◉Cell atlasMinimal free-living cell ContentsTreeOriginsFinal analysisSearch
Where am I in the cell?
How hurdles are presented

Each hurdle states the problem (what a cell concretely requires), why it is hard, and what has been shown by experiment, with a link to the related open problems listed in the books. Main categories, as on the open-problems page: chemistry before life; molecular machines; information and control; the integrated cell. Stages: 0 = not yet addressed by experiment; 1 = ingredients made under plausible early-Earth conditions; 2 = one step shown on its own; 3 = several steps shown working together; 4 = shown in a self-sustaining, reproducing system.

The second axis of the atlas

Origins, part by part

For each part of the cell: the question its origin raises, the hurdles a proposal has to meet, and two separate panels. One sets out the naturalistic proposal and what has been shown. The other gives the design analysis.

Naturalistic proposal

What researchers propose and what experiments have shown, with what would strengthen a natural account.

Design analysis

The observations, the hallmarks of design that apply, and the author’s provisional conclusion.

1,976open problems in the origin of the cell
16origins analyses, one per part
40hallmarks of design

Background: Why design? Why not design? · Prebiotic synthesis · Before heredity: what can selection explain? · Philosophy of origins · The 40 hallmarks · Final analysis

The parts of the cell

Each part shows its five key hurdles and the two panels. These are a selection: all 1,976 open problems are listed on Open problems in the origin of the cell.

Building carbon into molecules5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How did carbon capture, energy coupling and catalyst regeneration become compatible in a sustained network?

What needs explaining. Explain how carbon fixation became coupled to energy and reducing-power supply, while producing enough precursors to sustain the catalysts involved.

Hurdles

  1. H-CARB-1Carbon supplyThe problem. Every carbon atom in the cell comes from CO₂, which must be reduced before it can enter metabolism. In Aquifex this is done by the reverse Krebs cycle, whose carboxylating enzymes draw on low-potential electrons from ferredoxin.Why it is hard. CO₂ is a stable, fully oxidised molecule and is dilute in water. Reducing it needs a steady supply of strong reductant at the place where the products are used.Chemistry before lifeStage 1: Ingredients made under plausible early-Earth conditionsShown: CO₂ and simple carbon compounds are plentiful in volcanic and hydrothermal settings.In the books: 37 open problems on central carbon metabolism ↗
  2. H-CARB-2Catalysis and selectivityThe problem. The reverse Krebs cycle has about eleven steps, each catalysed by its own enzyme; the two carbon-fixing steps use iron–sulfur enzymes (pyruvate synthase and 2-oxoglutarate synthase).Why it is hard. Every step must win over competing side reactions. Metal ions speed up several of the steps, but they speed up unwanted reactions too, so the products drift away from the cycle.Molecular machinesStage 2: One step shown on its ownShown: Metal ions speed up several reverse Krebs cycle reactions without enzymes, but with limited selectivity (Muchowska et al. 2017).In the books: 37 open problems on central carbon metabolism ↗
  3. H-CARB-3Energy couplingThe problem. Several steps of the cycle run uphill: the carboxylations need reduced ferredoxin, and splitting citrate needs ATP. Energy has to be delivered to exactly those steps.Why it is hard. A favourable reaction elsewhere does not drive an unfavourable one unless something couples the two. In cells the coupling is done by the enzymes themselves.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 37 open problems on central carbon metabolism ↗
  4. H-CARB-4Network closureThe problem. A cycle fixes carbon only if each turn regenerates its own starting molecule and returns more than it uses up.Why it is hard. If one step leaks or one intermediate decays, the cycle runs down and stops. Every step has to work, at sufficient rate, at the same time.The integrated cellStage 0: Not yet addressed by experimentShown: A self-renewing carbon cycle has not been shown without enzymes.In the books: 37 open problems on central carbon metabolism ↗
  5. H-CARB-5IntegrationThe problem. The cycle's products (acetyl-CoA, pyruvate, oxaloacetate, 2-oxoglutarate) are the starting points for amino acids, sugars, lipids, nucleotides and cofactors. Supply has to match what those pathways consume.Why it is hard. Carbon fixation is useful only when the pathways it feeds exist, and those pathways need the enzymes, cofactors and energy that the rest of the cell provides.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 37 open problems on central carbon metabolism ↗
Naturalistic proposal

Scope. The modern pathways show what selected cells do today. They do not establish that the first living system used both routes, or required every modern enzyme at its beginning.

Experimental progress. A reconstructed biosynthetic network connects carbon assimilation with amino acid, nucleotide, and cofactor production. This provides a systems view of shared chemical inputs; a reconstruction is not an experimental demonstration of the first cell.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Carbon fixation and precursor reactions supply amino-acid, lipid, nucleotide and cofactor pathways. Carbon-transfer and redox reactions require compatible cofactors and regeneration.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Central Carbon Metabolism, with its scientific references.

How it was discovered

  1. 1937Hans Krebs and William Johnson describe the citric acid cycle, the hub of carbon metabolism (Nobel Prize 1953).Krebs & Johnson, Enzymologia 4:148–156 (1937)
  2. 1954Bassham, Benson and Calvin show how plants regenerate the molecule that captures CO₂, completing the Calvin cycle (Nobel Prize 1961).Bassham et al., Journal of the American Chemical Society 76:1760–1770 (1954)
  3. 1966Evans, Buchanan and Arnon propose that some bacteria run the citric acid cycle backwards to fix CO₂: the reductive TCA cycle that Aquifex uses.Evans, Buchanan & Arnon, PNAS 55:928–934 (1966)
  4. 2004Citryl-CoA lyase is discovered in a relative of Aquifex, showing that this group splits citrate in two steps instead of one.Aoshima, Ishii & Igarashi, Molecular Microbiology 52:763–770 (2004)
  5. 2014Braakman and Smith reconstruct the metabolic network of Aquifex aeolicus and use it to propose how carbon fixation first arose.Braakman & Smith, PLoS ONE 9:e87950 (2014)
  6. 2017Metal ions are shown to drive several consecutive reactions of the reverse Krebs cycle without enzymes.Muchowska, Varma & Moran, Nature Ecology & Evolution 1:1716–1721 (2017)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Making the twenty amino acids5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How did sufficiently selective amino-acid production, activation and coded incorporation become integrated?

What needs explaining. Explain amino-acid availability, stereochemical biases and connection to a coding-dependent synthesis system without assuming modern biosynthesis at the outset.

Hurdles

  1. H-AA-1Supply of all twentyThe problem. Proteins are built from twenty amino acids. Aquifex makes all twenty itself, through pathways of several dozen enzymes.Why it is hard. Experiments and meteorites yield mainly the simplest amino acids, mixed with many that life does not use; tryptophan and the basic amino acids are rare or absent.Chemistry before lifeStage 1: Ingredients made under plausible early-Earth conditionsShown: Spark-discharge experiments and meteorites both give a mixture of amino acids.In the books: 98 open problems on amino acid metabolism ↗
  2. H-AA-2Chemical compatibilityThe problem. Amino acids must accumulate in one place at useful concentrations, alongside the other building blocks.Why it is hard. In open water they are diluted, and they react with sugars, aldehydes and other molecules present in the same mixtures, which consumes them.Chemistry before lifeStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 98 open problems on amino acid metabolism ↗
  3. H-AA-3One-handednessThe problem. Proteins use only left-handed (L) amino acids. A single right-handed residue disrupts the regular folds proteins depend on.Why it is hard. Chemical syntheses give equal amounts of both hands. Meteorites show only small excesses of one hand, far from the purity proteins require.Chemistry before lifeStage 1: Ingredients made under plausible early-Earth conditionsShown: Meteorites carry small excesses of one handedness; how a full preference became established is not shown.In the books: 98 open problems on amino acid metabolism ↗
  4. H-AA-4Joining into functional chainsThe problem. Useful proteins are chains of about a hundred or more amino acids in a specific order. In cells they are joined on the ribosome, at the cost of energy for every bond.Why it is hard. Forming a peptide bond in water is uphill, and chains made without a template have random order, so a specific working sequence is not reproduced.Molecular machinesStage 2: One step shown on its ownShown: Wet–dry cycles give short peptide-like chains, without control of sequence (Forsythe et al. 2015).In the books: 98 open problems on amino acid metabolism ↗
  5. H-AA-5Sequence specificationThe problem. A working protein sequence has to be specified and made again in every generation. In cells the sequence is read from a gene through the genetic code.Why it is hard. Without coded information each chain is a one-off: there is no way to make the same useful sequence a second time.Information and controlStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 98 open problems on amino acid metabolism ↗
Naturalistic proposal

Scope. The first peptide-forming chemistry need not have used today’s complete twenty-amino-acid set or modern biosynthetic routes.

Experimental progress. A reconstructed biosynthetic network connects carbon assimilation with amino acid, nucleotide, and cofactor production. This provides a systems view of shared chemical inputs; a reconstruction is not an experimental demonstration of the first cell.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Ordered enzyme reactions convert precursors into amino acids used in protein synthesis. Carbon skeletons, nitrogen, cofactors and energy connect this supply to other systems.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Amino Acid Metabolism, with its scientific references.

