Physicochemical Foundations of Life that Direct Evolution: Chance and Natural Selection are not Evolutionary Driving Forces

Abstract

The current framework of evolutionary theory postulates that evolution relies on random mutations generating a diversity of phenotypes on which natural selection acts. This framework was established using a top-down approach as it originated from Darwinism, which is based on observations made of complex multicellular organisms and, then, modified to fit a DNA-centric view. In this article, it is argued that based on a bottom-up approach starting from the physicochemical properties of nucleic and amino acid polymers, we should reject the facts that (i) natural selection plays a dominant role in evolution and (ii) the probability of mutations is independent of the generated phenotype. It is shown that the adaptation of a phenotype to an environment does not correspond to organism fitness, but rather corresponds to maintaining the genome stability and integrity. In a stable environment, the phenotype maintains the stability of its originating genome and both (genome and phenotype) are reproduced identically. In an unstable environment (i.e., corresponding to variations in physicochemical parameters above a physiological range), the phenotype no longer maintains the stability of its originating genome, but instead influences its variations. Indeed, environment- and cellular-dependent physicochemical parameters define the probability of mutations in terms of frequency, nature, and location in a genome. Evolution is non-deterministic because it relies on probabilistic physicochemical rules, and evolution is driven by a bidirectional interplay between genome and phenotype in which the phenotype ensures the stability of its originating genome in a cellular and environmental physicochemical parameter-depending manner.

Keywords: Darwinism; Evolution; Origin of life; RNA; biophysics.

Figure 1

(A) The origin of life probably relies on simpler forms of organization than those observed in modern living organisms. The commonly accepted hypothesis postulates that the origin of life corresponds to the emergence of polymers, such as RNAs and proteins. These two polymers are interdependent as RNAs serve as templates for protein synthesis (1) and proteins are necessary for RNA synthesis (2). This interdependence, which can be represented in the form of feedforward and feedback loops between the proto-genome (RNAs) and the proto-phenotype (proteins) can only be maintained if proteins relax environment-dependent physicochemical constraints triggering for example RNA degradation (3). This interdependence is the foundation of life and evolution. (B) Variations in the extracellular environment induce constraints on cellular polymers sensitive to these variations, triggering the cellular response that results in the biogenesis of gene products, whose activity relaxes the initial constraints (1). However, if these variations exceed a certain amplitude or persist over time, they challenge the integrity of the targeted polymers, which ultimately lead to mutations. This process stops only when new sequences (directly or indirectly) relax the initial constraints (2). (C) A DNA molecule is subjected to environmental fluctuations of physicochemical parameters, which triggers the biogenesis of polymers (gene products) whose activities correspond to the phenotype. Cellular activities (the phenotype) allow a return to equilibrium by relaxing the initial constraints (1). Nevertheless, these activities also generate constraints directly or indirectly on their originating genome, meaning that a genome is adapted to the constraints generated by its own activities (2). (D) According to the current framework of evolutionary theory (left panel), there is no direct relationship between physiological and genetic adaptation because physiological adaptation is based on physicochemical principles of homeostasis as a function of environmental fluctuations, while genetic adaptation would be fueled by random mutations generating a diversity of phenotypes on which natural selection acts. In contrast, in the model proposed in this article (right panel), genetic adaptation is the consequence of physiological adaptation. Indeed, physiological adaptation can take place as long as fluctuations in environment-dependent physicochemical parameters do not exceed a certain threshold. Above this physiological threshold, the integrity of nucleic and amino acid polymers, in particular DNA, is challenged which leads to targeted mutations. This mutational process stops when the mutations generate a phenotype that maintains the integrity of the DNA with regard to environmental constraints (genetic adaptation).

Figure 2

(A) The amino acid or nucleotide composition of proteins (left) or DNA (right), respectively, determines the physicochemical properties of these polymers with consequences on their physical and chemical properties. IDR, intrinsically disordered region and ssDNA, single-stranded DNA. (B) The composition of a polymer determines its physicochemical properties, and therefore its folding and physical resistance to specific constraints. Depending on the composition of a given polymer, some physicochemical constraints induce reversible structural changes (blue arrows), and others induce irreversible damages (e.g., aggregation and breaks) (red arrows). (C) Chemical modifications of amino acids (left) or nucleotides (right) change the physicochemical properties of polymers. These chemical modifications are reversible (blue arrows) or induce irreversible damages (red arrows). (D) Any given polymer (blue) is stable in a physicochemical environment and unstable in another one (1 vs. 2). A different polymer (red) reacts differently under the same constraints (3 vs. 2). (E) Composition determines the physicochemical properties of nucleic or amino acid polymers. Nucleotide or amino acid composition is constrained by physicochemical parameters, which suggests that their composition must correspond to the same fundamental physicochemical parameters as the sequence of nucleic acid polymers determines the composition of proteins.

