In situ sequencing for RNA analysis in preserved tissue and cells

Journal name:
Nature Methods
Volume:
10,
Pages:
857–860
Year published:
DOI:
doi:10.1038/nmeth.2563
Received
Accepted
Published online

Tissue gene expression profiling is performed on homogenates or on populations of isolated single cells to resolve molecular states of different cell types. In both approaches, histological context is lost. We have developed an in situ sequencing method for parallel targeted analysis of short RNA fragments in morphologically preserved cells and tissue. We demonstrate in situ sequencing of point mutations and multiplexed gene expression profiling in human breast cancer tissue sections.

At a glance

Figures

  1. Procedure for targeted in situ sequencing.
    Figure 1: Procedure for targeted in situ sequencing.

    (a) mRNA is copied to cDNA by reverse transcription (i), which is followed by degradation of the mRNA strand by RNase H (ii). A padlock probe is hybridized to the cDNA strand. (iii) For gap-targeted sequencing, the padlock probe binds to the cDNA with a gap between the probe ends over the bases that are targeted for sequencing by ligation. This gap is filled by DNA polymerization and DNA ligation to create a DNA circle. (iv) For barcode-targeted sequencing, DNA circularization of a padlock probe carrying a barcode sequence is carried out by ligation only. (v) The DNA circle is amplified by target-primed RCA generating an RCP that is subjected to sequencing by ligation. An anchor primer is hybridized next to the targeted sequence (red fragment) before the ligation of interrogation probes, which consists of four libraries of 9-mers, with eight random positions (N) and one fixed position (A, C, G or T); each library is labeled with one of four fluorescent dyes. The interrogation probe with best match at the fixed position is incorporated by ligation, along with the corresponding fluorophore. (b) The sample is imaged, and each RCP displays the color corresponding to the matched base. The interrogation probe is washed away before the application of interrogation probes for the next base. These steps of ligation, imaging and washing are iterated until the desired number of bases has been read. Enlarged views of five RCPs are shown for each sequencing cycle. The illustration is based on actual data. Scale bar, 50 μm. (c) Base-calling is done by image analysis recording the fluorescence staining patterns across sequencing cycles.

  2. In situ sequencing of fragments of ACTB and HER2 mRNA in breast cancer tissue.
    Figure 2: In situ sequencing of fragments of ACTB and HER2 mRNA in breast cancer tissue.

    (a) Raw data showing the location of sequences called from a fresh-frozen breast cancer tissue section (blue, DAPI; red, general stain of sequence common to all probes). Scale bar, 25 μm. (b,c) Each diamond symbol represents a decoded sequence, color coded as shown. The white line was manually drawn to separate cancer cells from adjacent nonmalignant stroma. (d) The relative frequency of each sequence is quantified in normal and cancer tissue and is represented by a pie chart (the number in parentheses is the number of occurrences, and the total area is proportional to the total number of RCPs). The two most abundant sequences are from ACTB (light blue) and HER2 (brown) transcripts. Note that the other unexpected sequences differ by a single nucleotide and occur only once each.

  3. Gene expression profiling on an ER-negative breast cancer tissue section by barcode-targeted in situ sequencing.
    Figure 3: Gene expression profiling on an ER-negative breast cancer tissue section by barcode-targeted in situ sequencing.

    (a) Part of a hematoxylin-and-eosin (H&E)-stained fresh-frozen breast cancer tissue section that was subjected to sequencing. Scale bar, 100 μm. (b) Localization of each of the 31 detected barcodes is shown as a symbol on top of a fluorescence microscopy image showing the nuclei (blue). Each symbol represents a barcode sequence that corresponds to a specific transcript. (c) Map showing the local density of detected HER2 (magenta) and VIM (cyan) transcripts, plotted on top of the H&E stain image. White indicates coexpression of HER2 and VIM. Note that the molecular staining pattern aligns to the histological staining pattern. (d) Spatial location of 31 different transcripts in the boxed area in a. (e) Expression profiling of the same 31 transcripts in the HER2-positive region (in magenta) and VIM-positive region (cyan). Transcript counts in the VIM-positive region were normalized to those in the HER2-positive region.

References

  1. Levsky, J.M. & Singer, R.H. Trends Cell Biol. 13, 46 (2003).
  2. Wang, Z., Gerstein, M. & Snyder, M. Nat. Rev. Genet. 10, 5763 (2009).
  3. Bonner, R.F. et al. Science 278, 14811483 (1997).
  4. Dalerba, P. et al. Nat. Biotechnol. 29, 11201127 (2011).
  5. Navin, N. et al. Nature 472, 9094 (2011).
  6. Tang, F. et al. Nat. Methods 6, 377382 (2009).
  7. Hou, Y. et al. Cell 148, 873885 (2012).
  8. Xu, X. et al. Cell 148, 886895 (2012).
  9. Nilsson, M. et al. Science 265, 20852088 (1994).
  10. Banér, J., Nilsson, M., Mendel-Hartvig, M. & Landegren, U. Nucleic Acids Res. 26, 50735078 (1998).
  11. Larsson, C., Grundberg, I., Soderberg, O. & Nilsson, M. Nat. Methods 7, 395397 (2010).
  12. Shendure, J. et al. Science 309, 17281732 (2005).
  13. Drmanac, R. et al. Science 327, 7881 (2010).
  14. Kamentsky, L. et al. Bioinformatics 27, 11791180 (2011).
  15. Thévenaz, P., Ruttimann, U.E. & Unser, M. IEEE Trans. Image Process. 7, 2741 (1998).
  16. Sparano, J.A. & Paik, S. J. Clin. Oncol. 26, 721728 (2008).
  17. Wang, E.T. et al. Nature 456, 470476 (2008).

Download references

Author information

  1. These authors contributed equally to this work.

    • Rongqin Ke &
    • Marco Mignardi

Affiliations

  1. Science for Life Laboratory, Department of Biochemistry and Biophysics, Stockholm University, Stockholm, Sweden.

    • Rongqin Ke,
    • Marco Mignardi,
    • Jessica Svedlund &
    • Mats Nilsson
  2. Department of Immunology, Genetics, and Pathology, the Rudbeck Laboratory, Uppsala University, Uppsala, Sweden.

    • Rongqin Ke,
    • Marco Mignardi,
    • Johan Botling &
    • Mats Nilsson
  3. Science for Life Laboratory, Centre for Image Analysis, Uppsala University, Uppsala, Sweden.

    • Alexandra Pacureanu &
    • Carolina Wählby
  4. Imaging Platform, Broad Institute of Harvard and Massachusetts Institute of Technology, Cambridge, Massachusetts, USA.

    • Carolina Wählby

Contributions

R.K. and M.M. designed and performed the experiments. J.B. provided tissue sections and pathology examination of the tissue. C.W. designed the image analysis pipelines and performed the image analysis together with A.P. J.S. performed the correlation between in situ sequencing and RNA-seq data. R.K., M.M., C.W. and M.N. wrote the manuscript. All authors commented on and revised the manuscript. M.N. conceived the idea and supervised the project.

Competing financial interests

M.N. owns shares in the company Olink AB, Uppsala, Sweden, which holds patents whose value may be affected by publication of these results.

Corresponding authors

Correspondence to:

Author details

Supplementary information

PDF files

  1. Supplementary Text and Figures (4 MB)

    Supplementary Figures 1–15, Supplementary Tables 1–7 and Supplementary Notes 1 and 2

Additional data