Quantitative approach to numbers and sizes: Generation of primary neurospheres from the dorsal lateral ganglionic eminence of late embryonic mice

Background: The neurosphere assay is a powerful in vitro tool to investigate neural stem cells in the dorsal lateral ventricle (dLGE). In the dLGE, metrics of sizes and numbers of neurospheres generated using this assay has not been completely characterized. The objective of this protocol is to provide a stepwise method from a single isolation that predicts the average number of neurospheres generated and to estimate an approximation of its sizes after several days in vitro. The advantage of this protocol is that no expensive and specialized equipment is needed for tissue isolation. Estimates about the numbers and sizes of neurospheres will provide investigators with quantitative data to advise on how much starting dLGE tissue is required to generate the appropriate number of spheres for the implementation of downstream applications, including immunocytochemistry, self-renewal and differentiation assays. Methods: Our method is based on a simple dissection technique, where tissue surrounding the dorsal lateral ventricle from a single mouse embryo is trimmed away to enrich for neural stem cell and progenitor populations. Following this dissection, tissue is mechanically dissociated by trituration. Cells are then cultured in media containing epidermal growth factor and other supplements to generate healthy primary neurospheres. Results: Using this approach, we found reproducible number of primary neurospheres after 7 days in vitro (DIV). Furthermore, we observed that this method yields an average range of neurospheres sizes greater than 50 μm, but less than 100 μm after 7 DIV. Lastly, using an anti-GFAP antibody, we show that these neurospheres can be stained, confirming their use in future immunocytochemistry studies. Conclusions: Future use of this protocol provides metrics on the generation of primary neurospheres that will be useful for further advances in the area of stem cell biology.


Introduction
Neural stem cells are a tissue-specific subtype of self-renewing and multipotent cells that will produce several mature cell types. The neurosphere assay is an important tool that has been extensively employed to study neural stem cell biology 1 . Since its introduction some 25 years ago 2 , neurospheres have been used to study neurogenesis 3 , genes that regulate selfrenewal 4,5 , and molecular mechanisms that control neuronal and glial differentiation 3,6-9 . Although there are many neurosphere protocols, the expected number and size of neurospheres generated after a week in vitro is not entirely characterized.
We developed a simple dissection technique that helps to maximize the number of neurospheres that can be produced in culture. Furthermore, we characterize the expected sizes of neurospheres after a week in vitro. With some other techniques, a brain slicer or other means are used to obtain thick slices of brain tissue from late embryonic stages 10,11 . The area surrounding the ventricle is then microdissected from a given slice of tissue to enrich for neural stem/progenitor cells. This approach, while effective, can be painstaking and may require expensive specialized equipment. Additionally, many protocols do not provide metrics on expected numbers and sizes of neurospheres generated. Thus, it is unclear whether researchers can generate sufficient numbers of neurospheres in a particular range of sizes. In contrast, our approach requires no specialized equipment. The lateral ventricle is visualized with a stereomicroscope, and the surrounding tissue is simply trimmed away using a razor blade or scalpel. This method requires only half of a single brain, and generates reproducible numbers of neurospheres in a few days. Furthermore, using our method neurospheres small appears as early as 3 days. Another advantage of this protocol is that it can generate neurospheres with average sizes of 50 μm -100 μm after 5-7 days in vitro.

Mice
The animals were housed in the AAALAC-accredited East Campus Research Facility and Transgenic Mouse Core Facility in the Veterinary College of Cornell University. All animal procedures were performed in accordance with the guidelines outlined in the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals, eighth Edition. Animals were approved by Cornell University's Animal Care and Use Committee (IACUC; #01-75). Mice were maintained on a mixed 129Sv/ C57BL/6 background and housed on a reverse light-dark cycle. Food and water were continuously available. Male and female mice mated overnight. The following morning females were separated and checked for vaginal plug. Pregnant mice were euthanized using CO 2 asphyxiation followed by cervical dislocation consistent with the recommendations of the Panel on Euthanasia of the American Veterinary Medical Association and the Cornell IACUC. Embryonic day 17.5 (E17.5; E0.5 was defined as the first detection of the vaginal plug) male and female embryos were dissected. A total number of n=5 mice were used for primary neurospheres experiments.

