Waiting
Login processing...

Trial ends in Request Full Access Tell Your Colleague About Jove

Cancer Research

Spatial Profiling of Protein and RNA Expression in Tissue: An Approach to Fine-tune Virtual Microdissection

Published: July 6, 2022 doi: 10.3791/62651

Summary

Here, we describe a protocol for fine-tuning regions of interest (ROIs) for Spatial Omics technologies to better characterize the tumor microenvironment and identify specific cell populations. For proteomics assays, automated customized protocols can guide ROI selection, while transcriptomics assays can be fine-tuned utilizing ROIs as small as 50 µm.

Abstract

Multiplexing enables the assessment of several markers on the same tissue while providing spatial context. Spatial Omics technologies allow both protein and RNA multiplexing by leveraging photo-cleavable oligo-tagged antibodies and probes, respectively. Oligos are cleaved and quantified from specific regions across the tissue to elucidate the underlying biology. Here, the study demonstrates that automated custom antibody visualization protocols can be utilized to guide ROI selection in conjunction with spatial proteomics assays. This specific method did not show acceptable performance with spatial transcriptomics assays. The protocol describes the development of a 3-plex immunofluorescent (IF) assay for marker visualization on an automated platform, using tyramide signal amplification (TSA) to amplify the fluorescent signal from a given protein target and increase the antibody pool to choose from. The visualization protocol was automated using a thoroughly validated 3-plex assay to ensure quality and reproducibility. In addition, the exchange of DAPI for SYTO dyes was evaluated to allow imaging of TSA-based IF assays on the spatial profiling platform. Additionally, we tested the ability of selecting small ROIs using the spatial transcriptomics assay to allow the investigation of highly-specific areas of interest (e.g., areas enriched for a given cell type). ROIs of 50 µm and 300 µm diameter were collected, which corresponds to approximately 15 cells and 100 cells, respectively. Samples were made into libraries and sequenced to investigate the capability to detect signals from small ROIs and profile-specific regions of the tissue. We determined that spatial proteomics technologies highly benefit from automated, standardized protocols to guide ROI selection. While this automated visualization protocol was not compatible with spatial transcriptomics assays, we were able to test and confirm that specific cell populations can successfully be detected even in small ROIs with the standard manual visualization protocol.

Introduction

Advances in multiplexing techniques continue to provide better characterization tools for targets present in tumors. The tumor microenvironment (TME) is a complex system of tumor cells, infiltrating immune cells, and stroma, where spatial information is critical to better understand and interpret mechanisms of interaction between biomarkers of interest1. With emerging techniques such as the GeoMx Digital Spatial Profiler (DSP) and 10x Visium, multiple targets can be detected and quantified simultaneously within their spatial context. The use of immunofluorescence protocols that facilitate tissue visualization can further improve the spatial profiling capabilities of these technologies.

The Spatial Omics technology we focused on for this method development consists of spatial proteomics and transcriptomics assays where oligonucleotides are attached to antibodies or RNA probes via a UV-sensitive photocleavable linker. Histological slides are labeled with these oligo-conjugated antibodies or probes and then imaged on the spatial profiling platform. Next, ROIs of different sizes and shapes are selected for illumination, and the photocleaved oligonucleotides are aspirated and collected in a 96-well plate. The photocleaved oligonucleotides are prepared to be quantified with either the Nanostring nCounter system or Next Generation Sequencing (NGS)2,3 (Figure 1)4,5.

Cell distributions vary within tissues, and the ability to characterize specific locations of cells using selected markers and different ROI sizes is of great importance to fully understand the tissue environment and identify specific features. In the Spatial Omics technology mentioned here, the standard visualization protocol uses directly conjugated antibodies and is a manual protocol. The standard markers to distinguish between tumor and stroma are panCytokeratin (panCK) and CD456,7, but additional markers are necessary to target specific cell populations of interest. Furthermore, the use of directly conjugated fluorescent antibodies lacks amplification, which limits antibody selection to abundant markers. Additionally, manual assays are subject to more variability than automated workflows8. Therefore, it is desirable to have a customizable, automated, and amplified visualization protocol for ROI selection.

Here, the study demonstrates that, for spatial proteomics assays, TSA technology can be used for visualization protocols on an automated platform resulting in a more targeted and standardized assay. In addition, TSA based assays enable the use of low-expressing markers, increasing the range of targets that can be selected for visualization. A 3-plex assay for panCK, FAP, and Antibody X was developed using an automated platform where panCK and FAP were used to differentiate between tumor and stroma, respectively. Antibody X is a stromal protein frequently encountered in tumors, but its biology and impact on anti-tumor immunity are not fully understood. Characterizing the immune contexture in areas rich in Antibody X can elucidate its role in anti-tumor immunity and therapeutic response, as well as its potential as a drug target.

