The (elusive) role of the SMC5/6 complex

KEYWORDS:

• aging
• anaphase bridges
• BLM
• cancer
• micronuclei
• MMS21
• NSMCE2
• recombination
• SCE
• SMC5/6

This article refers to:

SMC5/6 is the most mysterious of SMC complexes, as exemplified by the fact that in contrast to SMC1/3 (cohesin) or SMC2/4 (condensin), it still lacks a generic name. The complex was discovered in fission yeast, by noticing that Rad18 and Spr18 formed a heterodimer resembling SMC complexes.¹ In addition to SMC5 and SMC6, the complex in mammals contains 4 additional proteins, known as non-SMC elements (NSMCE1-4). Besides the ATPase of SMC5/6, the only known enzymatic activities are a SUMO ligase in NSMCE2 (Mms21 in yeast) and an ubiquitin ligase in NSMCE1. Since these were the only recognizable activities in the complex, it led to the assumption that they were “the” mechanism by which the complex worked. However, some obvious facts pointed otherwise, such as that SUMO ligase-deficient yeast strains of Mms21 were viable, and deletion mutants were not.

The original observation was that Smc5/6 mutants were sensitive to certain genotoxic drugs such as alkylating agents. The logical way to start was by assuming that the complex participated in DNA repair. Consistently, early studies showed recruitment of the complex to a persistent nuclease-induced double-stranded break (DSB). Likewise, works in human cells using RNAi have reported deficient DNA repair and lower rates of HR measured by sister chromatid exchange (SCE). However, many pieces of evidence challenge that the role of SMC5/6 is on DNA repair. First, SCE rates are enhanced (rather than reduced) in SMC5/6 complex mutants from all organisms tested. In fact, the mms21 mutant was discovered as showing a 23-fold increase in spontaneous recombination. Second, NSMCE2 fails to form ionizing radiation (IR)-induced foci in mitotic cells neither it localizes to meiotic DSBs during meiosis.² Third, repair of meiotic DSB in yeast or IR-induced DSBs in mammalian cells are largely unaffected by SMC5/6 deficiencies. Finally, all members of the complex are essential in yeast, but HR is not. All of the above support that the role of the SMC5/6 complex is not on promoting DNA repair.

If not DNA repair, what could it be? What is obvious is that the complex is essential to facilitate segregation and to limit recombination; but how? One interesting idea is that SMC5/6 could “dissolve” joint molecules (JM) that are formed between 2 molcules of DNA, similar to the function proposed for BLM in mammals. The accumulation of JMs can explain the 2 hallmarks of SMC5/6 mutants, segregation problems and increased recombination, since JM resolution is done by their cleavage through structure-specific nucleases and subsequent recombination. But what are these JMs?

On one hand, a fraction of them might arise as byproducts of DSB-repair, since Rad51 deletion abrogates the accumulation of JMs in Smc5/6 mutants.^3,4 However, some of these structures could instead arise from replication forks. Supporting this view, Smc5/6 mutants enter anaphase before completing DNA replication.⁵ Whereas overall DNA replication proceeds normally in NSMCE2-deficient mouse cells,² replication might not be fully completed at certain sites, which would generate JMs and hamper segregation. But why would SMC5/6 be needed for only some breaks or replication forks? I believe that the complex is particularly relevant for repeated DNA sequences.

First, genomewide-mapping studies in yeast showed that the complex localizes preferentially to repeated DNA sequences. Moreover, Smc5/6 mutants show a particular deficiency in the segregation of rDNA,³ which is the biggest repeat in yeast. Likewise, NSMCE2 foci invariably form around pericentromeric DNA, the largest repeat of the murine genome. Regarding foci, from all reagents tested catalytic inhibitors of TopoII induce the greatest accumulation of NSMCE2,² consistent with the synthetic lethality described for Smc5/6 and topoisomerase mutants.⁶ How or if this SMC5/6-Topo interaction relates preferentially to DNA repeats is not known. Nevertheless, unraveling this 3-way connection (SMC5/6-Topo-repeats) seems key to understand the function of the complex.

The final question is how would the complex work. As mentioned, SUMOylation-deficient Mms21 strains are alive, whereas deletion mutants are not. Similarly, mice lacking NSMCE2 SUMO ligase activity have a normal lifespan. Hence, SUMOylation might only play a fine-tuning role. Whereas some residual ligase activity could exist in these mice, I find it rather unlikely, given that the mutation alters one of the metal-coordinating Cys from the catalytic domain. Moreover, recent work showed that SUMOylation is a highly redundant process, which works by modifying many components of a given pathway with the role of each SUMOylation being marginal. If not SUMOylation, what can it be? The ubiquitin ligase of NSMCE1 is an obvious candidate. However, my opinion is that the role of the complex might be more related to its SMC architecture than to the recruitment or modification of other proteins. It could, for instance, concentrate around JMs formed at repeated sequences in a manner that would help their “presentation” to other complexes involved in their dissolution.

Regardless of the mystery that surrounds SMC5/6, our recent work has shown that it is essential to suppress cancer and aging in mice.² Moreover, NSMCE2 hypomorphism in humans also leads to a genomic instability syndrome that limits lifespan.⁷ Hence, understanding the function of the SMC5/6 complex is still an important task, which beyond its academic interest might yield unanticipated insights with potential applications to human health.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.