Inching over hurdles: how DNA helicases move on crowded lattices

Abstract

Many of the genome maintenance transactions require continuous progression of molecular motors along single or double stranded DNA (dsDNA) molecule. DNA, however, is rarely found in the cell in its bare form. Structural proteins organize dsDNA and control its accessibility to molecular machines of DNA replication, repair, recombination, and transcription. Single-stranded DNA (ssDNA) is sequestered by ssDNA binding proteins, which protect it from degradation, modification and undesired transactions. Appreciation of how molecular machines compete with these stationary blocks and with each other for the access to DNA is important for our understanding of the mechanisms underlying genome maintenance. This understanding in turn establishes the molecular basis of various human diseases resulting from defects in molecular motors and their ability to navigate in crowded intracellular environments. By building upon our recent finding that it is possible for a helicase translocating on ssDNA to bypass a stationary bound protein without displacing it, we discuss potential outcomes of collisions between DNA helicases and ssDNA binding proteins. We then propose that the selective ability of some helicases to bypass or displace a specific ssDNA binding protein may be important for activation of these enzymes for particular DNA maintenance tasks.

Introduction: DNA transactions in a crowded cell

Cellular interior presents a crowded environment where numerous molecular machines zoom along their preferred tracks. One of such busy cellular highways is DNA. It is shared by molecular machines orchestrating readout, duplication and maintenance of the genetic material. Double-stranded DNA (dsDNA) securely stores the genetic information, but it also is a barrier for the readout of this information by the cellular machineries that require single-stranded DNA (ssDNA) as templates. Unwinding of dsDNA into ssDNA is catalyzed by a ubiquitous class of enzymes called DNA helicases (reviewed in ^1–4). Additionally, DNA-translocating molecular motors move along highly organized genetic material at different speeds in both directions colliding with each other and negotiating stationary obstacles (Figure 1). In eukaryotes, cellular DNA is organized in chromatin, where nucleosomes present a barrier that controls virtually all vital DNA transactions (Figure 1a). DNA translocating motors such as DNA and RNA polymerases as well as numerous chromatin remodeling complexes exert physical forces by pushing and pulling nucleosomes and twisting the DNA around them (reviewed in ⁵). Encounter between RNA polymerase (RNAP, a molecular machine of transcription) transcribing one of the duplex DNA strands and a nucleosome can result in quite different outcomes. RNAP can slide the nucleosome from its original position unwrapping the DNA from the histone core (Figure 1c). Alternatively, reformation of the histone-DNA contacts and of the stable double helix may slide the polymerase back to the arrested state. Such back-sliding is deleterious to transcription because the backtracked polymerase possesses no energy source to disrupt DNA-histone contacts ^{6, 7}

Figure 1. Nucleic acid motors and their cellular lattices.

(a) Nucleosomes organize the genomic DNA and present a barrier to various cellular DNA transactions. (b) Chromatin remodeling machines regulate access to genetic information by dislodging or repositioning nucleosomes. (c) Nucleosomes may also affect transcription by impeding progression of RNA polymerase. (d) Recombination specific chromatin remodeler, Rad54 is stimulated by interaction with nucleoprotein filament formed by Rad51 recombinase on ssDNA^{54, 55}. (e) Rad54 also controls postsynaptic steps in homologous recombination by acting on the products of Rad51-catalyzed strand invasion⁵⁵. Several bona fide DNA helicases also act at this step³⁸. (f) Srs2 helicase controls recombination by facilitating disassembly of Rad51 nucleoprotein filaments. (g) SMARCAL1 (also known as HARP) is an annealing helicase whose function involves re-formation of the DNA duplex by re-zipping ssDNA coated with ssDNA binding protein, RPA.

To counteract the negative influence of nucleosomal organization, chromatin remodeling complexes regulate access to nucleosomal DNA by dislodging, repositioning or redistributing nucleosomes and by exchanging histone variants (reviewed in ⁸) (Figure 1b). Cellular chromatin remodeling machines are numerous in their function and organization: while some work as large complexes that can encapsulate an entire nucleosome, others perform their tasks as individual motors. No matter how complex or simple, in their core all of these machines contain very similar Snf2-family motors structurally related to DNA and RNA helicases.

In addition to nucleosome-coated DNA, translocating molecular motors collide with other stationary bound proteins, elements of DNA secondary structure and DNA lesions. To read out the genetic information, cellular machines incorporating these motors compete for access to ssDNA. Replisome (a molecular machine of replication driven by replicative helicase and DNA polymerases) has to cope with obstacles presented by DNA repair complexes as well as by stalled and active transcription events. The processes of DNA replication and RNA transcription overlap both spatially and temporally, but progress at different speeds forcing the replisome and RNA polymerase to compete for access to or co-exist on the same DNA molecule ^{9, 10}.Similar to the transcription and replication, many cellular DNA transactions affect one another either positively or negatively and are greatly influenced by the crowded environment of the cell.

Helicases and related nucleic acid translocases

Since their discovery in 1970s, helicases were reputed as the canonical enzymes that “unwind” duplex nucleic acids in an energy-dependent manner¹¹. Following identification of conserved motifs¹² termed “helicase signature motifs”, and subsequent classification into superfamilies^{1, 12}, the amino acid sequences containing such motifs were ascribed to encode for “helicases”. However, only a fraction of enzymes identified through bioinformatics display bona fide strand separation activity. The most fundamental property of all enzymes possessing the helicase signature motifs is their ability to move on nucleic acid lattices (DNA or RNA) with a distinct polarity. Signature motifs are most conserved elements in the motor core of a molecular motor. They contain amino acids responsible for coupling chemical energy derived from nucleoside triphosphate hydrolysis to directional translocation along DNA or RNA molecule with or without unwinding of the helix (reviewed in^{1, 13}). What happens when a translocating helicase encounters nucleic acid duplex and to what extent translocation and duplex separation activities are interlinked remain controversial¹³. While some helicases may be capable of duplex unwinding just as a consequence of translocation, others were proposed to employ a completely different mechanism thus distinguishing the two activities (reviewed in ^{13, 14}). Moreover, when acting in the context of their endogenous molecular machineries, structurally related “helicases” are known to display marked preference for a particular translocation lattice, which could be single- or double-stranded DNA or RNA, nucleic acid of particular configuration or protein-nucleic acid complex. The largest and the most structurally diverse helicase superfamily II (SF2) is comprised of DNA and RNA helicases, motor components of chromatin remodeling machines and even polypeptide translocases. Although bioinformatics and biochemical studies may provide some clues regarding the preferred lattice for these helicases/translocases, in the absence of information on the cellular role and interacting partners for a particular enzyme, these clues may lead to erroneous conclusions. NS3 helicase from Hepatitis C virus, for example, is involved in replication of viral dsRNA genome in the cell cytoplasm. It is capable, however, of efficient unwinding of dsDNA and possesses remarkable structural similarity to bona fide DNA helicases of RecQ family ¹⁵. Despite its role in peptide export, another structurally related enzyme, SecA, is also capable of unwinding RNA duplexes¹⁶. Our recent observation that a SF2 helicase, XPD can move along protein-coated ssDNA further expands the repertoire of available translocation lattices¹⁷ (Figure 2).

Figure 2. Obstacle bypass by XPD helicase.

