A transcriptomic snapshot of early molecular communication between Pasteuria penetrans and Meloidogyne incognita

Background Southern root-knot nematode Meloidogyne incognita (Kofoid and White, 1919), Chitwood, 1949 is a key pest of agricultural crops. Pasteuria penetrans is a hyperparasitic bacterium capable of suppressing the nematode reproduction, and represents a typical coevolved pathogen-hyperparasite system. Attachment of Pasteuria endospores to the cuticle of second-stage nematode juveniles is the first and pivotal step in the bacterial infection. RNA-Seq was used to understand the early transcriptional response of the root-knot nematode at 8 h post Pasteuria endospore attachment. Results A total of 52,485 transcripts were assembled from the high quality (HQ) reads, out of which 582 transcripts were found differentially expressed in the Pasteuria endospore encumbered J2 s, of which 229 were up-regulated and 353 were down-regulated. Pasteuria infection caused a suppression of the protein synthesis machinery of the nematode. Several of the differentially expressed transcripts were putatively involved in nematode innate immunity, signaling, stress responses, endospore attachment process and post-attachment behavioral modification of the juveniles. The expression profiles of fifteen selected transcripts were validated to be true by the qRT PCR. RNAi based silencing of transcripts coding for fructose bisphosphate aldolase and glucosyl transferase caused a reduction in endospore attachment as compared to the controls, whereas, silencing of aspartic protease and ubiquitin coding transcripts resulted in higher incidence of endospore attachment on the nematode cuticle. Conclusions Here we provide evidence of an early transcriptional response by the nematode upon infection by Pasteuria prior to root invasion. We found that adhesion of Pasteuria endospores to the cuticle induced a down-regulated protein response in the nematode. In addition, we show that fructose bisphosphate aldolase, glucosyl transferase, aspartic protease and ubiquitin coding transcripts are involved in modulating the endospore attachment on the nematode cuticle. Our results add new and significant information to the existing knowledge on early molecular interaction between M. incognita and P. penetrans. Electronic supplementary material The online version of this article (10.1186/s12864-018-5230-8) contains supplementary material, which is available to authorized users.


Background
Following the publication of the Caenorhabditis elegans (Maupas, 1900) Dougherty, 1955 genome [104], some laboratories have become focused on using this nematode as a model for studying innate immunity [61,71,96,111]. Due to the short developmental time of this nematode, the studies have necessarily focused on the infection of adults and earlier developmental stages, and in particular the non-feeding dauer stage have been neglected. The life-cycle of plant-parasitic nematodes commences when an infective juvenile hatches from an egg as a second-stage juvenile, and migrate through the soil seeking a suitable host plant before feeding starts. This period of time offers the opportunity to study early responses of the nematode to bacterial infection [26].
Pasteuria penetrans (Thorne, 1940) Sayre and Starr, 1985, a Gram-positive soil bacterium of the Bacillus-Clostridium clade, is a hyperparasite of the root-knot nematodes (RKN), and represents a typical naturally coevolved pathogenhyperparasite system [15,86]. This is an excellent system to study the early stages of the nematode infection processes by bacterial parasites. The life-cycle and developmental stages of P. penetrans inside M. incognita are well documented and understood [28,82]. The bacterium completes its life-cycle in three phases, [1] Attachment and germination, [2] Rhizoid production and exponential growth; and [3] Sporogenesis [28]. There is a high degree of genetically regulated host specificity involved in this interaction. Pasteuria is highly selective and specific to their host; one population of Pasteuria will not recognize and infect other species in the same genus, and not even all populations of the same species [24]. The surface of nematode cuticle plays a decisive role in facilitating the specificity of the adhesion [27,99] and the attachment of P. penetrans endospores to an as of yet uncharacterized cuticle receptor is the primary and arguably the most crucial step of the bacterial infection [28]. After the RKN J2 s establish permanent feeding sites in their plant hosts, the endospores perceive some currently unknown cue(s) from the nematode and germinate [25,93]. The bacterium proliferates inside the worm's body, kills it, and converts the females into an "endospore sac" containing millions of endospores [25,82].
The recent development of genomic tools and technologies for the plant-parasitic nematodes has enabled researchers to investigate in detail at the molecular level the nematode's interactions with their hosts, symbionts and pathogens/hyperparasites. It is known that hosts respond to pathogen attack by altering their gene expression; in the infection of Daphnia by Pasteuria ramosa, it was observed that the incompatible/resistant hosts responded by an up-regulated gene response, whereas the down-regulated gene response was pronounced in the compatible/susceptible host [64]. However, in a recently published study that used RNA-Seq to understand the nematode's response to Pasteuria infection at three days post attachment, when nematode appeared less mobile and moribund because of the bacterial infection, it was found that 91% of the 445 differentially expressed genes were up-regulated [117]. This was contrary to the general understanding that a down-regulated gene response is exhibited by the susceptible hosts, which is probably caused by subversion of host immunity by the parasite [6]. Therefore, this large up-regulation in gene expression by Daphnia against Pasteuria infection over time, as reported by McTaggart et al. [64] and again by Zou et al. [117] using a population of M. incognita susceptible to Pasteuria penetrans at three days post attachment warranted further investigation.
In order to understand the nematode genes involved in the very first and crucial stage of the M. incognita -P. penetrans interaction, i.e. the recognition and adhesion of endospores, here we investigated the early transcriptional response of M. incognita at eight hours after their initial encounter and before the infective J2 s have invaded the plant root. Additionally, we also identified the functional role of five important differentially expressed genes in M. incognita -Pasteuria interaction by using RNAi based gene silencing and measuring their effects on endospore adhesion.