How it was discovered

  1. 1953Stanley Miller passes electric sparks through a mixture of gases and obtains several amino acids.Miller, Science 117:528–529 (1953)
  2. 1956Edwin Umbarger finds that the end product of a pathway can switch off its first enzyme: feedback inhibition in amino acid synthesis.Umbarger, Science 123:848 (1956)
  3. 1970Amino acids are found in the Murchison meteorite, showing that they can form outside living things.Kvenvolden et al., Nature 228:923–926 (1970)
  4. 1997Cronin and Pizzarello report a small excess of left-handed forms among some Murchison amino acids.Cronin & Pizzarello, Science 275:951–955 (1997)
  5. 2008A re-analysis of Miller’s stored samples from a volcanic-style experiment finds more than 20 amino acids.Johnson et al., Science 322:404 (2008)
  6. 2015Repeated wetting and drying joins amino acids into short peptide-like chains through ester intermediates.Forsythe et al., Angewandte Chemie 54:9871–9875 (2015)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Making DNA and RNA letters5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How were activated nucleotides produced with suitable stereochemistry, selectivity and compatibility with polymer copying?

What needs explaining. Explain precursor production, polymer-compatible nucleotide chemistry and supply or salvage before the complete modern network existed.

Hurdles

  1. H-NUC-1Feedstocks and synthesisThe problem. A nucleotide is a base, a ribose sugar and a phosphate joined in a fixed way. Cells make them through about forty enzymes.Why it is hard. Laboratory routes reach them from different starting chemicals in separate, purified steps carried out in sequence by the chemist; those conditions are not found together in one natural setting.Chemistry before lifeStage 2: One step shown on its ownShown: Separate laboratory routes give pyrimidine and purine nucleotides from simple starting materials, with steps carried out one after another by the chemist.In the books: 130 open problems on nucleotide metabolism ↗
  2. H-NUC-2Correct connectionsThe problem. Only one of many possible linkages is used: the base on carbon 1 of ribose in the β-furanose form, and phosphate links between carbons 3 and 5.Why it is hard. Chemical mixtures give many alternative linkages and ring forms, and the wrong ones are produced alongside the right one.Chemistry before lifeStage 2: One step shown on its ownShown: Some routes favour the correct sugar–base attachment (Becker et al. 2016).In the books: 130 open problems on nucleotide metabolism ↗
  3. H-NUC-3One-handednessThe problem. RNA and DNA use only D-ribose.Why it is hard. Synthesis gives both hands. Mixed-hand nucleotides interfere with template copying, so purity is needed before copying works well (Joyce et al. 1984).Chemistry before lifeStage 1: Ingredients made under plausible early-Earth conditionsIn the books: 130 open problems on nucleotide metabolism ↗
  4. H-NUC-4Activation and stabilityThe problem. Nucleotides must be activated before they join; cells make triphosphates (ATP, GTP, CTP, UTP) with kinases.Why it is hard. Activated nucleotides break down in water within hours to days, and RNA itself hydrolyses faster at the high temperatures where Aquifex lives.Molecular machinesStage 2: One step shown on its ownShown: Activated cyclic phosphates appear in the 2009 pyrimidine route.In the books: 130 open problems on nucleotide metabolism ↗
  5. H-NUC-5CopyingThe problem. Copying a sequence needs activated monomers, a template, a start, strand separation and continuous resupply.Why it is hard. Copying without enzymes has worked only for short stretches, and the copies stay bound to their templates.Information and controlStage 2: One step shown on its ownShown: Short RNA templates have been copied without enzymes, but not over repeated rounds.In the books: 130 open problems on nucleotide metabolism ↗
Naturalistic proposal

Scope. A nucleotide is not a genetic instruction by itself. The sequence, copying behavior, and functional consequences of polymers must also be explained.

Experimental progress. Experiments in Bacillus show that the nucleotide signal (p)ppGpp binds the regulator PurR and changes control of purine biosynthesis during amino acid starvation.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Carbon and nitrogen precursors supply RNA, DNA and energy-carrier synthesis. Salvage routes can recover material from nucleotide turnover where those pathways are present.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Nucleotide Metabolism, with its scientific references.

How it was discovered

  1. 1869Friedrich Miescher isolates “nuclein”, later known as DNA, from white blood cells.Miescher, Hoppe-Seyler’s Medicinisch-chemische Untersuchungen 4:441–460 (1871)
  2. 1953Watson and Crick propose the double-helix structure of DNA.Watson & Crick, Nature 171:737–738 (1953)
  3. 1960Joan Oró makes adenine, one of the bases of DNA and RNA, from ammonium cyanide.Oró, Biochemical and Biophysical Research Communications 2:407–412 (1960)
  4. 2009Powner, Gerland and Sutherland make activated pyrimidine nucleotides from simple starting materials by a new route.Powner, Gerland & Sutherland, Nature 459:239–242 (2009)
  5. 2016Becker and colleagues report a route that joins purine bases to ribose at the correct position.Becker et al., Science 352:833–836 (2016)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Manufacturing cofactors and vitamins5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How were useful cofactors made, retained and coupled to catalysts before modern biosynthetic enzymes existed?

What needs explaining. Explain how cofactor chemistry, catalyst specificity and cofactor regeneration became compatible before the modern biosynthetic network existed.

Hurdles

  1. H-COF-1Chemical supplyThe problem. Enzymes depend on about a dozen cofactors (NAD, FAD, coenzyme A, folate, thiamine, pyridoxal phosphate, S-adenosylmethionine, heme and others), each made by its own multi-step pathway.Why it is hard. Only fragments of a few of these molecules have been made under proposed early-Earth conditions; complete cofactors have not.Chemistry before lifeStage 1: Ingredients made under plausible early-Earth conditionsShown: Some cofactor parts, such as nicotinamide, have been made under proposed prebiotic conditions.In the books: 86 open problems on cofactor metabolism ↗
  2. H-COF-2Compatible conditionsThe problem. Each cofactor must be made, kept intact and used in the same environment.Why it is hard. Several are unstable to heat, light or oxygen; the conditions that make one tend to destroy another.Chemistry before lifeStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 86 open problems on cofactor metabolism ↗
  3. H-COF-3Binding and specificityThe problem. A cofactor works only when held in a specific pocket of its enzyme, which positions it next to the substrate.Why it is hard. Free cofactors are weak and unselective catalysts; their usefulness comes from the protein that binds them.Molecular machinesStage 2: One step shown on its ownShown: Laboratory-selected ribozymes can use a cofactor such as NAD+, shown one step at a time.In the books: 86 open problems on cofactor metabolism ↗
  4. H-COF-4RecyclingThe problem. Each cofactor cycles between forms (NAD⁺ and NADH, coenzyme A and acyl-CoA, the forms of folate) and must be regenerated after every use.Why it is hard. Regeneration needs other enzymes and a source of electrons or energy; without it the cofactor is used up after one turn.Molecular machinesStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 86 open problems on cofactor metabolism ↗
  5. H-COF-5Mutual dependenceThe problem. The enzymes that make cofactors are themselves proteins that need cofactors.Why it is hard. Cofactor pathways and the enzymes that depend on cofactors require each other, so neither can be present without the other.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 86 open problems on cofactor metabolism ↗
Naturalistic proposal

Scope. A modern cofactor pathway need not have been the first route to that molecule. Modern biosynthesis and prebiotic synthesis are different explanatory questions.

Experimental progress. A reconstructed biosynthetic network connects carbon assimilation with amino acid, nucleotide, and cofactor production. This provides a systems view of shared chemical inputs; a reconstruction is not an experimental demonstration of the first cell.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Cofactor production supports enzymes whose chemistry requires more than a protein chain. Precursor and regeneration pathways connect cofactor supply to metabolism.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Cofactor Metabolism, with its scientific references.

How it was discovered

  1. 1912Casimir Funk proposes the term “vitamine” for dietary factors needed in small amounts.Funk, Journal of State Medicine 20:341–368 (1912)
  2. 1932Warburg and Christian describe the “yellow enzyme”, the first flavoprotein.Warburg & Christian, Biochemische Zeitschrift 254:438–458 (1932)
  3. 1947Lipmann and colleagues identify coenzyme A, the carrier of acetyl groups (Nobel Prize 1953).Lipmann et al., Journal of Biological Chemistry 167:869–870 (1947)
  4. 1956Dorothy Hodgkin solves the structure of vitamin B12 by X-ray crystallography (Nobel Prize 1964).Hodgkin et al., Nature 178:64–66 (1956)
  5. 1972Woodward and Eschenmoser complete the laboratory synthesis of vitamin B12, after about a decade of work by about a hundred chemists.Eschenmoser & Wintner, Science 196:1410–1420 (1977)
  6. 1976Harold White proposes that nucleotide-based cofactors are relics of an earlier, RNA-based stage of metabolism.White, Journal of Molecular Evolution 7:101–104 (1976)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Turning hydrogen into usable energy5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How could geochemical energy sources connect to selective catalysts, membranes and useful energy currencies?