Collection of tissue
The mouse brain was sliced down the center into two hemispheres. Using one hemisphere, placed sagittal orientation, dissect out the dorsal lateral ventricle (see Figure 1). The dissections were guided using stereotaxic coordinates (A/P 1 mm, M/L 1 mm, D/V 2.3 mm) from Paxinos and Franklin (2007) atlas source.

Data acquisition and statistics
Images were taken with a Canon EOS Rebel XS camera. (Canon USA; Melville, NY). The optimum magnification is approximately 5x with 3888 x 2592 dimensions. Camera was connected to the trinocular port of the stereomicroscope (Carl Zeiss Stemi 305; White Plains, NY) using Mount Adaptor EF-EOS (6098B007AA; Canon; Melville, NY). The working distance was defined as the amount of room required between the top of the neurosphere and the bottom of the objective lens in order for the image to be in focus. The steromicroscope was used at a working distance of ~110 mm. Due to the variation in neurosphere size, 110 mm should be adjusted to focus on the desire region of the neurosphere to provide optimal focus. The field of view represents a length of 783 μm and a width of 522 μm. Data is derived from single random pictures of each well. Per animal, 3-4 wells were analyzed. A total of 5 individual animals were analyzed. Size measurements and neurosphere counts were analyzed in Adobe Photoshop (version 10.0.1) CS3 Extended (Adobe; San Jose, CA) using the measurement function (ImageJ is an open-access alternative that can be used to perform this function). Statistical analyses were carried out as previously described 12,13 . Briefly, data were analyzed using PRISM 8 (version 8.3.0) (GraphPad Software; San Diego, CA) by performing one-way ANOVA. If the main effect was significant (p < 0.05), Bonferroni's multiple comparison post hoc test were used to compare the different replicates. The null hypothesis was rejected at p < 0.05. Data is made available on figshare open access platform (Metrics of Primary Neurospheres). Error bars represent standard error of the mean (±SEM).

Neurosphere assay protocol
This protocol is designed to generate neurospheres from a single embryo. Multiply all values as needed to generate neurospheres from additional embryos. See Table 1 and Table 2 for premade solutions and materials needed. Have a petri dish (100 mm) prepared with ice-cold Hank's buffer kept on ice, which will be used to collect embryos after dissection. Afterwards, additional petri dishes will be needed to place in each of the dissected brains (35 mm).
2.1) Spray the abdomen with 70% ethanol, and make an incision to expose the uterus. Remove the uterus and transfer it to an empty petri plate.
2.2) Remove embryos from the uterus, spray desired number with 70% ethanol, and decapitate one or more embryos. Rinse each decapitated head in one petri plate containing icecold Hank's buffer, and then transfer to the second petri plate containing Hank's buffer on ice.
2.3) Use forceps to remove the skin and skull. Remove the brain and place in an empty petri dish.
2.4) Separate the two hemispheres with a razor blade, and place one half of a brain on its lateral surface.
2.5) Using a stereomicroscope, identify the location of the lateral ventricle on the medial surface (the dorsal region of the lateral ventricle contains the dLGE). The ventricle is visible as a T-shaped structure that is slightly darker than the rest of the brain. Using a razor blade or a scalpel, sequentially trim away the brain surrounding the ventricle on all four sides. 3.4) Add 10 mls of room temperature Hank's to trypsin/tissue mixture and incubate at 37°C for 5 min. with intermittent swirling. Spin culture at 300 RCF for 3 min. and remove the supernatant.
3.5) Repeat wash step 3.4 two additional times.
3.6) Aspirate the supernatant and add 4 mls of Hank's-low BSA.
3.7) Triturate the tissue gently and slowly approximately 10 times with an 18-gauge needle until tissue chunks appear relatively uniform in size. Avoid creating bubbles or foam.
3.8) Triturate the crude cell suspension gently and slowly approximately 7-10 times with a 21-gauge needle until tissue chunks appear relatively uniform in size.
3.9) Triturate the suspension approximately 4-5 times with a 23-gauge needle until suspension appears uniform.
3.10) Add 3 mls of Hank's-high BSA solution to a 15 ml tube. Slowly add the cell suspension to the bottom of the tube 3.11) Centrifuge at 300 RCF for 5 min.

