How Not To Be in the Wrong Place at the Wrong Time: An Education Primer for Use with “Deposition of Centromeric Histone H3 Variant CENP-A/Cse4 into Chromatin Is Facilitated by Its C-Terminal Sumoylation”

Abstract

Recent work by Kentaro Ohkuni and colleagues exemplifies how a series of molecular mechanisms contribute to a cellular outcome—equal distribution of chromosomes. Failure to maintain structural and numerical integrity of chromosomes is one contributing factor in genetic diseases such as cancer. Specifically, the authors investigated molecular events surrounding centromeric histone H3 variant Cse4 deposition—a process important for chromosome segregation, using Saccharomyces cerevisiae as a model organism. This study illustrates an example of a post-translational modification—sumoylation—regulating a cellular process and the concept of genetic interactions (e.g., synthetic dosage lethality). Furthermore, the study highlights the importance of using diverse experimental approaches in answering a few key research questions. The authors used molecular biology techniques (e.g., qPCR), biochemical experiments (e.g., Ni-NTA/8His protein purification), as well as genetic approaches to understand the regulation of Cse4. At a big-picture level, the study reveals how genetic changes can lead to subsequent molecular and cellular changes.

DURING the lifetime of a cell, separating two sets of chromosomes into two daughter cells is a task that must be carried out with exacting accuracy . Multiple steps must be well coordinated to achieve proper chromosome segregation (Figure 1). First, mitotic chromosomes (now fully duplicated) become attached to structures called microtubules. Microtubules are made up of tubulin subunits, which can polymerize or depolymerize depending on the demands of a cellular process. Microtubules emanate from another cellular structure—the centrosomes (or spindle pole bodies in yeast)—located at the opposite ends of a cell.

Figure 1.

Visual summary of the study by Ohkuni et al. (2020). This study illustrates how genetic changes can alter molecular and cellular mechanisms using a centromeric histone H3 variant Cse4 and its mutants.

Mechanisms of Chromosome Segregation

How are microtubules attached to the chromosomes? Microtubules’ attachment to the chromosomes is restricted to a specific chromosomal region called the centromere. “Centromeres” refer to DNA sequences that have an affinity for centromeric histones and other proteins required to build a proteinaceous structure called a kinetochore (Kitagawa and Hieter 2001; Verdaasdonk and Bloom 2011) (Figure 2). In Saccharomyces cerevisiae, Cse4 is a centromere-specific histone H3 variant that binds to the centromeres and mediates the kinetochore assembly crucial to the attachment of microtubules (Kitagawa and Hieter 2001; Verdaasdonk and Bloom 2011). When in excess, the centromeric histone H3 variant localizes to noncentromeric regions and perturbs proper chromosome segregation (Collins et al. 2004; Heun et al. 2006; Au et al. 2008; Shrestha et al. 2017). In Drosophila and human cells, the centromeric histone H3 variant residing in noncentromeric regions can support kinetochore assembly (Heun et al. 2006; Barnhart et al. 2011; McKinley and Cheeseman 2016). This observation may explain why mislocalization of the histone H3 variant leads to chromosome instability. Chaperone proteins are responsible for localization of the centromeric histone H3 variant. For example, in S. cerevisiae, the chaperone protein Scm3 is responsible for localizing Cse4 to centromeres (Camahort et al. 2007; Mizuguchi et al. 2007; Stoler et al. 2007). Cse4 can be chaperoned by the CAF-1 complex when Scm3 is absent and Cse4 is overexpressed (Hewawasam et al. 2018).

Figure 2.

A simple model to illustrate the interface between centromeric region, kinetochores and microtubule in S. cerevisiae. Multiple proteins (not shown in the figure), including centromeric histone H3 variant (Cse4 in S. cerevisiae), mediate kinetochore assembly. Visual models with molecular details can be found in review articles such as a review by Lampert and Westermann (2011).

