English translation of my 1984 paper on the origin of life on Earth. Part I
View on Lipids of Microorganisms from the Standpoint of Prebiotic and Biological Evolution* PART I
Leonid Andreev
Part II: https://systemity.livejournal.com/4661607.html
Original paper "View on Lipids of Microorganisms from the Standpoint of Prebiotic and Biological Evolution" published in: Voprosy Evolutsii Bakterij (Evolution of Bacteria), USSR Academy of Sciences, Center for Biological Research, Institute of Biochemistry and Physiology of Microorganisms, Pushchino, 1984, pp. 93-119 (https://www.dropbox.com/s/tx666ilga018yvt/OriginOfLifeRus.pdf).
This translation was first published in 2004 on the company website of Equicom, Inc. which is no longer online.
Contents:
• Specificity of lipids as an object of molecular biological studies
• Cellular level of lipid studies
• The concept of polylipids
• Interaction between polylipids and proteins
• Role of fatty acid residues
• Enzymes and coenzymes
• Principle of quasi-equilibrium of biosynthesis of bacterial lipids
• On the origin of life on Earth
• Macromolecular chronometry
• Molecular model of a protobiont
• Conclusions
Specificity of lipids as an object of molecular biological studies
Modern molecular biology incorporates a number of research areas dealing with substances and processes of general biological significance. Those areas of research, stimulated and inspired by the efficacy of the currently available physico-chemical methods, considerably differ in methodologies and - what is especially important - due to their specific developmental backgrounds, they have different levels of relationship with cellular biology.
When that relationship is lost or is not yet strong enough, it is often understood as an indication that some of the problems of physical chemistry of bioorganic molecules may be as broad and complex as the issues encountered with in the study of the functioning of live cells. This concerns particularly investigations of biological macromolecules that are functionally active outside the cells that synthesize them. That kind of misconception is not accidental. The awareness of the fact that, despite a theoretically possible large variety of structural and functional organizations of biopolymers, the Nature has only a limited number of their variants, impedes the advance of researchers in physics and chemistry of the functionality of such molecules, as it makes them divert to working on problems which require qualitatively different approaches and expertise and force them to study such properties of organisms which require deep empirical knowledge. This is a psychological reason that explains why many molecular biologists at least sympathize with, if not fully concede to, the thesis that the notion of ‘cell’ “has become a brake on the progress toward the understanding of live structures at the molecular level” and that “one may stop treating the cell as a biological unit but consider it as merely one of the stages of a complex chain of transformations” [16]*.
In practice, such views may seem to be justified as the attempts not to limit oneself to the “barest necessities” of evolutionary biology and “biological purposefulness” and, instead, consider the cell to be “merely one of the stages of a complex chain of transformations” may often be helpful in extensive fundamental investigations in molecular biology, leading to innovative approaches to various aspects of the functioning, systematics and evolution of live organisms. A classical example of such fundamental works in molecular biology that already by now have significantly contributed into the progress in evolutionary biology, is the method of macromolecular chronometry developed a quarter of a century ago [34] based on physico-chemical, rather than biological, logic.
_________________________________________________________
* Rendition is based on the Russian edition of the source book.
Notwithstanding the success in certain areas of molecular-biological studies concerned with the phenomenon of life, their independence from the general biological knowledge and cellular biology is only an illusion caused by the breathtaking new capabilities that look almost like a sudden materialization of something that only yesterday seemed to be not more than an idealized goal, set merely for justification of a fundamental research project. This is especially clearly exhibited in one of the areas of molecular biology - the study of lipids. The structure of lipids is less complex as compared to proteins, nucleic acids, or polysaccharides, which is why the study of lipids of microorganisms has fairly quickly advanced in accumulating molecular-biological information, which, in its turn, helped to see the narrowness and, oftentimes, senselessness of investigations that are disconnected from the problems of cellular biology. Some ten years ago, O’Leary wrote to the effect that not long ago the studies on microbial lipids were basically boiling down to investigating almost exclusively the chemical nature of those compounds and their biosynthesis.
Undertaken by many researchers within the past 20 years, those investigations unexpectedly appeared to be difficult, and the results they yielded were in many ways surprising. This field has already gathered sufficiently large amounts of information, and it continues to grow. Such a wealth of information and data on biosynthesis, although valuable in itself, is not much of an asset, as it says a lot but does not explain much. We have found ourselves in the role of observers solemnly gazing at the fascinating amount of data which shows how much we know about microbial lipids and how little we understand about them. It gets clearer to us what the lipids are and how they are organized, but - why are they needed? What does it all mean for the cell functioning? [12]**
Since the time these questions had been put, both the capabilities of theoretical and experimental lipidology and the information on lipids have dramatically expanded. But still again - in the preface to Membrane Lipids of Prokaryotes published in 1982 [20] - K. Blokh, a well-known specialist in the field, puts the same questions. He states that overabundance, rather than thrift, characterizes the composition of membrane lipids occurring in the Nature. In terms of the structural common denominator of all the components of the membrane lipid bilayer, it would suffice to have just one phospholipid that has an amphipathic structure and is capable of participating in formation of closed vesicles.
In the meantime, there is no cellular membrane in the Nature which would have only one phospholipid. “From the standpoint of compositional complexity of natural membrane lipids”, further writes K. Blokh, “it is remarkable that chemically homogenous liposomes imitate many of the properties of natural membranes, including transport, phase transitions, or the effect on membrane-bound enzymes. Clearly, there must be a great variety of membrane-dependent phenomena, expressed only in cells, which cannot be discovered by studying a single-component membrane model” [20]. Despite a relative simplicity of the lipid structure, in the Nature there are lots of structurally individual forms of lipids. For instance, bacteria-synthesized lipids consist of several classes, of which phospholipids are most widely occurring. There are over two dozens of various phospholipids that differ by a radical attached to the phosphate group. Rather than being an individual substance, each of them is a set of substances, which differs from others by the kind of fatty acid (some bacteria have up to several dozens of various fatty acids) with which the C1 and C2 hydroxyls of glycerol are esterified. Esterification of hydroxyl groups of glycerophosphate can be incomplete, in which case the so-called phospholipid lysoforms are produced. Hydroxyl groups of glycerophosphate can be bound to saturated and unsaturated alcohols, aldehydes and oxyacids.