Results
This approach generates consistent number of primary neurospheres Figure 1 shows a general overview of the neurospheres assay. A picture representation of primary neurospheres grown for 7 days in vitro is given in Figure 2A. The statistical analysis using a one-way ANOVA revealed no significant difference between the average numbers of neurosphere per field of view (F (4,13) = 0.666; p = 0.6268; N=5) ( Figure 2B). Figure 3 shows in the variation in sizes of neurospheres grown for 7 days in vitro ( Figure 3A). The statistical analysis using one-way ANOVA revealed a significant difference in the sizes of neurospheres between the replicates (F (4,129) = 11.666; p < 0.0001) ( Figure 3A). Using similar analyses, we found significant differences between the size classification of neurospheres that were less than 50 μm, between 50-100 μm, and greater than 100 μm (F (2,379) = 424; p < 0.0001) ( Figure 3B). Post hoc analysis using Bonferroni's multiple comparison revealed a significant difference between the primary neurospheres that were greater than 100 μm compared to neurospheres that were less than 50 μm (p <0.0001) or between 50-100 μm (p <0.0001) ( Figure 3B). Similarly, we found a substantial difference The comparison of the numbers of neurosphere that are less than 50 μm, between 50-100 μm, and greater than 100 μm. Key to statistics **, *** = p <0.01, 0.001, respectively, in comparison to NS less than 50 μm or replicate 1. #, ##, ###, = p < 0.05, 0.01, 0.001, respectively, in comparison to NS between 50-100 μm or replicate 2. $ = p <0.05 in comparison to replicate 3 (minimum of 5 independent samples; N=29, N=214, N=136, respectively to NS<50 μm, NS between 50 μm -100μm, NS>100 μm). between primary neurospheres that were greater than 100 μm compared to neurospheres that were between 50-100 μm (p <0.0001) ( Figure 3B). Numbers and sizes of neurospheres, alongside the raw images used to produce these values, are available as Underlying data 14 .

This protocol generates different sizes of neurospheres
Primary neurospheres at 7 days in vitro can be used for immunocytochemistry Neurospheres can be used for a variety of purposes, including immunocytochemistry. Figure 4 is a picture of a small ( Figure 4A-B; arrowhead) and a larger ( Figure 4C; arrowhead) primary neurosphere immunostained using an anti-GFAP antibody and counterstained with DAPI ( Figure B; arrowhead). Additional neurosphere staining using other antibodies can be found in previous published studies 3 from our lab.

Discussion
There are several key steps that are important to maximize the yield and health of neurospheres. The most important step is to incubate the final triturated culture in the prewarmed DMEM/ F12 with serum for 2-4 hours. This incubation time is necessary in order for the antibiotics in the media to inhibit growth of bacteria. A sign of bacteria contamination is reduced visibility of the media. Another marker of an unhealthy culture is a large number of differentiated neurons surrounding neurospheres. Indicators of differentiation are the large presence of axons and dendrites in your cultures. This can be caused by depletion of growth factors. If this is the case, it is recommend that you increase the concentration of EGF. Another cause of differentiation is too many cells in your prep. This leads to over crowdedness. It is recommend to split the culture to a lower density or decrease the number of neurons that are plated per 48 well.
Another essential step is to perform the trituration as gently as possible. Over-trituration, or trituration with great force, will result in increased cell death. After trituration, if a uniform suspension has not been achieved an alternative method used in previous protocols are strainers 15 .
Additional modifications to the approach may be required. We have provided general guidelines for the number of cells to be plated per well and the expected number of neurospheres per field of view. If too many cells are plated per well, differentiation may occur. If so, either reduce the number of cells plated per well or increase the concentration of EGF to 20 ng/ml. If desired, trypsin inhibitor can be substituted for the FBS during the generation of primary neurospheres to inactivate proteolytic activity.
Our approach utilizes a mechanical trituration to dissociate cells. By sequentially processing the cells through successively higher gauge needles, we obtain very few clumps of cells. This eliminates the need for a cell strainer, which can reduce yield, as well as the time needed to perform enzymatic dissociation. Although cell death can be increased with mechanical dissociation, in our hands, this does not seem to affect neurosphere yield or health.
One advantage of our approach is the ease with which tissue surrounding the lateral ventricle can be isolated from the rest of the brain. Although this dissection is relatively crude, it is easier and faster than other approaches, which may require a brain slicer and/or microdissection 10,11 . This method eliminates the need for specialized equipment while generating high numbers of neurospheres from a single brain. This method has been shown in other studies to generate neurospheres greater than 300 μm in size when grown for 12 days in vitro 3 . This project contains the following underlying data: • Spreadsheet containing numbers and sizes per field of view.xlsx (details on numbers and sizes of neurospheres produced from each mouse).

