参见作者原研究论文

本实验方案简略版
Nov 2020
Advertisement

本文章节


 

Generation and Maintenance of Homogeneous Human Midbrain Organoids
同种人类中脑类器官的生成与维持   

引用 收藏 提问与回复 分享您的反馈 Cited by

Abstract

Three-dimensional cell cultures (“organoids”) promise to better recapitulate native tissue physiology than traditional 2D cultures and are becoming increasingly interesting for disease modeling and compound screening efforts. While a number of protocols for the generation of neural organoids have been published, most protocols require extensive manual handling and result in heterogeneous aggregates with high sample-to-sample variation, which can hinder screening-based strategies. We have now developed a fast and efficient protocol for the generation and maintenance of highly homogeneous and reproducible midbrain organoids. The protocol is streamlined for use in fully automated workflows but can also be performed manually without the need for highly specialized equipment. It relies on the aggregation of small molecule neural precursor cells (smNPCs) in standard 96-well V-bottomed plates under static culture conditions without cumbersome matrix embedding. The result is ready-to-assay uniform 3D human midbrain organoids available in freely scalable quantities for downstream analyses in 3D cell culture


Graphic abstract:



Automated midbrain organoid generation workflow and timeline


Keywords: Organoids (类器官), Midbrain (中脑), Stem cells (干细胞), Neuroscience (神经系统科学), Drug development (药物开发), Automation (自动化), High Throughput Screening (HTS) (高通量筛选)

Background

While classical preclinical model systems including 2D cell culture and animal models have driven our understanding of human biology and disease, there is increasing evidence of human-specific processes that elude these standard model systems. Particularly, the human brain is unique in its structure and physiology and has proven difficult to approximate in research settings (Kim et al., 2020). Organoids, i.e., stem cell-derived, self-organized, three-dimensional microtissues, may mimic aspects of human organ structure and function better than previous model systems (Sato et al., 2009; Lancaster et al., 2013; Takasato et al., 2015; McCracken et al., 2014; Dye et al., 2015), thus promising to improve our understanding of uniquely human processes (Fatehullah and Barker 2016; Dutta et al., 2017).


Over recent years, different groups have focused on mimicking specific brain regions to better understand their tissue biology during development and disease. One region of specific interest is the midbrain because of its prominent early involvement in Parkinson’s disease, the second most common neurodegenerative disorder affecting 2-3% of individuals aged 65 years or older (Poewe et al., 2017). While a number of different midbrain organoid systems have been developed (Jo et al., 2016; Qian et al., 2016; Monzel et al., 2017), the high degree of homogeneity, reproducibility, and scale-up required for disease modeling and drug development have proven challenging.


We recently published the generation of automated midbrain organoids (AMOs), which were specifically developed to overcome these challenges (Renner et al., 2020). AMOs are highly reproducible and homogeneous with regard to their size, morphology, gene and protein expression, functionality (i.e., neural activity), and response to compounds. They are generated from small molecule neural precursor cells (smNPCs), a cell type optimized for easy, small molecule-based handling as well as fast and robust neural differentiation (Reinhardt et al., 2013). Combined with an organoid generation workflow that does not require any manual intervention (e.g., matrix embedding), this makes our protocol easily reproducible and results in highly homogeneous midbrain organoids. Thus, AMOs are ideally suited for all applications that require large numbers of comparable samples with clearly defined and predictable parameters, most notably compound screening. However, many other applications including disease modeling and developmental biology can also greatly benefit from homogeneous starting conditions without the need for industry-scale sample numbers. A predictable 3D model like AMOs facilitates the discrimination between normal biological variance and the influence of e.g., a single point mutation in a model for a hereditary disease. Thus, we developed the protocol to also be fully compatible with manual pipetting and standard lab equipment, making it available and useful to a large audience.


While the general workflow described herein can be applied to a number of different organoid protocols and starting cell types, we note that smNPCs are not capable of forming forebrain organoids. smNPCs are restricted to mid- and hindbrain fates, including the generation of motor neurons. For details, please see (Reinhardt et al., 2013). Moreover, smNPCs do not form neural rosettes. Our midbrain organoids are optimized for homogeneity to facilitate screening strategies at the cost of complexity. If a strong internal tissue polarity is desired, we recommend starting the culture with other, less restricted cell types, including pluripotent stem cell (PSC)-derived NPCs of an earlier developmental stage than smNPCs or even PSCs at the risk of increasing resulting organoid heterogeneity.

Materials and Reagents

  1. Screw cap tubes (Sarstedt, catalog numbers: 62.547.254 [50 ml] and 62.554.502 [15 ml])

  2. Serological pipets (Falcon, catalog numbers: 356543 [5 ml], 356551 [10 ml], and 356525 [25 ml])

  3. Pipet tips (StarLab, catalog numbers: S1120-1810 [20 μl], S1120-8810 [200 μl], and S1122-1830 [1,000 μl])

  4. Standard tissue culture-treated 6-well plates (Sarstedt, catalog number: 83.3920)

  5. Conical/V-bottomed 96-well plates (Thermo Fisher, catalog number: 277143)

  6. Vacuum filtration system (Corning, catalog number: 431097)

  7. Optional: Biomek liquid handler pipette tips AP96 P250 (Beckman Coulter, catalog number: 717252)

  8. Small molecule neural precursor cells [smNPCs (Reinhardt et al., 2013)]

  9. KnockOut (KO) DMEM/F-12 (Thermo Fisher, catalog number: 12660012)

  10. Matrigel (BD Biosciences, catalog number: 354263)

  11. DMEM/F-12 (Thermo Fisher, catalog number: 11320033)

  12. Neurobasal medium (Thermo Fisher, catalog number: 21103-049)

  13. B27 supplement without vitamin A (Thermo Fisher, catalog number: 12587010)

  14. N2 supplement (Thermo Fisher, catalog number: 17502-048)

  15. Penicillin-Streptomycin (Sigma-Aldrich, catalog number: P4333)

  16. GlutaMAX (Thermo Fisher, catalog number: 35050061)

  17. Bovine Serum Albumin (BSA, Thermo Fisher, catalog number: 15260037)

  18. Human Serum Albumin (HSA, Biological Industries, catalog number: 05-720-1B)

  19. Accutase (Sigma-Aldrich, catalog number: A6964)

  20. Polyvinyl Alcohol (PVA; Sigma-Aldrich, catalog number: 363170)

  21. Ascorbic acid (Sigma-Aldrich, catalog number: A4544)

  22. CHIR-99021 (Biomol, catalog number: Cay13122)

  23. Smoothened Agonist (SAG; Cayman Chemical, catalog number: Cay11914)

  24. Brain Derived Neurotrophic Factor (BDNF; PeproTech, catalog number: 450-02)

  25. Glial cell line-Derived Neurotrophic Factor (GDNF; PeproTech, catalog number: 450-10)

  26. Activin A (PeproTech, catalog number: 120-14E)

  27. Transforming Growth Factor Beta 3 (TGF-β3; PeproTech, catalog number: 100-36E)

  28. N6,2’-O-Dibutyryladenosine 3’,5’-cyclic monophosphate (dbcAMP; Sigma-Aldrich, catalog number: D0627)

  29. Basal medium (see Recipes)

  30. smNPC medium (see Recipes)

