Chromosome-Level Genome Assembly of Protosalanx chinensis and Response to Air Exposure Stress

Simple Summary

In the present study, we assembled a high-quality chromosome-level genome of Protosalanx chinensis, which is the first chromosome-level genome for Salangidae. These genomic data provide a fundamental resource for ecological and adaptation studies of Protosalanx chinensis, and offer a deeper understanding of the response to air exposure stress and species conservation.

Abstract

Protosalanx chinensis is a suitable particular species for genetic studies on nearly scaleless skin, transparency and high sensitivity to hypoxia stress. Here, we generated a high-quality chromosome-level de novo assembly of P. chinensis. The final de novo assembly yielded 379.47 Mb with 28 pseudo-chromosomes and a scaffold N50 length of 14.52 Mb. In total, 21,074 protein-coding genes were predicted. P. chinensis, Esox lucius and Hypomesus transpacificus had formed a clade, which diverged about 115.5 million years ago. In the air exposure stress experiment, we found that some genes play an essential role during P. chinensis hypoxia, such as bhlh, Cry1, Clock, Arntl and Rorb in the circadian rhythm pathway. These genomic data offer a crucial foundation for P. chinensis ecology and adaptation studies, as well as a deeper understanding of the response to air exposure stress.

1. Introduction

Protosalanx chinensis (family Salangidae, order Salmoniformes, Figure 1) is a small annual cold-temperature fish endemic to East Asia and has some specific morphological characteristics, including transparency and scaleless skin [1,2,3,4]. P. chinensis exhibits strong ecological plasticity, with populations found in both freshwater and seawater habitats, including the Yangtze River Basin and its associated lakes (Taihu Lake, Hongze Lake, etc.), as well as the offshore waters of the Yellow Sea, Bohai Sea and East China Sea [1,5,6]. Due to the important economic value of P. chinensis, it was widely translocated to many lakes and reservoirs in northern China in the mid-1980s, and gradually formed a stable population [7]. This deliberate act of artificial translocation has engendered a remarkable shift, resulting in a notable proliferation of P. chinensis across diverse aquatic habitats, effectively amplifying its presence [7]. Studies of the genomic and physiological characteristics of P. chinensis have helped us to better understand the environmental adaptations of P. chinensis.

P. chinensis is a valuable model for studying the molecular mechanisms underlying the evolution of hypoxia. P. chinensis is difficult to obtain alive and can die quickly when stressed by hypoxia during net fishing [1]. Previous studies on the genome of P. chinensis have focused only on phylogeny, sexual differentiation, and skeletal development [2,3]. A basic understanding of stress and the corresponding physiological state of P. chinensis is still lacking. An air exposure stress experiment is an effective experimental tool with which to understand the stress state. An air exposure experiment on gilthead seabream showed that stress-induced hormonal changes affected the liver’s metabolic organization and highlighted the crucial role of vasotocinergic and isotocinergic pathway [8]. A study of rainbow trout demonstrated changes in miRNAs in fish blood during air exposure and identified several miRNA markers [9]. As an economically valuable fish, P. chinensis has not yet been fully cultured in captivity and transported live, so there is potential value in studying its response to hypoxic stress.

With the development of genomic (particularly long-reads) sequencing, two draft genome assemblies of P. chinensis had recently been reported with assembly quality with a contig N50 of 17.2 Kb [2], and a contig N50 of 103 Kb [3], respectively (Protosalanx hyalocranius and Protosalanx chinensis were the same species). Although these two genome drafts provide preliminary genetic information of P. chinensis, these genomes are deficient due to the limitation of sequencing technology. Therefore, chromosome-level genomes and comparative genomics resource are essential to understanding ecological and evolutionary research, translocation adaptation, and genetic improvement.

Here, we generated a high-quality chromosome-level de novo assembly of P. chinensis. A set of protein-coding genes was annotated, and the evolutionary history of P. chinensis was analyzed. In the air exposure stress experiment, the expression pattern of differentially expressed genes (DEGs) was investigated. These genomic data offer a crucial foundation for P. chinensis ecology and adaptation studies, as well as a deeper understanding of the response to air exposure stress and species conservation.

2. Materials and Methods

2.1. Sample Collection and DNA and RNA Sequencing

We collected muscle samples from an adult P. chinensis individual in the Hongze Lake at Jiangsu, China, for sequencing (Figure 1). After the muscle samples were collected, they were rapidly frozen in liquid nitrogen and stored at −80 °C until DNA extraction. DNA was extracted from muscle tissue. RNA was extracted from the larvae. DNA was extracted following the phenol/chloroform DNA extraction method. After the extraction of the DNA and RNA, corresponding quality control was conducted according to different library construction types. The quality control included assessing the concentration, purity, and fragment integrity of the samples.