While customized automated TSA visualization panels proved to be successful for spatial proteomics assays, the application of these assays for spatial transcriptomics assays could not be confirmed. This is most likely due to the reagents and the protocol used for the automated visualization protocols, which seem to compromise RNA integrity. Recognizing that an automated labeling protocol for visualization markers can be used for spatial proteomics assays but not for spatial transcriptomics assays provides important guidance on Spatial Omics technology assay designs.

Additionally, the study demonstrates that the spatial transcriptomics assay can be used to profile targets in regions as small as 50 µm in diameter, or approximately 15 cells. Two different-sized ROIs were selected to test the ability of the assay to also detect transcripts in small ROIs. For each region of interest, oligos corresponding to 1,800 mRNA targets were collected and made into libraries according to the spatial profiling platform protocol. Libraries were individually indexed, subsequently pooled, and sequenced. This allowed the evaluation of both pooling efficiency and the capability of identifying specific cell populations in small ROIs.

This paper shows that for spatial proteomics assays, an automated protocol to guide ROI selection on specific markers of interest can be used to selectively target the interrogation of relevant tissue areas and characterize the spatial environment of the tissue. Furthermore, we demonstrate that smaller ROIs can be used for spatial transcriptomic assays to detect and characterize specific cell populations.

Subscription Required. Please recommend JoVE to your librarian.

Protocol

All human tissues were acquired from commercial biobanks or accredited tissue banks under warranty that appropriate Institutional Review Board approval and informed consent were obtained.

NOTE: The protocol is performed using the Discovery Ultra and the GeoMx Digital Spatial Profiler. See the Table of Materials for details about reagents, equipment, and software used in this protocol.