  31. Aggregation medium (see Recipes)

  32. Patterning medium (see Recipes)

  33. Maturation medium (days 6-X) (see Recipes)

  34. Split medium (see Recipes)

Equipment

  1. Water bath

  2. Humidified CO2 incubator

  3. Optional: CO2 incubator connected to an automated liquid handling system (ALHS)

  4. Sterile tissue/cell culture hood

  5. Fridge and freezers (4°C, -20°C, and -80°C)

  6. Centrifuge (e.g., Eppendorf, model: Centrifuge 5702)

  7. Stereomicroscope with camera (e.g., Leica MZ10 F [microscope] and Leica DFC425 C [camera], Leica Microsystems)

  8. Mechanical pipets

  9. Optional: Multichannel mechanical pipets

  10. Optional: Automated liquid handling system (e.g., Biomek FXP Laboratory Automation Workstation, Beckman Coulter)

Software

  1. Leica application suite v. 4.8 (LAS, Leica Microsystems)

  2. Fiji/ImageJ (Schindelin et al., 2012)

  3. Optional: Biomek Software v. 3.3 (controlling the automated liquid handler, Beckman Coulter)

Procedure

  1. Preparation of matrigel-coated plates for 2D smNPC cell culture

    Note: Before the final coating step, all plasticware and solutions containing matrigel must be kept on ice to prevent premature gelation during handling.

    1. Dilute matrigel 1:100 in cold (4°C) KnockOut DMEM/F-12

      Note: We recommend keeping a frozen stock of 1:4 diluted matrigel in KnockOut DMEM/F-12 and further diluting this 1:25 in KnockOut DMEM/F-12 immediately before use.

    2. Quickly add 1 ml 1:100 matrigel solution per well of a 6-well plate to cover the entire bottom of the well. Distribute the solution evenly across the surface of the well by gently rocking the plate. If you imagine a coordinate system with the XY plane parallel to the ground, first tilt the plate along the Y-axis about ten degrees in either direction 5 times, then tilt the plate along the X-axis about ten degrees in either direction 5 times or until the entire plate bottom is covered by matrigel solution. Seal the plate with Parafilm to prevent evaporation.

    3. Incubate at room temperature overnight. It is possible to store the plates at 4°C for approximately 2 weeks.

      Note: If plates are needed on the same day, it is also possible to incubate them at 37°C for 3 h and use them immediately.


  2. smNPC culture

    Note: Generally, smNPCs are seeded and cultured in 2 ml medium per well in 6-well plates. As the cells become more confluent, carefully monitor the cell culture medium. If the pH indicator shows acidification (i.e., the commonly used phenol red will turn yellow), refresh medium every day and / or increase media volume by 50-100% up to 4 ml per well. Figure 1 includes examples of smNPC cultures at different stages as a reference.

    1. Prepare smNPC medium (see Recipes, Table 2) from basal medium (see Recipes, Table 1) and prewarm it to 37°C in a water bath. Small molecules should be added to the medium immediately before use. Avoid heat-cycling the medium once reconstituted.

    2. Aspirate old medium.

    3. Add 1 ml accutase per well of a 6-well plate and incubate at 37°C for 10-15 min (check after 7-8 min whether cells are detached). Cells should detach without mechanical agitation.

    4. Resuspend the cells 2-3 times as gently as possible using a 1,000 μl manual pipette to maximize cell harvest and to disperse small remaining aggregates. Transfer suspension into a tube with 6 ml split medium (see Recipes, Table 6) to dilute and inactivate accutase.

    5. Centrifuge for 2 min at 220 × g.

    6. Aspirate the supernatant. Take care not to aspirate the pellet.

    7. Resuspend the pellet in smNPC medium and seed the cells at a ratio of approximately 1:10-1:20 (depending on cell line and experience).
      Note: Aspirate the supernatant from the matrigel-coated wells before adding the cell solution.

    8. Distribute cells evenly across the surface of the well by gently rocking the plate in an XY-motion (see Step A2). Avoid swirling/circular motions as these tend to concentrate the cells in the center of the well. Place plate into a 37°C CO2 incubator.

    9. Exchange medium every other day and maintain cells until they reach approximately 90-100% confluence before splitting them again/using them for organoid generation. See Figure 1d.



      Figure 1. Phase contrast images of 2D smNPC culture at different stages of confluence and differentiation. (a/b) Representative images illustrating the density of smNPCs 2 days after splitting at either 1:10 (a) or 1:20 (b). (c) Example image of a low-quality smNPC culture contaminated by large numbers of differentiated cells (very dark and large, located between colonies). Partially differentiated cultures may be rescued by maintaining high splitting ratios (at least 1:20) over several passages, as smNPCs usually proliferate faster than differentiated cells. These types of culture should not be used for generating organoids. (d) smNPCs at the optimal confluence for organoid generation or splitting.


  3. Organoid generation and maintenance

    Note: See Figure 2 for representative brightfield images of organoids at different ages.

    Day 0

    1. Detach smNPCs from the matrigel-coated cell culture plate using accutase as described in Steps B3 and B4.

    2. Collect cells in split medium (see Recipes, Table 6), count them, and transfer the appropriate number of cells (see note below) to a new Falcon tube. Then spin them down at 220 × g for 2 min.

      Note on cell numbers and volumes: Seeding requires 9,000 cells in 150 μl medium per well of a 96-well plate. This is equivalent to: 9,000 cells/0.15 ml = 60,000 cells/ml. Thus, prepare the cell suspension at a concentration of 60,000 cells/ml. Make sure to prepare 10-20% surplus volume of the cell suspension to account for pipetting inaccuracies/dead volume.