Biological materials
The single egg mass culture of an Indian isolate of M. incognita race 1 was increased on tomato plant (Solanum lycopersicum L. cv. Pusa ruby) in a glasshouse at ICAR-Indian Agricultural Research Institute, New Delhi, India. Nematode infected tomato roots were washed free of soil, egg mass were hand-picked and kept for hatching on a modified Baermann's funnel assembly [110]. The freshly hatched J2 s were used for all of the experiments, viz., endospore attachment, RNA isolation, dsRNA treatment etc. The unused/left over J2 s were autoclaved and discarded.

Endospore attachment
Pasteuria penetrans (Strain AII-329: Pasteuria collection, ICAR-IARI, New Delhi, India) endospores were produced on M. incognita cultured on adzuki bean (Vigna angularis (Willd.) Ohwi and Ohashi) in CYG growth pouches (Mega International, St Paul, MN, USA) as described by Rao et al. [88]. Transfer of germinated seeds in growth pouches, setting up of root infection and post infection maintenance of plants were conducted as described earlier [84].
The freshly hatched ca. 20,000 J2 s were collected in a 1.5 ml microcentrifuge tube and mixed with 200 μl of endospore suspension (2.5 × 10 3 endospores ml − 1 ). The attachment of endospores was pursued by centrifugation method [41] and resulted in attachment of approximately 30-35 endospores on the cuticle surface of each juvenile (n = 200). The attachment of endospores onto the juveniles was confirmed microscopically. Following endospore attachment, the J2 s were washed thrice in M9 buffer (1 mM MgSO 4 , 22 mM KH 2 PO 4 , 42.3 mM Na 2 HPO 4 and 85.6 mM NaCl; pH 7.0) to remove the free endospores. The endospore encumbered J2 s were incubated in fresh M9 buffer at room temperature (28°C) for 8 h on a slowly moving rotator. The adherence of the endospores onto J2 surface was again confirmed microscopically after 8 h, and the juveniles were found to have similar numbers of attached spores as seen earlier. Following attachment, the J2 s were incubated at room temperature (28°C) for 8 h on a slowly rotating incubator. The freshly hatched juveniles incubated in M9 buffer for 8 h without Pasteuria endospores served as a control.