© 2019 Shandra O.
This is an open access peer review report distributed under the terms of the Creative Commons , which permits unrestricted use, distribution, and reproduction in any medium, provided the original Attribution License work is properly cited.

Oleksii Shandra
Fralin Biomedical Research Institute, Virginia Tech Carilion, Roanoke, VA, USA Virginia Polytechnic Institute & State University, Blacksburg, VA, USA This method article written by Dr. Blackwood is aimed to provide a simple and reproducible protocol generating neurospheres from lateral ganglionic eminence in late embryonic mice and demonstrate its applicability for further immunocytochemical assays. Several aspects and protocol steps, however, require clarification: This protocol includes multiple sections with crucial steps. One suggestion that could make the protocol even greater for educational purposes is providing a brief paragraph to sections 1, 2 and 3 (prior to individual steps) that will explain what is important in this section and list all the necessary reagents and tools that will be used. This will help anyone who is using this protocol for the first time to prepare all required items for this protocol section easier rather than finding the specific reagents and volume from each individual step. Table 1 details equipment, reagents and catalogue numbers; however, the digital camera used for acquiring the images is not listed. The model of stereo stereomicroscope (step 2.5) should be added to table 1, as well as the optimum magnification required or used in this protocol. The type and model of humidified incubator in step 3.19 would also be helpful for reproducibility.
Section 3 called "primary neurosphere culture" is the main section explaining the method of neurosphere generation. In other publications, e.g. Raponi et al. (2007) , a filtration using a 40um-mesh-nylon cell strainer to dissociate cells after triturating. It may be valuable to add a note after trituration step that at this point similar step could be done if the purpose is to obtain uniform suspension from tissue.
The "Data acquisition and statistics" section requires more details. Specifically, the additional valuable information could be addressing the distance from the culture plate, whether this is an important factor for further size measurements and if so what was the distance between the camera and the well. What was the zoom factor used here?
The section "Primary neurospheres at 7 days can be used for immunocytochemistry" is in vitro aimed to demonstrate applicability of this protocol. The GFAP/DAPI is used as an example. While it is stated in the "Introduction" that neurospheres have been used to study molecular mechanisms controlling differentiation of both neural and glial cells, it is not very clear why GFAP was used for this example. Maybe author could clarify or explain why more specific markers were not used here instead of GFAP. From the Fig. 4 it looks as only 1 cell within the neurosphere was GFAP+, hence I believe using a more specific antibody could produce more representative result. Several references provided to this review detail markers that could also be used.
Lastly, I believe another interesting question that could be potentially valuable here is the step explaining the reader, which factors or characteristics of the neurosphere help to determine that the neurosphere is mature.