What needs explaining. Explain how electron-transfer chemistry became coupled to useful ion movement, a sufficiently selective membrane and energy-consuming reactions.

Hurdles

  1. H-EN-1Available energyThe problem. The cell needs a continuous energy supply, not a single event. Aquifex obtains it by oxidising hydrogen with oxygen or sulfur compounds, using hydrogenases.Why it is hard. Natural redox and pH differences exist, but a cell has to tap a sustained source and convert it into a form its chemistry can use.Chemistry before lifeStage 1: Ingredients made under plausible early-Earth conditionsShown: Natural redox and pH differences exist at hydrothermal systems; using them is a separate question.In the books: 32 open problems on electron transport and energy production ↗
  2. H-EN-2Coupling energy to synthesisThe problem. Cells make ATP with ATP synthase, a rotary motor built from 8 to 9 kinds of protein (about two dozen chains), driven by protons flowing back across a sealed membrane.Why it is hard. The gradient, the sealed membrane and the motor must exist together: none of the three makes ATP on its own.Molecular machinesStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 32 open problems on electron transport and energy production ↗
  3. H-EN-3A sealed boundaryThe problem. To hold a proton gradient the membrane must be nearly impermeable to protons while still admitting nutrients through dedicated transporters.Why it is hard. Simple fatty-acid membranes leak protons and small ions; the tight phospholipid membranes of cells block nutrients unless transport proteins are present.Molecular machinesStage 2: One step shown on its ownShown: Simple vesicles retain some molecules while letting others in.In the books: 32 open problems on electron transport and energy production ↗
  4. H-EN-4Ordered electron transferThe problem. The electron transport chain passes electrons through carriers in a set order. Complex I alone has 14 core subunits and a chain of about eight iron–sulfur clusters.Why it is hard. Electrons must move forward through every carrier and not leak to side reactions, which depends on exact spacing and orientation inside the proteins.Molecular machinesStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 32 open problems on electron transport and energy production ↗
  5. H-EN-5Self-built energy machineryThe problem. The cell must build its own hydrogenases, electron carriers and ATP synthase from genes, rather than relying on an external gradient.Why it is hard. Building these machines takes ATP, ribosomes and a membrane, which are themselves maintained by the energy the machines provide.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 32 open problems on electron transport and energy production ↗
Naturalistic proposal

Scope. Aerobic respiratory complexes and anaerobic alternatives are not all assumed to coexist. The origin of cellular energy conversion must be assessed for a specified chemical environment.

Experimental progress. Cryo-electron microscopy resolved a bacterial ATP synthase in different rotational states, supporting the structural explanation of membrane-coupled ATP production.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Membrane energy conservation depends on a maintained electrochemical gradient. Electron transfer and ion movement must couple to ATP production in the selected respiratory chain.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Electron Transport and Energy Production, with its scientific references.

How it was discovered

  1. 1929Karl Lohmann, and independently Fiske and Subbarow, discover ATP, the cell’s energy currency.Lohmann, Naturwissenschaften 17:624–625 (1929)
  2. 1961Peter Mitchell proposes that cells make ATP using gradients of ions across membranes (Nobel Prize 1978).Mitchell, Nature 191:144–148 (1961)
  3. 1994Abrahams, Leslie, Lutter and Walker solve the structure of the catalytic part of ATP synthase (Nobel Prize 1997, with Boyer).Abrahams et al., Nature 370:621–628 (1994)
  4. 1997Noji and colleagues watch a single ATP synthase head rotate under a microscope.Noji et al., Nature 386:299–302 (1997)
  5. 2012Lane and Martin propose that natural ion gradients at alkaline hydrothermal vents powered the first cells.Lane & Martin, Cell 151:1406–1416 (2012)
  6. 2013The complete structure of respiratory complex I, an electron-driven proton pump, is solved in a heat-loving bacterium.Baradaran et al., Nature 494:443–448 (2013)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Building and maintaining the membrane5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How did amphiphile production, compartment stability and compatible chemistry become coupled before modern lipid enzymes?

What needs explaining. Explain a growing boundary that can admit resources, retain useful components and support compatible membrane proteins and gradients.

Hurdles

  1. H-MEM-1Lipid supplyThe problem. Cell membranes are made of phospholipids (two fatty-acid chains, glycerol and a phosphate head), produced by about twenty enzymes.Why it is hard. Prebiotic syntheses yield mostly short fatty acids at low concentration, while stable vesicles need longer chains at concentrations of tens of millimolar.Chemistry before lifeStage 1: Ingredients made under plausible early-Earth conditionsShown: Fatty acids form in proposed prebiotic syntheses and are found in meteorites.In the books: 36 open problems on lipid and membrane metabolism ↗
  2. H-MEM-2Salt and pH toleranceThe problem. The membrane must stay intact in the salt, magnesium and pH of the cell's environment.Why it is hard. Fatty-acid vesicles fall apart in sea-water salt and with magnesium or calcium, and are stable only in a narrow pH range.Chemistry before lifeStage 2: One step shown on its ownShown: Mixtures of different amphiphiles make vesicles more tolerant of salt and pH.In the books: 36 open problems on lipid and membrane metabolism ↗
  3. H-MEM-3Permeability trade-offThe problem. The membrane must keep proteins, RNA and metabolites inside while letting nutrients and waste through.Why it is hard. Leaky membranes lose contents; tight membranes block nutrients. Cells resolve this with transport proteins, which need the membrane to exist first.Molecular machinesStage 2: One step shown on its ownShown: Fatty-acid vesicles let some nucleotides in while keeping larger molecules inside.In the books: 36 open problems on lipid and membrane metabolism ↗
  4. H-MEM-4Growth and divisionThe problem. Cells grow their membrane by inserting newly made lipids and divide it with a dedicated machinery (FtsZ and partner proteins).Why it is hard. Growth and division must keep the internal network intact; in the laboratory vesicles divide only when pushed by external forces.The integrated cellStage 3: Several steps shown working togetherShown: Vesicles grow and divide in repeated cycles in the laboratory, though without an internal metabolism (Zhu & Szostak 2009).In the books: 36 open problems on lipid and membrane metabolism ↗
  5. H-MEM-5Cellular membrane compositionThe problem. A cell membrane needs lipid-synthesis enzymes, which sit in the membrane they make, plus transport and energy-coupling proteins.Why it is hard. A phospholipid membrane blocks the nutrient entry that simple vesicles allow, so it is workable only when transporters are present at the same time.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 36 open problems on lipid and membrane metabolism ↗
Naturalistic proposal

Scope. Spontaneous lipid assembly is an established physical process. The open problem concerns the origin and sustained integration of boundary, metabolism, and heredity.

Experimental progress. Manipulating fatty-acid elongation in bacteria changes the chain-length distribution of products, illustrating how enzyme activity controls lipid building blocks.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. The membrane co-locates reactions and provides a functional setting for membrane proteins. Ion-gradient storage requires a sufficiently selective and intact boundary.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Lipid and Membrane Metabolism, with its scientific references.

How it was discovered

  1. 1925Gorter and Grendel conclude that cell membranes are a double layer of lipids.Gorter & Grendel, Journal of Experimental Medicine 41:439–443 (1925)
  2. 1965Alec Bangham shows that lipids spontaneously form closed vesicles in water (liposomes).Bangham, Standish & Watkins, Journal of Molecular Biology 13:238–252 (1965)
  3. 1972Singer and Nicolson describe the membrane as a fluid lipid layer with proteins floating in it.Singer & Nicolson, Science 175:720–731 (1972)
  4. 1978Hargreaves and Deamer show that simple fatty acids, not only modern phospholipids, can form vesicles.Hargreaves & Deamer, Biochemistry 17:3759–3768 (1978)
  5. 2003Hanczyc, Fujikawa and Szostak find that clay speeds up the formation of fatty-acid vesicles, which can grow and divide.Hanczyc, Fujikawa & Szostak, Science 302:618–622 (2003)
  6. 2009Zhu and Szostak show fatty-acid vesicles growing and dividing in repeated cycles while keeping most of their contents.Zhu & Szostak, Journal of the American Chemical Society 131:5705–5713 (2009)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Copying and repairing the genome5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How did copying, strand separation and error control first become linked?

What needs explaining. Explain template copying, sufficient fidelity and chromosome management, distinguishing plausible early copying from modern DNA replication.

