Student remote and distance research in neuroanatomy: Mapping Dscaml1 expression with a LacZ gene trap in mouse brain

Abstract

Undergraduate student engagement in research increases retention and degree completion, especially for students who are underrepresented in science. Several approaches have been adopted to increase research opportunities including curriculum based undergraduate research opportunities (CUREs), in which research is embedded into course content. Here we report on efforts to tackle a different challenge: providing research opportunities to students engaged in remote learning or who are learning at satellite campuses or community colleges with limited research infrastructure. In our project we engaged students learning remotely or at regional centers to map gene expression in the mouse brain. In this project we mapped expression of the Down syndrome cell adhesion molecule like 1 (Dscaml1) gene in mouse brain using a LacZ expression reporter line. Identifying where Dscaml1 is expressed in the brain is an important next step in determining if its roles in development and function in the retina are conserved in the rest of the brain. Students working remotely reconstruct brain montages and annotated Dscaml1 expression in the brain of mice carrying one or two copies of the gene trap. We built on these findings by further characterizing Dscaml1 expression in inhibitory neurons of the visual pathway. These results build on and extend previous findings and demonstrate the utility of including distance learners in an active research group for both the student learners and the research team. We conclude with best practices we have developed based on this and other distance learner focused projects.

Introduction

Engagement in research has been shown to increase undergraduate student success, especially for students who are members of underrepresented groups (Joshi, Aikens et al. 2019, Crews, Wilson et al. 2020, Malotky, Mayes et al. 2020, Schneider Burton and Vicente 2020). In the wake of the COVID-19 pandemic many undergraduate research opportunities were curtailed, adapted to computational or at home learning formats, or eliminated (Chandrasekaran 2020, Forrester 2021). Efforts to identify strengths and weaknesses of online research programs are ongoing and the lessons can be applied to non-pandemic situations to support students engaged in distance learning (Erickson, Cole et al. 2022). Here we report on our experience providing online and distance research opportunities in neuroanatomy and histology to students who were not able to access university facilities or who were based at regional distance learning centers or a community college that lacked lab facilities. Students were successfully able to initiate and complete research projects over this period and going forward the model will be utilized to continue offering distance learning research opportunities for undergraduate students.

Students engaged in a research project to map expression of a neural cell adhesion molecule in the mouse brain. Genes that encode cell adhesion molecules (CAMs) are essential for the proper development and function of neural tissues (Edelman and Jones 1998, McNeill 2000). CAMs connect cells to each other and their environment by connecting the inter- and intracellular compartments. Students mapped expression of one such CAM, the Down syndrome cell adhesion molecule like 1 (Dscaml1), an Ig – superfamily transmembrane protein. Dscaml1 is involved in diverse roles within the central nervous system. Its roles include the regulation of developmental cell death, and the spatial arrangements of neurons and dendrites (Fuerst, Bruce et al. 2009, Cui, Lao et al. 2013). In the retina, Dscaml1 is expressed in sub-populations of amacrine cells (All amacrine cells), rod bipolar cells, and rod photoreceptors, and plays a role in self-avoidance, thereby facilitating even distribution of cells across the surface of the retina. Dscaml1 is also highly expressed in the hippocampus, cerebellum, and cortex, but systematic mapping of its expression has not been completed and functional studies outside of retina are limited (Barlow, Micales et al. 2002, Zhang, Huang et al. 2015). Disruption of Dscaml1 expression has been associated with diseases, such as congenital stationary night blindness and Autism Spectrum Disorders (ASD) and a better understanding of its roles will be essential to developing interventions for these disorders (Karaca, Harel et al. 2015, Astuti, van den Born et al. 2018). Identifying where Dscaml1 is expressed in the central nervous system is an essential first step to understanding how Dscaml1 contributes to proper neural development and function. The complexity of the brain required significant effort. We utilized students learning remotely or at distance education centers by pairing them with students learning in the lab to tackle this challenge. The inclusion of distance education in this project opened up research opportunities for students learning off campus and aided students working on site in the lab by providing additional support for their projects and opportunities to develop scientific communication and teambuilding skills.

