An approach using Caenorhabditis elegans screening novel targets to suppress tumour cell proliferation

Tumour cell proliferation requires high metabolism to meet the bioenergetics and biosynthetic needs. Dauer in Caenorhabditis elegans is characterized by lower metabolism, and we established an approach with C elegans to find potential tumour therapy targets.


| INTRODUC TI ON
Cancer remains one of the leading causes of death due to the late diagnosis, poor prognosis, metastasis and frequent occurrence of drug resistance. 1 Usually, the treatment for tumour is surgery, chemotherapy, radiotherapy, complemented with immunotherapy and hormone therapy. 2,3 These various treatments for cancer may have not been able to fully meet the clinical needs, and they frequently bring a variety of side effects to patients. 4,5 Therefore, tumour still needs new treatment strategy.
The most fundamental trait of tumour cells involves their ability to sustain proliferation. 6 In comparison with the normal tissues, high proliferating tumour cells need large amount of energy and biosynthetic precursors of macromolecules as building blocks for new cells. 7 In clinical, patients who carry cancer cells with low proliferation can survive for relative long periods-in some cases indefinitely-without relapsing. 8,9 In addition, many of us may have in situ tumours, like breast, prostate and thyroid cancer, without recognizing it. 10,11 These findings suggest that these microscopic tumours are possibly dormant and need other proliferation signals to become lethal tumours. 10 Evidence suggests that metabolism rewiring can suppress cancer incidence, delay tumour progression and inhibit metastasis. 12,13 As an example, calorie restriction impairs tumour cell proliferation, alters expression of cell cycle proteins, modifies tumour suppressor gene function and disrupts metabolic pathways. 14 Therefore, tumour cell metabolism may be a potential target for tumour therapy.
Dormancy is a special process with extremely low metabolism in nature. 15 But there is no suitable model for screening dormancy-related targets to suppress tumour cell proliferation.
By contrast to tumour cells, dauer, a dormancy in C elegans, which is characterized by their slow growth, maintains extremely low metabolic levels. The C elegans spontaneously enter dauer during the development of L2/L3 in response to adverse environmental conditions, such as high temperature, limiting amounts of food. [16][17][18] Several signalling cascades including insulin-like pathway, TGFβ-like pathway, steroid hormone pathway and guanylyl cyclase pathway are documented to be critical in modulating nematode dauer formation. 16,[19][20][21] In addition, C elegans is a fine model to study tumour. In C elegans, GLP-1 signalling promotes a proliferative germ cell state and prevents germ cells from undergoing meiosis. 22 Thus, loss of GLP-1 signalling results in a severe proliferation defect and early meiotic entry, while constitutive activation yields a germ-line tumour with all germ cells maintaining the germ cells in the stem cell state. 23,24 The expanding germ-line tumour cells eventually perforate the gonad, invade throughout the worm body and lead to early animal death. 25 Thus, tumour mutant in combination with dauer in C elegans may present a valuable model to find the targets suppressing tumour cell proliferation.
Here, we established an approach with C elegans for screening novel targets that rewire tumour cell metabolism and suppress proliferation. We uncovered dauer formation signals could rewire metabolism and extend the lifespan of glp-1(−) mutants in a longevity-or dauer-independent manner.

| Classical dauer formation signals suppressed germ cell proliferation
Next, we asked whether these classical dauer formation signals extend the lifespan via suppressing the germ cell excessive proliferation in glp-1(−) mutants. We knocked down these dauer formation signals in these worms, then dissected the entire gonads at day 4 of adulthood and detected the germ cell number with DNAintercalating dye DAPI. Our results showed that glp-1(−) mutants had more undifferentiated germ cells than glp-1(+) worms (Figure 2A,B).
Furthermore, knockdown of daf-2, daf-1, daf-14, daf-11 and tax-2 resulted in reduced undifferentiated germ cells compared with the counterparts feeding with L4440 in glp-1(−) mutants (Figure 2A,B, Table S2), and all the P values were less than .001. Collectively, these data suggested that activation of classical dauer formation signals suppressed tumour cell proliferation.

