# Chapter 55. RNA Helicases and RNP Remodelers: Families, ATPase Cycles, Mechanochemical Logic, and Applications

## Scope Note

This chapter provides comparative, mechanism-centered ownership of enzymes commonly called RNA helicases and ribonucleoprotein (RNP) remodelers. It explains how conserved superfamily 1 and superfamily 2 nucleoside-triphosphatase cores bind RNA, cycle through nucleotide states, separate strands, translocate, displace proteins, or stabilize selected RNP conformations. It treats DEAD-box, DEAH/RHA, Ski2-like, Upf1-like superfamily 1, viral, and bacterial boundary cases together so that shared chemistry is not mistaken for identical mechanical behavior. Pathway chapters retain ownership of spliceosome progression, translation initiation, ribosome biogenesis, decay, innate immunity, and viral replication; this chapter owns family comparison, ATPase-cycle logic, kinetics, regulation, measurement, disease mechanisms, inhibition, and engineering.

## Executive Summary

An RNA helicase is an enzyme whose conserved nucleoside-triphosphatase core binds nucleic acid and couples nucleotide-dependent conformational changes to RNA rearrangement. The historical name emphasizes duplex unwinding, but many family members act primarily as RNP remodelers: they evict or reposition proteins, expose RNA to an enzyme, promote a conformational transition, or clamp an RNA-protein state. A protein can therefore be a bona fide helicase-family ATPase without behaving as a processive molecular motor on every substrate. The mechanistic claim must name the substrate, directionality, step size or local opening event, protein partners, nucleotide state, and observed product.

Most RNA helicases belong to superfamily 1 (SF1) or superfamily 2 (SF2). Their catalytic engine contains two RecA-like domains carrying conserved motifs that coordinate ATP, magnesium, RNA, and interdomain communication. ATP and RNA binding favor closure of the cleft between the domains; hydrolysis and product release redistribute conformations and affinities. This shared fold does not impose one shared mechanism. DEAD-box proteins usually bind directly to duplex regions, bend one strand, and create short local separation without requiring a long single-stranded loading tail. DEAH/RHA and many Ski2-like proteins generally bind a single-stranded segment and translocate directionally, allowing repeated steps to disrupt downstream RNA or RNP structure. SF1 Upf1-like motors use additional regulatory domains to tune processivity and RNA residence. Viral helicases can integrate the same core logic into replication proteins with protease, polymerase, membrane, or immune-antagonist functions.

ATP is not simply fuel poured into a miniature winch. Each nucleotide state changes an ensemble of enzyme-RNA conformations. Productive work depends on how these states are gated by RNA geometry, accessory factors, competing ligands, and pathway timing. Coupling efficiency can be low when ATP hydrolysis occurs without the scored remodeling event, and that apparent inefficiency may be physiologically useful: futile cycles can create repeated local sampling, while regulated ATPase checkpoints can delay progression until an RNP substrate is correctly assembled. The same ATPase can promote productive assembly on one substrate and discard a defective intermediate on another.

Mechanistic evidence requires matched measurements. ATP consumption alone does not demonstrate unwinding; endpoint strand separation alone does not demonstrate an ATP-driven motor; loss of an RNP protein does not identify direct displacement. Robust inference combines nucleotide binding and hydrolysis, RNA binding, direct product detection, structural state assignment, kinetic perturbation, and biological rescue. Bulk assays average asynchronous molecules and can conceal pauses or inactive fractions. Single-molecule experiments reveal stepwise motion and state heterogeneity, but applied force, tether geometry, fluorophore placement, and selected observation windows can create their own distortions. Structural snapshots constrain feasible mechanisms but do not by themselves establish the time order or flux through states.

Helicase dysfunction can alter neurodevelopment, cancer, infection, ribosome assembly, splicing, translation, and RNA surveillance. Yet disease association is not equivalent to loss of unwinding: variants can change ATPase kinetics, cofactor binding, localization, phase behavior, or substrate choice. Small molecules likewise use several mechanisms. Some occupy a nucleotide or allosteric pocket; rocaglates stabilize eIF4A on selected polypurine RNA sequences and thereby convert a general initiation factor into a sequence-selective clamp. Such compounds illustrate why “helicase inhibitor” should be replaced by a state-specific biochemical description whenever possible.

## Concept Inventory

- **RNA helicase:** an SF1 or SF2 nucleoside-triphosphatase that uses nucleotide-dependent changes in nucleic-acid binding and conformation to rearrange RNA or an RNA-containing complex.
- **RNP remodeler:** a factor that changes the composition, structure, accessibility, or kinetic state of a ribonucleoprotein complex; direct duplex unwinding is not required.
- **RecA-like domain:** one of two homologous folds that form the catalytic core of most SF1 and SF2 helicases.
- **ATPase cycle:** ordered or branched transitions among apo, ATP-bound, hydrolysis-transition, ADP-phosphate, and ADP-bound states, including RNA binding and release.
- **Mechanochemical coupling:** mapping of chemical transitions in nucleotide turnover onto RNA binding, translocation, strand separation, or RNP rearrangement.
- **Local strand separation:** destabilization of a short RNA duplex segment without sustained directional tracking along RNA.
- **Translocation:** directionally biased movement of an enzyme relative to a nucleic-acid strand.
- **Processivity:** average extent of repeated action before dissociation, defined relative to a stated substrate and product.
- **Polarity:** direction of motor movement on the tracking strand, conventionally described 3′→5′ or 5′→3′.
- **Loading strand:** single-stranded nucleic-acid segment initially occupied by a directional helicase before it encounters a barrier.
- **DEAD-box protein:** SF2 family named for motif II sequence Asp-Glu-Ala-Asp; members commonly promote local opening and ATP-dependent clamping.
- **DEAH/RHA protein:** SF2 family containing a related motif II and accessory domains; many members are directional RNA translocases.
- **Ski2-like helicase:** SF2 family with a characteristic helicase core and accessory architecture, represented by Ski2, Mtr4, and spliceosomal Brr2.
- **SF1 helicase:** superfamily that includes Upf1-like and many viral or DNA/RNA motors; family identity does not alone specify substrate.
- **Kinetic proofreading:** ATP-dependent temporal or energetic discrimination that increases rejection of suboptimal substrates at a cost in energy and yield.
- **Coupling efficiency:** relationship between nucleotide hydrolysis and scored mechanical output, such as base pairs separated or translocation steps completed.
- **Functional processivity:** persistence of pathway action, which may arise from cofactors or confinement even when the isolated core is weakly processive.
- **Molecular clamp:** stable nucleotide-dependent RNA-bound state that can exclude proteins or hold an RNP conformation without continuous translocation.
- **RNPase activity:** operational ability to remove or reorganize proteins on RNA; it does not by itself reveal direct protein displacement.

## What to Know Before Reading This Chapter

RNA duplexes are not uniform rods. Their stability depends on sequence, length, mismatches, bulges, ends, ions, temperature, and tertiary contacts. An RNP substrate adds proteins that can protect, bend, bridge, or expose RNA. Consequently, “unwinding activity” is conditional on substrate architecture. A 10-base-pair model duplex with a long 3′ tail, a buried helix in a spliceosome, a translating messenger RNP, and a viral replication intermediate are mechanically different objects even if each contains paired RNA. Chapters [2](chapter1002.md) and [3](chapter1003.md) provide the chemical and folding foundation.

ATP hydrolysis releases free energy, but free-energy availability does not determine mechanism. The enzyme must bind particular RNA and nucleotide states, bias transitions, and prevent rapid reversal or unproductive escape. A measured ATPase rate is the net flux through a chemical cycle under the assay conditions. An unwinding rate is a population-level or single-molecule measure of a structural output. The ratio between them can be informative only when active concentrations, stoichiometry, parallel reactions, and product trapping are controlled. Quantitative regimes and artifact controls are developed in [124](chapter1156.md).

The term **motor** is used here for enzymes that generate directionally biased motion or repeated mechanical steps. The term **remodeler** is broader and includes local separation, protein displacement, and nucleotide-dependent clamping. “Helicase” remains the accepted family name, but the chapter avoids inferring a motor mechanism from annotation alone.

## 55.1. Helicase superfamilies, RNA motors, and the boundary between unwinding and RNP remodeling

### Classification begins with conserved cores, not the substrate label

Helicases are classified primarily by sequence motifs, core architecture, and evolutionary relationships. Most RNA helicases fall within SF1 or SF2, whose two RecA-like domains form an ATP- and RNA-responsive cleft. Within those superfamilies, motif order and family-specific insertions distinguish DEAD-box, DEAH/RHA, Ski2-like, Upf1-like, viral NS3-like, and other groups. This classification is more reliable than the label “RNA helicase” in a database entry, because several families contain proteins that act on RNA, DNA, RNA-DNA hybrids, or more than one polymer depending on cofactors and cellular context.

The core motifs supply a common biochemical vocabulary. Motif I, also called the Walker A or P-loop motif, helps bind phosphate groups of ATP. Motif II contributes acidic residues to magnesium-dependent hydrolysis and gives DEAD and DEAH families their names. Motif VI participates in nucleotide sensing and interdomain communication. Additional motifs contact the sugar-phosphate backbone or couple the two RecA-like domains. The motifs should not be read as independent functional badges: a substitution can perturb folding, ATP binding, hydrolysis, RNA affinity, or communication among these steps.

Figure 55.1 maps the families onto their dominant mechanical modes while preserving overlap. Table 55.1 separates classification evidence from mechanistic evidence.

![Figure 55.1. One conserved engine supports several RNA-remodeling modes](../assets/figures/chapter1159_figure1.png)

**Figure 55.1. One conserved engine supports several RNA-remodeling modes.** “SF1 and SF2 cores conserve ATP- and nucleic-acid-coupled chemistry, but family-specific domains and partners produce different mechanical outputs. Family identity predicts a tendency, not the demonstrated product on a particular RNP.”

**Table 55.1. Family assignment and demonstrated mechanism are different claims.** Separate evidence for lineage from evidence for mechanical output.

| Claim | Minimum evidence | Stronger evidence | Common overreach |
| --- | --- | --- | --- |
| **SF1 or SF2 family** | Conserved motif order and core homology | Structure plus phylogenetic placement | Family predicts physiological substrate |
| **DEAD-box membership** | Motif II and family-wide sequence architecture | Closed-core RNA/nucleotide structure | All members have identical cofactors |
| **Directional translocation** | Polarity-dependent movement on a tracking strand | Steps or run lengths under several loads | ATPase activity proves motion |
| **Processive unwinding** | Length-dependent repeated strand separation per encounter | Single-molecule runs and product trapping | Endpoint amplitude equals processivity |
| **Direct RNP remodeling** | Defined RNP converted to defined product | Purified reconstitution and partner release | Cellular loss of crosslinking proves displacement |
| **Molecular clamping** | Long-lived nucleotide-dependent RNA state | State-specific kinetics and structural contact | Increased occupancy means increased flux |

### Unwinding is one possible remodeling outcome

Duplex unwinding is operationally defined as conversion of paired strands into separated products. A directional helicase can load on a single-stranded tail, translocate to the duplex junction, and repeatedly disrupt base pairs. A DEAD-box protein often reaches the same endpoint by binding directly over a short duplex and bending one RNA strand so sharply that local pairing becomes unfavorable. These paths differ in tail dependence, directionality, duplex-length response, ATP use, and sensitivity to obstacles.

RNP remodeling expands the product definition. An enzyme may displace a protein by competing for the RNA, by moving along RNA into the protein, by destabilizing the underlying RNA structure, or by inducing a conformation that lowers protein affinity. A helicase can also promote assembly: ATP-dependent clamping can stabilize an intermediate long enough for a partner to bind. The biological product may be an exposed splice site, a ribosomal assembly intermediate, a messenger RNA committed to export, or a transcript delivered to an exoribonuclease. None is adequately described as simply “unwound.”

Direct protein displacement requires particular evidence. If an ATPase converts an RNP into free RNA plus released protein in a purified system, direct remodeling is plausible. In cells, reduced crosslinking of a protein after helicase activation could instead reflect RNA decay, altered localization, translation, or competition by a third factor. Orthogonal measurements of RNA abundance, RNP composition, localization, and direct biochemical remodeling distinguish these alternatives.

### Family names predict tendencies, not guarantees

DEAD-box proteins tend to be local strand-separation enzymes, whereas DEAH/RHA, Ski2-like, and many SF1 proteins tend to translocate directionally. These are useful starting rules, not definitions. Some DEAD-box proteins show limited movement or repeated action on specialized substrates; accessory proteins can increase functional persistence; some directional motors are autoinhibited or mechanically silent until recruited. Viral proteins can act on both RNA and DNA model substrates in vitro even when viral RNA is their main biological substrate.

Do not overgeneralize: the ability of an isolated protein to separate a short fluorescent duplex does not prove that its physiological function is duplex unwinding. Family assignment, cellular recruitment, biological substrate, and product identity must converge. The broad family framework is established, but the precise mechanical output of many human helicases remains incompletely mapped.

## 55.2. RecA-like cores, ATPase cycles, RNA binding, and mechanochemical coupling

### The catalytic cleft links ATP and RNA

The two RecA-like domains form a composite active site: one domain alone does not contain the complete nucleotide- and RNA-coupling machinery. In many structures, an open core has relatively separated domains, whereas ATP and RNA together stabilize a closed conformation. Closure aligns catalytic residues, magnesium, water, and phosphate groups and also organizes a continuous RNA-binding surface. Hydrolysis, phosphate release, and ADP release then change which contacts are favored. The cycle is therefore a network of ligand-linked conformational equilibria rather than a rigid sequence of cartoon poses.

RNA can stimulate ATPase activity by promoting closure or organizing catalytic motifs. ATP can increase RNA affinity or change the footprint. The order of ligand binding varies by family and conditions, and measured order can be obscured by rapid equilibria. A nonhydrolyzable ATP analog may stabilize a useful structural state, but analogs differ in geometry and metal coordination; no analog is guaranteed to reproduce the ATP ground state. ADP-aluminum fluoride and related transition-state mimics likewise constrain interpretations rather than serving as exact temporal photographs.

Figure 55.2 follows ligand binding, domain closure, hydrolysis, mechanical output, and release as coupled transitions. Table 55.2 lists the biochemical perturbations that distinguish defects in ATP binding, chemistry, RNA binding, and coupling.

![Figure 55.2. Ligand-linked ATPase cycle and mechanochemical branch points](../assets/figures/chapter1159_figure2.png)

**Figure 55.2. Ligand-linked ATPase cycle and mechanochemical branch points.** “RNA and nucleotide states reshape a composite cleft between RecA-like domains. Productive remodeling competes with release, reversal, pausing, and futile ATP turnover.”

**Table 55.2. Diagnose where a helicase cycle is disrupted.** Link perturbations to distinguishable biochemical defects.

| Observation | Plausible defect | Discriminating measurement | Interpretation limit |
| --- | --- | --- | --- |
| **No ATPase, no remodeling** | ATP binding, hydrolysis, folding, or active fraction | Nucleotide binding, thermal stability, active-site titration | One catalytic mutant can affect several steps |
| **ATPase retained, remodeling lost** | Uncoupling, RNA geometry, product loss | Direct product assay, RNA binding, trap control | Basal ATPase may come from off-pathway RNA |
| **RNA binding retained, ATPase lost** | Catalytic chemistry or closure defect | Nucleotide-state binding and structural control | Binding can occur in an unproductive orientation |
| **ATPase accelerated, product reduced** | Futile cycling or premature release | Matched ATPase/product time courses | A regulator can change both active fraction and coupling |
| **Product accumulates, turnover slow** | Product release or enzyme recycling | Single-turnover versus steady-state comparison | Product inhibition can mimic slow release |
| **Apparent direction changes** | Strand assignment or substrate artifact | Reverse tail geometry and tracking-strand labels | Duplex orientation must be chemically explicit |

### Chemical steps are not automatically mechanical steps

An ATP molecule can be hydrolyzed without producing the scored RNA product. Such uncoupling may arise because the enzyme binds an unproductive RNA orientation, slips, releases before completing work, or cycles in the absence of a required cofactor. Conversely, an observed unwinding event can include thermal base-pair opening captured by the protein, so the enzyme need not supply the full duplex free energy in a single stroke. Mechanochemical coupling describes how chemical-state probabilities bias these molecular fluctuations.

The coupling ratio is often stated as ATP molecules hydrolyzed per nucleotide moved or base pair separated. In practice, both numerator and denominator are difficult. ATPase assays count all active cycles, including futile ones. Ensemble unwinding assays average productive and nonproductive molecules and can lose products to reannealing. A processive motor may hydrolyze ATP while pausing; a DEAD-box protein may use ATP binding to create separation and hydrolysis mainly to recycle the enzyme. Coupling ratios are therefore conditions- and model-dependent rather than immutable family constants.

Directionality emerges from asymmetric contacts and gated affinity changes. A translocase alternates relative positions of the RecA-like domains and nucleic-acid contacts so that forward states are favored over backward states. Direction is specified on the tracking strand. For a 3′→5′ motor, the enzyme advances from the tracking strand's 3′ side toward its 5′ side; this does not mean the duplex itself has one global direction. At a junction, movement can exclude the complementary strand or collide with an RNP obstacle.

### ATPase cycles can enforce decisions

RNA metabolism is full of branched pathways. ATP-dependent remodeling can make branch choice time-dependent. A correctly assembled substrate may proceed quickly to a productive transition; a suboptimal substrate may wait long enough for a proofreading ATPase to remodel, reject, or discard it. In splicing, Prp-family ATPases act at defined transitions and can increase fidelity by giving competing routes different kinetic opportunities. The energetic cost is real: greater discrimination can reduce yield or slow processing.

ATPase-dependent commitment does not require a motor to “inspect” sequence in a cognitive sense. Specificity can reside in RNA geometry, partner proteins, the lifetime of a preceding complex, and competition between reaction paths. Mutations that slow ATPase release may increase rejection by extending a checkpoint, or reduce rejection by trapping an unproductive state. Interpreting variants therefore requires a kinetic network, not simply a faster-versus-slower activity ranking.

## 55.3. DEAD-box proteins as local strand-separation enzymes and regulated RNP clamps

### Local opening by a bent RNA strand

DEAD-box proteins are SF2 ATPases named for the amino-acid sequence of motif II. A common mechanism begins when the two RecA-like domains and ATP bind across a short RNA duplex. One RNA strand follows a strongly bent path across the closed core. That geometry is incompatible with continued pairing, causing a limited number of base pairs to separate. The protein does not need to begin on a long single-stranded extension or walk through the entire duplex. Duplex stability near the binding site, accessibility, and protein concentration often matter more than remote duplex length.

The locally opened strands may dissociate, refold, or be captured by another factor. ATP hydrolysis and inorganic phosphate release promote reopening and recycling. This division of labor explains why ATP binding can be central to separation while hydrolysis is central to turnover. It also explains why some mutants or nucleotide analog states clamp RNA tightly without completing multiple cycles.

Figure 55.3 contrasts local opening with directional tracking and shows how the same DEAD-box cycle can yield strand separation, protein exchange, or clamping.

![Figure 55.3. DEAD-box local opening, protein exchange, and clamping](../assets/figures/chapter1159_figure3.png)

**Figure 55.3. DEAD-box local opening, protein exchange, and clamping.** “DEAD-box proteins commonly destabilize a short duplex by imposing an incompatible RNA path. The resulting open or clamped state can support strand release, protein exchange, or assembly.”

### eIF4A, Ded1/DDX3, and regulation by context

The translation-initiation factor eIF4A is a DEAD-box ATPase whose isolated activity is modest. Within the eIF4F complex and with cofactors such as eIF4B, eIF4H, and RNA-binding components, eIF4A helps mRNA leaders become accessible to the scanning machinery. This is not equivalent to a single eIF4A molecule processively clearing every structure between the cap and start codon. Repeated local remodeling, cofactor-mediated recruitment, and cooperation with the ribosome-initiation apparatus create functional persistence at a messenger RNP.

