Welcome to DU!
The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards.
Join the community:
Create a free account
Support DU (and get rid of ads!):
Become a Star Member
Latest Breaking News
General Discussion
The DU Lounge
All Forums
Issue Forums
Culture Forums
Alliance Forums
Region Forums
Support Forums
Help & Search
Synthetic connectivity, emergence, and self-regeneration in the network of prebiotic chemistry
The paper I'll discuss in this post is this one: Synthetic connectivity, emergence, and self-regeneration in the network of prebiotic chemistry (Wołos et al., Science 369, 1584 (2020) eaaw1955.
I was inspired to spend some time with this (full) paper by another post in this group, specifically this one: A New Chemical 'Tree of The Origins of Life' Reveals Our Possible Molecular Evolution (ScienceAlert) (Thank you eppur se muova!)
When I was a kid, planning an organic synthesis was largely a "seat of the pants" undertaking; one would look at a structure, and try to figure a series of disconnections that were available to bond formation using synthetic reactions of various types - the more you knew, the better you were able to plan these things - and one would talk to friends, advisors, bosses and also spend a long time in the library in the complex pathways to various leather bound "chemical abstracts" and their various kinds of indexes, find the relevant abstract, then search for the bound issue of the journal, only, often, to find that the particular paper was not the one you actually needed to accomplish the task. You'd work your way back until you had a plan using readily available starting materials.
Sometimes, your plan would actually work so far along the line, and you'd get ambitious and scale up, and then a flask would break, and boom...the pain...the pain...the pain...
You really had to think a lot in those days, and also consider the risks of wasting valuable materials.
You really had to think, and certain things had to be second nature.
Here's a problem, I pulled out of an old "exercise book," Ranganathan and Ranganathan, Academic Press, 1980, for problem 124. pg 42, on my bookshelf: The problem asked to propose a mechanism for the reaction that converted compound #4 into compound #10:
The paper, Chemistry of Santonic Acid. Oxidative and Reductive Modifications (Hortmann and Daniel, J. Org. Chem. 1972, 37, 26, 4446–4460) of course, gives an answer to the problem, a mechanism that seems quite reasonable, but the point of the exercise is to work it out yourself before being prodded by the paper.
Imagine someone asked you to make compound 10 with almost no information, other than what you learned in school? It's not too obvious...
This nice presentation, from K.C. Nicolaou shows classic "disconnection" thinking: [link:https://nicolaou.rice.edu/ppt_lectures/12_GERMANY_GENERAL.pdf|Total Synthesis of Natural Products of Biological
and Medicinal Importance]
It's been 30 years, maybe, since I planned a full synthesis; my career took a different path, but those were the "good old days." (Obviously not Lindsay Graham "Good Old Days" Up yours Lindsay.)
But when I was a kid, we used to go, like kids going to a big rock concert, to lectures by E.J. Corey, or D.H.R Barton, or Barry Trost, or K.C. Nicolaou or Roald Hoffman...
...sigh... Life is fun and then you die.
There are now, I'm sure a large number of computational programs to help design syntheses, and in fact, the paper referenced above, is about prebiotic chemistry, where the starting materials are simply the common "prebiotic" molecules spread across space. These programs probably use connectivity matrices of some kind or another - I don't know - my knowledge of programming is primitive at best, and to be honest, it's been decades since I thought about connectivity matrices.
I'm not sure if the link give more than the simple abstract or what is called the "structured abstract," to non-subscribers but the "structured abstract" has this conclusion:
CONCLUSION
Computer-generated reaction networks are useful in identifying synthetic routes to prebiotically relevant targets and are indispensable for the discovery of prebiotic chemical systems that are otherwise challenging to discern. As our network continues to grow by means of crowd-sourcing of newly validated prebiotic reactions, it will allow continued simulation of chemical genesis, beginning with molecules as simple as water, ammonia, and methane and leading to increasingly complex targets, including those of current interest in the chemical and pharmaceutical industries.