One major consequence of faulty chromosome segregation is aneuploidy, a condition in which cells carry an aberrant number of chromosomes. Thus, understanding the regulation of Cse4 (or its counterparts in other organisms) is fundamental to elucidating the mechanisms of aneuploidy. Why do we care about aneuploidy? In somatic cells undergoing mitosis in multicellular organisms, aneuploidy will often lead to cell death, but, in some cases, the event can be the precursor for cancer development. Likewise, in reproductive cells undergoing meiosis, aneuploidy can result in deleterious effects such as embryonic lethality or birth defects. In a recent study, Ohkuni et al. (2020) investigated whether sumoylation of the C-terminus of Cse4 plays a role in correct deposition of the protein on centromeric DNA sequences and chromosome segregation.

Why Is S. cerevisiae a Good Model Organism for this Study?

S. cerevisiae is a great model organism due to is affordability, short life cycle and the availability of genome-wide sequencing data (Duina et al. 2014). The following reasons describe why it is a suitable model organism specifically for this study. First, this study addresses research questions at the cellular and molecular levels. As a unicellular eukaryote, S. cerevisiae provides conclusive answers without the confounding factors from interacting tissues and cell types one might observe in multicellular eukaryotes. Second, genetic interaction and mutational studies reported in this study are feasible because S. cerevisiae is more amenable to genetic manipulation than are higher eukaryotes. Third, the centromeric histone H3 variant (Cse4 in S. cerevisiae and CENP-A in humans) is likely to be functionally conserved throughout evolution (Kitagawa and Hieter 2001). For instance, one study demonstrated that Cse4 can complement the function of CENP-A (Wieland et al. 2004).

Molecular Concepts Behind the Study

Functional regulation of a protein is influenced by a crosstalk between inherent properties of the protein, such as its amino acid sequence, and its interactions with other biomolecules. The protein itself harbors motifs or domains whose chemical nature is determined by their primary structure. These motifs or domains carry messages that are decoded by other proteins at the right time under appropriate conditions. In the context of this paper, we will explore two examples of how a protein’s function can be regulated.

Protein–protein interaction

To carry out multiple biological functions, each protein engages in transient, reversible interactions with myriad biomolecules including other proteins. These interactions—highly specific in most cases—are somewhat predetermined by chemical compatibilities between two interacting biomolecules. In proteins, amino acids exist in a three-dimensional environment poised to interact with amino acids from other proteins when they are in close proximity. For instance, a hydrophobic surface of one protein may interact with that of another, given that surrounding residues are chemically compatible and positioned to interact with each other. These protein–protein interactions are the driving force of many biological processes, from enzyme: substrate interactions to a protein complex escorting another protein to an intended location. In this paper, the authors asked whether lysine (K) residues at 215 and 216 positions (K215/216) are important for Cse4 to interact with its chaperones, the CAF-1 complex and Scm3.

Protein degradation

Protein function may be regulated through a proteasome-mediated degradation process. Although it appears as an excessive measure, degradation ensures that target proteins are excluded from cellular processes when their functions are no longer required. The proteasome harbors a multi-subunit protease—an enzyme that catalytically targets the peptide bonds to degrade proteins (Pickart and Cohen 2004). Since the degradation is irreversible, multiple layers of regulation have evolved to ensure that only proteins meant to be degraded are transported to the proteasome. This transport is triggered by a post-translational modification called ubiquitination.

Post-translational modification of proteins refers to an enzyme-mediated process in which a chemical group or a peptide (a small protein) becomes covalently linked to a target protein. Ubiquitin is a peptide that can be conjugated to target proteins in a process called ubiquitination. A series of enzymes, E1, E2, and E3, are responsible for ubiquitination of target proteins at K residues. Lysine residues on ubiquitin conjugated to a target protein can also be ubiquitinated in a process referred to as polyubiquitination (Swatek and Komander 2016). A ubiquitinated target protein may have as few as one ubiquitin or a chain of polyubiquitin attached to it depending on the cellular signal. Prior to its degradation, a target protein is first polyubiquitinated and this chemical signature allows the proteasome to selectively identify proteins to be degraded (Bard et al. 2018).

