Introduction

Cestode parasites of the genus Dibothriocephalus (formerly Diphyllobothrium) are widespread throughout the globe. Some cause diphyllobothriasis in humans, which is transmitted through fish (Delyamure et al. 1985; Dupouy-Camet and Peduzzi 2004; Dupouy-Camet and Yera 2010; Kuchta et al. 2013). The annual number of diphyllobothriasis diseases in the world is 20 million (Scholz et al. 2009; Dupoy-Camet and Year; 2010; Tsukamoto et al. 2019). Dibothriocephalus dendriticus is a parasitic tapeworm that is the main cause of diphyllobothriasis in some northern areas of Siberia and in the Baikal region (Pronin et al. 2012). During the last 50 years, D. dendriticus has been studied as a model organism with respect to ecology, life cycle, morphology, and cell biology (Gustafsson 1990). The life cycle of D. dendriticus includes three developmental stages. The first larval stage or procercoid develops in copepods that are the first intermediate hosts. The second larval or plerocercoid stage develops in the intermediate host, which includes fish of the suborder Salmonoidei and nonsalmonid fish (Rusinek 2008). The adult stage infects the intestine of the final host—mammals, including humans (Scholz et al. 2009). In the Lake Baikal basin, diphyllobothriasis caused by D. dendriticus is prominent among other parasitosis (Chizhova and Hoffman-Kadoshnikov 1960). The main source of human infection with diphyllobotriasis in the Lake Baikal is the most numerous fish species—Baikal omul Coregonus migratorius (Salmoniformes). The Republic of Buryatia is one of the regions with a high incidence of diphyllobotriasis—up to 650 cases per year, while in other regions of Russia the situation is even more serious: annually more than 20,000 people fall ill with diphyllobotriasis caused by various diphyllobothriids (Lysenko et al. 2002). Such a picture of the spread of this disease is caused by the lack of effective preventive measures, the lack of vaccines, and the use of antihelminthics with low treatment efficacy. Therefore, research is currently underway to find new ways to interrupt the spread of diphyllobothriasis and increase the effectiveness of treatment (Alroy et al. 2020; Lopez et al. 2021). One way to interrupt the spread of diphyllobothriasis is to break the developmental cycle of the parasitic worms of this species.

Reliable estimates of human cases to assess the true impact of broad tapeworms on human and wildlife health are still lacking (Scholz et al. 2019). Genomic and transcriptomic data can be used to develop future diagnostic tools and epidemiological studies, including screening of environmental DNA for the life cycle stages of the broad tapeworm in the environment (Bass et al. 2015).

Broad tapeworms are of great interest as subjects for high-throughput sequencing studies. Due to the fact that this family includes human parasites (as well as a long-accepted laboratory model), three representatives of the broad tapeworms have been included in the genome sequencing project conducted by the 50 Helminth Genomes Initiative (2014). In addition, life cycle transcriptomes of helminth parasites such as Taenia pisiformis (Yang et al. 2012), Angiostrongylus cantonensis (Wang et al. 2013), Fasciola gigantica (Zhang et al. 2017), and Schistosoma mansoni (Wangwiwatsin et al., 2020) have now been obtained, indicating a high level of interest in these organisms.

In this study, we sequenced the transcriptomes of the plerocercoid and adult gull tapeworm D. dendriticus. The data obtained allowed us to identify differentially expressed genes (DEGs) in the life cycle development of D. dendriticus to compare and determine the differences in their biological processes. This will contribute to further research aimed at identifying targets for new generation drugs and the development of specific vaccines.

Materials and methods

Samples of plerocercoid and adult D. dendriticus

Plerocercoids of D. dendriticus were obtained from the body cavity of the Baikal omul, Coregonus migratorius. Adult tapeworms of D. dendriticus were extracted from the intestines of the herring gull Larus argentatus. All samples were maintained in liquid nitrogen (− 196 °C).

RNA extraction

Total RNA from 0.5 to 1.0 g of tissue was isolated with TRIzol reagent (Ambion) and purified with simultaneous treatment with DNAase I on PureLink RNA Mini columns (Invitrogen). RNA quality was determined on a BA2100 Bioanalyzer using an RNA Nano Kit.

