The Mouse Stroke Unit Protocol with Standardized Neurological Scoring for Translational Mouse Stroke Studies

This article presents a protocol for executing the highly translational and independently proven mouse Stroke Unit (mSU) concept in large-stroke mice. It also provides a standardized protocol for executing the focal experimental stroke scale in mice, which evaluates stroke focal deficits detected even long-term.

The filament model of middle artery occlusion (fMCAo) is perhaps the most translational mouse stroke model, allowing for controlled ischemia with intravascular reperfusion/recanalization. However, it lacks alignment with current clinical advances for stroke care (e.g., Stroke Units), usually employs subjective or vague neurological scoring among laboratories, and exhibits high acute-phase mortality. Here, we address these limitations with validated video-guided protocols. We present the mouse Stroke Unit (mSU) protocol with instructional videos and a decision algorithm (Risk Stratification Score), bridging the gap between clinical and mouse stroke modeling. To increase accuracy and sensitivity of stroke neurological scoring, we present for the first time a video-standardized format of the focal Experimental Stroke Scale (fESS) and prove its value up to 6 months post-stroke. Additionally, protocols for mice Ladder-rung test, as well as the known Cylinder test, for unbiased, quantitative assessment of limbs´ motor function are presented. Results highlight mSU's translational efficacy. Focal ESS (fESS) excels over other known scales in detecting focal stroke deficits, capturing recovery, and maintaining sensitivity for up to 6 months post-stroke. Ladder-rung and Cylinder tests objectively quantify and monitor fore- and hind-limb motor deficits, long-term. In summary, integrating mSU, fESS, and motor function tests provides a robust framework for clinically relevant stroke investigations. Our protocols improve the translational value in mouse stroke research.

Large ischemic stroke is one of the major causes of morbidity and disability worldwide¹. Still, mouse stroke research faces a translational block in almost all preclinically tested treatments so far². The multiple reasons for this problem and the efforts to overcome it have been reviewed in detail in previous publication². These challenges arise from the limited integration of clinical advances into relevant animal models² and the weaknesses of behavioral and neurological scoring systems in detecting post-stroke deficits accurately and sensitively².

Recent discoveries³ support that the filament model of middle artery occlusion (fMCAo) in mice has significant translational advantages over other models with open skull injury and trauma^2,4. On top of that, this is the only one that models the reperfusion and mechanical thrombectomy of clinical stroke settings^5,6,7. However, the model faces high subacute mortality between 3-7 days post-stroke, which prohibits long-term studies of large strokes and was considered until recently the model´s inherent and insurmountable artifact. In addition, the detection of post-stroke neurological deficits is relatively difficult and biased in mice due to their small size, especially among non-clinicians or inexperienced researchers^8,9. To address this, different groups have developed various scales to detect deficits. The most used and known scales include the 3-point Bederson scale (BS)¹⁰, the 5-point modified Bederson scale (mBS)¹¹, the 5-point Longa score (LS) (which is similar to mBS)⁶, the modified Neurological Stroke Scale (mNSS)^12,13, the 18-point Garcia scale (GS)⁹ and the more detailed DeSimoni¹⁴ or differently known as "Neuroscore" (NS)^9,15 scale. Unfortunately, some of these scales are either too crude and are restricted only to the few acute-phase days following stroke (BS, mBS and LS)^8,9 or their interpretation is blurred by general deficits (NS).

With this background, we previously developed and validated the mouse Stroke Unit (mSU) support protocol, along with the experimental stroke scale (ESS) for mice¹³. The rationale for mSU was to adapt knowledge from the clinical routine (i.e., human stroke units) into preclinical research of stroke, while ESS critically consolidated previously existing but "insensitive" or redundant stroke scales into a practical, refined, sensitive and time-efficient one. The mSU consists of frequent and adapted monitoring of mouse´s basic clinical parameters with adapted application of animal support to enhance survival¹³.

Indeed, data from our laboratory and others verify the value of both methods. The mSU translates clinical basic supportive measures and advances⁷ from humans to mice, significantly reduces mortality of the fMCAo in mice from 60-70% to 10-15%¹³ thus allowing for studies of larger strokes, and can be effectively applied in independent laboratories since then^16,17,18,19. In addition, the ESS can distinguish between focal (focal component of ESS, fESS) and general (general component of ESS, gESS) post-stroke deficits and symptoms, models the human clinical scoring (differentiation between focal and general signs and symptoms in humans), is linearly related to the stroke lesion size and is long-term sensitive to deficits^13,20. Still, despite the proven value and efficacy of both mSU and ESS, the lack of clear-cut, visualized, standardized instructions for even inexperienced researchers leaves several open questions on mSU application and significant subjectivity on scoring focal deficits on fESS.

As such, our present article aims to provide video-assisted and clear instructions for mSU and fESS protocols. We strongly believe that it will support stroke researchers to increase the translational efficiency of their stroke studies, significantly reduce the loss of animals during the first 3-10 days after stroke, reduce experimentation costs, and eventually reproducibly evaluate neurological deficits for months after stroke¹³. Additionally, the add-on combination of the Ladder-rung test and Cylinder test can easily quantify the focal limb paresis (fore- and hindlimbs) due to fMCAo.

