Xiao Xiang's team from Shanghai Jiaotong University and Hong Liang's team applied high-throughput protein structure groups to reveal the metabolic characteristics of the common ancestor of deep-sea hydrothermal archaea bacteria

Recently, Xiao Xiang's team, School of Life Science and Technology/State Key Laboratory of Microbial Metabolism, Shanghai Jiao Tong University, and Hong Liang's team from the Academy of Natural Sciences, took the isolation and cultivation of deep-sea hydrothermal superthermophilic archaea and thermophilic bacteria with ancient evolutionary status and unique metabolic commonality as the starting point, and with the help of self-improved deep learning AlphaFold2 technology and neutron scattering experiment, the high-throughput protein structure group was applied for the first time to study the metabolism of early life, revealing the metabolic characteristics of the common ancestor of deep-sea hydrothermal archaea bacteria
。 The research results, titled "Proteome-wide 3D structure prediction provides insights into the ancestral metabolism of ancient archaea and bacteria", were published in
the international authoritative journal Nature Communications 。 Assistant researcher Zhao Weiyin of the School of Life Science and Technology of Shanghai Jiao Tong University and Zhong Bozitao, an undergraduate student of the Institute of Natural Sciences, are the co-first authors of this paper, and Professor Xiao Xiang of the School of Life Science and Technology and Professor Hong Liang of the Institute of Natural Studies are the co-corresponding authors
of this paper.

Tracing and reconstructing the evolution process and early life forms with existing life is an important idea
for studying the origin and evolution of life.
Usually, researchers will speculate about the metabolic characteristics of early life based on genome comparisons, but highly variable genomes often struggle to clearly reveal evolutionary contexts and patterns
.
In this study, the 3D structure of nearly 10,000 protein molecules was predicted by applying high-throughput AlphaFold2 deep learning technology, and the protein structure group was introduced, and the metabolic characteristics
of the common ancestor of archaea bacteria were reconstructed by comparing and analyzing the sequence, structure and function of ancient superthermophilic archaea and representative strains of thermophilic bacteria with similar metabolic functions from deep-sea hydrothermal sources.
Surprisingly, the metabolic modules of these structures conserved in archaeal bacteria highly overlap with experimentally validated pre-life chemical processes, a discovery that provides new perspectives and new methods
for the study of the origin of life.
In addition, the flexibility coefficients of different species in this study were quantitatively characterized by neutron scattering experiments, and the relationship
between macroscopic flexibility and heat tolerance was established in archaea and bacteria in different growth temperature ranges.

Figure 1 The workflow of this study and the basic characteristics
of A501 (archaea, red) and 3DAC (bacteria, blue).
High-throughput structure prediction was established, protein structure groups were introduced, and the similarities and differences between sequences, structures and functions of deep-sea hydrothermal archaea bacteria were compared

In this study, the superthermophilic archaea Thermococcus eurythermalis A501 and the thermophilic representative bacteria Zhurongbacter thermophilus 3DAC were selected to compare the similarities and differences
between their sequences, structures and functions.
Both strains come from deep-sea hydrothermal vents, which are considered to be one of
the cradles of life.
Considering the sequence and protein structure similarities and differences of functionally similar protein pairs in the two strains, protein pairs can be divided into 3 different groups: (i) direct homologous sequences with similar structure and function; (ii) non-direct homologous sequences with similar structure and function; (iii) Non-direct homologous sequences
that are structurally distinct but functionally similar.
Some protein pairs are structurally similar, but the gene sequences are very different, further demonstrating that protein structure and function are more
related than sequences.
By mapping different sets of protein pairs into metabolic pathways, we found that protein evolution in both strains does not occur at the individual protein level, but in metabolic modules
.
For example, the proteins involved in the middle half of glycolysis of both strains (from DHAP to acetyl-CoA) belong to group (i), while proteins in the glycolysis and lipid biosynthesis linkage pathway (DHAP to G13P2) belong to group
(ii).
The metabolic modules thus divided can be used to reveal different sources and different evolutionary histories
of proteins.
This has never been found
in previous sequence-based analyses.

Further analysis of conserved protein structure revealed the conserved metabolic module
of bacteria and archaeal common ancestor (ABCA).
These conserved metabolic modules include the middle half of glycolysis (from DHAP to acetyl-CoA), the biosynthesis of purines and pyrimidines, the metabolism of some essential amino acids (i.
e.
, Asp, Glu, Ser, Gly, and Thr), the biosynthesis of some essential cofactors (i.
e.
, NAD(P)+ and CoA), energy respiration with MBH and MBS, most aminoacyl-tRNA ligases and some ribosomal proteins, etc.
, which are speculated to exist in the common ancestor ABCA.
It was then passed on to bacteria and archaea
.

Figure 2 Through the comparison of protein structure, it was revealed that the conserved metabolic modules in the common ancestor of archaea bacteria highly coincided with pre-life processes

Interestingly, this study found that the structure-conserved metabolic modules are highly overlapping with pre-life chemical processes, such as the conserved metabolic modules of central carbon metabolism are highly consistent with proven pre-biosynthetic pathways, and the reverse process of inferring the origin of life from existing life intersects with the positive process from chemical reactions to the origin of life, and it also proves from the side that the conserved metabolic modules obtained by protein prediction structure are no coincidence
.

In this study, high-throughput protein structure was applied for the first time to study the metabolism of early life, a new method of protein structure group was established, and it was found that the origin and evolution of protein was not isolated but based on metabolic modules, and the conserved protein structure revealed the conserved metabolic modules of the common ancestor, and for the first time found that the structure-conserved metabolic modules highly coincided with the pre-life chemical processes, providing a new research idea for the study of ancestor metabolic reorganization and the origin of life.
This partially verifies the potential value
of high-throughput protein structure prediction in the study of the origin and evolution of life.

This research was supported by the National Natural Science Foundation of China Innovation Group (41921006), Youth Fund (42106087) and General Projects (11974239, 31630002), Shanghai Artificial Intelligence Laboratory, Shanghai Science and Technology Commission, Shanghai Education Commission Major Project, Shanghai Pujiang Talent Program (22PJ1406900), as well as Shanghai Jiao Tong University Interdisciplinary Research Fund (YG 2016QN13) and "Deep Blue Program" (SL2021PT103).

Link to paper: https://doi.
org/10.
1038/s41467-022-35523-8