1. Automated visualization protocol for spatial proteomics assays

  1. Program autostainer to apply fluorescent visualization antibodies
    1. In the autostainer software, click on the home button and choose Protocols. Then, click on Create/Edit Protocols and select RUO DISCOVERY Universal as the procedure.
    2. Click on First Sequence > Deparaffinization > Depar v2. For Medium Temperature, choose 72 Deg C, then click on Pretreatment and select Cell Conditioning and CC1 Reservoir. For Very High Temperature, choose 100 Deg C, then click on CC1 8 Min and continue clicking until CC1 64 Min is selected.
    3. Click on Inhibitor > DISCOVERY Inhibitor, and for Incubation Time, choose 8 Minutes. Then, click on Antibody > High Temp Ab Incubation; for Low Temperature, choose 37 Deg C; for Antibody, choose the antibody number from the dispenser label (Antibody 6 in this example); for Plus Incubation Time, select 32 Minutes.
      NOTE: This protocol has three different antibody numbers. The first antibody is FAP, the second is Pan-Cytokeratin, and the third is Antibody X.
    4. Click on Multimer HRP > Multimer HRP Blocker, then for Antibody Blocking, choose Gt Ig Block, and for Incubation Time, choose 4 Minutes. Next, on Multimer HRP Reagent, select OMap anti-Rb HRP, and for Incubation Time, choose 16 Minutes. Then, click on Cy5, and for Long Incubation Time, choose 0 Hr 8 Min.
    5. Click on Dual Sequence and choose Antibody Denaturation, then select Antibody Denature CC2-1, and for Very High Temperature, choose 100 Deg C. Ensure that incubation time is 8 min. Next, click on DS Inhibitor and select Neutralize.
    6. Click on DS Antibody, and for Very Low Temperature, choose 37 Deg C. Then, in Antibody, choose the antibody number from the dispenser label (Antibody 3 in this example), and for Plus Incubation Time, select 32 Minutes.
    7. Click on DS Multimer HRP and choose DS Multimer HRP Blocker. Then, for Antibody Blocking, select Gt Ig Block, and for Incubation Time, choose 4 Minutes. Next, on Multimer HRP Reagent, select OMap anti-Ms HRP, and for Incubation Time, choose 16 Minutes. Then, click on DS Rhodamine 6G, and for Long Incubation Time, choose 0 Hr 8 Min.
    8. Click on Triple Stain and select TS Antibody Denaturation, then select Antibody Denature CC2-2, and for Very High Temperature, select 100 Deg C. Ensure that incubation time is 8 min. Next, click on TS Inhibitor and select TS Neutralize.
    9. Click on TS Antibody, and for Very Low Temperature, select 37 Deg C. Then, in Antibody, select the antibody number from the dispenser label (Antibody 7 in this example), and for Plus Incubation Time, select 32 Minutes.
    10. Click on TS Multimer HRP and choose TS Multimer HRP Blocker. Then, for Antibody Blocking, select Gt Ig Block, and for Incubation Time, choose 4 Minutes. Next, on Multimer HRP Reagent, select OMap anti-Rb HRP, and for Incubation Time, choose 16 Minutes. Then, click on TS FAM, and for Long Incubation Time, choose 0 Hr 8 Min.
    11. Add a title to the protocol. Select a protocol number, add a comment, and click on Active followed by Save.
  2. Prepare slides, print labels, and start the autostainer run
    1. Bake vendor procured FFPE (Formalin-fixed, paraffin-embedded) human tissue sections in an oven set to 70 °C for 20-60 min. While slides are baking, print labels by clicking Create Label in the autostainer software. Next, click on Protocols, select the protocol number, and click on Close/Print. Add relevant information on the slide label and click on Print.
    2. Remove slides from the oven and let them cool down to room temperature (RT). Apply the previously printed protocol labels to the corresponding slides.
    3. Load refillable antibody dispensers with antibodies according to the antibody label number assigned in step 1.1. In this protocol, use Antibody 6 for FAP at 1 µg/mL, use Antibody 3 for Pan-Cytokeratin at 0.1 µg/mL, and use Antibody 7 for Antibody X at 2.5 µg/mL. Dilute each antibody in the specified diluent and prime the refillable antibody dispensers.
    4. Gather blocking, detection, and amplification pre-filled reagent dispensers mentioned above and place them on a reagent tray. Load slides in the slide trays (up to 30 slides per run) and click on Running followed by Yes. Note that DAPI is not used for nuclear counterstaining.
    5. Confirm the start of the run by checking the run duration on the autostainer software. The protocol takes ~11 h for completion and can run overnight.
    6. The next day, ensure run completion by observing green flashing lights on the slide tray slots. Then, take the slides off the instrument and rinse the slides vigorously in 1x Reaction Buffer until the liquid coverslip solution is completely removed.
  3. Spatial profiling platform protocol for proteomics assay
    1. Follow the spatial proteomics protocol indicated in the note below. Include the following changes to the protocol to enable the 3-plex automated labeling procedure.
    2. Perform steps 1.2.1-1.2.6 the day before starting the spatial proteomics protocol and move straight to the antigen retrieval step of the spatial proteomics protocol after performing step 1.2.6. Omit the visualization procedure outlined in the spatial profiling proteomics protocol.
    3. Replace SYTO 13 with SYTO 64 at 5000 nM to enable fluorophore integration. Dilute SYTO 64 in 1x TBS and incubate in a humidity chamber for 15 min.
    4. When setting up the scan parameters in the spatial profiling platform, select the filters and focusing channels according to the fluorophores used in the visualization panel to allow 3-plex integration. Use FITC for DISCOVERY FAM and set the exposure to 200 ms, use Cy3 for DISCOVERY Rhodamine 6G and set the exposure to 200 ms, use Texas Red for SYTO 64 and set the exposure to 50 ms (specify this as the focus channel), and finally, use Cy5 for DISCOVERY Cy5, set the exposure to 200 ms, and save changes.
      NOTE: Detailed instructions of the spatial profiling proteomics protocol are found on the official website Support tab by selecting Documentation > User Manuals. Here, look for the protein protocol for nCounter using Bond RX.

2. ROI selection for spatial transcriptomics assays

  1. Bake FFPE cell pellet sections in an oven set to 70 °C for 20-60 min. Follow the spatial profiling platform NGS protocol indicated in the note below and scan slides on the spatial profiling platform, selecting sizes of 50 µm and 300 µm. The following recommendation is highlighted for the spatial transcriptomics assay:
    1. When preparing the library, if various sizes of ROIs were selected, pool together ROIs of similar sizes to make one library per size. This is to ensure that sufficient sequencing depths are achieved for all ROIs without bias from size.
      NOTE: Detailed instructions of the spatial profiling platform are found on the official website Support tab by selecting Documentation > User Manuals. Here, look for the RNA protocol for NGS applications using Bond RX.

Subscription Required. Please recommend JoVE to your librarian.

