GCリッチな準種からの創発

ハイパーサイクルとは、互いの複製を触媒し合う閉じた一群の複製子である。このサイクルは複数の異なる配列間の 相互的な 触媒作用によって定義されるため、単一の鋳型コピーから核形成することはできない。1つの分子はサイクル全体の機能を備えておらず、その最初のコピーにはループを閉じるための触媒パートナーが存在しない。したがって、ハイパーサイクル的組織は突然出現することはできず、集団のレベルで徐々に組み上げられていく必要がある。その組み上げの最も有望な揺りかごとなるのが、GCリッチな準種である。

Why one template cannot start a cycle

An autocatalytic replicator is self-sufficient: a single strand that templates its own copying grows exponentially from one molecule, because the molecule and its product carry the same function. A hypercycle is not of this kind. Its defining property is a closed loop of catalytic support in which member i enhances the replication of member j, which enhances k, and so on until the loop returns to i. Strip out any single member and the remaining members no longer regenerate it. Consequently, the first copy of any one member is a dead end: it has no complementary partners, and no closed catalytic cycle exists to amplify it preferentially.

A GC-rich quasi-species supplies the raw material

A quasi-species is not a single sequence but a mutationally connected cloud of related sequences distributed around a dominant, master-like sequence. In a GC-rich pool this cloud is unusually well stocked. Every additional G–C pair contributes a third hydrogen bond, so GC-rich stems melt at higher temperature, fold more reliably and tolerate more point mutations before their secondary structure collapses. The reservoir therefore combines two things the nucleation problem requires at once: a high local diversity of variants with altered loops, stems and surface geometry, and enough structural stability that individual variants remain functional long enough to be tested by selection.

Mutation explores coupling, selection keeps it

Mutational exploration is what actually searches the space of interactions. A point substitution can reshape a hairpin, expose a new binding surface, or recruit a metal-ion-dependent catalytic centre, changing the weak coupling coefficient that describes how strongly member i assists the replication or catalysis of member j. Most such variants are neutral or harmful. Occasionally one weakly assists a neighbouring template, and the effect is amplified because it is heritable: the variant is copied and keeps assisting. Selection during this phase no longer acts only on a template's own replication rate. It acts on coupling: a variant is enriched when the net effect of its interactions increases the production of other templates in the pool. Reciprocal pairs are favoured twice over, because each partner now raises the other's yield.

From quasi-species organisation to hypercyclic organisation

As links accumulate, the mutant distribution changes shape. Diversity is no longer spread around a single master sequence; it condenses toward a small subset of mutually catalytic templates, and ring closure occurs only once enough links exist for the network to be closed rather than merely open or branched. Before closure the ensemble behaves as a quasi-species with some correlated variants. After closure, growth is cooperative and super-additive: the cycle multiplies as a unit, and because each member feeds the next, the coupled set can outgrow an isolated quasi-species competing for the same monomers and replicase. This is how a hypercycle could appear without any preformed multi-gene genome — by recruitment and integration of mutant genes already present in the GC-rich reservoir. The underlying fixed-point analysis of the two-membered case makes the same point in rate-coefficient form: whether the pair ends in competitive exclusion, selection of one member, or stable hypercyclic coexistence depends only on the relative magnitudes of self-enhancement and mutual enhancement, and the criterion generalises to three-membered and n-membered cycles.

When the route fails

Replication fidelity must stay above the error threshold. If the mutation rate is too high, the reservoir dissolves into a diffuse cloud in which no variant persists long enough to establish a link.
Functional compatibility matters: members must not interfere destructively with one another's folding or catalysis, otherwise the accumulating links cancel instead of reinforcing.
Parasite load must stay low. A selfish template that is efficiently replicated thanks to the emerging cycle, but catalyses nothing back, drains the shared resources and can destabilise the network before it closes.

Worked sketch. In a GC-rich RNA pool, a dominant GGC/GCC-like sequence and its mutants form a quasi-species. A mutant may weakly assist replication of a neighbouring sequence, and selection preserves both partners because their joint yield exceeds what either achieves alone. Further mutations generate reciprocal assistance, the branch that was missing is filled in, and a two-membered or larger cycle closes. Once closed, hypercyclic growth outcompetes the isolated quasi-species that gave rise to it.

GCリッチな準種からの創発

Three stages read left to right. First, a GC-rich master sequence sits at the centre of a mutational cloud, with an inner selection zone marking the variants that are progressively enriched while stable GC base pairs protect their folded structures. Second, the surviving variants are joined by weak, one-directional catalytic links, and one connection is still dotted because the network is not yet closed. Third, those links become reciprocal and the network closes into a hypercyclic ring in which each member catalyses the next, so the set as a whole can outgrow an isolated quasi-species.

