As the deepest vertebrate in the ocean, the hadal snailfish (Pseudoliparis swirei), which lives at a depth of 6,000~8,000m, is a representative case for studying adaptation to extreme environments. Despite some preliminary studies on this species in recent years, including their loss of pigmentation, visual and skeletal calcification genes, and the role of TMAO in adaptation to high-hydrostatic pressure, it is still unknown how they evolved and why they are among the few vertebrate species that have successfully adapted to the deep sea environment. Using genomic data from different trenches, we found that the hadal snailfish may have entered and fully adapted to such extreme environments only in the last few million years. Meanwhile, phylogenetic relationships show that they spread into different trenches in the Pacific Ocean within a million years. Comparative genomic analysis has also revealed that the genes associated with perception, circadian rhythms, and metabolism have been extensively modified in the hadal snailfish to adapt to its unique environment. More importantly, the tandem duplication of a gene encoding ferritin significantly increased their tolerance to reactive oxygen species, which may be one of the important factors in their adaptation to high-hydrostatic pressure.
This important study advances our understanding of the potential mechanisms of deep-sea adaptation and sheds light on the evolutionary history of hadal snailfish. Through comparative genomic analysis, the authors provide convincing evidence and propose hypotheses on the timing of trench colonization, population structure, and adaptations to the hadal snailfish genome in response to their environment.
Since its capture in 2014 the hadal snailfish (Pseudoliparis swirei) has captured the attention of biologists and public as the deepest known vertebrate on Earth (Fujii, Jamieson, Solan, Bagley, & Priede, 2010; Mackenzie E. Gerringer, Linley, Jamieson, Goetze, & Drazen, 2017; Mackenzie E. Gerringer, Linley, & Nielsen, 2021). The hadal zone, 6 kilometers below the sea surface where this species lives, and after which the species is named, is characterized by high-hydrostatic pressure (HHP), complete darkness, barrenness (A. Jamieson, 2015; A. J. Jamieson, 2011). This special organism that survives and thrives in the hadal zone provides us with an unusual case of adaptation to an extreme environment. After several years of anatomical and genomic research (M. E. Gerringer, 2019; M. E. Gerringer et al., 2017; K. Wang et al., 2019), it’s now known that the hadal snailfish has degraded vision and a lack of melanin in its skin (K. Wang et al., 2019). Previous studies have also revealed that having more unsaturated fatty acids in the cell membrane (Cossins & MacDonald, 1984; Fang, Barcelona, Nogi, & Kato, 2000) and high levels of trimethylamine N-oxide (TMAO) in the body (P. H. Yancey, Gerringer, Drazen, Rowden, & Jamieson, 2014) may have played important roles in resistance to HHP. However, much remains to be discovered about this species; our lack of knowledge can be categorized into three principal areas.
The first concerns the origin of hadal snailfish. They have been observed in several Trenches in the Northwest Pacific Ocean, including the Mariana (K. Wang et al., 2019), Yap (Mackenzie E. Gerringer et al., 2021), Kuril–Kamchatka (M. E. Gerringer et al., 2017) and Japan (Mackenzie E. Gerringer et al., 2021) Trenches. The question arises: Do they have the ability to migrate across trenches, or do they enter the hadal zone independently? Furthermore, it has been shown that the divergence time between hadal snailfish and Tanaka’s snailfish (a closely related species distributed in shallow areas) is about 20 million years ago (Mya) (K. Wang et al., 2019); but when did the hadal snailfish enter the hadal zone and how long did they take to complete their adaptation to this ecological niche? The second aspect concerns its morphological and physiological characteristics. For example, in such a dark environment, what do hadal snailfish rely on to sense the world: is it smell, taste or something else? Do they still have circadian rhythms in the absence of sunlight? Does darkness and HHP have any effect on their behavior? The last area concerns the mechanisms by which they tolerate HHP. If unsaturated fatty acids and TMAO are common in marine fish, why have only a few species such as hadal snailfish been observed to reach such depths? Some studies suggest that certain genetic alterations may confer tolerance to HHP (K. Wang et al., 2019), but if so, how do these alterations help hadal snailfish to adapt to this environment, and which alterations are most critical?
Unfortunately, these questions have not been well resolved because it is difficult for us to make long-term observations of deep-sea organisms in situ. During 2018 to 2019, we collected multiple samples of hadal snailfish from the Mariana Trench and Tanaka’s snailfish from the Yellow Sea. Based on more data and more refined genome, we have been able to trace the genomic signals left by adaptive evolution in an attempt to more fully understand the evolutionary processes and key changes in this special organism.
Improved genome assembly for Mariana hadal snailfish
A total of four hadal snailfish (Pseudoliparis swirei) and four Tanaka’s snailfish (Liparis tanakae) individuals were collected for this study (Supplementary file 1). Using a combination of ONT long reads, BGI short reads, and Hi-C sequencing technologies, we generated a chromosome-level genome assembly for hadal snailfish (Supplementary file 2). The genome assembly comprised 1,173 contigs (total length = 626.44 Mb, contig N50 = 4.22Mb), organized into 24 chromosomes with an anchoring rate of 98.24%. The new assembly filled 1.26 Mb of gaps that were present in our previous assembly and have a much higher level of genome continuity and completeness (with complete BUSCOs of 96.0 % [Actinopterygii_odb10 database]) than the two previous assemblies (Figure 1-figure supplements 1 and 2; Supplementary file 3 and 4) (Mu et al., 2021; K. Wang et al., 2019). Moreover, the genome redundancy caused by mis-assembly is also largely reduced in the new assembly (Figure 1-figure supplements 3), which ensures the reliability of the subsequent analysis. Meanwhile, we generated a high-quality chromosomal level genome assembly for Tanaka’s snailfish for a comparative evolutionary study. We noticed that there is no major chromosomal rearrangement between hadal snailfish and Tanaka’s snailfish, and chromosome numbers are consistent with the previously reported MTZ-ancestor (the last common ancestor of medaka, Tetraodon, and zebrafish) (Kasahara et al., 2007), while the stickleback had undergone several independent chromosomal fusion events (Figure 1-figure supplements 4).
Based on the new genome assemblies, we re-examined the genetic changes that occurred in the common ancestor of hadal snailfish in combination with the new resequencing and transcriptome data. After a thorough scan and careful inspection, we identified 51 absent genes, 20 unitary pseudogenes, 21 lineage-specific expanded genes, 33 genes with insertions and deletions (with a length ≥ 3 amino acids) in coding regions, and 33 de novo originated new genes (Supplementary file 8-12). Most of them have not been previously reported and we discuss them in the following sections.
Cross-trench distribution and high level of genetic diversity
Combining the eight new sequenced individuals (four hadal snailfish and four Tanaka’s snailfish) with five previously reported individuals (four hadal snailfish and one Tanaka’s snailfish), we have been able to form an initial perspective of the hadal snailfish at the population level. The principal component analysis (PCA), neighbor-joining tree and genetic clustering analysis shows that the eight hadal snailfish individuals can be divided into two populations, the first with seven individuals and the second with one individual (Figure 1A-C; Supplementary file 1). Interestingly, the first population includes samples from both the Mariana and Yap trenches. Using the mitochondrial data, we found that the divergence time of these individuals from different trenches appears to be only about 44,000 years (Figure 1-figure supplements 5). Combined with additional publicly available mitochondrial data, we noticed that the sample from the Kermadec Trench (M. E. Gerringer et al., 2017), about 6,400 kilometers away from the Mariana Trench, is also clustered with individuals from the first population, and the divergence time was estimated to be 1.0 Mya (Figure 1D; Figure 1-figure supplements 6). These results suggest that hadal snailfish have successfully spread to multiple trenches in the Pacific Ocean over the course of a million years. And this dispersal may have been caused by population expansion or deep circulation.
