Comparative transcriptomics suggests a highly species-specific nature of the phenotypic plasticity associated with the outbreaks of the two main pest locusts

Abstract

Background: Locust outbreaks cause devastation and provide material for fundamental research. They associate with a case of phenotypic plasticity whereby the shift between the two extremes of the polyphenism (i.e., gregarious phase versus solitarious phase) affects behaviour as well as most aspects of the locusts' biology. The phenotypic changes imply changes in gene expression, the changes in behaviour characterize the locusts' phase change, and the changes in the Central Nervous System (CNS) control the changes in behaviour. Thus, understanding and tackling the phenomenon requires studying the gene expression changes that the locusts' CNS undergoes between phases. The genes that change expression the same way in different locusts would be ancestrally relevant for the phenomenon in general and some of those that change expression in a species-specific way would be relevant for the phenomenon in species-specific way.

Methods: Here, we use available raw sequencing reads to build transcriptomes and to compare the gene expression changes that the CNS of the two main pest locusts (Schistocerca gregaria and Locusta migratoria) undergo when they turn gregarious. The differentially expressed genes resulting from this comparative study were compared with the content of the L. migratoria core transcriptional phase signature genes database. Our aim is to find out about the species-specificity of the phenomenon, and to highlight the genes that respond in the same way in both species.

Results: The locust phase change phenomenon seems highly species-specific, very likely due to the inter-specific differences in the material used, and in the biology and life conditions of the different locust species. Research on locust outbreaks, gregariousness and swarming would therefore benefit from considering each locust species apart, and caution is needed when extrapolating results between species-as no species seems representative of all locust species. Still, the 109 genes and 39 non-annotated sequences that we found to change expression level the same way in the two main pest locusts, especially those previously reported as core transcriptional phase signature genes in L. migratoria's CNS-related tissues (10 and 1, respectively), provide material for functional testing in search for important genes to better understand, or to fight against locust outbreaks in a non-species-specific way. The large set of genes that respond in a species-specific way provide material for comparing, understanding and tackling the locust's phase change phenomenon in a species-specific way. The still uncharacterized transcripts that change expression either in a species-specific or the same way between the two species studied here provide material for gene discovery. Functional testing and confirmation are needed in all cases.

Keywords: Locusta migratoria; Schistocerca gregaria; Comparison; Gregarious; Locust; Outbreak; Solitarious; Transcriptome.

Fig. 1

Heatmaps of the overall expression levels of the common expressed (A) common blasted differentially expressed (B) and common non-annotated differentially expressed genes (C) of the solitarious and gregarious Locusta migratoria and Schistocerca gregaria. LmS: Solitarious L. migratoria, LmG: Gregarious L. migratoria, SgS: Solitarious S. gregaria, SgG: Gregarious S. gregaria, Lm_Exp: Average solitarious and gregarious gene expression in L. migratoria, Sg_Exp: Average solitarious and gregarious gene expression in S. gregaria, Exp_Cons: Degree of conservation of the average solitarious and gregarious gene expression level between L. migratoria and S. gregaria, Dif_Cons: Degree of conservation of the solitarious versus gregarious differential gene expression level between L. migratoria and S. gregaria

Fig. 2

The biological processes that contain significantly different numbers of sequences in the Locusta migratoria and Schistocerca gregaria reference transcriptomes compared in this work—as by Fisher’s exact test. Y axis shows the proportion that the biological process represents among all the processes (number of sequences of a transcriptome that belong to a specific biological process/total number of sequences of the same transcriptome that belong to any biological process)

Fig. 3

The biological processes with 30 or over genes to which belong the sequences of the part of the Locusta migratoria and Schistocerca gregaria reference transcriptomes used here that have the same best BLAST hit in both transcriptomes

Fig. 4

Biological processes to which belong the genes that are significantly different between the solitarious and gregarious states both in Locusta migratoria and Schistocerca gregaria and that contain over one gene

Fig. 5

Biological processes that contain more than one gene and to which belong the genes that increase their levels of expression in the gregarious phase of both Locusta migratoria and Schistocerca gregaria

Fig. 6

Gene network analysis of the genes that show significant change in expression levels between the solitarious and the gregarious phases both in Locusta migratoria and in Schistocerca gregaria. A: Our gene list against all the String databases at the highest confidence level (0.9). B: Our gene list against all the String databases the high confidence level (0.7). C: Our gene list plus the first 10 genes shell in all the String databases at high (0.7) confidence level