| Three novel dauer-related genes extended the lifespan of glp-1(−) mutants
Most of the homologous of classic dauer signals in human had antitumour effect, as shown in Table 1. Hence, we used the glp-1(−) mutants to screen novel genes capable of suppressing tumour cell proliferation. Protein kinases play an indispensable role in regulating cell proliferation and function, which makes them perfect targets for screening tumour-related genes. [31][32][33] Worms encode about 438 kinase-coding genes. 34,35 To screen these genes, an RNAi library contains 287 kinase-coding genes were established ( Figure 3A). We optimized conditions for RNAi library screening and performed the screening (see Figure S1).
Sixty-one genes were identified to promote dauer formation after Interestingly, F47D12.9, W02B12.12 and gcy-21 were first identified to be involved in dauer formation. To confirm our finding, we knock down these three genes, respectively, as a consequence, the survival of glp-1(−) mutants was significantly increased, and all the P values were <.001( Figure 5A-C, Table S3). Furthermore, at day 4 of adulthood, these worms had less germ cell than glp-1(−) mutants feeding with L4440, ( Figure 5D, Table S4), this was associated with the decreased cell proliferation ( Figure 5E), and all the P values were <.001.

| The lifespan extension of dauer-related genes in glp-1(−) mutants was independent of longevity and dauer period
Most of classical dauer formation signals were reported to extend the lifespan of glp-1(+) worms; Kenyon et al 36 had demonstrated that longevity signals could inhibit tumour growth. The glp-1(−) mutants in our study were in the background of N2 (glp-1(+)) worms. To test whether these three dauer-related genes have inherent longevity effect, we knockdown these genes in glp-1(+) worms, respectively, and no longevity effect was observed ( Figure 6A-F, Table S5), which suggested the lifespan extension in glp-1(−) mutants was independent of longevity.

| Dauer-related genes rewired metabolism pathways in the glp-1(−) mutants
To investigate the molecular mechanisms underlying the antitumour effect of dauer-related genes, we compared their transcriptome profiles in worms. As shown in Figure S2A Next, we analysed the differentially expressed genes. RNArelated metabolism and processing were found to be significantly F I G U R E 2 Classical dauer formation signals suppress germ cell proliferation in glp-1(−) mutants. Adult animals were stained with the DNA-intercalating dye DAPI. Left panels, the whole gonad. Right panels, midpoints of the gonad arms. A, glp-1(−) mutants lack oocytes and have many undifferentiated germ cells in their gonads. Knockdown of daf-2, daf-1, daf-14, daf-11 and tax-2 by RNAi, respectively, they have far fewer undifferentiated germ cells and maintain the integrity of their gonads. Representative of n = 3 independent experiments. B, Knockdown of classical dauer formation signals reduced undifferentiated germ cells was detected and analysed using one-way ANOVA test followed by Bonferroni correction for post hoc test (*P < .05,**P < .01 for t test); data shown are mean ± SD

| NPR1/gcy-21 and TSSK6/W02B12.12 were unfavourable prognostic indicators in glioma patients
Gcy-21, F47D12.9 and W02B12.12 are homologous to NPR1, DCAF4L2 and TSSK6 in human, respectively. 40 To translate our findings from C elegans models to human individuals, we first investigated whether the antitumour effect of these three genes was parallel in human. Using the TCGA data sets, we found that NPR1, DCAF4L2 and TSSK6 were overexpressed in glioma samples compared with adjacent normal tissues in human ( Figure 9A). In Figure 9B, the cluster analysis revealed that NPR1, DCAF4L2 and TSSK6 were significantly upregulated in classical and mesenchymal subtypes of glioma.
Next, we evaluated the prognostic values of NPR1, DCAF4L2 and