Genetic diversity of hadal snailfish is about 3.48 times higher than Tanaka’s snailfish. The FST between the two species is close to 0.91, indicating a large genetic divergence (Figure 1-figure supplements 7). That most polymorphisms are unique to each species, and that they share only 6% of their SNPs, is consistent with this observation. In addition, we also estimated the demographic history of hadal snailfish at the population level and observed a significant expansion in the last 60,000 years (Figure 1-figure supplements 8).
Preserved rh1 gene suggests rapid adaptation to hadal zone
Based on the mitochondrial data, the closest known species related to the hadal snailfish were found to be from the genera Careproctus, Crystallias, Rhodichthys and Paraliparis, which contain many species living at approximately 1,000 m depth (Figure 1D; Supplementary file 13). The divergence time between hadal snailfish and these species was estimated to be about 9.9 Mya, close to the formation time (8 Mya) of the deepest trough of the Mariana Trench (Oakley, Taylor, Moore, & Goodliffe, 2009). We might therefore speculate that the ancestor of hadal snailfish adapted to the deep-sea environment around 1,000 m at about 9.9 Mya, and subsequently gradually adapted to greater depths in the formed or forming trench. The upper and lower limits of the time for hadal snailfish to enter the hadal zone were estimated to be 9.9 Mya (divergence time between hadal snailfish and its closest relatives) and 1.0 Mya (divergence time between different hadal snailfish individuals), respectively. Interestingly, the vision-related genes also confirm a rapid adaptation to hadal zone.
Fish that inhabit different depths of the sea rely on different vision-related genes (Musilova et al., 2019). Since light with longer wavelengths is absorbed more quickly than those with shorter wavelengths, (except for the shortest UV wavelengths) high energy light with shorter wavelengths, such as blue, is able to penetrate to greater depths (Figure 2A). The genes responsible for absorbing these shorter wavelengths (sws2 and rh2) are therefore much more important to deep-sea species in the photic zone (above 200m) than those that absorb longer wavelengths (lws). Similarly, the genes providing monochromatic vision in very dim light (rh1 and gnat1) have been proven to be important for deep-sea species of the disphotic zone (from 200m to 1,000m) (Musilova et al., 2019). We noticed that the lws gene (long wavelength) has been completely lost in both hadal snailfish and Tanaka’s snailfish; rh2 (central wavelength) has been specifically lost in hadal snailfish (Figure 2B and 2C); sws2 (short wavelength) has undergone pseudogenization in hadal snailfish (Figure 2-figure supplements 1); while rh1 and gnat1 (perception of very dim light) is both still present and expressed in the eyes of hadal snailfish (Figure 2D). A previous study has also proven the existence of rhodopsin protein in the eyes of hadal snailfish using proteome data (Yan, Lian, Lan, Qian, & He, 2021). The preservation and expression of genes for the perception of very dim light suggests that they are still subject to natural selection, at least in the recent past.
Highly expressed auditory genes
Do hadal snailfish compensate for the lack of vision when perceiving the external environment? The genes associated with the olfactory and auditory systems were investigated using both comparative genomic and transcriptomic methods. While the number of olfactory receptors was largely reduced (Figure 3-figure supplements 1), we found that the majority of the auditory genes were well preserved in hadal snailfish. Many of the auditory genes also tended to be significantly more upregulated in the brain of hadal snailfish than in Tanaka’s snailfish (Figure 3A; Supplementary file 14). The upregulated genes involve many aspects of the auditory system, including the development and tethering of otoliths (Kang, Stevenson, Yau, & Kollmar, 2008; Stooke-Vaughan, Obholzer, Baxendale, Megason, & Whitfield, 2015), the development (Iyer & Groves, 2021; Kozlowski, Whitfield, Hukriede, Lam, & Weinberg, 2005; Riley, 2021; L. Wang et al., 2008), maturation and maintenance of inner ear hair cells, the development and mechanosensitivity of stereocilia (Cirilo, Gunther, & Yengo, 2021; Kitajiri et al., 2010), and other factors (Giffen, Liu, Kramer, & He, 2019; Verdoodt, Van Camp, Ponsaerts, & Van Rompaey, 2021). (Figure 3A). Of these, the most significant upregulated gene is tmc1, which encodes transmembrane channel-like protein 1, involved in the mechanotransduction process in sensory hair cells of the inner ear that facilitates the conversion of mechanical stimuli into electrical signals used for hearing and homeostasis (Maeda et al., 2014), and some mutations in this gene have been found to be associated with hearing loss (Kitajiri, Makishima, Friedman, & Griffith, 2007; Riahi et al., 2014). Interestingly, tmc1 is also found to be the gene with the longest deletion specific to hadal snailfish (11 amino acids) in the regions that are generally highly conserved across vertebrate’s genomes (Figure 3-figure supplements 2), the functional implications of this alteration need further verification.
Moreover, the gene involved in lifelong otolith mineralization, cldnj, has three copies in hadal snailfish, but only one copy in other teleost species, encodes a claudin protein that has a role in tight junctions through calcium independent cell-adhesion activity (Figure 3B, Figure 3C) (Hardison, Lichten, Banerjee-Basu, Becker, & Burgess, 2005). This may be important for hadal snailfish because calcium carbonate, the inorganic component in otoliths, is thought not to accumulate efficiently below the carbonate compensation depth (CCD; > 4,000 m - 5,000 m) (A. Jamieson, 2015). It should be noted that the hadal snailfish survive at depths far beyond the limits of CCD, but their otoliths still maintain densities similar to those of sea-surface species (M. E. Gerringer et al., 2021). In our investigation, we found that the expression of cldnj was not significantly up-regulated in the brain of the hadal snailfish than in Tanaka’s snailfish, which may be related to the fact that cldnj is mainly expressed in the otocyst, while the expression in the brain is lower. However, due to the immense challenge in obtaining samples of hadal snailfish, the expression of cldnj in the otocyst deserves more in-depth study in the future. Expansion of cldnj was observed in all resequenced individuals of the hadal snailfish (Supplementary file 10), which provides an explanation for the hadal snailfish breaks the depth limitation on calcium carbonate deposition and becomes one of the few species of teleost in hadal zone.
Circadian rhythm decoupled from sunlight and dark adaptation
There is growing evidence that persistent darkness challenges the physiology and behavior of animals, leading to disrupted circadian rhythms, neurological damage and depressive-behavioral phenotypes (Fisk et al., 2018). Consistent with previous research in cavefish (Policarpo et al., 2021), we noticed that many of the circadian rhythm genes (per2a, cry1a, cry3, cry5 and gpr19) are lost or have undergone pseudogenization in the hadal snailfish (Figure 4A and B; Figure 4-figure supplements 1). Despite that, we noticed that the essential clock control genes are present and expressed in the hadal snailfish, indicating the rhythm cycle is retained, although it is likely to have been largely uncoupled from sunlight. Moreover, gpr19 deficiency has been reported to prolong the cycle of circadian locomotor activity rhythms (Yamaguchi et al., 2021), so hadal snailfish may have an extended rhythm cycle like cavefish (Cavallari et al., 2011; Yamaguchi et al., 2021).
In addition, in the teleosts closely related to hadal snailfish, there are usually two copies of grpr encoding the gastrin-releasing peptide receptor; we noticed that in hadal snailfish one of them is absent and the other is barely expressed in brain (Figure 4C), whereas a previous study found that the grpr gene in the mouse suprachiasmatic nucleus (SCN) did not fluctuate significantly during a 24-hour light/dark cycle and had a relatively stable expression (Pembroke, Babbs, Davies, Ponting, & Oliver, 2015) (Figure 4-figure supplements 1). It has been reported that grpr deficient mice, while exhibiting normal circadian rhythms, show significantly increased locomotor activity in dark conditions (Wada et al., 1997; Zhao et al., 2023). We might therefore speculate that the absence of that gene might in some way benefit the activity of hadal snailfish under complete darkness.