(a) Structure and domain organization of XPD helicase (PDB: 3crv³¹). HD1 and HD2 and helicase domains 1 and 2 respectively. These two domains form SF2 helicase motor core. Arch and FeS are the two family-specific auxiliary domains. (b) One possible configuration of ssDNA within translocating XPD helicase. After the primary binding site (indicated by yellow arrows) ssDNA passes through the hole between Arch and FeS domains into the secondary DNA binding site in the FeS domain. Green arrow illustrates energy transfer from Cy3 fluorophore to the FeS cluster resulting in distance-dependent Cy3 quenching. (c) Alternative configuration of the ssDNA-XPD complex, which does not involve ssDNA passage through the hole between two auxiliary domains. (d) Single-molecule XPD translocation assay exploits FeS-dependent fluorescence quenching (left panel). Representative fluorescence trajectory (middle panel) shows that XPD translocation along ssDNA decorated with Cy3 dye at the 3′-end results in gradual quenching of Cy3 intensity followed by its full recovery when XPD dissociates from the substrate. Distribution of individual translocation rates binned in 5% fluorescence change per second intervals (right panel). (e) Single-molecule experiment carried out in the presence of RPA2 protein. (f) Simultaneous visualization of XPD translocation (monitored by following FeS-dependent quenching of Cy3 dye) and RPA2 binding (monitored by following FRET between Cy3 dye incorporated at the 3′-end of ssDNA and Cy5 dye located at the N-terminus of RPA2 protein). Fluorescence of the Cy3 and Cy5 dyes is shown in green and red, respectively. A synergistic quenching followed by the recovery of Cy3 and Cy5 fluorescence suggested that XPD helicase can bypass bound RPA without dissociating it from the lattice. (g.) Translocation of XPD was observed through quenching of directly excited Cy5-RPA2. Gradual decrease and increase in Cy5 intensity reflected XPD approaching and moving away from Cy5 labeled RPA2. We interpreted this quenching pattern as translocation of XPD helicase over bound RPA2. Data shown in this figure are adapted from Honda, et al¹⁷.

Using total internal reflection fluorescence microscopy we monitored translocation by individual molecules of XPD, an FeS cluster-containing helicase (Figure 2a), on ssDNA by following iron-mediated quenching of Cy3 fluorophore incorporated into a surface-tethered ssDNA molecule (Figure 2d). A marked decrease in XPD translocation velocity in the presence of one of its cognate SSBs, RPA2, (Figure 2e) prompted us to probe whether this slower translocation is due to competition or co-existence. To visualize the two proteins simultaneously, RPA2 was labeled with Cy5, another fluorophore of a different color, so that the presence of RPA2 on the DNA could be monitored either through fluorescence resonance energy transfer (FRET) from Cy3 on DNA (Figure 2f) or via direct excitation by the laser light (Figure 2g). At the same time, XPD helicase was tracked by following iron-dependent quenching of Cy3 on DNA or Cy5 on RPA2. Simultaneous visualization of both the helicase and its obstacle brought us to the conclusion that this helicase can translocate on the protein-coated ssDNA without dismantling the protein-nucleic acid complex.

Forces exerted by the motor activity of a helicase may be directly or indirectly coupled to the disassembly of protein-nucleic acid complexes. Motors that strongly interact with their translocation lattices can remove obstacles bound to the nucleic acid. Some helicases can disrupt interactions as strong as those between biotin moiety incorporated into DNA and streptavidin bound to the biotin, which is one of the tightest protein-ligand interactions observed in nature. In fact, directional translocation of a helicase along nucleic acid track is commonly probed by following its ability to facilitate streptavidin dissociation ¹⁸. Often, a helicase can both unwind DNA duplex and disassemble the protein-nucleic acid complexes. For some helicases only one of these activities may be physiologically relevant. In other cases, both activities may be utilized in vivo either simultaneously or under different circumstances. It is important to distinguish, therefore, whether the ability of a helicase to displace proteins from nucleic acids is a physiologically relevant function or a mere consequence of its directional movement. Making such a distinction is not a trivial task and is often requires combining the in vivo data with carefully crafted in vitro analyses.