RNA extraction
Total RNA was extracted from about 20,000 M. incognita non-encumbered and endospore encumbered J2 s with TRIzol reagent (Thermo Fisher Scientific, Waltham, MA, USA) according to manufacturer's protocol. The RNA was treated with RQ1 RNase-Free DNase (Promega, Madison, WI, USA) to remove any genomic DNA contamination. The integrity of the isolated RNA was tested on a Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). The quality and concentration was determined by 1% agarose gel and NanoDrop-1000 spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA). The process was replicated twice.
cDNA synthesis, library preparation, RNA-sequencing The total RNA was subjected to downstream processing for cDNA synthesis and library preparation. The extracted RNA was assessed for quality using an Agilent 2100 bioanalyzer (Agilent Technologies, Santa Clara, CA, USA) and RNA with an RNA integrity number (RIN) of 8.0 was used for mRNA purification. The mRNA (messenger RNA) was purified from approximately 5 μg of intact total RNA using oligodT beads (Illumina® TruSeq® RNA Sample Preparation Kit v2). The purified mRNA was fragmented in the presence of bivalent cations and first strand cDNA was synthesized using Superscript II reverse transcriptase (Invitrogen, Carlsbad, CA, USA) and random hexamer primers (Invitrogen, Carlsbad, CA, USA). Second strand cDNA was synthesized in the presence of DNA polymerase I and RNaseH following standard protocol (Illumina). The cDNA was cleaned using Agencourt AMPure XP purification kit (Beckman-Coulter, Brea, CA, USA), amplified, quantified using a Nanodrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA) and checked for quality with a Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). In total, 4 libraries were prepared for non-encumbered and Pasteuria encumbered samples (2 each) as per the Illumina protocols. The cDNA libraries were then sequenced on the Illumina HiSeq platform by outsourcing to Bionivid Technologies Pvt. Ltd., Bangalore, India.

Transcriptome assembly, quantitation and identification of differentially expressed transcripts
All the Paired End fastq files were subjected to standard quality control using NGS QC toolkit, v2.3.3 [77]. Reads with adapter contamination were removed along with their mate pair. High Quality (HQ) reads from all the samples were merged together to generate a Primary Assembly using Trinity Assembler (Trinity RNA-Seq-v2.0.6) [36] with default k-mer length 25, minimum contig length 200 bp and minimum k-mer coverage as 5. Further amelioration of the transcripts was done by filtering on the basis of average depth (≥ 5) and coverage (≥ 70%) in the individual samples [5]. The ameliorated transcripts were then subjected to clustering using CD_HIT_EST (v4.6.1) to make the secondary and final assembly with sequence identity threshold as 0.8 and length difference cut off as 0.9. The redundant transcripts were removed by CD_HIT_EST to make the secondary assembly. We observed a higher percentage (~45%) of small transcripts (< 500 bp). These were further fileterd based on annotation obtained against NCBI NRDB protein database. Sequences < 500 bp, which remained un-annotated, were discarded and a final transcriptome assembly was generated.
The final assembly was used for quantitation of transcripts in each of the individual libraries by using RSEM method [58]. The assembly validated .bam (Binary Sequence Alignment/Map) file was processed using bedtools [87] and samtools [59] for quantitation (read count estimation) for each transcript in a library and also to calculate the total coverage and average depth of the transcriptome in each library. The differentially expressed transcripts were identified using DESeq R package [3] between the treatment (endospore encumbered J2 s) and control (non-encumbered J2 s) groups in replicate. The differential expression of transcripts was determined with log2fold change ≥2 & P value ≤0.05 obtained by DESeq analysis.

Transcript annotation
Homology based annotation for the final transcriptome was done against National Center for Biotechnology Information (NCBI) non-redundant (nr) protein database [2]. The filtration criteria used for blastx were: Evalue ≤0.001, Query Coverage ≥60 and Percentage Identity ≥40. The results were subjected to Gene Ontology (GO) and Pathway analysis using Blast2GO [22] and KAAS [69]. Additional analysis were performed to find the secreted peptides, neuropeptides and RNAi pathway genes present in the differentially expressed transcripts by using SignalP v4.1 [72], and by blast search against the local database of C. elegans neuropeptide sequences (Li and Kim, 2008) and M. incognita RNAi genes [1] at E value ≤0.001 and query coverage ≥60, respectively.