Methods to map gene expression include gene expression reporters, measuring mRNA or protein localization. Each of these has advantages and disadvantages. For example, in situ hybridization is a robust technique but detects mRNA in the nuclei, which can make identifying which cells express a gene difficult in tissues with mixed neuron populations. Immunohistochemistry provides information on protein localization but is limited by antibody availability and is challenging for synapse associated proteins such as DSCAML1. In this study we utilized a LacZ knock-in gene trap that we have previously characterized to complement existing in situ expression libraries (Fuerst, Bruce et al. 2009). The bacterial gene, β-galactosidase (LacZ), is incorporated into many prokaryotic and eukaryotic transgenic model systems and provides an easy to develop reagent to monitor gene expression (Skarnes, Auerbach et al. 1992). The β -gal enzyme cleaves β - galactoside carbohydrates into monosaccharides and is widely used in laboratory settings because of its ability to cleave the chromogenic chemical 5-bromo-4-chloro-3-indolyl- β-d-galactopyranoside (x-gal), resulting in deposition of a blue substrate that is easy to visualize. The staining procedure for this protocol is straightforward and well suited to inclusion of undergraduate researchers. To develop brain maps, we paired students working in the lab with students working remotely. Students in the lab fixed, stained, cut and imaged brain sections. Students working remotely manually stitched images of brain sections into montages, identified their corresponding location using the Allen brain atlas and mapped the regions of the brain in which Dscaml1 is expressed. Mapping of Dscaml1 expression was performed in sequential coronal brain sections of mouse brain in control and Dscaml1 mutant brain. Localization of staining in cell bodies and axons permitted identification of cells and associated axon tracts. We further characterized expression of Dscaml1 in inhibitory neurons within the central visual pathway to demonstrate the utility of the atlas students generated as part of this project.

Materials and Methods

Pedagogy: Remote Student Recruitment

Identifying remote students who are a good match for a research project is a key component of our ability to successfully integrate remote students into our team. To this end the support of faculty working at the community college or distance center was essential. Faculty teaching at such locations are typically focused on teaching and will be well situated to recommend students who would be able to make positive contributions to research projects. Writing to the department or unit chair and asking to be put in touch with faculty members they would recommend is a good start. Developing a professional working relationship with the contact person at the remote site will be very helpful. In our experience the remote contact has consistently been able to identify highly motivated students who are looking for research experiences not available locally. Our remote contact was also able to identify students who were interested in transferring from a two-year community college to a four-year institution to complete a bachelor’s degree and postgraduate studies. Such students benefit from having local contacts at the institution they transfer to, are able to pick up in person lab work very readily and are very knowledgeable about supporting the next wave of distance research students.

Pedagogy: Remote Workstations

We established remote worksites for students at a community college and at a distance learning center. The workstation consisted of a dedicated space with computers (community college) or computers and limited lab space (distance learning center). While this was not essential for our projects it permitted student access to higher power computers and a distraction free environment in which to work. It also helped to raise visibility of the project and remote research opportunities at the community college. Support from our faculty liaisons at the community college and regional center were essential in establishing and maintaining these facilities.

Pedagogy: Student Introductions and Project Identification

Students were introduced to the PI by the distance learning center contact. The PI reached out to the student and provided a manuscript relevant to their research and asked the student to attend lab meetings by zoom. During lab meetings the student leads on given research projects review progress and challenges encountered, with each student working on the project providing an update, which might range from their focus on exams over the last week to presentation of new data. During this process specific needs were identified that could be addressed by distance researchers, for example in image analysis or possible side projects. After the meeting the PI would reach out to the in-person and remote student to determine if the in-person student would like support from the remote student and if the remote student would be interested in contributing to the specific project. Incentive for the in-person students to work with the remote students are two-fold: support on the research project and an opportunity for leadership experience on a research team. As undergraduate researchers are often planning postgraduate education the opportunity to gain leadership experience is very valuable.

Pedagogy: Central Data Base

Establish a central online data center that students can use to access and share files and data. This has several benefits including the ease of providing images that remote students can work on, maintaining a centralized site for organizing results and quantification of results and for project continuity, specifically students matriculating into other programs. We used several different sharing platforms including Dropbox and Onedrive. We currently utilize Onedrive for sharing files among students and maintain a hard copy of data off the cloud to ensure that accidental deletion of material by students does not occur. Organization of the data base will be very helpful for the PI and students.

Pedagogy: Monitoring Progress

We hold weekly lab meetings, utilizing Zoom, to ensure that students and projects were progressing. Each student would provide a brief update on their work over the last week and plans for the upcoming week. Given the reality of undergraduate studies this was inconsistent for us because of student examination schedules but helped us to monitor how well students were working together and if more instructions were needed.

