Abiogenesis / personal lessonsLesson 4 of 6
Lesson browserAbiogenesis lessonsPrevious: Energy and chemical reactionsNext: Compartments and protocells
Abiogenesis · Lesson 4

Copying molecules and the RNA world

RNA can carry sequence information, and some RNA molecules catalyze reactions. The RNA-world hypothesis proposes an early stage in which RNA played central roles before the modern division of labor among DNA, RNA, and proteins.

This is a hypothesis about an early stage, not a claim that the first organic molecule was RNA or that RNA alone solves every origin problem. Use the distinction between catalysts and energy sources from lesson 3.

One kind of polymer can do two jobs

RNA is a chain of nucleotides. Each contains a sugar, a phosphate group, and one of the bases A, C, G, or U. The order of bases is its sequence. That order can affect how the strand folds and which molecules it binds.

Complementary pairing usually places A opposite U and C opposite G. This gives one sequence a way to guide another. The same pairing interactions, together with other contacts, let a single RNA fold into a three-dimensional structure. Some folded RNAs catalyze reactions; these are ribozymes.

Modern biology provides evidence that RNA can perform catalysis. For example, RNA forms the catalytic center responsible for peptide-bond formation in the ribosome, within a large RNA–protein assembly. This supports the chemical capabilities required by RNA-world ideas. It does not directly record an ancient RNA-only organism.

Alternative proposals involve earlier genetic polymers, mixtures of RNA and peptides, or chemical networks before sequence-based inheritance. They still face the need to explain how variation was retained and how it affected continued reproduction.

Base pairing is only part of copying

A template is a strand whose sequence guides the order of a new strand. Complementary building blocks can align with it, but they must also form covalent bonds in a new backbone. Suitable activation chemistry and conditions are required.

A template guides a complementary strand

Template, 3′ to 5′
ACGUAC
Complement grows 5′ to 3′
······

0 of 6 bases added. The complete complementary sequence is UGCAUG.

Step 0 / 6
The six-base sequence is invented. One step adds one complementary base in this teaching model. Real copying needs suitable activated building blocks and reaction conditions; pairing alone does not make a covalent backbone. The animation omits failures, strand separation, and copying the complement back into the original sequence.

The two strands have opposite directions, called antiparallel orientation. The labels 5′ and 3′ refer to positions on the nucleotide sugar. A new strand grows in the 5′-to-3′ direction. In the diagram, the template is written in the opposite direction so you can read aligned base pairs directly.

The first product is a complement, not an identical sequence. Copying that complement can restore the original sequence. A full replication cycle must therefore make usable strands, separate paired strands, and use the products as templates again.

Separation is a real constraint. Paired strands can bind tightly, preventing another copying round. Conditions that separate them can also damage RNA, disrupt membranes, or interfere with the next reaction. A diagram that ends when one complementary strand forms has not demonstrated repeated replication.

Follow the sequence and the chemical operations separately

Sequence information means an ordered distinction among building blocks. A strand containing the same numbers of A, C, G, and U can still differ from another strand if their order differs. The order affects potential pairings and folding, so counting total RNA alone does not establish that a particular sequence has been copied. A copying claim needs evidence connecting the product's ordered bases to those of its template.

In the animation, the template reads ACGUAC from its labeled three-prime end to its five-prime end. The growing complementary strand reads UGCAUG in the opposite direction, from five-prime to three-prime. Match the first A with U, the next C with G, and continue across all six positions. The letters obey the stated pairing rule. The direction labels tell you how the two chains align; they are not additional bases.

Use Next step to reveal one product position at a time. Before revealing it, predict its letter from the aligned template position. After six steps, compare every pair rather than merely checking that the new strand has six bases. Back hides the latest displayed addition and Reset returns to the initial illustration. These controls inspect a predetermined successful sequence; they do not sample real errors or simulate chemical reversibility.

Alignment is only one operation. Bases can associate through noncovalent interactions while still belonging to separate building blocks. A persistent new chain also needs covalent connections in its backbone. The copying chemistry must make those connections using appropriate reactants. The animation combines alignment and successful incorporation into one teaching step, so it cannot establish how fast incorporation occurs or how much material fails to become a complete strand.

The first product is a complement. To obtain another strand with the original sequence, a further copying round can use that complement as a template. The strands must be available for the required interactions, and paired products can obstruct access. Simply separating them once is not necessarily enough: the original partners may rejoin before new copying finishes. A repeated cycle must address competition between rejoining, copying, and degradation.

Orientation can also create an apparent mismatch in written sequences. A strand shown from three-prime to five-prime is written in the reverse order if you choose to list it from five-prime to three-prime instead. Reversing the written order does not change its chemical identity. Complementing changes which base is specified at each aligned position. Check both direction and pairing before treating different letter strings as evidence of a copying error.