Validation of RNA-Seq gene expression data by qRT PCR
Quantitative real time PCR (qRT PCR) was carried out to confirm the expression pattern of fifteen selected transcripts differentially expressing in the range of + 2.01 fold to − 5.06 fold in the RNA-Seq experiment. These 15 transcripts (Table 1) were chosen based on their predicted role in nematode -Pasteuria interaction, and included ten down-regulated transcripts and five up-regulated transcripts. cDNA was prepared from the same RNA samples that were used for RNA-Seq. Approximately 500 ng of RNA was reverse transcribed using cDNA synthesis kit (Superscript VILO, Invitrogen, Carlsbad, CA, USA) and qRT PCR was performed in a realplex 2 thermal cycler (Eppendorf, Hamburg, Germany) using SYBR Green Supermix Kit (Eurogentec, Liege, Belgium). Each reaction mixture contained a final volume of 10 μl, comprised of 5 μl of SYBR Green PCR Master mix (Eurogentec, Liege, Belgium), 750 nM of each primer and 1.5 ng of cDNA. To normalize the gene expression level 18S rRNA (Genbank accession: HE667742), a constitutively expressed gene was used as internal reference. Three biological and three technical replicates were maintained for each sample. The data were analyzed by ΔΔCt method [60] and results were expressed as log2-transformed fold change values and Dunnet's multiple comparison was performed for determining the statistical significance of the expression data. The primer details for qRT PCR are provided in Table 1.

Functional validation of role of transcripts in Meloidogyne -Pasteuria interaction
To determine the role of up-and down-regulated transcripts in the Meloidogyne -Pasteuria interaction, five out of the 15 transcripts, i.e., TR10010, TR14793, TR26363, TR16177 and TR10990 were selected and silenced by RNAi. Conserved domains in the transcripts were analyzed by NCBI Conserved Domain Database (https://www.ncbi.nlm.nih.gov/Structure/cdd/wrpsb.cgi) and specific primers were designed to amplify them ( Table 1). The dsRNA for each of the five transcripts was prepared as described earlier [84]. The effect of dsRNA treatment on the attachment of endospores on nematode cuticle was tested by soaking the J2 s in dsRNA solution at 28°C for 18 h [105], followed by mixing 100 μl of endospore suspension (2.5 × 10 3 ml − 1 ) with approximately 200 dsRNA treated J2 s as described earlier [41]. The level of transcript suppression after dsRNA feeding was quantified and analyzed by qRT PCR [84]. Freshly hatched J2 s and J2 s soaked in dsGFP were used as controls. The endospore attachment was quantified by observing~30 nematodes with a Zeiss Axiocam compound microscope (Carl Zeiss, Oberkochen, Germany) and photographed. The assays were performed in triplicate.

Transcriptome sequencing and differentially expressed genes
The RNA sequencing of the Pasteuria endospore encumbered M. incognita J2 s and non-encumbered J2 s generated 32 to 39 million reads per sample ( Table 2). The number of high quality reads in the raw sequence data was more than 99% for each of the samples ( Table 2). Quality filtering of the raw sequence data using NGS QC resulted in 32.69, 34.62, 39.38 and 37.29 million reads for each of the two samples of non-encumbered and endospore encumbered nematode J2 s, respectively ( Table 2). The quality filtered HQ sequence data from all the samples were used to generate a common reference assembly using the Trinity assembler resulting in a total of 161,705 transcripts. The minimum and maximum transcript lengths were 224 and 8320 bp, respectively with the N50 transcript length 1004 bp (Table 3). Improving the assembly by amelioration, removing the duplicates by CD_HIT_EST and filtering un-annotated < 500 bp transcripts resulted in a final merged assembly of 52,485 transcripts (minimum length: 224 bp, maximum length: 8320 bp) with N50 transcript length 1159 bp & GC% of 33.99 ( Table 3). The sequence data from individual samples were analyzed for differentially expressed transcripts between the endospore encumbered and non-encumbered J2 s. Eight hours post attachment of Pasteuria endospores, a total