Representative Results

Automated visualization protocol to guide ROI selection
In this paper, we present the use of an automated, custom TSA-based IF protocol to visualize the tissue and select specific ROIs. Visualization panel development using melanoma and human normal skin as control tissues consisted of epitope stability testing, fine-tuning of marker intensities, and bleedthrough assessment through leave one out controls. To test if epitope stability of the antibodies is affected by repetitive elution steps, FAP and Antibody X signals were qualitatively assessed after labeling in each position of the 3-plex. Antibody X signal remained consistent in all positions, while FAP signal intensity decreased after one elution (Figure 2). Previous internal studies showed that the panCK epitope is stable throughout a 3-plex assay. Therefore, FAP was chosen to be in the 1st position in the 3-plex, panCK was placed in the 2nd, and Antibody X in the 3rd position. Since FAP signal was weaker overall compared to Antibody X and panCK, FAP was paired with the brighter fluorophore, Cy5.

The automated 3-plex was optimized with leave one out controls, where each marker was omitted from the 3-plex protocol once to confirm that there was no bleedthrough into neighboring channels (Figure 3).

The fluorophores used on the automated staining platform were tested on the spatial profiling platform to confirm that the filters are compatible with the TSA fluorophores in the 3-plex visualization panel. Since the 3-plex protocol uses FITC, Cy3, and Cy5, and the DAPI channel is used for UV cleavage on the spatial profiling platform, a nuclear counterstain in the Texas Red channel had to be used. Two different nuclear dyes, SYTO 59 and SYTO 64, were tested, and the dyes showed a distinct nuclear signal in colon tissues. However, SYTO 59 also displayed some bleedthrough signal in the Cy5 channel while the Cy5 channel was clean when using SYTO 64 (Figure 4). Therefore, SYTO 64 was the preferred choice for nuclear detection.

Further testing of SYTO 64 showed that this dye was not photostable and photobleached quickly during imaging. Once the signal is bleached, the imaging system focuses on the background, which could be interpreted as a non-specific signal (Figure 5). Increasing the concentration of the nuclear dye to increase signal intensity helped minimize this issue.

Tissues labeled with the panCK/CD45 manual visualization protocol, which uses directly conjugated antibodies, were compared to tissues labeled with the panCK/FAP/Antibody X automated custom visualization protocol. Both protocols share panCK as a common marker, and the signal was comparable between the two protocols (Figure 6).

To ensure that a modified visualization protocol has no impact on the upstream spatial proteomics assay, we compared the count values obtained after Spatial Profiling Platform collection and Counting Platform processing when using the panCK/CD45 manual visualization protocol. This uses directly conjugated antibodies or the 3-plex automated custom visualization protocol, which uses TSA. Four colorectal cancer (CRC) tissues were labeled with the panCK/CD45 manual visualization protocol or the 3-plex automated custom visualization protocol in combination with the spatial proteomics assay. Similar ROIs were chosen for each protocol, and count data of the corresponding ROIs were compared using housekeeping normalization for S6 and Histone. Similar trends were observed between these ROIs as shown in the log2 heatmap (Figure 7A), with a Spearman R value of 0.88 for all markers included in the spatial proteomics assay (Figure 7B). Of the 31 antibodies used in the assays, 23 antibodies had a Spearman R value higher than or equal to 0.5.

As spatial proteomics and transcriptomic protocols differ, we evaluated if the automated visualization protocol could also be applied to spatial transcriptomics assays. The automated visualization protocol requires the IHC assay to be performed before RNA detection, whereas the original manual protocol detects RNA first before applying the visualization protocol. The 3-plex automated visualization protocol was tested side by side with the CK/CD45 manual visualization protocol in combination with the spatial transcriptomics assay. Sequencing data for the 3-plex automated visualization protocol Quartile 3 count (Q3), normalized on the spatial profiling software, presented lower values when compared to the CK/CD45 manual visualization protocol (Figure 8A), which is also reflected by the low Spearman R values of 0.15 (Figure 8B). Also, a loss of dynamic range was observed when using the 3-plex visualization automated protocol (Figure 8C).

This loss of counts with the automated visualization protocol indicates that this protocol compromises RNA integrity, and that detecting the RNA prior to performing the visualization assay is necessary to conserve RNA quality. Since this automated visualization panel requires the protein part to be performed prior to RNA detection, this protocol is not suitable for RNA assays.

Size of ROI selection in spatial transcriptomics assays
To better understand the impact of ROI size on RNA data output, RNA counts of different sized ROIs were compared. The spatial transcriptomics assay was performed on SUDHL1 and JURKAT cell lines to minimize differences inherent to tissues. Circular ROIs of 50 µm and 300 µm diameter were selected and compared to each other using Q3 normalization on the spatial profiling software. RNA counts for selected targets on different cell lines were comparable between the two ROI sizes after normalization (Figure 9A,B).

