Arsenic and SUMO wrestling in protein modification

This article refers to:

Arsenic-induced sumoylation of Mus81 is involved in regulating genomic stability

It is 20 years since the first discovery of small ubiquitin-like protein by Shen et al.¹ in a yeast-two hybrid system using human Rad52 as a bait and Boddy et al.² in a screening of human B cell cDNA library for PML-interacting clones. Simultaneously or shortly after this discovery, Mahajan et al.³ named the same protein that modifies nuclear pore complex-associating protein RanGAP1 as small ubiquitin-related modifier 1. The abbreviation of this nomenclature, SUMO1, is somehow coinciding to the traditional Japanese sport, SUMO wrestling, in which 2 big-sized men try to force the opponent out of the circular ring. Since then, 3 additional SUMO proteins, SUMO2, SUMO3 and SUMO4, had been identified in mammalian cells. All of these SUMO proteins are able to covalently attach to the lysine (K) residues in the consensus sequence ψKXE of specific target proteins, where ψ represents a hydrophobic amino acid and X represents any amino acid. In general, SUMO1 is most likely to form monosumoylation, whereas SUMO2 and SUMO3 are capable of forming poly SUMO chain on the target proteins.

Since several oncoproteins and tumor suppressors had been known to be sumoylated, it is a powerful impetus to determine whether genetic mutations in SUMO and/or its regulatory pathways are involved in cancer development. Unexpectedly, accumulating evidence suggested that the overall genetic integrity of the core components in the SUMO and its related pathway are not compromised in cancers, except SNPs in SUMO E2 conjugating enzyme UBC9 gene and chromosomal translocations in 2 desumoylases.⁴ Thus, it is very likely that changes in the dynamics of sumoylation may be responsible for the pathogenesis of tumors.

As a group I carcinogen, environmental arsenic had been linked to several human cancers in several large scale epidemiological studies. Meanwhile, in animal studies, arsenic is a well-established teratogen. Despite more than hundred proteins have been shown to be SUMO-modified, only few proteins are studied for sumoylation in response to arsenic. The most studied protein for arsenic-induced sumoylation is PML-RARa, the primary culprit for APL pathogenesis.⁵ Arsenic induces initial sumoylation of PML-RARa, followed by RNF4-mediated ubiquitination and degradation of this protein. Obviously, this unraveled the therapeutic efficacy, rather than the carcinogenic role of arsenic.

The studies by Hu et al.⁶ in this volume of Cell Cycle demonstrated that arsenic is highly capable of inducing SUMO-conjugation of several proteins important for DNA repair, cell cycle and transcriptional regulation. Using HeLa cells as a model, they provided evidence that arsenic increases protein sumoylation in manners of both dose- and time-dependent. The same conclusion was achieved in experiments using additional cell lines that are either PML-RARa negative or positive, indicating that such an induction of protein sumoylation by arsenic is not limited to a particular cell type, but rather, a general phenomenon. To define the nature of arsenic-induced sumoylation, they designed an elegant experimental strategy using the cells expressing His6-SUMO2 and treated with arsenic, followed by affinity separation by Ni-IDA resin and the subsequent proteomics analysis, which showed that the arsenic induces sumoylation of Mus81, UVBL, p21, ARID1a, and others. Mus81 is a member of the XPF (ERCC4) family of endonucleases that forms a complex with EME1 during Holliday junction resolution of DNA homologous repair. Additional biochemistry studies suggested that SUMO E3 ligase PIAS3 can promote Mus81 sumoylation at lysines 10 and 524 residues. By ectopic expression of the WT and lysine-mutated Mus81 in the cells tested, a decreased stability of the mutated Mus81 in response to arsenic was noted, implying that sumoylation may stabilize Mus81 protein.

Although the authors tried to answer the ultimate question on the functional consequence of Mus81 sumoylation, additional tests are certainly warranted. At this moment, a conclusion cannot be reached based on the different DNA damage responses between arsenic- and chromium(VI)-treated cells expressing the WT or sumoylation resistant Mus81 (Figs 6c and 6d in Hu et al.⁶). Furthermore, it is also difficult to translate such an information into the potential mechanism of arsenic-induced carcinogenesis. However, the data showing that arsenic induces sumoylation of protein other than Mus81 by the authors clearly point to the direction that an altered dynamic of SUMO modifications on cellular proteins is one of the key contributing factors to arsenic carcinogenesis.

Disclosure of potential conflicts of interest

No conflict of interest is disclosed.

Funding

The work in FC's laboratory is supported by NIH R01 ES020137 and partially supported by NIH P30 ES020957.