To showcase mSU and fESS efficiency, we provide medium- (14-days) and long-term (6-month) data on mice after fMCAo. Twelve-week-old male C57Bl/6J mice (n= 31) were used and housed under controlled temperature (22 ± 2 °C), with a 12 h light-dark cycle period and access to pelleted food and water ad libitum. Mice were divided into two cohorts followed for 14 days (n=10, cohort 1) and 6 months (n=15, cohort 2) respectively. These mice were subjected to a 60 min cerebral ischemia using the well-described model of fMCAo^13,20,21 under isoflurane anesthesia. Sham-operated animals (n=6, operated as the above two cohorts, but no ischemia was induced) followed for 6 months served as controls. Buprenorphine was used as a pre- and 3-day post-operative analgesic. The Ladder rung and cylinder tests were also performed for the 6-month cohort as a part of the neurological scoring.

All animals were checked daily for humane endpoints during the first 14 days post-stoke, defined as: 1) severe hypothermia (<33 °C) and/or immobility (e.g., score 4 on Test 6 of fESS or "spontaneous activity" test of gESS) that was not improved by "mouse Stroke Unit" treatment (i.e., passive heating and active feeding, see 1.4, 1.6, 1.8) within one hour, 2) signs of pain or anxiety behavior (e.g. score >2 in "anxiety/automatic behavior" test of gESS), even after post-operative analgesia as per the protocol.

In this protocol, post-stroke support of the mice in the form of mSU begins immediately after mouse recovery from the fMCAo operation. This has 3 phases: phase A (0-48 h post-reperfusion), phase B (>48 h and up to the "end of needed active support", usually at day 10-14, depending on individual animal´s phase B), and phase C (day 14 onwards). It has five significant interventions (visits/Risk Stratification Score, feeding, fluids, temperature, and local disinfections; see 1.1. to 1.8. below) tailored to each animal, according to each phase (A, B or C) and its daily actual clinical status assessed by the Risk Stratification Score (RSS), see Supplementary Figure 1 and Table 1 with three typical examples. The required materials and tools for mSU are shown in Figure 1a and described in Table of Materials. We also provide a template for animal monitoring during mSU (see Supplementary File 1).

The experiments reported in this article were conducted following National and European guidelines for the use of experimental animals and were approved by the Greek governmental committees (Athens, License No "843895_06-09-2022").