There are two factors that contribute into diversity of the lipid composition of bacteria. Firstly, the number of structurally individual lipids and the ratio between them may vary - apparently, indefinitely - in bacterial species; and, unlike lab-synthesized lipids, bacterial lipids are not randomized. In paraphrase, each group of lipids has an unpredictably large variety of distributions of molecular species. Secondly, the group and fatty acid compositions of phospholipids of most of bacteria vary extremely widely and according to very complex patterns dependent on the physiological condition of a cell population, which, in its turn, depends on a variety of environmental factors. Oftentimes, such variations in phospholipids composition can be considered as qualitative [5].
For the said reasons, an exhaustive description of lipids of even one species of bacteria is quite complicated and uneconomic in terms of the time and effort involved, given that all those details and nuances have no practical application within the context of the currently existing theories on the functional role of lipids in bacterial cells. This situation is clearly a result of the overestimation of the capability of exact sciences to solve the problems of cellular biology at the molecular level. Contemporary theories on the functional role of lipids in bacterial membranes are mostly of the physico-chemical, rather than biological, nature, and are based on interpreting such integral properties of lipids as the high degree of reduction (lipids as energy accumulators), spatial dissociation of hydrophilic and hydrophobic groups, capability for phase- isolation in aqueous
environment, dependence of the degree of spatial freedom of fatty acid radicals on their structure, etc. The physico-chemical approach to the study of natural lipids has helped to solve a number of important problems, including, first of all, production and utilization of artificial biological membranes. As well, it significantly contributed into the progress of technology of lipid studies, and many of the results obtained by physico-chemical methods are successfully used in biology and medicine.
Nevertheless, it would be fair to say that, in the long run, the study of biological role of lipids has suffered serious damage as a result of preoccupation with physico-chemical methods. This has become especially clear when physico-chemical lipidology, having failed to offer a more or less credible explanation to the astonishing diversity of the lipid spectra in living organisms, has changed its position toward downplaying the significance of this crucially important fact. There came a new wave of studies whose authors were determining the melting temperature of lipids isolated from various microorganisms, assuming that the cell lipid metabolism, as intricate as it is, is responsible mainly for maintaining a required level of membrane liquidity [21, 30]. For quite a long time, this inappropriately naïve idea kept attracting many researchers as
___________________________________________________________________
** Back translation from the Russian edition of the source book.
an affordable way to obtain integral quantitative characteristics of a cell lipid pool, although absolutely helpless in terms of explaining the uniqueness of lipid spectra in bacteria which avail themselves of thousands of ways to create a same physical effect. This phenomenon is an indication that phase transitions in bacterial membranes are by no means the main cause of purposeful changes in lipid composition - instead, they are one of the effects of those changes. The idea about the defining role of the physical condition of membranes is closely connected with the works on physiology and biochemistry of bacteria artificially rendered unable to synthesize certain lipids [28]. It was shown, for instance, that bacteria can grow and develop even when their membranes are artificially supplied with mixtures of lipids that they cannot not naturally synthesize.
It runs through all of those studies that specificity of lipid composition of bacteria is not life’s necessity. However, as that idea was further developing, it eventually has led to an absolutely opposite result - a realization that there must be lots of membrane-dependent phenomena that are exhibited only in cells and cannot be detected by experimenting with physico-chemical models, even those that imitate a live cell.
Proteins and nucleic acids can be studied in respect of their functionality, without going deep into fine details of the biology oforganisms they are isolated from. With lipids, such an approach seems to be practically senseless, and that is why it is not surprising that, despite the tremendous progress in deciphering the fine mechanisms of functioning of proteins and nucleic acids, the function of lipids remains as unclear as it was decades ago. Functionally, lipids are inseparable from organisms that synthesize them. This is the main peculiarity of lipids as objects of molecular-biological studies, which requires the whole methodology of lipid studies to be revised so as to restore its connection with the cellular level.
Cellular level of lipid studies
The cellular approach to the study of lipids means the establishing of functional relations between the lipid composition and its regulation, on the one hand, and biological (morphological, physiological, ecological, and other) characteristics of organisms that synthesize them, on the other. Taking a full advantage of that approach involves two serious but surmountable challenges. The first one is the need of solid knowledge in biology of microorganisms under study, which can be resolved through research cooperation. The second challenge comes from is a virtually complete lack of an adequate theoretical basis. As a minimum, we need hypotheses explaining the fantastic diversity of lipid spectra of bacteria, let alone the capability to predict, based on bacteria known properties, lipid compositions of specific bacteria. Properties of lipids should be studied in close connection with biological peculiarities of cells of bacteria that synthesize them, as this will help to bridge the gap between theory and experiment. Functional integrity of the lipid pool of living cells is especially valuable in the context of this problem.
Evolution of bacteria was accompanied with the restructuring of their membrane system with its inherent integrity of the structure and functions. As well, integrity is haracteristic of lipids of live cell membranes, which can be proved on various examples, some of which are discussed in [2, 4-7, 13]. Primarily, it is expressed in the character of adaptational changes in membrane lipids in the course of bacterium growth and development, in response to changes in the environment parameters. In such cases, the lipid pool of live cells responds as an integral system, acting by fairly complex rules that are certainly impossible to figure out based on lipid physico-chemical properties solely.