To determine whether strand separation, translocation or protein displacement activity of a helicase is important for its cellular function one needs to examine how interacting proteins and other components of the same pathway affect these activities. This is because helicases in the cell are rarely exposed to naked nucleic acids that are free of bound proteins. The proteins commonly found associated with the cellular DNA or RNA processing intermediates may compete with the helicase for the lattice or may target the helicase to its cognate DNA processing intermediate. Even after the helicase is delivered to the relevant protein-DNA complex, it has to bypass proteins bound to the ssDNA or displace them in order to advance along the DNA molecule.

Helicases and ssDNA binding proteins: displace, avoid or bypass?

Molecular machines orchestrating most of the cellular DNA transactions require ssDNA intermediates to readout the information. Being susceptible to chemical and enzymatic degradation, ssDNA rarely exists in the cell in its bare form. Instead, it is usually found in complex with the ssDNA binding proteins (recently reviewed in¹⁹). Single-stranded DNA binding proteins (SSBs) form dynamic complexes on ssDNA to protect it from degradation, modification, formation of unwanted secondary structures and to prevent undesired or untimely transactions. SSBs from all species bind ssDNA with high affinity utilizing oligonucleotide/oligosaccharide-binding (OB) folds and generally do not display specificity for particular sequences. SSBs are sometimes referred to as helix destabilization proteins reflecting their propensity to melt regions of secondary structure present in ssDNA.

Because the structurally and functionally conserved OB folds do not posses high degree of sequence conservation, it is often difficult to identify these proteins through bioinformatics analysis. Traditionally ssDNA binding proteins were designated as SSB proteins in bacteria ²⁰ and RPAs (replication protein A) in eukaryotes ²¹. Growing evidence suggests that all organisms contain not just generic SSB or RPA, but also more specialized proteins that contain the OB modules incorporated among other functional domains. For example, bacteria contain PriB which targets PriA helicase to the collapsed replication forks during reassembly of the replication machinery. Furthermore, human cells contain diverse proteins such as the telomere specific protein Pot1²², hSSB1 protein involved in recombinational DNA repair²³, and the RMI (RecQ-mediated genome instability) complex that functions together with BLM helicase²⁴.

SSBs also interact physically and functionally with many proteins involved in DNA replication, recombination and repair^{25, 26}. These interactions are important for coordinating timely recruitment of DNA maintenance machineries to the chromosomal sites requiring attention. Upon recruitment, these machines are expected to replace respective SSBs on ssDNA. If the helicase specifically recognizes junction between ssDNA and dsDNA (for example, if it contains additional DNA binding site specific for duplex DNA or for a particular structural elements(s) found at the junction), one can envision that its interaction with a specific structural feature on DNA and SSB located closest to this feature will facilitate helicase binding to the junction (schematically depicted in figure 3a). Interaction of this helicase with SSB bound at a distance from the junction should not produce similar effect because the helicase will have to compete with SSB there for binding to ssDNA

Figure 3. Scenarios for co-existence of a helicase and ssDNA binding protein on the DNA substrate.

(a) Targeting of a helicase to an ssDNA-dsDNA junction. (b) A helicase can step-over the region of ssDNA wrapped around an SSB. (c) Physical interaction between helicase and SSB may facilitate bypass and prevent dissociation of SSB. (d) Helicase can bypass bound SSB without displacing it from ssDNA if the two proteins interact with different features on DNA (for example, with phosphodiester backbone and nitrogen bases).