Pedagogy: Lab Visits

We organized lab visit days for students working remotely. These would involve the remote students visiting our campus and lab and observing some of the techniques as possible given safety considerations. Students would have an opportunity to meet each other and share a lunch depending on pandemic restrictions. Remote students reported a great deal of satisfaction and enjoyment in these visits.

Pedagogy: Student Incentive: Degree Requirements, Professional Development and Salary

While not specific to remote student research, student motivation to engage in research is varied and can include factors in addition to interest in the research projects, including degree requirements, salary and professional development opportunities such as presentation, meeting and grant opportunities. Student degree requirements can be fulfilled at our university through remote research, for example 1 credit hour per three hours spent working on a project per week. Examples include capstone and thesis requirements and upper division credit requirements. Additional student incentives included the opportunity to present research at local meetings and to apply for in house research grants to support the student’s research. The availability of summer internship support for students should also be explored. Our university had multiple opportunities for students to complete paid summer internships through support from the National Institutes for Health (NIH) and National Science Foundation (NSF) and foundations that support undergraduate research. Finally, salary for student work was occasionally provisioned as available. Considering availability of all opportunities can facilitate additional student engagement and resources for the lab group.

Pedagogy: Aligning University, College, Center and Faculty Goals

An additional faculty benefit of working with students at community colleges and distance learning centers relates to the institutional goals of each. Understanding what these are and aligning our research team with these helped raise the visibility of our research program, garner support for the research team and aid in faculty promotion reviews. In our case these were the following: University: The University of Idaho has a long-standing goal of increasing enrollment. Working with students at community colleges, many of whom transferred, supported the university goal of increasing enrollment. Community College: Our partner community college unit has a goal of preparing students for transfer and completion of a bachelor’s degree. We supported this mission by providing opportunities not otherwise available for the college’s students and by providing marketing stories for both institutions about student pathways, for example from a community college to completing medical school. Distance center: Our distance learning center had underutilized facilities and was not as well connected to the home campus. By providing space for students learning remotely the center was able to support research and education missions that it was not previously providing. In our case our research team’s small number of students nucleated a larger culture wherein multiple research teams now sponsor groups of students conducting research remotely out of our distance education center.

Mice and Tissue (in lab)

We utilized mice that were heterozygous (Dscaml1 ^GT/+ ; one copy of the gene trap) or homozygous (Dscaml1 ^GT/GT ; two copies of the gene trap) mice in this study (Fuerst, Bruce et al. 2009). GAD67-GFP mice were used to track inhibitory interneurons in the central visual pathway (Munsch, Yanagawa et al. 2005). Mice were housed in the University of Idaho Laboratory and Research Facility on a 12-hour light:dark cycle, and fed ad libitum. Tissue collection took place between 8 and 16 weeks of age. Mice were anesthetized and transcardially perfused with 1X phosphate buffered saline (PBS) (10X PBS: distilled H₂O, 80g NaCl, 2.0g KCL, 26.8g Na₂HPO₄ ^. 7H₂O, 2.4g KH₂PO₄, pH 7.4). Brains were fixed in 4% paraformaldehyde (PFA) for four to six hours, then washed three times with 1X PBS. Brains were cut into 100 µm coronal sections, using a vibratome, and slices were stored in 1X PBS at 4°C until staining. All procedures were approved by the University of Idaho Animal Care and Use Committee.

X-gal staining (in lab)

X-gal staining stock solution contained 100 mg of x-gal, ultrapure grade in 1 mL of dimethyl sulfoxide (DMSO). X-gal solution was diluted in either 1X PBS or Z-buffer (16.1g Na₂HPO₄·7H₂O, 5.5g NaH₂PO₄· H₂O, 0.75g KCl, 0.246g MgSO₄·7H₂O, 2.7 ml β-mercaptoethanol, H₂O to volume of 1L, pH 6.9). For dilutions: x-gal solution was diluted in either 1X PBS or Z-buffer to 1:100, 1:250, 1:500, 1:1000, and 1:5000 dilutions for optimization. Two slices were stained in each dilution condition.

Before the onset of the COVID-19 pandemic students worked to optimize the concentration of x-gal staining solution to use in our procedures. We performed LacZ staining at 1:100, 1:250, 1:500, 1:1000, and 1:5000 x-gal concentrations in 1X PBS at room temperature (22°C). We determined that a 1:250 dilution of x-gal stock was most suitable due to the staining intensity of known DSCAML1 localization and the limit of over-saturation of the surrounding tissue (Fig. 1). This dilution helps to conserve resources for undergraduate focused research projects and provided an introductory experiment for students to become acquainted with lab protocols.

