Research Article: Methods/New Tools | Novel Tools and Methods Laser capture microdissection of single neurons with morphological visualization using fluorescent proteins fused to transmembrane proteins https://doi.org/10.1523/ENEURO.0275-20.2021 Cite as: eNeuro 2021; 10.1523/ENEURO.0275-20.2021 Received: 23 June 2020 Revised: 12 July 2021 Accepted: 15 July 2021 This Early Release article has been peer-reviewed and accepted, but has not been through the composition and copyediting processes. The final version may differ slightly in style or formatting and will contain links to any extended data. Alerts: Sign up at www.eneuro.org/alerts to receive customized email alerts when the fully formatted version of this article is published. Copyright © 2021 Chang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution 4.0 International license, which permits unrestricted use, distribution and reproduction in any medium provided that the original work is properly attributed. Title Laser capture microdissection of single neurons with morphological visualization using fluorescent proteins fused to transmembrane proteins Abbreviated Title Stable fluorescent labeling for microdissection Authors Ching Ching Chang, Hai Tarng Chong and Ayumu Tashiro (田代 歩) School of Biological Sciences, Nanyang Technological University, Singapore Author Contributions CCC and AT designed research, analyzed data, and wrote paper. CCC and HTC performed research. 1 Corresponding should be addressed to Ayumu Tashiro ([email protected]) or 2 Chang Ching Ching ([email protected]). Number of Figures: 7 Number of Tables: 0 Number of Multimedia: 0 Number of words for Abstract: 250 Number of words for Significance Statement: 110 Number of words for Introduction: 472 Number of words for Discussion: 1372 Acknowledgements We thank Ms. Miaomiao Han for her technical assistance. Conflict of Interest Authors report no conflict of interest. Funding sources 1 Ministry of Education, Singapore, Academic Research Fund Tier 1 (2016-T1-001- 010, 2018-T1-002-053) and Tier 3 (MOE2017-T3-1-002). 2 3 Abstract 4 Gene expression analysis in individual neuronal types helps understand brain 5 function. Genetic methods expressing fluorescent proteins are widely used to label 6 specific neuronal populations. However, because cell type specificity of genetic 7 labeling is often limited, it is advantageous to combine genetic labelling with 8 additional methods to select specific cell/neuronal types. Laser capture 9 microdissection is one of such techniques with which one can select a specific 10 cell/neuronal population based on morphological observation. However, a major 11 issue is disappearance of fluorescence signals during tissue processing required for 12 high quality sample preparation. Here, we developed a simple, novel method in 13 which fluorescence signals are preserved. We use genetic labeling with fluorescence 14 proteins fused to transmembrane proteins, which shows highly stable fluorescence 15 retention and allows for the selection of fluorescent neurons/cells based on 16 morphology. Using this method in mice, we laser-captured neuronal somata and 17 successfully isolated RNA. We determined that ~100 cells are sufficient to obtain a 18 sample required for downstream applications such as quantitative PCR. Capability to 19 specifically micro-dissect targeted neurons was demonstrated by ~10-fold increase 20 in mRNA for fluorescent proteins in visually identified neurons expressing the 21 fluorescent proteins compared to neighboring cells not expressing it. We applied this 22 method to validate virus-mediated single-cell knockout, which showed up to 95% 23 reduction in knocked-out gene RNA compared with wild-type neurons. This method 24 using fluorescent proteins fused to transmembrane proteins provides a new, simple 25 solution to perform gene expression analysis in sparsely labelled neuronal/cellular 26 populations, which is especially advantageous when genetic labelling has limited 27 specificity. 3 28 Significance statement 29 Genetic labeling of specific cell types with reporter fluorescent proteins is widely 30 used for gene expression analysis. However, because the specificity of genetic 31 labeling is often limited, an additional method is required to collect samples from a 32 specific cell type. Here, we developed a novel method to allow for selection of a 33 cell/neuronal type of interest based on morphology before performing sample 34 collection using laser capture microdissection. Stable fluorescent signals from 35 fluorescent proteins fused to transmembrane proteins allows us to morphologically 36 select and laser-capture hundreds of single neurons. This method would be 37 applicable to genomic and transcriptomic analysis of different cell/neuronal types and 38 could facilitate our understanding of brain functions. 