Hadal, the deepest area of the ocean (6,000–11,000 m), is a unique harsh environment characterized by extremely high hydrostatic pressures, low temperature, and limited food sources (Somero, 1992; Jamieson et al., 2010). Hadal is an extremely hostile environment for most organisms; however, some species seem to be adapted very well to the hadal environment (Somero, 1992; Bartlett, 2002; Yancey and Siebenaller, 2015). With increasing sampling efforts regarding the hadal trenches, a wide range of organisms was found to thrive within the extreme environment (Nunoura et al., 2015; Tarn et al., 2016). How those hadal creatures adapted to the extreme environment is a topic of great interest to researchers.

The organisms that are endemic to hadal zones have evolved unique physiological and biochemical functions necessary for growth and survival in this hadal habitat. Hadal creatures were found to have more fluidity of cell membranes due to an increase of less rigid unsaturated fatty acid chains (Cossins and Macdonald, 1984, 1989). For example, compared with the bathyal roundnose grenadier Coryphaenoides rupestris (400–1,200 m depth), the abyssal grenadier Coryphaenoides armatus (2,000–5,000 m depth) was found to have consistently higher fluidity in membranes (Cossins and Macdonald, 1989). Hadal creatures were also found to own some external types of “assistance” to improve protein stability and functional adaptability under high hydrostatic pressures. The types of “assistance” include stress proteins, phospholipids, and micromolecular counteractants, for example, trimethylamine N-oxide (TMAO), which was reported to be of great significance for the adaptation of hadal species (Yancey, 2020).

With the development of molecular technology, researchers tried to explore the molecular adaptation mechanisms of the hadal creatures at the genetic level. For example, the whole genome sequence of Mariana hadal snailfish (Pseudoliparis Swirei) was recently analyzed. The bone Gla protein (BGLAP) gene possessed a frameshift mutation in P. swirei, which resulted in an incomplete ossification of the hadal snailfish to adapt to a high-pressure environment. The hadal snailfish also lost several important photoreceptor genes to adapt to the lightless hadal environment (Wang et al., 2019).

Amphipods (Arthropoda: Crustacea: Amphipoda) have a long evolutionary history, wide variety of species, and wide distribution ranging from 0 to 11,000 m, and are considered to be the dominant scavengers in the hadal food web (Eustace et al., 2016; Lacey et al., 2016; Copilaş-Ciocianu et al., 2020). These animals are among the few fauna species that can be readily obtained in large numbers and diversities. Therefore, hadal amphipods are commonly used in the study of adaptability to an extreme environment. By comparing hadal and littoral amphipods, researchers have found that TMAO, scyllo-inositol, and other pressure counteractants increase with depth (Downing et al., 2018). The hadal Hirondellea gigas was found to have a unique enzyme component, cellulase (HGcel), which can digest the wooden debris buried in the deepest seafloor (Kobayashi et al., 2012).

Among the reported hadal amphipods so far, one unique hadal amphipod, Alicella gigantea, attracts wide attention due to its significant gigantism. A. gigantea, also called “supergiant,” inhabits in the deep abyssal plains in the Northern Hemisphere of the North Atlantic Ocean (off the Canaries, Cape Verde, and in the Demerara Basin) and near the Hawai'ian Islands of the North Pacific Ocean (Barnard and Ingram, 1986; Hasegawa et al., 1986; De Broyer and Thurston, 1987). It is the largest known amphipod, whose adult body length ranges from 240 to 340 mm (Harrison et al., 1983; Barnard and Ingram, 1986; Jamieson et al., 2013). Chapelle and Peck (2004) counted the body lengths of over 2,000 amphipod species and found that the body sizes of most amphipod species were <40 mm. It has also been suggested that, with reduced temperature and increased hydrostatic pressure, there should be an increase in cell size and life span (Timofeev, 2001), which might be an explanation for the gigantism of A. gigantea. However, until now there has not been a genetic mechanism study on the gigantism of A. gigantea.

Bathycallisoma schellenbergi is a geographically widely distributed hadal amphipod species. It is found in the North Pacific, the Puerto Rico Trench in the Southwest Pacific, and the Java Trench in the Indian Ocean (Lacey et al., 2013). Compared with the supergiant A. gigantea, B. schellenbergi is quite a small hadal amphipod. For example, B. schellenbergi collected from the Tonga Trench had an average size of 25 mm (Jamieson, 2015). Similar to most hadal amphipods, B. schellenbergi is an obligate scavenger, feeding on carrion falling to the deep-sea floor (Sainte-Marie, 1992; Britton and Morton, 1994; Kaiser and Moore, 1999). It has some hadal adaptations, such as the acute chemoreceptor organs used to detect carrion and the ability to resist chronic hunger, which is conducive to the scavenging foraging strategy (Smith and Baldwin, 1982; Sainte-Marie, 1992). Therefore, B. schellenbergi can be used as a fine reference species to explore the gigantism mechanism of A. gigantea.