Relatively recently, scientists discovered a specialized ubiquitination process restricted to proteins that are first modified with SUMO (small ubiquitin-like modifier). Similar to ubiquitin, a series of enzymes facilitate covalent conjugation of SUMO peptide to target proteins when conditions demand. Sumoylation facilitates various molecular changes such as protein–protein interactions, cellular localization, or association of the target protein with DNA (Sarangi and Zhao 2015). In some cases, sumoylated proteins can be subsequently ubiquitinated by a specialized E3 ubiquitin ligase, STUbL (SUMO-targeted ubiquitin ligase) (Jentsch and Psakhye 2013). Not all sumoylated proteins are degraded by a STUbL. In other words, sumoylation is a prerequisite for degradation only in a few cases. In these special cases, initial sumoylation of proteins enables them to engage in a biological process, such as DNA damage repair. When the service of these sumoylated proteins is no longer necessary, STUbL-mediated degradation of these proteins halts their function (Jentsch and Psakhye 2013). In the context of this paper, Cse4 has been reported to be a target of a yeast STUbL complex, Slx5/Slx8 (Ohkuni et al. 2016, 2018).

What Is the Research Question?

Previous findings imply that correct deposition of Cse4 can be influenced by the amount of Cse4 and its interaction with the chaperones. Given this context, the authors asked which segment of Cse4 is responsible for regulating these processes. In 2018, they demonstrated that sumoylation of K65 within the N-terminus of Cse4 is responsible for STUbL-mediated degradation and prevents mislocalization of Cse4 on chromosomes (Ohkuni et al. 2018). This latter study asked the following questions: do K215/216 within the C-terminus of Cse4 play a similar role with regards to protein stability? Are these residues responsible for the interaction of Cse4 with its chaperones? Do these residues influence proper localization of Cse4 and chromosome segregation?

Methods/Experimental Approaches

The experimental approaches used by Ohkuni et al. (2020) are described below. For experiments with quantitative data, note that the authors performed multiple replicates and statistical analyses to ensure reproducibility of data.

How to conditionally overexpress a gene

Ohkuni et al. (2020) used the galactose-inducible system to overexpress CSE4 or its mutants and to control when to overexpress these genetic variants. In S. cerevisiae, galactose is a carbon source alternative to glucose. A transcription factor that binds to a cis-acting DNA sequence (GAL promoter) regulates galactose metabolic genes in response to galactose. In the absence of glucose and in the presence of galactose, the transcription factor binds to the GAL promoter, facilitating transcription of genes under the control of this promoter. The authors genetically engineered yeast strains so that they carry CSE4 or its genetic variants under the control of the GAL promoter. When expression of these genes is not desired, the researchers simply grow yeast strains in the presence of glucose. When they need to overexpress the genes, the researchers grow yeast strains in the absence of glucose and presence of galactose. (This link provided by the Buratowski laboratory further describes the galactose-inducible system in S. cerevisiae.) The authors also used another inducible system that can achieve similar outcomes. Figure 6 in their paper described a copper-inducible system (Cu-CSE4), in which expression of CSE4 (or its mutants) can be induced conditionally by the addition of copper to the growth medium.

How to study the significance of specific amino acid residues within a protein

How do we know if a specific amino acid is important for the function or regulation of a protein? We use a genetic approach to know its value by taking it away! The authors replaced K within the C-terminus of Cse4 with arginine (R) or alanine (A). Note that both K and R have positively charged side chains, whereas A does not. In K→ R mutants, Cse4 cannot be post-translationally modified with SUMO or ubiquitin and, yet, these mutants retain a positive charge at those positions. However, the authors noted that R can be post-translationally modified with a methyl group. To ensure that the consequences observed in K→ R mutants are due to lack of sumoylation but not due to extraneous addition of methyl groups, they decided to observe the effects in K → A mutants as well.

How to determine the sumoylation status of a protein

After the researchers mutated K residues, they determined whether the genetic changes impaired sumoylation of Cse4 (Figures 1 and 7 in Ohkuni et al. 2020). How do we determine the sumoylation status of a protein? We can isolate a pool of Cse4 proteins from cells, separate them according to their molecular weights in gel electrophoresis and detect which species of Cse4 are present or absent using antibodies. Ohkuni et al. (2020) isolated Cse4 by taking advantage of the interaction between the 8His epitope tag and Nickle-NTA (Ni-NTA) beads. Epitope tags are small peptides that are used universally in biological experiments regardless of the model organisms, and, thus, antibodies or other reagents that have affinity for these tags are commercially available at a reasonable price. Antibodies are proteins naturally produced by vertebrates. They specifically bind to a short stretch of amino acids, referred to as “epitopes”. For research purposes, antibodies can be produced to recognize a protein of interest. Because of their specificity, antibodies can be used to isolate proteins fused to an epitope tag. Epitope tags can also be used in an affinity purification that does not rely on antibodies. For example, Ni-NTA beads rely on the affinity between nickel ions and the 8His epitope tag. The latter is a small peptide made up of eight histidines, which can become negatively charged under basic conditions and have a high affinity for positively charged nickel ions on Ni-NTA beads. Therefore, proteins tagged with 8His can be separated from the rest of the cell lysate using Ni-NTA beads (Figure 3).