It should be noted that the above-mentioned missing genes are not sufficient to exhibit the full range of changes that occur in the nervous system of hadal snailfish. Previous studies suggest that HHP suppressed the compound action potential in nerve trunks of fishes from shallow areas but not from deep areas, and can perturb the function of G protein-coupled receptors (Siebenaller & Murray, 1995). From our transcriptome data, we also observed that the brain is one of the most divergent organs regarding expression levels between hadal snailfish and Tanaka’s snailfish (Figure 4-figure supplements 2). Specifically, there are 3,587 up-regulated genes and 3,433 down-regulated genes in the brain of hadal snailfish compared to Tanaka snailfish, and Gene Ontology (GO) functional enrichment analyses revealed that up-regulated genes in the hadal snailfish are associated with cilium, DNA repair, and microtubule-based movement, while down-regulated genes are enriched in membranes, GTP-binding, proton transmembrane transport, and synaptic vesicles (Supplementary file 15). In line with this observation, one of our previous studies showed that zebrafish brains have the highest number of differentially expressed genes than the other investigated organs when exposed to HHP (M. L. Hu et al., 2022). We also identified 15 de novo new genes in hadal snailfish that are highly expressed in the brain (Figure 4-figure supplements 2). The adaptation of the nervous system to HHP deserves more in-depth study in the future.
Possible survival strategy of storing energy
In a previous study, it was noticed that the individual hadal snailfish we investigated retained a large amount of intact food in its stomach and had larger eggs than might otherwise be expected (M. Gerringer, Popp, Linley, Jamieson, & Drazen, 2017; K. Wang et al., 2019). It appears that the hadal snailfish have a survival strategy of storing energy, which is often found in species that need to cope with occasional starvation. Here we find another clue that hints at the existence of this possibility: the pseudogenization of the gene gpr27 in hadal snailfish (Figure 4-figure supplements 3). Gpr27 is a G protein-coupled receptor, belonging to the family of cell surface receptors, involved in various physiological processes and expressed in multiple tissues including the brain, heart, kidney, and immune system. It has been reported that the knockout of gpr27 increases the expression of key enzymes in the carnitine shuttle complex (Nath et al., 2020), especially cpt1, which is essential for the beta-oxidation of lipid metabolism. The transcriptome data further confirm that the gene cpt1 is significantly upregulated in the liver of hadal snailfish. As lipid mobilization is thought to be a common metabolic response to short-term starvation in fish (Liao et al., 2017), the inactivation of gpr27 could help hadal snailfish to better survive periods of food deficiency. Although previous surveys have shown that various types of organisms live in the hadal zone, and that the hadal snailfish can survive by eating amphipods and occasionally polychaetes and decapod shrimp (Yan et al., 2021), short-term starvation is still possible because although energy sources are limited by complete darkness, but organisms usually tend to “over-reproduce”.
Reduced bone mineralization
Vitamin D synthesis is dependent on UV light, with phytoplankton being the origin of vitamin D in food (Björn & Wang, 2000). Whether and how vitamin D reaches the hadal zone through various pathways, for instance as particulate organic matter, is still unknown. By investigating the genes associated with vitamin D metabolic pathways, we found that these genes are well conserved in the genome of hadal snailfish and are similarly expressed in both hadal snailfish and Tanaka’s snailfish (Figure 4-figure supplements 4), suggesting that vitamin D may not be a limiting factor for hadal zone vertebrates.
Nonetheless, micro-CT scans have revealed shorter bones and reduced bone density in hadal snailfish, from which it has been inferred that this species has reduced bone mineralization (M. E. Gerringer et al., 2021); this may be a result of lowering density by reducing bone mineralization, allowing to maintain neutral buoyancy without expending too much energy, or it may be a result of making its skeleton more flexible and malleable, which is able to better withstand the effects of HHP. The gene bglap, which encodes a highly abundant bone protein secreted by osteoblasts that binds calcium and hydroxyapatite and regulates bone remodeling and energy metabolism, had been found to be a pseudogene in hadal fish (K. Wang et al., 2019), which may contribute to this phenotype. Here, we found two more lost genes specific to hadal snailfish, tmem251 and tmem263, that contribute to reduced bone mineralization (Figure 4D; Figure 4-figure supplements 5). These two genes encode transmembrane proteins, and loss-of-function mutations have now been found that may affect bone mineral deposition and thus bone development and body growth (Ain et al., 2021; Z. Wu et al., 2018). Furthermore, many genes that determine chondrocyte differentiation and bone mineralization were found to be differentially expressed in the bones of hadal snailfish and Tanaka’s snailfish. However, it should be noted that this result derives from a single bone sample of a hadal snailfish and needs further verification.
High-hydrostatic pressure adaptation at cellular levels
High-hydrostatic pressure exerts broad effects upon cells, including cell membrane fluidity (Casadei, Manas, Niven, Needs, & Mackey, 2002; Chong, Cossins, & Weber, 1983; Kato, Hayashi, Tsuda, & Taniguchi, 2002), protein structure stability (Abe, 2021; Gross & Jaenicke, 1994), and oxidative stress (Aertsen, De Spiegeleer, Vanoirbeek, Lavilla, & Michiels, 2005; Moserova et al., 2017). In regard to the effect of cell membrane fluidity, relevant genetic alterations had been identified in previous studies, i.e., the amplification of acaa1 (encoding acetyl-CoA acetyltransferase 1, a key regulator of fatty acid β-oxidation in the peroxisome, which plays a controlling role in fatty acid elongation and degradation) may increase the ability to synthesize unsaturated fatty acids (Fang et al., 2000; K. Wang et al., 2019). As for the stability of the protein structure, previous studies have suggested that the high level of trimethylamine N-oxide (TMAO) content could help the marine fishes in resistance to the inhibitory effects of high pressure on numerous proteins (Ma, Pazos, & Gai, 2014; Paul H Yancey, Blake, & Conley, 2002; P. H. Yancey et al., 2014). We also observed another gene duplication event associated with protein stability. The gene vbp1 (Figure 5-figure supplements 1) (Vainberg et al., 1998), encoding prefoldin subunit 3 that promotes protein folding, has two copies in hadal snailfish but one copy in other teleost fishes. But unfortunately, although it is widely known that high pressure leads to the accumulation of reactive oxygen species (ROS) (Abe, 2021; Aertsen, Van Houdt, Vanoirbeek, & Michiels, 2004; Le et al., 2020), it is still unknown how deep-sea fish cope with this challenge.
We further examined the known ROS-related genes in hadal snailfish, but found that they were not significantly altered in sequence or expression (Figure 5-figure supplements 1). Next, we identified 34 genes that are significantly more highly expressed in all organs of hadal snailfish in comparison to Tanaka’s snailfish and zebrafish, while only seven genes were found to be significantly more highly expressed in Tanaka’s snailfish using the same criterion (Figure 5-figure supplements 1). The 34 genes are enriched in only one GO category, GO:0000077: DNA damage checkpoint (Adjusted P-value: 0.0177). Moreover, five of the 34 genes are associated with DNA repair. Interestingly, however, when we analyzed the genes that were both expanded and highly expressed in most tissues, we identified only one gene, fthl27 (encoding a ferritin heavy chain-like protein), which has 14 copies (most of which are tandem duplicates) in hadal snailfish as opposed to 3 copies in Tanaka snailfish (Figure 5A; Figure 5-figure supplements 2). It has also been suggested that ferritin helps control ROS (ORINO et al., 2001; Salatino et al., 2019). The expansion of fthl27 was validated in all the eight resequencing individuals by reads mapping (Figure 5-figure supplements 3), indicating that the tandem duplication event occurred at least before the differentiation of these individuals. To test whether the fthl27 can resist oxidative stress, we cultured 293T cells with or without fthl27-overexpression plasmid in cell culture medium supplemented with H2O2 or ferric ammonium citrate (FAC) for 4h, and subsequently measured intracellular ROS levels as well as cell viability. The results showed that the intracellular ROS levels of fthl27-overexpression cells were significantly lower than that of the control group (Figure 5B; Figure 5-figure supplements 4). Meanwhile, the fthl27-overexpression cells were also found to had significantly higher cell viability (Figure 5C). Therefore, we hypothesize that the expansion and high expression of this gene may be an important mechanism of resistance to HHP induced ROS in hadal snailfish.