Ded1 in budding yeast and its DDX3-family relatives illustrate broader substrate and regulatory behavior. They influence structured leaders, RNP granules, export-associated transitions, and stress-responsive translation. DDX3X can partition into condensate-like assemblies and resolve inhibitory RNA-RNA interactions in stress granules, but condensate localization does not establish that phase separation is the catalytic mechanism. Direct RNA remodeling, partner binding, post-translational modification, and local concentration all contribute.

Other DEAD-box proteins are recruited to specific pathways: Dbp2 participates in messenger RNP assembly and release of 3′-end processing factors; DDX39B is integrated with export-associated complexes; DDX21 coordinates ribosomal RNA transcription and processing. Their conserved cores operate inside distinct regulatory architectures. The enzyme family supplies a chemical engine; low-complexity regions, terminal domains, protein partners, and localization determine which RNA encounter becomes productive.

### Clamping and exchange are mechanistic outcomes

A closed DEAD-box protein can bind RNA and ATP or an ATP-like state tightly enough to act as a clamp. Clamping can hold an RNA conformation, prevent a competing protein from rebinding, or recruit a partner that recognizes the closed ATPase. Hydrolysis and product release can then time disassembly. A mutation that reduces hydrolysis may therefore increase occupancy while reducing pathway flux. Conversely, a mutation that weakens closure may preserve basal ATP turnover yet eliminate the stable RNP intermediate.

“RNA chaperone” is also a functional, not purely enzymatic, description. DEAD-box cycles can help structured RNAs escape kinetic traps by repeatedly opening local helices, but they do not encode the final native fold. Direction comes from the RNA energy landscape and from partners that capture productive intermediates. In ribosome biogenesis, multiple helicases act at different assembly stages; depletion phenotypes identify blocked maturation states, while direct substrates and exact remodeling events often require crosslinking, structures, and reconstitution.

## 55.4. DEAH/RHA motors, directional translocation, processivity, and spliceosome remodeling

### A single-stranded path supports directional tracking

DEAH/RHA proteins are SF2 RNA motors that usually engage a single-stranded RNA segment and translocate with 3′→5′ polarity. Their helicase core is extended by family-specific domains that shape an RNA channel, regulate the catalytic cleft, and connect to cofactors. Repeated ATPase cycles shift contacts along the tracking strand. When the enzyme reaches a duplex or RNP barrier, continued translocation can separate strands, move RNA relative to proteins, or destabilize an assembly.

This mechanism gives substrate tails and geometry special importance. A directional motor needs physical access to a loading segment with the correct polarity and sufficient length. The same duplex can be remodeled efficiently with one tail and poorly with the opposite tail. Binding to a free oligonucleotide does not prove access inside a large RNP: proteins may occlude the loading strand, or a cofactor may position the motor internally. Processivity depends on competition between forward stepping, pausing, backtracking, and dissociation.

Structural studies of Prp43 reveal a tunnel-like path for single-stranded RNA across the helicase core and accessory domains, supporting translocation rather than direct local-duplex capture. However, a structure containing a nucleotide mimic and a short RNA constrains contacts in that state; it does not alone determine step size, rate-limiting transition, or whether the enzyme moves continuously in the full spliceosome. Those assignments require kinetics and state-resolved structural series.

### Spliceosomal ATPases remodel a changing machine

The spliceosome is assembled anew on each intron and passes through a series of RNP conformations. ATPases act before catalysis, between the two chemical steps, during proofreading, and during disassembly. Prp16 remodels the spliceosome after branching and can reject suboptimal branch-site or 5′-splice-site configurations. Prp22 promotes events associated with exon ligation and release, while Prp43 contributes to discard and disassembly. These proteins do not act on a naked duplex in vivo; they pull or reposition RNA within a massive protein-RNA complex.

Genetic suppression is a powerful but conditional probe. A slower ATPase allele can allow a weak substrate more time to react before rejection, whereas a hyperactive or prematurely recruited ATPase can discard potentially productive intermediates. Such phenotypes support competition between catalysis and remodeling, but they do not identify the physical contact by themselves. Biochemical reconstitution and structural trapping connect kinetic effects to particular RNP states.

Figure 55.4 presents a kinetic branch model in which catalytic progression competes with ATPase-driven rejection rather than portraying proofreading as direct sequence recognition.

![Figure 55.4. Spliceosomal proofreading as competition among rates](../assets/figures/chapter1159_figure4.png)

**Figure 55.4. Spliceosomal proofreading as competition among rates.** “Fidelity can arise when productive chemistry competes with ATPase-driven remodeling or discard. An allele changes the time available to react and can have substrate-specific effects.”

### Cofactors create timing and pathway specificity

G-patch proteins are common activators of DEAH-family ATPases. A G-patch can contact the helicase and stabilize an active RNA-bound architecture, increase ATPase or unwinding activity, or recruit the enzyme to a substrate. Prp43 is activated by different G-patch proteins in splicing and ribosome biogenesis. Sharing one motor across pathways is possible because cofactors and localization determine where and when the active state forms.

The helicase is therefore one component of a composite enzyme. An isolated-core rate can underestimate activity if an essential cofactor is missing; saturating cofactor in vitro can obscure physiological regulation. Cofactors can also change substrate specificity rather than merely multiply a rate. A mechanistic comparison should report the full protein construct, cofactor stoichiometry, RNA architecture, nucleotide and magnesium conditions, and whether the cofactor changes binding, chemistry, coupling, or release.

## 55.5. Ski2-like, SF1, viral, and bacterial RNA motors in processing, decay, and defense

### Ski2-like motors feed RNA into processing machines

Ski2-like SF2 helicases include cytoplasmic Ski2, nuclear Mtr4, and spliceosomal Brr2. Their accessory domains and partner complexes create specialized routes for RNA. In the cytoplasmic Ski complex, Ski2 is positioned to receive RNA associated with ribosomes or decay substrates and channel it toward the exosome. The architecture includes gates and regulatory contacts that reduce uncontrolled access, illustrating how processivity can be a property of a multiprotein machine.

Mtr4 associates with the nuclear RNA exosome and with adaptor complexes that select aberrant, pervasive, or processing-associated RNAs. Its arch domain provides binding surfaces for adaptors and exosome partners. Cryogenic electron microscopy of exosome-Mtr4 assemblies supports an RNA path from the helicase toward the ribonuclease chamber. The motor can unwind or feed RNA, but degradation requires the nuclease; conversely, the nuclease cannot efficiently receive many structured substrates without the motor and adaptors. Chapter [34](chapter1032.md) owns the consequences for RNA decay, while this chapter owns the coupled motor logic.

Brr2 contains tandem helicase-like cassettes, but only one is catalytically active; the inactive cassette has regulatory and structural roles. This is a useful boundary case against motif counting. A protein can retain a helicase-like fold that has become a scaffold or regulator during evolution. Autoinhibitory domains and spliceosomal contacts keep Brr2 from unwinding U4/U6 small nuclear RNA at the wrong time.