Integrity of the lipid pool is also expressed in the fact that evolutionary modifications of certain biological properties of bacteria were accompanied with coordinated changes in lipid compositions, which were similar in different bacteria, irrespective of their systematic position and ecology. Functional integrity of the lipid pool of live cells and functional inactivity of lipids isolated from the cell, as well as the diversity of lipid spectra, should be taken into account when designing a model claiming to explain the biological role of lipids.
The concept of polylipids
According to this concept, phospholipids of cell membranes form cellular polymeric structures [1, 5]. In the ideal case, a polylipid of the bacterial cytoplasmic membrane can be considered a molecule of the same size as a bacterial cell. Phospholipids form the skeleton of the native membrane - a so-called polylipid membrane. Monomer phospholipids of the polylipid membrane are bonded mainly by principal valence forces. Fig. 1 shows a schematic view of a polylipid membrane built from molecules of phosphatidyl ethanolamine (PE) and cardiolipin (CL). The basic monomer unit in such a membrane is a dimer in which PE molecules are connected through a metal atom in such a way that fatty acid residues of both PE molecules are oppositely oriented. Another type of bond between monomer phospholipids (PE dimers) is formed by substitution of protons of enolized ester carbonyls by ions of polyvalent metals (mostly, Mg). Such bonds, particularly with the involvement of Mg, have low conformational freedom and can be broken by the attachment of protons. Two layers of polylipids, organized in the way shown in Fig. 1, form the membrane’s polylipid bilayer whose stability is determined by potassium and sodium atoms in the intra-bilayer space. These atoms interact with ester carbonyls of phospholipids of both monolayers. This interaction involves also amino groups of phosphatidyl ethanolamine, whose degree of involvement determines the ratio between Mg and Na in the inter-bilayer space.
Proteins and nucleic acids can be studied in respect of their functionality, without going deep into fine details of the biology of organisms they are isolated from. With lipids, such an approach seems to be practically senseless, and that is why it is not surprising that, despite the tremendous progress in deciphering the fine mechanisms of functioning of proteins and nucleic acids, the function of lipids remains as unclear as it was decades ago. Functionally, lipids are inseparable from organisms that synthesize them. This is the main peculiarity of lipids as objects of molecular-biological studies, which requires the whole methodology of lipid studies to be revised so as to restore its connection with the cellular level.
Cellular level of lipid studies
The cellular approach to the study of lipids means the establishing of functional relations between the lipid composition and its regulation, on the one hand, and biological (morphological, physiological, ecological, and other) characteristics of organisms that synthesize them, on the other. Taking a full advantage of that approach involves two serious but surmountable challenges. The first one is the need of solid knowledge in biology of microorganisms under study, which can be resolved through research cooperation. The second challenge comes from is a virtually complete lack of an adequate theoretical basis. As a minimum, we need hypotheses explaining the fantastic diversity of lipid spectra of bacteria, let alone the capability to predict, based on bacteria known properties, lipid compositions of specific bacteria.
Properties of lipids should be studied in close connection with biological peculiarities of cells of bacteria that synthesize them, as this will help to bridge the gap between theory and experiment. Functional integrity of the lipid pool of living cells is especially valuable in the context of this problem. Evolution of bacteria was accompanied with the restructuring of their membrane system with its inherent integrity of the structure and functions. As well, integrity is characteristic of lipids of live cell membranes, which can be proved on various examples, some of which are discussed in [2, 4-7, 13]. Primarily, it is expressed in the character of adaptational changes in membrane lipids in the course
of bacterium growth and development, in response to changes in the environment parameters. In such cases, the lipid pool of live cells responds as an integral system, acting by fairly complex rules that are certainly impossible to figure out based on lipid physico-chemical properties solely.
Integrity of the lipid pool is also expressed in the fact that evolutionary modifications of certain biological properties of bacteria were accompanied with coordinated changes in lipid compositions, which were similar in different bacteria, irrespective of their systematic position and ecology. Functional integrity of the lipid pool of live cells and functional inactivity of lipids isolated from the cell, as well as the diversity of lipid spectra, should be taken into account when designing a model claiming to explain the biological role of lipids.
The concept of polylipids
According to this concept, phospholipids of cell membranes form cellular polymeric structures [1, 5]. In the ideal case, a polylipid of the bacterial cytoplasmic membrane can be considered a molecule of the same size as a bacterial cell. Phospholipids form the skeleton of the native membrane - a so-called polylipid membrane. Monomer phospholipids of the polylipid membrane are bonded mainly by principal valence forces. Fig. 1 shows a schematic view of a polylipid membrane built from molecules of phosphatidyl thanolamine (PE) and cardiolipin (CL). The basic monomer unit in such a membrane is a dimer in which PE molecules are connected through a metal atom in such a way that fatty acid residues of both PE molecules are oppositely oriented. Another type of bond between monomer phospholipids (PE dimers) is formed by substitution of protons of enolized ester carbonyls by ions of polyvalent metals (mostly, Mg). Such bonds, particularly with the involvement of Mg, have low conformational freedom and can be broken by the attachment of protons.
Two layers of polylipids, organized in the way shown in Fig. 1, form the membrane’s polylipid bilayer whose stability is determined by potassium and sodium atoms in the intra-bilayer space. These atoms interact with ester carbonyls of phospholipids of both monolayers. This interaction involves also amino groups of phosphatidyl ethanolamine, whose degree of involvement determinesthe ratio between Mg and Na in the inter-bilayer space.