One can also envision how physical interaction between an SSB and a helicase can facilitate helicase translocation past a DNA-bound SSB without completely removing it from the lattice. If the helicase releases part of its contacts with ssDNA upstream of the bound obstacle and then reforms these contacts downstream, it can potentially “step over” the SSB obstructing its progress (Figure 3b). Since such a stepping process would transiently reduce the number of interactions between the helicase and its lattice, it may drastically increase probability of the helicase to dissociate (or, in other words, will decrease its processivity). Transient interactions with SSB during such a step over may compensate for the lost contacts preventing complete dissociation of the helicase. A mechanism similar to this may be responsible, for example, for the continuous re-initiation of fork DNA unwinding by human BLM helicase in the presence of human RPA protein²⁷. The inverse of this situation seems likely to be a more general mechanism for bypassing of bound SSBs (Figure 3c): translocating helicase may form complex with the downstream SSB causing its partial or complete dissociation from the lattice; the helicase can then wedge into place previously occupied by SSB, after which the SSB can be transferred behind the helicase. This model is especially attractive considering the large ssDNA binding site of eukaryotic RPA (binding of human RPA to ssDNA occludes up to 30 nucleotides). Bacterial SSB also has a large ssDNA binding site. In one of its binding modes the tetramer of E. coli SSB wraps about 65 nucleotides of ssDNA. In such ‘closed’ wrapping mode where a long ssDNA enters and exits the protein at about the same location, a bypass strategy depicted in Figure 3b is more plausible. The recently discovered ability of E. coli SSB to diffuse on ssDNA adds another layer of possible complexity on how SSB proteins with multiple DNA binding sites may dynamically coordinate with helicases and other proteins that bind to ssDNA ^{28, 29}.

Sometimes a helicase can sneak by bound SSB even in the absence of physical interaction between the two proteins (Figure 2d). Recently we have demonstrated that archaeal XPD helicase can move along ssDNA coated with one of its cognate ssDNA binding proteins, RPA2, without dissociating from the lattice and without displacing RPA2¹⁷. The two proteins interact functionally. RPA2 targets XPD helicase to the ssDNA-dsDNA junction and may play an important role in helicase activity of XPD by destabilizing duplex ahead of the helicase. However, no physical interaction has been observed between XPD and RPA2³⁰. Very small ssDNA binding site (5 nucleotides) of monomeric RPA2 consisting of a single OB fold precludes possibility that it breaks and reforms partial contacts with ssDNA. Because translocation of XPD bypassing RPA2 proceeds gradually at a constant speed and without sudden “jumps” relative to both ssDNA and RPA2¹⁷ (representative translocation trajectories are shown in Figure 2d&e), we also excluded possibility that it steps over RPA2 by partially breaking contacts between helicase and ssDNA. Instead, we envisioned that XPD and RPA2 can coexist on the same ssDNA. How can this be achieved?

Structures of both archaeal XPD helicase^31–33 (Figure 2a) and archaeal SSB protein, which has similar configuration to that of RPA2³⁴(inset in figure 2e), have been solved in the absence of DNA. Lack of structural information on binary complexes makes it unclear how the two proteins do interact with DNA. The OB fold of RPA2 is expected to be similar to that of eukaryotic ssDNA binding proteins and to interact with nucleobases³⁵. XPD likely tracks along the phosphodiester backbone of DNA making minimal contacts, if any, with the bases as do other SF2 helicases¹. Observation that XPD-mediated dsDNA unwinding is not stalled by the lesions affecting bases³⁶ is suggestive of this translocation mode. An arrangement where the helicase contacts the DNA backbone while the SSB interacts with bases would allow both proteins to be bound simultaneously to the same region of ssDNA and would allow the helicase to move by the SSB.

