Monitoring Fine and Associative Motor Learning in Mice Using the Erasmus Ladder

Summary

This article presents a protocol that allows a non-invasive and automated assessment of fine motor performance, as well as adaptive and associative motor learning upon challenges, using a device called the Erasmus Ladder. Task difficulty can be titrated to detect motor impairment ranging from major to subtle degrees.

Abstract

Behavior is shaped by actions, and actions necessitate motor skills such as strength, coordination, and learning. None of the behaviors essential for sustaining life would be possible without the ability to transition from one position to another. Unfortunately, motor skills can be compromised in a wide array of diseases. Therefore, investigating the mechanisms of motor functions at the cellular, molecular, and circuit levels, as well as understanding the symptoms, causes, and progression of motor disorders, is crucial for developing effective treatments. Mouse models are frequently employed for this purpose.

This article describes a protocol that allows the monitoring of various aspects of motor performance and learning in mice using an automated tool called the Erasmus Ladder. The assay involves two phases: an initial phase where mice are trained to navigate a horizontal ladder built of irregular rungs ("fine motor learning"), and a second phase where an obstacle is presented in the path of the moving animal. The perturbation can be unexpected ("challenged motor learning") or preceded by an auditory tone ("associative motor learning"). The task is easy to conduct and is fully supported by automated software.

This report shows how different readouts from the test, when analyzed with sensitive statistical methods, allow fine monitoring of mouse motor skills using a small cohort of mice. We propose that the method will be highly sensitive to evaluate motor adaptations driven by environmental modifications as well as early-stage subtle motor deficits in mutant mice with compromised motor functions.

Introduction

A variety of tests have been developed to assess motor phenotypes in mice. Each test gives information on a specific aspect of motor behavior¹. For example, the open field test informs on general locomotion and anxiety state; the rotarod and walking beam tests on coordination and balance; footprint analysis is about gait; the treadmill or running wheel on forced or voluntary physical exercise; and the complex wheel is about motor skill learning. To analyze mouse motor phenotypes, investigators must perform these tests sequentially, which involves a lot of time and effort and often several animal cohorts. If there is information at the cellular or circuitry level, the investigator normally opts for a test that monitors a related aspect and follows from there. However, paradigms that discriminate different aspects of motor behavior in an automated way are lacking.

This article describes a protocol to use the Erasmus Ladder^2,3, a system that allows comprehensive assessment of a variety of motor learning features in mice. The main advantages are the reproducibility and sensitivity of the method, along with the ability to titrate motor difficulty and to separate deficits in motor performance from impaired associative motor learning. The main component consists of a horizontal ladder with alternate high (H) and low (L) rungs equipped with touch-sensitive sensors that detect the position of the mouse on the ladder. The ladder is made of 2 x 37 rungs (L, 6 mm; H, 12 mm) spaced 15 mm apart from each other and positioned in a left-right alternating pattern with 30 mm gaps (Figure 1A). Rungs can be moved individually to generate various levels of difficulty, that is, creating an obstacle (raising the high rungs by 18 mm). Coupled with an automated recording system and associating modifications of the rung pattern with sensory stimuli, the Erasmus ladder tests for fine motor learning and adaptation of motor performance in response to environmental challenges (appearance of a higher rung to simulate an obstacle, an unconditioned stimulus [US]) or association with sensory stimuli (a tone, a conditioned stimulus [CS]). Testing involves two distinct phases, each assessing improvement in motor performance over 4 days, during which mice undergo a session of 42 consecutive trials per day. In the initial phase, mice are trained to navigate the ladder to assess "fine" or "skilled" motor learning. The second phase consists of interleaved trials where an obstacle in the form of a higher rung is presented in the path of the moving animal. The perturbation can be unexpected to assess "challenged" motor learning (US-only trials) or announced by an auditory tone to assess "associative" motor learning (Paired trials).