A bilayer whose polylipid structure suffers a damage can transform itself into a lamellar-type bilayer in which phospholipids are positioned back-to-back with their nonpolar radicals, whereas their polar heads are directed outward, according to the Gorter-Grendel model proposed in 1925 and further developed by Dawson and Danielli. A lamellar-type bilayer is not a polylipid since it is organized according to physical, but not chemical, interaction of monomer units. The biological membrane of eubacteria is structured by the polylipid layer both lengthwise and crosswise. Qualitative composition and proportions of monomer phospholipids determine to the large extent mechanical and physico-chemical properties of the polylipid skeleton of the membrane and underlie the diversity of their molecular organization. Both in Gram-positive and Gram-negative bacteria, diphosphatidyl glycerol (or cardiolipin (CL)) is the cross-agent that splices individual polylipid chains. As it follows from the general principles of the model, polylipid monolayers can change their positions relative to each other, using alkali metal ions as a lubricant. However, the transfer of monomer phospholipids from one polylipid monolayer to another occurs only if the polylipid skeleton is destroyed and a lamellar-type bilayer is constructed.
Interaction between polylipids and proteins
The concept of polylipids, even in the above-presented form, allows the understanding of the cause of the qualitative and quantitative diversity of lipid spectra of such organisms as bacteria whose cytoplasmic membrane performs practically all of the membrane-dependent functions; it also explains the diversity of the forms of bonds between proteins and membranes. Based on the model of molecular organization of the biological membrane, we will discuss a possible mechanism for protein embedding into the membrane’s polylipid skeleton. When contacting the surface of the bilayer polylipid membrane, protein interacts with it through weak van der Waals forces emerging when the protein non-polar amino acid radicals come into contact with short fragments of fatty acid radicals “sticking-out” on the surface of the polylipid layer (cf. Fig. 1). When the cumulative amount of that interaction reaches a certain critical value, the pressure, directed onto the surface of the polylipid skeleton, destroys it in the site of protein embedding. This effect is determined by the limitedness of conformational freedom of the metal (magnesium)-lipid complexes, due to which even a small deformation of the
lipid skeleton causes a dramatic increase of reactive capacity of M-O (metal - oxide) bonds whose disruption results not only in the embedding of proteins but also in their covalent bonding with lipids.
Thus, a biological membrane polylipid layer serves as both an accumulator and amplifier of weak interactions that are determined by the spatial structure of the embedding proteins, on the one hand, and the lipid composition at the embedding site, on the other. Anisotropy of protein embedding into the membrane is determined by the asymmetry of the polylipid bilayer. During the embedding process, the weak quantum-mechanical interactions between lipids and proteins are transformed into interactions with the breaking and formation of primary valence bonds. The strength of the both types of primary valence bonds between monomer phospholipids is determined by numerous factors, including the concentration of cations - primarily, protons. However, it should be taken into consideration that the polylipid membrane is a dynamic structure wherein primary valence bonds are being broken and restored. The structure of the polylipid skeleton is determined by not only the quality and quantity of individual lipids, but also - and maybe even to a higher extent - by the qualitative and quantitative composition of proteins and other components of biomembranes.
At the initial stage of protein embedding into the membrane polylipid skeleton, non-polar interactions between amino acid and fatty acid residues induce the restructuring of the polylipid skeleton to achieve a thermodynamically advantageous state of the maximal contact. Prior to the contact with a macromolecule to be embedded, the embedding site represents, due to high lateral flexibility of monomer phospholipids, a polymer with randomized distribution of monomers. However, after initiation of the contact, the distribution becomes more or less specific. Regardless of the extent of specificity - hence, the energy of non-polar interactions for protein embedding with the breaking of primary valence bonds - the protein contact with the polylipid membrane results in the decrease of the contents of individual phospholipids that are “complementary” to a given protein and the increase of those that are “non-complementary”. As a result, the changes in the individual lipid contents in the membrane appear to be closely connected with the changes in the protein composition of the membrane.
Role of fatty acid residues
The proposed model of polylipids brings us closer to solving the problem of the purposefulness of bacterial biosynthesis of fatty acids with different structures of fatty acid residues. Calculations indicate that at the dimer configuration of phospholipids in the membrane, only the terminal sections of fatty acid chains containing more than 12 carbon atoms are actually accessible for direct contact with amino acid residues of proteins. In most Gram-negative bacteria, fatty acids contain 14 to 18 carbon atoms. With the lamellar structure of the phospholipid layer, this difference range is negligible, and, therefore, it is unclear why a 5-15% change of the average length of fatty acid residues should correlate with substantial changes in morphology, physiology and biochemistry of bacteria. Based on the proposed model, however, the increase from 14 to 18 carbon atoms means a many-fold expansion of the area for lipid-protein contacts. The polylipid model demonstrates a great extent of rationality in the organization of lipid-protein interactions in Gram-positive bacteria whose fatty acid residues are in structural accord with the residues of such non-polar amino acids as valine, leucine, isoleucine, and alanine. The same amino acids are biosynthetic predecessors of iso- and anteiso-fatty acids with 14 -17 carbon atoms [27], and their addition to the growth medium typically results in the increase of the contents of respective fatty acids [26]. According to the proposed model, an increase of the share of those amino acids in the amino acid spectrum of the average statistical protein should cause same changes in fatty acid spectra of lipids.
The concept of polylipids taken out of the context of other aspects of regulation of lipid spectra of bacterial cells, cannot answer the numerous questions that modern bacteriology has for lipidology. For instance, it is hard to explain why there are distinctive groups of bacteria that have fundamentally different fatty acid metabolism. Examples of such groups are: (1) all Gram-negative bacteria except for very few genera and species, e.g. P. maltophilia and P. putrefaciens of Pseudomonas genus; Thermus genus which belongs to the following group; and some others; (2) Gram-positive bacteria with fatty acids of iso- and anteiso-structures, as, for example, such phylogenetically different groups of bacteria [25, 29] as Bacillus, Propionibacterium, and Arthrobacter; (3) mycobacteria, nocardia, rhodococci, and animal-pathogenic corynebacteria; (4) lactobacilli and alike. These groups differ also in cytostructural organization, metabolism, physiology, and ecology. A notable fact is that eubacteria of different groups differ in lipid metabolism more drastically than, for instance, all eukaryotes do. The cause of those differences may partially be explained by a controversy in the functioning of the prokaryote membrane system [3].