    3. Resuspend the pellet in the appropriate amount of aggregation medium (see Recipes, Table 3).

      Note: Step 4 and all following feeding steps can be done either by hand or using the automated liquid handling system (ALHS)/pipetting robot.

    4. Seed 9,000 cells in 150 μl aggregation medium per well of a 96-well V-bottomed plate and incubate at 37°C 5% CO2 for 48 h.


    Day 2

    Note: All media changes involving organoids should be performed carefully to avoid damaging the samples or accidentally removing them from the wells. To facilitate manual media aspiration, we recommend slightly tilting the plate and performing all liquid handling by pipetting (not by vacuum pumps or similar devices). Leave a small amount of old medium (≤ 10 μl) in the well so as to not damage the organoids by extensive direct contact with a pipet tip. A small number of organoids (< ca. 50) can be handled with standard single channel mechanical pipets. For larger numbers of samples (more than half a 96-well plate), we recommend using multichannel pipets (in our experience, a standard mechanical pipet with 200-μl capacity works well). Finally, further scale-up for high-throughput applications makes it feasible to use a pipetting robot / automated liquid handler. The same precautions as for manual handling apply – it is advisable to leave a small amount of old medium (≤ 10 μl) per well rather than damage organoids by more ambitious removal of liquids. While dependent on the exact setup, here, dead volumes are generally higher than for manual pipetting, and we advise preparing 5-10 ml surplus of all liquids.

    1. Replace medium with 150 μl patterning medium per well (see Recipes, Table 4).


    Day 4
    1. Replace medium with 150 μl patterning medium per well.


    Day 6
    1. Replace medium with 150 μl maturation medium plus activin A per well (see Recipes, Table 5).


    Day 8 + X
    1. Replace medium with 150 μl maturation medium (same as day 6 without activin A) per well.

    2. Feed aggregates every 2 days with the same medium until they reach the desired age for analysis/follow-up experiments.

      Note: Maturation dynamics vary from cell line to cell line and need to be tested empirically by immunostaining or qPCR. Generally, organoids contain a large number of MAP2+ neurons by day 30. Glial cells including GFAP+ astrocytes arise later, around day 40 onward. For more details, please refer to the detailed analyses in our previous publication (Renner et al., 2020).



      Figure 2. Representative brightfield images illustrating organoid growth and morphology over time. Seeded on day 0 as a single cell smNPC suspension, cells aggregate to a spherical shape within one day. Over time, they maintain their homogeneous round shape and grow in size.

      Note: Sometimes a few dead cells can be found at the bottom of the well (especially at earlier stages. Here, see day 15). As long as the aggregates themselves maintain their structure, this is not an issue. The dead cells will be washed away during the next feeding procedure.

Data analysis

During culture, monitor organoids regularly for normal growth and morphology using a stereomicroscope (see Figure 2 as a reference).

    For a number of other possible downstream analyses and QC steps, including e.g., qPCR, whole mount staining and clearing, organoid viability, and functional analyses (calcium imaging and multi-electrode arrays), please refer to our previous publication (Renner et al., 2020).

Notes

As outlined above, this protocol facilitates either manual or fully automated sample generation. Since each organoid is generated in a separate well and remains separate from all other organoids from seeding to maturation, they represent true biological replicates. Starting with a single cell suspension rather than partially dissociated colonies also maximizes homogeneity between different wells and plates of the same batch. Generating one organoid per well also greatly simplifies the downstream application of compounds and subsequent analysis. In our runs, we routinely generated organoid batches with a coefficient of variation (i.e., standard deviation divided by the mean) of less than 5% with regard to the area of their largest cross-section and around 5% for their viability. All these factors render AMOs ideally suitable for compound screening and toxicity testing since scale-up of fully automated workflows is comparably trivial. While many scientific lines of questioning (e.g., mechanistic analyses) often do not require sample numbers in the hundreds or thousands, the high homogeneity and reproducibility of AMOs still offer advantages, and we recommend testing our protocol with manual pipetting using a single or multichannel pipet.

Recipes

Note: All media without growth factors and small molecules can be stored for up to 2 weeks at 4°C in the dark. Growth factors and small molecules should always be added immediately before use.

  1. Basal medium (Table 1)


    Table 1. Composition of basal medium

    Reagent Amount for 100 ml Supplier Product no. Dilution factor Stock conc.
    DMEM/F-12 48.6 ml Thermo Fisher 11320033 2
    Neurobasal 48.6 ml Thermo Fisher 21103049 2
    N2 250 μl Thermo Fisher A1370701 400
    B27 without Vit A 500 μl Thermo Fisher 12587010 200

    Ascorbic Acid
    (in water)

    50 μl Sigma-Aldrich A4544 2,000 200 mM
    PenStrep 1,000 μl Sigma-Aldrich P4333 100
    GlutaMax 1,000 μl Thermo Fisher 35050061 100


  2. smNPC medium (Table 2)


    Table 2. Composition of smNPC medium

    Reagent Supplier Dilution factor Product no. Stock conc. Final conc.
    Basal medium

    CHIR-99021
    (in DMSO)

    Biomol 2,000 Cay13122 6 mM 3 μM
    SAG (in DMSO) Cayman Chemical 20,000 Cay11914 10 mM 0.5 μM


  3. Aggregation medium (Table 3)


    Table 3. Composition of aggregation medium

    Reagent Amount for 50 ml Supplier Dilution factor Product no. Stock conc. Final conc.
    smNPC medium 45 ml
    Polyvinyl alcohol (PVA, in water)* 5 ml Sigma-Aldrich 10 363170 4% (w/v) 0.4% (w/v)

    *PVA is purchased as a crystalline powder. We recommend preparing a 4% stock solution, which can be stored at 4°C for several months. To prepare the stock solution, dissolve PVA in water by heating to 60-70°C overnight with constant stirring. After the solution has cooled down, sterile filter it with e.g., a vacuum filtration system before use.


