteins translated at earlier time points.

Quantification of differentially expanded cysteine peptidases

We performed a thorough exploration of cysteine cathepsin isoforms identified in the egg transcriptome/proteome. This was preceded by a manual curation of F. hepatica genome assembly (PRJEB25283) using previously described sequences12,13,50. In total, we identified 17 cysteine cathepsin isoforms, 16 of which were detected in the RNA-seq data. Two cathepsin L isoforms (CL0 and CL1_3a) and one cathepsin B isoform (CB9) belonged to the most expressed peptidase transcripts (Table 1). RNA-seq identified ten transcribed cathepsin L isoforms, five cathepsin B isoforms, and one cathepsin F isoform across all three developmental stages of the eggs (Fig. 6). A translation of five cathepsin L members was detected by LC–MS/MS (named CL0, CL2, CL1_3a, CL1_3b, and CL1_2; see Fig. 3), whereby the first four belonged to the ten most abundantly expressed peptidases (Table 1). Isoform CL1_3b was quantified only in LC–MS/MS data and it was the only differentially expressed cathepsin downregulated at T5 time point (Fig. 8).

Figure 6
figure 6

Relative expression of all transcripts coding cysteine cathepsins. Relative abundances (in TPM) are shown on a logarithmic scale. Isoforms marked by an asterisk were differentially expressed. Isoform CL1_3b was quantified only on protein level by LC–MS/MS. We used the Cwiklinski et al.50 nomenclature. Alternative cathepsins’ names used in McNulty et al.13 phylograms are shown in parentheses.

Differential expression analyses focused on peptidases

Differential expression and translation of peptidases throughout the egg development was evaluated using the same approach as in the analyses of transcriptome and proteome (see above). The results are depicted in heatmaps (Figs. 7 and 8). Most cysteine peptidases were expressed in the freshly laid eggs (T0) (Fig. 7) but several cysteine peptidases (n = 12) were upregulated at later developmental stages, e.g., calpain B, separin, and ataxin (at stage T5) and cathepsin F (at T10). Nine cathepsin genes were found differentially expressed according to the RNA-seq data. Expression of cathepsins L isoforms decreases as the egg matures. After the initial upregulation of aspartic peptidase cathepsin D (one of the most expressed peptidases), its expression gradually decreased (from 3850 TPM at T0 to 2400 TPM at T10; Table 1). All isoforms of proteasome subunits formed by threonine peptidases were overexpressed at the T10 time point. Serine peptidases did not show a constant trend in terms of differential expression. For instance, furin, prolyl oligopeptidase, and dipeptidyl peptidase were expressed mainly directly after laying, while serine carboxypeptidase and lon protease homolog were upregulated in more mature eggs. Finally, metallopeptidases were overexpressed mainly in the T10 group, except for leukotriene hydrolase and zinc carboxypeptidase, which were upregulated at the time point T0. As for mass spectrometry data, only ten peptidases (or their isoforms) were found differentially translated (Fig. 8). Cathepsin L1_3a, upregulated in eggs soon after laying, was downregulated at T5, and then strongly expressed at T10 (for relative abundances, see Fig. 6). An overexpression of proteasome subunits was observed also in older eggs. Curiously, most of the differentially translated metallopeptidases were not detected in more mature eggs, but the trend of elevated production of metallopeptidases could be observed in 5-days-old eggs.

Figure 7
figure 7

Expression dynamics of genes coding peptidases in F. hepatica eggs at different points of development. The heat maps show statistically significant changes in RNA levels of differentially expressed genes at a single gene resolution. Genes which are not expressed at a particular time point are coloured in grey. Differences in expression levels are shown as Z-score values. Gene names shown at x-axis begin with the peptidase family code; catalytic classes of peptidases are shown above the heatmap.

Figure 8
figure 8

Statistically significant changes in the abundance of proteins with proteolytic activity based on LC–MS/MS quantification. The heat map shows statistically significant changes in protein abundance at a single gene resolution. Genes which are not expressed at a particular time point are coloured in grey. Differences in protein expression are shown as Z-score values. Gene names shown at x-axis begin with the peptidase family code; catalytic classes are shown above the heatmap (C—cysteine peptidases, M—metallopeptidases, S—serine peptidases, T—threonine peptidases).

Biochemical characterisation of proteolytic activities, pH optima, and inhibition assays

Proteolytic activities in protein extracts from F. hepatica eggs were investigated also using a biochemical approach. Enzymatic activities were measured using a continuous kinetic assay with a complex library of internally quenched FRET substrates in buffers of a wide pH range (2–10). Proteolytic activities of extracts derived from eggs at different times after laying have different pH profiles (Fig. 9). The highest proteolytic activity of T0 eggs was detected in acidic conditions with a sharp optimum at pH 3.5 and two local optima at pH 5.5 and 7.5. At T5, on the other hand, the greatest activity was observed within pH range 8.0–9.0, and for T10 in alkaline pH around 8.0. Local optima were identified at pH 3.5 and 6.0 for T5 egg extract and around pH 4.5 and 5.5 for T10 egg extract. These findings indicate changes in the composition of peptidases during egg maturation.