Characterization of differentially expressed transcripts
The annotation of 582 differentially expressed transcripts was done by blast search against NCBI nr database. Out of 582 transcripts, 246 transcripts showed nematode genes as the top hits (Fig. 2a) Fig. 2b. The transcripts showing highest down-regulation in the RNA-Seq experiment, TR38275 (− 7.0 folds) and TR22780 (− 6.9 folds) were annotated as heat shock protein 90 (HSP 90) and papain family cysteine protease, respectively (Additional file 1). The biggest group of down-regulated proteins were different ribosomal proteins (79 transcripts), whereas, hypotheticals represented the second largest group of the down-regulated proteins (49 transcripts). The heat shock proteins represented the third biggest group of down-regulated proteins (23 downregulated transcripts) and included proteins such as HSP 70, HSP 20 and HSP 12.6. Eleven protease/proteinase (including two cysteine proteinases) and six kinases/ phosphatases also showed down-regulation. Some other interesting proteins showing down-regulation were plectin (2 transcripts) and secretory proteins (3 transcripts), two fructose bisphosphate aldolase class-I family, one fatty acid and retinol binding protein, one signal transduction protein possibly involved in cell surface receptor  signaling pathway (GO:0007166), three ubiquitin related proteins, three homologues of phospholipase B like 2 and one of phospholipase A2. In addition, M. incognita genes showing homology to tropomyosin (2 transcripts), actin-2 and tubulin (1 transcript each) were also found to be substantially down-regulated (Additional file 1). The top three highly up-regulated transcripts in Pasteuria encumbered J2 s were TR25864, TR1903 and TR25239 that showed 9.4, 9.3 and 8.9 fold up-regulation in the RNA-Seq experiment. All of these three transcripts were uncharacterized or hypothetical proteins. In fact, hypotheticals, uncharacterized and unknown proteins were the dominant group under the up-regulated transcripts with 96 out of total 229 up-regulated transcripts falling into these categories. Cytochrome oxidase and major sperm protein domain containing proteins were the second most enriched transcript groups with 13 transcripts each. Blast based annotation showed that five transcripts matched to hormone receptor-like in 38 genes, four to transformation transcription domain-associated protein, two to venom allergen-like protein and one transcript matched to TK/FER protein kinase, UDP-glucosyl transferase and glutathione-S-transferase each (Additional file 1).
Analysis of the pathways represented by transcripts differentially expressed in the encumbered J2 s was performed by KAAS, an automatic genome annotation and  In addition, pathways involved in recycling of cellular contents such as endocytosis (KO04144), phagosome (KO04145), lysosome (KO04142) and autophagy-animal (KO04140) were also found to be active in the encumbered J2 s. The search for differentially regulated secreted peptides revealed that a total of 376 differentially expressed transcripts contained secretion signal, out of which 90 transcripts showed up-regulation, whereas 286 showed down-regulation. Of these 376 transcripts, only 320 could be annotated ( Table 6, Additional file 2).

Validation of differentially expressed transcripts by qRT PCR
A total of fifteen transcripts, identified as differentially expressed in the RNA-Seq data, were taken up for validation by qRT PCR. Expression of 12 out of 15 transcripts was validated as significantly up-or down-regulated at 8 h post endospore encumbrance (Table 7). Unlike the RNA-Seq data, the fold change expression of the transcripts encoding heat shock protein 20, glutathione S-transferase and tropomyosin was not found to be significant. Data revealed that the selenium binding  On the other hand, silencing of aspartic protease (118 ± 12 endospores/J2) and ubiquitin (101 ± 8 endospores/J2) resulted in approximately three fold higher incidence of endospore attachment. However, there was no change on endospore attachment in the major sperm protein silenced J2 s (31 ± 6 endospores/J2), as compared to the controls (Fig. 3).