Figure 55.5 compares free motor action with channeling by the Ski-exosome and Mtr4-exosome assemblies. Table 55.3 places Ski2-like, SF1, viral, and bacterial examples on the same substrate-output grid.

![Figure 55.5. Free helicase action versus RNA channeling to the exosome](../assets/figures/chapter1159_figure5.png)

**Figure 55.5. Free helicase action versus RNA channeling to the exosome.** “Ski2 and Mtr4 couple ATP-dependent RNA routing to exosome-mediated decay. Partner complexes select substrates, gate access, and align the motor with the nuclease.”

**Table 55.3. Representative RNA-motor families and boundary cases.** Compare substrate entry, polarity, partners, and biological output without collapsing family diversity.

| Family or example | Typical substrate entry | Dominant tendency | Partner-created function | Boundary case |
| --- | --- | --- | --- | --- |
| **DEAD-box eIF4A/Ded1/DDX3** | Direct local RNA engagement | Local opening or clamp | Translation and granule remodeling | Functional persistence without core processivity |
| **DEAH Prp16/Prp22/Prp43** | Single-stranded RNA segment | 3′→5′ translocation | Spliceosome remodeling and discard | Cofactor-dependent loading inside an RNP |
| **Ski2/Mtr4** | Gated RNA channel | Directional feeding/unwinding | Exosome substrate delivery | Motor is not the nuclease |
| **Brr2** | Spliceosome-confined U4/U6 context | Regulated translocation/unwinding | Timed spliceosome activation | Second helicase-like cassette is inactive |
| **SF1 UPF1** | Single-stranded RNA with regulatory domains | Translocation and RNP remodeling | Nonsense-mediated decay | Model-duplex helicase output incompletely predicts NMD |
| **Viral NS3** | Single-stranded tail or replication intermediate | Directional motor | Replication-complex action | DNA model-substrate behavior requires cautious transfer |
| **Bacterial DEAD-box factors** | Local structured RNA or RNP | Local remodeling | Ribosome assembly and decay | Stress phenotype can be indirect |

### SF1 Upf1-like motors combine translocation with regulatory domains

UPF1 is an SF1 RNA helicase central to nonsense-mediated mRNA decay and other RNA surveillance pathways. Its helicase core is surrounded by regulatory domains that affect RNA binding and partner interactions. UPF2 can relieve autoinhibition and remodel the RNA-bound state; ATPase cycling controls RNA residence and movement. Single-molecule and biochemical analysis of Upf1-like proteins shows that nucleic-acid grip influences processivity.

Recent work complicates a simple “more unwinding equals more nonsense-mediated decay” model. UPF1 variants that retain ATPase activity but show reduced canonical helicase activity can still support substantial decay, while ATPase autoinhibition and activation alter RNA-binding kinetics and pathway efficiency. The relevant outputs include loading, translocation, release, partner exchange, ribosome-associated remodeling, and access of decay enzymes. These findings do not make helicase activity irrelevant; they show that a model-duplex assay does not capture every cellular mechanical function.

Sen1 and related SF1 helicases provide another boundary: directional translocation can remove RNA polymerase or resolve transcription-associated RNA-DNA structures, connecting RNA surveillance with transcription termination and genome stability. Their biological substrate can include RNA-DNA hybrid contexts even when the tracking interactions are RNA-centered. The substrate should be described chemically rather than assigned by family name.

### Viral helicases integrate replication and immune evasion

Many positive-strand RNA viruses encode SF1 or SF2 helicases. Hepatitis C virus NS3 is a classic SF2 motor whose helicase domain translocates 3′→5′ on nucleic acid and whose N-terminal region also contains a protease domain in the full protein. Optical-tweezers and single-molecule fluorescence studies have measured stepping, pauses, bursts, and strand switching, while structural series support a ratchet-like translocation mechanism. Results on DNA model substrates illuminate motor principles but must be transferred cautiously to viral RNA replication complexes.

Flaviviral NS3 proteins and coronavirus nsp13-like helicases participate in replication and can influence capping, recombination, or innate immune antagonism. Their activity is shaped by the viral polymerase, membranes, other nonstructural proteins, and structured genomic RNA. An inhibitor that blocks purified helicase ATPase activity may fail in cells because it does not reach the replication organelle, while a compound that suppresses infection may act through protease, polymerase, membrane, or host pathways. Direct target engagement and resistant-virus genetics are essential.

### Bacterial RNA helicases emphasize environmental and pathway diversity

Bacteria encode several DEAD-box proteins and, in some lineages, other RNA motor families. These enzymes contribute to ribosome assembly, RNA decay, translation, stress adaptation, and cold growth. Low temperature stabilizes RNA secondary structure and slows spontaneous rearrangement, increasing demand for local RNA remodeling. A cold-sensitive growth phenotype and an RNA-folding defect can support this role, but broad stress phenotypes can also result from impaired ribosome biogenesis or altered decay.

Bacterial degradosomes illustrate family-specific integration. A DEAD-box helicase can help a ribonuclease access structured RNA without becoming a long-range processive motor. In other systems, RhlE-like proteins or specialized helicases act in ribosome assembly. Organellar helicases may retain bacterial ancestry but evolve eukaryotic targeting sequences and new partners. The appropriate comparison is not “simple bacterial versus complex eukaryotic”; it is how a conserved catalytic core is embedded in each RNA-processing network.

## 55.6. Substrate specificity, cofactors, localization, regulation, and pathway commitment

### Specificity is distributed across an encounter

The conserved helicase core often recognizes RNA backbone and shape more strongly than a unique sequence. Biological specificity can instead arise from recruitment by an RNA-binding protein, docking on a large RNP, recognition by terminal domains, a required single-stranded geometry, localization to an organelle or condensate, or activation only after a preceding pathway step. Specificity is therefore distributed across the molecular encounter.

This distributed model explains why crosslinking maps can be broad while functional effects are selective. Binding is not equivalent to remodeling, and remodeling is not equivalent to pathway commitment. A helicase may sample many transcripts but complete an ATP-coupled cycle only on those presented by a cofactor. Conversely, abundant nonspecific RNA can stimulate ATPase activity in vitro and obscure a rare biologically privileged substrate.

Accessory domains can recognize G-quadruplexes, proteins, or RNP architectures. For example, DHX36 is strongly associated with G-rich nucleic-acid structures, whereas DDX39B participates in export-associated assemblies and Mtr4 recognizes adaptor networks. These preferences combine direct binding and context; no single domain should be assumed to dictate all cellular targets.

### Regulation changes state occupancy and access

Regulation can operate at expression, localization, post-translational modification, cofactor binding, autoinhibition, oligomerization, or phase partitioning. Phosphorylation or methylation may change partner binding without altering the purified core ATPase. A low-complexity region may promote condensate recruitment, thereby raising local concentration and changing substrate competition. Stress can redistribute DDX proteins among translating messenger RNPs and granules. Such changes are spatial and network-level forms of regulation, not substitutes for enzymatic mechanism.