Construction of RNA-seq library and sequencing

The TruSeq Stranded mRNA Library Preparation Kit (Illumina) was used to prepare directional barcoded transcriptome libraries with double-indexed UD according to the manufacturer’s protocol, with modifications made to obtain longer embeddings (200–500 bp). One microgram of total RNA was collected, and the mRNA fragmentation time was 4 min. After amplification of the libraries, additional length selection was performed on AMPureXP magnetic particles—0.65 vol of AMPureXP was added to the diluted library. The quality and molarity of the obtained libraries were determined on a BA2100 bioanalyzer; the libraries were mixed in equimolar amounts to a total concentration of 2 nM. The obtained libraries were sequenced on an Illumina NextSeq550 High Throughput Sequencer using a NextSeq 550 High Output v2 Kit (300 cycles) in 150 bp paired-end reads.

De novo assembly

The resulting fastq reads were preprocessed, removing adapter sequences with Scythe v0.9944 BETA (2016) and low-quality reads with Sickle v1.210 (Joshi and Fass, 2011) (Phred > 30 quality score). The preprocessed reads were subjected to a de novo assembly procedure using Trinity v.2.8.5. (Grabherr et al. 2011) with default parameters. The build was performed on a 36-core server with 256 GB RAM.

Next, we quantitatively assessed the completeness of the obtained data by comparing our set of transcripts to a single copy set of highly conserved orthologs. To do this, we used the BUSCO (Benchmarking Universal Single-Copy Orthologs) v2 pipeline (Simão et al., 2015) and compared it to a predefined set of 4584 eukaryotic single-copy orthologs from the OrthoDB v9.1 database (Zdobnov et al., 2016). We counted the number of complete (length within two standard deviations of the average length of a given BUSCO), duplicate (complete BUSCOs represented by more than one transcript), fragmented (partially recovered BUSCOs), and missing (unrecovered) BUSCOs.

Further analysis of the transcripts was performed using a joint de novo assembly of the plerocercoid transcriptome and the adult gull tapeworm D. dendriticus transcriptome, also generated using the Trinity v.2.8.5. software package with default parameters. To obtain sets of non-redundant transcripts, we applied the following filtering steps: First, we used TransDecoder to identify all likely coding regions in the collected transcripts, and then filtered by selecting the best open reading frame (ORF) per transcript according to the TransDecoder pipeline (–single_best_orf). All transcripts with ORFs less than 200 bp in length were removed before further analysis. In addition, redundancy in the remaining transcript sets was further reduced by clustering similar sequences with the program CD-Hit (Li and Godzik, 2006) using an amino acid sequence identity threshold of 1.00. Unigene function was annotated using the BLASTp tool (Altschul et al. 1990) in the following databases: UniProtKB/SwissProt (http://www.uniprot.org) (2017), KEGG (Kyoto Encyclopedia of Genes and Genomes, http://www.genome.jp/kegg) (2022), and eggNOG (Orthologous groups of genes, http://eggnogdb.embl.de) (Huerta-Cepas et al. 2017) with the e-value parameter < 10−5.

Differential gene expression analysis

Differential expression analysis of adult and plerocercoid D. dendriticus transcripts was performed using the R package DESeq2 (Love et al. 2014), Bowtie2 v2.3.5.1 (Langmead and Salzberg 2012), and Trinity v2.8.5 software components. DESeq2 provides statistical methods for determining differential expression in digital gene expression data using a model based on a negative binomial distribution (Anders and Huber 2010). Filtering of differentially expressed (DE) transcripts was performed using a predetermined threshold of P < 0.001. Some differentially expressed genes (DEGs) were annotated using the UniProtKB/SwissProt database, using the tool BLAST and the parameter e-value 1e − 3.

Gene Ontology and KEGG enrichment analysis of differentially expressed genes

EggNOG mapper and Panther (Mi et al. 2013) were used to predict and classify the functions of DEGs. Enrichment analysis of differentially expressed genes using Genetic Ontology (GO) was performed using the GOseq R package (Young et al. 2010), in which gene length error was corrected. GO terms with a corrected P-value of less than 0.05 were considered as significantly enriched differentially expressed genes. KEGG is a database resource for understanding the higher-level functions and utility of a biological system, such as a cell, an organism, and an ecosystem, based on molecular-level information, especially large molecular datasets generated by genome sequencing and other high-level data. We used the KOBAS software (Xie et al. 2011) to test the statistical enrichment of genes with differential expression in the KEGG pathways.