Figure 1: X-gal concentration optimization experiment.

Optimal x-gal concentration was determined to be 1:250 after testing different concentrations. This concentration was used for all further experiments. The tested x-gal concentrations were 1:100 (manufacturer’s recommended concentration), 1:250, 1:500, 1:1000, and 1:5000. Scale bar = 1 mm.

Students next tested different temperature conditions (4°C, 22°C, and 37°C), using a 1:250 dilution in either 1X PBS or the manufacturer-recommended Z-buffer. There was no strong difference between Z-buffer and 1X PBS at 4°C (Fig 2A). However, sections stained at 4°C and 22°C showed more prominent LacZ staining in cell bodies compared to sections stained at 37°C. Additionally, sections stained in 1X PBS showed more prominent cellular staining than sections stained in Z-buffer at 22°C and 37°C (Fig. 2B, C). Sections stained in Z-buffer reached full saturation in one-third of the staining timeframe as sections in 1X PBS at both 22°C and 37°C.

Figure 2: X-gal staining temperature and buffer optimization experiment.

Optimal staining protocol was a 1:250 x-gal dilution in 1X PBS at 22°C. A) X-gal staining in 1X PBS or Z-buffer at 4°C. B) X-gal staining in 1X PBS or Z-buffer at 22°C. C) X-gal staining in 1X PBS or Z-buffer at 37°C. Scale bar = 1 mm.

Optimized Protocol

Based on optimal results from x-gal dilution experiments, brain slices were stained in a 1:250 x-gal solution diluted with either 1X PBS or Z-buffer at 4°C, room temperature (22°C), or 37°C. Three sections were stained in each condition. Slices were imaged every 12 hours for 4°C and 22°C. Sections were imaged every four hours for 37°C. Sections in 1X PBS were stained for a total of 144 hours at 4°C, 72 hours at 22°C, and eight hours at 37°C. Sections in Z-buffer were stained for a total of 144 hours at 4°C, 24 hours at 22°C, or and eight hours at 37°C.

Imaging (in lab)

Image acquisition of brain montages was performed with a Zeiss Axio Lab.A1 microscope with an Axiocam 105 color camera using ZEN 2.3 software (Zeiss). Images used for optimization experiments were acquired using a dissection scope with camera attachment.

Image montage (remote)

Images were processed using Adobe Photoshop for montage compilation and the publicly available NIH software FIJI ImageJ was used for image analysis (Schindelin, Arganda-Carreras et al. 2012). Any modifications to images were done uniformly across the entire image/montage. Brain regions were determined using the Allen Mouse Brain Atlas (Lein, Hawrylycz et al. 2007).

Neuroanatomy mapping (remote)

Students identified the location of a given coronal section using the Allen Brain Atlas as a reference. Brain structures were identified and labeled on sections using the Atlas as a reference.

IHC (in lab)

Brain sections were stained with antibodies to LacZ at 1:500 dilution (Abcam ab9361) in blocking solution (1x PBS, 0.4% triton, 7.5% normal donkey serum). Sections were first blocked in blocking solution overnight and then incubated in primary antibody solution for four days at 4° C. Sections were washed in PBS four times for one hour at 4° C. Sections were incubated for one day in secondary antibodies at 4° C (Jackson ImmunoResearch) diluted 1:1000 in blocking solution. Sections were washed four times for one hour each wash and mounted. Sections were imaged on an Olympus fluorescent microscope.

Results

We systematically mapped LacZ production in the heterozygous and homozygous Dscaml1 gene trap mouse brain. Students working in the lab stained and imaged coronal sections of mouse brain and provided these images to students working remotely. Images were montaged into full brain slices by students working remotely. The Allen Mouse Brain Atlas was used as a reference for identifying where in the brain a given coronal section was located (Lein, Hawrylycz et al. 2007). Students used the Mouse brain atlas to identify and label brain regions on stained brain slices (Fig. 3A and Fig. 3C). Compared to the Allen Mouse Brain Atlas in situ hybridization data, x-gal localization data adds information on the axon tracts of Dscaml1 expressing cells (Fig. 3B and Fig. 3D). Students used the Allen brain atlas to identify brain regions with Dscaml1 expression and placed roman numerals on the structures linked to the name of the brain region.