Hurdles

  1. H-GEN-1Faithful copyingThe problem. A genome of over a million bases must be copied with fewer than about one error per million bases, which cells achieve with proofreading polymerases and repair enzymes.Why it is hard. Copying without enzymes makes errors at rates of several percent per base, too high to keep a sequence of useful length.Information and controlStage 2: One step shown on its ownShown: Short templates have been copied without enzymes, but not over repeated rounds.In the books: 112 open problems on dna processing ↗
  2. H-GEN-2DNA building blocksThe problem. Cells make DNA building blocks from RNA building blocks with ribonucleotide reductase, a complex radical enzyme.Why it is hard. Prebiotic routes to deoxyribose and deoxynucleotides give low yields, and the enzyme that makes them in cells is itself a large protein.Chemistry before lifeStage 1: Ingredients made under plausible early-Earth conditionsShown: Prebiotic routes to some DNA building blocks have been reported.In the books: 112 open problems on dna processing ↗
  3. H-GEN-3Copying and separationThe problem. Replication needs a helicase to open the double helix, a primase, a polymerase, a sliding clamp, a ligase and gyrase to relieve twisting.Why it is hard. Separating long double strands by heat also destroys other components; enzymatic separation needs the replication proteins themselves.Molecular machinesStage 2: One step shown on its ownShown: Copying short templates and separating strands by heating have each been shown on their own.In the books: 112 open problems on dna processing ↗
  4. H-GEN-4Error thresholdThe problem. To keep its information, a genome must be copied with fewer errors than about one per genome length per generation.Why it is hard. Above that error rate the information degrades with every copy. Staying below it needs proofreading and repair proteins, which the genome itself must encode.Information and controlStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 112 open problems on dna processing ↗
  5. H-GEN-5Replication–division couplingThe problem. Each daughter cell must receive a complete genome, so copying must finish and the copies be separated before division.Why it is hard. Cells coordinate this with dedicated proteins (DnaA, SMC and partitioning proteins); without them daughters can be left without genes.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 112 open problems on dna processing ↗
Naturalistic proposal

Scope. A requirement of modern DNA cells is not automatically a requirement of the earliest hereditary system. Origin hypotheses must specify their proposed starting chemistry.

Experimental progress. Single-molecule experiments show dynamic behavior of bacterial replication polymerases, helping distinguish functional coordination from rigid synchronization.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Template copying coordinates strand separation with repeated nucleotide addition. DNA twisting and entanglement must be managed during replication and chromosome handling.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article DNA Processing, with its scientific references.

How it was discovered

  1. 1953Watson and Crick point out that DNA base pairing suggests how the molecule could be copied.Watson & Crick, Nature 171:964–967 (1953)
  2. 1956Arthur Kornberg discovers DNA polymerase, the enzyme that builds new DNA on a template (Nobel Prize 1959).Kornberg et al., Biochimica et Biophysica Acta 21:197–198 (1956)
  3. 1958Meselson and Stahl show that each new DNA molecule keeps one old strand and gains one new one.Meselson & Stahl, PNAS 44:671–682 (1958)
  4. 1968Okazaki and colleagues find that one DNA strand is copied in short pieces that are later joined.Okazaki et al., PNAS 59:598–605 (1968)
  5. 1976Gellert and colleagues discover DNA gyrase, which relieves the twisting that copying creates.Gellert et al., PNAS 73:3872–3876 (1976)
  6. 1993Tomas Lindahl shows that DNA decays on its own in water, so constant repair is needed (Nobel Prize 2015).Lindahl, Nature 362:709–715 (1993)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Reading genes into RNA5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How did templated copying, adaptors, amino-acid activation and peptide synthesis become a heritable coded translation system?

What needs explaining. Explain the coupling of message, adaptor charging, peptide synthesis and inherited maintenance, distinguishing prebiotic chemistry from a complete translation system.

Hurdles

  1. H-RNA-1Functional RNAsThe problem. Translation needs ribosomal RNAs of about 1,500 and 2,900 nucleotides, dozens of transfer RNAs and other structured RNAs.Why it is hard. Catalytic RNAs obtained in the laboratory are short, work only in selected conditions, and are far from the size and accuracy of these RNAs.Molecular machinesStage 2: One step shown on its ownShown: RNA chains form on clay surfaces, and catalytic RNAs can be selected in the laboratory.In the books: 144 open problems on rna processing and translation ↗
  2. H-RNA-2Code assignmentsThe problem. The genetic code assigns 64 codons to 20 amino acids and stop signals. In cells the assignments are made by about twenty enzymes (aminoacyl-tRNA synthetases), each recognising one amino acid and its tRNAs.Why it is hard. No chemical rule fixes which codon goes with which amino acid; the assignments exist only in the enzymes that implement them.Information and controlStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 144 open problems on rna processing and translation ↗
  3. H-RNA-3Adapters and chargingThe problem. Each tRNA must carry the right amino acid. Synthetases attach them with about one error in ten thousand, helped by editing sites.Why it is hard. Catalytic RNAs from the laboratory attach amino acids with low specificity, which would scramble every protein.Molecular machinesStage 2: One step shown on its ownShown: Laboratory-selected ribozymes can attach amino acids to RNA.In the books: 144 open problems on rna processing and translation ↗
  4. H-RNA-4Mutual dependenceThe problem. The ribosome contains over fifty proteins as well as its RNAs, and the synthetases are proteins too; all of them are made by translation.Why it is hard. The machinery that makes proteins is itself made of proteins, so it cannot be built until it already exists.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 144 open problems on rna processing and translation ↗
  5. H-RNA-5Self-reproduction of translationThe problem. The whole translation apparatus must be rebuilt in every generation, using about four energy-rich bonds for every peptide bond.Why it is hard. Translation depends on monomer supply, energy and accurate copying of its own genes, all of which in turn depend on translation.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 144 open problems on rna processing and translation ↗
Naturalistic proposal

Scope. The ribosome animation illustrates selected bacterial translation steps. It is not a reconstruction of the first translation apparatus or a complete account of all RNA processing.

Experimental progress. Time-resolved cryo-electron microscopy captured bacterial ribosome–EF-G intermediates during GTP-associated translocation, directly examining movement within the translation cycle.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. mRNA sequence becomes amino-acid sequence through adaptors, charging enzymes and ribosomes. Translation produces components required for further gene expression; closure belongs to the interacting system.

Hallmarks that apply.

  • H07 Translation: mRNA sequence becomes amino-acid sequence through adaptors, charging enzymes and ribosomes.
  • H11 Semiotic closure: Translation produces components required for further gene expression; closure belongs to the interacting system.
  • H33 Multi-level error monitoring: Charging, decoding, maturation and rescue handle errors at different stages.

The author’s conclusion. Within the author’s framework, the convergence of coding, translation and inherited construction supports a provisional preference for design as an explanation of the integrated architecture. This is a qualitative abductive judgment, not an experimentally established origin or a calculated probability. Its force depends on whether the comparison with agency is causally informative and whether alternative pathways can explain the same organization.

From the book article RNA Processing and Translation, with its scientific references.

How it was discovered

  1. 1959An enzyme that makes RNA from a DNA template, RNA polymerase, is first described.Weiss & Gladstone, Journal of the American Chemical Society 81:4118–4119 (1959)
  2. 1961Brenner, Jacob and Meselson show that a short-lived messenger RNA carries genetic instructions to the ribosome.Brenner, Jacob & Meselson, Nature 190:576–581 (1961)
  3. 1969Burgess and colleagues discover the sigma factor, which tells RNA polymerase where genes start.Burgess et al., Nature 221:43–46 (1969)
  4. 1982Thomas Cech finds an RNA that cuts and rejoins itself without a protein: RNA can be a catalyst (Nobel Prize 1989).Kruger et al., Cell 31:147–157 (1982)
  5. 1983Sidney Altman’s group shows that the RNA part of RNase P does the cutting (Nobel Prize 1989).Guerrier-Takada et al., Cell 35:849–857 (1983)
  6. 1986Walter Gilbert names the “RNA world”: a proposed stage when RNA both stored information and catalysed reactions.Gilbert, Nature 319:618 (1986)
  7. 2001A ribozyme made by laboratory selection copies up to 14 nucleotides of another RNA.Johnston et al., Science 292:1319–1325 (2001)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Building proteins and folding them5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How did templated copying, adaptors, amino-acid activation and peptide synthesis become a heritable coded translation system?

What needs explaining. Explain the coupling of message, adaptor charging, peptide synthesis and inherited maintenance, distinguishing prebiotic chemistry from a complete translation system.

Hurdles

  1. H-RNA-1Functional RNAsProduce and retain RNAs that fold and catalyze reactions in compatible conditions.chemistryStage 2: One step shown on its ownRNA chains form on clay surfaces, and catalytic RNAs can be selected in the laboratory.
  2. H-RNA-2Code assignmentsEstablish reproducible relationships between nucleotide patterns and amino acids.informationStage 0: Not yet addressed by experiment
  3. H-RNA-3Adapters and chargingLink amino acids to suitable adapters with enough accuracy to support useful peptides.coordinationStage 2: One step shown on its ownLaboratory-selected ribozymes can attach amino acids to RNA.
  4. H-RNA-4Mutual dependenceExplain how RNA catalysts and peptides first came to work together, without assuming a series of earlier versions.coordinationStage 0: Not yet addressed by experiment
  5. H-RNA-5Supply and heredityConnect translation to monomer production, energy supply, and reliable reproduction of its own components.coordinationStage 0: Not yet addressed by experiment
Naturalistic proposal

Scope. The ribosome animation illustrates selected bacterial translation steps. It is not a reconstruction of the first translation apparatus or a complete account of all RNA processing.