Results and discussion

Transcriptome sequencing and assembly of transcripts

The present study is the first attempt at a comparative analysis of the adult and plerocercoid transcriptomes of gull tapeworm D. dendriticus. This study provides the basis for understanding the analysis of differential gene expression of D. dendriticus depending on its developmental stage. In order to comprehensively evaluate the transcriptomes studied, a procedure was carried out to identify fish representatives of C. autumnalis.

Transcriptome assembly can be performed using either a reference genome or a de novo method. However, the genetic proximity of the reference genome is important for its use. In our study, the best choice was the approach of assembling the transcriptomes of the adult and plerocercoid stages of D. dendriticus using the de novo method. Six transcriptome libraries were prepared for sequencing—three from the pleocercoids and three from the adults. After filtering the reads from the raw data, 8,648,999,940 reads from adult D. dendriticus and 76,564,854 reads from D. dendriticus plerocercoid tissues were obtained. These reads were collected using the Trinity program with N50 1841 bp for the adult and N50 1090 bp for D. dendriticus plerocercoids. The average length of the assembled contigs was approximately 1148 and 1184 bp, respectively. The distribution of the length of all contigs showed that the most numerous contigs were grouped in 25,470-bp- and 19,960-bp-long groups for the adult and D. dendriticus plerocercoid, respectively.

The resulting assemblies contained 108,334 and 130,762 transcripts for the adult and plerocercoid D. dendriticus, respectively. To assess and compare the gene set completeness of our transcriptomes, we used the BUSCO assay, which showed that most of the eukaryote genes were successfully recovered in our two assemblies. Specifically, of the 4584 orthologs, 95.4% were fully recovered and 1.7 to 2% were partially recovered (Table 1). It should be noted that 2.6 to 2.9% of the 4584 orthologs were classified as missing, indicating the high quality of the protein-coding transcripts in our assemblies. The obtained results indicate a good quality of sequencing and transcriptome assembly, which allowed us to proceed to the next stage of whole de novo assembly of plerocercoid and adult D. dendriticus transcriptomes. A total of 175,621 transcripts were identified. To eliminate redundant transcripts, we used the TransDecoder tool to predict ORF. After obtaining all predicted ORFs of our total group, they were filtered to select the best ORF for each transcript. This resulted in 52,633 D. dendriticus transcripts. Next, we grouped the filtered sequences using the CD-hit program with 100% identity to remove repetitive sequences. The final number of unigenes in the resulting D. dendriticus assembly was thus 35,129. To obtain more detailed information about these unigenes, we searched for sequence similarities in three databases using the BLASTp tool: Swiss-Prot, KEGG, and eggNOG. As shown in Table S1, the maximum match was found in eggNOG, for which 14,070 unigenes (40%) were annotated, while for Swiss-Prot, 16,432 unigenes (47%) were annotated, and for KEGG, 10,258 unigenes (29%) were annotated (Table 2).

Table 1 Evaluation of the completeness of the set of eukaryotic gene transcripts of an adult and plerocercoid D. dendriticus
Full size table
Table 2 Summary of the statistics for the annotation of unigenes of D. dendriticus
Full size table

Identification and analysis of DEGs

To evaluate the expression level of adult and plerocercoid D. dendriticus transcripts, we created a single transcriptome assembly from reads of both samples. The resulting D. dendriticus assembly was subjected to analysis to identify differentially expressed transcripts using the DESeq2 method, which allowed us to identify the transcripts we were looking for. We considered a gene model as differentially expressed if the difference between the normalized means at the 1e − 3 confidence level was greater or less than zero. The obtained results were ordered by their log2 fold change and presented in Table S1. A total of 8022 DE transcripts were identified, including 3225 upregulated and 4797 downregulated DE transcripts from the plerocercoid and adult animals. A heat map was constructed to visualize the quantitative differences in DEG expression levels between plerocercoid and adult D. dendriticus (Fig. 1). The heat map was generated based on FPKM (expected number of fragments per kilobase of transcript sequence per million sequenced base pairs) RNA-seq data and represents the expression levels of the transcripts. The yellow area indicates high levels of expression, and the purple area indicates low levels of expression.