Computer-generated reaction networks are useful in identifying synthetic routes to prebiotically relevant targets and are indispensable for the discovery of prebiotic chemical systems that are otherwise challenging to discern. As our network continues to grow by means of crowd-sourcing of newly validated prebiotic reactions, it will allow continued simulation of chemical genesis, beginning with molecules as simple as water, ammonia, and methane and leading to increasingly complex targets, including those of current interest in the chemical and pharmaceutical industries.
So there you have it...
The "structured abstract" includes this graphic:
The caption:
Network of prebiotic chemistry.
Computer simulation of plausible prebiotic reactions creates a network of molecules that are synthesizable from prebiotic feedstocks and establishes multiple unreported—but now experimentally validated—syntheses of prebiotic targets as well as self-regenerating cycles. In this schematic illustration, light blue nodes represent abiotic molecules, dark blue nodes represent molecules along newly discovered prebiotic syntheses of uric and citric acids, and red nodes represent other biotic molecules.
Computer simulation of plausible prebiotic reactions creates a network of molecules that are synthesizable from prebiotic feedstocks and establishes multiple unreported—but now experimentally validated—syntheses of prebiotic targets as well as self-regenerating cycles. In this schematic illustration, light blue nodes represent abiotic molecules, dark blue nodes represent molecules along newly discovered prebiotic syntheses of uric and citric acids, and red nodes represent other biotic molecules.
From the introduction:
Research on the chemical origins of life (OL) is coming of age. The pioneering efforts of Miller (1), Oparin (2), Oró (3, 4), and Orgel (5) by the 1960s; Eschenmoser (6) in the 1990s; and Sutherland (7, 8), Carell (9), Moran (10), and others (11–15) in recent years have systematized the knowledge about reactions that can be performed under consensus prebiotic conditions, as well as the plausible synthetic routes leading to life’s key molecules. On the other hand, we still have only a fragmentary understanding of whether and how other types of molecules formed on primitive Earth and how this entire prebiotic molecular space evolved into systems of chemical reactions (12, 16) and compartments (17, 18) housing them. Such analyses require consideration of very large numbers of putative reaction pathways but are finally becoming possible, owing to recent advances in the study of chemical reaction networks and computer-assisted organic synthesis (19, 20). In this study, we use such large-scale in silico network analyses to map the space of molecules synthesizable from basic prebiotic feedstocks, quantifying the structure of this space as well as the abundances and thermodynamic properties of its members. We then demonstrate three notable forms of chemical emergence: (i) that the molecules created within the network can themselves enable new types of prebiotic reactions, including multicomponent transformations that lead to complex and useful organic scaffolds; (ii) that within just a few synthetic generations, simple chemical systems (such as self-regenerating cycles) begin to emerge; and (iii) that the network contains prebiotic routes to surfactant species (both peptide-based and long-chain carboxylic acids), thus outlining a path to biological compartmentalization. We support these results by experimental validation of previously unappreciated prebiotic syntheses (e.g., of acetaldehyde, diglycine, as well as malic, fumaric, citric, and uric acids) and entire reaction systems—notably, we demonstrate a self-regenerating cycle of iminodiacetic acid (IDA) that complements prebiotic autocatalysis on the basis of the formose cycle (21). The web application underlying our calculations is made freely available to the community (https://life.allchemy.net) in the hope that synthetic network analyses will become a useful addition to the toolkit of OL research by supporting accelerated discovery of prebiotic routes, including environmentally friendly syntheses of useful targets from basic feedstocks.
Allchemy’s “Life” module uses 614 reaction rules (“transforms”) involving C, O, N, S, and P elements, grouped within 72 broader reaction classes. Inclusion of these rules in our set is contingent on the existence of literature-described examples that document their execution under generally accepted prebiotic conditions...
Allchemy’s “Life” module uses 614 reaction rules (“transforms”) involving C, O, N, S, and P elements, grouped within 72 broader reaction classes. Inclusion of these rules in our set is contingent on the existence of literature-described examples that document their execution under generally accepted prebiotic conditions...