Autoinhibition prevents untimely RNA engagement or ATP consumption. Partner binding can open an RNA channel, stabilize domain closure, or expose a loading site. Productive regulation should be dissected into at least four questions: Does the regulator change RNA affinity? Does it change ATP turnover? Does it change the ratio of ATP use to remodeling? Does it change release or recycling? A tenfold increase in ATPase rate is not necessarily activation of the biological output.

Figure 55.6 organizes specificity as sequential gates from localization to recruitment, catalytic commitment, product capture, and release.

![Figure 55.6. Distributed specificity gates a productive helicase encounter](../assets/figures/chapter1159_figure6.png)

**Figure 55.6. Distributed specificity gates a productive helicase encounter.** “Physiological specificity is often distributed across where a helicase is located, how it is recruited, which RNA geometry it can engage, and whether a downstream factor captures the remodeled product.”

### Commitment is a competition among rates

An RNA enters a pathway when forward flux becomes more likely than return or diversion. A helicase can create commitment by exposing a nuclease site, releasing an export factor, separating a snRNA duplex, ejecting a surveillance-protective protein, or delivering RNA to a degradation channel. The outcome depends on the relative rates of remodeling, capture by the downstream factor, refolding, rebinding, and enzyme dissociation.

Box 55.1 uses the same kinetic competition to explain why a helicase can appear to promote opposite fates in different experiments.

> **Box 55.1. How one helicase can promote opposite RNA fates**
>
> - Misconception prevented: A helicase has one intrinsic “pro-RNA” or “anti-RNA” biological function.

Genetic epistasis can order factors in a pathway but rarely identifies a direct mechanical step. Acute depletion followed by time-resolved RNA structure, RNP composition, localization, and processing measurements is stronger than a late steady-state transcriptome alone. Rescue with wild-type, ATP-binding-defective, hydrolysis-defective, RNA-binding-defective, and cofactor-binding-defective alleles can separate catalytic from scaffolding roles, provided expression and localization are matched.

## 55.7. Kinetics, structures, single-molecule measurements, ensemble assays, and interpretation limits

### Match each observable to the claim

Helicase enzymology usually combines an ATPase measurement with an RNA-remodeling measurement. ATP turnover can be followed by phosphate detection, radiolabeled nucleotide separation, chromatography, or enzyme-coupled optical assays. Unwinding can be followed by native gels, fluorescence resonance energy transfer (FRET), fluorescence quenching, molecular beacons, or force spectroscopy. Protein displacement can be measured by fluorescence, electrophoretic mobility, crosslinking, or compositional analysis. Each output has distinct artifacts.

In a gel-based unwinding assay, separated strands can reanneal after the helicase acts. An excess unlabeled trap strand can capture a released strand, but the trap can also bind the enzyme or change reaction kinetics. FRET loss can report strand separation, fluorophore rearrangement, protein-induced bending, dye quenching, or cleavage. ATPase-coupled assays can be limited by the reporter relay or inhibited by test compounds. Direct and orthogonal product measurements should agree before coupling is quantified.

Steady-state ATPase parameters average binding, hydrolysis, product release, RNA release, and inactive fractions. Single-turnover unwinding places active enzyme in excess over substrate and can reveal productive amplitudes and rates without requiring repeated enzyme recycling. Pre-steady-state nucleotide experiments can detect rapid binding or phosphate-release phases. Varying duplex length and tail length can distinguish local opening from translocation, but only if duplex stability and end effects are controlled.

### Structures are state constraints, not movies

X-ray crystallography, cryogenic electron microscopy, and nuclear magnetic resonance define contacts and conformational states. Structures have revealed closed DEAD-box cores on bent RNA, RNA channels through DEAH and Ski2-like enzymes, regulatory domain contacts in UPF1, and motors embedded in spliceosomes or exosomes. Comparing nucleotide and substrate states can motivate an ATPase cycle. Mutational tests can then ask whether a contact affects binding, catalysis, coupling, or assembly.

Structural occupancy is selected by purification, ligand concentrations, mutations, analogs, freezing, and classification. A state that yields a high-resolution map may be rare during normal flux. Missing density can represent motion, compositional heterogeneity, or damage rather than absence. Local resolution and model uncertainty matter particularly for RNA paths and flexible accessory domains. Time order should be inferred from kinetics or time-resolved perturbation, not from arranging static structures according to visual preference.

### Single molecules expose distributions

Single-molecule FRET can reveal repeated opening and closing, dwell-time distributions, and heterogeneous responses among molecules. Optical or magnetic tweezers apply force to a tethered nucleic acid and report extension changes as a motor unwinds, translocates, or allows refolding. Nanopore and high-resolution tracking approaches can add sequence or positional information. The HCV NS3 system demonstrated that individual trajectories contain pauses and bursts concealed by ensemble averages.

Force changes the RNA energy landscape. A duplex held near its mechanical transition opens more readily than the same duplex without load. Attachment points determine which movement changes extension. Fluorophores can alter a local duplex, and only molecules that survive selection filters enter analysis. Hidden-state models can summarize noisy trajectories, but the number of inferred states depends on model choice and time resolution. Single-molecule does not mean assumption-free.

Figure 55.7 aligns structural, bulk kinetic, and single-molecule evidence on one ATPase cycle. Table 55.4 links common observations to justified conclusions and alternatives. Box 55.2 gives a practical interpretation sequence for apparent uncoupling.

![Figure 55.7. Evidence triangulation across structures, bulk kinetics, and single molecules](../assets/figures/chapter1159_figure7.png)

**Figure 55.7. Evidence triangulation across structures, bulk kinetics, and single molecules.** “Structures constrain possible states, bulk kinetics measures population flux, single-molecule trajectories expose distributions, and cellular perturbations establish biological consequence. No one layer defines the whole cycle.”

**Table 55.4. Observations, justified conclusions, and alternatives.** Prevent platform outputs from being promoted into unsupported mechanisms.

| Observation | Justified conclusion | Important alternatives | Orthogonal next test |
| --- | --- | --- | --- |
| **RNA-stimulated ATPase** | RNA affects nucleotide-cycle flux | Nonspecific polyanion stimulation | Substrate-series direct remodeling |
| **FRET decrease** | Reporter geometry changed | Strand separation, bending, quenching, cleavage | Gel or sequence-resolved product assay |
| **Released RBP in cells** | RNP occupancy changed | RNA loss, localization, competition | Purified RNP remodeling plus RNA abundance control |
| **Closed ATP-RNA structure** | Selected closed state is physically feasible | Analog- or preparation-stabilized rare state | Kinetics and alternative nucleotide states |
| **Single-molecule steps** | Discrete signal transitions occurred | Missed events, state-model dependence, tether effects | Change force, label geometry, and time resolution |
| **Slow surface dissociation** | Long surface residence | Rebinding, avidity, mass transport | Solution chase and lower ligand density |
| **Variant loses unwinding** | Model-substrate output is impaired | Instability, binding loss, uncoupling | Matched ATPase, binding, structure, and rescue |

> **Box 55.2. Diagnose apparent ATPase-unwinding uncoupling**
>
> - Misconception prevented: ATP turnover without a fluorescent unwinding signal always proves a biologically futile enzyme.