1. The "mouse Stroke-Unit" (mSU)

1. Assess animal´s clinical status daily ("visits") and calculate its RSS score
   1. Print the Supplementary File 1, preferably before animal´s surgery, for recording data and monitoring.
   2. Assess each animal clinically at standard time points (visits): early in the morning (at around 08:00), midday (between 14:00 to 16:00), and late at night (at around 22:00 - 23:00).
      NOTE: Adapt these standard visits according to the animal´s follow-up phase: check the animal´s status 2x/daily (morning-night) for Phase A, 3x/daily for Phase B, and once daily for Phase C. Adapt further the frequency of visits according to the RSS score (see 1.1.6). Work preferably in a team of >2 persons for workload reduction. The morning and night visits are crucial for successful support, monitoring, and mortality reduction because the nocturnal (active) phase of the mice increases metabolic and energy demands^13,22,23 and is linked to increased nocturnal post-stroke mortality.
   3. Weigh the mouse using a sensitive standard scale during the morning visit.
   4. Measure the temperature with a contactless thermometer placed over its ventral body center during the morning visit. Hold and immobilize the mouse with its back. Take the reading within the first few seconds after immobilization (see Figure 1f).
      NOTE: Do not use a rectal thermometer due to potential intestinal rupture after repeated measurements with concomitant peritonitis.
   5. Assess and record the general clinical condition of the mouse (mobility, wounds, urethra cap, reactivity, fur, hydration status, etc.) in the morning visit for all phases, daily up to day 14.
   6. Calculate the RSS score in the morning visit by summing the points according to the animal´s weight, temperature, and fESS score as described in Supplementary Figure 1. For automated calculation of the RSS in a spreadsheet table, use and adapt the formula found in Table 1. Euthanize the animal with an institutional approved method, if humane endpoints are present (see above).
   7. Adapt both the frequency of visits and intensity of support, daily according to RSS score and animal´s Phase post-stroke.
      NOTE: Phase B is the most critical for mortality. Increase the frequency of visits and intensity of support for RSS ≥3 (for definition and instructions, see Table 1). If the animal continues to show an RSS >3 in Phase C, visit and support it as in Phase B. Use the RSS and general descriptives (such as recovering, stable, low/medium ill, and critically ill) to describe its clinical condition among the support team. After Day 14, reduce the animal visits to only those required by the experimental protocol (e.g. for behavioral/neurological assessments).
2. Preparation of the gel food for nutritional support
   1. Pulverize common food pellets using a blender (Figure 1a).
   2. Add commercial sugar at 5-10% to increase the caloric content of the food.
   3. Add water to create a gel-like liquid texture (Figure 1b).
   4. Use this gel food for both passive and active nutritional support.
      NOTE: Optionally, use ready-to-use commercial gel food. Prepared food can stay in the fridge for 2-3 days.
3. Passive nutritional support of the mice
   1. Fill a Petri dish with gel food and place it on the cage floor (passive nutritional support by provision of gel food for easy access to the mouse). Leave 2-3 food pellets next to the Petri dish.
   2. Refresh the Petri dish with gel food daily. During the first few days, mice may fill or cover the Petri dish with cage bedding. Some will begin eating from it (Figure 1c).
4. Active nutritional support (active feeding)
   NOTE: Active feeding occurs during Phase B. The standard feeding frequency is 3x/daily and is adapted to more times according to RSS score (e.g., even 4-5 times/day). Occasionally, for mice that constantly have an RSS score of 0-1 (i.e., small stroke) and eat actively by themselves downgrade the frequency to 1-2/daily.
   1. Pre-fill 1mL syringes with gel food.
   2. Grasp and hold the mouse gently but firm from its neck fur, as shown in Figure 1d (red arrowhead).
   3. Stabilize the mouse´s cheek with the middle finger, as shown by red arrows in Figure 1d: while holding the mouse firmly, slide the finger sideways towards the mouse´s nose against its cheek. Hold position.
   4. Place the tip of the syringe in mouse´s mouth between its teeth and direct the tip towards the interior side of its cheek (red arrow, Figure 1e).
   5. Give approximately 40-50 µL of gel food each time (a mouthful).
   6. Gently return the mouse to the cage grate. Allow time for chewing.
   7. Repeat feeding after 1-2 min, until each mouse has eaten 1-1.5 mL of the gel food.
      NOTE: Feed 5-10 mice per run (1-2 cages); the others eat and swallow the gel food as each mouse is given a mouthful. The process confers no stress on mice. It has also been observed that some animals eventually position themselves for grasping and feeding.
5. Normal saline supplementation
   1. Weigh the mouse during the morning visit (see 1.1.2 and 1.1.3).
   2. Calculate its weight loss/gain compared to the previous measurement by subtracting the values.
   3. Supplement its weight deficit (in grams) with mL of normal saline (subcutaneously).
      NOTE: Avoid fluid overload and edema at all costs. Do not exceed 1-2 mL per dose and 3-4 mL of fluids per day. Begin after 36 h post-fMCAo.
6. Glucose supplementation
   1. Check the calculated RSS score of the mouse and its overall movement.
   2. Inject 0.5 - 1 mL of 5% glucose subcutaneously if RSS >3 and the mouse shows concomitant signs of hypothermia (<34°C) or immobilization/bradykinesia. Inject preferably only during the night visit.
      ​NOTE: Avoid hyperglycemia after stroke at all costs. Post-stroke hyperglycemia is toxic⁷. As such, supplement glucose only after active feeding and in cases of severe clinical condition or to prevent hypoglycemia during night activity. Active feeding (see 1.4) should be the main support measurement, as it provides smoother and longer-lasting energy support and translates clinical guidelines ⁷. Do not supplement glucose before Day 3.
7. Temperature management (passive)
   1. Measure the animal´s surface body temperature pre-operatively at baseline and post-operatively at every visit (Figure 1a). Measure as shown in Figure 1f and described in 1.1.4. Note down the values.
      NOTE: Hypothermia in mice is defined as body temperature <35 °C ^24,25. Previous data¹³ indicate that any temperature below 34 °C is a warning sign for support after stroke (see below).
   2. When hypothermia between 33 and 34 °C is observed, warm the mouse passively in a cage on a heating plate (or any other passive-warming device) until it reaches a body temperature >34 °C. Actively feed the mouse to provide energy for active heat production by the animal (longer-lasting supply of energy).
      NOTE: Keep the animal on passive heating for 1 to a few hours. Avoid passive hyperthermia at all costs.
   3. When hypothermia is observed, <33 °C, escalate support with additional glucose administration subcutaneously (fast but short-lasting supply of energy; see 1.6 above).
8. Cleaning of genitalia and/or neck wound (local "disinfections")
   NOTE: Occasionally, a wound dehiscence or urethral plug can be observed. Wound dehiscence is rare. Urethra plug is a hard plug at the tip of the mouse genitals (Figure 1g) and is observed more common, especially in animals with higher stroke severity. When plug blocks the urethra, a urinary obstructive syndrome may occur and can lead to morbidity and death²⁶. Therefore, it should be cleaned well.
   1. Soak the urethral plug, if present, with normal saline to facilitate its removal.
   2. Remove the urethral plug gently via a small swab or a simple tweezer at each animal check/visit. Disinfect the urethra using a gauge or swab, soaked with an antiseptic solution.
   3. If a wound dehiscence is present, disinfect it locally with any common surgical antiseptic solution (e.g., Octanisept) and routinely suture the wound using preferably a 5/0 silk suture.