Figure 3

(A) In an RNA world, an RNA molecule is replicated thanks to the product of replication (e.g., ribozyme, left panel). This process is enhanced by amino acids or small peptides (right panel). (B) In an RNP world, replication (RNA production) and translation (protein production) could have been performed simultaneously as in modern prokaryotes, thereby avoiding the nascent RNA to interact back to the template and increasing the probability that the neo-synthetized protein interacts with the RNA replication product. (C) Cotranslational replication in an RNP world could have be performed owing to amino acids attached to proto-tRNAs that enhanced the polymerization of proto-tRNAs and that were simultaneously incorporated into the nascent protein. (D) Two different proto-cells (red and green circles) growing in different chemical environments (e.g., N-poor vs. N-rich environment) could have developed different proto-genetic codes. Their cooperation in fluctuating environments could have led to horizontal transfer, leading to the emergence of the universal genetic code. (E) Nucleic acid polymers are often represented as linear strings of letters that are translated into proteins with physicochemical properties unrelated to those of nucleic acid polymers. If the genetic code has been constrained over evolutionary time to match nucleic acid polymers and their cognate amino acid polymers to the same fundamental physicochemical constraints (e.g., temperature, element availability), nucleic and amino acid polymers share more physicochemical properties than previously anticipated. (F) In an RNP world, cooperation between interdependent polymers (i.e., RNAs and proteins) relies on their activities toward the biogenesis of nucleotides and amino acids necessary for their synthesis. The phenotype of a proto cell in an RNP world corresponds to the polymerization of nucleotides and amino acids. The polymerization products produce nucleotides and amino acids. (G) RNAs give rise to proteins, which allow the biogenesis of RNAs and metabolites through transformation of molecules captured from the environment (blue lines). RNAs probably played a role, at least early in evolution, in metabolite biogenesis (blue broken lines). Metabolites are required, in turn, to give rise to RNAs and proteins (red lines). (H) Metabolic-dependent chemical modifications of nucleic and amino acid polymers contribute to the cell activities and to the cellular physiological adaptation in response to environment-dependent constraints (1). However, metabolic-dependent chemical modifications can also induce irreversible damages of nucleic and amino acid polymers (2).

Figure 4

(A) Variations in the extracellular environment increase the expression of target genes which is a process associated with physical constraints on DNA generated by RNA polymerases (RNAP). These physical constraints are transient if the gene activity relaxes the initial constraints (e.g., through the biogenesis of gene products) (1). If not, physical constraints persist, and conflicts between RNAP and DNA polymerases (DNAP) induce DNA damages. DNA damage is repaired by homologous recombination, which favors GC over AT nucleotides, but which can also induce gene copy number variation (CNV). Both of these processes, in turn, generate more gene products, and therefore relax the initial constraint (2). (B) Cellular stresses activate the expression of specific target genes while increasing the production of intracellular ROS. One strand of transcriptionally induced genes is exposed to ROS, promoting deamination of methyl cytosine, which gives rise to thymine. C > T mutations change Arg codons into Trp and Cys codons. Trp- and Cys-containing proteins can play a role in protecting cells from ROS, therefore, relaxing the initial constraints (2). (C) Transcriptional activation of genes induces neo-insertion of repeated sequences within transcription-dependent ssDNA. Neo-insertion of repeated sequences can facilitate co-regulation of two genes by bringing them closer to each other in space, and it also promotes recombination. Both processes coordinate the production of gene products and contribute to relaxing the initial constraints. (D) Cellular stress induces chemical modifications of target genes, which affects chromatin organization and transcription fidelity (“epi-mutations”). Chemical modifications that induce biogenesis of new RNAs and proteins could allow survival of the cells in which these modifications took place. Chemical modifications in surviving cells lead to mutations that increase the survival probability of daughter cells. (E) Cellular stress induces physical constraints simultaneously on target genes and on proteins produced from these genes (1 and 2). By disrupting translation, for example, by inducing nascent protein unfolding, a stress can induce translation stopping and cotranslational cleavage of mRNAs. RNA fragments generated during translation or transcription hybridize on the complementary DNA strand and locally generate DNA or chromatin modifications, thus, increasing the probability of mutations occurring in the targeted regions (3). This process stops only when the gene and its products obtain physicochemical properties that relax the initial constraints (4). (F) Somatic cells are constrained by environmental fluctuations, and their activities can have consequences for germ cells, for example, through the transfer of metabolites or small RNAs from somatic to germline cells (1). These compounds can change the activity of germ cells, affect the development of the body after fertilization, and cause mutations in germ cells (2 and 3). These mutations could lead to the emergence of somatic cells (3 and 4) whose activities maintain the integrity of the germ cell genome of the following generations. (G) Somatic cell genes that are constrained by environmental parameters produce extracellular vesicles containing RNAs that can either induce reversible epigenetic mutations or irreversible genetic mutations.