Figure 1
Figure 1: Digital Spatial Profiling workflow. (A) Slides are labeled with antibodies (spatial proteomics assay) or probes (spatial transcriptomics assay) as well as visualization markers. (B) Slides are placed on the spatial profiling platform to image and select ROIs. (C) Oligos are cleaved by UV light and collected in a 96-well plate. This process is repeated for every ROI selected. (D) The samples are processed, and data is generated using the counting platform or a sequencer. This figure has been modified from Nanostring technologies website4,5. Please click here to view a larger version of this figure.

Figure 2
Figure 2: Epitope stability. (A-C) Antibody X and (D-F) FAP tested on human normal skin and melanoma, respectively. Each marker in the (A,D) 1st, (B,E) 2nd, and (C,F) 3rd position of a 3-plex. Antibody X is stable in all positions, while (D) FAP shows decreasing signaling intensity after (E-F) one and two elutions, respectively. Antibody X in green, FAP in red, DAPI in grey. Scale bars are at 200 µm. Please click here to view a larger version of this figure.

Figure 3
Figure 3: Leave one out controls for the 3-plex automated custom visualization protocol on human pancreas tissues. For all images, Antibody X is depicted in green, panCK in cyan, and FAP in red. The first row shows the (A) 3-plex and the single channels for (B) Antibody X in FITC, (C) panCK in Cy3, and (D) FAP in Cy5. The second row shows the leave one out control 3-plex for (E) Antibody X and (F-H) the single channels for each marker. The third row shows the leave one out control 3-plex for (I) panCK and (J-L) the single channels for each marker. The fourth row shows the leave one out control 3-plex for (M) FAP and (N-P) the single channels for each marker. No bleedthrough was observed in any of the leave one out controls as indicated by lack of signal in the corresponding channels when tested (F) without FAP, (K) without panCK, and (P) without Antibody X. Scale bars are at 500 µm. Please click here to view a larger version of this figure.

Figure 4
Figure 4: SYTO dye testing. Texas Red channel displays the nuclear signal in Colon for (A) SYTO 59 with bleedthrough into (B) Cy5 channel (red). Texas Red channel shows the nuclear signal in Colon for (D) SYTO 64 without bleedthrough into (E) Cy5 channel. Texas Red and Cy5 channels in Colon displayed together to show (C) bleedthrough of SYTO 59 and (F) no bleedthrough for SYTO 64. Scale bars are at 50 µm. Please click here to view a larger version of this figure.

Figure 5
Figure 5: SYTO 64 photostability. Colon samples labeled with SYTO 64 (A) before and (B) after photobleaching. Image B shows the non-specific signal that becomes more visible after nuclear signal loss. Scale bars are at 50 µm. Please click here to view a larger version of this figure.

Figure 6
Figure 6: Comparison of panCK labeling. panCK signal on colorectal cancer (CRC) samples with (A) panCK/CD45 manual visualization protocol and (B) panCK/FAP/Antibody X automated visualization protocol. No differences in panCK signal were observed between the two protocols. Scale bars are at 200 µm. Please click here to view a larger version of this figure.

Figure 7
Figure 7: Comparison of count values for the spatial proteomics assay using two different visualization protocols on similar ROIs of four CRC tissues. (A) Log 2 Heatmap of all the markers included in the spatial proteomics assay comparing the manual panCK/CD45 protocol (black) to the automated 3-plex protocol (grey). (B) Log 2 Spearman correlation for all the markers presents an R value of 0.88. Statistical analysis was performed using GraphPad Prism 7 Software or with Python programming language. Please click here to view a larger version of this figure.

Figure 8
Figure 8: Comparison of count values for the spatial transcriptomics assay using two different visualization protocols. (A) Log 2 Heatmap of selected targets included in the spatial transcriptomics assay comparing the manual panCK/CD45 protocol (black) to the 3-plex automated protocol (grey). (B) Log 2 Spearman correlation for all the markers presents an R value of 0.15. (C) The mean (blue) and standard deviation where the dynamic range is lost in the 3-plex automated protocol. Statistical analysis was performed using GraphPad Prism 7 Software or with Python programming language. Please click here to view a larger version of this figure.