QUESTION 1

Why can a hypercycle not nucleate from a single template copy?

Because a single RNA sequence encodes only its own replication, not the complementary catalytic partners needed to close a loop.

Because GC-rich sequences melt at temperatures too low to allow copying.

Because mutation rates inside a single molecule are zero.

Because isolated RNA molecules are chemically unable to serve as templates at all.

QUESTION 2

What does a GC-rich quasi-species contribute to the gradual emergence of a hypercycle?

A mechanism that removes parasites from the pool immediately.

A ready-made set of protein enzymes that can already catalyse each other.

A guarantee that every mutant produced is catalytically active.

A reservoir of related, mutationally connected variants plus stable base pairing that protects their folded structures.

QUESTION 3

During the transition from quasi-species to hypercycle, what does selection act on?

Only on the replication rate of the master sequence in isolation.

Only on the physical length of the RNA molecules.

Only on the average GC content of the whole pool.

On coupling: variants that raise the replication or catalysis of neighbouring templates are enriched, not only the fastest self-replicator.

QUESTION 4

Which sequence of stages correctly describes the emergence of a hypercycle from a GC-rich pool?

Weak coupling → hypercyclic closure → GC-rich reservoir → reciprocal coupling.

Reciprocal coupling → GC-rich reservoir → hypercyclic closure → weak coupling.

Hypercyclic closure → GC-rich reservoir → weak coupling → reciprocal coupling.

GC-rich reservoir → weak coupling → reciprocal coupling → hypercyclic closure.

QUESTION 5

Which condition most directly destabilises an emerging hypercycle before its ring closes?

A generous supply of nucleotide precursors.

A high load of parasitic templates that are replicated by the emerging cycle but catalyse nothing back.

A high replication fidelity that preserves functional variants.

Strong GC base pairing in the reservoir.

Case Study: Nucleating a Two-Membered Hypercycle In Vitro

Read the scenario and answer the questions.

You propagate a GC-rich RNA population in a continuously stirred flow reactor supplied with nucleotide precursors and Qβ replicase. Deep sequencing after 30 transfers reveals one dominant GCC/GGC-like sequence surrounded by a cloud of single- and double-mutant variants, the most abundant differing from the master at three to eight per cent of positions. Serial transfer continues for a further 120 passages at a lower dilution rate. At the end, two variants, X and Y, each decline sharply in abundance when the other is removed from the reactor, and in mixed culture the total RNA yield of the X plus Y pair exceeds the sum of the two monocultures measured under identical conditions. A third variant, Z, appears in parallel cultures.

1. Identify which features of the experiment correspond to (i) the GC-rich quasi-species stage and (ii) the beginning of hypercyclic organisation, and explain why neither X nor Y alone could have nucleated a hypercycle.

Answer:
Model solution. (i) The quasi-species stage is documented by the deep-sequencing snapshot at 30 transfers: a single dominant GC-rich master sequence surrounded by a mutationally connected cloud of one- and two-point variants. This cloud is the reservoir — it supplies many candidate folds and binding surfaces while GC base pairs keep those structures stable. (ii) The hypercyclic stage is signalled by the later observations: X and Y each collapse when the other is removed, and their joint yield is greater than the sum of their separate yields. Mutual dependence plus super-additive growth is the signature of coupling, not of independent replication. Neither X nor Y alone could nucleate the cycle because the loop requires reciprocal catalysis: a single variant supplies only half (or less) of the closed cycle, so on its own it has no partner to regenerate it and is selected simply as an ordinary replicator. The cycle is assembled by recruitment of both variants from the reservoir, not started by either one in isolation.

2. Variant Z is replicated efficiently in the presence of X and Y but contributes no catalysis of its own. Predict the consequence for the population over the next 200 transfers and name two quantities the experimenter could measure to test the prediction.

Answer:
Model solution. Z is a parasite: it exploits the shared products of the X–Y cycle without returning any catalytic benefit, so it gains a direct replication advantage inside the reactor. Expect Z to rise in frequency while X and Y decline, with total RNA yield initially unchanged or even increased and then falling as the catalytic partners are diluted out. If the parasite load becomes too high, the two-membered cycle is destabilised before or shortly after closure. Measurable quantities: (1) the frequency trajectories of X, Y and Z obtained by deep sequencing across passages, which should show Z rising and X and Y falling; (2) the joint yield of X and Y (for example total RNA per unit replicase, or the degree of mutual dependence when each is removed in turn), which should weaken as Z takes over. Comparing the mixed X plus Y plus Z culture against the X plus Y control quantifies the parasitic burden directly.