Leonid Andreev
Part II: https://systemity.livejournal.com/4661607.html
Original paper "View on Lipids of Microorganisms from the Standpoint of Prebiotic and Biological Evolution" published in: Voprosy Evolutsii Bakterij (Evolution of Bacteria), USSR Academy of Sciences, Center for Biological Research, Institute of Biochemistry and Physiology of Microorganisms, Pushchino, 1984, pp. 93-119 (https://www.dropbox.com/s/tx666ilga018yvt/OriginOfLifeRus.pdf).
This translation was first published in 2004 on the company website of Equicom, Inc. which is no longer online.
Contents:
• Specificity of lipids as an object of molecular biological studies
• Cellular level of lipid studies
• The concept of polylipids
• Interaction between polylipids and proteins
• Role of fatty acid residues
• Enzymes and coenzymes
• Principle of quasi-equilibrium of biosynthesis of bacterial lipids
• On the origin of life on Earth
• Macromolecular chronometry
• Molecular model of a protobiont
• Conclusions
Specificity of lipids as an object of molecular biological studies
Modern molecular biology incorporates a number of research areas dealing with substances and processes of general biological significance. Those areas of research, stimulated and inspired by the efficacy of the currently available physico-chemical methods, considerably differ in methodologies and - what is especially important - due to their specific developmental backgrounds, they have different levels of relationship with cellular biology.
When that relationship is lost or is not yet strong enough, it is often understood as an indication that some of the problems of physical chemistry of bioorganic molecules may be as broad and complex as the issues encountered with in the study of the functioning of live cells. This concerns particularly investigations of biological macromolecules that are functionally active outside the cells that synthesize them. That kind of misconception is not accidental. The awareness of the fact that, despite a theoretically possible large variety of structural and functional organizations of biopolymers, the Nature has only a limited number of their variants, impedes the advance of researchers in physics and chemistry of the functionality of such molecules, as it makes them divert to working on problems which require qualitatively different approaches and expertise and force them to study such properties of organisms which require deep empirical knowledge. This is a psychological reason that explains why many molecular biologists at least sympathize with, if not fully concede to, the thesis that the notion of ‘cell’ “has become a brake on the progress toward the understanding of live structures at the molecular level” and that “one may stop treating the cell as a biological unit but consider it as merely one of the stages of a complex chain of transformations” [16]*.
In practice, such views may seem to be justified as the attempts not to limit oneself to the “barest necessities” of evolutionary biology and “biological purposefulness” and, instead, consider the cell to be “merely one of the stages of a complex chain of transformations” may often be helpful in extensive fundamental investigations in molecular biology, leading to innovative approaches to various aspects of the functioning, systematics and evolution of live organisms. A classical example of such fundamental works in molecular biology that already by now have significantly contributed into the progress in evolutionary biology, is the method of macromolecular chronometry developed a quarter of a century ago [34] based on physico-chemical, rather than biological, logic.
_________________________________________________________
* Rendition is based on the Russian edition of the source book.
Notwithstanding the success in certain areas of molecular-biological studies concerned with the phenomenon of life, their independence from the general biological knowledge and cellular biology is only an illusion caused by the breathtaking new capabilities that look almost like a sudden materialization of something that only yesterday seemed to be not more than an idealized goal, set merely for justification of a fundamental research project. This is especially clearly exhibited in one of the areas of molecular biology - the study of lipids. The structure of lipids is less complex as compared to proteins, nucleic acids, or polysaccharides, which is why the study of lipids of microorganisms has fairly quickly advanced in accumulating molecular-biological information, which, in its turn, helped to see the narrowness and, oftentimes, senselessness of investigations that are disconnected from the problems of cellular biology. Some ten years ago, O’Leary wrote to the effect that not long ago the studies on microbial lipids were basically boiling down to investigating almost exclusively the chemical nature of those compounds and their biosynthesis.
Undertaken by many researchers within the past 20 years, those investigations unexpectedly appeared to be difficult, and the results they yielded were in many ways surprising. This field has already gathered sufficiently large amounts of information, and it continues to grow. Such a wealth of information and data on biosynthesis, although valuable in itself, is not much of an asset, as it says a lot but does not explain much. We have found ourselves in the role of observers solemnly gazing at the fascinating amount of data which shows how much we know about microbial lipids and how little we understand about them. It gets clearer to us what the lipids are and how they are organized, but - why are they needed? What does it all mean for the cell functioning? [12]**
Since the time these questions had been put, both the capabilities of theoretical and experimental lipidology and the information on lipids have dramatically expanded. But still again - in the preface to Membrane Lipids of Prokaryotes published in 1982 [20] - K. Blokh, a well-known specialist in the field, puts the same questions. He states that overabundance, rather than thrift, characterizes the composition of membrane lipids occurring in the Nature. In terms of the structural common denominator of all the components of the membrane lipid bilayer, it would suffice to have just one phospholipid that has an amphipathic structure and is capable of participating in formation of closed vesicles.
In the meantime, there is no cellular membrane in the Nature which would have only one phospholipid. “From the standpoint of compositional complexity of natural membrane lipids”, further writes K. Blokh, “it is remarkable that chemically homogenous liposomes imitate many of the properties of natural membranes, including transport, phase transitions, or the effect on membrane-bound enzymes. Clearly, there must be a great variety of membrane-dependent phenomena, expressed only in cells, which cannot be discovered by studying a single-component membrane model” [20]. Despite a relative simplicity of the lipid structure, in the Nature there are lots of structurally individual forms of lipids. For instance, bacteria-synthesized lipids consist of several classes, of which phospholipids are most widely occurring. There are over two dozens of various phospholipids that differ by a radical attached to the phosphate group. Rather than being an individual substance, each of them is a set of substances, which differs from others by the kind of fatty acid (some bacteria have up to several dozens of various fatty acids) with which the C1 and C2 hydroxyls of glycerol are esterified. Esterification of hydroxyl groups of glycerophosphate can be incomplete, in which case the so-called phospholipid lysoforms are produced. Hydroxyl groups of glycerophosphate can be bound to saturated and unsaturated alcohols, aldehydes and oxyacids.