A folded ribozyme introduces a related accessibility problem. Its sequence may need a particular structure to catalyze a reaction, while copying requires access to that sequence as a template. Conditions favoring one operation can obstruct another. This does not rule out RNA-based systems, but it explains why demonstrating catalysis and demonstrating copying separately leave an integration problem. The experiment must show how the relevant forms and conditions participate in a repeatable process.

Longer sequences are harder to copy exactly

Let q be the probability of copying one base correctly and L the strand length. If errors are independent and every position has the same correctness probability, the probability of an exact copy is qL. This is ordinary repeated multiplication, not a reconstruction of an early replicator.

Small per-base error rates accumulate

20 bases

81.8% exact copies

100 bases

36.6% exact copies

Exact-copy probability is q raised to L, where q is per-base correctness and L is length. The model assumes independent errors with the same probability at every position. It does not model reaction yield, strand separation, functional tolerance of mutations, or an evolutionary error threshold.

Exact sequence identity is not the same as useful function. Some changes leave activity intact, some destroy it, and some improve it in a particular environment. An evolutionary model also needs relative reproduction rates, population sizes, resources, and the effects of mutations. The exact-copy calculation alone does not establish a universal maximum genome length.

Variation must affect descendants

Suppose two RNA variants differ in their ability to help copying. If the more effective variant leaves more descendants and its sequence is inherited, the population can change by natural selection. If all molecules make random new sequences unrelated to their templates, better performance is not reliably passed on.

A catalyst that helps every nearby template equally raises another problem. A shorter or nonfunctional template might copy faster while benefiting from the useful catalyst. Such parasitic templates can undermine a shared copying system. Compartments can sometimes limit this sharing, but introduce their own costs.

Derive the exact-copy probability before interpreting it

Begin with a two-position example rather than a long strand. Suppose each position is correct with probability ninety percent, independently of the other. Both positions are correct with probability 0.9 times 0.9, or 0.81. Exactly one error has probability 0.9 times 0.1 plus 0.1 times 0.9, or 0.18. Two errors have probability 0.1 times 0.1, or 0.01. These three mutually exclusive outcomes sum to one.

For an exact copy of a longer strand, every position must be correct. Under the same independence and equal-correctness assumptions, multiply the per-position probability once for each position. That repeated product is q raised to L. Here q is a probability between zero and one, not the percentage number ninety-nine. For ninety-nine percent correctness, use 0.99 rather than 99 in the multiplication.

Select the one-error-per-hundred-bases preset. The twenty-base panel gives about 81.8 percent exact copies, while the hundred-base panel gives about 36.6 percent. One error on average per hundred positions does not mean that every hundred-base product has exactly one error. Some have none, some have one, and some have more. The average number of errors and the probability of having zero errors are different summaries of a distribution.

Then select Higher fidelity. At 99.9 percent per position, the hundred-base exact-copy fraction is about 90.5 percent. Small improvements per position can matter across many positions because each multiplication compounds the effect. At the Frequent errors preset, the hundred-base fraction is below 0.1 percent, not zero. The interface uses a threshold label so a small positive probability is not mistaken for impossibility.

The model is conditional on its simplified copying assumptions. It does not calculate the probability that copying starts, reaches full length, separates from its template, or produces a functional catalyst. For a separate invented calculation, suppose half of attempted products reach full length and 36.6 percent of those completed products are exact. Then about 18.3 percent of all attempts produce exact full-length copies. The completion fraction is an additional assumption, not an output of the slider.

Exactness is also different from function. A changed base might preserve a useful fold, disrupt it, or alter its activity under particular conditions. You cannot infer the proportion of functional products from q raised to L without information about which changes matter. Nor does an exact copy necessarily function if its environment prevents folding or destroys it before it acts. Sequence measurements and activity measurements answer different questions.

Real errors need not be independent or equally likely at every position. Neighboring sequence, particular base pairs, and reaction conditions can change error behavior. The calculation isolates one consequence of accumulated copying risk; it does not establish a universal genome-size limit or a complete evolutionary error threshold. To make an evolutionary prediction, you would also need how variants reproduce, survive, compete for resources, and pass their differences to descendants.

Successful RNA experiments still have defined inputs

Horning and Joyce reported a laboratory-evolved RNA polymerase ribozyme in 2016. It synthesized several functional RNAs and amplified short templates under supplied conditions. This demonstrates substantial RNA-catalyzed copying capability without protein enzymes in those reactions.

The ribozyme was produced through laboratory selection, and the experiment supplied substrates and controlled reaction conditions. It was not a general, autonomous system that copied its own complete machinery indefinitely from simple environmental feedstocks. A historical explanation must account for the emergence of useful sequences as well as the inputs to their reactions.

When evaluating a copying result, ask what was copied, how long it was, how many cycles occurred, how errors were measured, and what the researcher supplied at each stage. These questions distinguish a demonstrated reaction from a complete evolutionary system.