Figure 3: Annotated and in situ mouse brain comparisons.

A) Individual brain regions were mapped and compared to experimental x-gal staining results in the caudal diencephalon region. B) In situ hybridization images from Allen Mouse Brain Atlas compared to experimental x-gal staining results in the caudal diencephalon region. Arrow = CA3 region with intense x-gal staining when compared to in situ. Arrowheads = CA1, CA2, and dentate gyrus (respectively) when compared to in situ, that shows less intense staining in the lacZ model than in situ. C) Individual brain regions were mapped and compared to experimental x-gal staining results in the septo-striatal region. D) In situ hybridization images from the Allen Mouse Brain Atlas compared to experimental x-gal staining results in the septo-striatal region. E) Table of region identification for (A). F) Table of region identification for (C). In situ images credit: Allen Institute. © 2004 Allen Institute for Brain Science. Allen Mouse Brain Atlas. Available from: https://mouse.brain-map.org/gene/show/77592 (Dscaml1 – RP_060412_04_G05 – coronal). Scale bar = 1mm.

This process was repeated to generate brain images by manually stitching brain sections to generate whole slice montages. Dscaml1 expression was mapped in the whole brain by staining and measuring x-gal staining in 55 and 47 sections of Dscaml1 ^GT/+ and Dscaml1^GT/GT brain, respectively. A total of three brains was used for each map, which was necessary because of inconsistent cutting of sections. Each image was manually annotated with roman numerals linked to the name of a given brain region. A full-size image and annotation for each individual brain slice shown in Figures 4, is presented in a separate excel page for each genotype (See Supplemental Data Set 1). Notable regions of expression were the CA1 and CA3 regions of the hippocampus, dentate gyrus, nerve layers of the olfactory bulb, and the cerebellum (Fig. 4 and Supplementary Data Set 1).

Figure 4: Coronal brain sections of Dscaml1^GT/+ LacZ expression.

Serial images of whole brain coronal sections heterozygous for gene trap insertion, individually montaged, to demonstrate LacZ expression starting rostrally and ending caudally. For annotated reference of each section mapping expression in individual images, refer to the extended data sheets.

Dscaml1 ^GT/GT coronal brain sections had increased saturation due to both Dscaml1 alleles expressing the LacZ gene trap insertion. The gene trap also eliminates expression of full length Dscaml1 and is a null allele (Fuerst, Bruce et al. 2009). Staining appeared more intense in mice homozygous for the gene trap, consistent with mice carrying two copies of the gene trap compared to the heterozygous brain sections. Regions of increased LacZ intensity were the hippocampus regions, dentate gyrus, and the main olfactory bulbs. A full-size image and annotation for each individual brain slice shown in Figures 5, is presented in a separate excel page for each genotype (See Supplemental Data Set 2). Roman numerals on the full-size images are linked to the name of the structure for convenience.

Figure 5: Coronal brain sections of Dscaml1^GT/GT LacZ expression.

Serial images of whole brain coronal sections homozygous for gene trap insertion, individually montaged, to demonstrate LacZ expression starting rostrally and ending caudally. For annotated reference of each section mapping expression in individual images, refer to the extended data sheets. Roman numerals on the full-size images are linked to the name of the structure for convenience.

To demonstrate the utility of the atlas that students generated we focused on identification of Dscaml1 positive neurons specifically within the central optic pathway, including the lateral geniculate nucleus. Diffuse staining was observed in the lateral geniculate nucleus (for example, supplementary data set 1 panel 40 structure xvi). Based on known neuron populations of the geniculate nucleus we tested if Dscaml1 was expressed in inhibitory neurons by staining for LacZ in a strain carrying a GFP reporter expressed in inhibitory neurons. Double labeling of LacZ and a GFP reporter expressed in inhibitory neurons was observed (Figure 6).

Figure 6: Dscaml1 expression in inhibitory neurons of the central optic pathway.

A) Brain section stained with antibodies to LacZ. LacZ immunohistochemistry overlapped with GFP expression in inhibitory neurons (A and B). C and D) higher magnification of the dorsal and ventral lateral geniculate nucleus (dLGN and vLGN) and Olivary pretectal tract (OPT). Arrow heads point to individual double stained neurons.