2. The focal component of Experimental Stroke Scale (fESS)

1. Prepare the testing field
   1. Clean a bench with 70% alcohol to remove external stimuli (optical or odors) and prepare the testing materials, as shown in Figure 2a.
   2. Print the ESS scoring sheet provided as a Supplementary file 2 and follow the tests one by one.
2. Forelimb paresis (Test 1)
   1. Raise and hold the mouse gently from its tail-base.
   2. Check and assess visually for symmetric forward extension and symmetric mobility of the forelimbs as the mouse moves (Figure 2c).
   3. Score depending on the visual findings as follows: 0 for normal, symmetric extension and movement of the forelimbs; 1 for "slight asymmetry" defined as mild flexion and/or reduced movement of the contralesional limb or limb-stiffness; 2 for "marked flexion" of the contralesional limb; 3 for adherence of the contralesional limb to the trunk; score 4 for no body or limb movement.
      NOTE: The test detects paresis of the contralesional forelimb. It is also part of BS, LS, mNSS and NS scales. Score 1 may be challenging to define due to the animal´s escape maneuvers.
3. Hindlimb paresis (Test 2)
   1. Continue holding the mouse by its tail base, as described Test 1.
   2. Check visually for symmetric extension and movement of the hindlimbs and toes (Figure 2c).
   3. Score depending on the findings as follows: 0 for normal, symmetric position/extension and movement of hindlimbs and toes; 1 for clear asymmetry, when significantly decreased movement and asymmetric position/extension of the contralateral (right) hindlimb (Figure 2d).
      NOTE: The test detects paresis of the contralesional hindlimb. It is also part of mNSS scale. This test may be tricky to score accurately.
4. Trunk symmetry (Test 3)
   1. Continue holding the mouse by its tail base (following 2.3.) and assess its trunk movements.
   2. Score depending on the findings as follows: 0 for normal, no trunk flexion or symmetrical mild trunk flexion/movement to both sides (as an "escape" maneuver); 1 for marked (>30°) head/trunk flexion predominantly contralesional (to the right), within first 10s from holding.
      NOTE: The test detects asymmetry by unilateral motor lesions. It is also part of mNSS scoring.
5. Circling behavior (Test 4)
   1. Place the mouse gently on a flat surface and leave it to move freely for at least 1 min. Observe its movement pattern.
   2. Score depending on the findings as follows: 0 for exploratory, vivid, linear movement without circling behavior; 1 for the tendency to turn to one side but not in closed circles; 2 for inconstant movement in closed circles to one-side; 3 for constant movement in closed circles to one side; 4 for pivoting, swaying or no movement at all.
      NOTE: The test detects sensorimotor and spatial navigation deficits. It is also part of NS.
6. Body symmetry (Test 5)
   1. Continue from Test 4 and let the mouse move freely on the flat surface. Observe its nose-to-tail axis.
   2. Score depending on the findings as follows: 0 for normal posture with elevated trunk from the bench, and straight tail; 1 for "slight asymmetry" if slight body bending to one side and distally bent tail; 2 for "moderate asymmetry" if obvious body bending or leaning to the contralesional side with proximal bent tail and/or limbs stretch out of the body; 3 for "prominent asymmetry" if marked body bending, one side inconstantly lies on the bench and the tail is clearly proximally bent (more than score 2); 4 for "extreme asymmetry" if the body is bent, one side constantly lies on the bench and the tail is highly bent.
      NOTE: It detects sensorimotor unilateral truncal and tail deficits. It is also part of NS scale.
7. Gait assessment (Test 6)
   1. Continue from Test 5, and let the mouse move freely on the flat surface. Observe now its gait and check for gait motor deficits such as paresis, limb stiffness, spastic gait, limbing.
   2. Score depending on the findings as follows: 0 for normal gait if exploratory, symmetric and quick with flexible limbs that move "hidden" under the body; 1 for stiff, inflexible and slower gait of a freely moving mouse without limping; 2 for clear limping with asymmetric (left-right) movements; 3 for trembling, drifting or falling after gentle push; 4 for no spontaneous walk (walks no longer than 3 steps when stimulated by gentle push).
      NOTE: It detects combined gait deficits. It is also part of NS.
8. Compulsory circling (Test 7)
   1. Hold the mouse by its tail base to elevate it from the surface and let it walk only on its forelimbs. Observe its movement.
   2. Score depending on the findings as follows: 0 for normal linear movement; 1 for tendency to turn to one side; 2 for turning to one side in closed circles; 3 if pivoting to one side with clumsy movements or sluggishly fails to complete a circle; 4 if no advancement and the front part of the trunk lies on the bench.
      NOTE: The test complements the circling behavior test (test 4, see 2.5). It probably detects sensorimotor and spatial deficits. It is also part of NS scoring system.
9. Posture against lateral forces (Test 8)
   1. Place the mouse on the palm, with its tail nose axis (longitudinal) parallel to the sagittal palm axis.
   2. Swing the mouse forward to backward at a steady speed to induce forces perpendicular to its longitudinal axis.
      NOTE: This forces the mouse to retain its upright position and balance against the lateral forces during the swing. Its body normally does not contact the palm or fingers.
   3. Observe visual and feel any palm-contacts of the left-right sides of the mouse as it balances against the swing. Check for any asymmetric fall, touch on the palm or indication of lateral weakness.
   4. Score depending on the findings as follows: 0 if the mouse stands in a normal upright position with its back parallel to the palm and can balance against swing without heavily touching the palm; 1 if the mouse flattens its body during the swing to gain stability; 2 if a part of the mouse lies or touches briefly on the palm during swing; 3 if the mouse lies on one side, barely able to recover the upright position; 4 if the mouse lies on a prone position, not able to recover the upright position.
      NOTE: The test detects postural deficits against lateral acceleration, but its scoring may be subjective. It is also part of mNSS, NS.
10. Forelimb targeted placing (Test 9)
    1. Hold the mouse from its back or from its tail-base. Direct it perpendicular towards the beam, which functions as a "target to reach". Observe for symmetric "reach-out" and grasping.
    2. Once it grasps the beam steadily, pull it gently away and observe for symmetric "release" of the forelimbs.
    3. Score depending on the findings as follows: 0 for symmetric "reach-out", grasping and "release"; 1 for any asymmetry in "reach out" and/or "release", indicative of contralesional forelimb weakness.
11. Sensorimotor deficit of forelimbs (Test 10)
    1. Hold the mouse firmly by its back and face it ventrally (see video or Figure 1e).
    2. Stimulate the forepaws successively with a fine tip (e.g. wooden toothpick) from a blind angle. Observe the forepaw´s reaction.
    3. Score depending on the findings as follows: 0 for normal, prompt grasping movement both sides; 1 for absent or clearly delayed grasping contralesionally.
       NOTE: Both 2.10 and 2.11 test for forelimb fine sensorimotor deficits. They are also part of mNSS.
12. Beam walking (Test 11)
    1. Place a wooden beam of 1 m length (Table of Materials) to connect two elevated surfaces (e.g. benches, tables) at a height of 1-1.5 m above ground. Stabilize beam´s edges to avoid vibrations or instability.
    2. Place the mouse on the middle of the beam with its nose-tail axis parallel to the beam (Figure 2e). Allow for balance or free walk on the beam. Observe its balance or limb paresis and falls-off.
       NOTE: The mouse should gain balance on the beam before scoring it. Hold a thick glove below the animal (at a distance of approximately 15-20cm) to protect it if falling off.
    3. Score depending on the findings as follows: 0 if mouse balances and walks with steady posture towards one side of the beam; 1 if it grasps the side of the beam and does not walk on it; 2 if one limb falls off the beam; 3 if two limbs falls off the beam; 4 if the mouse attempts to balance on the beam but falls off (after 40s); 5 if it falls off between 20-40s; score 6 if it falls with no attempt to balance or hang on the beam (<20s).
       NOTE: The test detects combined motor deficits of forelimbs, hindlimbs and trunk. It is also part of mNSS scale.
13. Whisker´s sensation (Test 12)
    1. Place the mouse on an open bench top (as in tests 4 and 5). Allow some time for accommodation.
    2. Stimulate its left and right whiskers successively, with a thin, long-tipped, cotton swab, from a "blind angle". Observe for a prompt turning response of the head towards the stimulation.
    3. Score depending on the findings as follows: 0 for normal, fast head-turn upon stimulation, bilaterally; 1 for slower contralesional response upon stimulation; 2 for absent contralesional response (no turn); 3 for absent contralesional and slower ipsilesional response; 4 for bilateral absent response.
       NOTE: Avoid visual detection of the cotton because this would falsely cause an ipsilateral head turn. The test is technically correct only when the mouse is "surprised" by the whisker stimulation. The test is also part of NS and detects sensorimotor deficits of whiskers (barrel cortex).
14. Climbing (Test 13)
    1. Use a rubber-coated, 45° angled, sloped surface (Figure 2a) and place the mouse at its center. Observe the mouse´s effort to climb.
    2. Score depending on the findings as follows: 0 (normal) if the mouse quickly and effortless climbs to the top of the surface; 1 if it climbs with strain, with limb weakness; 2 if it holds onto the slope without slipping or climbing; 3 if it slides down the slope despite effort to prevent fall; 4 if it slides with no effort to prevent fall.
       NOTE: The test is also part of NS and detects general motor strength.
15. Calculate the score of fESS by summing all scores from tests 1 to 13
    NOTE: fESS score of 0 indicates a normal mouse, the maximal score is 42. Large infarcts with striatal and cortical lesions usually induce scores >7-9. Examine and score several normal mice before starting evaluating stroked ones, to accommodate with variations of normal behavior and reduce false positive fESS scoring. Use the provided video protocol constantly for standardization of scoring through the entire experiment.