Figure 9
figure 9

The pH profiles of proteolytic activity in extracts from F. hepatica eggs. Activities were measured using a library of FRET substrates in extracts from eggs collected 0, 5, and 10 days after laying (T0, T5, and T10, respectively). Data were normalised against the highest proteolytic activity recorded for each protein sample. Mean values are listed with the standard deviation.

In order to identify which peptidase classes contribute to the global proteolytic activity in F. hepatica eggs, the protein extracts were treated with five class-specific peptidase inhibitors which selectively target metallopeptidases (EDTA)54, cysteine (l-trans-epoxysuccinyl-leucylamido(4-guanidino)butane, E-64)55, serine (4-(2-aminoethyl)benzenesulfonyl fluoride, AEBSF)56, aspartic peptidases (pepstatin)57, and threonine catalytic subunits of proteasome (benzyloxycarbonyl-l-leucyl-l-leucyl-leucinal, MG132)58. Reactions were performed at three different pH levels (3.5, 6, 8.5), which were selected according to the previously established pH optima of proteolytic activities of the egg extracts. Data are visualised in Fig. 10.

Figure 10
figure 10

Inhibition of proteolytic activity in F. hepatica egg extracts using class-specific inhibitors. Proteolytic activity of protein extracts from F. hepatica eggs at 0, 5 and 10 days post laying (T0, T5 and T10, respectively) was assayed in the presence of class-specific inhibitors (EDTA, E-64, AEBSF, pepstatin, MG132) and a combination of all five inhibitors (MIX). Inhibitor of threonine peptidases (MG132) was used only at pH 8.5 due to its non-specific activity against cysteine peptidases in acidic environments. Inhibition of activity is expressed as the percentage of remaining proteolytic activity relative to a non-inhibited control (100%). Mean values are listed with standard deviation.

Cysteine peptidase inhibitor E-64 suppressed proteolysis at T0 in acidic pH 3.5 by 80%. A treatment of T5 and T10 samples with E-64 reduced the substrate turnover at this pH by 48% and 15%, respectively. It seems, therefore, that during the process of egg maturation the activity of cysteine peptidases gradually decreases. On the other hand, pepstatin, which inhibits aspartic peptidases, reduced the cleavage of substrate by 13%, 41%, and 44% at T0, T5, and T10, respectively, which indicates that egg development is accompanied by a growing presence of aspartic peptidases.

At mildly acidic pH 6, treatment with E-64 resulted in a 57% decrease of proteolytic activity in the T0 extract, which suggests a dominant role of cysteine peptidases. Inhibition by aspartic peptidase inhibitor pepstatin and serine peptidase inhibitor AEBSF caused only a slight decline in enzymatic activity (by 7% and 13%, respectively). Treatment of T5 extract with each of the five peptidase class inhibitors used in our study suppressed substrate cleavage by approx. 15–20%, while a combination of inhibitors reduced activity by approx. 65%, suggesting a complex contribution of different peptidase classes at T5. Analogous situation was observed in the T10 sample, where combined inhibitors of all five catalytic classes resulted in a 60% decrease in proteolytic activity.

Activity of metallopeptidases in extracts from eggs collected at all three time points was the strongest in alkaline environment (pH 8.5). EDTA provided increased inhibition by 25%, 54%, and 63% at T0, T5, and T10, respectively, which indicates a growing role of metallopeptidases during the process of egg development (similar to what was observed for aspartic peptidases at pH 3.5). A lower level of inhibition of activity by serine peptidase inhibitor AEBSF was observed in alkaline conditions in the T0 sample and in the T5 sample in neutral environment (by 20% and 16%, respectively), suggesting a role of serine peptidases. Moreover, alkaline pH creates optimal conditions for the activity of proteasome, whose subunits were identified among the most abundant peptidase-coding transcripts and translated peptidases (Table 1). Inhibition of activity at pH 8.5 with threonine peptidase inhibitor MG132 targeting proteasome resulted in a decreased activity by 25%, 40%, and 50% at T0, T5, and T10, respectively, which indicates that this group of peptidases is prevalent during the process of egg maturation. Incomplete inhibition in pH 6 and 8.5 indicates the presence of peptidases insensitive to the general class-specific inhibitors used in this study.

To conclude, our results showed that cysteine peptidases, which are active in acidic environments and prevalent especially in freshly laid eggs, are later partly replaced by aspartic peptidases. Egg maturation is accompanied by the initiation of metallopeptidase activity, which achieves its optimum in neutral and mildly alkaline environments. Threonine catalytic subunits of proteasome are active throughout the eggs’ development.

Read more here: Source link