1.
Thank you for your comments. We have revised accordingly and provided a brief Response 2.1: paragraph to section 1, 2, and 3 prior to the individual steps.
Section 1:Set-up prior to tissue dissection NOTE: Breeding and euthanasia of all animals should be performed in accordance with an institutionally approved animal care and use protocol. Sterilize all surgical instruments packed in aluminum foil in an autoclave at 121°C (15 psi) for 30mins. This includes a scissors, forceps, and razor blades. Before starting all premade solutions should be warmed to 37°C.
Section 2: Tissue dissection NOTE: Make freshly prepared 70% ethanol spray prepared. Have a petri dish (100 mm) prepared with ice-cold Hank's buffer kept on ice that will be used for embryos after dissection. Afterwards, additional petri dishes will be needed to place in each dissected brain (35 mm) Section 3:Primary neurosphere culture NOTE: Before starting warmed to 37°C the following solutions: trypsin/EDTA serum media, Hank's-low, and Hank's-high. In this section you will need the 18-gauge, 21-gauge, 23-gauge needle will be needed for trituration steps. Trituration should be performed gently and slowly to avoid killing cells. Hemocytometer will be needed to count cells. Table 1 details equipment, reagents and catalogue numbers; however, the digital Comment 2.2 camera used for acquiring the images is not listed. The model of stereo stereomicroscope (step 2.5) should be added to table 1, as well as the optimum magnification required or used in this protocol. The type and model of humidified incubator in step 3.19 would also be helpful for reproducibility.
. We have revised the Table 1 to add the Camera and Humidifier. We also included Response 2.2 information for the optimum magnification to Data acquisition and statistics section.
"Images were taken with a Canon EOS Rebel XS camera. (Canon USA; Melville, NY). The optimum magnification is approximately 5x with 3888 x 2592 dimensions. Camera was connected to the trinocular port of the stereomicroscope (Carl Zeiss Stemi 305; White Plains, NY) using Mount Adaptor EF-EOS (6098B007AA; Canon; Melville, NY). Section 3 called "primary neurosphere culture" is the main section explaining the Comment 2.3 method of neurosphere generation. In other publications, e.g. Raponi et al. (2007) , a filtration using a 40um-mesh-nylon cell strainer to dissociate cells after triturating. It may be valuable to add a note after trituration step that at this point similar step could be done if the purpose is to obtain uniform suspension from tissue.
. Thank you have revised the "primary neurosphere culture" section accordingly. Response 2.3 "After trituration, if a uniform suspension has not been achieved an alternative method used in previous protocols are strainers ." The "Data acquisition and statistics" section requires more details. Specifically, the Comment 2.4 additional valuable information could be addressing the distance from the culture plate, whether this is an important factor for further size measurements and if so what was the distance between the camera and the well.
We have revised the Data acquisition and statistics" section according to your Response 2.4: recommendation. The scale bar should be used to advise the size of the neurosphere. Additionally, 1 14 recommendation. The scale bar should be used to advise the size of the neurosphere. Additionally, we have included information about the working distance to the "Data acquisition and statistics" section.
"The working distance was defined as the amount of room required between the top of the neurosphere and the bottom of the objective lens in order for the image to be in focus. The stereomicroscope was used at a working distance of ~110 mm. Due to the variation in neurosphere size, 110 mm should be adjusted to focus on the desire region of the neurosphere to provide optimal focus." The section "Primary neurospheres at 7 days can be used for Comment 2.5: in vitro immunocytochemistry" is aimed to demonstrate applicability of this protocol. The GFAP/DAPI is used as an example. While it is stated in the "Introduction" that neurospheres have been used to study molecular mechanisms controlling differentiation of both neural and glial cells, it is not very clear why GFAP was used for this example. Maybe author could clarify or explain why more specific markers were not used here instead of GFAP. From the Fig. 4 it looks as only 1 cell within the neurosphere was GFAP+, hence I believe using a more specific antibody could produce more representative result. Several references provided to this review detail markers that could also be used.
To provide more evidence that this protocol can be used to stain neurospheres, Response 2.5: we have included an additional picture to the panel ( Figure 4C). Larger neurospheres will contain more GFAP-expressing neurons that can be labeled. Furthermore, we have reference our previous published report that demonstrated other immunocytochemistry using other antibodies.
Lastly, I believe another interesting question that could be potentially valuable here Comment 2.6: is the step explaining the reader, which factors or characteristics of the neurosphere help to determine that the neurosphere is mature.
Thank you for the recommendation. We have revised the manuscript to expand Response 2.6: this point in our discussion. "A sign of bacteria contamination is reduced visibility of the media. Another marker of an unhealthy culture is a large number of differentiated neurons surrounding neurospheres. Indicators of differentiation are the large presence of axons and dendrites in your cultures.This can be caused by depletion of growth factors. If this is the case, we recommend that you increase the concentration of EGF. Another cause of differentiation is too many cells in your prep. This leads to over crowdedness. We recommend splitting the culture to a lower density and decreasing the number of neurons that are plated per 48 well."

Yes
No competing interests were disclosed.

Competing Interests:
Reviewer Expertise: My area of expertise is behavioral neuroscience, including in vivo electrophysiology, and various learning and conditioning methods I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias You can publish traditional articles, null/negative results, case reports, data notes and more The peer review process is transparent and collaborative Your article is indexed in PubMed after passing peer review Dedicated customer support at every stage For pre-submission enquiries, contact research@f1000.com