The more sequenced individuals provide us with more details about the evolutionary history about the hadal snailfish. For example, given that the divergence time of the hadal snailfish and the other species of the family Liparidae living at a depth of 1,000 m was about 9.9 million years ago (Mya), and the divergence time between different sequenced hadal snailfish individuals was about 1.1 Mya, it is known that the hadal snailfish entered the hadal zone between 1.1 and 9.9 Mya. Then consider the fact that the genes that are responsible for detecting light in dark environment are well preserved in the hadal snailfish, it is likely that this species have only entered a completely light-free environment in the last millions of years, after the full completion of the Mariana Trench (Oakley et al., 2009). In addition, the phylogenetic relationships between different individuals clearly indicate that they have successfully spread to different trenches within 1.0 Mya (Figure 1-figure supplements 6).
The comparative genomic analysis revealed that the complete absence of light had a profound effect on the hadal snailfish. In addition to the substantial loss of visual genes and loss of pigmentation, many rhythm-related genes were also absent, although some rhythm genes were still present. The gene loss may not only come from relaxation of natural selection, but also for better adaptation. For example, the grpr gene copies are absent or down-regulated in hadal snailfish, which could in turn increased their activity in the dark, allowing them to survive better in the dark environment (Wada et al., 1997). The loss of gpr27 may also increase the ability of lipid metabolism, which is essential for coping with short-term food deficiencies (Nath et al., 2020).
The most interesting question about the hadal snailfish is why this is currently one of the very few observed vertebrate species capable of surviving and reproducing at such depths. TMAO, which is able to maintain protein function under high pressure, is thought to be a limiting factor in determining the depth at which fish can survive (P. H. Yancey et al., 2014). Results from our previous analysis suggested that positive selection on fmo3 may be important in promoting TMAO synthesis; but after an updated genome and rigorous FDR correction, we found that these genes had not undergone any particular or significant alteration in the amino acid sequences, although this does not exclude the possibility of a small number of amino acids (three species-specific mutations in all the five copies) having an effect on enzyme activity (Figure 5-figure supplements 5). Since the expression of all five copies of fmo3 are similar among the various tissues of Tanaka’s snailfish and hadal snailfish (Supplementary file 16), it seems more likely that the mechanism associated with TMAO degradation is altered in hadal snailfish, although we do not yet have any additional evidence to support this because the genes associated with TMAO degradation are still unclear.
However, the levels of TMAO are not sufficient for us to understand why only the hadal snailfish can tolerate such high hydrostatic pressure, since this substance is widely present in marine fishes. In contrast, the tandem duplication events of two genes may play a more critical role in the adaptation of the hadal snailfish. The first event is the tandem duplication of cldnj, a gene essential for otolith formation (Figure 3B) (Hardison et al., 2005). The two more copies may help the hadal snailfish to maintain the densities of their otoliths far beyond the limits of CCD. Since the dissolution rate of calcium carbonate increases with higher pressure, the otoliths stability may be one of the reasons limiting fishes to dive to even deeper regions. The second event is the massive expansion and high expression of fthl27. Our cellular experiments proved that this gene could help cells to resist ROS burst and protect it from various damages caused by oxidative stress. Meanwhile, comparative transcriptomic analysis observed multiple genes associated with DNA repair are significantly more highly expressed in all tissues of hadal snailfish than in other fishes, which also coincides with the presence of oxidative stress (Figure 5-figure supplements 1).
In summary, we provide chromosome-level genomes of hadal snailfish and Tanaka’s snailfish, as well as additional transcriptome and resequencing data. We report here further advances in our understanding of the origin, specific characteristics and adaptive mechanisms of the hadal snailfish.
Materials And Methods
Sample collection and identification
All the experimentation in this study were conducted in accordance with the pre-approved guidelines of the Ethics Committee of the Institute of Deep-Sea Science and Engineering, Chinese Academy of Sciences (Sanya, China). The hadal snailfish samples were collected form one site in the Mariana Trench (142°26′E, 11°07′N) at depth of 7,254m using the deep-sea landers Tianya with a surfacing time of 3 hours (Supplementary file 1). These specimens were identified as conspecific with Pseudoliparis swirei by morphological observations. Tanaka’s snailfish specimens were collected in the southern Yellow Sea in 2018 and identified as Liparis tanakae (Gilbert and Burke, 1912) on the basis of morphological observations.
Genome sequencing and assembly
Genomic DNA was extracted from the muscle of four hadal snailfish collected from the Mariana Trench, and four Tanaka’s snailfish collected from the southern Yellow Sea. We generated a total of 47.8 gigabases (gb) of Nanopore reads, 148.6 gb of BGI short reads, and 123.3 gb of Hi-C reads for hadal snailfish; and 39.0 gb of Nanopore reads, 130.3 gb of BGI short reads, and 99.5 gb of Hi-C reads for Tanaka’s snailfish.
The genome sizes of hadal snailfish and Tanaka’s snailfish were estimated by k-mer distribution analysis (K = 27) of SOApec v2 (Luo et al., 2012) to be 633.2 Mb and 539.9 Mb, respectively. We then assembled the hadal snailfish and Tanaka’s snailfish genomes based on the filtered Nanopore sequencing data using wtdbg2 v2.4.1 (Ruan & Li, 2020) based on default parameters, followed by two rounds of error correction using NextPolish v1.0 (J. Hu, Fan, Sun, & Liu, 2020) based on the filtered BGI sequencing data, and finally assembled them into chromosomal versions using 3D-DNA v180114 (Dudchenko et al., 2017) based on Hi-C data. Finally, BUSCO v 4.1.2 (Manni, Berkeley, Seppey, Simao, & Zdobnov, 2021) was used with the library ‘actinopterygii_odb10’ to analyze and evaluate the completeness of the gene set in our draft genome.
A total of 11 transcriptomes from six tissues (eye, stomach, heart, liver, muscle, skin) were extracted from three hadal snailfish, while a total of 26 transcriptomes from 10 tissues (brain, spinal cord, eye, bone, cholecyst, stomach, heart, liver, muscle, skin) were extracted from three Tanaka’s snailfish. RNA was subsequently extracted using TRIzol (Invitrogen) and purified using the RNeasy Mini Kit (Qiagen). Transcriptome reads were obtained from the Illumina HiSeq 2000 sequencing platform. The RNA sequences were filtered by Fastp v0.20 (S. Chen, Zhou, Chen, & Gu, 2018) and assembled without reference using SPAdes (Bushmanova, Antipov, Lapidus, & Prjibelski, 2019) with default parameters. Subsequently, TransDecoder v5.5.0 (https://github.com/TransDecoder/TransDecoder) was used to identify coding regions of the transcripts.
Both de novo and homology-based predictions were used to identify repetitive elements in hadal snailfish and Tanaka’s snailfish. First, we constructed a de novo transposable element library using RepeatModeler v1.0.11 (Saha, Bridges, Magbanua, & Peterson, 2008), and then used RepeatMasker v4.0.7 (N. Chen, 2004) to detect repeats. For homologous annotations, the genome sequences were compared with data from Repbase using RepeatMasker v4.0.7 and RepeatProteinMask v1.36 to predict transposable elements. For tandem repeat sequences, we used Tandem Repeats Finder v4.07 (Benson, 1999) to make predictions.