Experimental progress. Time-resolved cryo-electron microscopy captured bacterial ribosome–EF-G intermediates during GTP-associated translocation, directly examining movement within the translation cycle.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. mRNA sequence becomes amino-acid sequence through adaptors, charging enzymes and ribosomes. Translation produces components required for further gene expression; closure belongs to the interacting system.

Hallmarks that apply.

  • H07 Translation: mRNA sequence becomes amino-acid sequence through adaptors, charging enzymes and ribosomes.
  • H11 Semiotic closure: Translation produces components required for further gene expression; closure belongs to the interacting system.
  • H33 Multi-level error monitoring: Charging, decoding, maturation and rescue handle errors at different stages.

The author’s conclusion. Within the author’s framework, the convergence of coding, translation and inherited construction supports a provisional preference for design as an explanation of the integrated architecture. This is a qualitative abductive judgment, not an experimentally established origin or a calculated probability. Its force depends on whether the comparison with agency is causally informative and whether alternative pathways can explain the same organization.

From the book article RNA Processing and Translation, with its scientific references.

How it was discovered

  1. 1958Hoagland, Zamecnik and colleagues find transfer RNA, the adapter that carries amino acids to the ribosome.Hoagland et al., Journal of Biological Chemistry 231:241–257 (1958)
  2. 1961Nirenberg and Matthaei show that the RNA sequence UUU codes for phenylalanine: the first codon read (Nobel Prize 1968).Nirenberg & Matthaei, PNAS 47:1588–1602 (1961)
  3. 1966All 64 codons of the genetic code have been assigned.Crick, Cold Spring Harbor Symposia on Quantitative Biology 31:1–9 (1966)
  4. 1973Christian Anfinsen concludes that a protein’s sequence determines how it folds (Nobel Prize 1972).Anfinsen, Science 181:223–230 (1973)
  5. 1997The structure of the GroEL–GroES chaperonin shows the protected chamber in which proteins can fold.Xu, Horwich & Sigler, Nature 388:741–750 (1997)
  6. 2000The atomic structure of the large ribosomal subunit shows that RNA, not protein, forms peptide bonds (Nobel Prize 2009).Nissen et al., Science 289:920–930 (2000)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Dividing into two cells5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How did growth, compartment division and inheritance become coordinated in early reproducing systems?

What needs explaining. Explain growth, genome inheritance and physical separation that yield viable descendants, while allowing that early division could differ from modern divisomes.

Hurdles

  1. H-DIV-1Coordinated growthThe problem. To split into two cells of equal size, the surface must grow by about 26% more than the volume alone would require.Why it is hard. Membrane and contents grow by different processes; if they are not matched, cells swell, shrink or divide unevenly.The integrated cellStage 2: One step shown on its ownShown: Fatty-acid vesicles grow by taking up new amphiphiles.In the books: 64 open problems on cell division and structure ↗
  2. H-DIV-2Content retentionThe problem. Division must separate the membrane into two closed compartments without losing contents.Why it is hard. Cells constrict and seal the membrane with a ring of FtsZ and partner proteins; without such a machine, division tears or leaks.Molecular machinesStage 2: One step shown on its ownShown: Vesicles can divide while keeping most of their contents (Zhu & Szostak 2009).In the books: 64 open problems on cell division and structure ↗
  3. H-DIV-3Genome partitioningThe problem. Each daughter must receive at least one complete genome.Why it is hard. With only one or a few genome copies, random splitting leaves some daughters without genes; cells prevent this with active partitioning systems.Information and controlStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 64 open problems on cell division and structure ↗
  4. H-DIV-4Reproductive continuityThe problem. Each daughter must contain enough of every essential component (ribosomes, enzymes, membrane) to start growing again.Why it is hard. Missing even one essential component in a daughter ends that line.The integrated cellStage 3: Several steps shown working togetherShown: Growth and division repeat over several cycles in the laboratory, without an internal metabolism.In the books: 64 open problems on cell division and structure ↗
  5. H-DIV-5Coordinated timingThe problem. Division must wait until the genome has been copied and separated, and growth must keep pace with both.Why it is hard. Cells time these events with dedicated regulators (DnaA, FtsZ and nucleoid occlusion). Without timing, daughters are empty or overfilled.Information and controlStage 2: One step shown on its ownShown: Vesicles carrying RNA can take membrane from empty ones (Chen et al. 2004), a physical link between contents and growth; timing growth, copying and division together has not been shown.In the books: 64 open problems on cell division and structure ↗
Naturalistic proposal

Scope. How a modern bacterium divides and how a first compartment could divide without proteins are separate questions. An origins explanation must connect these stages rather than equate them.

Experimental progress. Gene-addition experiments in engineered minimal cells identify contributors to division and morphology. These cells depend on provided culture conditions and are not a reconstructed prebiotic cell.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Division components must assemble at an appropriate location and geometry. Chromosome handling and physical division contribute different requirements for viable descendants.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Cell Division and Structure, with its scientific references.

How it was discovered

  1. 1989De Boer, Crossley and Rothfield describe the Min system, which keeps division at the middle of the cell.de Boer, Crossley & Rothfield, Cell 56:641–649 (1989)
  2. 1991Bi and Lutkenhaus see FtsZ form a ring at the site where the cell will divide.Bi & Lutkenhaus, Nature 354:161–164 (1991)
  3. 1998Löwe and Amos solve the structure of FtsZ and find it resembles tubulin.Löwe & Amos, Nature 391:203–206 (1998)
  4. 2004Vesicles carrying RNA are shown to grow by taking membrane from empty neighbours.Chen, Roberts & Szostak, Science 305:1474–1476 (2004)
  5. 2009Bacteria without a cell wall (L-forms) are shown to divide without FtsZ.Leaver et al., Nature 457:849–853 (2009)
  6. 2021Adding back a few genes restores regular division in the minimal cell JCVI-syn3A.Pelletier et al., Cell 184:2430–2440 (2021)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Pulling nutrients in from outside5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How did selective permeability and active transport develop while maintaining a useful compartment?

What needs explaining. Explain selective membrane passage, energy coupling and regulation under the resources available to the proposed early system.

Hurdles

  1. H-TR-1Getting nutrients inThe problem. A cell's membrane blocks ions, sugars, amino acids and nucleotides, so each nutrient needs a transporter.Why it is hard. Tight membranes are needed to hold gradients and keep contents, yet they stop nutrients; leaky membranes let nutrients in but lose contents.Molecular machinesStage 2: One step shown on its ownShown: Fatty-acid vesicles let some nucleotides in while keeping larger molecules inside.In the books: 152 open problems on transport systems ↗
  2. H-TR-2SelectivityThe problem. Transporters must tell apart very similar molecules and ions; the potassium channel lets K⁺ through about a thousand times more readily than Na⁺, although Na⁺ is smaller.Why it is hard. That discrimination depends on precise atomic positions inside the protein (Doyle et al. 1998).Molecular machinesStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 152 open problems on transport systems ↗
  3. H-TR-3Powered uptakeThe problem. Importing nutrients against their concentration needs energy: ATP-driven pumps of several subunits, or coupling to an ion gradient.Why it is hard. The pump and the energy supply must work together; transport also consumes part of the energy the cell produces.Molecular machinesStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 152 open problems on transport systems ↗
  4. H-TR-4Insertion and orientationThe problem. Transport proteins must be threaded into the membrane the right way round, by the Sec translocon and YidC.Why it is hard. The insertion machines are themselves membrane proteins that needed inserting.Molecular machinesStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 152 open problems on transport systems ↗
  5. H-TR-5System balanceThe problem. Influx, efflux and internal use must be balanced for every substance.Why it is hard. Imbalance leads to swelling, shrinking, or build-up of toxic levels; balance needs many transporters working together.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 152 open problems on transport systems ↗
Naturalistic proposal

Scope. A transporter demonstrated in one organism is an example of a mechanism. Its inclusion in this catalogue does not make that exact protein universal or indispensable in every environment.

Experimental progress. A structure of the BtuCD–F vitamin B12 transporter bound to an ATP analogue reveals a nucleotide-associated state of an ABC uptake system.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Binding sites and channels select transported molecules or ions. ATP or an ion gradient can power transport, depending on the transporter.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Transport Systems, with its scientific references.