Figure 5

(A) The environment generates physicochemical constraints which, by destabilizing genomes, induce molecular innovations, allowing the emergence of new physicochemical properties at different scales of life organization. These molecular innovations and new properties are transmitted across generations if they release the initial constraints (blue lines) but can also generate new constraints (red lines), allowing the adaptation of the genome to its own activities and generating an evolutionary dynamic. (B) Any biochemical reaction leads to the synthesis of a final product (1) and “waste”, by-products, or secondary metabolites. These compounds are not necessarily essential to the cell’s survival, but they are the result of vital cellular activities. Therefore, these compounds are not “random products” even though they can have cellular toxic effects by interacting with cellular polymers (2), which can lead to mutations. (C) UV radiations induce genomic instability favoring genomes that produce pigments that, in turn, protect them from the initial constraint (1). Likewise, UV radiation-absorbing pigments induce genomic instability favoring genomes that produce photosynthetic reaction centers that, in turn, protect them from the initial constraint (2). Likewise, O₂ production by photosynthesis induces genomic instability favoring genomes that produce oxidases and cellular components that, in turn, protect them from the initial constraint (3). (D) Two unicellular organisms (green and red), which are adapted to different environment-dependent constraints, cooperate by exchanging various components in a fluctuating environment. (E) Not all enzymatic reactions generated from a genome can take place simultaneously (left panel). The selective compaction of different regions of a genome according to the cellular metabolic state (and therefore its environment) through chemical modifications of histones protecting some genome parts and repressing their potentially toxic expression while allowing the expression of genes whose products contribute to maintain the cellular homeostasis. (F) Somatic cells “buffer” environment-dependent constraints, maintaining the stability of the genome from the germ cells. When “protected” by somatic cells (i.e., when the phenotype of the organism is adapted to its environment), germ cells give rise to gametes that generate after fertilization the same somatic cells, i.e., phenotype. (G) Cold induces physiological adaptation by activating cellular respiration that increases cellular heat production (1). However, an increase in cellular respiration increases ROS production, which can lead to activation of UCP proteins. This simultaneously balances the ROS genesis and increases heat production in muscle cells (2) or in brown adipose tissue (3). The loss of UCP1 in bird ancestor may have led to muscle hyperplasia for heat production with consequences on bird body plan (4).

Figure 6

(A) A genome (G) generates a phenotype (P) that maintains the stability of its originating genome in a stable environment and both (genome and phenotype) are reproduced identically (left panel). In an unstable environment (corresponding to variations in physicochemical parameters above a physiological range), the genome generates a phenotype (P*) that no longer maintains the stability of its originating genome and instead triggers mutations whose rate, nature, and location dependent on the initial constrains and the phenotype (right panel). This process occurs until new genetic variants (G**) generate a phenotype (P**) that maintains the stability of its originating genome. (B) In a given environment, a genome (G) gives rise to a range of phenotypes (P), and similar phenotypes can correspond to a range of genomic sequences. Each phenotype can be adapted to a range of environmental physicochemical constrains. (C) Evolutionary driving forces rely on physicochemical processes whose probabilistic nature generates genetic and phenotypic diversity, as symbolized by arrows from 1 to 5. If a path (exemplified by Path 1) does not stabilize its originating genome, the genome and the corresponding phenotype will not be reproduced. While some paths might be neutral in terms of natural selection (Paths 2 and 3), some paths (Paths 4 and 5) could lead to the emergence of phenotypes that can be under natural selection.