3. General component of ESS (gESS)

NOTE: Assess the gESS to detect signs of "general mouse illness", e.g., inflammation or infection, usually present during the first 1-2 weeks post-fMCAo. Use the scoring sheet in Supplementary file 2. Assess gESS preferably after completing the fESS.

1. Place the mouse on a flat surface (bench) and leave it to move freely as in 2.5. Assess sequentially the following scores.
2. Observe its fur. Score according to gESS scoring sheet (scores 0-2, see Supplementary file 2).
3. Check the mouse´s startle response: surprise the mouse with a sudden clapping from a blind angle, when it is relaxed. Check and score the reaction of mouse´s ears according to gESS scoring sheet (scores 0-2, see Supplementary file 2).
4. Assess visually the eyes of the relaxed mouse: check for open/closed eyes or mucus presence. Score according to gESS scoring sheet (scores 0-4, see Supplementary file 2).
5. Observe the mouse´s general spontaneous activity and alertness. Score according to gESS scoring sheet (scores 0-4, see Supplementary file 2).
6. Observe the mouse´s general spontaneous behavior for any presence of anxiety, hyperexcitability, aggression, or spinning along its axis during any handling. Score according to gESS scoring sheet (scores 0-4, see Supplementary file 2).
7. Calculate the score of gESS by adding the scores of all above steps (observational scoring, steps 3.2. to 3.6).
   NOTE: The gESS score does not detect the focal deficits induced by stroke lesions nor is affected by mSU, as previously shown ¹³. All tests are adapted from the NS scale ^9,15.