How it was discovered

  1. 1956Monod’s group shows that bacteria use a specific protein, a “permease”, to take up lactose.Rickenberg et al., Annales de l’Institut Pasteur 91:829–857 (1956)
  2. 1957Jens Skou discovers the sodium–potassium pump, the first known ion-pumping enzyme (Nobel Prize 1997).Skou, Biochimica et Biophysica Acta 23:394–401 (1957)
  3. 1986Higgins and colleagues recognise a large family of ATP-driven transporters, now called ABC transporters.Higgins et al., Nature 323:448–450 (1986)
  4. 1998The structure of the bacterial potassium channel KcsA shows how it lets potassium through but not sodium (Nobel Prize 2003).Doyle et al., Science 280:69–77 (1998)
  5. 2002The first structure of a complete ABC transporter, the vitamin B12 importer BtuCD, is solved.Locher, Lee & Rees, Science 296:1091–1098 (2002)
  6. 2003The structure of lactose permease shows how a sugar is carried in together with a proton.Abramson et al., Science 301:610–615 (2003)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Holding the internal balance5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How were stable internal conditions maintained in compartments while sustaining material and energy flow?

What needs explaining. Explain the coupling of transport and reactions that maintains viable internal conditions under environmental fluctuations.

Hurdles

  1. H-BAL-1Compatible operating rangesThe problem. Enzymes, RNAs and membranes each work only within limited ranges of pH, salt and temperature, and those ranges must overlap.Why it is hard. Conditions that suit one component often damage another, especially at the high temperatures where Aquifex lives.Chemistry before lifeStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 77 open problems on cellular homeostasis ↗
  2. H-BAL-2Exchange without collapseThe problem. The cell must take in resources while keeping its internal pH, water balance and ion gradients.Why it is hard. Every opening for nutrients is also a route for losing ions and water; cells control this with pumps, channels and compatible solutes.The integrated cellStage 2: One step shown on its ownShown: Simple fatty-acid vesicles let some nucleotides in while keeping larger molecules inside.In the books: 77 open problems on cellular homeostasis ↗
  3. H-BAL-3FeedbackThe problem. When conditions drift, the cell must detect the change and trigger a correction.Why it is hard. Feedback needs a sensor, a signal and a response, each a dedicated molecule, linked in the right way.Information and controlStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 77 open problems on cellular homeostasis ↗
  4. H-BAL-4Energy budgetThe problem. Maintenance (pumping ions, repairing damage, replacing molecules) consumes a large share of a cell's energy, and enough must remain for growth and division.Why it is hard. If maintenance costs exceed supply, the cell cannot grow; the energy system and the maintenance systems depend on each other.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 77 open problems on cellular homeostasis ↗
  5. H-BAL-5Self-maintained internal conditionsThe problem. The cell must keep its own internal pH, ion levels and water balance, with pumps, channels and regulators specified by its genes.Why it is hard. An environment cannot hold the inside of a cell constant; the cell has to do it with encoded machinery from the start.Information and controlStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 77 open problems on cellular homeostasis ↗
Naturalistic proposal

Scope. Homeostasis overlaps the other categories because it is an outcome of their coordination. Shared components should not be counted as additional independent proteins merely because they serve several functions.

Experimental progress. Structural and functional experiments on MgtE link magnesium-dependent channel gating with magnesium homeostasis, providing a concrete example of regulated ion uptake.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Viable pH, osmotic and redox ranges depend on regulated transport and reactions. Feedback can connect disturbance to compensating activity.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Cellular Homeostasis, with its scientific references.

How it was discovered

  1. 1961Peter Mitchell proposes that cells store energy as gradients of ions across membranes (Nobel Prize 1978).Mitchell, Nature 191:144–148 (1961)
  2. 1964Laurent, Moore and Reichard discover thioredoxin in E. coli as the hydrogen donor for making DNA building blocks.Laurent, Moore & Reichard, Journal of Biological Chemistry 239:3436–3444 (1964)
  3. 1975The three-dimensional structure of E. coli thioredoxin is determined.Holmgren et al., PNAS 72:2305–2309 (1975)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Surviving oxygen damage5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How did persistence under fluctuating conditions, chemical damage control and repair become compatible in early systems?

What needs explaining. Explain protective chemistry and inducible responses appropriate to actual historical stresses, without treating modern exposure-specific pathways as first-cell necessities.

Hurdles

  1. H-STR-1Environmental exposureThe problem. The cell's setting brings heat, ultraviolet light, reactive oxygen and toxic metals.Why it is hard. Each of these damages proteins, nucleic acids and membranes, and the specific agents depend on the setting assumed.Chemistry before lifeStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 54 open problems on stress response ↗
  2. H-STR-2PersistenceThe problem. Proteins, RNA and energy carriers must last long enough to be useful.Why it is hard. High temperature speeds up the breakdown of RNA, ATP and peptide bonds; at the temperatures where Aquifex lives, many such molecules last minutes to hours.Chemistry before lifeStage 2: One step shown on its ownShown: Mineral surfaces and vesicles can slow the breakdown of some fragile molecules.In the books: 54 open problems on stress response ↗
  3. H-STR-3Protection trade-offsThe problem. Protection must shield the cell without cutting it off from the resources it needs.Why it is hard. Barriers that keep out damaging agents also keep out nutrients; protection must be selective.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 54 open problems on stress response ↗
  4. H-STR-4TurnoverThe problem. Damaged molecules must be removed and replaced faster than damage accumulates.Why it is hard. Replacement costs energy and materials and depends on the synthesis machinery being intact.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 54 open problems on stress response ↗
  5. H-STR-5Encoded protectionThe problem. Protective enzymes (superoxide dismutase, peroxidases, chaperones and repair enzymes) must be specified by genes and made when needed.Why it is hard. Protection must be present before damage occurs, and the instructions for it must pass to every daughter cell.Information and controlStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 54 open problems on stress response ↗
Naturalistic proposal

Scope. The severity of a hurdle depends on exposure. Modern oxidative defenses cannot be imposed unchanged on every proposed early-Earth setting.

Experimental progress. Experiments in Escherichia coli show that hydrogen peroxide can damage selected iron–sulfur enzymes and interfere with metabolism, linking protective chemistry to continued enzyme function.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Stress recognition and regulation can change protective activity. Damage can require refolding or degradation, depending on the lesion.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Stress Response, with its scientific references.

How it was discovered

  1. 1969McCord and Fridovich discover superoxide dismutase, the first enzyme known to remove a reactive form of oxygen.McCord & Fridovich, Journal of Biological Chemistry 244:6049–6055 (1969)
  2. 1973Yost and Fridovich purify an iron-containing superoxide dismutase from E. coli, the type Aquifex uses (sodB).Yost & Fridovich, Journal of Biological Chemistry 248:4905–4908 (1973)
  3. 1989Jacobson and colleagues describe alkyl hydroperoxide reductase (AhpC), which protects DNA from oxidative damage.Jacobson et al., Journal of Biological Chemistry 264:1488–1496 (1989)
  4. 1995FtsH is shown to be a membrane-bound, ATP-dependent protease that removes specific proteins.Tomoyasu et al., EMBO Journal 14:2551–2560 (1995)
  5. 2014A review of the evidence dates the first lasting rise of oxygen in the atmosphere to about 2.4 billion years ago. Before that, reactive oxygen came mainly from sunlight and radiation acting on water.Lyons, Reinhard & Planavsky, Nature 506:307–315 (2014)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Loading metals into enzymes5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How were useful metal centres assembled, retained and coupled to catalysts under plausible environmental conditions?

What needs explaining. Explain metal-centre chemistry, controlled assembly and delivery to compatible protein recipients before the full modern maturation network existed.

Hurdles

  1. H-MET-1Available atomsThe problem. Iron–sulfur clusters need iron, sulfide and cysteine ligands together.Why it is hard. Free iron and sulfide precipitate as insoluble iron sulfide minerals rather than forming defined clusters.Chemistry before lifeStage 1: Ingredients made under plausible early-Earth conditionsShown: Iron, sulfide and simple ligands are expected in volcanic and hydrothermal settings.In the books: 9 open problems on metal cluster assembly ↗
  2. H-MET-2Productive assemblyThe problem. Cells build clusters on scaffold proteins, using a cysteine desulfurase to release sulfur and carrier proteins to deliver the finished cluster.Why it is hard. Without this machinery, iron and sulfur form many structures, most of them inactive.Molecular machinesStage 2: One step shown on its ownShown: Iron–sulfur clusters form on short peptides under laboratory conditions (Bonfio et al. 2017).In the books: 9 open problems on metal cluster assembly ↗
  3. H-MET-3StabilityThe problem. Clusters must stay intact while they work.Why it is hard. Iron–sulfur clusters are destroyed by oxygen and by heat unless the protein around them protects them.Molecular machinesStage 2: One step shown on its ownIn the books: 9 open problems on metal cluster assembly ↗
  4. H-MET-4Insertion into enzymesThe problem. Each cluster must be placed in a specific site of its target protein, with the right ligands, which tune its electrical properties.Why it is hard. A cluster outside its protein has the wrong properties; insertion is done by dedicated carrier proteins.Molecular machinesStage 1: Ingredients made under plausible early-Earth conditionsIn the books: 9 open problems on metal cluster assembly ↗
  5. H-MET-5Continuous rebuildingThe problem. Clusters must be repaired and rebuilt continuously, and the machinery that does it must be passed on to every daughter cell.Why it is hard. That machinery is made of proteins whose genes and synthesis themselves depend on iron–sulfur enzymes.Information and controlStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 9 open problems on metal cluster assembly ↗
Naturalistic proposal

Scope. Different metal centers are alternatives or specialized additions, not a single mandatory package. The origin of each assembly route requires a specified catalyst and environment.

