(C) PLOS One
This story was originally published by PLOS One and is unaltered.
 . . . . . . . . . .



Formamide denaturation of double-stranded DNA for fluorescence in situ hybridization (FISH) distorts nanoscale chromatin structure [1]

['Anne R. Shim', 'Department Of Biomedical Engineering', 'Northwestern University', 'Evanston', 'Illinois', 'United States Of America', 'Center For Physical Genomics', 'Engineering', 'Chemistry Of Life Processes Institute', 'Jane Frederick']

Date: 2024-06 

As imaging techniques rapidly evolve to probe nanoscale genome organization at higher resolution, it is critical to consider how the reagents and procedures involved in sample preparation affect chromatin at the relevant length scales. Here, we investigate the effects of fluorescent labeling of DNA sequences within chromatin using the gold standard technique of three-dimensional fluorescence in situ hybridization (3D FISH). The chemical reagents involved in the 3D FISH protocol, specifically formamide, cause significant alterations to the sub-200 nm (sub-Mbp) chromatin structure. Alternatively, two labeling methods that do not rely on formamide denaturation, resolution after single-strand exonuclease resection (RASER)-FISH and clustered regularly interspaced short palindromic repeats (CRISPR)-Sirius, had minimal impact on the three-dimensional organization of chromatin. We present a polymer physics-based analysis of these protocols with guidelines for their interpretation when assessing chromatin structure using currently available techniques.

Funding: This work was supported by funding from the National Science Foundation under the NSF EFRI (EFMA-1830961 awarded to PI V.B. and co-PI I.S.), the National Institutes of Health through the National Cancer Institute (R01CA228272 awarded to V.B. and I.S., R01CA225002 awarded to V.B., U54CA268084 awarded to PI V.B. and co-PI I.S., and U54CA261964 awarded to core lead V.B.), the Center for Physical Genomics and Engineering at Northwestern University, and philanthropic support from K. Hudson and R. Goldman, S. Brice and J. Esteve, M. E. Holliday and I. Schneider, the Christina Carinato Charitable Foundation, and D. Sachs. This material is based upon work performed by J.F. which was supported by the National Science Foundation Graduate Research Fellowship under Grant No. DGE-1842165. Any opinions, findings, conclusions, or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation. The funders had no role in the study design, data collection, data analysis, the decision to publish, or the preparation of the manuscript.

Prior work with both fluorescence and electron microscopy has demonstrated that the ultrastructure of the genome is sensitive to such labeling methods [ 26 – 28 ], but how these protocols relate to the structure of live cells is poorly understood. To address this knowledge gap, we investigate global nuclear chromatin structure before and after labeling with a gold-standard loci-specific method, 3D DNA FISH [ 29 ], to identify whether existing protocols are disrupting the native chromatin structure. Further, we probe which specific reagents and steps in the 3D FISH sample preparation protocol are deleterious to chromatin structure across the nucleus. We find that formamide exposure is the predominant cause of widespread alterations to chromatin domains. Finally, we evaluate other DNA-sequence labeling methods that do not require denaturation of double-stranded DNA, such as RASER-FISH [ 30 ] and CRISPR-Sirius [ 31 ], and show that these protocols are far less damaging to chromatin packing domains when compared to 3D FISH.

As a reflection of the considerable interest in understanding how aberrant genome organization leads to disease, there has been a massive expansion in imaging technologies and sample preparation protocols that allow chromatin to be visualized at super-resolution and smaller length scales [ 10 , 20 – 25 ]. The most common methods employ fluorescence in situ hybridization (FISH) techniques or CRISPR/dCas9-based labeling strategies to identify the locations of specific loci of interest. DNA (~1 nm) and histones (~10 nm) are well below the diffraction limit of light (~200 nm) and therefore require labeling with fluorescent proteins or other large probes [ 10 ]. These specialized labeling techniques could perturb chromatin structure, which poses a paradoxical challenge: it is only with labels that a locus can be identified, but the chromatin structure of the labeled locus is not necessarily representative of its unlabeled, native state.

