Animals must receive adequate nutrition to support their normal growth and development, lipids attract more attention due to their crucial role in metabolism/nutrition and their complex physiological process (Cai et al., 2017; Anderson et al., 2018; Cruz et al., 2020). Lipids are vital components for energy, and cell membrane structures and work as messenger molecules; they also play crucial roles in animal growth and reproduction (Wu et al., 2010; Zhu et al., 2013; González-Félix M. L. et al., 2018; Gao et al., 2019). Adequate lipids depend on efficient ingestion and digestion, which relies on various types of digestive lipases (Navarro-Guillén et al., 2015; Kulminskaya & Oberer, 2020). As the third-largest enzyme group, lipase is a key enzyme for lipids digestion (Kurtovic et al., 2009; Basheer et al., 2011; Bouchaâla et al., 2015; Cerk et al., 2018), especially for hydrolyzing triacylglycerides (TAGs), glycerophospholipids (GPs) and esters of cholesterol (Navvabi et al., 2018; Morshedi et al., 2021). Here we focused on the two types of digestive lipase (E.C. 3.1.1.3) genes, pancreatic lipase gene (pl) and bile salt-activated lipase gene (bsal), they mainly catalyze the hydrolysis of triacylglycerols into glycerol and fatty acids (Sæle et al., 2010; Holmes and Cox, 2011; Achouri et al., 2020; Qiu et al., 2020).

In mammals, the most important digestive lipase genes are the pancreatic lipase gene (PL) and two genes closely related to the PL gene, namely pancreatic lipase-related protein 1 (PLRP1) and pancreatic lipase-related protein 2 (PLRP2) (Kurtovic et al., 2009; Zhu et al., 2021). Meanwhile, the bile salt-activated lipase (BSAL) also have a wide range of role in the hydrolyzing of TAGs, GPs, cholesterol esters, and lipid-soluble vitamins (Holmes & Cox, 2011; Navvabi et al., 2018). In contrast, in teleost fishes, the most important digestive lipase may be the bile salt-activated lipase (bsal). Firstly, it has a certain controversy about the existence of a real pancreatic lipase in fishes (Rønnestad et al., 2013; Anderson et al., 2018; Yanes-Roca et al., 2018), especially the lipase purified from fishes seemed to require bile salts for activation (Tocher, 2003). Additionally, until now, no reports have confirmed whether the main digestive lipase is pancreatic lipase or bile salt-activated lipase in fish. However, the bile salt-activated lipase, also called carboxyl ester lipase (cel), has been suggested to be the most important digestive lipase in fish species (Murashita et al., 2014; Fuentes-Quesada & Lazo, 2018; Sæle et al., 2018; Qiu et al., 2020; Morshedi et al., 2021; Sterzelecki et al., 2021). Secondly, the absence of the pl gene was shown in the zebrafish genome (Sæle et al., 2018). Although, most fishes had a plrp gene (a copy of pl), according to some reports (Sæle et al., 2010; Rønnestad et al., 2013) and annotation in NCBI, these plrp genes might produce inactive pancreatic lipases in teleosts, these pancreatic lipases might not possess the lipolytic function due to mutations in the domain (Sæle et al., 2010; Sæle et al., 2018). Moreover, pancreatic lipase is a colipase-dependent lipase, colipase is required for lipid binding and allows the pancreatic lipase to carry out the function of hydrolyzing lipids (Smichi et al., 2017; Anderson et al., 2018; González-Félix ML. et al., 2018; Achouri et al., 2020). The absence of the colipase gene (clps) also can be found in zebrafish (Danio rerio), medaka (Oryzias latipe), Japanese pufferfish (Takifugu rubripes), and three-spined stickleback (Gasterosteus aculeatus) (Sæle et al., 2010; Smichi et al., 2017; Anderson et al., 2018; Sæle et al., 2018).