The repeat masked genome was used for the gene annotation. We used a combination of ab initio gene predictions, homologous gene predictions, and direct gene models produced by transcriptome assembly to identify protein-coding genes structure on the genome as follows.
Step 1: Augustus v3.2.1 (Stanke, Diekhans, Baertsch, & Haussler, 2008) was used to generate ab initio predictions with internal gene models.
Step 2: The protein sequences from seven species, medaka, Atlantic cod, flatfish, stickleback, zebrafish, turbot, and fugu and the transcriptome predicted protein sequences were used to align genomic sequences with BLAT v. 35 (Supplementary file 6) (Kent, 2002).
Step 3: The psl files obtained in the previous step were integrated and the protein sequences that were aligned to the overlapping region of the genome were scored and sorted based on the alignment results using a custom script to filter out the best aligned protein sequences in this region. Then, GeneWise v2.4.1 (Birney, Clamp, & Durbin, 2004) was used to predict gene models with the aligned sequences as well as the corresponding query proteins. The custom scripts have been deposited in GitHub (https://github.com/wk8910/bio_tools/tree/master/42.prediction).
Step 4: The Evidence Modeler (EVM) v1.1.1 (Haas et al., 2008) was used to integrate the prediction results with different weights for each.
The integrated gene set was translated into amino acid sequences using InterProScan v5 (Jones et al., 2014) to annotate motifs and domains in protein sequences by searching publicly available databases (including Pfam, PRINTS, PANTHER, ProDom and SMART) and the genes were further annotated using the KEGG databases.
Variant calling using resequencing data
Short reads of seven Mariana hadal snailfish, one Yap hadal snailfish and five Tanaka’s snailfish (Supplementary file 1) were mapped to the hadal snailfish genome assembled in this study with BWA v0.7.12-r1039 (Li, 2013); then SAMtools v1.4 (Li et al., 2009) was used to sort and obtain BAM files. To analyze population genetics, we focused on SNPs and small indels (1–10 bp) (Zhang et al., 2021). The SNPs were called by FreeBayes v0.9.10-3-g47a713e (Garrison & Marth, 2012) with parameters “--gvcf -- min-coverage 5 --limit-coverage 200” and filtered by following three thresholds: (1) SNPs with missing rate £ 30%; (2) The highest sequencing depth of SNP position < 200 x; (3) The lowest sequencing depth for each allele ≥ 5. Subsequently, we calculated the distribution of heterozygosity in genome-wide regions with 500 kb non-overlapping sliding windows.
Inference of phylogeny history
SNP tree, PCA and diversity statistics. PLINK v1.90b6.6 (Z.-L. Chen et al., 2019) was used to perform PCA and other population divergency statistics, including nucleotide diversity and genetic differentiation (FST). A neighbor-joining tree was constructed with PHYLIP v3.697 (Felsenstein, 1993) for paired genetic distance matrices.
Admixture analysis. Different K values (from 1 to 5) were tested using Admixture v1.3.0 (Alexander, Novembre, & Lange, 2009) to infer ancestral populations in all hadal snailfish and Tanaka’s snailfish individuals accessions.
Demographic analysis. The demographic history of hadal snailfish and Tanaka’s snailfish was inferred with pairwise sequential Markovian coalescent (PSMC) (Li & Durbin, 2011) analysis, based on a substitution rate of 1.9174e-09 per generation for hadal snailfish and 5.6790e-09 per generation for Tanaka’s snailfish. The analysis was performed using the following parameters: −N25 −t15 −r5 −p ‘4+25 × 2+4+6’. Thes mutation rates were estimated by r8s v1.81. The generation time of one year for Tanaka snailfish and three years for hadal snailfish.
Mitochondrial genome phylogenetic reconstruction and divergence time estimation. The mitochondria of eight hadal snailfish and five Tanaka’s snailfish were assembled by NOVOPlasty v4.3.1 (Dierckxsens, Mardulyn, & Smits, 2017) with default parameters and annotated by MITOS (http://mitos2.bioinf.uni-leipzig.de/index.py). Subsequently, mitochondrial data from currently published species of the Liparidae were combined, nucleic acid sequences of 13 coding genes on mitochondria were aligned by MUSCLE v3.8.425 (Edgar, 2021) using default parameters, and alignments of the coding sequences were generated with pal2nal v14 using default parameters. The maximum likelihood (ML) tree was constructed by RAxML-8.2.12 (Stamatakis, 2014) using the following parameters: -f a -m GTRGAMMA -p 15256 -x 271828 -N 100. Finally, divergence times were estimated by MCMCtree v4.9j (Yang, 2007) with 1 soft-bound calibration time-point (snailfish-stickleback: ~32-73 Ma)-based on previous studies. For Notoliparis kermadecensis, we combined all the above mitochondrial data and performed the same above analysis based on co1 and cytb gene sequences to obtain the ML tree and divergence times.
Gene loss and duplication
Here, we applied an improved read mapping-based method to identify gene loss and duplication, which is effective in reducing false positives and false negatives caused by genome assembly and annotation errors as well as multi-species sequence alignments. The custom scripts have deposited in GitHub (https://github.com/wenjie-xu-nwpu/hadal_snailfish). Although this method may have limitations for identifying gene loss and duplication in species with long divergence times, the divergence times of hadal snailfish and Tanaka’s snailfish are about 20 million years (K. Wang et al., 2019), and at least 88% of the reads in all hadal snailfish individuals can be well compared to Tanaka’s snailfish genome, indicating that this method is applicable to this study.
For gene loss, the following methods were used for identification. (1) Short reads of eight hadal snailfish and five Tanaka’s snailfish (~30 x) were compared to the stickleback and Tanaka’s snailfish genome using BWA v0.7.12-r1039 (Li, 2013) and subsequently sorted by SAMtools v1.4 (Li et al., 2009) to obtain the BAM files. (2) We obtained the reads depth for each sites in the gene coding region based on the annotation information of the reference genome, and subsequently classified the depths we had on individual loci into 3 types (“HIGH” for greater than half of the average coverage, “LOW” for less than 3, and “MID” for the rest). We defined sites with “HIGH” for Tanaka’s snailfish and “LOW” for hadal snailfish as hadal snailfish-specific lost sites (SLSs). Then, the genes with SLSs accounting for at least 40% of the coding sequence length were selected as the candidate specific loss genes. (3)The protein sequences of the genes selected in the previous step were used as a reference to search through the genome of hadal snailfish by BLAT v. 35 (Kent, 2002) and predict the gene structure using GeneWise v2.4.1 (Birney et al., 2004) to determine the genes that were completely lost or partially lost in this species. (4) The synteny alignment between the hadal snailfish, Tanaka’s snailfish and stickleback were plotted for partial or fully lost of the gene.
For gene duplication, the following methods were used for identification. (1) Short reads of eight hadal snailfish and five Tanaka’s snailfish were compared to the stickleback and Tanaka’s snailfish genome using BWA v0.7.12-r1039 (Li, 2013) and subsequently sorted by SAMtools v1.4 (Li et al., 2009) to obtain the BAM files. (2) The homologous sites whose average value of reads depth of all hadal snailfish individuals were greater than 1.5 the average value of the Tanaka’s snailfish individuals were retained and defined as hadal snailfish specific high-copy sites (HCSs). Then, the genes with HCSs accounting for at least 50% of the coding sequence length were selected as the candidate high-copy genes. (3) We searched for the location of this gene on the hadal snailfish genome by BLAT v. 35 (Kent, 2002) and predicted the gene structure by GeneWise v2.4.1 (Birney et al., 2004) to determine its copy number. (4) Finally, the expansion of this gene was determined by constructing a gene tree of the protein sequences of this gene family from nine species, hadal snailfish, Tanaka’s snailfish, medaka, Atlantic cod, flatfish, stickleback, zebrafish, turbot, and fugu.