Experimental progress. Structures of IscS complexes show different binding arrangements with sulfur-accepting partners, providing a molecular basis for directed sulfur transfer.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Metalloenzymes may require separate machinery to build and insert active centres. Metal and sulfur handling must produce a compatible cofactor rather than an arbitrary aggregate.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Metal Cluster Assembly, with its scientific references.

How it was discovered

  1. 1962Mortenson, Valentine and Carnahan isolate ferredoxin, an iron–sulfur protein, from Clostridium pasteurianum.Mortenson, Valentine & Carnahan, Biochemical and Biophysical Research Communications 7:448–452 (1962)
  2. 1988Günter Wächtershäuser proposes that life began on iron–sulfur minerals (the “iron–sulfur world”).Wächtershäuser, Systematic and Applied Microbiology 10:207–210 (1988)
  3. 1991The hyp genes needed to put the nickel–iron centre into hydrogenases are identified in E. coli.Lutz et al., Molecular Microbiology 5:123–135 (1991)
  4. 1993NifS is shown to be a cysteine desulfurase that supplies sulfur for iron–sulfur clusters.Zheng et al., PNAS 90:2754–2758 (1993)
  5. 1995The first crystal structure of a nickel–iron hydrogenase is solved.Volbeda et al., Nature 373:580–587 (1995)
  6. 1998The isc genes for general iron–sulfur cluster assembly, including the scaffold IscU, are described.Zheng et al., Journal of Biological Chemistry 273:13264–13272 (1998)
  7. 2017Ultraviolet light drives the formation of iron–sulfur clusters bound to short peptides under proposed prebiotic conditions.Bonfio et al., Nature Chemistry 9:1229–1234 (2017)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Switching genes on and off5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How did a detectable change become linked to a useful response and an inheritable regulatory architecture?

What needs explaining. Explain how sensing and response became linked to a useful physiological outcome rather than merely to chemical reactivity.

Hurdles

  1. H-REG-1Signal–response couplingThe problem. Responding to a change means linking a sensor (for example a histidine kinase) to a regulator and to the genes it controls.Why it is hard. Each link is a specific molecular interaction; the sensor, the regulator and the DNA binding site must all match.Information and controlStage 2: One step shown on its ownShown: Laboratory-selected RNAs can change their activity when they bind a small molecule.In the books: 270 open problems on codes, signaling, regulation and adaptation ↗
  2. H-REG-2SpecificityThe problem. Each sensor must respond to its own signal and talk only to its own partner, although many similar systems exist side by side.Why it is hard. Cross-talk between similar sensors and regulators would trigger the wrong responses.Information and controlStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 270 open problems on codes, signaling, regulation and adaptation ↗
  3. H-REG-3TimingThe problem. Responses must be fast enough to follow environmental change and stable enough not to switch back and forth.Why it is hard. Achieving both needs feedback and finely set binding strengths.Information and controlStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 270 open problems on codes, signaling, regulation and adaptation ↗
  4. H-REG-4CostThe problem. Regulators are proteins that must be made and maintained even while the condition they respond to is absent.Why it is hard. Their material and energy costs come out of the same budget as growth.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 270 open problems on codes, signaling, regulation and adaptation ↗
  5. H-REG-5Encoded responsesThe problem. The link between signal and response must be written in genes (regulator sequences and binding sites) so every daughter cell responds the same way.Why it is hard. Regulation is information: it has to be stored, copied and read like the rest of the genome.Information and controlStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the books: 270 open problems on codes, signaling, regulation and adaptation ↗
Naturalistic proposal

Scope. Present regulatory interdependence does not establish that every named modern signaling pathway existed in the first cells. Each proposed precursor network needs its own mechanism and evidence.

Experimental progress. Genetic and structural analyses connect phosphate transport-associated proteins with PhoR–PhoB signaling in Escherichia coli. This investigates a particular regulatory mechanism, rather than every bacterial phosphate response.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Production can change with physiological demand. Regulatory feedback links activity to later activity through identifiable signals.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Codes, Signaling, Regulation and Adaptation Systems, with its scientific references.

How it was discovered

  1. 1961Jacob and Monod propose the operon: genes switched on and off together by a repressor (Nobel Prize 1965).Jacob & Monod, Journal of Molecular Biology 3:318–356 (1961)
  2. 1966Gilbert and Müller-Hill isolate the lac repressor, the first gene-regulating protein to be purified.Gilbert & Müller-Hill, PNAS 56:1891–1898 (1966)
  3. 1969Cashel and Gallant find the “magic spot” (p)ppGpp, the alarm signal of starving bacteria.Cashel & Gallant, Nature 221:838–841 (1969)
  4. 1986Two-component systems, in which a sensor protein passes a signal to a response protein, are recognised as a family.Nixon, Ronson & Ausubel, PNAS 83:7850–7854 (1986)
  5. 2002Riboswitches are discovered: messenger RNAs that switch genes on or off by binding a small molecule directly.Winkler, Nahvi & Breaker, Nature 419:952–956 (2002)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Conserved, but we cannot say what they do5 key hurdles

Open this part of the cell in the tree

What needs to be explained

How can the origin of a protein be assessed when its role in the cell is not yet known?

What needs explaining. Account for 109 proteins that related organisms keep, including bacteria that have discarded every gene they could, but whose role in the Aquifex scenario cannot yet be named. 35 have no characterised activity at all. Others have a known chemical activity (38 carry an enzyme number) but no known place in a required system.

Hurdles

  1. H-UNK-1Function firstThe problem. Many proteins kept by every free-living relative have no known function.Why it is hard. The origin of a role cannot be assessed until the role is known.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the atlas: 207 proteins of the 1,051 whose function is not known ↗
  2. H-UNK-2Real requirementThe problem. It must be shown which of them are truly needed.Why it is hard. Being kept by related organisms suggests a protein is needed, but does not prove it; no knockout data exist for these organisms.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the atlas: 207 proteins of the 1,051 whose function is not known ↗
  3. H-UNK-3Place in the cellThe problem. Each protein must be assigned to a pathway or system, with the partners it works with.Why it is hard. Without that, the dependency map of the cell is incomplete.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the atlas: 207 proteins of the 1,051 whose function is not known ↗
  4. H-UNK-4Hidden dependenciesThe problem. Unknown roles may hide further dependencies between systems.Why it is hard. Each new dependency adds to what must be present together in the first cell.The integrated cellStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the atlas: 207 proteins of the 1,051 whose function is not known ↗
  5. H-UNK-5Protein familyThe problem. Each protein should be placed in a protein family, to see whether related proteins have known functions.Why it is hard. Many have no recognisable relatives outside closely related organisms, so their function cannot be inferred from family.Information and controlStage 0: Not yet addressed by experimentShown: not yet addressed by experiment.In the atlas: 207 proteins of the 1,051 whose function is not known ↗
Naturalistic proposal

Many conserved proteins of unknown function turn out, once studied, to belong to known protein families or to do housekeeping work such as modifying tRNAs or repairing damaged metabolites.

A naturalistic account treats them like any other protein: as members of families with a shared history, whose roles can be found by experiment and whose origins can then be assessed in the same way as the rest of the cell.

Until their roles are known, the atlas makes no claim about their origin either way.

Design analysis

The design analysis does not use these proteins as evidence. A hallmark can only be assessed once a protein's function, its partners and the effect of losing it are known, which is the test attached to each hallmark.

What they do show is that the inventory of what a free-living cell needs is larger than what can currently be explained in detail. Even the laboratory minimal cell JCVI-syn3.0 kept 149 genes of unknown function.

Each role discovered here will either add a coordinated function to the cumulative case or remove a claimed requirement. Both outcomes are informative.

How it was discovered

  1. 1998The complete genome of Aquifex aeolicus is published, including many predicted proteins with no known function.Deckert et al., Nature 392:353–358 (1998)
  2. 2004Galperin and Koonin argue that conserved proteins of unknown function should be priority targets for experiments.Galperin & Koonin, Nucleic Acids Research 32:5452–5463 (2004)
  3. 2016The synthetic minimal cell JCVI-syn3.0 keeps 473 genes, of which 149 have no known biological function.Hutchison et al., Science 351:aad6253 (2016)
Mineral Ion Metabolism5 key hurdles

In the tree these proteins are spread over other groups.

What needs to be explained

How did selective ion handling arise while preserving useful gradients and avoiding metal toxicity?

What needs explaining. Explain selective acquisition and controlled use of useful ions amid competing ions, with tolerable toxicity and a plausible supply environment.

