to BioTechniques free email alert service to receive content updates.
Microdissection of spatially identified single nuclei in a solid tumor for single cell whole genome sequencing
 
Yan Guo*1,2, Yi Yang*1, Juan Zhou1, Daniel M. Czajkowsky1, Bingya Liu1,3, and Zhifeng Shao1,2, 3
1Ministry of Education Key Laboratory of Systems Biomedicine, Shanghai Jiao Tong University, Shanghai, P R China
2State Key Laboratory of Oncogenes and Related Genes, Shanghai Jiao Tong University, Shanghai, P R China
3Shanghai Key Laboratory of Gastric Neoplasms, Shanghai Institute of Digestive Surgery, Shanghai Jiao Tong University, Shanghai, P R China


*Y.G. and Y.Y. contributed equally to this work
BioTechniques, Vol. , No. , April 2012, pp. 1–3
Full Text (PDF)
Supplementary Material
Method Summary

Here we present a method to extract single nuclei for whole genome sequencing applications using a combination of through-focus analysis, laser capture microdissection, and whole genome amplification. Our method should enable studies on genome variation in single cells from spatially defined positions.

Abstract

The relative spatial distribution of cells in a solid tumor contributes to development of malignancy, yet the details of this process remain poorly understood. To elucidate these mechanisms, the ability to extract and analyze the entire DNA content of individual cells whose precise location in the tumor is known is required, yet such methodology has not yet been described. Here we detail a procedure to directly extract complete individual nuclei from fixed-frozen tissue sections using through-focus analysis coupled with laser microdissection, followed by whole genome amplification. We show that this technique is suitable for routine evaluation of genomic variation such as SNP analyses of the specifically selected nuclei. Our method should provide a means for whole genome variation studies of single cells from spatially defined positions within tumor tissues.

Solid tumors are complex mixtures of cell types carrying different genomic variations. As such, analyses of entire genomes from individual cells is an indispensible means by which to examine this complexity (1-4). By evaluating correlations among, and diversity between, the genomic variations of single cells from a solid tumor, critical details underlying tumorogenesis, such as clonal origins or newly acquired functions (5), as well as important functional differentiations, can be elucidated.

Until now, single cell whole genome technologies have been mostly limited to studies where a tissue sample is first enzyme-digested to produce single cell suspensions (4, 6-10). Unfortunately this destroys any evidence of the original spatial relationship. Such spatial information is ultimately critical, since the location of a cell within the tumor can have a profound effect on the acquisition of the different genetic lesions leading to the malignant phenotype. Additionally, rare cells that are only morphologically identifiable, such as those at the metaplasia stage, are difficult to isolate from cells in suspension, and thus might be lost altogether from such an analysis.

A particularly effective alternative to working with cells in suspension is the use of laser capture microdissection (LCM) (11), however the isolation of intact whole nuclei using this methodology is an unsolved technical challenge (10). The main difficulty is that sectioning frequently cuts through individual nuclei, necessitating the collection of many serial sections, which invariably leads to sample loss. These losses are difficult to quantify, leading to erroneous results for “whole genome” examination. Therefore, developing effective procedures to identify and isolate intact nuclei using LCM is expected to facilitate broader and more reliable applications of single cell genomic studies of tumorogenesis and pathogenesis.

Here, we show that by using optimal thicknesses and through-focus evaluations, complete single nuclei in frozen-sectioned tissue can be easily identified and selected by laser microdissection in such a manner that the whole genome can then be reliably amplified (Supplementary Figure S1).

To demonstrate this approach, we selected a human high-grade (III) gastric cancer (intestinal type) and a paired normal gastric tissue (collected from Ruijin Hospital, Shanghai Jiao Tong University School of Medicine). Fixed frozen sections were labeled with 4′,6-diamidino-2-phenylindole (DAPI), while the general features of the sample were examined using adjacent sections stained with hematoxylin and eosin.

Intact DAPI-labeled single nuclei could be identified in three dimensions by adjusting the focal distance (Supplementary Video S1). A typical example is shown in Figure 1. To maximize the number of intact nuclei in these sections, we have systematically examined the intactness of visualized nuclei for different section thicknesses. We found that ~14 µm is the optimal thickness for single nucleus isolation for these gastric tumor samples (Supplementary Figure S2). Below this thickness, most nuclei are not intact, and with greater thickness, the ability to select intact single nuclei was reduced due to overlapping nuclei in the field of view.




Figure 1. Determination of a single complete nucleus in the tissue section. (Click to enlarge)


After appropriate sections were prepared, the slide was loaded onto a laser microdisscetion system (PALM Microlaser Technologies, Carl Zeiss, Bernreid, Germany). The intact DAPI-stained nuclei were first individually located by the through focus procedure and then encircled with partial cutting using the UV laser of the LCM. This step does not lead to complete dissection of the nuclei, but only serves to demarcate the chosen nuclei for subsequent removal. The coverslip was next removed by dipping the slide into 100% ethanol, and after drying, the sample was reloaded onto the stage. The collection caps were then filled with 10 µL cell lysis and fragmentation buffer, and the dissection was completed, followed by pressure-catapultion into the cap. The success of the procedure was verified by in situ examination of the caps (Supplementary Figure S3).

Following the isolation of the single nuclei, the material was processed for whole genome amplification. The collected material was first centrifuged at 20,800× g for 10 min at 4°C, followed by amplification using commercial kits (GenomePlex; Sigma-Aldrich, St. Louis, MO, USA). The amplified product was then purified using the GenElute PCR Clean-up kit (Sigma-Aldrich).