| D ISCUSS I ON
The primary characteristic of tumour cells is highly proliferation and active energy metabolism. 6 Previous studies indicated that reduced metabolism, such as calorie restriction, could inhibit tumorigenesis and progression. 12,14 Dormancy is a special process with extremely low metabolism in nature. Based on these two opposite metabolism states, we proposed that the activation of dormancy could suppress tumour progression. By using C elegans, we found that related Intriguingly, our approach screened out three novel dauer-related genes, namely F47D12.9, W02B12.12 and gcy-21, which could significantly extend the lifespan of glp-1(−) mutants via suppressing tumour cell proliferation. The functions of these three genes in C elegans were not yet clear and these genes were conserved in human beings. We found that their homologous genes NPR1, DCAF4L2 and TSSK6 were highly expressed in glioma patients and patients with high NPR1 and TSSK6 predicted a short survival.
Studies reported that NPR1/gcy-21 is the receptor for the cardiac hormone atrial natriuretic peptide, and it has been reported to be highly expressed in cancer cells such as lung cancer, prostate cancer and ovarian cancer. 54 The high expression and signalling of NPR1 are important for tumour growth; its deficiency protects C57BL/6 from lung, skin and ovarian cancers. 54 Nojiri et al 55  is a member of the serine/threonine protein kinase family, and it has been described as testis-specific due to effects on fertility. 57  however, P53 caused progeria in mouse. 60 Our results suggested NPR1/gcy-21 and TSSK6/W02B12.12 might be potential targets for glioma therapy via suppressing tumour cell proliferation ( Figure 5).
As an example, NPR1 interacts with atrial natriuretic peptide, which is an endogenous and physiological peptide, and atrial natriuretic peptide has significant antitumour effect in multiple cancers such as pancreatic cancer, breast cancer, small cell lung cancer, prostate cancer and colorectal cancer. 61,62 Notably, NPR1 and TSSK6 were highly correlated with the unfa- Overall, few molecular biological works have been performed for the TA B L E 2 The homologous genes of dauer-related genes in regulating tumour cell proliferation and metabolism regulation of these two genes till now; the detailed mechanisms are under investigation. These findings just provide clues for glioma target therapy, which requires a lot of work to verify its function and reveal its mechanisms in human glioma cell and mouse.
In this study, we found that dauer-related genes could extend the lifespan of glp-1(−) mutants and this extension was independent of dauer stage and longevity (Figures 6 and 7), which suggests that and W02B12.12 reduced undifferentiated germ cells was detected and analysed using one-way ANOVA test followed by Bonferroni correction for post hoc test (*P < .05,**P < .01 for t test); data shown are mean ± SD across evolutionary time (metabolism, organelle structure and function, gene regulation, protein biology, etc) have made C elegans an excellent organism with which to study general metazoan biology. [67][68][69] But no model organism can be used to answer all the research question, and working with C elegans has some limitations, for instance, C elegans lacks many genes that regulate cascades. 70
Nematodes were cultured at 20°C, except as noted below for temperature-sensitive.

| Worm RNAi by feeding
RNAi was performed essentially as per. 71 Single colonies of HT115 bacteria containing L4440 plasmids with cloned fragments corresponding to target genes were from RNAi feeding libraries. Each RNAi reagent was verified by DNA sequencing except for screening.
Young adult hermaphrodites were placed onto NGM plates seeded with dsRNA-expressing or empty vector control bacteria (RNAi feeding plate). After overnight incubation, worms were transferred to an identical fresh RNAi feeding plate and allowed to lay eggs for 2 hours.
F I G U R E 8 Dauer-related genes rewired metabolism in glp-1(−) mutants. A, KEGG pathway enrichment analysis of the identified differential expressed genes. Heatmap represents the differential expressed genes enriched in related pathway. The red colour represents the upregulated genes, and the blue colour represents the downregulated genes. The deeper the colour, the more genes enrich in related genes. Heatmap using enriched genes set with options: z-score based row and column normalization

| Lifespan analysis
Lifespan was measured following previous protocols. 36 For lifespan assays, we picked about 90 animals as L4 larvae at t = 0, and we transferred worms to OP50-seeded or RNAi feeding plates (roughly 30 animals per plate) approximately every other day until the end of the reproductive period. Dead worms were counted every other day until all worms were dead. Worms that were missing or that died from extruded internal organs or from internally hatched progeny were censored and statistically incorporated into the lifespan analyses. All lifespan experiments were performed at 25°C. We generated lifespan curves using GraphPad Prism 7, and we determined P values using the Mantel-Cox log-rank test. Each lifespan experiment was repeated at least three times.