### A minimum mechanistic evidence ladder

A strong claim that an enzyme directionally unwinds an RNP substrate should satisfy several layers. First, purified and active components bind the relevant RNA architecture. Second, ATP or another stated nucleotide is required and hydrolyzed. Third, a direct assay establishes the RNA or RNP product with product traps and no-enzyme controls. Fourth, tail polarity, substrate geometry, length dependence, and time courses support directionality or processivity. Fifth, catalytic-site perturbations preserve overall folding while disrupting the predicted step. Sixth, cofactors reconstruct physiological specificity. Seventh, cellular phenotypes are rescued by mechanistically informative alleles and linked to the same substrate.

No one experiment must contain every layer, and some physiological RNPs cannot yet be fully reconstituted. The evidence grade should then match the gap. Genetic dependence plus structural proximity can establish a likely remodeling role; it does not justify a measured step size. An isolated-core single-molecule trajectory can establish motor behavior on the model substrate; it does not by itself establish the cellular target.

## 55.8. Evolution, disease variants, inhibitors, antiviral targets, and biotechnology

### Evolution conserves the engine while rewiring control

SF1 and SF2 helicases trace to ancient nucleic-acid ATPases. Conserved core motifs maintain ATP- and polymer-coupling chemistry, whereas insertions, terminal domains, inactive cassettes, cofactor interfaces, and localization signals evolve rapidly. Gene duplication allows one core to specialize in ribosome assembly, translation, splicing, decay, immunity, or viral replication. Some paralogs remain partially redundant; others become essential because their recruitment or regulation is unique.

Family trees should not be interpreted as direct mechanism trees. Closely related paralogs can have different expression, phase behavior, and partners, while distantly related motors can converge on similar pathway outcomes. Comparative biochemistry across bacteria, archaea, eukaryotes, organelles, and viruses helps distinguish ancestral motor properties from recently evolved regulatory layers.

### Disease variants affect more than ATP hydrolysis

Variants in human helicases are associated with neurodevelopmental disorders, cancer, bone marrow failure, ribosomopathies, infection susceptibility, and inflammatory disease. A missense change in the ATPase core can alter nucleotide affinity, catalytic rate, RNA binding, or interdomain coupling. A variant outside the core can change localization, cofactor binding, condensate partitioning, or protein stability. Haploinsufficiency, dominant-negative trapping, gain of substrate occupancy, and altered pathway choice are distinct disease mechanisms.

DDX3X is a useful caution. It participates in translation, RNA export-associated processes, stress granules, innate signaling, and development, and it can behave differently across tumor types. A cancer-associated variant cannot be labeled “loss of helicase” from reduced proliferation or altered translation alone. Biochemical state, expression, localization, interacting partners, sex-linked paralog context, and cell lineage must be considered. Similar care applies to SKIV2L-related inflammatory phenotypes: the relevant defect may involve failure to clear immunostimulatory RNA rather than a generic loss of duplex unwinding.

### Inhibitors can block, trap, or redirect states

The ATP pocket is conserved and competes with millimolar cellular ATP, making selective orthosteric inhibition difficult. Allosteric pockets and family-specific interfaces offer alternatives, but biochemical selectivity must include related helicases and other ATPases. Aggregation, metal chelation, fluorescence interference, RNA binding, and general translation suppression can masquerade as direct inhibition.

Rocaglates illustrate state trapping. Rather than simply turning off eIF4A, these compounds stabilize eIF4A on selected polypurine sequences, converting the RNA-protein complex into a roadblock for scanning. This “molecular staple” mechanism produces sequence-dependent translation effects and should not be summarized as a uniform fall in helicase activity. DDX3-directed compounds such as RK-33 have shown cellular and preclinical effects, but target engagement, off-target translation stress, and context-dependent DDX3 biology remain central evaluation requirements.

Viral helicases are attractive because they are often essential and structurally distinct from host enzymes. Resistance mapping, biochemical potency against full replication complexes, antiviral activity in relevant cells, cytotoxicity, and pharmacokinetics must converge. Inhibiting an isolated NS3 helicase domain is a starting observation, not a drug mechanism. Host helicases can also be antiviral targets when viruses depend on them, but the therapeutic window may be narrower and immune consequences may be bidirectional.

### Biotechnology uses motors as controlled sources of direction and remodeling

Helicases can improve isothermal amplification, nanopore sensing, nucleic-acid sequencing, RNA structure analysis, and synthetic RNP control. A directional motor can feed nucleic acid through a pore or displace secondary structure ahead of a polymerase. A DEAD-box-like local remodeler could be engineered to expose a programmable site without global denaturation. Fusion to RNA-recognition domains can redirect recruitment, although specificity, ATP cost, off-target remodeling, and cellular regulation remain challenges.

Engineering should preserve the distinction between catalytic core and control architecture. Increasing basal ATPase activity may waste energy and reduce product yield if coupling falls. Increasing RNA affinity may trap product. Greater processivity may destroy local specificity. The desired property is a complete cycle matched to the application: bind the intended substrate, perform the required remodeling, release at the correct time, and avoid competing RNAs.

Table 55.5 compares intervention and engineering strategies by the state they perturb and the evidence needed to establish mechanism.

**Table 55.5. Interventions act on different states of the cycle.** Compare inhibitor and engineering mechanisms by state, not by one potency number.

| Strategy | State or interface perturbed | Desired output | Required evidence | Major risk |
| --- | --- | --- | --- | --- |
| **ATP-site inhibitor** | Conserved nucleotide pocket | Reduce cycle flux | Nucleotide competition and family selectivity | Broad ATPase inhibition |
| **Allosteric inhibitor** | Family-specific pocket | Block closure or coupling | State-specific kinetics and target engagement | Hidden paralog activity |
| **Molecular staple** | Protein-RNA composite site | Trap selected RNA-bound state | Ternary complex and sequence dependence | Off-target translation roadblocks |
| **Cofactor-interface inhibitor** | Recruitment or activation surface | Pathway-selective blockade | Partner disruption without global unfolding | Alternative cofactors |
| **Engineered targeting fusion** | RNA-recognition plus helicase core | Remodel selected RNA | Direct product, off-target mapping, release | Nonspecific RNA engagement |
| **Nanopore motor optimization** | Stepping and grip | Controlled translocation | Force-dependent run length and stall profile | Trapping or irregular stepping |

## Experimental Foundations and Evidence

The field rests on convergence among classical genetics, purified-component biochemistry, structural biology, and single-molecule measurements. Temperature-sensitive spliceosomal and ribosome-biogenesis alleles first exposed ordered ATP-dependent transitions. Recombinant proteins and model duplexes separated ATP hydrolysis from strand separation and revealed family-specific tail requirements. Structures established the shared RecA-like core and showed sharply different RNA paths. Modern cryogenic electron microscopy embeds motors in native-like spliceosomes, ribosomes, and exosomes, while transcriptomics and crosslinking identify cellular substrates.