Figure 3.

Visual representation of steps involved in an experiment to assess Cse4 sumoylation status. Cells lysate is prepared by breaking open cells and releasing biomolecules into an appropriate solution compatible with Ni-NTA bead purification. Purification process separates 8His-HA-Cse4 (both nonmodified and post-translationally modified forms) from the rest of the lysate. Western blot analysis using a α-SUMO antibody and α-HA antibody is used to detect sumoylated Cse4 and total Cse4, respectively.

Typically, cells have multiple copies of Cse4 proteins, only a subset of which may be sumoylated at a given time (Figure 3). When a Cse4 protein is sumoylated (monosumoylated or polysumoylated), its molecular weight will increase due to the weight of SUMO protein (∼11 kDa for one SUMO peptide in S. cerevisiae). After the pool of purified 8His-tagged Cse4 is separated in a denaturing gel, Cse4 proteins of different sizes will be present at different locations within the gel because gel electrophoresis separates proteins based on their molecular weights. Smaller proteins travel faster compared to the larger ones when an electric current is applied to the polyacrylamide gel in the presence of a denaturing agent sodium dodecyl sulfate (SDS). These proteins are transferred to a membrane, where they can be visualized using primary antibodies that recognize Cse4 or SUMO, followed by the detection with secondary antibodies that bind Cse4- or SUMO-specific primary antibodies (Figure 3). We are able to indirectly detect the presence of Cse4 or SUMO since secondary antibodies are covalently tagged with fluorescent molecules or chemical groups that will produce detectable signals. This method, in which gel electrophoresis separates the proteins according to their weight followed by antibody detection is called SDS-PAGE/Western blotting, or is sometimes just referred to as Western blotting.

Ohkuni et al. (2020) performed Western blot analysis using SUMO antibody (α-Smt3) on purified Cse4 samples (see their Figures 1 and 7). In these analyses, due to the specificity of the antibody, we will observe only the sumoylated species (for example, bands indicated by arrows in “8His-HA-CSE4” lane, Figure 1A) although unmodified Cse4 or Cse4 with other PTMs are present in the gel. The molecular weight of the 8His-HA tag also allows the authors to distinguish sumoylated Cse4 bands from nonspecific ones, which might otherwise occupy the same locations (see their Figure 1A). In parallel, they performed Western blot analysis using Cse4 antibody [α-HA (Cse4)] on the same purified samples. In such an analysis, we expect to observe all forms of Cse4 (modified or otherwise). Modified Cse4 will run more slowly through the gel than nonmodified Cse4. In theory, we should be able to observe all forms of Cse4 when Cse4 antibody is used. However, in practice, modified Cse4 species are generally lower in abundance compared to unmodified Cse4 and thus, are much harder to detect on a blot with the Cse4-specific antibody. Due to this technical reason, it is a common practice to crop the image of the blot to focus only on the unmodified version. In the lower panel of Figures 1A and 7A, the authors focused on only unmodified Cse4 with α-HA (Cse4) antibody.

How genetic interaction is used to understand molecular mechanisms

One experimental approach used in the reviewed study is to determine which genes and Cse4 exhibit a type of genetic interaction called synthetic dosage lethality (Figures 2, 6, and 7 in Ohkuni et al. 2020). What does “genetic interaction” mean? Living systems evolved to sustain homeostasis in the presence of minor perturbations either from the environment or from intrinsic factors such as functional deficiency of a gene. For example, in S. cerevisiae, ∼80% of genes are nonessential, suggesting that cells evolved molecular mechanisms to compensate for the loss of one gene, including having functionally redundant pathways (Winzeler et al. 1999; Hartman et al. 2001; Tong et al. 2001). How do we know which genes are functionally redundant or can alleviate the loss or a change in expression of a gene of interest? Determining genetic interactions is an approach that can answer this research question.