The authors, however, dodge the elephant in the room, chirality:
Our transforms account for reaction by-products and specify the scope of admissible substituents, structural motifs incompatible with a given reaction (some 400 potentially conflicting groups are considered for each reaction), typical conditions accepted in prebiotic chemistry, solvents, temperatures, and more. They do not consider stereochemistry [because homochirality probably appeared as a result of chemical evolution of racemic mixtures (25)] or reaction kinetics [because kinetic data are only sparingly reported in the studies of prebiotic chemistries (26)]. On the other hand, yields for each type of reaction are approximated on the basis of statistics collected from relevant publications and are categorized as trace (?3%), low (>3% to ?10%), moderate (>10% to and high (?80%) (SM section S2).
The authors begin their task by running the alchemy program for two hours on a "standard laptop computer," using a network of reactions that involved only carbon, hydrogen, oxygen, and sulfur and which limited the molecules "discovered" in silico to those with a molecular weight of less than 300 Daltons. This generated tens of thousands of molecules, 82 of which were "biotic," which they define as consisting of the following classes of molecules: "...amino acids and peptides, nucleobases and nucleosides, carbohydrates, and metabolites..."
Again, to beat a dead horse, lacking chirality, the molecules are only weakly "biotic" in nature as I see it, and remain so without the spontaneous generation of chirality through either kinetic or other means.
In this paper, the captions are rather long and pretty much tell much of the story:
Fig. 1 Biotic and abiotic molecules in the networkof prebiotic chemistry:
The caption:
(A) Scheme illustrating the synthetic algorithm in which SMARTS-coded (22) reaction transforms act on the current pool of reactants to produce the next generation of compounds. Afterward, these products are combined with original reactants and the procedure repeats until a user-specified number of generations is reached. (B) Six generations of a synthetic network originating from six primordial substrates—H2O, N2, HCN, NH3, CH4, and H2S—and leading to possible biotic products [amino acids and peptides, nucleobases and nucleosides, carbohydrates, and metabolites found in living organisms (28); red circles] and abiotic products [other small molecules; blue circles] with molecular mass not exceeding 300 g/mol. Circle size corresponds to the molecule’s incoming connectivity, kin (i.e., the number of reactions that produce this molecule as product). (C) Forty-one biotic molecules within the network’s seven generations [six generations are shown in (B); for the full network with all seven generations, see https://tol.allchemy.net]. Of the biotic class, glycine is in the second generation (G2); urea, adenine, butenedioic acid, and oxalic acid are in G3; glyceraldehyde, isoguanine, aspartic acid, hypoxanthine, cytosine, phenylalanine, succinic acid, malic acid, glyoxylic acid, and aldotetrose are in G4; xanthine, alanine, serine, guanine, uracil, lactic acid, oxaloacetic acid, and aldohexose are in G5; malonic acid, pentofuranose, glycerol, pyruvic acid, cytidine, and ketoheptose are in G6; and ketohexose furanose, threonine, methionine, proline, glutamic acid, citric acid, acetic acid, thymine, adenosine, guanosine, uridine, and uric acid are in G7. Various di-, tri-, and tetrapeptides are also present within G4 to G7 (fig. S55). In addition, histidine is in G8, arginine in G11, valine in G12, and leucine in G16. The molecules shown are colored according to the lowest-yielding step within the shortest pathway: Red, at least one step is predicted to generate only traces of product (?3%); orange, at least one step is low yielding (?10%); and green, all steps are predicted to proceed in moderate or high yields. (D) Allchemy’s screenshot of the G4 tree, with the two shortest prebiotic synthesis pathways of succinic acid and of glycine colored according to the yields of individual steps [color coding as in (C)]. (E) Schemes of the pathways. Numbers below reaction arrows correspond to the transform labels in SM section S2; this section also contains details of prebiotically plausible reaction conditions (e.g., CuCN, KCN, H2O, and irradiation for conversion of hydrogen cyanide into formaldehyde), along with pertinent literature references. Raw Allchemy screenshots of the pathways are provided in SM section S4. If two numbers are given below a single arrow, it means that the software recognizes the product of the first reaction as highly reactive and prone to the second reaction in a tandem sequence [e.g., hydrolysis of a nitrile to a carboxylic acid (#38) creates 2-aminomalonic acid, which readily undergoes elimination of carbon dioxide (#2) under hydrolysis conditions; formation of imines (#10) creates methyleneamine, which then undergoes addition of cyanide (#18)].