Figure 2A shows gel electrophoresis of amplified products of two single intact nuclei isolated from paired normal gastric tissue. The size range of DNA fragments obtained from this procedure was as expected. The amount of amplified products from the two nuclei was 2.6 and 3.0 µg, respectively. This represents an amplification of 105-fold with good reproducibility. To demonstrate that the amplified product was a faithful reproduction of the original material, we randomly cloned DNA fragments from the amplified product and selected 24 clones for regular sequencing. All of the cloned sequences corresponded to sequences in the human genome, illustrating the fidelity of the procedure (Supplementary Figure S4). Furthermore, we demonstrated that the amplified products can also be used for routine SNP analysis. The well-characterized rs2294008T allele of the PSCA gene has previously been correlated with gastric cancer from bulk analysis (12, 13). Using PCR restriction fragment-length polymorphism (PCR-RFLP), we found that this predisposing mutation of the rs2294008T allele was also reliably detected in the amplified fragments from single nuclei using our isolation method (Figure 2B), thus confirming the effectiveness of this procedure for SNP analysis.




Figure 2. Evaluation of the quality of single nucleus whole genome amplification products. (Click to enlarge)


In conclusion, we note that the UV cutting and catapulting functions of the LCM system are essential for the demarcation of intact nuclei, as well as their removal in the “cleanest” possible manner. Devices that rely on adhesion are less effective in this regard, since contamination by the surrounding area can be significant for small selected areas. It remains to be seen whether repeated adhesion operations could be reliably performed to collect many single nuclei into the same cap.

At the moment, single nucleus sequencing studies may incur significant costs, especially with large samples. Still, we expect these costs would be lowered greatly by incorporating exome isolation methodologies and will likely be reduced as the technology advances in the future.

Single nucleus collection and sequencing strategies such as the one described here are thus expected to facilitate elucidation of the origins and development of genomic variation within cancer tissues.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (grant no. 30900271), the State Key Development Program for Basic Research of China (grant no. 2010CB529205), and funds from the State Key Laboratory of Oncogenes & Related Genes (no. 91-10-09). D. M. C. was supported by the Chinese Academy of Sciences Fellowships for Young International Scientists (Grant 2009YA1-1).

Competing interests

The authors declare no competing interests.

Correspondence
Address correspondence to Zhifeng Shao Ministry of Education Key Laboratory of Systems Biomedicine, Shanghai Jiao Tong University, Shanghai 200240, P R China. Email: [email protected] or [email protected]


References
1.) Navin, N., J. Kendall, J. Troge, P. Andrews, L. Rodgers, J. McIndoo, K. Cook, A. Stepansky. 2011. Tumour evolution inferred by single-cell sequencing. Nature 472:90-94.

2.) Konings, P., E. Vanneste, S. Jackmaert, M. Ampe, G. Verbeke, Y. Moreau, J.R. Vermeesch, and T. Voet. 2012. Microarray analysis of copy number variation in single cells. Nat. Protocols 7:281-310.

3.) Geigl, J.B., and M.R. Speicher. 2007. Single-cell isolation from cell suspensions and whole genome amplification from single cells to provide templates for CGH analysis. Nat. Protocols 2:3173-3184.

4.) Geigl, J.B., A.C. Obenauf, J. Waldispuehl-Geigl, E.M. Hoffmann, M. Auer, M. Hörmann, M. Fischer, Z. Trajanoski. 2009. Identification of small gains and losses in single cells after whole genome amplification on tiling oligo arrays. Nucleic Acids Res. 37:e105.

5.) Navin, N., and J. Hicks. 2011. Future medical applications of single-cell sequencing in cancer. Genome Med. 3:31.

6.) Zhang, K., A.C. Martiny, N.B. Reppas, K.W. Barry, J. Malek, S.W. Chisholm, and G.M. Church. 2006. Sequencing genomes from single cells by polymerase cloning. Nat. Biotechnol. 24:680-686.

7.) Ishii, S., K. Tago, and K. Senoo. 2010. Single-cell analysis and isolation for microbiology and biotechnology: methods and applications. Appl. Microbiol. Biotechnol. 86:1281-1292.

8.) Arai, F., C. Ng, H. Maruyama, A. Ichikawa, H. El-Shimy, and T. Fukuda. 2005. On chip single-cell separation and immobilization using optical tweezers and thermosensitive hydrogel. Lab Chip 5:1399-1403.

9.) Hu, X., P.H. Bessette, J. Qian, C.D. Meinhart, P.S. Daugherty, and H.T. Soh. 2005. Marker-specific sorting of rare cells using dielectrophoresis. Proc. Natl. Acad. Sci. USA 102:15757-15761.

10.) Kvist, T., B.K. Ahring, R.S. Lasken, and P. Westermann. 2007. Specific single-cell isolation and genomic amplification of uncultured microorganisms. Appl. Microbiol. Biotechnol. 74:926-935.

11.) Hillebrands, J.L., F.A. Klatter, B.M. van den Hurk, E.R. Popa, P. Nieuwenhuis, and J. Rozing. 2001. Origin of neointimal endothelium and alpha-actin-positive smooth muscle cells in transplant arteriosclerosis. J. Clin. Invest. 107:1411-1422.

12.) Study Group of Millennium Genome Project for Cancer, H. Sakamoto, K. Yoshimura, N. Saeki, H. Katai, T. Shimoda, Y. Matsuno, D. Saito. 2008. Genetic variation in PSCA is associated with susceptibility to diffuse-type gastric cancer. Nat. Genet. 40:730-740.

13.) Wu, C., G. Wang, M. Yang, L. Huang, D. Yu, W. Tan, and D. Lin. 2009. Two genetic variants in prostate stem cell antigen and gastric cancer susceptibility in a Chinese population. Mol. Carcinog. 48:1131-1138.