There are two factors that contribute into diversity of the lipid composition of bacteria. Firstly, the number of structurally individual lipids and the ratio between them may vary - apparently, indefinitely - in bacterial species; and, unlike lab-synthesized lipids, bacterial lipids are not randomized. In paraphrase, each group of lipids has an unpredictably large variety of distributions of molecular species. Secondly, the group and fatty acid compositions of phospholipids of most of bacteria vary extremely widely and according to very complex patterns dependent on the physiological condition of a cell population, which, in its turn, depends on a variety of environmental factors. Oftentimes, such variations in phospholipids composition can be considered as qualitative [5].
For the said reasons, an exhaustive description of lipids of even one species of bacteria is quite complicated and uneconomic in terms of the time and effort involved, given that all those details and nuances have no practical application within the context of the currently existing theories on the functional role of lipids in bacterial cells. This situation is clearly a result of the overestimation of the capability of exact sciences to solve the problems of cellular biology at the molecular level. Contemporary theories on the functional role of lipids in bacterial membranes are mostly of the physico-chemical, rather than biological, nature, and are based on interpreting such integral properties of lipids as the high degree of reduction (lipids as energy accumulators), spatial dissociation of hydrophilic and hydrophobic groups, capability for phase- isolation in aqueous
environment, dependence of the degree of spatial freedom of fatty acid radicals on their structure, etc. The physico-chemical approach to the study of natural lipids has helped to solve a number of important problems, including, first of all, production and utilization of artificial biological membranes. As well, it significantly contributed into the progress of technology of lipid studies, and many of the results obtained by physico-chemical methods are successfully used in biology and medicine.
Nevertheless, it would be fair to say that, in the long run, the study of biological role of lipids has suffered serious damage as a result of preoccupation with physico-chemical methods. This has become especially clear when physico-chemical lipidology, having failed to offer a more or less credible explanation to the astonishing diversity of the lipid spectra in living organisms, has changed its position toward downplaying the significance of this crucially important fact. There came a new wave of studies whose authors were determining the melting temperature of lipids isolated from various microorganisms, assuming that the cell lipid metabolism, as intricate as it is, is responsible mainly for maintaining a required level of membrane liquidity [21, 30]. For quite a long time, this inappropriately naïve idea kept attracting many researchers as
___________________________________________________________________
** Back translation from the Russian edition of the source book.
an affordable way to obtain integral quantitative characteristics of a cell lipid pool, although absolutely helpless in terms of explaining the uniqueness of lipid spectra in bacteria which avail themselves of thousands of ways to create a same physical effect. This phenomenon is an indication that phase transitions in bacterial membranes are by no means the main cause of purposeful changes in lipid composition - instead, they are one of the effects of those changes. The idea about the defining role of the physical condition of membranes is closely connected with the works on physiology and biochemistry of bacteria artificially rendered unable to synthesize certain lipids [28]. It was shown, for instance, that bacteria can grow and develop even when their membranes are artificially supplied with mixtures of lipids that they cannot not naturally synthesize.
It runs through all of those studies that specificity of lipid composition of bacteria is not life’s necessity. However, as that idea was further developing, it eventually has led to an absolutely opposite result - a realization that there must be lots of membrane-dependent phenomena that are exhibited only in cells and cannot be detected by experimenting with physico-chemical models, even those that imitate a live cell.
Proteins and nucleic acids can be studied in respect of their functionality, without going deep into fine details of the biology oforganisms they are isolated from. With lipids, such an approach seems to be practically senseless, and that is why it is not surprising that, despite the tremendous progress in deciphering the fine mechanisms of functioning of proteins and nucleic acids, the function of lipids remains as unclear as it was decades ago. Functionally, lipids are inseparable from organisms that synthesize them. This is the main peculiarity of lipids as objects of molecular-biological studies, which requires the whole methodology of lipid studies to be revised so as to restore its connection with the cellular level.
Cellular level of lipid studies
The cellular approach to the study of lipids means the establishing of functional relations between the lipid composition and its regulation, on the one hand, and biological (morphological, physiological, ecological, and other) characteristics of organisms that synthesize them, on the other. Taking a full advantage of that approach involves two serious but surmountable challenges. The first one is the need of solid knowledge in biology of microorganisms under study, which can be resolved through research cooperation. The second challenge comes from is a virtually complete lack of an adequate theoretical basis. As a minimum, we need hypotheses explaining the fantastic diversity of lipid spectra of bacteria, let alone the capability to predict, based on bacteria known properties, lipid compositions of specific bacteria. Properties of lipids should be studied in close connection with biological peculiarities of cells of bacteria that synthesize them, as this will help to bridge the gap between theory and experiment. Functional integrity of the lipid pool of living cells is especially valuable in the context of this problem.
Evolution of bacteria was accompanied with the restructuring of their membrane system with its inherent integrity of the structure and functions. As well, integrity is haracteristic of lipids of live cell membranes, which can be proved on various examples, some of which are discussed in [2, 4-7, 13]. Primarily, it is expressed in the character of adaptational changes in membrane lipids in the course of bacterium growth and development, in response to changes in the environment parameters. In such cases, the lipid pool of live cells responds as an integral system, acting by fairly complex rules that are certainly impossible to figure out based on lipid physico-chemical properties solely.