  4. Patterning medium (days 2-4) (Table 4)


    Table 4. Composition of patterning medium

    Reagent Supplier Dilution factor Product no. Stock conc. Final conc.
    Basal medium

    BDNF
    (in PBS + 0.1% (v/v) HSA)

    PeproTech 10,000 450-02 10 μg/ml 1 ng/ml

    GDNF
    (in PBS + 0.1% (v/v) HSA)

    PeproTech 10,000 450-10 10 μg/ml 1 ng/ml
    SAG (in DMSO) Cayman Chemical 10,000 Cay11914 10 mM 1 μm


  5. Maturation medium (days 6-X) (Table 5)


    Table 5. Composition of maturation medium

    Reagent Supplier Dilution factor Product no. Stock conc. Final conc.
    Basal medium

    Activin A (day 6 only!)
    (in PBS + 0.1% (v/v) HSA)

    PeproTech 2,000 120-14E 10 μg/ml 5 ng/ml

    BDNF
    (in PBS + 0.1% (v/v) HSA)

    PeproTech 5,000 450-02 10 μg/ml 2 ng/ml

    GDNF
    (in PBS + 0.1% (v/v) HSA)

    PeproTech 5,000 450-10 10 μg/ml 2 ng/ml

    TGFβ-3
    (in PBS + 0.1% (v/v) HSA)

    PeproTech 1,000 100-36E 1 μg/ml 1 ng/ml
    dbcAMP (in water) Sigma-Aldrich 500 D0627 50 mM 100 μm


  6. Split medium (Table 6)


    Table 6. Composition of split medium

    Reagent Amount for 50 ml Supplier Product no. Dilution factor Stock conc.
    DMEM/F-12 49.3 ml Thermo Fisher 11320033
    BSA 667 μl Thermo Fisher 15260037 75 7.5%

Acknowledgments

This work was funded by the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation programme (grant agreement No [669168]). HR is supported by the International Max Planck Research School - Molecular Biomedicine, Münster, Germany. This protocol is based on our previous publication (Renner et al., 2020).

Competing interests

The work presented here is the subject of the patent application EP 18 19 2698.0-1120 to the European Patent Office, where the authors are inventors.

References

  1. Dutta, D., Heo, I. and Clevers, H. (2017). Disease Modeling in Stem Cell-Derived 3D Organoid Systems. Trends Mol Med 23(5): 393-410.
  2. Dye, B. R., Hill, D. R., Ferguson, M. A., Tsai, Y. H., Nagy, M. S., Dyal, R., Wells, J. M., Mayhew, C. N., Nattiv, R., Klein, O. D., White, E. S., Deutsch, G. H. and Spence, J. R. (2015). In vitro generation of human pluripotent stem cell derived lung organoids. Elife 4: e05098.
  3. Fatehullah, A., Tan, S. H. and Barker, N. (2016). Organoids as an in vitro model of human development and disease. Nat Cell Biol 18(3): 246-254.
  4. Jo, J., Xiao, Y., Sun, A. X., Cukuroglu, E., Tran, H. D., Goke, J., Tan, Z. Y., Saw, T. Y., Tan, C. P., Lokman, H., Lee, Y., Kim, D., Ko, H. S., Kim, S. O., Park, J. H., Cho, N. J., Hyde, T. M., Kleinman, J. E., Shin, J. H., Weinberger, D. R., Tan, E. K., Je, H. S. and Ng, H. H. (2016). Midbrain-like Organoids from Human Pluripotent Stem Cells Contain Functional Dopaminergic and Neuromelanin-Producing Neurons. Cell Stem Cell 19(2): 248-257.
  5. Kim, J., Koo, B. K. and Knoblich, J. A. (2020). Human organoids: model systems for human biology and medicine. Nat Rev Mol Cell Biol 21(10): 571-584.
  6. Lancaster, M. A., Renner, M., Martin, C. A., Wenzel, D., Bicknell, L. S., Hurles, M. E., Homfray, T., Penninger, J. M., Jackson, A. P. and Knoblich, J. A. (2013). Cerebral organoids model human brain development and microcephaly. Nature 501(7467): 373-379.
  7. McCracken, K. W., Cata, E. M., Crawford, C. M., Sinagoga, K. L., Schumacher, M., Rockich, B. E., Tsai, Y. H., Mayhew, C. N., Spence, J. R., Zavros, Y. and Wells, J. M. (2014). Modelling human development and disease in pluripotent stem-cell-derived gastric organoids. Nature 516(7531): 400-404.
  8. Monzel, A. S., Smits, L. M., Hemmer, K., Hachi, S., Moreno, E. L., van Wuellen, T., Jarazo, J., Walter, J., Bruggemann, I., Boussaad, I., Berger, E., Fleming, R. M. T., Bolognin, S. and Schwamborn, J. C. (2017). Derivation of Human Midbrain-Specific Organoids from Neuroepithelial Stem Cells. Stem Cell Reports 8(5): 1144-1154.
  9. Poewe, W., Seppi, K., Tanner, C. M., Halliday, G. M., Brundin, P., Volkmann, J., Schrag, A. E. and Lang, A. E. (2017). Parkinson disease.Nat Rev Dis Primers 3: 17013.
  10. Qian, X., Nguyen, H. N., Song, M. M., Hadiono, C., Ogden, S. C., Hammack, C., Yao, B., Hamersky, G. R., Jacob, F., Zhong, C., Yoon, K. J., Jeang, W., Lin, L., Li, Y., Thakor, J., Berg, D. A., Zhang, C., Kang, E., Chickering, M., Nauen, D., Ho, C. Y., Wen, Z., Christian, K. M., Shi, P. Y., Maher, B. J., Wu, H., Jin, P., Tang, H., Song, H. and Ming, G. L. (2016). Brain-Region-Specific Organoids Using Mini-bioreactors for Modeling ZIKV Exposure. Cell 165(5): 1238-1254.
  11. Reinhardt, P., Glatza, M., Hemmer, K., Tsytsyura, Y., Thiel, C. S., Hoing, S., Moritz, S., Parga, J. A., Wagner, L., Bruder, J. M., Wu, G., Schmid, B., Ropke, A., Klingauf, J., Schwamborn, J. C., Gasser, T., Scholer, H. R. and Sterneckert, J. (2013). Derivation and expansion using only small molecules of human neural progenitors for neurodegenerative disease modeling. PLoS One 8(3): e59252.
  12. Renner, H., Grabos, M., Becker, K. J., Kagermeier, T. E., Wu, J., Otto, M., Peischard, S., Zeuschner, D., TsyTsyura, Y., Disse, P., Klingauf, J., Leidel, S. A., Seebohm, G., Scholer, H. R. and Bruder, J. M. (2020). A fully automated high-throughput workflow for 3D-based chemical screening in human midbrain organoids. Elife 9: e52904.
  13. Sato, T., Vries, R. G., Snippert, H. J., van de Wetering, M., Barker, N., Stange, D. E., van Es, J. H., Abo, A., Kujala, P., Peters, P. J. and Clevers, H. (2009). Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche. Nature 459(7244): 262-265.
  14. Schindelin, J., Arganda-Carreras, I., Frise, E., Kaynig, V., Longair, M., Pietzsch, T., Preibisch, S., Rueden, C., Saalfeld, S., Schmid, B., Tinevez, J. Y., White, D. J., Hartenstein, V., Eliceiri, K., Tomancak, P. and Cardona, A. (2012). Fiji: an open-source platform for biological-image analysis. Nat Methods 9(7): 676-682.
  15. Takasato, M., Er, P. X., Chiu, H. S., Maier, B., Baillie, G. J., Ferguson, C., Parton, R. G., Wolvetang, E. J., Roost, M. S., Chuva de Sousa Lopes, S. M. and Little, M. H. (2015). Kidney organoids from human iPS cells contain multiple lineages and model human nephrogenesis. Nature 526(7574): 564-568.