Figure 9
Figure 9: Comparison of RNA count values for the spatial transcriptomics assay on cell pellets. Comparable results were observed for 50 µm (black) and 300 µm (grey) diameter ROIs for selected targets on (A) SUDHL1 and (B) JURKAT cell lines. No significant difference was observed using t-test. Mean and standard deviation shown for n = 3 ROIs per size. Statistical analysis was performed using GraphPad Prism 7 Software. Please click here to view a larger version of this figure.

Subscription Required. Please recommend JoVE to your librarian.

Discussion

To date, directly conjugated fluorescent antibodies in a manual protocol are most commonly used as visualization panels for spatial proteomics or spatial transcriptomics assays9,10. However, the use of directly conjugated fluorescent antibodies can be challenging for less abundant markers, limiting the selection of suitable antibodies. This protocol shows that labeling of visualization markers can be automated on an automated staining platform using TSA technologies to support spatial proteomics assays. This allows for more flexibility in the customization of visualization panels, which enables a more targeted ROI selection and acquisition of relevant biological data from specific areas of a tissue. TSA technology also allows the visualization of low abundance markers, and thereby increases the number of potential suitable antibodies. Additionally, automation of the visualization protocol ensures reproducibility and quality of the labeling procedure.

Developing a 3-plex IF assay requires careful assessment of marker/fluorophore signal intensities on appropriate control tissues. It is recommended to titrate antibody concentrations to obtain the best signal-to-noise ratio, which will help to minimize bleedthrough into adjacent channels. Furthermore, epitope stability for each marker of interest should be tested to ensure that the epitope is not affected by repeated elution steps. Once the order of antibodies in the panel is confirmed, the assay needs to be fine-tuned using leave one out controls. Here, a specific marker is omitted in the panel, and the corresponding channel in the acquired image is reviewed for bleedthrough and cross-talk.

It was shown here that TSA-based assays can be imaged on the spatial profiling platform, and that DAPI alternatives need to be used for nuclear staining. We tested SYTO dyes and determined that when deciding on which SYTO dye to use, the signal needs to be carefully assessed to avoid bleedthrough into adjacent channels. Another limitation of the use of SYTO dyes is that some SYTO dyes photobleach quicker, such as SYTO 64 and SYTO 83 (internal data), which could affect nuclear quantification. Therefore, light exposure of the stained slides should be avoided as much as possible. Other groups have shown the use of SYTO 1311,12, which is more photostable (internal data) and could be used in a panel as long as the markers of interest are assigned to fluorophores that are compatible with the nuclear dye of choice.

While automated TSA based fluorescent assays worked well for visualization in combination with spatial proteomics assays, it was determined here that combining a 3-plex automated visualization protocol with a spatial transcriptomics protocol resulted in loss of dynamic range, which could be due to the harsher protocol conditions of adding extra antigen retrieval/elution steps that might affect RNA integrity. Therefore, the 3-plex automated visualization protocol should only be adapted for proteomics assays, which is a limitation and requires alternative visualization approaches for RNA-based assays. As an alternative to same-slide visualization, serial sections stained with markers of interest could be overlaid to facilitate ROI selection13. In addition, the literature describes the use of in situ hybridization on adjacent slides to guide ROI selection14. However, these approaches could be limited by tissue availability, section-to-section variability, and cell-to-cell registrations.

We also tested the usability of different ROI sizes for spatial transcriptomics assays. Using small ROIs of 50 µm enables the researcher to acquire data of specific regions allowing a more detailed readout of selected cells or areas of interest. Limitations of selecting small ROIs are that the number of cells present might not be abundant enough to get count values that can be distinguished from the background. One option is to select pure cell populations to capture less abundant targets. However, ROI selection for single cells is limited by high signal-to-noise ratio15. Furthermore, it is important to consider ROI sizes for library preparation if different sized ROIs are collected in a single experiment. If ROI sizes differ significantly, it may be beneficial to pool ROIs according to their sizes and construct separate libraries to prevent large ROIs from being over-represented in one library. This will also ensure that each ROI will get sufficient sequencing coverage according to its size.

Other factors to consider when performing Spatial Omics experiments are that the data could show variations due to slide-to-slide variability, and lower expression of certain targets can bring up background and differences in correlation values. Adequate controls and normalization approaches need to be considered when planning experiments and evaluating data. For spatial proteomics assays on nCounter, normalizing to housekeeping targets or to the negative control IgGs are the preferred methods according to internal experience and vendor recommendations where the correlation between the housekeeping targets or to the negative control IgGs needs to be evaluated16. However, depending on the tissues used, the best normalization needs to be defined for every study17,18. For example, literature states that for breast cancer Spatial Omics studies, GAPDH, which is commonly used for normalizing to housekeeping targets, has been less concordant than S6 and Histone19. Therefore, S6 and Histone have been the recommended housekeeping targets for normalization in breast cancer tissues19.