Fig. 1
figure 1

Heat map showing normalized gene transcript abundance in adults and plerocercoids of D. dendriticus. Differentially regulated genes (listed to right) were identified using a predetermined threshold of P < 0.001 between adults (CH) and plerocercoids (L) of D. dendriticus. Legend above the heat-map shows stage of parasite development (adults, CH; plerocercoids, L)

Full size image

Among the identified DEGs, TOP50 transcripts were identified and annotated by UniProtKB/SwissProt (Table 3). Among them, 13 adult transcripts and 37 D. dendriticus plerocercoids transcripts were identified. It should be noted that 74% of the TOP50 DE were adult transcripts, while only 26% belonged to plerocercoids. The predominance of the adult DE level is most likely related to the reproductive growth of the individual. Prediction of these TOP50 DEs also revealed that only eight and five DEGs were annotated for the plerocercoid and adult, respectively, while most of the predicted DEGs from the TOP50 are novel proteins that were not previously annotated. This could be due to specific growth and development factors of D. dendriticus. Among the annotated TOP50 DEGs, three can be distinguished: fatty acid–binding protein and ferritin, whose DEGs are related to plerocercoid; Kunitz-type serine protease inhibitor, whose DEG is related to adult D. dendriticus. To perform a functional annotation of the identified DEGs, we performed a full annotation using the egg-NOG-mapper (Tables S2 and S3) which allowed us to define Gene Ontology (GO) terms and compare differences in term annotation between plerocercoid and adult. A total of 1542 unique terms were available; 3025 models had at least one annotation, of which 2054 models annotated biological processes. We found little variation in terms for biological processes (Fig. 2). Thus, biological regulation terms account for 24.2 and 23.2% for the plerocercoid and adult, developmental process terms account for 5.3 and 4.7%, immune system process terms account for 0.8 and 0.2%, localization terms account for 17.3 and 14.2%, locomotion terms are 2.4 and 1.2%, metabolic process terms are 35.1 and 42.2%, multicellular organismal process terms are 4.4 and 4%, response to stimulus terms are 11.5 and 10%, and signaling terms are 7.8 and 5%, respectively. The greatest difference between the metabolic process terms can be justified by the difference in the life cycle of the samples under study, which directly affects various developmental factors.

Table 3 TOP50 transcripts of differentially expressed genes of adult and plerocercoids of D. dendriticus, identified and annotated by UniProtKB/SwissProt
Full size table
Fig. 2
figure 2

Gene Ontology (GO) analysis of gene transcripts of plerocercoids (A) and adults (B) of D. dendriticus in terms for biological processes

Full size image

When compared to the terms of molecular functions and cellular components, the annotation was very consistent between the samples studied, with metabolic processes being the most represented in all cases (Fig. 3). We hypothesize that the overall pattern of GO annotation of the GO terms of the studied specimens of plerocercoid and adult D. dendriticus (Fig. 3) mainly reflects the “home state” of the specimens and is therefore very similar, which is also consistent with the previously presented studies.

Fig. 3
figure 3

Gene Ontology (GO) analysis of gene transcripts plerocercoids (A, C) and adults (B, D) of D. dendriticus in terms molecular functions (A, B) and cellular components (C, D)

Full size image

The enriched KEGG pathways of DEGs

KEGG enrichment results showed that DEG was specifically enriched in 207 KEGG pathways (Fig. 4, Table S4). Most of these pathways are related to biosynthesis and metabolism, such as folate biosynthesis, arginine biosynthesis, galactose metabolism, selenium compound metabolism, and phosphonate and phosphinate metabolism. The 20 most enriched KEGG metabolic pathways are listed in Table 4. Metabolic pathways (445 DEGs), tight junction (83 DEGs), and phagosome (70 DEGs) are the most represented. In addition, a parasite-specific mRNA surveillance pathway (51 DEGs) was found among the 20 most enriched KEGG pathways.

Fig. 4
figure 4

Histogram of differentially expressed transcripts of larval and adult stages of Dibothriocephalus dendriticus compared with the KEGG database. The column color (C1, C2…) represents the network of protein modules identified by i-KOBAS. For each individual module, if more than 5 terms are detected, the 5 with the highest enrichment factor are displayed

Full size image
Table 4 The 20 most enriched KEGG metabolic pathways of D. dendriticus differentially expressed genes
Full size table