Ask what repeats after the first successful copying round

A report of more RNA can describe several different outcomes. It might mean longer chains assembled from supplied building blocks, more copies of a particular template, or a mixture of many newly formed sequences. These are not interchangeable. To establish template copying, identify the products and their relationship to the template. To establish continuing reproduction, follow whether those products can participate in further rounds without the researcher replacing the essential copying machinery each time.

The Horning and Joyce study tested an RNA polymerase ribozyme improved through laboratory selection. Its ability to make functional RNA products and amplify short templates supports specific capabilities of RNA catalysis. It does not show that an arbitrary RNA molecule can copy itself, or that the improved ribozyme arose spontaneously from an unselected mixture of simple feedstocks. How a capable sequence emerges is distinct from what an already supplied sequence can accomplish.

Keep the copying catalyst and the copied template separate when interpreting an experiment. A polymerase can act on another sequence without copying its own entire sequence. Amplifying a short template therefore does not automatically renew a longer polymerase. A complete RNA-based reproductive system needs a way to reproduce or replenish every essential inherited component, not just the easiest target to copy in the experiment.

Selection requires more than unequal final counts. Suppose a useful RNA catalyst helps copying throughout a shared mixture. A shorter template that lacks the catalytic function might still benefit and copy quickly. Its increase could undermine production of the catalyst on which the system depends. This is the parasite problem discussed above. The relevant advantage must remain associated with descendants that retain the function, rather than being available equally to all competing templates.

Compartments can change that association by keeping some catalysts close to their templates and descendants. They also add requirements for transport and inheritance. A membrane that excludes necessary building blocks or loses essential strands during division can defeat the intended benefit. This is why the next lesson treats compartments as part of a coupled system rather than a universal solution that can simply be placed around a successful test-tube reaction.

A stronger repeated-cycle test records what enters and leaves each round. Track the target sequence, copying catalyst, full-length yield, error spectrum, and retained activity. Report fresh activated building blocks, externally supplied catalysts, and imposed condition changes. Feeding an open system is not inherently illegitimate, but the identity of the feed distinguishes replenishing consumable resources from rebuilding the system whose reproduction is being claimed.

Finally, separate the historical hypothesis from the laboratory capability. Modern catalytic RNA and copying experiments make RNA-world mechanisms scientifically testable. They do not uniquely reconstruct the earliest genetic material or prove an RNA-only stage at a particular ancient site. The useful conclusion identifies the demonstrated capability and the next missing connection, such as strand reuse, catalyst renewal, or inheritance through division. That directs a further experiment instead of treating a promising component as a complete origin pathway.

Exercises

Write the complement and the next copy

A template is written 3′ ACGU 5′. Write its aligned complement, then the aligned complement of that product.

Hint

Pair A with U and C with G. Keep the strand directions opposite.

Worked answer

Copying twice restores the sequence

Original: 3′ A C G U 5′

First complement: 5′ U G C A 3′

Copy of original: 3′ A C G U 5′

The strands shown opposite one another are antiparallel. This diagram omits the energy and chemistry needed to build and separate them.

The aligned complement is 5′ UGCA 3′. A complement of that strand is 3′ ACGU 5′, restoring the original sequence. The first copy is not identical to the original. Two rounds also require backbone formation and strand separation; letters alone omit those chemical steps.

Compare exact-copy probabilities

At 99% correctness per base, compare 20-base and 100-base strands. Assume independent errors. Does an inexact sequence necessarily lose all function?

Hint

Use 0.99 raised to each length, not 99% minus the number of bases.

Worked answer

Compare complete copies, not individual bases

20 bases

81.8% exact

100 bases

36.6% exact

At 99% correctness per base, independent errors make the exact-copy probability 0.99 raised to the strand length.

For 20 bases, 0.99²⁰ is about 0.818, or 81.8%. For 100 bases, 0.99¹⁰⁰ is about 0.366, or 36.6%. Longer sequences are less likely to be copied exactly under these assumptions. Some imperfect copies may still function, so these values are not the proportions of viable replicators.

Distinguish copying from evolution

A reaction produces more RNA after every round of feeding. What additional observations would support Darwinian evolution rather than repeated chemical production alone?

Hint

Look for inherited variation and a link between that variation and reproduction.

Worked answer

The copied variant must affect later reproduction

Variant A

AAAAAAAA

4 starting copies; 8 after copying

Variant B

BBBB

4 starting copies; 4 after copying

Invented counts. Both variants start with four copies. After a copying round there are eight A copies and four B copies. This would support selection only if the difference is inherited and causes different reproductive success under the test conditions.

You need variants whose differences pass to descendants and cause different reproductive success. Merely making more polymer does not establish those conditions. Track sequences and reproduction across rounds, and test whether changing the proposed useful sequence changes its performance. A single unequal count can also reflect chance or biased inputs.

Sources and limits

These papers support the experiments and hypotheses discussed here. The diagrams use teaching models, not data extracted from these papers. The text above explains the concepts needed for the exercises; opening the papers is optional.

Lesson browserAbiogenesis lessonsPrevious: Energy and chemical reactionsNext: Compartments and protocells