The project we describe here was one of several remote online research projects that we have found to be successful. Other anatomy project ideas that are ideal for remote students include online segmentation of electron microscopy stacks (Figure 7 A – C) and segmentation of neuron morphology (Figure 7D). In these projects students manually trace or fill in structures such as the dendrite arbors of specific cells in image stacks. Reconstruction of structures within electron micrograph stacks and manual segmentation of neurons is time consuming but provides very valuable data, making it an ideal project for students working remotely and as an opportunity to integrate computer science students who may assist with automating or simplifying tasks as part of the research team. Such data sets are increasingly available to the public and there is a wide range of projects that can be readily devised using existing resources.

Figure 7: Examples of additional research projects developed to include remote and distance learning students.

A) Single electron micrograph with rod terminal colored in. B) Reconstruction of rod terminal volume from multiple images including that in (A). C) Reconstruction of a rod bipolar cell and mitochondria from the same image stack. D) Fluorescent image of a cone bipolar cell and cone terminals (top) and segmentation of the dendritic arbor (bottom).

These projects teach students essential research skills such as working with NIH’s imageJ software and downstream statistical analysis and benefits the research team by essential data sets to studies (de Andrade, Kunzelman et al. 2014, Li, Mitchell et al. 2016, Li, Woodfin et al. 2016, Tropea 2017, Camerino, Engerbretson et al. 2020, Simmons, Camerino et al. 2020). Of the remote students we have worked with 91% graduated within starting college (Table 1). These numbers are favorable compared to other transfer students to our university, who have a 40% six-year graduation rate, and overall university six-year graduation rate of 59%, but is similar to the six-year graduation rate of other undergraduate students performing mentored research in biology and biological engineering departments (Table 1). We observed a high rate of matriculation into graduate programs (Ph.D., M.D., M.S., etc., 57%) compared to other undergraduate researchers (34%) and a high rate of undergraduate authors (57% vs. 30%).

Table 1:

Graduation Rates and Outcomes for Remote Research Students and Comparator Groups.

Student Category	On Campus %	Project Remote %	Overall Transfer to UI	Overall UI
6-year graduation rate from UI	97%	91%	40%	59%
B.S. is terminal degree	50%	28%	NA
Pursued advanced degree beyond BS (e.g., PhD, MD)	34%	57%
Author on Publication	30%	57%

We also surveyed students who worked in our research team. Both students working remotely and in person report positive experiences including the following:

• Remote: “It was something that allowed me to gain experience that I otherwise would not have access to. This allowed me to break into other research opportunities as well.”

• Remote: “My remote research experience as an undergraduate student was a valuable stepping-stone to an opportunity that changed the direction of my education. It got me hooked on the collaborative and investigative process that motivated me to pursue a research career. Since being in the lab, I have managed teams of undergraduate students across multiple institutions throughout my research education. This provided me with essential skills to be able to run my own lab one day, where I will continue to offer research opportunities like I had to remote students that want to be part of a research team.”

• Remote: “I consider myself fortunate to have remotely participated in research with Dr. Fuerst’s lab while I was attending community college, where research opportunities were limited. I worked remotely with two other students at the same location. This provided opportunities for in-person collaboration and problem-solving, along with the guidance from weekly lab meetings with Dr. Fuerst and lab members working in the lab at the University of Idaho. This experience paved the way for a seamless transition to working in the lab on site when I transferred to the University of Idaho.”

• Remote: “Remote research was valuable for me because it opened my eyes to a career that I didn’t know I would be interested in – that being MD-PhD.”

• Remote: “Being able to be a part of research remotely allowed me to work on a meaningful project that has vast impacts while living in a more remote rural community that typically wouldn’t afford me those opportunities. As a student, being able to do on-campus research remotely gave me a chance to be on a great team at an excellent institution with access to resources without having to sacrifice time with my family.”

• In person: “It was helpful to be able to collaborate with the NIC students on my research project even though they were far away because they provided a different perspective and unique ideas for how I could solve problems that came up in my project or how I could make adjustments to my posters/presentations. It was also beneficial to all of us that we could directly and clearly see what methods we were each using for our projects through the screen sharing tool on Zoom. The use of Zoom to meet and collaborate with them was also personally beneficial because I already had an understanding of how to use Zoom when the pandemic began and my courses in school became virtual, so I didn’t have to spend that extra time to learn how to use Zoom while trying to navigate my classes at the same time.”