The Erasmus ladder has been developed relatively recently^2,3. It has not been extensively used because setting up and optimizing the protocol required focused effort and was specifically designed to assess cerebellar-dependent associative learning without exploring in detail its potential to reveal other motor deficits. To date, it has been validated for its ability to unveil subtle motor impairments linked to cerebellar dysfunction in mice^3,4,5,6,7,8. For instance, connexin36 (Cx36) knockout mice, where gap junctions are impaired in olivary neurons, display firing deficits due to lack of electrotonic coupling but the motor phenotype had been hard to pinpoint. Testing using the Erasmus ladder suggested that the role of inferior olivary neurons in a cerebellar motor learning task is to encode precise temporal coding of stimuli and facilitate learning-dependent responses to unexpected events^3,4. Fragile X Messenger Ribonucleoprotein 1 (Fmr1) knockout mouse, a model for Fragile-X-Syndrome (FXS), exhibits a well-known cognitive impairment along with milder defects in procedural memory formation. Fmr1 knockouts showed no significant differences in step times, missteps per trial, or motor performance improvement over sessions in the Erasmus Ladder but failed to adjust their walking pattern to the suddenly appearing obstacle compared to their wild-type (WT) littermates, confirming specific procedural and associative memory deficits^3,5. Furthermore, cell-specific mouse mutant lines with defects in cerebellar function, including impaired Purkinje cell output, potentiation, and molecular layer interneuron or granule cell outputs, exhibited problems in motor coordination with altered acquisition of efficient step patterns and in the number of steps taken to cross the ladder⁶. Neonatal brain injury causes cerebellar learning deficits and Purkinje cell dysfunction that could also be detected with the Erasmus Ladder^7,8.

In this video, we present a comprehensive step-by-step guide, which details the setup of the behavioral room, the behavioral test protocol, and subsequent data analysis. This report is crafted to be accessible and user-friendly and is designed specifically to assist newcomers. This protocol provides insight into different phases of motor training and expected motor patterns that mice adopt. Finally, the article proposes a systematic workflow for data analysis using a powerful non-linear regression approach, complete with valuable recommendations and suggestions for adapting and applying the protocol in other research contexts.

Protocol

In the current study, adult (2-3 months old) C57BL/6J mice of both sexes were used. Animals were housed two to five per cage with ad libitum access to food and water in an animal unit under observation and maintained in a temperature-controlled environment on a 12 h dark/light cycle. All procedures were conducted in accordance with the European and Spanish regulations (2010/63/UE; RD 53/2013) and were approved by the Ethical Committee of the Generalitat Valenciana and the animal welfare committee of the Universidad Miguel Hernández.

1. Behavioral room setup

1. Reserve the behavioral testing room every day at the same time and establish the list and order of mice to be used, as well as arrangements for their hosting.
2. Keep the experimental mice outside the testing room so they do not hear the sounds of the air compressor and Erasmus Ladder tones when not being tested.
3. Check that all components of the Erasmus Ladder system are in order and ready to use: the network router, the computer with the software (see Table of Materials), the air compressor, two goal boxes, and the ladder with the rungs properly positioned.
4. Extensively clean the goal boxes, ladder, and rungs with water after each animal and with water and 70% ethanol at the end of each training day.

2. Behavioral test protocol

1. Create an experiment and enter the protocol into the software (Supplementary Figure S1).
   1. Turn on the software.
   2. To create an experiment, choose File | New experiment | New or Set up | Experiment protocol.
      NOTE: The default protocol, used in this study, is named EMC and was designed at Erasmus University Medical Center, Rotterdam.
   3. Give the experiment a name and click OK.
   4. Check that the default EMC protocol selected consists of 4 days of undisturbed sessions (42 undisturbed trials per day) and 4 days of challenge sessions (42 daily mixed trials: undisturbed, CS-only (tone), US-only (obstacle), Paired (obstacle announced by tone) (see Figure 1B). In the right side panel, also check the light cue (3 s maximum duration), air cue (45 s maximum duration), and tailwind (Yes in all the trial types), used to encourage the mouse to cross the ladder, and the tone (250 ms, Yes only in CS-only and Paired trials).
   5. To create a different protocol, choose Set up | Experiment protocol | New | From scratch or Copy from the EMC protocol and simply modify it, editing the table lines related to number of sessions (days of experiment) and number and type of trials per day.
      NOTE: Resting time, cues type and activation, duration, intensity, and interval can also be adapted according to the experimental questions.
   6. To open the session list and name the subjects, choose Setup | Session List.
   7. Click on Add Subjects and Variables.
   8. Enter each specific Mouse Identifier, Birth Date, Sex, Genotype, and relevant categories, following the ordered list of mice.
2. Start the session (Supplementary Figure S2).
   1. Before starting, check that the software is open, then turn on the Ladder.
   2. Check that the air compressor is connected and switched on.
   3. To open the Acquisition window, open the Experiment created.
   4. Choose Acquisition | Open Acquisition.
   5. Place the mouse with the identifier indicated by the software in the starting goal box (right side of the ladder).
   6. Select the mouse identifier to acquire in the first session.
   7. Click on Start Acquisition.
   8. Press the red ladder menu knob 3x. Check that the session starts and automatically controls and records mouse movements until the end of the last trial of the session.
3. End the session.
   1. Check that at the end of the 42^nd trial, the display shows the messages Sending Data and Acquired.
   2. Return the mouse to the home cage.
   3. Clean the ladder and the goal boxes.
   4. Place the next mouse and repeat from step 2.2.6 onwards.
4. Perform the selected type of session every day until the end of the protocol. Repeat steps 2.2 and 2.3 every day according to the selected protocol.
5. Export the data (Supplementary Figure S2).
   1. To visualize the recorded data, choose from the Analysis menu, Trial Statistics, Session statistics, and Group Statistic & Charts.
      NOTE: Data can be downloaded as a spreadsheet with data for individual trials and the means of the same trial types within a session. Sessions can also be filtered by variables chosen for specific analysis.
   2. Click on the Export button at the top-right corner choosing the file format (spreadsheet) and folder location.
   3. Right-click on the automatically generated charts and select Save to File as *.jpg.