Discussion
Present RNA-Seq experiment provides a snapshot of the early molecular response by the M. incognita J2 s after exposure to P. penetrans endospores. Our results show that 582 nematode transcripts were differentially expressed at 8 h post Pasteuria endospore attachment of which 229 were up-regulated, and 353 were down-regulated. The expression patterns of 12 out of 15 differentially expressed (ten down-and five up-regulated) transcripts identified by the RNA-Seq were significantly validated to be true by qRT PCR. The functional role of five transcripts predicted to be important in Meloidogyne -Pasteuria interaction were validated by RNAi. The silencing of transcripts coding for fructose bisphosphate aldolase (TR10010) and glucosyl transferase (TR14793) resulted in approximately six times lower endospore attachment as compared to the controls, whereas, silencing of aspartic protease (TR16177) and ubiquitin (TR10990) coding transcripts resulted in approximately three fold higher incidence of endospore attachment. No change on endospore attachment was detected in major sperm protein (TR26363) silenced J2 s. As established in several natural host-parasite interactions, viz., Daphnia -Pasteuria, Anopheles -Plasmodium and Bumblebee -Trypanosoma, a successful infection results from a compatible molecular interaction between the host and parasite genotypes [6,55,56,64]. Daphnia magna Straus, 1820 responds to P. ramosa Metchnikoff, 1888 infection by rapid gene expression at 4 h, and the response tapers off with the passage of time at 8 and 12 h [64]. However, unlike Daphnia, which becomes infected after Pasteuria is orally ingested and adheres to the esophageal region, Meloidogyne becomes infected when Pasteuria spores adhere to the juvenile body surface and germinate when the nematode starts to feed on plants. In a study on Meloidogyne -Pasteuria interaction at three days post endospore adherence, Zou et al. [117] found that nematodes respond by an up-regulated gene expression, and identified 445 differentially expressed genes out of which 406 were up-regulated while 39 got down-regulated. Thirty seven immune responsive genes encoding collagens, cytochrome P450, lysozymes were among the other identified active proteins. Interestingly, Zou et al. [117] found that biosynthesis of unsaturated fatty acid pathways was up-regulated, while cytochrome P450 related genes were down-regulated. However, the findings from the Daphnia -P. ramosa system suggest that there may be an even earlier transcriptional response; hence, we focused on understanding the early transcriptional response by the nematode at eight hours after endospore attachment.
Our study showed that 153 pathways were active in Pasteuria encumbered M. incognita J2 s at 8 h post infection. As compared to our finding, 98 pathways were enriched at 3 days post Pasteuria attachment [117], of which 58 pathways were found to be active at both the time points (Additional file 2). ErbB signaling, Rap1 signaling, FoxO signaling, Hedgehog signaling, thermogenesis, longevity regulating pathways were some of the pathways active only at 8 h post endospore attachment; whereas pathways like cytochrome P450 (drug and xenobiotic metabolism), MAPK signaling, calcium signaling, HIF-1 signaling, mTOR signaling, PI3K-Akt signaling, Wnt signaling, TGF-beta signaling, Hippo signaling and regulation of actin cytoskeleton were active at both the time points. Some of the pathways active at 8 h assist cell-cell interaction and cellular adhesion in model organisms including C. elegans, for example, ErbB signaling [113], TGF-beta signaling [39], Hippo kinase cascade [114] and Rap1 signaling [8]. Other pathways like Wnt signaling [97], FoxO signaling [40], PI3K-Akt signaling [33], AMPK signaling [68], MAPK signaling [47], mTOR signaling [43] and HIF-1 signaling [116] operate in bacterial pathogenesis and stress responses. Taken together with the findings of Zou et al. [117], it appears that M. incognita immune pathways like cytochrome P450, MAPK signaling, HIF-1 signaling, mTOR signaling, PI3K-Akt signaling and TGF-beta signaling pathways were active upon infection by Pasteuria from 8 h to 3 days, in addition to the regulation of actin cytoskeleton pathway (Additional file 2). Our study indicates that the nematode's transcriptional responses after Pasteuria infection can be broadly understood as: (a) transcripts involved in nematode immunity, (b) transcripts involved in altering the cuticular surface coat property and thereby affecting endospore attachment, (c) transcripts involved in modulating the behaviour of the endospore encumbered juveniles.

