A small part of H4 just wants CENP-A to go away

Science Spotlight

A small part of H4 just wants CENP-A to go away

From the Biggins Laboratory (Basic Sciences Division)

Dec. 19, 2016

Our genetic information is stored in DNA, which in our cells is wound around protein complexes called nucleosomes to form a structure that resembles “beads on a string”.  Nucleosomes are composed of eight proteins called histones, with two copies each of histones H3, H4, H2A, and H2B. Ultimately, the “beads on the string” structure is folded around itself extensively to form larger structures called chromosomes that in human cells resemble a letter X when the cell is undergoing cell division to generate new cells.  To be pulled into so-called “daughter” cells, chromosomes make attachments to fibers called microtubules.  The centromere, originally defined as the constriction point in the X shape, is the place on each chromosome that attaches to microtubules to be segregated during cell division.  A more modern definition of the centromere includes the presence of a protein called CENP-A in the associated nucleosomes, replacing the similar and more widely distributed histone protein H3 found in nucleosomes outside the centromere.  Because CENP-A defines the centromere or point of attachment with microtubules, its precise location in the “beads on a string” must be highly regulated to prevent chromosome loss or breakage.  In fact, there is some evidence that mislocalization of CENP-A is associated with tumorigenesis1.  Former postdoctoral fellow Gary Deyter and graduate student Erica Hildebrand in the Biggins Lab (Basic Sciences Division) found that a specific residue in histone H4 contributes to the stability and localization of CENP-A.  The results of their investigation were recently published in Genetics.

Previous research had identified a protein important for CENP-A degradation in budding yeast, which is a widely used model organism due to the high degree of similarity between yeast and human proteins involved in cell division.  When cells lack this protein, called Psh1, they are sensitive to high levels of CENP-A because it doesn’t get degraded and becomes incorporated at many places on each chromosome rather than just at the centromere.  This causes defects in the cell’s ability to accurately use its DNA.  CENP-A, as an H3 variant, tightly associates with the H4 histone protein.  Deyter et al hypothesized that interactions between H4 and CENP-A may be important for CENP-A stability and therefore proper chromosome function and maintenance.  To test this, Deyter performed a genetic screen using a library of yeast strains to test whether mutation of individual amino acids in H4 may cause sensitivity to over-expression of CENP-A. He observed that yeast with arginine (R) 36 of H4 mutated to alanine (A) died specifically when CENP-A was overexpressed.

Normal chromosome with CENP-A protein only at the centromere.  Abnormal chromosome with H4-R38A mutation lead to CENP-A nucleosomes throughout the chromosome.

[left] Normally, CENP-A nucleosomes are only incorporated at centromeres and Psh1 promotes the degradation of extra CENP-A. [right] When arginine 36 (R36) of histone H4 is mutated to alanine, changing the charge of the protein at that position from positive to neutral, Psh1 cannot bind CENP-A containing nucleosomes and they become incorporated throughout the DNA.

In line with the hypothesis that this residue is important for CENP-A stability, he found that CENP-A accumulated in cells with the H4-R36A mutation versus wild-type cells.  Additionally, the extra CENP-A protein was incorporated into the DNA of cells with arginine 36 of H4 mutated to alanine (see graphic). Further genetic analysis demonstrated that the positive charge of the amino acid at that position is what is important because cells with the arginine mutated to a similarly positive-charged amino acid, lysine, behaved just like wild-type cells.  However, changing it to a negatively charged amino acid such as glutamic acid caused sensitivity just as the mutation to alanine, which is neutral, had done. 

An average gene does not have CENP-A nucleosomes.  In a cell with H4-R36A nucleosomes CENP-A nucleosomes appear in the gene body and Psh1 E3 ligase accumulates in the 3' untranslated ends of genes.

[left] In a wild-type cell at a given gene, H3 nucleosomes wrap the DNA in the promoter (line on far left), in the gene-body (box), and in the 3' untranslated region (line, right of the box). Psh1 (light green oval) degrades CENP-A nucleosomes before they are incorporated into the DNA. [right] In a cell where arginine 36 of H4 is mutated to alanine, Psh1 accumulates in the 3' untranslated region at the ends of genes and CENP-A containing nucleosomes are abnormally incorporated into DNA within and surrounding genes.

Deyter and Hildebrand found that the interaction between CENP-A and Psh1 was decreased in cells with the H4-R36A mutation.  This suggests that Psh1 is not able to interact with CENP-A to degrade it when this part of H4 is mutated.  To investigate the effect on Psh1 function further, Hildebrand and Barber analyzed the localization of Psh1 at specific locations in the DNA and found that Psh1 was unusually enriched at the ends of some genes.  Overall, their results suggest that arginine 36 of histone H4 is required for proper homeostasis of CENP-A protein and localization of the Psh1 degradation enzyme.  

Arginine 36 is the only amino acid in a specific surface-exposed patch of H4 that appears to directly contact the DNA.  Mutating the adjacent residue, arginine 35, had no effect on the cell's sensitivity to high levels of CENP-A.  It is possible that the association of arginine 36 with DNA is essential for CENP-A recognition by proteins such as Psh1 that regulate CENP-A stability and therefore the proper, exclusive localization of CENP-A at the centromere.


Deyter GMR, Hildebrand EM, Barber AD, Biggins S.  2016.  “Histone H4 facilitates the proteolysis of the budding yeast CENP-ACse4 centromeric histone variant.”  Genetics.  Epub ahead of print. DOI: 10.1534/genetics.116.194027

This research was supported by the National Institutes of Health, the American Cancer Society, and the National Science Foundation. SB is an investigator of the Howard Hughes Medical Institute.  

Extra citation:

1.  Amato A, Schillaci T, Lentini L, Di Leonardo A.  2009.  "CENP-A overexpression promotes genome instability in pRb-depleted human cells."  Molecular Cancer.  8:119.