With the BGI MGISEQ platform, a short insert WGS library was generated according to the manufacturer’s recommendations. A PacBio HiFi library was constructed using a QIAGEN Blood & Cell Culture DNA Midi Kit following the manufacturer’s instructions (QIAGEN, Hilden, Germany) and then sequenced on the PacBio Sequel II system. A Hi-C library was generated using the Mbo I restriction enzyme and sequenced on the BGI MGISEQ platform. We constructed one PacBio HiFi library with an insert fragment size of approximately 15 kb, and one Hi-C library with an insert fragment size of approximately 300 bp.

Fifteen RNA libraries were constructed using the TRIzol Total RNA Isolation Kit (Takara, San Jose, CA, USA) after which the concentration and purity of the extracted RNA were measured to ensure quality. Subsequently, the RNA was fragmented into appropriate lengths using digestion enzymes. This fragmented RNA was then reverse-transcribed to synthesize cDNA, which underwent end repair, addition of specific adapter sequences, and PCR amplification for library construction. Finally, the constructed RNA library was sequenced on the BGI MGISEQ platform.

2.2. Sequencing QC and Genome Assembly

We used the SOAPnuke v2.1.7 [10] pipeline to filter out the low-quality and adaptor reads. After that, we calculated the K-mer (k = 21) frequency distribution with Jellyfish v2.2.6 [11] and analyzed the result using GenomeScope v1.0 [12]. HiFi reads with about 62× coverage were sequenced using the PacBio Sequel platform and cleaned with SMRTLink v8. The contig assembly was carried out using Hifiasm v0.16.1-r375 [13], followed by a removal of the redundant sequences with the Purge-Haplotigs [14] program. Subsequently, the contigs were further connected to the chromosome level using the Juicer v1.5 [15] and 3D-DNA v180922 [16] pipelines. The BUSCO completeness score of the P. chinensis genome was calculated using BUSCO v5.2.2 [17] based on the actinopterygii (odb10) dataset.

2.3. Identification of Repetitive Sequences

We identified the repetitive sequences using a combination of de novo and homolog-based methods. For de novo annotation, we used RepeatModeler v1.0.4 (http://www.repeatmasker.org/RepeatModeler/, accessed date: 23 February 2023) and LTR-FINDER v1.0.7 [18] software to construct a primary library. This customed library was used to screen repeat sequences via the program RepeatMasker v4.0.7 [19]. For the homolog-based prediction, we utilized RepeatMasker v4.0.7 [19], RepeatProteinMasker v4.0.7 [19] and Tandem Repeat Finder v4.10.0 [20] based on the Repbase database.

2.4. Genome Annotation

Gene prediction was conducted through a combination of homology-based prediction, ab initio prediction and transcriptome-based prediction methods. Next, 96.4 Gb RNA-seq data were directly mapped to P. chinensis assembly with Hisat2 v2.1.0 [21] to identify putative exon regions and splice junctions. StringTie v1.3.5 [22] was then used to assemble the mapped reads into gene models and validated using PASA v2.5.2 [23]. Finally, we identified the candidate coding regions by employing TransDecoder v5.5.0 (https://github.com/TransDecoder/TransDecoder, accessed date: 23 February 2023). For homology-based annotation, we downloaded the assemblies and gene annotation files of four actinopterygii species (Danio rerio, Oryzias. latipes, P. hyalocranius and Salmo salar) from the NCBI database. Combined with the above RNA-seq and homolog data, we predicted the homology-like coding sequences using GeMoMa v1.8 [24]. A total of 1200 high-quality coding genes were used to train the predictors using August v3.2.1 [25] and SNAP v2006-07-28 [26] (Korf, 2004) and then ab initio prediction was performed. Lastly, we integrated all the protein-coding genes predicted using the above three strategies with the EVidenceModeler (EVM) pipeline v1.1.1 [23].

2.5. Phylogenetic and Gene Family Analysis

We used OrthoFinder v2.3.11 [27] to cluster protein-coding genes. Single-copy orthologous genes (1:1:1) were aligned using MAFFT v7.310 [28]. Referring to the methods used in previous studies on P. chinensis [3], we used PhyML v3.3 [29] with the HKY85 model to construct a maximum-likelihood phylogenetic tree with 100 pseudoreplicates. All branches had 100/100 bootstrap support, showing phylogeny consistent with a previous study [3]. We estimated the species divergence time using MCMCTREE in PAML v4.9 [30]. Four divergence time points from TimeTree (http://timetree.org.cn, accessed date: 23 February 2023) were used to calibrate the divergence times: (a) Callorhinchus milii and Latimeria chalumnae (421.5–461.6 MYA), (b) L. chalumnae and Lepisosteus oculatus (416.4–422.2 MYA), (c) L. oculatus and Anguilla Anguilla (372.4–383.4 MYA) and (d) **-stone towards unraveling the intricate mechanisms governing the species’ survival and adaptation strategies, further expanding our understanding of the complex interplay between genetics, physiology, and environmental stressors.