Nematode immune responses triggered by Pasteuria
In addition to the immune pathways discussed above, at 8 h post endospore encumbrance, the ribosomal pathway (KO03010) was identified as the most affected with 62 differentially expressed transcripts. Besides, the ribosomal proteins represented the largest down-regulated group (79 transcripts) along with three ubiquitin related transcripts and one ZIP or ZRT/IRT-like protein (TR38155). The suppression of these transcripts suggest a reduction in the mRNA translational activity in Pasteuria encumbered M. incognita J2 s. This is consistent with the earlier reports of "effector-triggered" or "surveillance immunity", where a down-regulation of host mRNA translation upon attack by bacterial pathogens has been reported in plants and in C. elegans [20,29,63,67]. The bacterial pathogens disable the process of host mRNA translation, thereby preventing the production of anti-microbial molecules, and improve the chances of the infection [20]. It appears that a similar strategy is being used by Pasteuria while infecting M. incognita. The RNAi mediated knockdown of nematode ubiquitin also increased the endospore attachment on cuticular surface. Several bacterial pathogen effectors interact with eukaryotic ubiquitination pathways to exploit host functions [80]. The involvement of ubiquitin proteasome system, targeting proteins for degradation, has been established in C. elegans as an inducible response to infection [4,20,65]. Silencing of ubiquitin may lead to anomalous immune response where the cells fail to mount a sufficient immune response to remove the pathogen.
Approximately 13 autophagy related transcripts (e.g. endocytosis, KO04144; phagosome, KO04145; lysosome, KO04142 and autophagy-animal, KO04140) were down-regulated in the endospore encumbered M. incognita J2 s. It is well established that autophagy plays key role in pathogen defense [20]. Further, a down-regulation of 23 heat shock protein coding transcripts, for example, HSP70, HSP20 and HSP12.6 was observed. The HSPs are highly conserved group of proteins and are involved in protection against biotic and abiotic stress [19,32,44,81,91,95,115]. Similarly, HSP70s and HSP20 are known to play crucial role during disease stress response and serves as an endogenous danger signal [10,23,98]. Although down-regulation of HSPs during biotic stress is rare [78]; small HSPs (e.g. HSP17, HSP21 etc.) and HSP70 were found to be down-regulated in Arabidopsis when challenged by Pseudomonas syringae van Hall, resulting in suppression of host defense responses [10]. The down-regulation of autophagy related genes and HSPs by P. penetrans indicates pathogen induced host defense suppression.
In addition to the above mentioned pathways, several differentially expressed transcripts related to nematode immunity against bacterial pathogens were identified, such as, aspartic protease (TR16177), phospholipase A2 (TR14120), glutathione S-transferase (TR40461), selenium binding proteins (TR31579) and hormone receptor-like in 38 (Hr38) (TR5128, TR39260). Aspartic proteases function in the intracellular and extracellular degradation of proteins including processing of peptide hormones, antigens and immunoglobulins in parasitic nematodes [51,100,101]. In the present study, the RNAi induced inhibition of aspartic protease led to an increase in Pasteuria endospore adhesion on nematode cuticle indicating the defensive role of aspartic protease against bacterial pathogens. The inhibition of nematode phospholipase A2 by Pasteuria possibly prevents the bacterial infection structures from degradation, thereby allowing further infection [89]. Glutathione S-transferase (GST) was up-regulated in Pasteuria infected J2 s. In C. elegans and parasitic helminthes, GST is associated with an immune response and xenobiotic metabolism [11,12,18,46,62,107,108]. The up-regulation of GST may indicate nematode's efforts to detoxify the potential Pasteuria effectors [86]. The RNAi mediated knockdown of M. incognita selenium binding protein (SeBP) significantly increased P. penetrans endospore attachment possibly through altering the cuticular surface coat property [83]. Lastly, the nuclear receptors (NRs) are ligand-dependent transcription factors that play pivotal roles in cell growth, differentiation, metabolism, reproduction and morphogenesis and immunity [13,34]. Up-regulation of hormone receptor-like in 38 (Hr38) transcript, the Drosophila ortholog of the mammalian NGFI-B subfamily of orphan nuclear receptors [16] in M. incognita may be an indication of detection of the foreign nucleic acid particle (of Pasteuria), thereby triggering an immune response in M. incognita.

Modification of nematode cuticle biochemistry by Pasteuria
Mutations in genes involved in the building of complex cuticular surface components in C. elegans are known to affect bacterial adhesion [37]. Several differentially expressed transcripts identified in this study are predicted to interfere with attachment of endospores on the nematodes by altering the cuticle surface structure. These include TR10010 (fructose bisphosphate aldolase), TR14793 (glucosyl transferase), TR11544 (fatty acid and retinol binding (FAR) protein), and TR24724 (TK/FER kinase).
The fructose bisphosphate aldolase is a key enzyme of the glycolytic pathway [79] and its inhibition may contribute to accumulation of sugar molecules on the cuticle surface [30]. Similarly, glucosyl transferase catalyzes the transfer of sugar moieties to a wide range of acceptor molecules [73,112]. RNAi induced in vitro inhibition of both these transcripts reduced the endospore attachment on nematode surface. Silencing fructose bisphosphate aldolase might have resulted in accumulation of sugar on the specific carbohydrate recognition sites on nematode mucin-like glycoprotein that binds to Pasteuria endospores [25,85], thereby leading to reduction in attachment of endospores. Additionally, inhibition of glucosyl transferase may also result in mis-folding of the native glycoprotein molecules, thus resulting in decreased endospore attachment. Our findings are in concurrence with earlier observations on C. elegans, where the bus-8 mutant worms, defective in expressing glucosyl transferase were resistant to infection by Microbacterium nematophilum Hodgkin, Kuwabara and Corneliussen, 2000, due to failure of the bacterium to bind to the host surface [76]. The fatty acid and retinol binding (FAR) protein, uniquely present in nematodes, has been found to inhibit bacterial attachment onto nematode body surface by sequestering pharmacologically active lipids [45]. It has been demonstrated that the FAR protein is involved in the protection of M. incognita from attachment of P. penetrans endospores [84]. The TK/FER kinase regulates cadherin and integrin dependent cellular adhesion [9] and is also involved in nascent cell-cell adhesion via phosphorylation pathway [50]; and our results suggest it may also be involved in M. incognita and P. penetrans infection processes.