Another graphic:
Fig. 2 The network’s molecular content and synthetic connectivity.
The caption:
(A) Distribution of biotic (blue markers) and abiotic (gray markers) molecules in a plane defined by molecular mass and heat of formation calculated using the PM6-D3H4X (66) semiempirical method implemented in the MOPAC2016 software (67). To simplify presentation, abiotic compounds were clustered into 1202 groups according to their structural similarity (quantified by Tanimoto coefficients between molecules’ ECFP4 fingerprints). Each cluster is represented by a circle of diameter proportional to the number of members, and position is determined by the group’s centroid (i.e., a group’s “representative” molecule, defined as the molecule with maximum average Tanimoto similarity to other members of the cluster). A similar correlation is observed when larger and unclustered samples of abiotic compounds are considered (see fig. S11 for distributions of >11,000 compounds; also see table S1 for additional thermodynamic considerations). (B) Distribution of biotic and abiotic compounds in a plane defined by the logP values calculated from Wildman and Crippen’s method (32) and the number of hydrogen bond acceptors. Biotic compounds are, on average, less hydrophobic than abiotic compounds for a given number of hydrogen bond acceptors. Further details of the underlying feature selection are described in the SM section S6.2. (C and D) Graphical illustration of condition changes along synthetic pathways leading to 30 randomly chosen (C) biotic versus (D) abiotic compounds in the G7 network. The horizontal axes quantify the numbers of steps in each pathway: For an n-step synthesis, the first step will correspond to location 1/n, the second step to 2/n, and the final step to n/n = 1 (i.e., all pathways stop at the scale’s value of 1). Conditions on the vertical axis: A, acidic; MA, moderately acidic; N, neutral; MB, moderately basic; and B, basic. (E and F) Full condition variability statistics are summarized in histograms for the syntheses of (E) 82 biotic molecules and (F) 36,603 abiotic molecules. The difference in the two distributions is statistically significant with P value < 0.001, as evaluated by ?2, Kolmogorov-Smirnov, and bootstrap tests (SM section S6.1). (G) Distribution of node degrees (k) for G5, G6, and G7 networks. Connectivity of a given node is the sum of the numbers of its incoming and outgoing connections (k = kin + kout; for distributions of kin and kout, see fig. S58). Linearity of these dependencies shown on the doubly logarithmic scale indicates a power law —P(k) ? k^(?? ), where ? ~ 1.8—and a scale-free network architecture. (H) Cumulative distribution of ? (k)=?kki=0 versus k provides evidence for preferential attachment. In this expression, denotes the average increase of the degrees of nodes with k = i between the fifth and sixth generations (green curve) and between the fifth and seventh generations (blue). The plot traces such evolutions of all nodes present in the network’s fifth generation (compounds in G5 with only a single incoming connection are not considered). The linearity of the dependence on the doubly logarithmic scale indicates another power law, ? (k)=k?, and an exponent greater than unity (? ~ 1.6 to 1.8) confirms preferential attachment. Notably, the slopes of both power laws are close to the values we previously found (35) for the scale-free network of all organic chemistry, indicating that prebiotic and modern organic syntheses are both governed by the same rules of synthetic reactivity.
The authors note that the elimination of some classes of reaction classes does not necessarily preclude arriving at the full or nearly full set of biotic molecules as a generated in the full sized class reactions, whereas removing sets of reaction classes has a fairly profound effect on the the number of abiotic molecules generated.
Interestingly, the software is said to find new pathways to biotic molecules.
Because the individual reaction rules used to generate the network are derived from the OL literature, we trivially expect the network to contain the known prebiotic pathways leading to all of these biotic compounds. Indeed, all such syntheses are present in the network, as illustrated in Fig. 3, A and B, for adenine, a popular target of prebiotic studies (for syntheses of other targets, see SM section S4). Notably, in addition to cataloging known routes, the network also contains previously unreported syntheses of biotic molecules. As a case in point, consider a computer-generated subnetwork of reactions leading to succinic acid and also involving syntheses of lactic, pyruvic, malic, fumaric, and glyoxylic acids (all biotic molecules are depicted in green in Fig. 3C). Analysis of the network in comparison with known literature reveals that most routes to these biotic molecules are a patchwork of steps reported in different publications (corresponding to different colors of the arrows), some of which are not concerned with OL issues (steps marked as NOL for non-OL), but all performed under prebiotic conditions.