• In person: “I always thought working with remote students was cool because every week someone would present something that wasn’t what we were working on in the lab per se. We got to see a variety of things other staining and imaging, like 3D reconstructions.”

• In person: “It tested our collaboration skills. Since we could not all be together we had to find other ways to collaborate, stay on the same page, and work as a team. We had to make sure our communication skills were excellent to ensure the research kept progressing. It was a challenging but valuable experience. I will take these skills moving forward to help me build and maintain connections ultimately preventing distance from becoming a limiting factor.”

Discussion

Here we present our experience providing robust neuroanatomy research opportunities to undergraduate students engaged in distance and remote learning. We utilized existing data sets and partnered students working in and out of a lab setting. Students working off campus were able to make significant contributions to characterizing expression of Dscaml1 in the brain.

The data set students generated complement existing in situ libraries. Utilization of the LacZ gene trap as a method to detect gene expression throughout the whole cell, allows for the unique detection in fiber tracts to identify specific cell populations and a more graded staining intensity, something that isn’t observe when using in situ hybridization. The gene expression atlas will serve as a resource for mechanistic experiments at understanding the contributions that Dscaml1 makes to neural development.

In the last seven years our research team has incorporated fourteen students learning at a community college 85 miles from the main campus and lab. Engagement in research can promote an easier transfer to main campus or University for students who start at a branch campus or community college and can help to increase enrollment. Students who took part in our program had a high rate of transfer to complete a four-year degree. The six-year graduation rate for students who started research with our team before transfer was 91% (11/12 within six years of starting college) vs. a six-year graduation rate of 59% for the University as a whole and a 40% six-year graduation rate for transfer students. There was likely a high degree of self-selection in terms of highly motivated students seeking research opportunities, as evidenced by the high matriculation into graduate programs, and our graduation rates were similar to a survey of 247 undergraduate research students working in the same or similar departments. Notable differences include a higher rate of student publication rates for our program, possibly reflecting the collaborative nature of work in which remote students contribute to core projects (57% vs. 30%).

Students who received remote research opportunities perceived these to have a positive impacts on their education. They reported learning to collaborate as a team across locations, to lead and develop research projects, troubleshoot ideas, manage data accumulation, organization and timelines, and developed skills both in and out of the lab that can be used in future research and/or career opportunities as examples of positive outcomes. Most of the remote research students we worked with transfered to in-person lab positions with minimal additional training in lab techniques, as they already had understanding of the projects and research goals. To ease this transfer, in-person students assisted in lab on-boarding of the transfer students and this allowed students to further develop a teambuilding learning environment. This system has allowed students to explore and/or solidify their post baccalaureate education and career goals, something that remote students may not have had the opportunity to do otherwise.

When COVID-19 restricted lab access we had a well-established system of developing and supporting distance research projects. Based on these experiences we developed best practices for online and remote research to guide current and future work:

• Pair students working remotely: Students working remotely or off-site benefit from working with a colleague that they are working alongside with. Integrating students across years so that incoming students learn from more experienced students is especially valuable given the lack of research staff and administration to consult with. Continuity of an off-campus or remote cohort greatly reduces onboarding of incoming students.

• Pairing students off and on site: Integrating students working remotely with on campus students and projects helps promote a sense of belonging and provides an additional contact point for remote students.

• On campus visit: Plan an on campus visit for students working off site. Seeing the lab and meeting the rest of the research team will promote cohesion of the research group.

• Holding regular meetings: Including remote students in lab meetings will ensure projects stay on task and provides a forum to troubleshoot problems. Set goals for the coming meeting and review progress of goals set at the previous meeting.

• Have a backup plan: Students may be required to complete a research project as part of their degree. If progress of remote students is contingent on in lab progress their ability to complete a meaningful project may be impaired. Having a ready to analyze or backup data set can help ensure students are able to complete a meaningful project.

• Remote students can make valuable contributions: Having additional support to quantify data sets can be very valuable. Students working remotely often have different expectations than on campus students and in our experience eagerly tackle work intensive projects.

• Be patient: Students working remotely may be under resourced and make inconsistent progress.

Here, we present results of remote research opportunities for undergraduate students who mapped Dscaml1 expression in the mouse brain. The use of a LacZ gene trap insert is a valuable tool for mapping unique gene expression profiles, outside of the most widely used techniques, and provides us with an additional useful resource. Other imaging data sets, including publicly available data sets, can easily be adapted to this learning format to give remote students a research experience and promote retention and degree completion.