Each evidence class has a characteristic reach. Genetics establishes necessity and kinetic competition in a cellular pathway. Reconstitution can establish sufficiency and directness. Structures establish physical compatibility and state-specific contacts. Bulk kinetics estimates population flux and active amplitudes. Single-molecule assays expose distributions and intermediate lifetimes. Cellular sequencing and imaging establish breadth, location, and downstream consequence. Strong mechanistic synthesis uses the narrowest justified conclusion from each rather than asking one assay to answer all layers.

## Biological Contexts Across Systems

In bacteria, helicases help ribosome assembly, translation, RNA decay, and adaptation to temperature or nutrient stress. In archaea, Ski2-like and other helicases connect RNA processing to exosome-related machinery. Eukaryotes expand paralog number and regulatory domains, embedding motors in the spliceosome, ribosome-assembly factors, export complexes, surveillance pathways, translation initiation, organelles, and granules. Plants and fungi include lineage-specific duplications and stress programs; metazoans add developmental and tissue-selective expression. Viruses encode compact motors or exploit host helicases.

The common principle is not that every system uses the same motor. It is that structured RNA and stable RNPs create kinetic barriers whose regulated rearrangement consumes energy. Organism-specific conclusions should remain labeled. A yeast Prp ATPase mechanism may be deeply conserved in spliceosome architecture yet differ in accessory proteins and regulatory timing in mammals. A bacterial cold-shock helicase phenotype may not predict mammalian stress-granule behavior. Comparative conservation is strongest for core chemistry and must be demonstrated for pathway wiring.

## Technology, Computational, Clinical, or Engineering Links

Computational analysis can classify helicase families, predict domain architecture, map disease variants onto structures, and fit kinetic networks. Molecular simulations can suggest allosteric paths or ligand pockets, but they depend on starting states and force fields and do not replace biochemical flux measurements. Machine-learning models trained on sequence or structure may predict RNA-binding regions or compound affinity; validation should test family selectivity, nucleotide state, RNA context, and cellular target engagement.

Clinically, helicases are potential biomarkers and drug targets in cancer, infection, developmental disease, and inflammatory syndromes. Their pleiotropy is both opportunity and risk. A pathway can depend strongly on one helicase state in a specific tumor or virus-infected cell, but systemic inhibition may impair translation, hematopoiesis, neurodevelopment, or innate immunity. State-selective or context-selective targeting is more promising than assuming that conserved ATPase blockade will be safe.

Engineering applications benefit from explicit kinetic design. A motor used to control polymer movement through a nanopore needs predictable stepping and stall behavior. A helicase added to amplification must relieve structures without degrading primers or inhibiting polymerase. A synthetic RNA remodeler must release product rather than remain clamped. These design questions map directly onto the biochemical framework in [124](chapter1156.md).

## Recent Consensus

The field agrees that “RNA helicase” names an evolutionary and biochemical class broader than processive duplex-unwinding enzymes. DEAD-box proteins commonly act through local strand separation and ATP-dependent clamping, while many DEAH/RHA, Ski2-like, SF1, and viral proteins use directional translocation. The two-RecA-domain core couples RNA and nucleotide states, but family-specific domains and cofactors determine substrate access, processivity, timing, and pathway outcome.

There is also broad agreement that ATP hydrolysis, unwinding, translocation, and RNP remodeling are distinct observables. ATPase activity alone does not establish mechanical work, and a model-duplex endpoint does not define cellular function. Structures, ensemble kinetics, single-molecule trajectories, and cellular perturbations are complementary. Disease variants and compounds must be described by their effects on particular biochemical states, recruitment events, or partner interactions rather than by a generic “helicase activity” label.

## Open Questions, Controversies, Deprecated Models, and Common Misconceptions

Open questions:

- How many human helicases act primarily as directional translocases, local strand separators, clamps, protein-displacement factors, or noncatalytic scaffolds on their physiological substrates?
- Which accessory factors determine productive coupling rather than simply increasing basal ATPase activity?
- How are ATPase cycles synchronized with irreversible chemistry or degradation in large RNP machines?
- Which single-molecule states correspond to the dominant pathway flux inside cells rather than to rare behaviors selected by assay geometry?
- Can disease variants be classified by kinetic mechanism well enough to support variant-specific therapy?
- Which allosteric or RNA-composite pockets allow selective inhibition without broad disruption of essential host RNA metabolism?

Controversies:

- Cellular crosslinking, condensate localization, and broad transcriptomic changes can support a helicase's involvement in an RNA network, but the extent to which each phenotype reflects direct remodeling versus secondary consequences remains case-specific.
- The meaning of processivity can differ across isolated enzymes, cofactor-bound complexes, and confined RNP machines. Functional persistence in a pathway need not equal long uninterrupted motion by one core.

Deprecated or weakened claims:

- Helicase-family annotation was once often treated as evidence for long-range processive duplex unwinding. Comparative biochemistry now shows that local opening, clamping, and RNP exchange are common primary outputs.
- Static structures were sometimes arranged as a complete mechanochemical movie without kinetic testing. Structural state order and pathway flux now require independent temporal evidence.

Common misconceptions:

- “All RNA helicases unwind duplex RNA processively.” Many DEAD-box proteins open short local helices, and some family members primarily remodel protein-RNA contacts or clamp RNA.
- “ATPase activity proves helicase activity.” ATP can be hydrolyzed in futile or uncoupled cycles; direct RNA or RNP product measurement is required.
- “A nonhydrolyzable ATP analog is an exact ATP-state mimic.” Analogs impose distinct geometries and affinities and must be interpreted as state-stabilizing reagents.
- “One ATP is hydrolyzed for every base pair opened.” Coupling ratios depend on family, substrate, cofactors, pauses, slipping, futile cycles, and measurement design.
- “More ATPase activity means a more active remodeler.” A regulator can accelerate futile cycling, and a fast mutant can lose pathway specificity or product capture.
- “A helicase mutation outside the catalytic core is nonmechanistic.” Accessory regions can govern recruitment, autoinhibition, localization, partner binding, and release.
- “A helicase inhibitor simply turns the enzyme off.” Compounds can compete, allosterically block, trap a state, staple an enzyme to RNA, perturb RNA directly, or inhibit an assay reporter.
- “Single-molecule data directly reveal behavior in cells.” Tethers, forces, fluorophores, selection, and model fitting constrain which behaviors are observed.