Two genes are considered to have a genetic interaction if the combination of the mutations in those genes yield a phenotype different from the phenotype exhibited by each single mutant (Boone et al. 2007). Suppose that knocking out gene A or gene B individually in a yeast strain does not kill the cells. However, knocking out both genes in the same strain results in a lethal phenotype. In this example, genes A and B are considered to have a genetic interaction, which may indicate that they function in two different molecular pathways. Synthetic dosage lethality is a special type of genetic interaction that describes the relationship between an overexpressed gene and functional deficiency of another gene (Yan et al. 1991; Li and Herskowitz 1993; Friedman et al. 1994; Kroll et al. 1996; Boone et al. 2007). Synthetic dosage lethality is thought to exist if overexpression of a gene necessitates the function of another gene (Yan et al. 1991; Li and Herskowitz 1993; Friedman et al. 1994; Kroll et al. 1996; Boone et al. 2007). For example, overexpression of CSE4 does not have any observable growth phenotype; however, knocking out PSH1 results in lethality when CSE4 is overexpressed. Since Psh1 is one of the ubiquitin ligases responsible for degradation of Cse4, the thought is that eliminating the function of the ligase exacerbates the functional consequences of excess Cse4 (Hewawasam et al. 2010; Ranjitkar et al. 2010). The authors performed multiple synthetic dosage lethality experiments to understand mechanisms behind deposition and degradation of Cse4.

How to assess protein stability

The authors inquired if Cse4 with mutations at K215/216 are more stable than the wild-type (Figures 1 and 2 in Ohkuni et al. 2020). The reasoning is that perhaps these mutants can no longer be recognized by the proteasomal degradation system, and, therefore, their stability may differ from that of the wild type. To address this question, the authors measured the half-life (t_1/2) of each protein, the time it takes to degrade half the amount of the protein present at the start of the experiment. Scientists first inhibit global protein synthesis by using a drug called cycloheximide. This step is necessary because how much a protein is present at a given time point—the steady-state level of a protein—is the net result of protein synthesis and degradation. To examine how fast a protein is degraded, we first need to remove the other variable (i.e., protein synthesis). Ohkuni et al. (2020) also turned off transcription of galactose-inducible CSE4 by using glucose. Under these conditions, they monitored the protein expression in a time-course study and approximated the half-life using Western blot analysis.

How to measure protein–protein interaction

When the authors mutated K215/216 of Cse4, they predicted that the mutations might interrupt the protein’s interaction with its chaperones, Scm3 and the CAF-1 complex (Figures 3, 6, and 8 in Ohkuni et al. 2020). To test this prediction, they used coimmunoprecipitation, followed by Western blot analysis. In this experiment, one first isolates the protein of interest using an antibody that specifically recognizes this protein. The antibody used can be the one that directly recognizes the protein or a covalently attached epitope tag. In their Figure 3A, Ohkuni and colleagues used Flag and HA tags for Cac2 (a subunit of the CAF-1 complex) and Cse4, respectively. Using the antibody against the Flag tag, they were able to isolate Cac2. If the conditions are right, proteins associated with Cac2 can also be purified along with Cac2 because noncovalent interactions between Cac2 and its coassociating partners are chemically robust enough to survive the process of purification. This process is known as coimmunoprecipitation. As an example, we will take a closer look at Figure 3 in the Ohkuni et al. (2020) paper.

The authors have prior knowledge that Cac2 and Cse4 interact with each other, but they wanted to see if mutation of K215/216 would weaken or eliminate this interaction. They immunoprecipitated Cac2 via the Flag tag and separated all precipitated proteins using SDS-PAGE. Then, they detected Cse4 (HA-Cse4) using HA antibody in a Western blot analysis to determine the affinity of Cse4 to Cac2 in wild-type or mutant cells. Similarly, they detected Cac2 (Cac2-Flag) using Flag antibody to ensure that the precipitated protein was present approximately equally in all samples.