Identification of unitary pseudogenes
Unitary pseudogenes are non-functional genes that decay at their original location (Tutar, 2012), and we suggest that some missing homologs will exist in hadal snailfish genome as unitary pseudogenes during their adaptation to the special environment of the hadal zone.
We obtained pseudogenes in hadal snailfish by following five steps. (1) Using the stickleback genome sequence as a reference, we performed synteny alignment for three species (hadal snailfish, Tanaka’s snailfish and stickleback) with Last v956 (Kiełbasa, Wan, Sato, Horton, & Frith, 2011) using the parameters ‘-E 0.05’, generating a total of 382 Mb (of which 290 Mb was informative for all species) of one-to-one alignment sequences with Multiz v1 (Blanchette et al., 2004) using the default parameters. (2) Genes with at least 70% of the coding sequences of stickleback or Tanaka’s snailfish present in the MAF and not present in the corresponding regions of hadal snailfish were selected as alternative unitary pseudogene datasets. (3) We used BLAST v2.9.0 (Altschul, Gish, Miller, Myers, & Lipman, 1990) to determine if this gene was present in other regions of the hadal snailfish genome. (4) The hadal snailfish corresponding region was extended left and right by 10 KB, and the genes of stickleback and Tanaka’s snailfish were used as references for predict the gene structure using GeneWise v2.4.1 (Birney et al., 2004). (5) Screening for pseudogenes that were consistent in all hadal snailfish individuals.
De novo originated new genes
First, the short reads of eight hadal snailfish and five Tanaka’s snailfish were compared to the hadal snailfish genome using BWA v0.7.12-r1039 (Li, 2013) and subsequently sorted by SAMtools v1.4 (Li et al., 2009) to obtain the BAM files. In the second step, we defined a single sequenced sample with reads depths < 10 at a single locus as a deletion locus. Based on the annotation file of hadal snailfish, we screened all Tanaka’s snailfish individuals for genes with deletions > 50%. Next, For the genes specifically present in hadal snailfish selected in the previous step, we used BLAST v2.9.0 (Altschul et al., 1990) to align them with the genomes of eight other fishes (Tanaka’s snailfish, medaka, Atlantic cod, flatfish, stickleback, zebrafish, turbot, and fugu) and screened for genes with a matching region < 0.4. Genes with transcripts per million (TPM) maxima less than 1 in each tissue of hadal snailfish were filtered out. The fully annotated genes (presence of start and stop codons) in the results were defined as novel genes of hadal snailfish.
Lineage-specific changes in amino acid sequences
For 17 species, Tanaka’s snailfish, stickleback, pacific bluefin tuna, medaka, platy fish, Atlantic cod, flatfish, zebrafish, turbot, fugu, spotted gar, coelacanth, chicken, mouse, human, brownbanded bamboo shark and elephant shark (Supplementary file 6), we identified one-to-one orthologues for each species and hadal snailfish by the Reciprocal Best-Hits (RBH) method, and subsequently selected genes present in 15 species, including hadal snailfish, as conserved gene sets. Next, the protein sequences of the selected genes were aligned using MAFFT v7.471 (Katoh & Standley, 2013), and a custom script was used to select regions that were consistent in other species and had contiguous specificity at sites greater than 3 bp in hadal snailfish, and that had at least 90% sequence identity for each 5 bp region before and after this variant region (B. Wu et al., 2021). Finally, genes with consistent variants in all hadal snailfish individuals were selected.
We performed protein structure simulation using AlphaFold2 (Cramer, 2021) for the amino acid sequences of target genes in hadal snailfish and Tanaka’s snailfish. Finally, the highest scoring prediction was selected as the best structure and visualized by UCSF Chimera (Pettersen et al., 2004).
Comparative transcriptome analysis
For the RNA sequences of hadal snailfish and Tanaka’s snailfish, we used Fastp v0.20.0 (S. Chen et al., 2018) to filter out low quality and contaminated reads, and then used Hisat2 v 2.1.0 (Kim, Paggi, Park, Bennett, & Salzberg, 2019) to align them to the respective reference genomes. StringTie v1.3.6 (Pertea, Kim, Pertea, Leek, & Salzberg, 2016) was then used to count the number of reads paired for each gene with the help of gene annotation information of the species, and then TPM values were calculated for each gene in both species. Next, we identified 17,281 one-to-one orthologues of hadal snailfish and Tanaka’s snailfish using the RBH method. Subsequently, we identified the genes that were differentially expressed (DEGs) between the same tissues of two species using the R package DESeq2 with |log2 (foldchange)| ≥ 1 and corrected P < 0.05. For genes that are up- regulated or down-regulated in multiple tissues, we first found by stochastic simulation that a gene is differentially expressed between two species in one organ does not affect the probability that this gene is differentially expressed in any other organ. Subsequently, we counted the genes that were up-regulated or down-regulated in each tissue to obtain a list of genes that were co-altered in multiple tissues.
Cell culture, transfection and ROS detection
We selected human embryonic kidney (HEK) 293T cells as an in vitro model and chose DMEM medium at 37°C and 5% CO2 in a saturated humid environment, changing the medium 2 or 3 times a week. Cells were inoculated in 6-well plates at a density of 4.0 × 105 cells/well. After a day when the cell density reached 50-60%, pcDNA3.1 and pcDNA3.1-fthl27 were transfected into cells with polyethylenimine (PEI), respectively. After 24h of transfection, 6-well plates (2 ml total volume of medium per well) were treated with 200 μl of 0.3 mM H2O2 or 0.4 mM FAC, both for 4 hours. Subsequently, the ROS Reactive Oxygen Species Kit (Aladdin R272916-1000T) was used to measure ROS. ROS levels were measured using a DCFH-DA molecular probe, and fluorescence was observed through a fluorescence microscope with an optional FITC filter, with the background removed to observe changes in fluorescence. Bright field photos and fluorescent photos were merged using ImageJ (https://github.com/imagej/ImageJ). Subsequently, we added AlamarBlue reagent to the cells in complete medium, incubated them for 2 hours, and then assayed them with a fluorescence zymograph with excitation light wavelength between 530-560 nm and emission light wavelength of 590 nm, and recorded the relative fluorescence units.
The project was supported by the National Key R&D Program of China (2022YFC3400300); the National Natural Science Foundation of China (32122021 to K.W. and 41876179 to S.H.); the 1000 Talent Project of Shaanxi Province to Q.Q., K.W. and S.H.; Fundamental Research Funds of Northwestern Polytechnic University; Strategic Priority Research Program of Chinese Academy of Sciences (Grant No. XDB42000000), open Foundation from Marine Sciences in the First-Class Subjects of Zhejiang (No.OFMS011). The authors thank Dr. Yang Zhou, Dr. Yuan Yuan, and Dr. Tao Qin for their advice and discussions during this project.
The sequence data files of the hadal snailfish have been deposited in the NCBI BioProject database with accession numbers PRJNA852951 (genome data) and PRJNA855356 (transcriptome data). The genome assembly file is under accession number JANBZZ000000000.
The following data sets were generated
NCBI BioProject ID PRJNA852951. Pseudoliparis swirei isolate: HS2019 Genome sequencing and assembly.
NCBI BioProject ID PRJNA855356.Transcriptome sequencing data of Pseudoliparis swirei.
The following previously published data sets were used
NCBI BioProject ID PRJNA472845. Pseudoliparis amblystomopsis Raw sequence reads.
NCBI BioProject ID PRJNA512070. Pseudoliparis sp. Yap Trench Genome sequencing and assembly.
Genome Reference Consortium (2019) Ensembl release 97 ID BROAD S1. Stickleback assembly and gene annotation.
Genome Reference Consortium (2019) Ensembl release 97 ID GRCz11. Zebrafish assembly and gene annotation.
Genome Reference Consortium (2019) Ensembl release 102 ID LepOcu1. Spotted gar assembly and gene annotation.
Genome Reference Consortium (2019) Ensembl release 97 ID GRCm39. Mouse assembly and gene annotation.