Discussion

De novo sequencing of transcriptomes is a powerful tool for identifying and analyzing the genes of organisms for which the genome is not yet available. In addition, transcriptome data allow the identification of differentially expressed genes, which is an advantage of this sequencing method. Recently, there has been an increased interest in studying the life-cycle characteristics of helminth parasites such as Taenia pisiformis, Angiostrongylus cantonensis, Fasciola gigantica, and Schistosoma mansoni. In this study, we performed a comparative analysis of plerocercoid and adult D. dendriticus transcripts. Transcriptome assembly and analysis yielded and annotated 35,129 unigenes, noting that 16,568 (47%) unigenes were not annotated in known databases, which may indicate a unique set of expressed transcripts for D. dendriticus. Compared with other studies, we obtained fewer unigenes than for T. pisiformis (68,588) and A. cantonensis (72,957), but more than for the other two parasitic helminth species. If we estimate the number of differentially expressed genes, in our study of D. dendriticus, their maximum number of 8022 was determined, which allows us to speak of the peculiarities of the development of the two stages of D. dendriticus manifested by a large number of DEGs. The analysis of DEGs has shown that among the most differentially expressed genes, there are important genes characteristic of each stage. Thus, several genes are characteristic of D. dendriticus plerocercoids, including fatty acid–binding protein and ferritin.

Since tapeworms, such as Platyhelminthes, have lost the ability to synthesize their own lipids due to their parasitic lifestyle, the presence of fatty acid binding proteins (FABPs) plays a particularly important role in the survival of these organisms (Smyth and McManus 1989). FABPs are likely involved in the removal of fatty acids from the inner surface of the cell membrane and their subsequent placement at specific cellular sites (Alvite and Esteves 2012). In addition, FABPs may be involved in the uptake, transport, and storage of hydrophobic ligands, the targeting of ligands to specific organelles or pathways, the release of toxic compounds, and the regulation of gene expression (Alvite and Esteves 2012). To date, several homologous FABP proteins have been isolated and characterized in different parasites, namely, S. japonicum (Sj-FABPc) (Becker et al. 1994), S. bovis (SbFABP) (GenBank accession number AY615730), F. hepatica (Fh15) (Rodríguez-Pérez et al. 1992), F. gigantica (FgFABP) (Estuningsih et al. 1997), Echinococcus granulosus (EgFABP1 and EgFABP2) (Esteves et al. 1993; Esteves et al. 2003), Mesocestoides vogae (MvFABPa and MvFABPb) (Alvite et al. 2008), and T. solium (TsFABP) (GenBank accession number ABB76135).

Iron is an important element necessary for many cellular processes. Iron-containing proteins take part in several key biochemical pathways, such as DNA synthesis, electron transfer reactions, and energy metabolism (Beard 2001). Free intracellular iron reduced to the ferrous (Fe2+) form and catalyzes the generation of reactive oxygen species (ROS) which are cytotoxic reactive radicals that can cause harm to lipids, DNA, and proteins (Emerit et al. 2001). In ferritin, ferric iron (Fe3+) is kept in a nontoxic, soluble, and biologically non-reactive state (Meyron-Holtz et al. 2011). This capability suggests that ferritins function as protective proteins. Ferritin proteins have been reported in a wide range of organisms from prokaryotes, eukaryotes, and plants (Thiel 1987; Pulliainen et al. 2005; Lopez-Soto et al. 2009; Mohamed et al. 2010; Levi and Rovida 2015). Ferritin proteins have also been reported in Schistosoma mansoni (Dietzel et al. 1992), S. japonicum (Glanfield et al. 2007; Jones et al. 2007; Glanfield et al. 2010), T. saginata (Benitez et al. 1996), E. granulosus (Ersfeld and Craig 1995), Paragonimus westermani (Kim et al. 2002), and Clonorchis sinensis (Tang et al. 2006). During development in the mammalian host, the F. hepatica parasites feed on blood, hepatocytes, and bile. Therefore, iron compounds are required for parasite nutrition and egg production, which is consistent with reports on schistosomes (Dietzel et al. 1992; Jones et al. 2007), P. westermani (Kim et al. 2002), and C. sinensis (Tang et al. 2006). It is also expected that ferritin molecules in F. gigantica provide a protective mechanism against the harmful effects of iron. Unregulated degradation of ferritins could potentially lead to cellular toxicity owing to uncontrolled release of iron. Ferritin molecules in F. gigantica are also expected to provide a protective mechanism against the deleterious effects of iron. Therefore, ferritins could represent a potential drug target and/or vaccine candidate because of the vital roles they play in iron metabolism. Cabán-Hernández et al. (2012) have shown in their qPCR studies of Fasciola hepatica that expression of ferritin increases with parasite development and is consistent with the observed reactivity of FhFtn-1 with 2- and 4-week-old infection sera. At this time, the juvenile parasite is actively migrating through the liver parenchyma and feeding on blood and hepatocytes, thereby acquiring iron compounds necessary for its nutrition. Our findings are in agreement with the literature, as D. dendriticus actively feeds and grows at the plerocercoid stage.