For spatial transcriptomics assays, normalization strategies need to be defined depending on the study design (e.g., tissue types and ROI selection). Some strategies that are used in RNA-Seq, such as upperquartile or trimmed mean of M values (TMM) normalization, could be applied to normalize sequenced data19,20.

Spatial transcriptomics assays provide information on thousands of targets and are becoming more commonly used to explore the whole transcriptome, where specific TME compartments are selected and analyzed in conjunction with single-nucleus RNA-seq20 or single-cell RNA sequencing21. We demonstrated that visualization protocols to guide ROI selection to assess the spatial distribution of targets could be automated for spatial proteomics assays but not for spatial transcriptomics assays. Additionally, we present that ROIs as small as 50 µm can be selected to allow a more in-depth investigation of a particular tissue region. Future applications of this technology will improve understanding of the tumor microenvironment.

Subscription Required. Please recommend JoVE to your librarian.

Disclosures

Veronica Ibarra-Lopez, Sangeeta Jayakar, Yeqing Angela Yang, Zora Modrusan, and Sandra Rost are employees and stockholders of Genentech, a member of the Roche Group. Other companies that are part of Roche produce reagents and instruments used in this manuscript. Ciara Martin is a full-time employee of NanoString Technologies Inc, which produces reagents and instruments used in this manuscript.

Acknowledgments

The authors acknowledge Thomas Wu for processing NGS files. We thank James Ziai for the results discussions and manuscript review and Meredith Triplet and Rachel Taylor for internal manuscript revision.

Materials

Name Company Catalog Number Comments
10x Tris buffered saline (TBS) Cell Signaling Technologies 12498S Diluted to 1x TBS in DEPC treated water
Antibody X (not disclosed) antibody blinded due to confidentiality
DEPC-treated water ThermoFisher AM9922 Another can be used
DISCOVERY Cell Conditioning ( CC1) Ventana 950-500
DISCOVERY Cy5 Kit Ventana 760-238 Referred as Cy5
DISCOVERY FAM Kit Ventana 760-243 Referred as FAM
DISCOVERY Goat Ig Block Ventana 760-6008 Referred as Gt Ig Block
DISCOVERY OmniMap anti-Ms HRP Ventana 760-4310 Referred as OMap anti-Ms HRP
DISCOVERY OmniMap anti-Rb HRP Ventana 760-4311 Referred as OMap anti-Rb HRP
DISCOVERY Rhodamine 6G Kit Ventana 760-244 Referred as Rhodamine 6G
DISCOVERY ULTRA Automated Slide Preparation System Ventana 05 987 750 001 / N750-DISU-FS Referred as autostainer on the manuscript
FAP [EPR20021] Antibody Abcam Ab207178
GeoMx Digital Spatial Profiler NanoString GMX-DSP-1Y Referred as spatial profiling platform on the manuscript
Humidity chamber Simport M920-2 Another can be used
Pan-Cytokeratin [AE1/AE3] Antibody Abcam Ab27988
ProLong Gold Antifade Mountant ThermoFisher P36934
Python Python Statistical analysis
Reaction Buffer (10x) Ventana 950-300
Statistical analysis software GraphPad Prism 7 Statistical analysis
SYTO 64 ThermoFisher S11346
ULTRA Cell Conditioning (ULTRA CC2) Ventana 950-223
Ventana Antibody Diluent with Casein Ventana 760-219 Referred as specified diluent on the manuscript
Ventana Primary antibody dispenser Ventana Catalog number depends on dispenser number