In addition to the primary ssDNA binding site spanning the helicase motor core, XPD contains a secondary ssDNA binding site in the FeS domain³⁷. This site was proposed to accommodate the displaced strand during unwinding of forked DNA substrates. In the absence of ssDNA/dsDNA junction, ssDNA within the ssDNA-XPD complex may be distributed between the two binding sites. It is not known what types of interactions are responsible for binding of ssDNA in the secondary DNA binding site or whether the absence of binding to this site would impede XPD translocation on ssDNA. The path connecting the two sites and the path of DNA within XPD predicted based on the crystal structures of apo-protein^{31, 38} lays through the narrow hole between FeS domain and another auxiliary domain (Arch domain) inserted into the motor core of XPD (Figure 2b). In the apo-structure, this hole is topologically closed and provides an opening of only 15-20Å wide, which is too narrow to thread through the ssDNA plus OB fold (about 30Å wide. If ssDNA is indeed threaded through the opening between Arch and FeS domains, which is likely but not necessarily the only possible path, a conformational change that forces the two domains apart would be a prerequisite for binding of XPD to ssDNA. This is because XPD binds to internal sites on ssDNA¹⁷, not to single stranded end, to unwind bubble structure in vitro³⁶. Binding to the internal sites on the bubble are also important for its in vivo function which should entail binding to the single-stranded regions of repair or transcription bubbles³⁹. This binding mode is inconsistent with ssDNA being threaded through the hole between FeS and Arch domains of XPD starting from the terminus. Therefore, in order to bind ssDNA in the hole between FeS and Arch domains as predicted based on crystal structures, XPD has to undergo a significant conformational change moving the FeS and Arch domains apart. This conformational alteration may even provide opening large enough to accommodate ssDNA-RPA2 complex. The requirement for the conformational change may explain the slower translocation rate on RPA2-coated ssDNA relative to XPD movement on the protein free lattice.

It is also formally possible that ssDNA takes a completely different path around the FeS domain (figure 2c). If this is indeed the case, ssDNA has to take a rather sharp turn when bound to XPD. Bending of ssDNA by bound RPA2 may assist in pre-configuring ssDNA for binding in the bent conformation reflecting slightly increased affinity of XPD helicase for RPA2-coated DNA^{17, 30}. New structural or mechanistic information will be necessary to resolve how XPD interacts with protein free DNA and with DNA-SSB complexes.

The number and ubiquity of recently indentified eukaryotic SSBs containing just one or two OB folds, resembling RPA2 organization²⁵ and the large number of SF2 helicases involved in DNA repair and maintenance raises the question whether the bypassing an SSB protein by a translocating helicase could represent a biological activity important for the cellular function of these molecular motors possibly no less important than removal of the stationary blocks.

Encounter by the helicase of bound SSB may also facilitate SSB dissociation followed by transfer of released ssDNA to other proteins or to complementary DNA strand. SMARCAL1 helicase, for example, uses its interaction with RPA protein to facilitate it removal from ssDNA during helicase-mediated strand annealing⁴⁰ (Figure 1g).

Another class of ssDNA binding proteins that physically and functionally interact with DNA helicases is comprised of RecA-like strand exchange proteins. Formation of nucleoprotein filaments by bacterial RecA, archaeal RadA and eukaryotic Rad51 proteins is the central step in homologous recombination and recombinational DNA repair of broken chromosomes (reviewed in⁴¹). In the presence of ATP these proteins form recombinationally active extended helical filaments on ssDNA. Translocation activity of several recombinational helicases and translocases is directed to disassembly of such filaments from ssDNA or to recycling of Rad51 protein remaining bound to dsDNA after completion of strand exchange reaction. Due to the nature of Rad51(RecA) filaments which form extensive contacts with DNA sequestering it in the middle of the complex⁴², it is unlikely that any motor protein can bypass bound RecA or Rad51. In contrast, several DNA helicases are known to displace Rad51 from DNA. Displacement of the strand-exchange protein can be achieved by exerting force and competing with it for binding as observed for UvrD helicase that dismantles preformed RecA or Rad51 nucleoprotein filaments^{43, 44} or for E. coli Rep helicase that prevents RecA filament formation using its repetitive translocation activity ⁴⁵. Anti-recombinase activities of helicases may also require specific protein-protein interactions. Interaction between Rad54 translocase and Rad51 is important for post-recombinational recycling of Rad51⁴⁶. Yeast Srs2 helicase and Rad51 protein provide another example of such specific interaction. Srs2 helicase has long been known to act as an anti-recombinase^{47, 48}. However, the mechanism of its action has been questioned until very recently when it was unambiguously shown to antagonize homologous recombination by disassembling the presynaptic complexes formed by Rad51 recombinase⁴³. Moreover, Srs2 does not merely plow away bound Rad51 by exerting force during translocation or by competing with for binding to the ssDNA. Instead, specific interaction between the two proteins results is acceleration of ATP hydrolysis by bound Rad51 leading to its dissociation⁴³. Intuitively, the most important parameters that define helicase ability to bypass or displace the obstacles are the shape, strength and number of contacts between the motor, its lattice and the obstacle.