4 39 Introductions 40 Gene expression analysis of specific neuronal types has been successfully 41 performed using genetic labeling with reporter fluorescent proteins and fluorescence 42 activated cell sorting (FACS) (Xia et al., 2017, Daigle et al., 2018). However, one 43 technical limitation is that genetic labeling often marks multiple neuronal types and/or 44 other cell types (van Praag et al., 2002, Balthasar et al., 2004, Lagace et al., 2007, 45 Harris et al., 2014) whereas FACS cannot separate such unwanted, labelled cell 46 types. In addition, enzyme-mediated cell dissociation procedures could affect viability 47 and gene expression (Hussain et al., 2018, Ho et al., 2018, Reichard et al., 2018). 48 An alternative method that allows for separating cell/neuronal types without tissue 49 dissociation steps would be useful as a complementary method. 50 51 Laser capture microdissection is a technique that dissects a small piece of tissue or 52 a single cell by laser and collect them with gravity or assistance of laser. 53 Instruments for laser capture microdissection are often equipped with a fluorescence 54 light source (Emmert-Buck et al., 1996, Bonner et al., 1996, Fend et al., 2000). 55 Therefore, one can observe the morphology of genetically labeled fluorescent cells 56 and select a cell type of interest to capture. However, in practice, this is not a simple 57 task. Commonly for laser capture microdissection, sections are prepared from freshly 58 frozen tissue, rather than fixed tissue, to acquire high quality RNA. In such sections, 59 fluorescent proteins diffuse out from cells and their fluorescence signals disappear 60 during tissue processing procedures (Singh et al., 2019). Although an optimized 61 protocol can improve the issue to some extent, instability of fluorescence signals 62 makes it difficult to use fluorescence as a guide to distinguish cell types for laser 63 capture microdissection. More stable fluorescence labeling method would be 64 valuable for solving this technical limitation. In this line, studies (Rossner et al., 2006, 65 Chatzi et al., 2019) demonstrated that nuclear targeted fluorescent proteins 66 preserves their signals. However, because only nuclei, but not processes/neurites, is 67 visualized, its application to morphological identification of cell types is limited. 68 69 In the course of an experiment for some other purpose, we happened to find that the 70 fluorescence signals from transmembrane proteins fused with fluorescent proteins 71 was extremely stable and kept visualizing dendritic morphology in air-dried, non-fixed 72 brain sections. We later found that similar preservation of fluorescence signals from 5 73 transmembrane proteins fused with fluorescent proteins has been previously 74 described in the context of flow cytometry of primary human fibroblasts (Kalejta et al., 75 1997). This observation made us reason that genetic labeling with fluorescent 76 proteins fused to transmembrane proteins allows us to identify a cell type of interest 77 based on their morphology and selectively capture them using laser capture 78 microdissection. In this study, we tested this possibility and validated that the use of 79 labeling with fluorescent proteins fused to transmembrane proteins allows us to 80 select neurons over non-neuronal cells and perform downstream gene expression 81 analyses. 82 83 Materials and methods 84 A flow chart of the methodology is shown in Figure 1. We freshly froze brains after 85 retrieving them from the skull. We sectioned the brains, mounted sections on the 86 glass slides and dehydrated/fixed them with ethanol. We distinguished neurons from 87 glial cells based on their dendritic morphology under a fluorescence microscope, and 88 then harvested the cell bodies of the selected neurons with laser capture 89 microdissection. A total of 100-300 microdissected neurons were pooled for RNA 90 extraction. As examples of downstream gene expression analyses, we performed 91 quantitative PCR (qPCR) analysis after cDNA amplification. 92 93 Mice 94 We used 6- to 11-week-old C57BL/6 transgenic mice of either sex, housed in a 12 95 hours dark/light cycle with ad libitum access to water and food. Transgenic lines 96 used were floxed NR1 mice (B6.129S4-Grin1tm2Stl/J) (Tsein et al., 1996) and Pomc- 97 ChR2 mice, which were produced by crossing two transgenic lines, STOCK 98 Tg(Pomc1-cre)16Lowl/J (Balthasar et al., 2004) (back-crossed with C57BL/6 mice) 99 and B6.Cg-Gt(ROSA)26Sortm32(CAG-COP4*H134R/EYFP)Hze/J (Madisen et al., 100 2012). All three transgenic lines were originally purchased from Jackson laboratory 101 (USA) and bred in our local facilities.