How to determine the amount of Cse4 protein in different compartments of a cell

If the interaction between Cse4 and its chaperones is interrupted, less than normal amounts of Cse4 should be present on chromatin since the chaperones can no longer place Cse4 effectively on the DNA. To address this question, the authors used an experimental approach that distinguishes the nonchromatin pool from the chromatin-bound pool of Cse4 (Figure 4 in Ohkuni et al. 2020). First, they enzymatically and chemically broke open yeast cells to extract proteins. Then, they separated soluble fraction (proteins in solution) from the insoluble fraction (chromatin-bound proteins) using 30% sucrose combined with centrifugation. In this experiment, cellular extract containing both soluble and insoluble materials are placed on top of a 30% sucrose solution. After centrifugation, the insoluble portion will settle at the bottom of the tube since it is denser than the soluble counterpart. Then, the authors performed SDS-PAGE gel electrophoresis and Western blot analysis to detect Cse4 in each fraction. They also wanted to ensure that their fractionation worked reasonably well—chromatin fraction was relatively devoid of soluble proteins and vice versa. To this end, they used Pgk1, a protein found exclusively in the soluble fraction, and H3, a canonical histone exclusively detected in the chromatin fraction as markers for those fractions.

How to measure association of a specific protein with a defined region of chromatin

While studying chromatin-associated proteins, sometimes, it is essential to understand if these proteins are properly localized (Figures 4–6, and 8 in Ohkuni et al. 2020). To determine if Cse4 was deposited at the correct chromatin location, the authors use “ChIP-sequencing” and “ChIP-qPCR.” The former approach identifies all genome-wide locations of Cse4, while the latter approach assesses whether Cse4 is deposited at a chromosomal location specified by the researcher. Similar to coimmunoprecipitation, an antibody specific to the protein of interest is used in ChIP (chromatin immunoprecipitation). In this method, before immunoprecipitation, cells are treated with a cross-linking agent (usually formaldehyde) to preserve transient noncovalent interactions that occur naturally between biomolecules (Figure 4). This experimental trick allows us to capture protein–DNA interactions that might otherwise have been lost due to their dynamic nature (Figure 4). When the protein of interest is immunoprecipitated after cross-linking, DNA regions associated with the protein will also be isolated (Figure 4).

Figure 4.

Visual illustration of steps involved in an experiment to determine the association of Cse4 with chromatin. Cells are first treated with formaldehyde to cross-link biomolecules, followed by preparation of cell lysate. Chromatin immunoprecipitation of Cse4 separates Cse4 along with the associating chromatin from the rest of the lysate. Then, all biomolecules, except DNA, are removed from the samples. The remaining DNA is used as input for qPCR or genomic sequencing.

What do we do with these DNA regions? They are useful to answer two types of questions: (1) Does the protein of interest interact with a chromosomal region of interest that the researcher has in mind? (2) Which regions within the genome does the protein of interest interact with? In this study, the authors asked both questions. The former question is answered by using ChIP-qPCR and the latter by ChIP-sequencing (Figure 4). In ChIP-qPCR, the authors immunoprecipitated Cse4 and co-associating DNA from the cell lysate (Figure 4). Then, they eliminated proteins, RNA, and lipids from the samples so that all they had left with was DNA (Figure 4). They used this isolated DNA as a template in a quantitative polymerase chain reaction (qPCR) (Figure 4). In this method, scientists use short pieces of DNA called primers specific to the DNA sequence they would like to target. For each region of interest in the genome, scientists design a pair of primers that will anneal to a stretch of DNA sequences bracketing the region. These primers are then used in qPCR that measures the extent of amplification compared to a control. (DNA Learning Center from Cold Spring Harbor Laboratory provides a great interactive tool demonstrating each step of PCR.) qPCR records the number of PCR cycles it takes to reach a plateau of DNA amplification. For example, if two copies of DNA are present in sample A and four copies of the same DNA are present in sample B, it will take one fewer PCR cycle to reach the plateau for sample B than for sample A. Thus, by measuring the number of cycles, one can infer the relative amount of DNA templates in one sample compared to another. In their Figure 4C–F, for example, the authors asked how much Cse4 or its variants associated with noncentromeric regions (PHO5, SLP1, SAP4, and RDS1 promoters). They performed ChIP followed by qPCR and observed that in Figure 4C there was less PHO5 promoter region associated with cse4 mutants compared to the wild type.