In the light of the above information, the main digestive lipase gene seemed to show a complete difference between mammals and fishes. In this study, we identified the two types of digestive lipase genes (pl and bsal) in several vertebrates. We predicted the amino acid sequences, structures, protein models, as well as the phylogenetic relationship for these lipase genes. These results could be helpful for the understanding of the function and evolution of digestive lipase genes in the vertebrate.

Lipid is an important source of energy and a constituent of cell membranes throughout the life cycle in animals, it is considered a key nutrient during early life stages due to its high growth and development (Tocher 2003; Benedito-Palos et al., 2014; Anderson et al., 2018; Gilannejad et al., 2020). Lipases are versatile enzymes that catalyze the hydrolysis of ester linkages in lipids, such as TAGs (Karimi et al., 2010). Lipases are ubiquitous throughout animals. There are two important lipases in animals, one is pancreatic lipase (pl) and another is bile salt-activated lipase (bsal), which play an important role in hydrolyzing the ester bonds in TAGs during digestive processes (Karimi et al., 2010; Rueda-López et al., 2017). Although the reports about these two lipases are abundant in mammals, the main digestive lipase is controversial in fishes (Rønnestad et al., 2013; Anderson et al., 2018; Yanes-Roca et al., 2018). In this study, we understood the different lipid digestion genes among mammals and fishes, and this is the first report, which comprehends the digestive lipases from whole-genome in the fishes. The results showed that the pancreatic lipase gene (pl) just existed in part of fish species. Our findings in the present study indicated that zebrafish, medaka, common carp, and Nile tilapia lost the pl gene in the genome. Consistent with this result, a previous study also reported that pl gene was absent in zebrafish (Sæle et al., 2018). Moreover, we also found that the pl genes of Atlantic salmon and rainbow trout became pseudogenes. The syntenic analysis had ensured the correction of the whole genome identified in the present study, the neighboring genes were conserved in all fish species. Although most of the fish species possessed the pl gene, and they only had one copy of pl gene in the genome, the PL gene had occurred in gene tandem duplication and obtained three copies of the PL gene in mammals. The three copies of PL gene had been reported and confirmed in mammals (Xiao et al., 2011; Aloulou et al., 2014; Zhu et al., 2021). In contrast, as we can see, the BSAL gene not only existed in mammals but also existed in all fishes, even the bsal had occurred gene duplication in most of the fish species. Therefore, the copy number of lipase genes exhibited different models between mammals and fishes.

Meanwhile, the amino acid sequence analysis showed that the Ile and Leu were replaced by Ser, Lys, or Arg in the lid domain of pancreatic lipase in most fishes species, it also might influence the lipolytic ability of pancreatic lipase in fish (Smichi et al., 2017). However, the bsal gene occurred the gene duplication and obtained two to five copies in the fish species in which the Ile and Leu were absent, the phylogenetic analysis also showed that these fish species were grouped together (Figure 3). Thus, we inferred that most of the fish species had the pl gene, but this gene might have no or low capacity to digest the lipids, the annotation of these pl genes in NCBI also are inactive pancreatic lipase-related protein genes. In order to compensate for the low or inactive lipolytic ability of pancreatic lipase in these fish species, the bile salt-activated lipase gene (bsal) might occur the gene duplication. Some studies also believed that the bsal gene was the main and most important digestive lipase in Atlantic cod (Gadus morhua), Pacific bluefin tuna (Thunnus orientalis), California halibut (Paralichthys californicus), and yellowfin seabream (Acanthopagrus latus) (Sæle et al., 2010; Murashita et al., 2014; Fuentes-Quesada & Lazo, 2018; Morshedi et al., 2021). Some specific fish species, like spotted gar and catfishes, had the pl gene which might have the great capacity to digest the lipids, so the bsal did not occur gene duplication, they only had one copy of bsal gene. Similarly, the BSAL also did not occur in gene duplication in mammals. In addition, the amino acid of the lid domain in pancreatic lipase was conserved in these species, it also means that the function of pancreatic lipase might be fully-functioning (Ollis et al., 1992; Verger 1997; Smichi et al., 2017).