Chromatin organizes into hierarchical structures associated with maintaining cell-type-specific gene expression. At the nucleosomal scale (<1 Kbp [ 1 ]), epigenetic modifications control access for transcriptional proteins to genes of interest; at the level of topologically associating domains (TADs, 100 Kbp– 1 Mbp [ 2 , 3 ]), architectural proteins maintain proximity between linearly distant enhancers and promoters; and at the level of A/B compartments (1–100 Mbp [ 4 , 5 ]), the phase separation between gene-rich and gene-poor regions of chromatin may increase interactions between genes and proteins [ 6 ]. Recent work has established the existence of chromatin domains with sizes between the nucleosome and TAD length scales that are functionally important for regulating transcription [ 7 – 9 ]. Within chromatin packing domains, the chromatin polymer has a fractal-like behavior which is quantified by the power-law scaling parameter, D (M~r D , where M is mass, quantified for chromatin by the number of base pairs, and r is the radius of the spherical volume under investigation) which is related to the contact scaling s observed in high-throughput conformation capture and oligo-paint measures of TADs [ 8 , 10 , 11 ]. Changes in chromatin scaling behavior have been correlated with altered gene expression patterns [ 12 – 14 ], phenotypic plasticity [ 14 ], carcinogenesis [ 15 – 17 ], chemotherapeutic efficacy [ 12 , 14 ], and reduced survival in cancer patients [ 14 , 18 , 19 ].

Results

As chromatin packing domains have an average radius of ~80 nm [11], the domain conformation within these subdiffractional structures cannot be effectively resolved by most conventional optical techniques. However, though the structure of individual domains cannot be resolved, it is possible to obtain information about their sub-diffractional organization in live cells by utilizing label-free nanoscopic sensing modalities such as Partial Wave Spectroscopic (PWS) microscopy [8,32]. In each diffraction-limited pixel, PWS microscopy captures the signal originating from the heterogeneity of the nanoscale molecular density distribution [32]. Due to its sensitivity to length scales between 20–200 nm (kin to sub-Mbp length scales) [33], the resulting signal is sensitive to changes in conformation within chromatin packing domains down to the size of the chromatin chain [11,21]. As a result, PWS microscopy offers several advantages over traditional super-resolution microscopy that are vital for this study: 1) it is label-free, allowing for careful study of nanoscale chromatin organization without perturbation; 2) both live and fixed cells can be imaged, which provides an opportunity to compare labeling protocols requiring fixation to the original, native chromatin structure; and 3) due to the fast acquisition time (<5 seconds) and a wide field of view (~10,000 μm2), it is a high-throughput technique that provides increased statistical power by simultaneously imaging hundreds of cells within minutes [8,32].

Importantly, PWS microscopy is sensitive to length scales containing chromatin packing domains [32,33], which are spatially separable regions that follow power-law scaling behavior [8,11]. Analysis of PWS microscopy images allows for the identification of chromatin packing domains and the measurement of the statistics of the chromatin chain conformation within these domains [8,32]. In polymer physics, the space a polymer occupies (the radius of gyration, r g ) can be related to the size of the polymer (the number of bond segments, N) through the Flory exponent, ν, which describes the scaling behavior ( ) [34]. As chromatin is a complex polymer, within a chromatin packing domain the mass of the chromatin contained within the domain M is related to the size of the domain r through the scaling exponent D (M~rD). Chromatin packing domains were first identified in electron microscopy images by finding DNA-dense domain centers, and then analyzing where the domain stops following power-law scaling to determine the domain boundary [8,11]. Because the only factor that identifies a chromatin packing domain is the ability to follow power-law scaling, every packing domain can have a unique mass (M), radius (r), or scaling behavior (D) [8,11]. Given that D is the inverse of ν, in the absence of spatial constraints D can theoretically range from 5/3, if the chromatin within a packing domain exhibits an expanded coil state (self-avoiding walk) [35], to 3, for chromatin organized into a compact globule state [4,36]. However, polymers in a Θ solvent, such as chromatin, have the properties of a random walk (D ~ 2) [37,38], therefore values below this are not a feasible model for chromatin organization within the nucleus [39,40], limiting the theoretical range to 2 < D < 3 [41,42]. Consistent with this finding, past studies quantified that the typical physiological range of D is 2.2–2.8, depending on the cell line [8,11,43].

[END]
---
[1] Url: https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0301000

Published and (C) by PLOS One
Content appears here under this condition or license: Creative Commons - Attribution BY 4.0.

via Magical.Fish Gopher News Feeds:
gopher://magical.fish/1/feeds/news/plosone/