Among the most highly expressed DEGs of the adult stage of D. dendriticus is the Kunitz-type serine protease inhibitor, in two putative isoforms. It should be noted that a recent study also showed increased expression of Kunitz-type protease inhibitors in T. pisiformis. Kunitz inhibitors are a class of serine protease inhibitors found in all multicellular animals and their prototype is the bovine pancreatic trypsin inhibitor (Rawlings et al. 2004, 2008). Currently, there are insufficient data on Kunitz inhibitors of parasitic helminths (Bozas et al. 1995; Hawdon et al. 2003; Gonzalez et al. 2009). However, further studies on these proteins could increase the understanding of the basis of their interaction with host proteins and their biological role in general.

The analyses of GO and KEGG metabolic pathways revealed that a large number of the DEGs of D. dendriticus are associated with the biosynthesis of various substances such as arginine and folate, as well as with various metabolic pathways such as galactose metabolism, selenocompound metabolism, and phosphonate and phosphinate metabolism.

Mononuclear phagocytes are among the first host cells to come into contact with parasites and can not only exert antimicrobial effects but also serve as a catalyst for parasite multiplication (Lugo-Villarino et al. 2019). Therefore, some parasites have developed the ability to regulate macrophage activation via host L-arginine metabolism and can manipulate specific metabolic pathways (Holzmuller et al. 2018). This mechanism is one of the current studies to control parasite infections by preventing the dysregulation of L-arginine through specific immunization or with drugs.

Another important finding of our study was the presence of an mRNA surveillance pathway among the TOP20 enriched KEGG pathways of D. dendriticus, specific for several parasite species and detected for the first time in parasitic tapeworms. Extensive studies on the mechanisms of the mRNA surveillance pathway have been performed in parasites such as Plasmodium falciparum (Erath et al. 2019) and Caenorhabditis elegans (Son et al. 2017). More than 60% of coding mRNAs were found. P. falciparum has an unbalanced genome composition, containing over 80% AT, and most of the genes with polyadenosine tracks are expressed throughout its lifespan (Djuranovic et al. 2020). Sites containing polyadenosine tracks are known to negatively affect gene expression (Wigington et al. 2014) and can halt ribosome activity, cause reading frame shifts, and activate mRNA surveillance mechanisms (Guydosh and Green 2017). It was found that the translation mechanisms of P. falciparum are able to translate genes containing polyadenosine tracks without error and without activating mRNA surveillance mechanisms (Guydosh and Green 2017; Erath et al. 2019). However, the molecular mechanisms of this AT-rich transcriptome, including the biological role of this feature, are currently unknown. Moreover, it is not entirely clear why the AT-rich genome composition trait has been conserved during long evolutionary development and what advantages this provides D. dendriticus for its vital functions. This study provides the first overview of differential gene expression in broad tapeworms by comparing the two developmental stages of D. dendriticus. Despite the lack of genomic data on this animal and its closest relatives, the transcriptome data obtained are very important for understanding the developmental features of the D. dendriticus life cycle. In the DEG analysis, we identified several important genes that could subsequently become targets for the development of next-generation vaccines and antiparasitic drugs.

Conclusion

The present work focuses on a comparative analysis of the transcriptomes of adult and plerocercoid D. dendriticus and the identification of their DEGs. A total of 8022 DEGs were identified in the plerocercoid and adult stages. The predominance of DEGs in the adult stages could be due to the greater biological activity of this stage, which could be due to feeding and hatching characteristics. The results contribute significantly to a better understanding of the molecular, biological, and cellular mechanisms triggered by the plerocercoid and adult D. dendriticus to regulate their life activity in the host. Finally, the identification of a large number of differentially regulated D. dendriticus transcripts offers the possibility to develop new molecular markers for diphyllobothriosis as well as potential new generation vaccines and anthelmintics.