Most of the replicative helicases (for example bacterial DnaB, viral E1 or archaeal and eukaryotic MCM helicases) are ring-shaped hexamers (reviewed in¹). These enzymes are highly processive and very fast. During replication initiation, these helicases are recruited to the replication origin, where they assemble into the functional hexameric helicase rings, which then translocate away from the origin and unwind the double helix ahead of DNA polymerase and the rest of the replication machinery. The ring morphology defines helicase processivity and is imperative to complete and timely replication of the genome. In addition, the ring shape would not allow the replicative helicase to bypass obstacles on DNA before they are cleared off the lattice. When such a helicase encounters DNA damage or stalled protein-nucleic acid complex (such as backtracked RNA polymerase), it halts forward progression causing the replication fork to stall or collapse.

The strength of the interaction is important when helicase needs to exert force to displace an obstacle. SF1 helicases UvrD and Srs2, which are capable of exerting sufficiently large forces to dismantle RecA and Rad51 nucleoprotein filaments, form extensive contacts with bases on ssDNA. Although the strength of interactions between the two motor domain in SF1 helicases and ssDNA lattice changes during different steps of mechano-chemical cycle, both ATP-bound and empty form of these helicases remain firmly associated with the lattice^{49, 50}. In contrast, SF2 helicase NS3, which contacts DNA primarily through the interactions involving phosphodiester backbone uses a Brownian ratchet mechanism, whereby it switches between ATP-free tightly bound state and more loose ATP-bound state⁵¹. It is not surprising, therefore, that heterologous bacterial SSB greatly improves NS3 ability to unwind DNA duplexes by virtue of binding behind the NS3 helicase and preventing backsliding⁵². Similar mode of translocation expected for XPD helicase may help it bypass modified DNA bases³⁶ and bound SSBs¹⁷, because this activity requires flexibility and limited contacts with the lattice rather than ability to exert force.

Human Rad3 family helicases and ssDNA binding proteins: is there a specific match?

In addition to XPD helicase, which is conserved in eukaryotes, bacteria and archaea, some eukaryotes possess a number of FeS containing helicases structurally similar to XPD. These enzymes form a distinct group called the Rad3 family named after the yeast homolog of XPD (reviewed in⁵³). All studied Rad3 family enzymes are bona fide helicases involved in various aspects of DNA repair and maintenance of genomic integrity. Human Rad3 helicases include XPD, which participates in nucleotide excision repair and RNA transcription as an integral part of TFIIH complex; Bach1 (also known as FancJ) helicase, which is involved recombinational repair of inter-strand DNA cross-links; Rtel1 (regulator of telomere maintenance), which controls homologous recombination by unwinding toxic recombinational intermediates; and ChlR1, which plays role in sister chromatid cohesion. These helicases contain the FeS and Arch domains and are likely to utilize DNA translocation and unwinding mechanisms similar to that of XPD. Human XPD and Bach1 helicases were shown to interact with human RPA protein. It is likely that activity of other Rad3 helicases is also affected by interactions with RPA or other human SSBs (such as SSB1, Pot1 and RMI). It is also foreseeable that some of these helicases may interact physically or functionally with more than one SSB, and that these selective interactions may target the helicase to its cognate intermediate and activate the helicase for desired DNA maintenance task.

Many mammalian helicases associate with multiple distinct DNA maintenance machineries, which may utilize either DNA translocation or strand separation activities of the helicase. Ability to recognize a specific nucleoprotein complex and to navigate through the molecular traffic jam is likely to be important for allowing DNA translocating motors adapt to a diverse set of cellular functions they are expected to perform.