Therefore, we preliminarily considered the main digestive lipase gene was completely different between mammals and most fishes. We inferred that the PL in the ancestor of mammals possessed the lipolytic ability and obtained two other similar copies (PLRP), so the BSAL was not necessary to obtain an extra copy and retained one BSAL gene in mammals. However, the pl gene might lose the lipolytic ability in some fish ancestors, instead, the bsal not only played a major function in lipolytic ability in these fish species but also occurred gene duplication. Sequence alignment also revealed the amino acid residues of the bile salt-binding site were conserved in most of bsal genes (Figure 4). In some of the fish species, the bsal copies had been translocated to another chromosome, and the amino acid residues of the bile salt-binding site also changed (Figures 2, 4), it might be some mistakes happened when the bsal copies were translocated to another chromosome after the gene duplication occurred.

Moreover, the phylogenetic results demonstrated that the digestive lipase genes evolved independently between mammals and fishes (Figures 3, 4). The pancreatic lipase genes diverged into two groups between mammals and fishes, all PL genes of mammals were grouped together as the outgroup, and all pl genes of fish were grouped together (Figure 3). The same situation also could be seen in bile salt-activated lipase genes (Figure 4). The study suggested that the main digestive lipase gene might diverge into two types from the common ancestor of mammals and fishes. The spotted gar is a representative of holosteans, it diverged from the teleosts before the teleost genome duplication (TGD), and the genome of spotted gar has conserved from bony vertebrate ancestors (Amores et al., 2011; Pasquier et al., 2017; Martin & Holland, 2017). So the genome of the spotted gar was considered to be closest to vertebrate ancestors in this study. Therefore, we inferred that the pl gene and bsal gene both had the lipolytic ability in the vertebrate ancestors, and then the main digestive lipase gene evolved into different types between the mammals and fishes (Figure 9). The PL gene might play an important role in lipids digestion in mammals (Aloulou et al., 2014; Zhu et al., 2021), and the PL had occurred the gene tandem duplication in order to satisfy the lipids digestion ability (Figure 1). In contrast, the bsal might play a major function in lipids digestion in the ancestor of teleosts. Previous studies suggested that the massive gene deletions appeared after the third WGD (TGD) in teleost fishes and a large number of genes have been lost mostly through pseudogenization in rainbow trout after the fourth WGD (Berthelot et al., 2014; Inoue et al., 2015). So, we inferred that the pl gene might be absent (eg. zebrafish and medaka) or become a pseudogene (eg. rainbow trout and Atlantic salmon) with the WGD events that happened in the ancestor of some fish species. Although most of the fish species retained the pl gene, it might become an inactive pl gene, e.g., mandarin fish, seabass, and pufferfish. So, the bsal gene might be the main digestive lipase gene in fish and it is retained in all the teleost fishes. Moreover, the bsal gene might have extra two to five gene copies in order to satisfy the lipids digestion ability in all the teleost fishes. The bsal gene seemed to be more important in the teleost fishes, it might be related to the ability to hydrolyze polyunsaturated fatty acids (PUFAs) from triacylglycerol, but the pancreatic lipase can not hydrolyze those PUFAs (Bogevik et al., 2008; Anderson et al., 2018; Cai et al., 2017; Gilannejad et al., 2019, 2020). It also has been reported that bsal is far more efficient in hydrolyzing PUFAs from TAGs than pl (Sæle et al., 2010). As we all know, PUFAs are very important in fish larval development (Benedito-Palos et al., 2014; Fuentes-Quesada & Lazo, 2018; Arantzamendi et al., 2019). Thus, bsal might be a suitable digestive lipase for most fishes.