DOWNLOAD MATERIALS LIST

References

  1. Nerurkar, S. N., et al. Transcriptional spatial profiling of cancer tissues in the era of immunotherapy: The potential and promise. Cancers. 12, (9), 2572 (2020).
  2. Decalf, J., Albert, M. L., Ziai, J. New tools for pathology: a user's review of a highly multiplexed method for in situ analysis of protein and RNA expression in tissue. Journal of Pathology. 247, (5), 650-661 (2019).
  3. McGinnis, L. M., Ibarra-Lopez, V., Rost, S., Ziai, J. Clinical and research applications of multiplexed immunohistochemistry and in situ hybridization. Journal of Pathology. 254, (4), 405-417 (2021).
  4. NanoString. GeoMx Digital Spatial Profiler. Available from: https://nanostring.com/products/geomx-digital-spatial-profiler/geomx-dsp-overview (2022).
  5. NanoString. GeoMx Digital Spatial Profiler. Available from: https://nanostring.com/wp-content/uploads/BR_MK0981_GeoMx_Brochure_r19_FINAL_Single_WEB.pdf (2022).
  6. McCart Reed, A. E., et al. Digital spatial profiling application in breast cancer: a user's perspective. Virchows Arch: An International Journal of Pathology. 477, (6), 885-890 (2020).
  7. McNamara, K. L., et al. Spatial proteomic characterization of HER2-positive breast tumors through neoadjuvant therapy predicts response. Nature Cancer. 2, (4), 400-413 (2021).
  8. Kim, S. W., Roh, J., Park, C. S. Immunohistochemistry for pathologists: Protocols, pitfalls, and tips. Journal of Pathology and Translational Medicine. 50, (6), 411-418 (2016).
  9. Muñoz, N. M., et al. Molecularly targeted photothermal ablation improves tumor specificity and immune modulation in a rat model of hepatocellular carcinoma. Communications Biology. 3, (1), 783 (2020).
  10. Coleman, M., et al. Hyaluronidase impairs neutrophil function and promotes Group B Streptococcus invasion and preterm labor in nonhuman primates. mBio. 12, (1), 03115-03120 (2021).
  11. Gupta, S., Zugazagoitia, J., Martinez-Morilla, S., Fuhrman, K., Rimm, D. L. Digital quantitative assessment of PD-L1 using digital spatial profiling. Laboratory Investigation; A Journal of Technical Methods and Pathology. 100, (10), 1311-1317 (2020).
  12. Carter, J. M., et al. Characteristics and spatially defined immune (micro)landscapes of early-stage PD-L1-positive triple-negative breast cancer. Clinical Cancer Research. 27, (20), 5628-5637 (2021).
  13. Busse, A., et al. Immunoprofiling in neuroendocrine neoplasms unveil immunosuppressive microenvironment. Cancers. 12, (11), 3448 (2020).
  14. Kulasinghe, A., et al. Profiling of lung SARS-CoV-2 and influenza virus infection dissects virus-specific host responses and gene signatures. European Respiratory Journal. 59, (1), (2021).
  15. Li, X., Wang, C. Y. From bulk, single-cell to spatial RNA sequencing. International Journal of Oral Science. 13, (36), (2021).
  16. Merritt, C. R., et al. Multiplex digital spatial profiling of proteins and RNA in fixed tissue. Nature Biotechnology. 38, (5), 586-599 (2020).
  17. Van, T. M., Blank, C. U. A user's perspective on GeoMxTM digital spatial profiling. Immuno-Oncology Technology. 1, 11-18 (2019).
  18. Introduction to GeoMx Normalization: Protein. White Paper. Nanostring. Available from: https://nanostring.com/wp-content/uploads/MK2593_GeoMx_Normalization-Protein.pdf (2020).
  19. Bergholtz, H., et al. Best practices for spatial profiling for breast cancer research with the geomx spatial profiler. Cancers. 13, (17), 4456 (2021).
  20. Hwang, W. L., et al. Single-nucleus and spatial transcriptomics of archival pancreatic cancer reveals multi-compartment reprogramming after neoadjuvant treatment. BioRxiv. (2020).
  21. Jerby-Arnon, L., et al. Opposing immune and genetic mechanisms shape oncogenic programs in synovial sarcoma. Nature Medicine. 27, (2), 289-300 (2021).
This article has been published
Video Coming Soon
PDF DOI DOWNLOAD MATERIALS LIST

Cite this Article

Ibarra-Lopez, V., Jayakar, S., Yang, Y. A., Martin, C., Modrusan, Z., Rost, S. Spatial Profiling of Protein and RNA Expression in Tissue: An Approach to Fine-tune Virtual Microdissection. J. Vis. Exp. (185), e62651, doi:10.3791/62651 (2022).More

Ibarra-Lopez, V., Jayakar, S., Yang, Y. A., Martin, C., Modrusan, Z., Rost, S. Spatial Profiling of Protein and RNA Expression in Tissue: An Approach to Fine-tune Virtual Microdissection. J. Vis. Exp. (185), e62651, doi:10.3791/62651 (2022).

Less
Copy Citation Download Citation Reprints and Permissions
View Video

Get cutting-edge science videos from JoVE sent straight to your inbox every month.

Waiting X
Simple Hit Counter