Identification of Suitable Meloidogyne spp. Housekeeping Genes

We utilized the publicly available M. hapla RNAseq database (https://brcwebportal.cos.ncsu.edu/haplatome/1-0_hapla.php) to discover potential HK genes for RKN (Cha, 2016). Genes with more than five reads per sample across all the treatments were analyzed. Differently expressed genes (DEG) were determined in R v.3.3.3 (Team, 2006) using package “edgeR” (McCarthy et al., 2012). Specifically, the count of the genes was normalized using function “calcNormFactors”, and then the dispersion was estimated using function “estimateDisp”; finally, pairwise comparison was accomplished using function “exactTest”. Genes with comparision value 0 across all treatments were considered non-DEG. The non-DEG with matching loci within the M. hapla genome were subject to SignalP 4.1 to determine whether they encode secreted proteins ( Petersen et al., 2011), putative secreted protines were omitted as candidates due to the possibility of direct involvement in the plant–nematode interaction. The details of candidate HK genes are in listed in Table A1. The gene functions were annotated utilizing both KEGG (Kanehisa and Goto, 2000) and WormBase parasite website (Howe et al., 2017). Seven genes of the non-DEG pool were selected based upon ease of primer design and high expression level to validate stability of gene expression throughout the life cycle of M. incognita in growth chamber assays.

To test whether the genes can be widely applied to all RKN, a growth chamber experiment was set up to validate candidate HK genes on M. incognita. Tomato seeds (variety “Rutgers”) were surface sterilized and planted in potting soil. After 3 wk, seedlings were transferred into individual pots with sand. Second stage juveniles (J2) of M. incognita were hatched from freshly collected eggs, and 300 J2 were inoculated into each pot. The pots were kept in a growth room with 16 hr light/8 hr dark at 26°C. The freshly collected eggs and J2 were also stored in −80°C for RNA extraction.

Roots were gently washed in sterile water, dried and immediately frozen in liquid nitrogen. A total of 11 time points were collected with five time points during early infection, (1, 2, 4, 5, and 7 days post inoculation-dpi) and six time points during nematode maturity (21 dpi) (Table 1). To test diurnal effects on RKN gene expression, at the nematode maturity stage, root samples were collected three time points in dark and three time points in light (Table 1). Each of the total 11 time points had four biological replicates, and 3-wk old non-infected tomato plants were used as negative control.

The homologues of M. hapla HK gene of M. incognita were obtained through WormBase parasite website. The transcript sequences of M. incognita V3 were used as template in IDT primerQuest tool for RT-qPCR primer design with the default parameters. To avoid any possible DNA amplification which would alter transcript level detection, primers were designed to span an exon–exon junction if possible or target adjacent exons. We checked the specificity of the primers against the whole transcriptome of M. incognita using Primer-BLAST on the NCBI website. The corresponding gene ID and annotation of the candidate HK genes are included in Table 2, and two genes with high to low DEG value, plus a known actin gene (Duarte et al., 2016) were used as controls (Table A2). Table 2.

Validated Candidate housekeeping genes and primers for M. incognita.

RNA was extracted from the root samples, eggs and naïve, pre-parasitic J2 using RNeasy Plant Mini kit (Qiagen, USA) according to the manufacturer’s protocol. cDNA was synthesized with the iScript Reverse Transcription Supermix (Bio-RAD) for RT-qPCR according to the manufacturer’s instructions.

RT-qPCR was carried out with a StepOne Plus Real Time PCR detection system (Applied Biosystem, USA) using SYBR Green as detection agent. PCR reactions were conducted with three technical replications for each sample in MicroAmp Fast 96 well reaction plate (0.1 ml) (Applied Biosystems). The total volume of the reaction was 10 μl with 0.5 μl forward and reverse primers each at 10 μM, 2.5 μl cDNA template, 1.5 μl of Nuclease free water and 5 μl SYBR Green qPCR MasterMix (iTaq Universal SYBR Green Supermix, Bio-RAD). PCR was performed at 95°C for 20 sec followed by 40 cycles at 95°C for 3 sec, and 60°C for 30 sec, the melting curve was generated at 95°C for 15 sec, 60°C for 1 min, and 95°C for 15 sec. Ct value was calculated using the StepOne software V2.3 (Applied Biosystems).

To ensure no off-target amplicons were produced, the melt curve was checked carefully to only include samples with single peaks at the correct melting temperature for downstream analyses. The mean C_T (cycle threshold) value of each sample was calculated in excel from three technical replicates. The data were organized in excel according to the suggested format by NormFinder (Jensen and Ørntoft, 2004). According to the algorithm of NormFinder, candidate HK genes were compared pairwise within and across groups resulting in a stability value; the most stable HK genes are those with the lowest stability value. In our study, each time point was treated as a group factor. To test the robust of NormFinder analyses, the C _T value was also submitted to geNorm, another similar HK gene algorithm, for stability calculation (Vandesompele et al., 2002).

Candidate HK genes were identified utilizing RNAseq database and then annotated with KEGG and WarmBase parasite. After filtering out genes with low count and DEG >0, we identified 48 candidate HK genes (Fig. 1). While only 39 out of the 48 genes were mapped to M. hapla genome, the 39 genes did not have the potential to encode secreted protein (Fig. 1) and are less likely to be directly involved in the plant–nematode interaction. The remaining 39 genes were submitted to KEGG for functional annotation; only 15 out of 39 genes were annotated (Fig. 2), and majority of the genes belong to genetic information processing and metabolism (Fig. 2). The annotation incorporated the results of KEGG and Wormbase was listed in Table A1.

Figure 1:

Figure 2:

Seven out of the 39 candidate HK genes were validated in M. incognita based on high expression level and ease of primer design. Primer specificity was confirmed by single peak on the melt curve. Uninfected tomato plant root samples were not amplified by the primer sets, indicating that all primers were specific to RKN. We used two algorithms to validate suitable HK genes through all developmental stages, including a 24 hr period at 3-wk post inoculation. We determined genes with low stability values as stably expressed genes. Gene stability is slightly different between the two algorithms and throughout the life cycle of RKN.

The geNorm algorithm defines any gene with stability value less than 0.15 as stable genes; RKN genes under the cut-off value were selected as HK genes. According to geNorm, Actin1, Disu, ELF, Poly, Skinase were all suitable HK genes throughout all the developmental stages of RKN (Fig. 3). However, for naïve juveniles, Actin2, Disu, Minc04440, PTP, and Skinase were better reference genes for transcript quantification (Fig. 3). In addition, over a 24 hr period (diurnal) at the adult stage, Actin1, ELF, Poly, Skinase were determined suitable HK genes (Fig. 3). In summary, Disu, Poly, and Skinase had stability values below the threshold for any developmental stage. Normfinder yielded similar yet slightly different results than geNorm. When including all stages, Disu, ELF, Poly, Skinase were suitable reference genes (Fig. 4), but Actin2, Disu, Minc04440, Poly, PTP, and Skinase were stably expressed gene at naïve juvenile stage (Fig. 4). Lastly, Disu, ELF, Poly, and Skinase were better HK genes at the mature stage over the 24 hr period (Fig. 4). Overall, Disu, Poly, and Skinase were suitable HK genes for any developmental stage as suggested by Normfinder. Importantly, our negative control gene, Colla, was highly variable based on RNAseq data, and was also determined to be highly variably expressed throughout all developmental stages by both algorithms.

Figure 3:

Figure 4: