Super-resolution microscopy reveals the number and distribution of topoisomerase IIα and CENH3 molecules within barley metaphase chromosomes

Topoisomerase IIα (Topo IIα) and the centromere-specific histone H3 variant CENH3 are key proteins involved in chromatin condensation and centromere determination, respectively. Consequently, they are required for proper chromosome segregation during cell divisions. We combined two super-resolution techniques, structured illumination microscopy (SIM) to co-localize Topo IIα and CENH3, and photoactivated localization microscopy (PALM) to determine their molecule numbers in barley metaphase chromosomes. We detected a dispersed Topo IIα distribution along chromosome arms but an accumulation at centromeres, telomeres, and nucleolus-organizing regions. With a precision of 10-50 nm, we counted ~ 20,000-40,000 Topo IIα molecules per chromosome, 28% of them within the (peri)centromere. With similar precision, we identified ~13,500 CENH3 molecules per centromere where Topo IIα proteins and CENH3-containing chromatin intermingle. In short, we demonstrate PALM as a useful method to count and localize single molecules with high precision within chromosomes. The ultrastructural distribution and the detected amount of Topo IIα and CENH3 are instrumental for a better understanding of their functions during chromatin condensation and centromere determination. Supplementary Information The online version contains supplementary material available at 10.1007/s00412-023-00785-8.


Introduction
Determining the ultrastructures of cell nuclei and chromosomes and quantifying their molecular components are required to understand the dynamics of such basic biological processes as transcription, replication, and cell division. Often, fluorescence dyes are used to label the protein of interest. In contrast to classical fluorescence microscopy techniques, super-resolution microscopy such as structured illumination microscopy (SIM) and single-molecule localization microscopy (SMLM), including photoactivated localization microscopy (PALM) are well suited to achieve ultrastructural imaging by breaking the diffraction limit of light (Schermelleh et al. 2019;Khater et al. 2020). In PALM, fluorophores are excited in such a way that only one molecule of many within the diffraction-limited spot is in its "On" state and can reach precisions up to ~10-40 nm (Betzig et al. 2006).
Meanwhile, SIM, implemented in different microscopic platforms, has been widely used in cell biology (Heintzmann and Huser 2017). It was applied to scrutinize chromosome scaffold proteins (Poonperm et al. 2015) and unveiled the localization of the replication protein A on chromosome axes during meiotic recombination in mammals (Yoon et al. 2018). Moreover, SIM revealed that the mammalian genome, in interphase nuclei, is organized into functional chromatin domains of ~200-300 nm in diameter (Miron et al. 2020).
PALM uncovered dynamic clusters of cohesin and the insulator protein CTCF (Hansen et al. 2017) and contributed to elucidating the mammalian higher-order chromatin structure (Nozaki et al. 2017). PALM was also successfully applied to bacteria to monitor transcription (Stracy and Kapanidis 2017). The development of a three-dimensional (3D) assay for transposase-accessible chromatin-PALM enabled to study chromatin domains and genome topology changes in single cells (Xie et al., 2020).
On the other hand, SMLM, like PALM studies, is still limited in plants. Schubert and Weisshart (2015) determined the number of RNA polymerase II molecules in differentiated Arabidopsis nuclei by PALM. Besides, PALM was used to analyze microtubules and microtubule-associated proteins in Arabidopsis root epidermal cells (Vavrdová et al. 2020).
Recently, we compared different super-resolution microscopy methods and proved their superiority over diffractionlimited fluorescence microscopy to analyze chromosomal chromatin. The achieved lateral SIM resolution of ~100 nm and PALM localization precision of up to ~10 nm demonstrated that the combination of both techniques provides a comprehensive overview of Topoisomerase IIα (Topo IIα) localization in barley metaphase chromosomes at the ultrastructural level (Kubalová et al. 2021b).
Topoisomerases are involved in transcription, DNA replication, and chromatin organization (Björkegren and Baranello 2018;Meijering et al. 2022;Pommier et al. 2022). Topo IIα is a dimeric enzyme (~175 kDa in human) owning catalytic and non-catalytic functions. The former depends on ATPase activity, whereas the latter relies solely on the C-terminal domain (CTD) (Fukui and Uchiyama 2007;Edgerton et al. 2016). The catalytic function ensures that supercoiled or catenated DNA becomes resolved via DNA strand passage. Topo IIα introduces double-strand breaks into dsDNA, thus allowing other DNA fibers to pass through. Afterward, the break becomes sealed without any loss of DNA information. This action is important for several biological processes such as DNA replication, transcription, chromosome condensation, and segregation (Nitiss 2009). Although CTD is dispensable for decatenation (Dickey and Osheroff 2005), it is essential for the targeting of Topo IIα in mitotic chromosomes (Lane et al. 2013). First, Topo IIα resolves inter-chromatid entanglements, then it generates intra-chromatid entanglements to promote thickening. Only the latter process requires the CTD (Shintomi and Hirano 2021). The majority of Topo IIα concentrates in the inner centromere and is associated with the control and activation of cell cycle checkpoints as demonstrated in human, mouse, and muntjac cells (Coelho et al. 2008;Lane et al. 2013;Gomez et al. 2013Gomez et al. , 2014Nielsen et al. 2020).
Besides, Topo IIα localizes in chromosome arms. Topo IIα was found as a component of the chicken and human mitotic chromosome scaffolds Samejima et al. 2012;Chu et al. 2020). The application of light and electron microscopy uncovered twisted double-stranded protein scaffolds in both human metaphase chromatids. These scaffolds are composed of alternating Topo IIα enzymes, condensins, and kinesin family member 4 (KIF4) proteins (Ono et al. 2004;Samejima et al. 2012;Poonperm et al. 2017;Chu et al. 2020). The importance of Topo IIα was demonstrated via its depletion, disrupting the scaffold structure (Poonperm et al. 2015).
Although most data originate from mammalian research, it was reported that plant Topo IIα acts in mitotic and meiotic recombination (Singh et al. 2004). In onion (Zabka et al. 2014) and tobacco (Singh et al. 2017), Topo IIα is involved in cell cycle regulation and removes meiotic bivalent interlocks in Arabidopsis (Martinez-Garcia et al. 2018). Thus, Topo IIα possesses several roles while residing on mitotic chromosomes, each requiring a precise location and number of available molecules.
Centromeres, occurring as distinct primary constrictions (monocentromeres) or distributed along chromosomes (holocentromeres) , are fundamental for correct chromosome segregation during mitotic and meiotic cell divisions. Thus, they secure the proper distribution of genetic material into daughter cells. Tandemly repeating DNA sequences of different lengths among species are typical for these regions. In contrast to non-centromeric chromatin, most centromeres contain a specific histone H3 variant, termed CENH3 (or CENP-A) (Palmer et al. 1991;Talbert et al. 2002;Cleveland et al. 2003;Ali-Ahmad and Sekulić 2020). In most eukaryotes, CENH3 specifies the position of a proteinaceous complex, the kinetochore, where spindle fibers attach pulling the chromosomes towards both daughter cells (Musacchio and Desai 2017). Vast numbers of proteins residing at the centromere and in the kinetochore detect the fidelity of the spindle fiber attachment and eventually trigger cell cycle checkpoints (Cleveland et al. 2003;Hindriksen et al. 2017). Thus, the vital function of centromeres based on the presence of a certain CENH3 amount is required.
The monocentromeres of most plant species, like rye, barley, Aegilops speltoides, and Cuscuta japonica are determined by CENH3, and their ultrastructures were analyzed by SIM (Wanner et al. 2015;Schubert et al. 2016Schubert et al. , 2020Oliveira et al. 2020). SIM revealed that barley CENH3 is localized mainly in the interior, rather than at the surface of the monocentromeres. Only a low amount is present in the pericentromeres. Barley encodes two CENH3 variants, αCENH3 and βCENH3, interacting with a fraction of Cerebra, a centromeric retroelement 1 3 (CR)-like repeat, and besides with a GC-rich centromeric satellite (Houben et al. 2007;Sanei et al. 2011;Schroeder-Reiter et al. 2012). α and βCENH3 colocalize and form together two distinct globular intermingling structures at the primary constriction of mitotic and meiotic metaphase chromosomes (Ishii et al. 2015;Wanner et al. 2015).
SIM has also been used to quantify the relative amount of immuno-labeled CENH3 during the mitotic and meiotic cell cycles of rye ). However, SIM investigations cannot determine the absolute number of molecules. The specific number and localization of proteins are required to understand their function in the chromatin organization of interphase nuclei and during cell divisions. Moreover, these data are necessary to improve polymer simulations (Câmara et al. 2021;Kubalová et al. 2021a) explaining chromatin condensation along chromosome arms and at centromeres.
In this work, we show that PALM/SMLM is a useful method to count and localize single molecules with high precision. We applied SIM and PALM to localize and quantify the number of Topo IIα and CENH3 molecules based on immuno-labeled somatic barley metaphase chromosomes. The observed accumulation of both proteins within centromeres shows their need to arrange plant centromeres. Furthermore, Topo IIα is present along chromosome arms probably necessary to condense chromatin.

Topo IIα occurs dispersed at arms but accumulates at centromeres, telomeres, and NORs of barley metaphase chromosomes
To analyze the distribution of Topo IIα at the ultrastructural level, we stained flow-sorted barley chromosomes with specific antibodies and applied 3D-SIM. Both Topo IIα peptide antibodies raised in rabbits and guinea pigs (Topo IIrb12 and Topo IIgp13, respectively) (Kubalová et al. 2021b)   To figure out whether Topo IIα and CENH3-positive chromatin colocalize at centromeres, the chromosomes were labeled with CENH3-specific antibodies in addition (Fig. 1). Similar as shown previously in Arabidopsis and cereals ), CENH3 labels cluster-or ring-like ultrastructures as shown on the third chromosome of Fig. 1. Topo IIα and CENH3 detecting the inner centromere are differently positioned but intermingle among each other (enlarged region of the fifth chromosome in Fig. 1). Movie 3 visualizes the spatial Topo IIα and CENH3 distribution and colocalization in the same rotating 3D-SIM image Z stack.
Twenty-eight percent of the ~20,000-40,000 Topo IIα molecules per chromosome localize within the pericentromere To determine the absolute Topo IIα molecule numbers per chromosome, we performed 3D-PALM (Fig. 2b). The PALM imaging confirmed the molecule distribution patterns visualized via SIM. Both different Topo IIα antibodies delivered similar molecule numbers indicating the reliability of the antibodies and the 3D-PALM method. The chromosome in Fig. 2b showed after 3D-PALM in the Gauss visualization mode a similar Topo IIα distribution as via 3D-SIM. The red crosses within the Gauss mode indicate the exact position of the localized and counted molecules within the enlarged centromeric region of a single PALM slice (Fig. 2 bottom). Running through the PALM-Gauss image Z stack exhibits the spatial distribution of Topo IIα (Movie 4).

Topo IIα enzymes surround ~13,500 CENH3-containing nucleosomes at centromeres
To colocalize Topo IIα and CENH3, we immunolabeled flow-sorted barley chromosomes with specific antibodies simultaneously and applied 3D-SIM. While Topo IIα is mainly evident in the pericentromeres, CENH3 concentrates within the core of the primary constrictions and intermingles with Topo IIα-labeled chromatin ( Fig. 1; Movie 3). The fluorescence signals of anti-CENH3 were detected only at centromeres, forming one CENH3-positive region per chromatid. Compared to wide-field and deconvolution microscopy, the increased resolution achieved via SIM allowed the detection of looped CENH3-labeled chromatin fibers (Fig. 3a). Besides, CENH3 chromatin may form ring-like structures) (Fig. 1), similar as found in other cereals and Arabidopsis ).  Fig. 1. c Boxplot representing the CENH3 molecule number variability in 13 barley centromeres analyzed by 3D-PALM. Numbers indicate lower whisker (10,000), 25% quantile (11,700), median (12,800), mean (13,515), 75% quantile (15,520), and upper whisker (19,500) Afterward, 3D-PALM was applied on isolated chromosomes exclusively labeled with CENH3-Alexa488 antibodies (Fig. 3b, c). The labeling pattern was consistent with the SIM imaging. The distinctly localized single molecules (cross presentation) accumulate especially within bright spots of the Gauss display. A mean number of ~13,500 CENH3 molecules per chromosome, i. e., ~6750 per sister centromere, were counted. Compared to Topo IIα, the CENH3 molecule numbers varied less in the 13 barley centromeres analyzed.
3D-PALM allows the detection of single Topo IIα molecules with a lateral (XY) and axial (Z) precision of up to ~10 nm (Kubalová et al., 2021b). We achieved a similar high precision detecting the number of CENH3-containing nucleosomes. About 87% of them were localized laterally and 88% axially with a precision of 10-50 nm (Suppl. Figure 1).
Assuming an octamer structure, each centromeric nucleosome octamer of ~11 nm in diameter contains two CENH3 histones (Nechemia-Arbely et al. 2017). Our achieved PALM localization precision does not allow us to separate both CENH3 histones within a nucleosome. Consequently, each barley metaphase centromere should contain ~27,000 CENH3 molecules.

3D-PALM is useful for quantifying single molecules
In this study, we investigated the distribution and the absolute numbers of plant Topo IIα and the centromeric variant of histone H3 (CENH3) in barley metaphase chromosomes. To obtain the numbers and positions of these molecules, we combined two super-resolution microscopic techniques, SIM and PALM. Flow-sorted chromosomes were used because flat and cytoplasm-free specimens can deliver the most informative data. Two different polyclonal peptide antibodies against Topo IIα revealed the localization of this protein with a lateral and axial precisions of ~10-60 nm (Kubalová et al., 2021b). A similar precision we reached for CENH3-specific signals allowing to localize ~88% of single molecules with a distance of ~10-50 nm.
Low numbers of CENH3 molecules were detected in fission yeast with 26 by PALM (Lando et al. 2012), 84 in Drosophila using CENH3-EGFP fluorescence intensity measurements (Schittenhelm et al. 2010), and 25-62 in chicken DT-40 cells by SMLM or a confocal microscopy-based fluorescence ratio method (Ribeiro et al. 2010;Johnston et al. 2010). Contrary, in HeLa cells Black et al. (2007) determined at most 30,000 CENH3 molecules, i.e., ~15,000 CENH3-containing nucleosomes per centromere by using immunoblotting of extracts from randomly cycling cells with known amounts of CENH3 as quantitation standards. More recently, Bodor et al. (2014) used an indirect fluorescence method to show that human centromeres contain ~400 CENH3 molecules. This high data variability may be caused by using indirect fluorescence-based molecule counting methods containing possibly erroneous steps.
In our opinion, SMLM with its high precision to localize single molecules is currently the most reliable method to count molecules, albeit limitations apply (Shivanandan et al. 2014). It should be regarded that the number of molecules detected depends on sample integrity, staining efficiency, imaging parameters, and image processing parameters. Consequently, an exact molecule number cannot be provided. Multi-emitter algorithms might be able to single out blinking molecules, but cannot rule out the underestimation of the true molecule number (Dempsey et al. 2011;Oddone et al. 2014).
The amount of ~27,000 CENH3 molecules per centromere we detected in barley is in the range Black et al. (2007) revealed in human chromosomes and is clearly higher than determined by the other above-mentioned authors.

Topo IIα distribution in barley differs from that in non-plant species
Here, we demonstrate that the most prominent Topo IIα fluorescence signals are present at pericentromeres, NORs of chromosomes 5H and 6H, and some telomeres of mitotic barley metaphase chromosomes. We assume that the telomeric signals of Topo IIα are chromosome-specific because both NOR-bearing chromosomes identifiable after flow-sorting always showed identical Topo IIα-labeling patterns. The high density of Topo IIα in barley pericentromeres and some telomeres is consistent with protein accumulations identified by scanning electron microscopy (Wanner and Schroeder-Reiter 2008).
At barley chromosome arms, Topo IIα is distributed in a reticulate manner. In human HeLa and Chinese hamster cells, Topo IIα, together with condensin, form a line-like protein scaffold inside each chromatid (Maeshima and Laemmli, 2003;Kireeva et al. 2004;Poonperm et al. 2015;Walther et al. 2018). These scaffold proteins were shown to be linked via bridges between the sister chromatids in pig, muntjac, and human (Chu et al. 2020). But it has also been reported that in HeLa cells, the chromatid axes occur as isolated compaction centers rather than forming a continuous line-like scaffold (Sun et al. 2018) and appear to consist of a helical structure that serves to organize chromatin loops into the metaphase chromatid (Phengchat et al. 2019). We suppose that the reticulate scaffold formation in barley may be due to different lengths of major and minor loops forming the 400 nm thick helically organized chromonema building condensed metaphase chromatids. The loop sizes fit experimental Hi-C data induced via a dispersed helical scaffold. Due to the intermingling of ~80 nm lower-order chromatin fibers, the helical chromonema structure is not visible by 3D-SIM on homogeneously stained chromatids but via differential oligo-FISH labeling (Kubalová et al. 2021a).
One Topo IIα enzyme for every 20-50 kb of DNA was estimated to be present in mitotic HeLa metaphase chromosomes (Gasser et al. 1986;Fukui and Uchiyama 2007). In barley chromosome arms, we determined ~27,400 Topo IIα molecules, ~28% of them around centromeres. The genome size of barley containing seven chromosomes is 4.65 Gb DNA (Monat et al. 2019). This corresponds to 664.3 Mb per chromosome. Thus, one molecule of Topo IIα resides at approximately every 24 kb (664.3 Mb /27,400 = 24 kb) of the whole barley chromosome, a value similar to that found for HeLa cells. Due to the accumulation of Topo IIα (~7,700 molecules) in the barley pericentromeres spanning ~100 Mb DNA (Monat et al. 2019), we assume one molecule per 13 kb (100 Mb/7700 =13 kb), and along chromosome arms a lower density with one molecule every 32.5 kb (564.3 Mb/17,400=32.5 kb).
Besides the detection of the accumulation of Topo IIα in pericentromers, NORs and telomeres, and a less prominent amount along chromosome arms, PALM revealed the clustering of Topo IIα in these regions, possibly representing chromatin fiber looping centers. Given the role of Topo IIα in chromosome condensation and its reticular distribution along the chromosomal arms, it would be of interest to apply PALM to condensins, the key components in chromosome organization (Hirano 2012). Walther et al. (2018) detected by fluorescence correlation spectroscopy ∼195,000 condensin I and ∼35,000 condensin II complexes in HeLa chromosomes. Determining the number and distribution of condensins also in barley chromosomes would improve the understanding of the mitotic condensation process.
A positive correlation exists between kinetochore and chromosome size, and the adequate number of attached microtubule spindle fibers are important for correct chromosome segregation during cell division. Larger chromosomes require more microtubules, and thus larger kinetochores to move them with the same velocity as small ones (Nicklas 1965;Plačková et al. 2022). The microtubule-binding capacity increases with kinetochore size in Indian muntjac chromosomes (Drpic et al. 2018) and in rat-kangaroo PtK1 cells, and it was demonstrated that the chromosome size determines the number of microtubules (McEwen et al. 1998). In grass species, the anti-CENH3 signal size is strongly correlated with genome size. Species with large genomes and few chromosomes have the largest centromeres (e.g., rye), while species with small genomes and many chromosomes have the smallest centromeres (e.g., rice) (Zhang and Dawe 2012). Although not as obvious as between species, a positive correlation between kinetochore size and chromosome size was also observed in human (Irvine et al. 2004) and maize (Wang et al. 2021), and within bimodal karyotypes as demonstrated for Agavoideae species (Plačková et al. 2022).
We determined ~27,000 CENH3 molecules per centromere for the relatively large barley chromosomes. For comparison, it will be interesting to elucidate the CENH3 amount in small chromosomes by PALM/SMLM.

Microscopy
The fluorescence signals of Topo IIα and CENH3 were imaged by wide-field (WF), deconvolution (DCV) of WF, and superresolution 3D-SIM, using an Elyra PS.1 microscope system equipped with a 63×/1.4 Oil Plan-Apochromat objective and the software ZENBlack (Carl Zeiss GmbH). Images were captured separately for DAPI and Alexa488 using 405 nm and 488 lasers for excitation and appropriate emission filters. Reconstruction of SIM images was done with the ZENBlack software structured illumination processing module. 3D-PALM was performed with the 488 laser and the images were processed with the ZENBlack software PALM processing module. The localization precision in 3D-PALM was calculated via simulations of the experimental point-spread function (Weisshart et al. 2016;Kubalová et al. 2021b). The localization precision is the standard deviation of the data fit. Therefore, it describes the certainty of the localized position or likewise the area within which the molecule is positioned with high likelihood. Determining the resolution is not straightforward and would require the spacing of the labeled molecules. If the spacing of molecules is at least twice as fine as the localization precision, the latter represents according to the Nyquist criterion the resolution. Otherwise, the resolution is twice the spacing. As spacing normally is not known, one has to resort to taking the profile between two structures as described in Kubalová et al. (2021b) to determine the resolution at a specific site. The resolution can be quite different in various areas of the image.
3D rendering of SIM and PALM image stacks to produce movies was performed with the Imaris 9.7 software (Bitplane).

Acknowledgements
We thank Petr Cápal (Olomouc, Czech Republic) for the flow-sorting of chromosomes and Ingo Schubert for critical reading of the manuscript.
Author contribution VS and IK designed the work. IK performed slide preparations. AH provided the CENH3 antibodies. VS performed microscopy. VS and KW analyzed the data. VS and IK wrote the main manuscript text. All authors reviewed the manuscript.
Funding Open Access funding enabled and organized by Projekt DEAL. This work was supported by the Deutsche Forschungsgemeinschaft (Schu 762/11-1). Costs for open access publishing were partially funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation, grant 491250510).
Data availability There are no additional data and material available.

Ethics approval
No humans or animals involved to collect research data.

Conflict of interest
The authors declare no competing interests.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.