Aug 23, 2024

Methods for human brainstem tissue processing and capture for spatial transcriptomics using 10x-Genomics Visium Spatial platform. V.2

DOI
https://dx.doi.org/10.17504/protocols.io.dm6gpzm3dlzp/v2
  • 1University of Sydney
courtney.wright Wright
Icon indicating open access to content
QR code linking to this content
DOI: https://dx.doi.org/10.17504/protocols.io.dm6gpzm3dlzp/v2
Protocol Citation: Sandy S. Pineda, Hongyun Li, Zhixiang Wu, Amy Burke, Ping Wu, YuHong Fu, Glenda Halliday 2024. Methods for human brainstem tissue processing and capture for spatial transcriptomics using 10x-Genomics Visium Spatial platform.. protocols.io https://dx.doi.org/10.17504/protocols.io.dm6gpzm3dlzp/v2Version created by courtney.wright Wright
License: This is an open access  protocol  distributed under the terms of the  Creative Commons Attribution License,  which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: Working
We use this protocol and it's working
Created: August 23, 2024
Last Modified: August 23, 2024
Protocol  Integer ID: 106313
Keywords: ASAPCRN, human post-mortem, spatial transcriptomics, 10X Genomics, sample preparation, human tissue samples for spatial transcriptomic, capture for spatial transcriptomic, genomics visium spatial platform, spatial transcriptomic, 10x genomics visium platform, using 10x genomics visium platform, methods for human brainstem tissue processing, genomics visium platform, 10x genomic, human brainstem tissue processing, fixed human tissue sample, tissue processing, preparing formalin
Funders Acknowledgements:
Michael J Fox Foundation
Grant ID: ASAP-020529
Michael J Fox Foundation
Grant ID: ASAP-020505
Abstract
This protocol details the process of preparing formalin fixed human tissue samples for spatial transcriptomics using 10X Genomics Visium platform
Materials
Consumables
  • RNAse AWAY (Invitrogen #10594063)
  • RNA/DNAse free plastic ware
  • Lo-bind RNAse/DNAse free tubes
  • Poly-prep slides (Millipore-Sigma, P0425)
  • haematoxylin
  • eosin



Hardware
  • Rotary microtome - HistoCore MULTICUT R Rotary (Leica Biosystems)
  • NovaSeq6000sequencing system (Illumina ´Corporation, USA).
Before start
To Note

  • All surfaces and equipment were decontaminated in advance with RNAse AWAY, surface decontaminant

  • The protocols described below require RNA/DNAse free plastic-ware and Lo-bind RNAse/DNAse free tubes.

  • All buffers and water used in these protocols are RNAse/DNAse free.

  • Consult all available 10x Genomics guides for detailed equipment models and suggested swaps (CytAssist workflow guides: CG000518, CG000520, CG000521, CG000495). Links in the references tab.
Tissue preparation
3h
  1. Selected formalin fixed & paraffin embedded (FFPE) blocks of human post-mortem, midbrains and pons were cut in a rotary microtome at 6 µm thickness.
  2. Two or three tissue sections were cut and placed on each poly-prep slides. Slides were dried in an oven for 03:00:00 at 42 °C and placed in desiccant, to avoid any moisture forming.
  3. Mounted slides were transported to the Garvan Cellular Genomics Platform for further processing (Garvan Institute of Medical Research, Sydney, Australia).
  4. Subsequently, tissue sections were deparaffinised, stained with Haematoxylin and Eosin (H&E) following the standard CytAssist workflow (CG000520).
  5. Slides were then coverslipped and imaged using a DM6B upright microscope (Leica Microsystems)) at 10x magnification (CytAssist, Spatial applications imaging CG000521).
  6. Images were captured, stitched and exported in low and high-resolution.


3h
RNA hybridisation
  1. Slides were then submerged in water to gently remove the coverslip and were immediately de-stained and de-crosslinked.
  2. RNA probes were then added and hybridised overnight.
  3. Slides containing the de-crosslinked and hybridised sections, along with the Visium CytAssist Spatial GeneExpression 11x11mm slides (Visium v5 slide-FFPE v2), were transferred to the CytAssist instrument to continue with the Gene Expression Capture workflow, all steps were followed as described in the 10X user guide GC000495.
  4. After capturing the probes, cDNA libraries were pre-amplified and cleaned, followed by a qPCR test, to determine the number of optimal cycles to use to construct the final libraries. Amplification cycles ranged from 13-20, for all the samples processed. cDNA libraries were constructed individually and then pooled in equimolar amounts prior to sequencing.
  5. Library pools were sequenced using a NovaSeq6000 sequencing system (Illumina corporation, USA).
  6. FASTQ files were inspected using FastQC (v0.12.1) to ensure reads were of consistent high quality, including high Q30 rates.
Image alignment and analysis
  1. Images, both in high resolution TIFF + CytAssist TIFF, were loaded into 10X’s Loupe browser (v7) using the built-in Visium CytAssist Image Alignment tool and each slide was manually aligned. Since multiple tissue sections were placed in each Visium window, following fiducial alignment, individual tissue sections on each slide were selected one at the time using the Loupe browser’s spot selection tool and each tissue section alignment was exported as a separate JSON file.
  2. Space Ranger v2.1.1 was run once for each tissue section with default settings and using the transcriptomic reference GRCh38-2020-A and Visium_Human_Transcriptome_Probe_Set_v2.0_GRCh38-2020-A.csv probe set.
  3. The Space Ranger web summaries were inspected for additional quality control metrics, including high rates of valid barcodes and UMIs, high sequencing saturation, and high mapping rates to the probe set.
  4. Space Ranger filtered matrixes were used for downstream analysis using a combination of Seurat v5.1.0 and Semla v1.1.6 {Hao, et. al. 2024; Larsson, et. al. 2023}. Briefly, for each sample an individual Semla object was created.
  5. Known anatomical annotations were applied to each tissue section using a combination of known Immunohistochemistry markers (from nearby-sections) that roughly delineate each region in the midbrain (SNM, SNL, SND, SNV, VTA, SNR, RN and region), and pons (LC, region).
  6. The FeatureViewer function in Semla was used to manually select spots belonging to each anatomical region. All cases were manually annotated and subsequently merged. In conjunction with Semla, a Seurat object with all the cases was merged separately and metadata from Semla was added utilising the AddMetadata function.     
Protocol references
CytAssist workflow guides

CG000518 - CG000518_Demonstrated_Protocol_VisiumCytAssistSpatialProtocolsFFPE_TissuePreparationGuide_RevD.pdf (10xgenomics.com)

CG000520 - CG000520_Demonstrated_Protocol_VisiumCytAssist_Deparaffin_H_E_RevC.pdf (10xgenomics.com)

CG000521 - CG000521_VisiumCytAssistImagingGuidelinesTN_RevD.pdf (10xgenomics.com)

CG000495 - My Document (10xgenomics.com)


  • Hao et. al. 2024 Dictionary learning for integrative, multimodal and scalable single-cell analysis. PMID: 37231261
  • Larsson et. al. 2023. Semla: a versatile toolkit for spatially resolved transcriptomics analysis and visualization. PMID: 37846051