Integrity of the lipid pool is also expressed in the fact that evolutionary modifications of certain biological properties of bacteria were accompanied with coordinated changes in lipid compositions, which were similar in different bacteria, irrespective of their systematic position and ecology. Functional integrity of the lipid pool of live cells and functional inactivity of lipids isolated from the cell, as well as the diversity of lipid spectra, should be taken into account when designing a model claiming to explain the biological role of lipids.
The concept of polylipids
According to this concept, phospholipids of cell membranes form cellular polymeric structures [1, 5]. In the ideal case, a polylipid of the bacterial cytoplasmic membrane can be considered a molecule of the same size as a bacterial cell. Phospholipids form the skeleton of the native membrane - a so-called polylipid membrane. Monomer phospholipids of the polylipid membrane are bonded mainly by principal valence forces. Fig. 1 shows a schematic view of a polylipid membrane built from molecules of phosphatidyl ethanolamine (PE) and cardiolipin (CL). The basic monomer unit in such a membrane is a dimer in which PE molecules are connected through a metal atom in such a way that fatty acid residues of both PE molecules are oppositely oriented. Another type of bond between monomer phospholipids (PE dimers) is formed by substitution of protons of enolized ester carbonyls by ions of polyvalent metals (mostly, Mg). Such bonds, particularly with the involvement of Mg, have low conformational freedom and can be broken by the attachment of protons. Two layers of polylipids, organized in the way shown in Fig. 1, form the membrane’s polylipid bilayer whose stability is determined by potassium and sodium atoms in the intra-bilayer space. These atoms interact with ester carbonyls of phospholipids of both monolayers. This interaction involves also amino groups of phosphatidyl ethanolamine, whose degree of involvement determines the ratio between Mg and Na in the inter-bilayer space.
Proteins and nucleic acids can be studied in respect of their functionality, without going deep into fine details of the biology of organisms they are isolated from. With lipids, such an approach seems to be practically senseless, and that is why it is not surprising that, despite the tremendous progress in deciphering the fine mechanisms of functioning of proteins and nucleic acids, the function of lipids remains as unclear as it was decades ago. Functionally, lipids are inseparable from organisms that synthesize them. This is the main peculiarity of lipids as objects of molecular-biological studies, which requires the whole methodology of lipid studies to be revised so as to restore its connection with the cellular level.
Cellular level of lipid studies
The cellular approach to the study of lipids means the establishing of functional relations between the lipid composition and its regulation, on the one hand, and biological (morphological, physiological, ecological, and other) characteristics of organisms that synthesize them, on the other. Taking a full advantage of that approach involves two serious but surmountable challenges. The first one is the need of solid knowledge in biology of microorganisms under study, which can be resolved through research cooperation. The second challenge comes from is a virtually complete lack of an adequate theoretical basis. As a minimum, we need hypotheses explaining the fantastic diversity of lipid spectra of bacteria, let alone the capability to predict, based on bacteria known properties, lipid compositions of specific bacteria.
Properties of lipids should be studied in close connection with biological peculiarities of cells of bacteria that synthesize them, as this will help to bridge the gap between theory and experiment. Functional integrity of the lipid pool of living cells is especially valuable in the context of this problem. Evolution of bacteria was accompanied with the restructuring of their membrane system with its inherent integrity of the structure and functions. As well, integrity is characteristic of lipids of live cell membranes, which can be proved on various examples, some of which are discussed in [2, 4-7, 13]. Primarily, it is expressed in the character of adaptational changes in membrane lipids in the course
of bacterium growth and development, in response to changes in the environment parameters. In such cases, the lipid pool of live cells responds as an integral system, acting by fairly complex rules that are certainly impossible to figure out based on lipid physico-chemical properties solely.
Integrity of the lipid pool is also expressed in the fact that evolutionary modifications of certain biological properties of bacteria were accompanied with coordinated changes in lipid compositions, which were similar in different bacteria, irrespective of their systematic position and ecology. Functional integrity of the lipid pool of live cells and functional inactivity of lipids isolated from the cell, as well as the diversity of lipid spectra, should be taken into account when designing a model claiming to explain the biological role of lipids.
The concept of polylipids
According to this concept, phospholipids of cell membranes form cellular polymeric structures [1, 5]. In the ideal case, a polylipid of the bacterial cytoplasmic membrane can be considered a molecule of the same size as a bacterial cell. Phospholipids form the skeleton of the native membrane - a so-called polylipid membrane. Monomer phospholipids of the polylipid membrane are bonded mainly by principal valence forces. Fig. 1 shows a schematic view of a polylipid membrane built from molecules of phosphatidyl thanolamine (PE) and cardiolipin (CL). The basic monomer unit in such a membrane is a dimer in which PE molecules are connected through a metal atom in such a way that fatty acid residues of both PE molecules are oppositely oriented. Another type of bond between monomer phospholipids (PE dimers) is formed by substitution of protons of enolized ester carbonyls by ions of polyvalent metals (mostly, Mg). Such bonds, particularly with the involvement of Mg, have low conformational freedom and can be broken by the attachment of protons.
Two layers of polylipids, organized in the way shown in Fig. 1, form the membrane’s polylipid bilayer whose stability is determined by potassium and sodium atoms in the intra-bilayer space. These atoms interact with ester carbonyls of phospholipids of both monolayers. This interaction involves also amino groups of phosphatidyl ethanolamine, whose degree of involvement determinesthe ratio between Mg and Na in the inter-bilayer space.
A bilayer whose polylipid structure suffers a damage can transform itself into a lamellar-type bilayer in which phospholipids are positioned back-to-back with their nonpolar radicals, whereas their polar heads are directed outward, according to the Gorter-Grendel model proposed in 1925 and further developed by Dawson and Danielli. A lamellar-type bilayer is not a polylipid since it is organized according to physical, but not chemical, interaction of monomer units. The biological membrane of eubacteria is structured by the polylipid layer both lengthwise and crosswise. Qualitative composition and proportions of monomer phospholipids determine to the large extent mechanical and physico-chemical properties of the polylipid skeleton of the membrane and underlie the diversity of their molecular organization. Both in Gram-positive and Gram-negative bacteria, diphosphatidyl glycerol (or cardiolipin (CL)) is the cross-agent that splices individual polylipid chains. As it follows from the general principles of the model, polylipid monolayers can change their positions relative to each other, using alkali metal ions as a lubricant. However, the transfer of monomer phospholipids from one polylipid monolayer to another occurs only if the polylipid skeleton is destroyed and a lamellar-type bilayer is constructed.
