Structural insights into regulation of CNNM-TRPM7 divalent cation uptake by the small GTPase ARL15

Curation statements for this article:
  • Curated by eLife

    eLife logo

    eLife assessment

    In this potentially important study, Mahbub and colleagues examine how the small GTPase ARL15 regulates ion transport. Using a complementary array of techniques, the authors gathered solid evidence for the binding of ARL15 to CNNM proteins, resulting in a proposal how this may affect the function of the TRPM7 channel. Additional experiments are required to fully substantiate the claims.

This article has been Reviewed by the following groups

Read the full article See related articles

Abstract

Cystathionine-β-synthase (CBS)-pair domain divalent metal cation transport mediators (CNNMs) are an evolutionarily conserved family of magnesium transporters. They promote efflux of Mg 2+ ions on their own and influx of divalent cations when expressed with the transient receptor potential ion channel subfamily M member 7 (TRPM7). Recently, ADP-ribosylation factor-like GTPase 15 (ARL15) has been identified as CNNM-binding partner and an inhibitor of divalent cation influx by TRPM7. Here, we characterize ARL15 as a GTP and CNNM-binding protein and demonstrate that ARL15 also inhibits CNNM2 Mg 2+ efflux. The crystal structure of a complex between ARL15 and CNNM2 CBS-pair domain reveals the molecular basis for binding and allowed the identification of mutations that specifically block binding. A binding deficient ARL15 mutant, R95A, failed to inhibit CNNM and TRPM7 transport of Mg 2+ and Zn 2+ ions. Structural analysis and binding experiments with phosphatase of regenerating liver 2 (PRL2 or PTP4A2) showed that ARL15 and PRLs compete for binding CNNM to coordinate regulation of ion transport by CNNM and TRPM7.

Article activity feed

  1. Author Response

    Reviewer #2 (Public Review):

    Mahbub et al further elucidate the structural and functional consequences of the ARL15-CNNM2 interaction for divalent cation transport. They show that ARL15 has low GTP binding affinity and could not detect GTPase activity, questioning whether ARL15 functions as a GTPase. Although the interaction of ARL15 and CNNMs has been demonstrated by multiple groups before, this study addresses some of the key questions that are central within the TRPM-CNNM-PRL-ARL15 field. Particularly, the authors have identified residues in both ARL15 and CNNM proteins which are required for their binding to one another. In addition, they have also illustrated how PRL proteins compete with ARL15 for their binding to CNNMs. Lastly, the functional consequences of ARL15 binding to CNNMs are shown by TRPM7-mediated Zn2+ transport assays.

    We thank the reviewer for the many positive comments.

    However, the current dataset also comes with limitations. Previous studies demonstrated that PRLs interact with the CBS domains of CNNMs and lock them in their so-called "flat" confirmation. It remains unclear how ARL15 affects the structure of the CBS domains, especially in the presence of ATP. The subcellular localisation of these interactions has not been examined. Moreover, the consequences of ARL15 on TRPM7 activity are not completely elucidated. It remains unclear whether this functional effect is CNNM-dependent. Moreover, how the zinc uptakes translate to other divalent ion transport, such as magnesium, has not been examined. These questions should be answered to confirm the model as presented in Figure 7.

    We agree that CBS-pair domain dimerization is important. Structural studies of a prokaryotic CNNM homolog from our group showed large conformational changes in an ATP-binding mutant (Chen et al., Nat Comm, 2021).

    While most crystal structure of PRL-CNNM complexes do indeed show the flat conformation, it is unclear if that is a consequence of crystal packing or PRL binding. We do not see an effect of ATP on PRL binding affinity. The CBS-pair domain dimerization interface appears to be very adaptable; our recent structure of PRL-CNNM proteins from flies shows a completely different dimerization interface (Fakih et al, JBC, 2023).

    In contrast, the ARL15-CNNM interaction is affected by ATP. As suggested by the reviewer, we have examined ARL15 binding to a CNNM2 mutant (T568I) that is unable to bind ATP. These results confirm the roughly two-fold improvement in affinity is due to ATP binding to the CNNM2 CBS-pair domain and resulting conformational changes.

    As requested by all the reviewers, we have added experiments to Figure 7 that investigate the effect of ARL15 on Mg2+ transport.

  2. eLife assessment

    In this potentially important study, Mahbub and colleagues examine how the small GTPase ARL15 regulates ion transport. Using a complementary array of techniques, the authors gathered solid evidence for the binding of ARL15 to CNNM proteins, resulting in a proposal how this may affect the function of the TRPM7 channel. Additional experiments are required to fully substantiate the claims.

  3. Reviewer #1 (Public Review):

    ARL15 forms a complex with the TRPM7 channel and CNNM transporters and is involved in the regulation of the TRPM7 function. To understand the regulatory mechanism, the authors performed biochemical and structural characterizations. In this work, they determined the crystal structure of ARL15 in complex with CNNM2 CBS domain, performed the mutational analysis based on the structure, and successfully revealed the binding mechanism between ARL15 and CNNM.

    However, the detailed mechanism of TRPM7 inhibition by ARL15 remains unclear because the structure of TRPM7 in complex with ARL15 is still unknown. Furthermore, despite the structure determination of ARL15 in complex with CNNM, the effect of ARL15 on CNNM function is still unclear.

    Nevertheless, the structural information on the ARL15-CNNM complex provided by the authors is valuable for the related research field, and the structure-based CNNM mutants specifically targeting disruption of binding to either ARL15 or PRL would also be useful.

  4. Reviewer #2 (Public Review):

    Mahbub et al further elucidate the structural and functional consequences of the ARL15-CNNM2 interaction for divalent cation transport. They show that ARL15 has low GTP binding affinity and could not detect GTPase activity, questioning whether ARL15 functions as a GTPase. Although the interaction of ARL15 and CNNMs has been demonstrated by multiple groups before, this study addresses some of the key questions that are central within the TRPM-CNNM-PRL-ARL15 field. Particularly, the authors have identified residues in both ARL15 and CNNM proteins which are required for their binding to one another. In addition, they have also illustrated how PRL proteins compete with ARL15 for their binding to CNNMs. Lastly, the functional consequences of ARL15 binding to CNNMs are shown by TRPM7-mediated Zn2+ transport assays.

    However, the current dataset also comes with limitations. Previous studies demonstrated that PRLs interact with the CBS domains of CNNMs and lock them in their so-called "flat" confirmation. It remains unclear how ARL15 affects the structure of the CBS domains, especially in the presence of ATP. The subcellular localisation of these interactions has not been examined. Moreover, the consequences of ARL15 on TRPM7 activity are not completely elucidated. It remains unclear whether this functional effect is CNNM-dependent. Moreover, how the zinc uptakes translate to other divalent ion transport, such as magnesium, has not been examined. These questions should be answered to confirm the model as presented in Figure 7.

  5. Reviewer #3 (Public Review):

    The authors studied the interaction between Arl15 and CNNMs using various biochemical and biophysical approaches. Significantly, they solved the crystal structure of Arl15 and the CBS-pair domain of CNNM2 and demonstrated that PRLs and Arl15 could compete for binding to CNNMs. The study should advance our understanding of how cellular divalent ions are regulated via Arl15, CNNMs, and TRPM7, although some issues regarding the guanine nucleotide-binding of Arl15 need to be addressed.