LINC complexes, consisting of SUN proteins in the inner nuclear membrane and KASH proteins in the outer nuclear membrane, transfer forces between the cytoskeleton and the nucleoskeleton. One role of LINC complexes is to mediate nuclear positioning. The canonical C. elegans SUN protein UNC-84 interacts with two different KASH proteins-UNC-83 (a functional homolog of mammalian Nesprin-4 and KASH5) to move nuclei via microtubule motors and ANC-1 (the ortholog of the giant mammalian Nesprin-1 and -2) to anchor nuclei to actin. It is not known how UNC-84 chooses between binding UNC-83 to move nuclei or binding ANC-1 to anchor nuclei. Based on structural data of human SUN-KASH complexes, we first hypothesized that key conserved residues in the SUN domain of UNC-84 are required to interact with the C-termini of KASH proteins. Both UNC-84(S1034E) and UNC-84(C994E) failed to recruit UNC-83 to the nuclear envelope and nuclear migration was blocked. Second, the structure predicted that the length of the C-terminus of KASH proteins is limited by the size of the binding pocket in the SUN domain. The addition of a single alanine to the end of UNC-83 completely blocked UNC-83 localization and nuclear migration in vivo. Finally, the structure identified an intermolecular di-sulfide bond between conserved cysteine residues in SUN and KASH proteins. The conserved cysteine at position -23 of KASH proteins is present in ANC-1, but not in UNC-83. This led us to hypothesize that cysteine bonds are dispensable for nuclear migration, but required for nuclear anchorage. To test this hypothesis, we used CRISPR/Cas9-mediated genome editing to mutate conserved cysteines in SUN and KASH proteins. Hypodermal nuclei expressing UNC-84(C953A) migrated normally, but clustered unevenly in the adult syncytium. Similar anchorage defects in the adult animal were observed after mutating the conserved cysteine in the KASH domain of ANC-1, or replacing the KASH domain of ANC-1 with the KASH domain of UNC-83, which does not have a cysteine. Based on these results, our model is that di-sulfide bonds between SUN proteins and various KASH partners could be a component of a molecular switch between actively moving and anchored nuclei. Interestingly, careful quantification of the nuclear anchorage phenotype shows that
unc-84 mutants are not as severe as
anc-1 mutants, suggesting ANC-1 has activities independent of UNC-84. We are currently testing potential ANC-1 partners to see if they enhance nuclear anchorage defects of
unc-84.