| Fluorescence microscopy
For DAPI nuclear staining, gonads were dissected, fixed and stained with DAPI as described. 36

| Quantitative RT-PCR
More than 400 well-fed synchronized worms on day 1 were collected into 1.5-mL tube and washed at least three times with M9 buffer.
Total RNA was extracted with Trizol reagent (Invitrogen act-1 as control, and product size was 120 bp. All these genes were annealed at 60°C. The folds of changes were shown as means ± SD.
in three independent experiments with each triplicate.

| Library preparation for Transcriptome sequencing
A total amount of 3 μg RNA per sample was used as input material for the RNA sample preparations. Sequencing libraries were generated using NEBNext ® UltraTM RNA Library Prep Kit for Illumina ® (NEB) following the manufacturer's recommendations and index codes were added to attribute sequences to each sample. PCR products were purified (AMPure XP system) and library quality was assessed on the Agilent Bioanalyzer 2100 system.

| Data filtering
Raw data (raw reads) of fastq format were firstly processed through in-house perl scripts. The raw data obtained by sequencing contain a small number of reads with sequencing linker or low quality sequencing. In order to ensure the quality and reliability of data analysis, the original data need to be filtered. The content of the filtering is as follows. Remove low quality reads (Qphred ≤ 20 the number of bases in the read length face = 50% above reads). In this step, clean data (clean reads) were obtained by removing reads containing adapter, reads containing ploy-N and low quality reads from raw data. At the same time, the contents of Q20, Q30 and GC in the clean data were calculated. All the downstream analyses were based on the clean data with high quality.
Differential expression analysis of two conditions/groups (two biological replicates per condition) was performed using the DESeq2 R package (1.16.1). DESeq2 provides statistical routines for determining differential expression in digital gene expression data using a model based on the negative binomial distribution. The resulting P-values were adjusted using the Benjamini and Hochberg's approach for controlling the false discovery rate. Genes with an adjusted P-value < .05 found by DESeq2 were assigned as differentially expressed.

| GO and KEGG enrichment analysis of differentially expressed genes
Gene Ontology (GO) enrichment analysis of differentially expressed genes was implemented by the clusterProfiler R package, in which gene length bias was corrected. GO terms with corrected P value <.05 were considered significantly enriched by differential expressed genes.
KEGG is a database resource for understanding high-level functions and utilities of the biological system, such as the cell, the organism and the ecosystem, from molecular-level information, especially large-scale molecular datasets generated by genome sequencing and other high-through put experimental technologies (http://www.genome.jp/kegg/). We used clusterProfiler R package to test the statistical enrichment of differential expression genes in KEGG pathways.

| Sample collection
We included all available tumour samples from TCGA (https://tcgad ata.nci.nih.gov). In addition, 77 glioma tumour samples and three non-neoplastic normal brain tissues were obtained from Fudan

| Immunohistochemistry
The immunohistochemical analysis was performed on the 4-mmthick fraction mounted on slides and sectioned from each tumour sample. Then, each slide was deparaffinized in 60°C, followed by treatment with xylene and graded alcohol. After the antigen retrieval and being blocked with 5% bovine serum albumin, tissue slides were incubated with antibodies overnight at 4°C against NPR1 multiplying the intensity score with the extent of score of stained cells, ranging from 0 (the minimum score) to 12 (the maximum score).

ACK N OWLED G EM ENTS
This work is supported by grants from National Natural Science

Foundation of China (81872245 and 81803601), Fundamental
Research Funds for Minhang Hospital (2020MHJC12) and grants from Shanghai Sailing Program (17YF1416700).

CO N FLI C T O F I NTE R E S T
No conflict of interest was declared.

DATA AVA I L A B I L I T Y S TAT E M E N T
The authors declare that all the data supporting the findings of this study are available within the article and its Supplementary Information files and from the corresponding authors on reasonable request.