Indeed, to explore the hadal adaptation mechanisms and the possible causes of gigantism in A. gigantea, its whole genome has to be analyzed. However, the estimated huge genome size of A. gigantea (34.02 Gb) makes such a task cumbersome (Ritchie et al., 2017). It has been also well-known that the hadal amphipods possess a striking large genome, ranging from 4.04 Gb in Paralicella caperesca to 34.02 Gb in A. gigantea (Ritchie et al., 2017). Therefore, transcriptome sequencing and analysis could be an effective option for genome-wide comparative adaptation studies of hadal species. Some researchers have identified a genome-wide positive selection by combining the next-generation sequencing of the transcriptome with branching site modeling to reveal the underlying mechanisms of molecular adaptation (Tsaur and Wu, 1997; Roux et al., 2014; Dungan et al., 2016), which provides technical support for the hadal species adaptation studies at the genetic level. As of now, only H. gigas from the Mariana Trench has been analyzed by transcriptome (Lan et al., 2017), and the transcriptome analysis of other hadal amphipod species is lacking.

In this study, two different sized amphipods, A. gigantea and B. schellenbergi, were selected as the research objects for the transcriptome and evolutionary analysis, which could improve the adaptation mechanisms of hadal amphipods at the genetic level. Meanwhile, the gigantism mechanism of the “supergiant” A. gigantea was also explored in this study.

Hadal zone is an extreme environment and is also the least known area. How the organisms endemic to hadal zones adapted to an extreme environment attract a wide interest of several researchers. Due to technical limitations, previous studies were conducted mainly at biochemical levels, such as the studies on the osmotic factors, lipids, or proteins (Yancey, 2020), and there are very few studies regarding the adaptive mechanisms of the hadal species at the genetic level. With the development of the next-generation sequencing technology, transcriptome studies of many species have been conducted. However, due to the restriction by the difficulties of the hadal species samplings, only one hadal amphipod species, H. gigas, has been analyzed by using transcriptome sequencing (Lan et al., 2017). In this study, the two hadal amphipods, A. gigantea and B. schellenbergi, were selected for transcriptome sequencing and comparative evolutionary analysis. Compared to the published H. gigas transcriptomes (Lan et al., 2017), similarly predicted gene number and much more contigs were obtained from the de novo assembly for A. gigantea and B. schellenbergi (Supplementary Table 1), which probably indicated comparably higher incomplete contigs produced in this study. The higher contig number obtained in our assembly might result from the biological variability (Smith-Unna et al., 2016). This study not only would improve the adaptation biology studies of hadal amphipods from the perspective of phylogeny but also could explore the underlying reasons for the gigantism of A. gigantea.

According to the phylogenetic tree constructed from the orthologous genes of the hadal and shallow-water arthropods, we found that A. gigantea, B. schellenbergi, and H. gigas all belong to the lysianassoidea clades (Figure 3A). The result is consistent with a recent study on a large-scale molecular phylogeny of amphipoda, in which ecologically diverse deep-sea species could be gathered together in a clade of lysianassoids (Copilaş-Ciocianu et al., 2020). It should be noticed that A. gigantea and B. schellenbergi have a close distance in the phylogenetic tree (Figure 3A) and also shared the more common gene families (Figure 3B), which indicated that A. gigantea and B. schellenbergi have a close relationship. However, it should be noticed that the number of single-copy orthologous genes used in the phylogeny tree construction was only 512, thus our results could not fully determine the species differentiation time among the three hadal amphipods, A. gigantea, B. schellenbergi, and H. gigas. However, this result could also provide the reference data for species phylogeny relationship identification studies.

A large-scale phylogeny study of amphipoda also indicates a close relationship between the deep-sea lysianassoids and shallow-water gammaroids (Copilaş-Ciocianu et al., 2020). Thus, the hadal species belonging to Lysianassoidea could be compared against shallow-water species belonging to Gammaroidea for a positive selection analysis. Among the taxa of shallow-water marine amphipods, which possess the available transcriptome data of Melita plumulosa (Hadzioids) (Hook et al., 2014), P. hawaiensis (Talitroids) (Kao et al., 2016), Grandidierella japonica (Corophioids) (Hiki et al., 2019), Gondogeneia Antarctica (Kang et al., 2015), E. marinus (Cogne et al., 2019), and Eogammarus possjeticus (Chen et al., 2019), three species (G. antarctica, E. marinus, and E. possjeticus) belong to the Gammaroidea taxa, which could be chosen as the reference species for a positive selection analysis. G. antarctica, inhabiting another extreme environment (the Antarctic Pole) evolved despite a constant cold environment (Kang et al., 2015), which might disturb our evolutionary analysis. E. possjeticus was widely distributed in the coastal and estuarine areas, and the transcriptome data of E. possjeticus were released in 2019 (Chen et al., 2019). However, muscle tissues were used for RNA sequencing (RNA-seq) in that study. Therefore, we finally chose the four Gammaroidean species, including a shallow-water marine species (E. marinus) and three freshwater species (G. minus, G. fossarum, and G. chevreuxi), as the reference species for evolutionary analysis.