Instead of performing qPCR, scientists can also subject the chromatin-immunoprecipitated samples to genome-wide sequencing (Figure 4). This method can document the identity (sequence information) of all genomic DNA fragments associated with Cse4 wild-type or mutant proteins. In addition to sequence information, genome-wide sequencing can also reveal the quantity of each genomic DNA fragment associated with Cse4 proteins. This is the information that Ohkuni et al. used in their Figure 5A to assess whether the K215/216 of Cse4 are responsible for depositing the protein in various genomic regions when overexpressed. “Enrichment (IP/input)” on the y-axis represents the amount of each DNA region present in the ChIP sample relative to the input (non-ChIP DNA sample). Enrichment values are illustrated in a violin plot in which the width of the shape represents the frequency of each numerical value.

How to measure the rate of chromosome mis-segregation

The authors inquired if mutation of K215/216 led to faulty chromosome segregation (Figure 6 in Ohkuni et al. 2020). This question was addressed using “chromosome loss assay” developed by the Hieter laboratory and relies on an artificial chromosome (reporter chromosome) dispensable for yeast survival (Spencer et al. 1990) (Figure 5). This chromosome is transformed into a yeast strain with a loss-of-function mutation in the ADE2 gene (denoted as “ade2”). ADE2 encodes an enzyme in the adenine biosynthetic pathway. In the absence of this enzyme, an intermediate of the pathway with red pigment builds up, rendering ade2 mutants red (Jones and Fink 1982). However, the artificial chromosome carries a gene whose protein product suppresses this red phenotype, resulting in white yeast colonies (Figure 5). Under conditions in which chromosome numbers are unstable (aneuploidy-inducing conditions), the probability of losing the artificial chromosome becomes higher and we occasionally observe red cells among white colonies (Figure 5). Using this assay, the authors quantified the frequency of the appearance of red yeast cells and estimated the relative frequency of chromosome mis-segregation in cse4 mutants compared to the wild-type control (see their Figure 6E).

Figure 5.

Visual depiction of chromosome loss assay to measure the rate of chromosome mis-segregation. In this assay, a yeast strain with ade2 mutation is used. This mutation alone results in formation of red colonies. However, this phenotype can be suppressed by a gene called SUP11, which is carried on an artificial chromosome. Under conditions which induce aneuploidy, this artificial chromosome can be lost, resulting in red revertants. Thus, the frequency of red colonies can be used to infer the rate of aneuploidy under a given condition (Spencer et al. 1990).

Narrative of the study in a big picture

Kinetochore assembly must be tightly regulated so that daughter cells do not end up with an aberrant number of chromosomes. Under physiological conditions, kinetochores are assembled at defined chromosomal locations called centromeres and a centromeric histone H3 variant Cse4 in S. cerevisiae is one of the molecules important for this process. Mis-localization of Cse4 disrupts proper chromosome segregation. Expression level of Cse4 and its interaction with chaperone proteins can influence localization of Cse4 on chromosomes. In this study, the authors determine whether K215/216 within the C-terminus of Cse4 are responsible for Cse4 stability and/or Cse4’s interaction with chaperones. To this end, they mutated those K residues and asked the following questions:

1. Are K215/216 responsible for post-translational modification (sumoylation) of Cse4?

2. Does mutation of K215/216 impair Cse4 stability?

3. Does mutation of K215/216 impair Cse4’s interaction with its chaperone proteins?

4. Are K215/216 responsible for localization of Cse4 at centromeric and noncentromeric sites when the protein is overexpressed?

5. Does mutation of K215/216 lead to higher frequency of chromosome mis-segregation?

6. Is sumoylation of K65 (in the N-terminus) functionally distinct from sumoylation of K215/216 (in the C-terminus)?

Discussion questions

1. In Figure 1, why did the authors examine the stability of cse4 K215/216R and cse4 K215/216A mutants after they determined that mutation at those K residues decreased protein sumoylation?

2. In Figure 1A (also in Figure 7A), which bands represent sumoylated Cse4 and which bands are sumoylated proteins that nonspecifically bind to beads?

3. a. In Figure 2A, what type of genetic interaction did CSE4, cse4 K215/216R and cse4 K215/216A strains exhibit when PSH1 was knocked out?