3. Data analysis

NOTE: A list of parameters is automatically measured by the Erasmus Ladder based on the instantaneous recording of the activities of the touch-sensitive sensors. For analysis, output parameters selected by the user are organized and processed in the spreadsheets. Along with the software-generated graphs, users can generate graphs using the graphing software of choice to visualize specific changes in different parameters over sessions.

1. Choose specific parameters to analyze basal motivation or anxiety states, sensory responses, motor performance, and fine motor learning over the first 4 days.
   1. Select and plot control parameters, including resting time in the goal box and time to exit the goal box after the resting period in response to light and air cues (Figure 2A).
      NOTE: Resting times or response to cues are relatively constant in WT mice. Other parameters such as frequency of exits are basically negligible in WT mice-the animals rarely leave the rest box without the cues or come back once in the ladder, resulting in frequencies of exit equal to 1 per trial. If an animal goes out before cues are applied, an airflow gets activated forcing the mouse to return to the goal box; this is not counted as a trial by the software.
   2. Select and plot the time on ladder after cues, measured as the time spent crossing the ladder once the mouse leaves the goal box (Figure 2B).
      NOTE: A power non-linear regression is a robust method for evaluating learning. The Pearson or Spearman coefficients (R) will provide a measure of whether the data fitting is good (R values close to one when the animals learn/improve over sessions; R values close to 0 imply that the data are constant and mice do not learn).
   3. Select and plot stepping pattern parameters such as the percentage of trials with missteps as a sensitive learning parameter (Figure 2C).
      1. Define a correct step as a step from a high rung to another high rung (H-H), irrespective of the length of the step. Consider step types that involve a lower rung as missteps.
      2. Divide correct steps and missteps into short and long steps, backsteps, and jumps depending on the length and directionality of the step between the rungs pressed (see Figure 1A).
2. Select and plot specific parameters to evaluate challenged motor learning (US-only trials) and associative learning (paired trials) over the last 4 days.
   1. Select and plot the time on ladder after cues (Figure 3).
   2. Select and plot the percentage of trials with missteps (Figure 4A).
   3. Select and plot the pre and post perturbation step times, defined as a ms precision difference between rung activation just before (control step) and after the obstacle (adapted step) on the same side of the ladder (Figure 4B).
      NOTE: Pre vs post-perturbation step times analysis should be performed to compare data inside each type of session. The parameter measures the ability of the mice to predict and overcome the obstacles during associative learning.
3. Analyze the data with dedicated statistical software (e.g., SigmaPlot). Perform a power non-linear regression analysis of data collected from the same trial type across sessions to describe the learning process more efficiently and use Two-Way Repeated Measures (RM) ANOVA to compare between trial types.

Results

The Erasmus Ladder device, setup, and protocol applied are presented in Figure 1. The protocol consists of four undisturbed and four challenge sessions (42 trials each). Each trial is one run on the ladder between the starting and ending goal boxes. At the beginning of the session, a mouse is placed in one of the starting boxes. After a set time of 15 ± 5 s ("resting" state), the light is turned on (cue 1, for a maximum of 3 s). A light air cue (cue 2, 45 s maximum) is then appl...

Discussion

The Erasmus Ladder presents major advantages for motor phenotype assessment beyond current approaches. Testing is easy to conduct, automated, reproducible, and allows researchers to assess various aspects of motor behavior separately using a single mouse cohort. In the current study, reproducibility allowed the generation of robust data with a small number of WT mice taking advantage of the features of the device, experimental design, and analysis methods. For instance, when compared to traditional beam-walk assays, the ...

Materials

Name	Company	Catalog Number	Comments
C57BL/6J mice (Mus musculus)	Charles Rivers
Erasmus Ladder device	Noldus, Wageningen, Netherlands
Erasmus Ladder 2.0 software	Noldus, Wageningen, Netherlands
Excel software	Microsoft
Sigmaplot software	Systat Software, Inc.