Genome Reference Consortium (2019) Ensembl release 97 ID GRCh38.92. Human assembly and gene annotation.
Genome Reference Consortium (2019) Ensembl release 97 ID ASM223471v1. Medaka assembly and gene annotation.
Genome Reference Consortium (2019) Ensembl release 97 ID X maculatus-5.0. Platyfish assembly and gene annotation.
Genome Reference Consortium (2019) Ensembl release 97 ID gadMor1. Atlantic cod assembly and gene annotation.
Genome Reference Consortium (2019) Ensembl release 97 ID ASM318616v1. Turbot assembly and gene annotation.
Genome Reference Consortium (2019) Ensembl release 97 ID FUGU5. Fugu assembly and gene annotation.
Genome Reference Consortium (2019) Ensembl release 97 ID LatCha1. Coelacanth assembly and gene annotation.
Genome Reference Consortium (2019) Ensembl release 97 ID Gallus gallus-5.0. Chicken assembly and gene annotation.
Genome Reference Consortium (2019) Ensembl release 97 ID Callorhinchus milii-6.1.3. Elephant shark assembly and gene annotation.
Fisheries Resources Institute, Japan Fisheries Research and Education Agency (2021) NCBI CBI Assembly ID GCA_021601225.1. Scaffold-level assembly of Pacific bluefin tuna.
Tsinghua University (2017) NCBI CBI Assembly ID GCF_001970005.1. Scaffold-level assembly of Flatfish.
Phyloinformatics Unit, Division of Bio-Function Dynamics Imaging, Center for Life Science Technologies, RIKEN (2018) NCBI CBI Assembly ID GCA_003427335.1. Scaffold-level assembly of Brownbanded bamboo shark.
K.W. and S.H. designed and managed the project. B.W., H.X., H.Z., X.G., J.B. and L.H. collected the samples. M.H. and C.Z. performed genome assembly and Hi-C chromosomal assembly. W.X. performed the repeat annotation and gene annotation. W.X., C.Z., B.W., M.H., J.Z. and C.F. performed the evolutionary analyses. X.G. performed experiments with fthl27. W.X., K.W., S.H., C.F., W.W. and Q.Q. wrote and revised the manuscripts.
- Molecular responses to high hydrostatic pressure in eukaryotes: Genetic insights from studies on saccharomyces cerevisiaeBiology 10
- Induction of oxidative stress by high hydrostatic pressure in Escherichia coliApplied and Environmental Microbiology 71:2226–2231
- An SOS response induced by high pressure in Escherichia coliJournal of bacteriology 186:6133–6141
- Biallelic TMEM251 variants in patients with severe skeletal dysplasia and extreme short statureHum Mutat 42:89–101https://doi.org/10.1002/humu.24139
- Fast model-based estimation of ancestry in unrelated individualsGenome research 19:1655–1664
- Basic local alignment search toolJournal of molecular biology 215:403–410
- Tandem repeats finder: a program to analyze DNA sequencesNucleic acids research 27:573–580
- GeneWise and genomewiseGenome research 14:988–995
- Vitamin D in an ecological contextInternational Journal of Circumpolar Health 59:26–32
- Aligning multiple genomic sequences with the threaded blockset alignerGenome research 14:708–715
- rnaSPAdes: a de novo transcriptome assembler and its application to RNA-Seq dataGigaScience 8
- Role of membrane fluidity in pressure resistance of Escherichia coli NCTC 8164Applied and Environmental Microbiology 68:5965–5972
- A blind circadian clock in cavefish reveals that opsins mediate peripheral clock photoreceptionPLoS Biol 9https://doi.org/10.1371/journal.pbio.1001142
- Using Repeat Masker to identify repetitive elements in genomic sequencesCurrent protocols in bioinformatics 5:4–14
- fastp: an ultra-fast all-in-one FASTQ preprocessorBioinformatics 34:i884–i890
- A high-speed search engine pLink 2 with systematic evaluation for proteome-scale identification of cross-linked peptidesNature communications 10:1–12
- A differential polarized phase fluorometric study of the effects of high hydrostatic pressure upon the fluidity of cellular membranesBiochemistry 22:409–415
- Functional Role of Class III Myosins in Hair CellsFront Cell Dev Biol 9https://doi.org/10.3389/fcell.2021.643856
- Homeoviscous theory under pressure: II. The molecular order of membranes from deep-sea fishBiochimica et Biophysica Acta (BBA)-Biomembranes 776:144–150
- AlphaFold2 and the future of structural biologyNature Structural & Molecular Biology 28:704–705
- NOVOPlasty: de novo assembly of organelle genomes from whole genome dataNucleic acids research 45:e18–e18
- De novo assembly of the Aedes aegypti genome using Hi-C yields chromosome-length scaffoldsScience 356:92–95
- MUSCLE v5 enables improved estimates of phylogenetic tree confidence by ensemble bootstrappingbioRxiv
- Biochemical implications and geochemical significance of novel phospholipids of the extremely barophilic bacteria from the Marianas Trench at 11,000 mDeep Sea Research Part I: Oceanographic Research Papers 47:1173–1182
- PHYLIP (phylogeny inference package), version 3.5 c
- Light and Cognition: Roles for Circadian Rhythms, Sleep, and ArousalFront Neurol 9https://doi.org/10.3389/fneur.2018.00056
- A Large Aggregation of Liparids at 7703 meters and a Reappraisal of the Abundance and Diversity of Hadal FishBioScience 60:506–515https://doi.org/10.1525/bio.2010.60.7.6
- Haplotype-based variant detection from short-read sequencingarXiv preprint arXiv:1207.3907
- Comparative feeding ecology of abyssal and hadal fishes through stomach content and amino acid isotope analysisDeep Sea Research Part I: Oceanographic Research Papers 121:110–120
- On the Success of the Hadal SnailfishesIntegr Org Biol 1https://doi.org/10.1093/iob/obz004
- Habitat influences skeletal morphology and density in the snailfishes (family Liparidae)Front Zool 18https://doi.org/10.1186/s12983-021-00399-9
- Pseudoliparis swirei sp. nov.: A newly-discovered hadal snailfish (Scorpaeniformes: Liparidae) from the Mariana TrenchZootaxa 4358https://doi.org/10.11646/zootaxa.4358.1.7
- Revision of the depth record of bony fishes with notes on hadal snailfishes (Liparidae, Scorpaeniformes) and cusk eels (Ophidiidae, Ophidiiformes)Marine Biology 168https://doi.org/10.1007/s00227-021-03950-8
- Expression of Protein-Coding Gene Orthologs in Zebrafish and Mouse Inner Ear Non-sensory Supporting CellsFront Neurosci 13https://doi.org/10.3389/fnins.2019.01117
- Proteins under pressure: the influence of high hydrostatic pressure on structure, function and assembly of proteins and protein complexesEuropean journal of biochemistry 221:617–630
- Automated eukaryotic gene structure annotation using EVidenceModeler and the Program to Assemble Spliced AlignmentsGenome biology 9:1–22
- The zebrafish gene claudinj is essential for normal ear function and important for the formation of the otolithsMech Dev 122:949–958https://doi.org/10.1016/j.mod.2005.03.009
- NextPolish: a fast and efficient genome polishing tool for long-read assemblyBioinformatics 36:2253–2255https://doi.org/10.1093/bioinformatics/btz891
- Gene expression responses in zebrafish to short-term high-hydrostatic pressureZool Res 43:188–191https://doi.org/10.24272/j.issn.2095-8137.2021.352
- Transcription Factor Reprogramming in the Inner Ear: Turning on Cell Fate Switches to Regenerate Sensory Hair CellsFront Cell Neurosci 15https://doi.org/10.3389/fncel.2021.660748
- The hadal zone: life in the deepest oceans
- Ecology of deep oceans: hadal trencheseLS
- InterProScan 5: genome-scale protein function classificationBioinformatics 30:1236–1240