The PSG detection results of A. gigantea and B. schellenbergi showed that the number of PSGs shared by the two hadal amphipods exceeded the number of unique PSGs possessed by each species (Figure 4A), which further indicated a close genetic relationship between the two species. The PSGs of A. gigantea and B. schellenbergi were enriched by the GO and KEGG analysis, and it was found that they were mainly related to “starvation response” and “meiosis” (Figure 4B and Table 1). Hadal is an environment with limited food supplies, and most of the falling organic matter will be decomposed by bacteria and consumed by animals (Jamieson et al., 2010). However, amphipods can survive long periods of starvation (Jamieson, 2015). In the transcriptome analysis of the hadal amphipod H. gigas, researchers found that the key genes directly involved in the “energy metabolism” pathway were positively selected, which were suggested to be a new genetic adaptation strategy for H. gigas to survive in the limited food supply environment (Lan et al., 2017). In our study, we found that the PSGs in the KEGG pathway “glycerolipid metabolism” were enriched both in A. gigantea and in B. schellenbergi (Table 1). Such a result is consistent with the findings of Lan et al. (2017). Meiosis is a biological process that depends on the cytoskeleton (Bourns et al., 1988), and high hydrostatic pressures in the hadal environment will affect the cytoskeleton, thus affecting meiosis (Ishii et al., 2004). The PSGs involved in the “meiosis” pathway would be related to the environmental adaptations of high hydrostatic pressure.

As mentioned earlier, A. gigantea and B. schellenbergi have a closer genetic relationship compared with another hadal amphipod, H. gigas (Figure 3A). However, their body sizes are greatly different from the H. gigas body size (Figure 1B). As shown in Figure 1C, the growth fitting curves of both A. gigantea and B. schellenbergi are linear, but the slope of A. gigantea is much steeper than that of B. schellenbergi. It is well-known that the body size and weight fitting curve of crustaceans are the common references for growth and development studies. Researchers studied the correlations between the body size and weight of a shallow-water crustacean, Pederson cleaner shrimp (Ancylomenes pedersoni), and found that the precisely measured total length increased linearly with the carapace length while the wet mass increased exponentially with the carapace length (Gilpin and Chadwick, 2017). This is not consistent with the linear body size and weight fitting curve of our hadal amphipods, which might be due to a harsh hadal environment. With the shortage of food supplies, the wet mass of hadal amphipods could not increase exponentially with respect to their body length. However, the growth fitting slope curve of A. gigantea is much steeper than that of the B. schellenbergi, indeed suggesting that, with the same body length, A. gigantea might gain more weight and grow much faster than B. schellenbergi.

In combination with the PSG detection using A. gigantea as the foreground and other small-sized amphipods as the background, the results showed that 58 genes were considered to be under positive selection in A. gigantea (value of p < 0.05; Supplementary Table 5) (14 PSGs with FDR <0.05, see Table 2). Two PSGs (ITPK and IMPA2) are involved in the Inositol phosphate metabolism pathway [process from phosphatidylinositol (PI) to myo-inositol], where ITPK converts inositol 1,4,5-trisphosphate [Ins (1,4,5) P₃] to inositol 1,3,4,5-tetrakisphosphate [Ins (1,3,4,5) P₄] (Nalaskowski et al., 2003), and IMPA2 converts inositol 3-phosphate (Ins3P) to myo-inositol (Figure 5B; McAllister et al., 1992). It is well-known that myo-inositol is often added to animal feeds to meet the metabolic needs of animals (Wang et al., 2020). Moreover, IMPDH1 (PSG as shown in Figure 5A and Table 2) is involved in nucleotide metabolism by converting IMP to XMP (Slee and Bownes, 1995). XMP can be further converted to GMP (Shivakumaraswamy et al., 2020), which can promote the growth performance of organisms (Hossain et al., 2016). Therefore, we can conclude that the two PSGs (ITPK and IMPA2) are involved in the inositol phosphate metabolism pathway, and together with IMPDH1 would ultimately be related to growth and proliferations in A. gigantea.