Interaction between polylipids and proteins
The concept of polylipids, even in the above-presented form, allows the understanding of the cause of the qualitative and quantitative diversity of lipid spectra of such organisms as bacteria whose cytoplasmic membrane performs practically all of the membrane-dependent functions; it also explains the diversity of the forms of bonds between proteins and membranes. Based on the model of molecular organization of the biological membrane, we will discuss a possible mechanism for protein embedding into the membrane’s polylipid skeleton. When contacting the surface of the bilayer polylipid membrane, protein interacts with it through weak van der Waals forces emerging when the protein non-polar amino acid radicals come into contact with short fragments of fatty acid radicals “sticking-out” on the surface of the polylipid layer (cf. Fig. 1). When the cumulative amount of that interaction reaches a certain critical value, the pressure, directed onto the surface of the polylipid skeleton, destroys it in the site of protein embedding. This effect is determined by the limitedness of conformational freedom of the metal (magnesium)-lipid complexes, due to which even a small deformation of the
lipid skeleton causes a dramatic increase of reactive capacity of M-O (metal - oxide) bonds whose disruption results not only in the embedding of proteins but also in their covalent bonding with lipids.
Thus, a biological membrane polylipid layer serves as both an accumulator and amplifier of weak interactions that are determined by the spatial structure of the embedding proteins, on the one hand, and the lipid composition at the embedding site, on the other. Anisotropy of protein embedding into the membrane is determined by the asymmetry of the polylipid bilayer. During the embedding process, the weak quantum-mechanical interactions between lipids and proteins are transformed into interactions with the breaking and formation of primary valence bonds. The strength of the both types of primary valence bonds between monomer phospholipids is determined by numerous factors, including the concentration of cations - primarily, protons. However, it should be taken into consideration that the polylipid membrane is a dynamic structure wherein primary valence bonds are being broken and restored. The structure of the polylipid skeleton is determined by not only the quality and quantity of individual lipids, but also - and maybe even to a higher extent - by the qualitative and quantitative composition of proteins and other components of biomembranes.
At the initial stage of protein embedding into the membrane polylipid skeleton, non-polar interactions between amino acid and fatty acid residues induce the restructuring of the polylipid skeleton to achieve a thermodynamically advantageous state of the maximal contact. Prior to the contact with a macromolecule to be embedded, the embedding site represents, due to high lateral flexibility of monomer phospholipids, a polymer with randomized distribution of monomers. However, after initiation of the contact, the distribution becomes more or less specific. Regardless of the extent of specificity - hence, the energy of non-polar interactions for protein embedding with the breaking of primary valence bonds - the protein contact with the polylipid membrane results in the decrease of the contents of individual phospholipids that are “complementary” to a given protein and the increase of those that are “non-complementary”. As a result, the changes in the individual lipid contents in the membrane appear to be closely connected with the changes in the protein composition of the membrane.
Role of fatty acid residues
The proposed model of polylipids brings us closer to solving the problem of the purposefulness of bacterial biosynthesis of fatty acids with different structures of fatty acid residues. Calculations indicate that at the dimer configuration of phospholipids in the membrane, only the terminal sections of fatty acid chains containing more than 12 carbon atoms are actually accessible for direct contact with amino acid residues of proteins. In most Gram-negative bacteria, fatty acids contain 14 to 18 carbon atoms. With the lamellar structure of the phospholipid layer, this difference range is negligible, and, therefore, it is unclear why a 5-15% change of the average length of fatty acid residues should correlate with substantial changes in morphology, physiology and biochemistry of bacteria. Based on the proposed model, however, the increase from 14 to 18 carbon atoms means a many-fold expansion of the area for lipid-protein contacts. The polylipid model demonstrates a great extent of rationality in the organization of lipid-protein interactions in Gram-positive bacteria whose fatty acid residues are in structural accord with the residues of such non-polar amino acids as valine, leucine, isoleucine, and alanine. The same amino acids are biosynthetic predecessors of iso- and anteiso-fatty acids with 14 -17 carbon atoms [27], and their addition to the growth medium typically results in the increase of the contents of respective fatty acids [26]. According to the proposed model, an increase of the share of those amino acids in the amino acid spectrum of the average statistical protein should cause same changes in fatty acid spectra of lipids.
The concept of polylipids taken out of the context of other aspects of regulation of lipid spectra of bacterial cells, cannot answer the numerous questions that modern bacteriology has for lipidology. For instance, it is hard to explain why there are distinctive groups of bacteria that have fundamentally different fatty acid metabolism. Examples of such groups are: (1) all Gram-negative bacteria except for very few genera and species, e.g. P. maltophilia and P. putrefaciens of Pseudomonas genus; Thermus genus which belongs to the following group; and some others; (2) Gram-positive bacteria with fatty acids of iso- and anteiso-structures, as, for example, such phylogenetically different groups of bacteria [25, 29] as Bacillus, Propionibacterium, and Arthrobacter; (3) mycobacteria, nocardia, rhodococci, and animal-pathogenic corynebacteria; (4) lactobacilli and alike. These groups differ also in cytostructural organization, metabolism, physiology, and ecology. A notable fact is that eubacteria of different groups differ in lipid metabolism more drastically than, for instance, all eukaryotes do. The cause of those differences may partially be explained by a controversy in the functioning of the prokaryote membrane system [3].