简介

[摘要]三-三维细胞培养物(“类器官”)承诺更好综述自然组织的生理比传统的2D培养和正在成为疾病建模越来越有趣和化合物筛选工作。尽管已经发布了许多用于生成神经器官的协议,但是大多数协议都需要大量的人工处理,并且会导致样本间差异很大的异类聚集体,这可能会阻碍基于筛选的策略。我们现在已经开发了一种快速高效的协议,用于生成和维护高度 均质且可重现的中脑类器官。该协议已简化为可在全自动工作流程中使用,但也可以手动执行,而无需高度专业化的设备。它依赖于小分子神经前体细胞(的聚集smNPCs在标准的96孔V形底)ED下板的静态培养条件而无需麻烦的基质嵌入。结果是现成可测定的统一3D人类中脑类器官,可以自由扩展的量用于3D细胞培养的下游分析。

图形摘要:


自动化的中脑类器官生成工作流程和时间表


[背景]尽管包括二维细胞培养和动物模型的经典临床前模型系统已经推动我们的人类生物学和疾病的认识,越来越多的避开了这些标准模型系统人类特有的过程的证据。特别地,人脑在其结构和生理学方面是独特的,并且已经证明在研究环境中难以近似(Kim等人,2020)。类器官,即,干细胞衍生的,自组织的,三维微组织,可能人体器官的结构和功能比以前的模型系统更好的模拟方面(佐藤等人。,2009;斯特等人。,2013; Takasato等人。,2015;麦克拉肯。等人,2014;染料等。,2015) ,从而PROMIS荷兰国际集团改进我们人类独有的过程的理解(Fatehullah和Barker 2016;杜塔。等人,2017) 。

在最近几年,不同的群体都集中在模仿特定的大脑区域,以便更好地理解发展和疾病过程中的组织生物学。特别感兴趣的一个区域是中脑,因为它早期参与了帕金森氏病,帕金森氏病是第二大最常见的神经退行性疾病,影响2-3%的65岁或65岁以上的人(Poewe等人,2017)。尽管已经开发出许多不同的中脑类器官系统(Jo等人,2016; Qian等人,2016; Monzel等人,2017),但疾病建模需要高度的同质性,可再现性和规模化事实证明,药物开发具有挑战性。

我们最近发布了自动中脑类器官(AMO)的生成,它们是专门为克服这些挑战而开发的(Renner et al 。,2020)。AMOS是高度可重现的和均匀的,关于其大小,形态,基因和蛋白质表达,功能性(即,神经活动),并响应于化合物。它们是由小分子神经前体细胞(smNPC )产生的,该细胞类型经过优化,可轻松实现基于小分子的处理以及快速而强大的神经分化(Reinhardt等人,2013)。结合无需任何人工干预(例如,基质嵌入)的类器官生成工作流程,这使得我们的协议易于再现,并产生高度均一的中脑类器官。因此,AMO非常适合所有需要大量具有明确定义和可预测参数的可比样品的应用,尤其是化合物筛选。但是,许多其他应用(包括疾病建模和发育生物学)也可以从均质的起始条件中大大受益,而无需行业规模的样本数量。像AMO一样的可预测3D模型有助于区分正常的生物学差异和遗传疾病模型中例如单点突变的影响。因此,我们开发的协议,也可以用手动移液和标准实验室设备完全兼容,使其可以和有用到大量的观众。

虽然一般的工作流程在此描述中可以应用到许多不同的器官样的协议和起始的细胞类型,我们注意到,smNPCs不能形成前脑组织体。smNPC仅限于中脑和后脑的命运,包括运动神经元的产生。有关详细信息,请参见(Reinhardt等,2013)。而且,smNPC不会形成神经花环。我们对中脑类器官进行了优化,以提高同质性,从而以复杂性为代价促进筛选策略。如果需要强大的内部组织极性,我们建议从其他受限制较少的细胞类型开始培养,包括比smNPC甚至PSC发育阶段更早的多能干细胞(PSC)衍生NPC,否则可能会增加类器官异质性的风险。 。

关键字:类器官, 中脑, 干细胞, 神经系统科学, 药物开发, 自动化, 高通量筛选

材料和试剂


1.螺纹盖管(Sarstedt ,目录号:62.547.254 [50 ml]和62.554.502 [15 ml])     

2.血清移液器(猎鹰,目录号:356543 [5 ml],356551 [10 ml]和356525 [25 ml])     

3.吸管提示(STARLAB ,产品目录号:S1120-1810 [20微升],S1120-8810 [200微升]和S1122-1830 [1000微升])     

4.标准组织培养-处理的6孔板(Sarstedt的,目录号:83.3920)     

5.锥形/ V形底编96孔板(赛默飞世,目录号:277143)     

6.真空过滤系统(Corning,目录号:431097)     

7.可选:Biomek液体处理机移液器吸头AP96 P250(Beckman Coulter,目录号:717252)     

8.小分子神经前体细胞[ smNPC (Reinhardt et al 。,2013)]     

9. KnockOut (KO)DMEM / F-12(Thermo Fisher,目录号:12660012)     

10.基质胶(BD Biosciences,目录号:354263) 

11. DMEM / F-12(Thermo Fisher,目录号:11320033) 

12.神经基础培养基(Thermo Fisher,目录号:21103-049) 

13.不含维生素A的B27补充剂(赛默飞世尔(Thermo Fisher),目录号:12587010) 

14.氮气补充剂(赛默飞世尔,目录号:17502-048) 