- Sparc protein is required for normal growth of zebrafish otolithsJ Assoc Res Otolaryngol 9:436–451https://doi.org/10.1007/s10162-008-0137-8
- The medaka draft genome and insights into vertebrate genome evolutionNature 447:714–719
- High pressure-induced changes of biological membrane: Study on the membrane-bound Na+/K+-ATPase as a model systemEuropean journal of biochemistry 269:110–118
- MAFFT multiple sequence alignment software version 7: improvements in performance and usabilityMolecular biology and evolution 30:772–780
- BLAT—the BLAST-like alignment toolGenome research 12:656–664
- Adaptive seeds tame genomic sequence comparisonGenome research 21:487–493
- Graph-based genome alignment and genotyping with HISAT2 and HISAT-genotypeNature biotechnology 37:907–915
- A novel mutation at the DFNA36 hearing loss locus reveals a critical function and potential genotype-phenotype correlation for amino acid-572 of TMC1Clin Genet 71:148–152https://doi.org/10.1111/j.1399-0004.2007.00739.x
- Actin-bundling protein TRIOBP forms resilient rootlets of hair cell stereocilia essential for hearingCell 141:786–798https://doi.org/10.1016/j.cell.2010.03.049
- The zebrafish dog-eared mutation disrupts eya1, a gene required for cell survival and differentiation in the inner ear and lateral lineDev Biol 277:27–41https://doi.org/10.1016/j.ydbio.2004.08.033
- Hydrostatic pressure can induce apoptosis of the skinScientific reports 10:1–16
- Aligning sequence reads, clone sequences and assembly contigs with BWA-MEMarXiv preprint arXiv:1303.3997
- Inference of human population history from individual whole-genome sequencesNature 475:493–496
- The sequence alignment/map format and SAMtoolsBioinformatics 25:2078–2079
- Short-term starvation in silver pomfret (Pampus argenteus): molecular effects on lipid mobilization and utilizationAquaculture Research 48:4874–4885https://doi.org/10.1111/are.13307
- SOAPdenovo2: an empirically improved memory-efficient short-read de novo assemblerGigaScience 1
- Microscopic insights into the protein-stabilizing effect of trimethylamine N-oxide (TMAO)Proceedings of the National Academy of Sciences 111:8476–8481
- Tip-link protein protocadherin 15 interacts with transmembrane channel-like proteins TMC1 and TMC2Proc Natl Acad Sci U S A 111:12907–12912https://doi.org/10.1073/pnas.1402152111
- BUSCO Update: Novel and Streamlined Workflows along with Broader and Deeper Phylogenetic Coverage for Scoring of Eukaryotic, Prokaryotic, and Viral GenomesMol Biol Evol 38:4647–4654https://doi.org/10.1093/molbev/msab199
- Caspase-2 and oxidative stress underlie the immunogenic potential of high hydrostatic pressure-induced cancer cell deathOncoimmunology 6
- Whole genome sequencing of a snailfish from the Yap Trench (~7,000 m) clarifies the molecular mechanisms underlying adaptation to the deep seaPLoS Genet 17https://doi.org/10.1371/journal.pgen.1009530
- Vision using multiple distinct rod opsins in deep-sea fishesScience 364:588–592
- Genetic deletion of gpr27 alters acylcarnitine metabolism, insulin sensitivity, and glucose homeostasis in zebrafishFASEB J 34:1546–1557https://doi.org/10.1096/fj.201901466R
- Sedimentary, volcanic, and tectonic processes of the central Mariana Arc: Mariana Trough back-arc basin formation and the West Mariana RidgeGeochemistry, Geophysics, Geosystems 10https://doi.org/10.1029/2008gc002312
- Ferritin and the response to oxidative stressBiochemical Journal 357:241–247
- Temporal transcriptomics suggest that twin-peaking genes reset the clockElife 4
- Transcript-level expression analysis of RNA-seq experiments with HISAT, StringTie and BallgownNature protocols 11:1650–1667
- UCSF Chimera—a visualization system for exploratory research and analysisJournal of computational chemistry 25:1605–1612
- Contrasting Gene Decay in Subterranean Vertebrates: Insights from Cavefishes and Fossorial MammalsMol Biol Evol 38:589–605https://doi.org/10.1093/molbev/msaa249
- Whole exome sequencing identifies new causative mutations in Tunisian families with non-syndromic deafnessPLoS One 9https://doi.org/10.1371/journal.pone.0099797
- Comparative assessment of Fgf’s diverse roles in inner ear development: A zebrafish perspectiveDev Dyn 250:1524–1551https://doi.org/10.1002/dvdy.343
- Fast and accurate long-read assembly with wtdbg2Nat Methods 17:155–158https://doi.org/10.1038/s41592-019-0669-3
- Empirical comparison of ab initio repeat finding programsNucleic acids research 36:2284–2294
- H-Ferritin affects cisplatin-induced cytotoxicity in ovarian cancer cells through the modulation of ROSOxidative medicine and cellular longevity 2019
- The effects of pressure on G protein-coupled signal transductionBiochemistry and Molecular Biology of Fishes :147–174
- RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogeniesBioinformatics 30:1312–1313
- Using native and syntenically mapped cDNA alignments to improve de novo gene findingBioinformatics 24:637–644
- Otolith tethering in the zebrafish otic vesicle requires Otogelin and alpha-TectorinDevelopment 142:1137–1145https://doi.org/10.1242/dev.116632
- PseudogenesComparative and functional genomics 2012
- Prefoldin, a chaperone that delivers unfolded proteins to cytosolic chaperoninCell 93:863–873
- On the pathophysiology of DFNA9: Effect of pathogenic variants in the COCH gene on inner ear functioning in human and transgenic miceHear Res 401https://doi.org/10.1016/j.heares.2020.108162
- Generation and characterization of mice lacking gastrin-releasing peptide receptorBiochemical and biophysical research communications 239:28–33
- Morphology and genome of a snailfish from the Mariana Trench provide insights into deep-sea adaptationNat Ecol Evol 3:823–833https://doi.org/10.1038/s41559-019-0864-8
- Eya4 regulation of Na+/K+-ATPase is required for sensory system development in zebrafishDevelopment 135:3425–3434https://doi.org/10.1242/dev.012237
- The Genomes of Two Billfishes Provide Insights into the Evolution of Endothermy in TeleostsMol Biol Evol 38:2413–2427https://doi.org/10.1093/molbev/msab035
- A Novel Loss-of-Function Variant in Transmembrane Protein 263 (TMEM263) of Autosomal Dwarfism in ChickenFront Genet 9https://doi.org/10.3389/fgene.2018.00193
- Gpr19 is a circadian clock-controlled orphan GPCR with a role in modulating free-running period and light resetting capacity of the circadian clockSci Rep 11https://doi.org/10.1038/s41598-021-01764-8
- Insights into the vision of the hadal snailfish Pseudoliparis swirei through proteomic analysis of the eyeProteomics 21https://doi.org/10.1002/pmic.202100118
- Unusual organic osmolytes in deep-sea animals: adaptations to hydrostatic pressure and other perturbantsComparative Biochemistry and Physiology Part A: Molecular & Integrative Physiology 133:667–676
- Marine fish may be biochemically constrained from inhabiting the deepest ocean depthsProc Natl Acad Sci U S A 111:4461–4465https://doi.org/10.1073/pnas.1322003111
- PAML 4: phylogenetic analysis by maximum likelihoodMolecular biology and evolution 24:1586–1591
- Haplotype-resolved genome assembly provides insights into evolutionary history of the tea plant Camellia sinensisNat Genet 53:1250–1259https://doi.org/10.1038/s41588-021-00895-y
- Role of the GRP/GRPR System in Regulating Brain FunctionsACS Chemical Neuroscience