   • b In Figure 2C, why did the authors detect Tub2 (α-Tub2) along with Cse4? Why was the amount Tub2 relatively equal across multiple samples?

   • c In Figures 2, C and D, did the half-life of mutant Cse4 protein encoded by cse4 K215/216R or cse4 K215/216A change compared to the wild-type protein when PSH1 was absent?

   • d Based on your answers in a, b, and c, and the knowledge that Psh1 is an E3 ubiquitin ligase responsible for Cse4 degradation, what molecular mechanisms can you predict to explain why CSE4 overexpression is lethal in the absence of PSH1?

4. In Figures 2, A, B, 6E, and 7C, what is the purpose of “Vector” strains? What type of information will we miss without the use of these strains?

5. In Figure 3A, the authors measured the amount of Cse4 and its variant in whole cell extract (WCE) in addition to the amount that is associated with Cac2. Why is the measurement in WCE necessary to draw the conclusion that the protein product of cse4 K215/216R/A associates less with Cac2 compared to the wild type?

6. a. Based on Figure 4A, what is the distribution Cse4 and its mutants in soluble and chromatin fractions? How does the use of Pgk1 and H3 help to draw a conclusion about these proteins’ localization on chromatin?

   • b. How did the authors derive the graph displayed in Figure 4B using the data in Figure 4A?

7. Why did the authors decide to examine PHO5, SLP1, SAP4, and RDS1 promoter regions in Figures 4, C–F for ChIP-qPCR experiments? What is the significance of these chromatin regions?

8. In ChIP-qPCR experiments (for example, in Figures 4, C–F), the data were reported as “% input”. Discuss why the data were reported this way instead of the amount of DNA immunoprecipitated.

9. In Figure 5A, how do we read these violin plots? Discuss why the data were presented as “enrichment (IP/input)” values. What is the main conclusion for wild type and Cse4 mutants in this figure?

10. How are the chromosomal regions examined in Figures 5, B–E different from those regions examined in Figures 4, C–F?

11. What type of chromosomal region do “CEN1” and “CEN3” represent? Explain how they are different from chromosomal regions described in Figures 4 and 5.

12. In Figures 4, 5 and 6, the authors examined the localization of Cse4 and its mutants at various chromosomal regions using ChIP-qPCR or ChIP-sequencing in psh1Δ genetic background. Explain the reason for why they performed these experiments in the absence of PSH1.

13. a. In Figure 6E, under which condition was SCM3 expressed? Under which condition was CSE4 or its mutant expressed?

    • b. In Figure 6E, what phenotype did cells without CSE4 overexpression exhibit when SCM3 was absent? Discuss why they exhibited such a phenotype.

    • c. In Figure 6E, which strain was more efficient in rescuing the phenotype of Scm3Off: cse4 K215/216R or CSE4?. What is the interpretation that the authors provided for this observation?

14. In Figure 6F, which strains showed higher frequency of chromosome mis-segregation? Provide an interpretation for why these strains are more likely to have chromosome mis-segregation.

15. a. In Figures 7 and 8, the authors compared HA-cse4 K65R, HA-cse4 K215/216R and HA-cse4 K65/215/216R to the wild-type strain as well as to each other. Discuss the significance of these mutants in the context of regulation of Cse4. What is the overarching research question asked in Figures 7 and 8?

    • b. Discuss how these mutants differ in the following aspects. Based on these observations, what was one major conclusion that the authors drew?

      • i. sumoylation status

      • ii. genetic interaction with PSH1 knockout

      • iii. physical interaction with the chaperones (Cac2 and Smc3)

      • iv. localization in centromeric and noncentromeric regions

16. In Figure 9, Ohkuni et al. (2020) provide a model describing the molecular mechanisms of Cse4 regulation and how the K residues in the C-terminus of the protein contribute to the deposition of Cse4 in centromeric and noncentromeric regions.

    1. Which figures provide evidence that Cse4 is sumoylated at K215/216? Explain.

    2. Which figures provide evidence that sumoylated Cse4 interacts with Scm3 or the CAF-1 complex? Explain.

    3. What observations led the authors to the conclusion that K65 and K215/216 are involved in distinct regulatory pathways (protein stability for K65 and protein localization for K215/216)? Which figures illustrate those observations?