Fig. 3 Examples of known and newly identified syntheses within the network of prebiotic chemistry.
A) Ten prebiotic synthetic pathways leading to adenine, all previously described in the OL literature, are highlighted in the network (for clarity, only a subnetwork of C-, O-, and N-based chemistries up to G4 is shown). Identical synthetic connections common to several pathways are indicated by arcs of different curvatures. (B) Chemical schemes of adenine’s prebiotic syntheses, along with those from pertinent literature (3, 49, 61, 69–77). Colored circles over the arrows correspond to the colors of pathways shown in (A). Circle segments are used to indicate to which multiple pathways a given step belongs. The first (trimerization of HCN) and last (formation of amides, imides, amidines, or guanidines followed by cyclization) steps are common to all pathways. There are three main strategies in the syntheses of adenine; formamidine (1) and formamide (2) participate as second substrates in three key, two-component reactions. h?, light. (C) Subnetwork of reactions that lead to succinic acid and involve syntheses of lactic, pyruvic, malic, fumaric, and glyoxylic acids (biotic molecules are in green). Previously unreported connections now verified by experiment are denoted with red arrows. Previously reported connections share the same color if they come from the same source publication. NOL indicates reactions reported outside of origins research but performed under prebiotic conditions. (D to F) Syntheses of (D) citric acid, (E) diglycine, and (F) uric acid. Gray arrows and structures denote reactions that have been described previously [the fourth reaction in (D), hydrolysis of malonitrile, is described in (C)]; black structures and red arrows represent the software-predicted reactions that we verified experimentally. When several reactions were performed in one pot, some intermediates were not isolated (but were still confirmed spectroscopically); these are enclosed in square brackets. For all experimental details, see the main text and experimental procedures in SM sections S7 to S12. rt, room temperature.
Now it gets interesting, because was is discussed is an abiotic example of something that characterizes many metabolic processes, cyclic reactions, the most famous of which is the citric acid (Krebs) cycle.
Turning our attention to other classes of prebiotic chemistry, we considered the synthesis of citric acid (CA), illustrated in Fig. 3D. Recently, a prebiotic mimic of the CA cycle was reported (10), but it contained only analogs of CA and not CA itself. Our network analysis suggested that CA could emerge under prebiotic conditions in water from two equivalents of oxaloacetic acid [the synthesis of which has already been reported in the OL literature (10)] via a tandem aldol self-condensation (H2O, pH 7.5, 4°C) and decarboxylation sequence followed by a second decarboxylation. This second decarboxylation, promoted by either 0.054 or 0.081 M FeCl3, worked better at room temperature than at 70°C, the temperature used for related compounds in previous work (10). Under our milder conditions, we obtained CA in ~5% yield, whereas under harsher conditions, the citroylformic acid substrate gradually decomposed, reducing the yield to ~2% (table S8).
Reference 10 is this paper:
Synthesis and breakdown of universal metabolic precursors promoted by iron. (Muchowska, K.B., Varma, S.J. & Moran, J. Nature 569, 104–107 (2019))
And finally, there is the issue of catalysis, which, to return to a point I made earlier, may include asymmetric catalysis.
Emergence of catalysts and reaction types
We first discuss the finding that compounds created within the network can themselves act as catalysts of additional chemical reactions, all operative under prebiotic conditions, thereby substantially expanding the accessible prebiotic chemical space. To show this, we queried the network for known organocatalysts or bi- and tridentate metal chelators capable of binding metal cations present on primitive Earth [e.g., Cu(II), Zn(II), and Mn(II) (40)] and also used in modern organometallic catalysts. Figure 4A lists eight such catalysts enabling different reaction types and collectively more than doubling the size of the network. All of these reactions were previously carried out under prebiotic conditions (41–48), but their relevance to OL was unnoticed.