Alteration of nematode behavior and locomotion by Pasteuria
The endospore encumbered stressed nematode juveniles are slower in movement and finding their hosts [25,84,106]. The role of tropomyosin, actin and tubulin in regulation of muscle contraction has been established for several plant-parasitic, animal-parasitic and free living nematodes [38,54]. As observed in our RNA-Seq data, down-regulation of these muscle-associated proteins upon endospore attachment may result in perturbed locomotion of the nematodes. The RNA-Seq data showed perturbation in a large number of neuropeptides in the Pasteuria encumbered juveniles in our study (Additional file 2). The neuropeptides of major groups, viz., FLPs, NLPs and ILPs affect numerous behavioral responses via various signaling pathways [14,17,21,52,109]. The association of FLPs and NLPs with nematode locomotion has been observed in C. elegans [48,70,90] and ILPs in the fruit fly movement [31]. Disruption of flp-18 in M. incognita by RNAi is known to reduce the plant parasitism [75]. The nematode behavioral alteration could also result from the effect of Pasteuria on the neuropeptides, as suggested by our results.
Apart from the above mentioned pathways and transcripts, two interesting observations from our study need mention. Firstly, we found that a major sperm protein (MSP) was up-regulated in infected juveniles, but RNAi mediated silencing of MSP coding transcript (TR26363) did not affect the adhesion of Pasteuria endospores onto nematode cuticle surface. MSP is involved in motility machinery and crawling movement of nematode sperms [92,94], and is regulated at the onset of sexual differentiation in nematodes [49]. In C. elegans, MSP promotes the oocyte maturation and MAPK activation [53,66]. It is well known that Pasteuria is largely confined to the reproductive system of M. incognita leading to complete destruction of its reproductive ability [7]. The sexual differentiation and gonad development process in M. incognita starts in the second-stage juveniles (J2 s) [74] and the up-regulation of MSP by Pasteuria might indicate an early interference with the nematode's reproductive system.
Secondly, it has been observed that Pasteuria infected M. incognita have an increased life span of 10-12 days [25,82], which could be a consequence of reduced mRNA translation in microbial infection. This corroborates with other studies in which the down-regulation of protein synthesis has resulted in increased lifespan [35], possibly by two different mechanisms. Firstly, reduced mRNA translation decreases the synthesis of normal as well as damaged proteins, resulting in lower accumulation of toxic proteins [42]. Secondly, as protein synthesis is a high cellular energy-consuming process [57], reduction of mRNA translation might increase energy availability and allow diversion of critical resources towards cellular maintenance and repair, thus promoting longevity [102,103]. Our study shows a major reduction in protein synthesis in the Pasteuria encumbered nematodes thereby possibly extending lifespan when compared to the healthy nematodes.
Lastly, the hypothetical proteins represented the second largest group of the down-regulated proteins with 49 transcripts. There could be several interesting candidates within this group that may be directly or indirectly involved in the RKN -Pasteuria interaction.

Conclusion
Here we presented a transcriptomic analysis of M. incognita and P. penetrans interaction at eight hour post initial encounter. The transcriptome profile revealed that Pasteuria infection causes a reduction of the protein synthesis machinery of the nematode. We identified several differentially expressed transcripts putatively involved in nematode innate immunity, endospore attachment process and post-attachment behavioral modification of the juveniles. RNAi based functional validation of fructose bisphosphate aldolase (TR10010), glucosyl transferase (TR14793), aspartic protease (TR16177) and ubiquitin (TR10990) coding transcripts resulted in altered incidence of endospore attachment, whereas silencing of major sperm protein (TR26363) did not result in any alteration in Pasteuria attachment. Our results add new and significant information to the existing knowledge on early molecular interaction of M. incognita and P. penetrans.