4. Quantification of forelimb deficits with the Cylinder Test

NOTE: Cylinder test is previously described in detail^9,15,27 and will not be described in detail here. The cylinder test is an independent behavioral test that unbiased quantifies deficits of the forelimbs (see 4.2).

1. Run the test following the instructions provided here¹⁵. Capture videos and analyze them as in ¹⁵. Measure right and left first limb-contacts to the walls. Figure 3a,b shows images of the setup.
2. Calculate the lateralization of limb usage due to contralesional (right) paresis as a % percentage. Use the formula: % preference for left steps = [(left - right) first contacts / (right + left) contacts] x 100.
   NOTE: Run a baseline cylinder test for each mouse before inducing a stroke. Run tests at timepoints according to each experimental plan.

5. Quantification of fore- and hindlimb deficits with the Ladder-rung Test

NOTE: Construct the apparatus for mice out of Plexiglas (shown in Figure 3c): a 1 m long corridor with 20 cm high walls, plastic rungs (3mm diameter) placed at regular intervals of 15 mm (Figure 3c1) and a "refuge" box at the distal end. Place animal bedding in the refuge box for familiarity and some food pellets or peanut butter as a reward. The width of the corridor should be 10 cm and should be elevated approximately 30cm from ground.

1. Train the mice on the apparatus 1-3 days before the baseline of the experiment. Run the training as in steps 5.1.1 to 5.1.4.
   NOTE: Training is necessary in order to accommodate animals to a steady-speed, normal walking through the rung corridor, without stops, turnings or explorative behavior. Perform training for 2 consecutive days (2 sessions). Each training session consists of 3 runs (run = a single walk through the entire corridor towards the refuge box). Collection of baseline data is necessary for follow-up.
   1. Clean the rungs and walls before each run with 70% alcohol, to avoid odors and "exploratory" distractions/stops of the mouse.
   2. Position the mouse at the beginning of the ladder-rung corridor, facing the refuge box, to start a run. Let the mouse walk freely towards the refuge box at the end of the corridor. Do not force the animal to walk.
      NOTE: Runs during training are normally highly variable in terms of duration and walking pattern. The goal of the training is a stable walk towards the refuge box, usually within <10 s, with no stops/turns, which is achieved usually by the end of second training day. Handle the animal always gently to avoid stress induction which will distract proper training. Remove animals from the corridor and redo the run if they stay in the corridor and explore locally for >1 min, to avoid habituation.
   3. Perform the 1^st training run. Let the animal rest in the refuge box for 5 min after the 1^st run.
      NOTE: This prolonged resting time allows the mouse to familiarize itself with the refuge box and perceive it as a "resting/rewarding" target to reach. Achieving a "rewarding perception" of the refuge box is crucial for successful runs.
   4. Remove the mouse from the refuge box and perform the second and third training runs as in 5.1.2. Allow shorter resting time in the refuge box at the end of these runs (30-60 s), to avoid habituation.
      NOTE: Training is completed on day 2. By day 2 almost all animals learn to walk through the corridor at a steady pace. Consider excluding animals from further experimentation if they are not learning.
2. Run the experiment on trained mice. Test mice for 2 runs each time during all time points of the experiment and baseline.
   1. Perform runs as described in steps 5.1.1 and 5.1.2. Allow a maximum of 5 min of resting time between the 2 runs in the refuge box.
   2. Use a smartphone (or equivalent camera) and a "selfie-stick" to capture videos (data) of the animals at a slight ventral and "blind" angle from some distance (as shown in Figure 3c2). Make sure that you capture the limbs as the animal walks in the corridor. Do not visually distract the mouse during its run and follow its limbs during its run (Figure 3c2).
   3. Record both runs but use and analyze data from the second run.
   4. Measure the number of fault steps (falls) between the rungs for each limb (left/right, fore-/hindlimb) (Figure 3c3), by reviewing the video using any media-player software with slow-motion options. Calculate the absolute or % difference between the right and left limb fault steps, separately for fore- and hindlimbs.
      NOTE: The second run gives more reliable results. A healthy mouse usually has 0-1 arbitrary fault steps per limb (right or left) during its second run, which normally gives a difference range of 0 ±1. A stroked mouse will make fault steps with its right (contralesional) paretic limb (fore- or hindlimb), resulting in a positive "rightward" lateralization. A severely stroked animal may be slower or not complete the run during the first 7-10 days after stroke, however, it will achieve proper runs afterwards and show severe deficits.