In order to get more information about the digestive lipases in fishes, 3D structures were used to analyze the digestive lipases. The results showed that the 3D structures of pancreatic lipases were conserved in mammals and fishes, and the lid domain, β-9 loop, and active site catalytic triad of Ser-His-Asp were present in all species (Figure 7). A previous study also indicated that the active sites and 3D structure of three PL genes were highly conserved in mammals (Aloulou et al., 2014). Although there had some controversies and doubts about pancreatic lipase in fish (Rønnestad et al., 2013; Yanes-Roca et al., 2018; Anderson et al., 2018), the 3D structure of pancreatic lipases in the present study demonstrated the pancreatic lipase really existed in fish, and the structures of pancreatic lipases were conserved from holosteans to teleosts. Therefore, we believed that the pl gene also really existed in fish, and the “plrp” gene in some fish species was a copy of pl gene, it might have a similar function to the pl gene. As we can see, the loop structure of bile salt-activated lipase appeared in the bile salt-binding site in most of the fish species (Figure 8). The function of this loop is related to the activation of Bsal (Holmes & Cox, 2011). Specifically, the active site catalytic triad of Ser-His-Asp is centrally located within the Bsal structure and is partially covered by this loop (Hui & Howles, 2002). The bile salt activates Bsal by binding to the loop domain (a relatively short amino acid residue: GANFLXNYLY) (Wang et al., 1997; Murray et al., 2006; Holmes & Cox, 2011; Cai et al., 2017), bile salt trigger the lipase conformational changes and frees the active site allowing the access of water-insoluble substrates (lipids) (Karimi et al., 2010; Kurtovic et al., 2011). Although the amino acid residues of the bile salt-binding site (GANFLDNYLY) were correct in Bsal.1 of mandarin fish, the loop structure does not appear in the 3D structure. Lipases favor their interactions with TAGs at the interface through various interfacial phenomena and processes (Rueda-López et al., 2017), so the absence of loop structure might affect the lipolytic activity of Bsal.1 in mandarin fish. Meanwhile, the loop structure appeared in the Bsal.2 of the mandarin fish (see Supplementary Figure S4), but the amino acid residues of the bile salt-binding site had changed (DVAVLGNSLY) (Figure 6), which might also cause the bile salts could not bind to the loop domain. The special Bsal structure of the mandarin fish might be related to its special feeding habits, which only prey on other fish larvae but do not eat any zooplankton since the first feeding stage (Doi & Aoyama, 2004; Liang et al., 2008). It has been reported that the marine carnivorous fishes possess higher lipase activities than the herbivorous and omnivorous fish species (Zambonino-Infante et al., 2008; Wu et al., 2010; Heras et al., 2020), because they have adapted to ingest zooplankton rich in PUFAs from the larval stage (Rueda-López et al., 2017). The ancestor of the mandarin fish also is marine fish, which evolved in freshwater in East Asia (Li, 1991). So, we inferred that the structure and function of bile salt-activated lipase were changed in the ancestor of the mandarin fish in order to adapt to the freshwater environment.

In conclusion, we searched and identified the two types of important digestive lipase genes in mammals and fishes. The results showed that the PL gene occurred in the tandem duplication but the BSAL gene retained the one copy in mammals. In contrast, only one copy of pl gene was found in most of the fishes, the pl gene was absent or became a pseudogene in some fish species. However, the bsal gene existed in all fish species, even two to five copies of bsal gene were found in most fishes. Although the 3D structures of pancreatic lipase were conserved in all species, the amino acid sequences analysis showed that the Ile and Leu were absent in the lid domain and the β-9 loop might influence the lipolytic ability of pancreatic lipase in most of the fish species. But the key amino acid residues (bile salt-binding site and catalytic triad) and 3D structures of bile salt-activated lipase were conserved in mammals and fishes. By combining the phylogenetic results, we inferred that the main digestive lipase gene evolved into two types between mammals and fishes. The pancreatic lipase might play an important role in lipids digestion in mammals and the PL occurred the gene tandem duplication. In contrast, the bile salt-activated lipase might play a major function in the lipids digestion in the fishes and the bsal occurred gene duplication in most of the fishes.