Hurdles

  1. H-MIN-1Accessible ionsThe problem. Cells need potassium, magnesium, iron, zinc, manganese, nickel, cobalt and molybdenum, each at its own level.Why it is hard. Those ions must be available in the same place as organic synthesis, at usable concentrations.Chemistry before lifeStage 1: Ingredients made under plausible early-Earth conditionsShown: Hydrothermal and mineral settings that supply metal ions next to organic chemistry have been sampled and modelled.In the books: 30 open problems on mineral ion metabolism ↗
  2. H-MIN-2SelectivityThe problem. Each enzyme must bind its own metal, not a competing one.Why it is hard. Without proteins, binding strength follows a fixed chemical order, so the wrong metals would win.Molecular machinesStage 0: Not yet addressed by experimentShown: Without proteins, binding follows the Irving–Williams order, so copper and zinc would crowd out weaker metals. No experiment shows an early system choosing the right metal.In the books: 30 open problems on mineral ion metabolism ↗
  3. H-MIN-3Compatible concentrationsThe problem. Ion levels must suit several processes at once.Why it is hard. The same ion can help one process and harm another; magnesium, for example, is needed for RNA chemistry but breaks fatty-acid membranes.Chemistry before lifeStage 2: One step shown on its ownShown: Magnesium helps RNA copying but breaks fatty-acid vesicles. Adding citrate let both work in one experiment (Adamala & Szostak, Science 2013).In the books: 30 open problems on mineral ion metabolism ↗
  4. H-MIN-4Holding ions in placeThe problem. Useful ions must be kept near the enzymes that use them, without being locked in insoluble minerals.Why it is hard. Mineral surfaces hold ions, but not in a form a cell can carry with it or pass on.Molecular machinesStage 0: Not yet addressed by experimentShown: Mineral surfaces can hold ions near reactions, but not in a form a cell could carry with it or pass on.In the books: 30 open problems on mineral ion metabolism ↗
  5. H-MIN-5Encoded uptake and distributionThe problem. The cell must take up, store and distribute ions with transporters and carrier proteins specified by its genes.Why it is hard. An environment cannot do this for a cell that leaves it; the machinery must be encoded and inherited.Information and controlStage 0: Not yet addressed by experimentShown: No route has been shown from ions buffered by the environment to uptake and distribution encoded in genes.In the books: 30 open problems on mineral ion metabolism ↗
Naturalistic proposal

Scope. There is no single universal concentration range for all cell types. Mineral requirements depend on the selected catalysts, environment, and chemical speciation.

Experimental progress. Structural and functional experiments on MgtE link magnesium-dependent channel gating with magnesium homeostasis, providing a concrete example of regulated ion uptake.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Transport and binding sites distinguish useful ions from competing or toxic alternatives. Ion availability must remain compatible with catalysis and osmotic balance.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Mineral Ion Metabolism, with its scientific references.

How it was discovered

  1. 1808Humphry Davy isolates sodium, potassium, magnesium and calcium by electrolysis, the ions every cell must now manage.Davy, Philosophical Transactions of the Royal Society 98:333–370 (1808)
  2. 1948Irving and Williams find that metals bind to molecules in a fixed order of strength, with copper and zinc above iron and manganese. Cells have to work against this order.Irving & Williams, Nature 162:746–747 (1948)
  3. 1957Jens Skou discovers the sodium–potassium pump, the first known enzyme that moves ions against their gradient (Nobel Prize 1997).Skou, Biochimica et Biophysica Acta 23:394–401 (1957)
  4. 1962Mortenson, Valentine and Carnahan discover ferredoxin, an iron–sulfur protein, in a soil bacterium.Mortenson et al., Biochemical and Biophysical Research Communications 7:448–452 (1962)
  5. 1998MacKinnon's group solves the structure of a potassium channel, showing how it lets potassium through but not the smaller sodium ion (Nobel Prize 2003).Doyle et al., Science 280:69–77 (1998)
  6. 2007The structure of the magnesium channel MgtE shows how magnesium inside the cell closes the channel, a built-in feedback.Hattori et al., Nature 448:1072–1075 (2007)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Quality Control Systems5 key hurdles

In the tree these proteins are spread over other groups.

What needs to be explained

How did the first catalysts persist, avoid harmful aggregation and support turnover before modern quality-control networks?

What needs explaining. Explain how useful folding, defect handling and turnover became coordinated with production without presupposing every modern surveillance pathway.

Hurdles

  1. H-QC-1Lasting catalystsThe problem. Enzymes must stay folded and active long enough for useful work, especially at high temperature.Why it is hard. Stability depends on precise sequence and structure; unstable catalysts are lost before they pay back their cost.Molecular machinesStage 0: Not yet addressed by experimentShown: Catalysts used in the lab are often selected or engineered for stability. No unaided route to long-lasting catalysts has been shown.In the books: 144 open problems on quality control systems ↗
  2. H-QC-2Recognising damageThe problem. The cell must tell damaged or misassembled molecules from good ones.Why it is hard. Recognition needs dedicated machinery: proteases, chaperones and RNA surveillance systems.Information and controlStage 0: Not yet addressed by experimentShown: Telling damaged molecules from good ones needs recognition. No simple system that does this has been shown.In the books: 144 open problems on quality control systems ↗
  3. H-QC-3Folding and aggregationThe problem. Proteins must reach their working shape and avoid clumping; cells use chaperones such as GroEL–GroES and DnaK.Why it is hard. Large proteins often misfold or aggregate without help, and the helpers are themselves large proteins.Molecular machinesStage 2: One step shown on its ownShown: Small proteins refold unaided (Anfinsen), and modern chaperones work in the test tube. No simpler early folding helper has been shown.In the books: 144 open problems on quality control systems ↗
  4. H-QC-4Cost of recoveryThe problem. Unfolding and recycling damaged proteins spends ATP.Why it is hard. Quality control must recover more than it costs.The integrated cellStage 0: Not yet addressed by experimentShown: Modern recycling spends ATP to unfold and break down proteins. No early energy budget for recovery has been tested.In the books: 144 open problems on quality control systems ↗
  5. H-QC-5Error handling at scaleThe problem. Longer proteins and genomes need correspondingly more accurate error handling: proofreading, editing by synthetases, ribosome rescue and proteases.Why it is hard. The error-handling machines are themselves long proteins made by the system they protect.The integrated cellStage 0: Not yet addressed by experimentShown: No experiment follows error handling becoming better as polymers grew longer.In the books: 144 open problems on quality control systems ↗
Naturalistic proposal

Scope. Present quality-control systems are diverse and partly redundant. Their modern sophistication should not be treated as an experimentally established starting condition for life’s origin.

Experimental progress. The structure of a GroEL–GroES complex explains how a chaperonin can create a protected folding environment through large conformational changes.

What would strengthen a natural account. Steps shown to work together under plausible early-Earth conditions, without a chemist choosing the reagents, purifying the products or changing conditions between steps, would strengthen a natural account of origins in proportion to what they explain. Loss of a claimed dependency or discovery of functional alternatives would weaken the corresponding design argument. Gaps in the record do not supply a numerical likelihood for either explanation.

Design analysis

Observations. Chaperones assist selected folding reactions and help limit aggregation. Maturation, rescue and degradation handle distinct defective products.

Hallmarks that apply.

The author’s conclusion. The coordinated features give the author a positive reason to consider design within the wider cellular architecture. For this local system alone, they do not establish that design is uniquely the best historical explanation. The conservative inference is that the system contributes to a cumulative design case, with its weight conditional on the comparative evidence and on how much plausible natural pathways explain.

From the book article Quality Control Systems, with its scientific references.

How it was discovered

  1. 1962Ferruccio Ritossa sees new gene activity in fruit-fly chromosomes after a heat shock, the first sign of the heat-shock proteins.Ritossa, Experientia 18:571–573 (1962)
  2. 1973Christian Anfinsen sums up his finding that a small protein refolds on its own: the sequence holds the information for the fold (Nobel Prize 1972).Anfinsen, Science 181:223–230 (1973)
  3. 1978Laskey and colleagues call nucleoplasmin a 'molecular chaperone': a protein that helps others assemble without becoming part of them.Laskey et al., Nature 275:416–420 (1978)
  4. 1989Goloubinoff, Gatenby and Lorimer show that the bacterial GroEL and GroES proteins refold an unfolded enzyme in the test tube, using ATP.Goloubinoff et al., Nature 342:884–889 (1989)
  5. 1996Keiler, Waller and Sauer find that tmRNA rescues stalled ribosomes and tags the unfinished protein for destruction.Keiler et al., Science 271:990–993 (1996)
  6. 1997Xu, Horwich and Sigler solve the structure of the GroEL–GroES complex, showing the closed chamber in which proteins fold.Xu et al., Nature 388:741–750 (1997)

Each hurdle above is sorted by kind and by how far experiments have got (see “How hurdles are sorted” at the top).

Discovery histories for each pathway and protein are added in phase 3.