The mSU, as described above, begins after the fMCAo operation. Our representative results in the two independent mouse cohorts of fMCAo stroke and sham (Figure 4a) confirms the previously proven value of mSU, particularly during the critical period between days 3 and 10¹³. In our cohorts, mortality occurred for 3/15 animals of Cohort 1 (stroked, 6 months follow-up), 2/10 of Cohort 2 (stroked, 14 days follow-up), and 0/6 of sham-operated animals (Figure 4a) as part of stroke pathophysiology (all found dead during the morning visits), and was included in the scoring. No animals were euthanized per predefined humane endpoints. This acute mortality (10-15%) within phase A (initial 24-48 h post-fMCAo) (Figure 4a) was attributed to hemispheric edema and herniation resulting from large territorial strokes¹³, and was expected as part of translational stroke modeling. The delayed mortality during phase B (1/15 for cohort 1, 1/10 for cohort 2) added a maximum of 5-10% under the mSU protocol (Figure 4a) and usually occurs for animals with very large strokes that cannot be rescued despite mSU support. This also directly translates the severity-related human post-stroke mortality²⁹. No application of mSU or low adherence to its principles results in increased Phase B mortality that can reach even 60-90%¹³. Phase C is normally devoid of further mortality.

Upon application of the mSU, mice with large strokes survive beyond the critical phase B and show significant neurological deficits. Our data showcase that the standardized use of fESS detects and quantifies both fMCAo-induced deficits and spontaneous improvement in mice (Figure 4b), and outperforms or matches previous scales like the focal Neuroscore (NS), the mNSS and the Bederson (BE) scales^9,10,12,14. While not directly statistically comparable, the fESS shows similar or higher sensitivity in the acute phase, captures spontaneous neurological improvement after fMCAo, and remains deficit-sensitive up to 6 months post stroke, compared to the other scales. This supports its use for translational and sensitive scoring after mouse stroke and the present video-protocol ensures increased scoring objectivity.

In addition to the fESS, the Ladder-rung and Cylinder tests can complement long-term neurological scoring and monitoring by quantifying fore- and hindlimb paresis. Each test requires approximately 5-7 min per animal for video acquisition and 5-15 min for manual video analysis. Representative results are increased limb drops between the rungs as the mouse walks in the Ladder-rung test (resulting in right % lateralization for left stroke), and a "preferred" wall-touching with the healthy left forelimb for the Cylinder test (resulting in left % lateralization for left stroke). Our representative data for the 6-months´ cohort show that the Ladder-rung test can detect forelimb paresis in the acute/subacute phases but loses its sensitivity due to spontaneous improvement thereafter (Figure 4c, p<0.05 for time and p<0.01 for group difference, mixed-effects model), however remains more robust in detecting and quantifying hindlimb paresis for up to 6 months (Figure 4d, p<0.05 for group difference, mixed-effects model). In parallel, the Cylinder-test retains an excellent sensitivity in detecting forelimb paresis up to 6 months post-fMCAo (Figure 4e, p<0.001 for group difference, mixed-effects model). Combined, both tests effectively detect, quantify, and monitor long-term fore- and hind-limb paresis.

Figure 1. Post-fMCAo mSU support protocol. (A) Tools and materials used for mSU. (B) pulverizing of normal food (with food blender or other device) and making the gel-food for active (with syringe) and passive (petri dish) feeding support. (C) gel-food and solid pellets should be placed (and replaced freshly daily) in the cage of the animals few days prior (for accommodation) and after fMCAo; it is expected that the animals will hide the gel-food under bedding. (D) active grasping for syringe feeding (red arrows point at left cheek stabilization with fingers, arrowhead points at fur-grasping). (E) insertion of syringe in mouth, pointing at the right cheek (arrow). (F) Showcase of body surface temperature measurement (red area and arrow point at the center of ventral body for temperature measurement). (G) urethra plug (arrow). (H) Suggested timeline for chronic stroke experiments under mSU support (bsl: baseline, h: hours, d: days and m: months as evaluation timepoints), with suggested scoring using ESS, Ladder-rung and Cylinder tests. This timeline has been used here for cohorts 1 (6m follow-up) and 2 (14d follow-up). Please click here to view a larger version of this figure.

Figure 2. The experimental stroke scale. (A) Tools for ESS testing, (B) representative brain slices with stroke (left half, Nissl stain) and their corresponding overlap of the corresponding Allen Brain map, at bregma +1.0 and -0.3 mm anteroposteriorly (This figure has been modified from Allen Mouse Brain Atlas, mouse.brain-map.org and atlas.brain-map.org)³⁰. (C) Animal suspension by its tail-base for forelimb, hindlimb and trunk symmetry tests of fESS, grey dotted circles show normal symmetry of fore- and hindlimbs. (D) Clear asymmetry of the paretic right hindlimb (asymmetric position/extension). (E) normal position of an animal on the beam. Please click here to view a larger version of this figure.

Figure 3. Setup of Ladder-rung and Cylinder tests. (A) setup of Cylinder test, a mouse is in the cylinder while vertical mirrors facilitate the 360° view of its forelimbs. (B) close view of the animal with a left forelimb contact (usage) to the cylinder wall. (C) view and design details of the Ladder-rung test for mice (see also text for dimensions and construction instructions); (C1) shows a normal mouse in the Ladder-rung corridor from above, (C2) shows a normal mouse with correct stepping on the rungs (no falls), (C3) shows a forelimb fall between the steps (images are captured from videos). Please click here to view a larger version of this figure.

