Biochemical characterization of Caenorhabditis elegans histone chaperone LIN-53 in centromeric nucleosome assembly

Abstract

To ensure continuity of life, genetic information carried in cells must be passed on through cycles of cell division across generations. In many eukaryotes, the centromere is a conserved structure found on chromosomes that governs the process of chromosome segregation during cell division. It serves as the attachment site for kinetochore proteins and spindle microtubules that drive equal separation of chromosomes to daughter cells. The presence of a histone H3 variant, CENP-A, specifies the centromeric identity on the chromatin and is required for proper centromere function. Following the synthesis of new DNA strands during replication, CENP-A molecules have to be replenished on centromeric chromatin by tightly regulated assembly pathways.

In nematode Caenorhabditis elegans, the centromeric localization of CENP-AHCP-3 is known to be interdependent with the centromere licensing factor KNL-2 and histone chaperone LIN-53. Due to the absence of CENP-AHCP-3-specific assembly factor in C. elegans, LIN-53 has been proposed as the histone chaperone for CENP-AHCP-3 that escort CENP-AHCP-3 to the centromeric chromatin in a KNL-2 dependent manner. However, as LIN-53 homologs have been implied as a H3-H4 chaperone in other organisms, the mechanistic detail of how LIN-53 may recognize and differentiate CENP-AHCP-3 from histones H3 remains unclear.

To investigate how LIN-53 and CENP-AHCP-3 can spatially interact with each other in vivo, the subcellular localization of both proteins was determined by biochemical fractionation using C. elegans embryonic extracts. It is found that LIN-53 was ubiquitously found in cytoplasm and nuclei of C. elegans embryos whereas CENP-AHCP-3 was mostly associated with chromatin under physiological conditions. Co-immunoprecipitation analysis revealed that LIN-53 could physically interact with both KNL-2 and chromatin-bound CENP-AHCP-3.

To further investigate the direct physical interactions between LIN-53, CENP-AHCP-3 and other histone subunits, their recombinant forms were expressed and purified from E. coli by various chromatography methods. Pull-down assays demonstrated that LIN-53 could individually interact with histone H3, H4 and CENP-AHCP-3. In vitro reconstituted (H3-H4)2 tetramers were also successfully pulled-down by recombinant LIN-53. Using isothermal titration calorimetry, binding affinities between LIN-53 and histone H3 and H4 were individually examined. LIN-53 showed comparable binding affinities to H3 and H4 to the conserved H3-H4 chaperone UNC-85. Together these results are consistent with LIN-53’s duality in function by chaperoning CENP-AHCP-3 at centromere and H3-H4 at canonical nucleosome sites. However, the question on how LIN-53 can determine its recognition specificity towards H3, H4 and CENP-AHCP-3 remains.

This study has provided necessary tools and created new directions for further investigations, such as biochemical properties of histone multimers and nucleosomes reconstituted from purified histones, and functional domains of CENP-AHCP-3 and LIN-53 by constructing protein truncates. The chaperone activity of LIN-53 during the process of canonical or centromeric nucleosome assembly can be monitored and compared by plasmid supercoiling assay and fluorescence resonance energy transfer assay. Understanding the modes of interaction on how LIN-53 can both assemble canonical and centromeric histones in different contexts may highlight the functional importance of histone chaperones in maintaining genome integrity, and provide insights to the holocentromere assembly features as compared to that in monocentric system.