We first discuss the finding that compounds created within the network can themselves act as catalysts of additional chemical reactions, all operative under prebiotic conditions, thereby substantially expanding the accessible prebiotic chemical space. To show this, we queried the network for known organocatalysts or bi- and tridentate metal chelators capable of binding metal cations present on primitive Earth [e.g., Cu(II), Zn(II), and Mn(II) (40)] and also used in modern organometallic catalysts. Figure 4A lists eight such catalysts enabling different reaction types and collectively more than doubling the size of the network. All of these reactions were previously carried out under prebiotic conditions (41–48), but their relevance to OL was unnoticed.
Fig. 4 Chemical emergence in the network of prebiotic chemistry.
The caption:
A) (Left) Eight types of chemical reactions enabled by seven molecules created in the original G7 network (note that OAc and IDA repeat twice in nine entries shown). These molecules are either organocatalysts or components of catalytic complexes with prebiotically plausible metal cations [e.g., Zn(II), Cu(II), and Mn(II)]. All of the reactions shown had been previously performed under prebiotic conditions, but their relevance to origins research was not noted. (Middle) Colored arrows illustrate how many additional compounds can be created in our prebiotic network upon addition of each of the reactions shown. There is one arrow for entries 4 and 5 because two different catalysts enable the same reaction type (A3 coupling). (Right) Examples of molecules that are made synthesizable via these reactions. The gray part of the arrow at the bottom indicates the sum of these molecules (21,529), and its green extension represents the additional 34,957 molecules that are created when all of the reactions (1 through 9) are added to the generative set simultaneously. (B) The red arrow corresponds to the selective hydrolysis of 2-aminopentanedinitrile to 2-amino-4-cyanobutanamide catalyzed by formaldehyde, a reaction proposed by the software and validated experimentally. The remaining steps illustrate how the software navigated the synthesis of the aminopentanedinitrile substrate from the HCN primordial feedstock. These steps are shown in gray to indicate that they have already been executed by others and described in the OL literature. (C) The red arrow corresponds to the synthesis of 1-(1H-imidazol-4-yl)ethane-1,2-diol from ketotetrose, ammonia, and formaldehyde catalyzed by copper(II) acetate. This transformation was proposed by the software and chosen for experimental validation because it establishes an unreported prebiotic route to the histidine amino acid. All downstream and upstream steps (in gray) have been described earlier in OL literature. Previously, histidine was generated along an inefficient bypass (also shown in the scheme) from aldotetrose and formamidine (50).
IDA here is imidodiacetic acid, which is a derivative of the simplest amino acid, glycine, a "Siamese" glycine if you will. This molecule is an excellent chelator of metals, and as such, can coordinate metals which may act as catalysts.
The cyclic reaction detailed in the next figure, includes IDA in its pathway.
Fig. 5 Emergence of self-regenerating cycles within the network of prebiotic chemistry.
The caption:
(A) Self-regenerating cycle in which one molecule of IDA (orange) can produce up to two copies of itself. When the cycle was executed experimentally under prebiotic conditions (indicated next to the arrows), and upon pH changes from basic to slightly acidic to basic, it regenerated 126% of the IDA substrate, confirming autocatalysis. Dashed arrows trace the bypass route (through 4 and 5) that may also be used to regenerate IDA. (B) Plot quantifying the experimentally observed cycle yields for different combinations of the key parameters: the concentration ratio of the IDA and NCA reagents used in the first aminolysis step, the pH during the Strecker reaction, and the concentration of NaOH used for the final hydrolysis [5 M is not a likely prebiotic condition and actually produced suboptimal yield (table S13), but we tested it solely to map the phase space of the cycle]. Circle color corresponds to the yield scale on the right. For all experimental details, see SM section S9. (C) Another noteworthy cycle candidate pending experimental validation and producing up to three copies of each incoming (2-cyanoethyl)glycine molecule (orange). For an average 80% yield of each step, the overall cycle yield would still be ~114%.
The final piece of the puzzle is lipids, which of course make up cellular membranes, both for cells themselves and for intracellular organelles.