15.青霉素-链霉素(Sigma-Aldrich,目录号:P4333) 

16. GlutaMAX (赛默飞世尔,目录号:35050061) 

17.牛血清白蛋白(BSA,Thermo Fisher,目录号:15260037) 

18.人血清白蛋白(HSA,生物工业,目录号:05-720-1B) 

19. Accutase (Sigma-Aldrich,目录号:A6964) 

20.聚乙烯醇(PVA ;Sigma-Aldrich,目录号:363170) 

21.抗坏血酸(Sigma-Aldrich,目录号:A4544) 

22. CHIR-99021(Biomol公司,目录号:Cay13122) 

23.平滑化激动剂(SAG ;Cayman Chemical,目录号:Cay11914) 

24.脑源性神经营养因子(BDNF ;PeproTech ,目录号:450-02) 

25.胶质细胞系衍生的神经营养因子(GDNF ; PeproTech ,目录号:450-10) 

26.激活素A(PeproTech ,目录号:120-14E) 

27.转化生长因子Beta 3(TGF-β3 ;PeproTech ,目录号:100-36E) 

28. N6,2 ' -O-二丁3 ' ,5 ' 5'-环一磷酸(dbcAMP ; Sigma-Aldrich公司,目录号:D0627) 

29.基础培养基(请参见食谱) 

30. smNPC介质(请参阅食谱) 

31.聚合介质(请参阅食谱) 

32.图案化介质(请参见食谱) 

33.成熟培养基(6-X天)(请参阅食谱) 

34.拆分培养基(请参见食谱) 



设备


水浴
加湿的CO 2培养箱
可选:连接到自动液体处理系统(ALHS)的CO 2培养箱
无菌组织/细胞培养罩
冰箱和冰柜(4°C,-20°C和-80°C)
离心机(例如,Eppendorf ,型号:Centrifuge 5702)
带相机的立体显微镜(例如,Leica MZ10 F [显微镜]和Leica DFC425 C [相机] ,Leica Microsystems)
机械移液器
可选:多通道机械移液器
可选:自动化液体处理系统(例如,Biomek FX P实验室自动化工作站,Beckman Coulter)


软件


Leica应用套件4.8版(LAS,Leica Microsystems)
斐济/ ImageJ (Schindelin等,2012)
可选:Biomek软件v。3.3 (控制自动液体处理器,Beckman Coulter)


程序


用于二维smNPC细胞培养的基质胶包被板的制备
注意:在最后的涂覆步骤之前,所有包含基质胶的塑料和溶液必须放在冰上,以防止在处理过程中过早胶凝。


在冷的(4°C )KnockOut DMEM / F-12中将基质胶1:100稀释
注:我们建议保持冻结股票的1:4稀释的基质胶在淘汰赛的DMEM / F-12和进一步稀释本1:25在淘汰赛即将使用前DMEM / F-12。


在6孔板的每个孔中快速添加1 ml 1:100 matrigel溶液,以覆盖孔的整个底部。分发该轻摇板均匀分布在井的表面的解决方案。如果您想像一个XY平面与地面平行的坐标系,请先沿Y轴在任一方向上倾斜大约10度5次,然后沿X轴在任一方向上倾斜大约10度5次或直到整个板的底部被基质胶溶液覆盖。用Parafilm密封板以防止蒸发。
在室温下孵育过夜。可以将板在4°C下保存大约2周。
注意:如果在同一天需要培养板,也可以将其在37°C下孵育3小时并立即使用。


smNPC文化
注意:通常,将smNPCs接种并在6孔板中的每孔2 ml培养基中培养。随着细胞变得更加融合,请仔细监控细胞培养基。如果pH指示剂显示酸化(即,常用的酚红将变为黄色),则每天刷新培养基和/或将培养基体积增加50-100%,直至每孔最多4 ml。图1包括了处于不同阶段的smNPC培养的示例,以作为参考。


从基础培养基(请参见表1)中制备smNPC培养基(请参见表2),并在水浴中将其预热至37°C。小分子应在使用前立即添加到培养基中。重构后,避免对介质进行热循环。
吸出旧培养基。
在6孔板的每孔中加入1 ml Accutase ,并在37°C下孵育10-15分钟(在7-8分钟后检查细胞是否脱落)。细胞应在没有机械搅拌的情况下分离。
重悬细胞2-3次尽可能温和地使用1000 μ升手动移液管以最大化细胞收获并以分散小剩余的聚集体。将悬浮液转移到装有6 ml分离培养基的试管中(请参见配方,表6),以稀释和灭活Accutase 。
离心机为在220 2分钟×克。
吸出上清液。小心不要吸出沉淀物。
将沉淀重悬于smNPC培养基中,并以大约1:10-1:20的比例播种细胞(取决于细胞系和经验)。
注意:在添加细胞溶液之前,从基质胶包被的孔中吸出上清液。
通过在XY-运动轻轻摇动平板分发细胞均匀地分布在井的表面(见S TEP A2)。避免回旋/圆周运动,因为它们会使细胞集中在孔的中心。将板放入37°C的CO 2培养箱中。
交换介质隔日维持细胞,直到他们达到约90-100%confluenc Ë分裂之前再次/使用他们的组织体的产生。见图URE 1D。




2D的图1.相衬图像smNPC在confluenc的不同阶段培养ë和分化。(a / b)代表性图像显示了在以1:10(a)或1:20(b)分裂后两天后smNPC的密度。低的(C)实施例的图像-质量smNPC培养由大量分化的细胞(非常暗和大,位于集落之间)污染。由于smNPC通常比分化细胞增殖得更快,因此可以通过在多个传代中保持较高的分裂比(至少1:20)来挽救部分分化的培养物。这些类型的培养物不应用于生成类器官。(d)smNPCs在所述最佳confluenc Ë为类器官代或分裂。


类器官的产生和维护
注意:有关不同年龄的类动物体的代表性明场图像,请参见图2。


第0天


分离smNPCs从基质胶包被的使用细胞培养板ACCUTASE如上述步骤B3和B4中描述。
在分解介质收集细胞(见配方,表6),计算它们,和细胞的适当数量(见下面注释)转移到一个新˚F爱尔康管。然后以220 × g的速度旋转2分钟。
注意对细胞数量和体积:播种要求9000个细胞在150微升每孔的培养基在96孔板中。这相当于:9,000个细胞/0.15毫升= 60,000个细胞/毫升。因此,以60,000个细胞/ ml的浓度制备细胞悬液。确保准备好10-20%的细胞悬液多余体积,以解决移液不准确/死体积的问题。