On the other hand, four PSGs (aPKc, PPP1R3B, GYG1, and Slc2a1), which are associated with insulin signaling and involved in the glycogenesis pathway, were identified in A. gigantea (Figure 5B). PPP1R3B, a key factor, connects the insulin signaling and glycogenesis signaling pathway (Luo et al., 2011). Slc2a1, an important factor, plays a role in glucose transport (Zhao and Keating, 2007), and together with GYG1 was involved in the glycogenesis process. Therefore, the PSGs involved in the glycogenesis suggested that A. gigantea might undergo active energy metabolism, which could help the “supergiant” to survive in the harsh hadal environment with scarce food sources. Similar to the previous studies on H. gigas (Lan et al., 2017), these PSGs involved in the “energy metabolism” pathway might be also related to starvation resistance in the hadal amphipods. In addition, researchers also found that larger animals showed better resistance to starvation compared to smaller animals (Cushman et al., 1993; Arnett and Gotelli, 2003). The insulin production-related PSGs shown in larger-sized A. gigantea (Figure 5) possibly explain the reason for larger animals possessing a better resistance to starvation.

To explore the gigantism of A. gigantea, aPKC, the most significant PSG (FDR <5.70E-07) identified in the “supergiant” A. gigantea (Figure 5A and Table 2), invokes our attentions. APKC, which encodes a member of the PKC family of serine/threonine protein kinases, was reported to affect insulin regulation (Zhao et al., 2017). PRKCI, an aPKC isoform (PRKC iota), was reported to be related to the gigantism of capybara (Hydrochoerus hydrochaeris), the world's largest living rodent (Herrera-Álvarez et al., 2021) as it is involved in cell survival, differentiation, and proliferation by accelerating G1/S transition (Ni et al., 2016). Therefore, we could expect that the most significant PSG, aPKC identified in A. gigantea might ultimately be related to cell proliferation and growth of the “supergiant” amphipod.

From the evolutionary rate comparisons between the hadal and shallow-water amphipods, it was clearly shown that the evolutionary rates of GO categories, such as “lipid synthesis,” “meiosis,” and “DNA repair,” increased in the hadal amphipods (Figures 6A,B,D,E). However, the evolutionary rates of “lipid synthesis” and “meiosis” are consistent with the abovementioned PSG enrichment results. The hadal environment has an extremely high hydrostatic pressure, which can lead to DNA damage (Abe et al., 1999; Rothschild and Mancinelli, 2001; Aertsen et al., 2004). Therefore, researchers suggested that hadal organisms might require a high frequency of DNA repair (Dixon et al., 2004), which is also consistent with our results. Notably, the most significantly enriched GO category in A. gigantea was “regulation of growth,” (Figure 6C) which indicated that size control or growth regulation mechanisms might hide under the growth of the “supergiant” amphipod, and this could explain why A. gigantea is so huge. In mammals, growth is regulated by growth hormones, excessive growth hormone secretion can cause gigantism (Lodish et al., 2016), and possibly, the existence of a similar regulatory mechanism in the “supergiant” amphipods.

In this research, a comparative evolutionary study regarding two different-sized hadal amphipods, A. gigantea and B. schellenbergi, was conducted. Many PSGs involved in “glycerolipid metabolism,” “response to starvation,” and “meiosis” were found in the two hadal species, suggesting that these pathways might be the most important adaptation mechanisms for the hadal creatures. Moreover, seven PSGs (especially the most significant PSG, aPKC) solely identified in the A. gigantea showed to be related to inositol phosphate metabolism, insulin signaling, and glycogenesis signaling. Together, the evolutionary rate of the GO term “growth regulation” was significantly higher in the “supergiant” A. gigantea than in B. schellenbergi and other small-sized amphipods. These points might be the possible gigantism mechanisms of A. gigantea.

All sequencing data associated with this project were deposited in the National Center for Biotechnology Information (NCBI) Sequence Read Archive database [BioProject Accession Numbers: PRJNA739006 (Alicella gigantea) and PRJNA739007 (Bathycallisoma schellenbergi)].

QX conceived the experiments, led the whole project, and contributed to edits to the manuscript. WL and FW analyzed the data. BP designed the lander vehicle for sample collection. JC and BP collected the samples. JC extracted the RNA. SJ performed RNA-seq. WL, FW, and QX wrote the manuscript. All authors contributed to the article and approved the submitted version.

This work was supported in part by the Funding Project of the National Key Research and Development Program of China (2018YFC0310600), the National Key Research and Development Program of China (2018YFD0900601), the National Natural Science Foundation of China (Grant No. 31772826), and the Major Scientific Innovation Project from Shanghai Committee of Education (2017-01-07-00-10-E00060).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.