The caption:
Emergence of surfactants
Finally, the third class of emergence was the formation of surfactant molecules capable of spontaneously forming vesicles that could potentially house reactions and systems such as those described above. As illustrated in Fig. 6A, straight-chain saturated fatty acid and ?-hydroxy acid surfactants can form through repeated four-step cycles of aldehyde homologation. Breaking the cycle, the last step of fatty acid synthesis, may then occur via nitrile or thioamide hydrolysis to carboxylic acid. The aldehyde homologation cycle was proposed by Patel et al. (8) as a prebiotic method to make hydrophobic amino acids, but its straightforward extension to fatty acid surfactants was not noted in that report. In another and synthetically much shorter approach, peptide surfactants with variable glycine or alanine tails and aspartic acid head groups are available within only a few synthetic generations. Previously, such peptides had been synthesized by modern, nonprebiotic synthetic methods and had been shown (in the context of nanotechnology, not OL research) to form nanotubes and nanovesicles (64).
Finally, the third class of emergence was the formation of surfactant molecules capable of spontaneously forming vesicles that could potentially house reactions and systems such as those described above. As illustrated in Fig. 6A, straight-chain saturated fatty acid and ?-hydroxy acid surfactants can form through repeated four-step cycles of aldehyde homologation. Breaking the cycle, the last step of fatty acid synthesis, may then occur via nitrile or thioamide hydrolysis to carboxylic acid. The aldehyde homologation cycle was proposed by Patel et al. (8) as a prebiotic method to make hydrophobic amino acids, but its straightforward extension to fatty acid surfactants was not noted in that report. In another and synthetically much shorter approach, peptide surfactants with variable glycine or alanine tails and aspartic acid head groups are available within only a few synthetic generations. Previously, such peptides had been synthesized by modern, nonprebiotic synthetic methods and had been shown (in the context of nanotechnology, not OL research) to form nanotubes and nanovesicles (64).
Fig. 6 Biomimetic routes to surfactants and additional pharmaceutically relevant scaffolds.
The caption:
(A) Prebiotic synthesis of fatty acid and ?-hydroxy fatty acid surfactants by iteration of a known prebiotic sequence of four reactions homologating an aldehyde [for reactions 1 to 4, see (8)] followed by a previously unidentified breaking of the cycle via straightforward nitrile or thioamide hydrolysis. (B) A much shorter (i.e., fewer reaction steps) synthesis of peptide surfactants with variable glycine or alanine tails and aspartic acid head groups via sequential addition of Leuchs’ anhydride. (C) Implications of recently reported (78) prebiotically plausible methyl isocyanide formation from HCN, ultimately allowing for Passerini-type reactions (black arrows). Addition of methyl isocyanide to our reaction set substantiates prebiotically plausible syntheses of some useful scaffolds (red arrows): ?-acyloxycarboxamides via a classic Passerini reaction (3-CR, three-component reaction), peptide mimics via the four-component Ugi reaction, or heterocyclic derivatives of pyrazine via a less obvious three-component reaction, which has been reported in non-OL literature under prebiotic conditions (79). In the schemes shown, R1, R2 = alkyl, aryl, or hydrogen; R3 = any carbon; R4 = alkyl, aryl; and R5 = alkyl, aryl. AMP, adenosine monophosphate.
The authors consider, in conclusion that their work may not only assist in approaching the intellectually satisfying question of the origins of life, but may well have practical import as well.
Taken together, the above analyses and synthetic examples lead us to suggest that computational reaction network algorithms are useful for identifying new synthetic routes to prebiotically relevant targets and indispensable for the discovery of prebiotic chemical systems that are otherwise challenging to discern. Naturally, the prebiotic reaction networks should and will grow as distinct prebiotically plausible transformations are experimentally validated. As such transformations are added to our generative set (by means of the crowd-sourcing module illustrated in fig. S5, panel xii), we expect that network analyses will be able to trace prebiotic syntheses starting from primitive feedstocks to increasingly complex scaffolds, including those found in modern drugs (e.g., Fig. 6C). In other words, we envision a fruitful junction between prebiotic chemistries, performed in water and often under environmentally friendly conditions, and environmentally friendly pharmaceutical synthesis…
6 replies
= new reply since forum marked as read
= new reply since forum marked as read