将沉淀物重新悬浮在适量的聚集介质中(请参见配方,表3)。
注意:步骤4和所有后续进料步骤可以手动完成,也可以使用自动液体处理系统(ALHS)/移液机器人完成。


种子9000个细胞150微升每孔的96孔V形底的聚合介质ED板孵育在37℃,5%CO 2 48小时。


第二天


注意:所有涉及类器官的介质更换均应谨慎进行,以免损坏样品或意外将其从井中移出。为了便于手动媒体抽吸,我们建议稍微倾斜荷兰国际集团的板并执行荷兰国际集团,通过移液所有液体操作(未通过真空泵或类似装置)。离开旧培养基(≤10少量微升在孔),以便不通过用吸管尖广泛直接接触损坏类器官。少量的类器官(<约50个)可以用标准的单通道机械移液器处理。对于较大的样品数(超过一半的96孔板),我们建议使用多通道移液管(根据我们的经验,一个标准的机械吸管200 -微升容量的效果很好)。最后,进一步规模-开高-通量应用使得它可行使用移液机器人/自动液体处理器。相同的预防措施为人工处理应用-最好是离开旧培养基的少量(≤10微升),每孔而不是损伤类器官由多个远大除去液体。尽管取决于确切的设置,但此处的死体积通常要比手动移液高,我们建议准备5-10 ml多余的所有液体。


更换介质150微升每孔的图案化介质(见配方,表4)。


第四天


更换介质150微升每孔的图案化介质。


第六天


代替用150介质微升成熟培养基加激活素A,每孔(见配方,表5)。


第8天+ X


更换介质150微升每孔成熟培养基(同6天无活化素A)。
饲料每2天在相同的培养基中聚集一次,直到达到分析/跟进实验所需的年龄为止。
注意:成熟动态因细胞系而异,需要通过免疫染色或qPCR进行经验测试。通常,类器官在第30天时会包含大量MAP2 +神经元。包括GFAP +星形胶质细胞在内的神经胶质细胞出现的时间较晚,大约在40天左右。有关更多详细信息,请参阅我们以前的出版物中的详细分析(Renner等人,2020年)。


图2.代表性明场图像illustrat荷兰国际集团随时间类器官的生长和形态。接种上作为单个细胞0天smNPC悬浮液,细胞聚集于球状形状在一天之内。随着时间的流逝,它们保持其均一的圆形并不断增大。


注意:有时在井底会发现一些死细胞(尤其是在早期阶段。在这里,请参阅第15天)。只要聚集体本身保持其结构,这不是问题。在下一次喂食过程中,死细胞将被洗掉。


数据分析


在培养过程中,使用体视显微镜定期监测类器官的正常生长和形态(参见图2作为参考)。


  对于一些其他可能的下游分析和QC步骤,包括的例如,定量PCR,整装染色和结算,类器官活力,和功能分析(钙成像和多-电极阵列),请参阅我们之前的出版物(雷纳等人。 (2020年)。


笔记


如上所述,该协议有助于手动或全自动样品的生成。由于每个类器官都在单独的孔中生成,并且从播种到成熟都与所有其他类器官保持分离,因此它们代表了真正的生物学复制品。从单个细胞悬液而不是部分解离的菌落开始,还可以使同一批次的不同孔和板之间的同质性最大化。每孔生成一个类器官也极大地简化了化合物的下游应用和后续分析。在我们的运行中,我们经常产生的类器官批次与变异(系数即标准差除以在小于5%的平均值)相对于他们的最大的跨区域-部分,5%左右为他们的生存能力。所有这些因素使得摩非常适合于化合物筛选和毒性测试,因为规模-高达完全自动化的工作流程是相当微不足道的。尽管许多科研问题(例如机械分析)通常不需要成百上千的样本数量,但AMO的高度均质性和可重复性仍然具有优势,我们建议使用单道或多道移液器通过手动移液测试协议。


菜谱


注意:所有不含生长因子和小分子的培养基都可以在黑暗中于4°C下保存2周。生长因子和小分子应始终在使用前立即添加。


基底介质(表1)


表1 。基础培养基的组成


smNPC mediu米(表2)


表2 。组成smNPC媒体


聚合中介UM(T一BLE 3)




表3 。聚集介质的组成


* PVA以结晶粉末形式购买。我们建议您准备4%的原液,该原液可在4°C下保存几个月。要制备储备溶液,请在恒定搅拌下加热至60-70°C过夜,将PVA溶解在水中。溶液冷却后,在使用前用例如真空过滤系统进行无菌过滤。


图案化介质(2-4天)


表4 。构图介质的组成






成熟度(第6至X天)(表5)


表5 。成熟培养基的组成


拆分介质(表6)


表6 。拆分培养基的组成


致谢


这项工作由欧洲研究委员会(ERC)根据欧盟的Horizon 2020研究与创新计划(授权协议[669168])资助。人力资源得到了德国明斯特国际马克斯·普朗克研究学院-分子生物医学的支持。该协议基于我们以前的出版物(Renner等,2020)。


利益争夺


此处介绍的工作是欧洲专利局的专利申请EP 18 19 2698.0-1120的主题,在此是作者的发明人。


参考


Dutta,D.,Heo,I.和Clevers,H.(2017年)。干细胞衍生的3D类器官系统中的疾病建模。趋势医学杂志23(5):393-410。
Dye,BR,Hill,DR,Ferguson,MA,Tsai,YH,Nagy,MS,Dyal,R.,Wells,JM,Mayhew,CN,Nattiv,R.,Klein,OD,White,ES,Deutsch,GH和Spence,JR(2015)。我Ñ体外产生人多能干细胞衍生的肺类器官。Elife 4 :e05098。
Fatehullah,A.,Tan,SH和Barker,N.(2016年)。类器官作为人类发育和疾病的体外模型。Nat Cell Biol 18(3):246-254。
Jo,J.,Xiao,Y.,Sun,AX,Cukuroglu,E.,Tran,HD,Goke,J.,Tan,ZY,Saw,TY,Tan,CP,Lokman,H.,Lee,Y., Kim,D.,Ko,HS,Kim,SO,Park,JH,Cho,NJ,Hyde,TM,Kleinman,JE,Shin,JH,Weinberger,DR,Tan,EK,Je,HS和Ng,HH(2016 )。来自人类多能干细胞的中脑样类器官包含功能性多巴胺能和产生神经黑色素的神经元。细胞干细胞19(2):248-257。
Kim,J.,Koo,BK和JA Knoblich(2020)。人类类器官:人类生物学和医学的模型系统。Nat Rev Mol Cell Biol 21(10):571-584。
兰开斯特,马萨诸塞州,伦纳,男。马丁,加利福尼亚州,温泽尔,D。脑类器官模拟人脑发育和小头畸形。自然501(7467):373-379。              
McCracken,KW,Cata,EM,Crawford,CM,Sinagoga,KL,Schumacher,M.,Rockich,BE,Tsai,YH,Mayhew,CN,Spence,JR,Zavros,Y.和Wells,JM(2014)。在多能干细胞来源的胃类器官中模拟人类发育和疾病。自然516(7531):400-404。
Monzel,AS,Smits,LM,Hemmer,K.,Hachi,S.,Moreno,EL,van Wuellen,T.,Jarazo,J.,Walter,J.,Bruggemann,I.,Boussaad,I.,Berger, E.,Fleming,RMT,Bolognin,S.和Schwamborn,JC(2017)。从神经上皮干细胞衍生人中脑特异性类器官。干细胞报告8(5):1144-1154。              
Poewe,W.,Seppi,K.,Tanner,CM,Halliday,GM,Brundin,P.,Volkmann,J.,Schrag,AE和AE,Lang(2017)。帕金森综合症。Nat Rev Dis Primers 3:17013。              
钱谦(X.,Nguyen,HN),宋(MM),哈迪诺(Hdiono),C.,奥格登(Ogden),SC,哈马克(Hammack),C.,姚(Y.B.),哈默斯基(Halersky),GR,雅各布(Jacob),F. Jeang,W.,Lin L.,Li,Y.,Thakor,J.,Berg,DA,Zhang,C.,Kang,E.,Chickering,M.,Nauen,D.,Ho,CY,Wen, Z.,克里斯蒂安,KM,施,PY,马希尔,BJ,吴,H.,金,P.,唐H.,宋H.和明,GL(2016)。使用微型生物反应器对大脑区域特定的类器官进行建模ZIKV暴露。单元格165(5):1238-1254。
Reinhardt,P.,Glatza,M.,Hemmer,K.,Tsytsyura,Y.,Thiel,CS,Hoing,S.,Moritz,S.,Parga,JA,Wagner,L.,Bruder,JM,Wu,G 。,Schmid,B.,Ropke,A.,Klingauf,J.,Schwamborn,JC,Gasser,T.,Scholer,HR and Sterneckert,J.(2013)。仅使用小分子人类神经祖细胞进行神经退行性疾病建模的推导和扩展。PLoS One 8(3):e59252。
Renner,H.,Grabos,M.,Becker,KJ,Kagermeier,TE,Wu,J.,Otto,M.,Peischard,S.,Zeuschner,D.,TsyTsyura,Y.,Disse,P.,Klingauf, J.,Leidel,SA,Seebohm,G.,Scholer,HR和Bruder,JM(2020)。全自动的高通量工作流程,用于人类中脑类器官中基于3D的化学筛选。Elife 9 :e52904。
Sato,T.,Vries,RG,Snippert,HJ,van de Wetering,M.,Barker,N.,Stange,DE,van Es,JH,Abo,A.,Kujala,P.,Peters,PJ和Clevers, H.(2009年)。单个Lgr5干细胞可在体外建立隐窝-绒毛结构,而无需间充质位。自然459(7244):262-265。              
Schindelin,J.,Arganda-Carreras,I.,Frise,E.,Kaynig,V.,Longair,M.,Pietzsch,T.,Preibisch,S.,Rueden,C.,Saalfeld,S.,Schmid,B ,Tinevez,JY,White,DJ,Hartenstein,V.,Eliceiri,K.,Tomancak,P.和Cardona,A.(2012)。斐济:一个用于生物图像分析的开源平台。Nat Methods 9(7):676-682。
Takasato,M.,Er,PX,Chiu,HS,Maier,B.,Baillie,GJ,Ferguson,C.,Parton,RG,Wolvetang,EJ,Roost,MS,Chuva de Sousa Lopes,SM和Little,MH( 2015)。来自人iPS细胞的肾脏类器官含有多种谱系,并能模拟人的肾生成。自然526(7574):564-568。
  • English
  • 中文翻译
免责声明 × 为了向广大用户提供经翻译的内容,www.bio-protocol.org 采用人工翻译与计算机翻译结合的技术翻译了本文章。基于计算机的翻译质量再高,也不及 100% 的人工翻译的质量。为此,我们始终建议用户参考原始英文版本。 Bio-protocol., LLC对翻译版本的准确性不承担任何责任。
Copyright Renner et al. This article is distributed under the terms of the Creative Commons Attribution License (CC BY 4.0).
引用: Readers should cite both the Bio-protocol article and the original research article where this protocol was used:
  1. Renner, H., Grabos, M., Schöler, H. R. and Bruder, J. M. (2021). Generation and Maintenance of Homogeneous Human Midbrain Organoids. Bio-protocol 11(11): e4049. DOI: 10.21769/BioProtoc.4049.
  2. Renner, H., Grabos, M., Becker, K. J., Kagermeier, T. E., Wu, J., Otto, M., Peischard, S., Zeuschner, D., TsyTsyura, Y., Disse, P., Klingauf, J., Leidel, S. A., Seebohm, G., Scholer, H. R. and Bruder, J. M. (2020). A fully automated high-throughput workflow for 3D-based chemical screening in human midbrain organoids. Elife 9: e52904.
提问与回复
提交问题/评论即表示您同意遵守我们的服务条款。如果您发现恶意或不符合我们的条款的言论,请联系我们:eb@bio-protocol.org。

如果您对本实验方案有任何疑问/意见, 强烈建议您发布在此处。我们将邀请本文作者以及部分用户回答您的问题/意见。为了作者与用户间沟通流畅(作者能准确理解您所遇到的问题并给与正确的建议),我们鼓励用户用图片的形式来说明遇到的问题。

如果您对本实验方案有任何疑问/意见, 强烈建议您发布在此处。我们将邀请本文作者以及部分用户回答您的问题/意见。为了作者与用户间沟通流畅(作者能准确理解您所遇到的问题并给与正确的建议),我们鼓励用户用图片的形式来说明遇到的问题。