Figure 4. Long-term survival and neurological deficits after fMCAo. (A) Replication of the mSU effectiveness in two independent animal cohorts: in the first one the mice were observed for 6 months and in the second animals were observed for 14 days following stroke. (B) Scores of different, well-established, neurological scales for both cohorts. (C) Forelimb and (D) hindlimb motor deficits of the right side (= right - left false steps, for fore- and hind-limbs respectively) detected by Ladder-rung test in our 6-months animal cohort, vs sham operated animals. (E) Forelimb asymmetry results (as % preference for usage of the healthy left forelimb) detected by Cylinder-test in our 6-months animal cohort, vs sham operated animals. Please click here to view a larger version of this figure.

Table 1. Software interface: Example of 3 animals with their daily calculation of RSS scores. Actions and RSS apply for Phase B and C. Please click here to download this Table.

Supplementary Figure S1. Definition of the daily Risk Stratification Score (RSS) and the resulting support actions for each animal. Animals with scores 0-1 have a minimal mortality risk while those with 5-6 a maximum one. Please click here to download this File.

Supplementary File 1. Proposed template for animal monitoring and documentation during the fMCAo operation and the mSU follow-up. Please click here to download this File.

Supplementary File 2. Proposed scoring sheet for the focal and general ESS. Please click here to download this File.

The present protocol is a comprehensive guide to the mSU support protocol, designed to reduce artifactual mortality between days 3-10 in mice subjected to large territorial strokes by the fMCAo model¹³. Additionally, the present protocol includes a standardized video guide for the focal component of the ESS scale (fESS) to reduce the extended subjectivity and bias of neurological focal scoring in mice after stroke. Alongside, we introduce the Ladder-rung test for quantifying long-term (up to 6 months) contralesional fore- and hind-limb motor deficits.

The mSU is user-friendly and effective. Our video aided protocol makes mSU clear for application even for non-experienced researchers. The 3 phases of mSU (namely phase A, B and C, see section 1 of protocol above and Figure 1h) are defined empirically for tailoring its assessment and supporting measures. The mSU tailors its interventions daily according to each phase and depending on each mouse´s Risk Stratification Score (RSS) score. This score reflects its clinical severity, gives an estimation for its actual risk for mortality, and guides its tailored support per mouse. Complementary to that, the mere clinical observation of the animal should always add information for tailored mSU support. Eventually, mSU translationally implements key components of human Stroke Unit (SU) support, including temperature control, fluid balance, infection prevention/treatment, nutritional support, and normoglycemia^1,2,7, during the critical phase of the first 3-10 post-fMCAo days on mice¹³. Importantly, the effectiveness of mSU to reduce the artifactual mortality from 60-90% down to <15%^13,31 has been successfully reproduced in independent research groups^17,18,32,33.

Several critical steps are essential for the success of the mSU protocol. First, while mSU relies on close clinical observation and optimal animal support, it is advisable to have a team of at least two researchers working in shifts to reduce workload, although one researcher can also manage it alone. Second, the most critical timepoints for support are around 22:00 and 08:00, corresponding to the beginning and end of the nocturnal increase mouse activity, when energy demands are higher^13,22,23. Third, in line with current clinical stroke guidelines⁷, glucose supplementation should be carefully tailored and minimal to avoid neurotoxic post-stroke hyperglycemia. Fourth, active feeding should be carried out in blocks of 5-10 animals administering mouthfuls of 40-60µl in steps, to balance reduced workload with effective feeding. Finally, the intensity of mSU has to be tailored to each mouse´s needs and RSS score, based on stroke severity. This practical means that mice with small strokes may need shorter and less intensive support, while those with larger strokes may need support even beyond day 14 to mitigate risk of death.

The neurological scoring in mice has always been highly subjective and challenging due to their small size and difficulty to detect deficits. Due to these, all previous stroke scales (BS, mBS, LS, mNSS, GS, NS, see also introduction) either detect crude signs (e.g., the 3- to 5-point scales of BS, mBS and LS), mix focal with general post-stroke symptoms (e.g., NS), fail to capture subtle deficits beyond the acute phase of stroke ^8,9 (e.g., the 3-point BS ¹⁰, the 5-point mBS ¹¹ or the 5-point LS⁶), or fail to quantify long-term spontaneous improvement after stroke. To improve all these, we previously¹³ developed the ESS (fESS/gESS) by critically combining or omitting components from previous scales, creating a tool capable of assessing commonly affected areas by the fMCAo model²⁰ (Figure 2b). Now, we additionally provide the first video-standardization of fESS to offer visual guidance and resolve the long-standing limitation of subjectivity across laboratories and researchers. Our data support that the fESS outperforms the mBS and mNSS scales or equals the NS (Figure 4b) in detecting focal, stroke-related, deficits in mice while also capturing the ongoing long-term post-stroke spontaneous recovery. The provided video-standardization serves now as a reliable training tool for consistent evaluation of post-stroke deficits.

Supplementary to the fESS, we recommend using the Ladder-rung test and the previously described Cylinder test¹⁵ as a battery of tests to quantify forelimb and hindlimb paresis or corresponding improvements over the long-term (up to 6-months). For intra-animal limb analysis, baseline evaluations for both tests are necessary. The Ladder-rung test, previously used in rats ^34,35, has been adapted and described here for mice. Practically, the test critically depends on a stable walking pattern in the rung-corridor, with no turnings or stops. For this, we recommend analyzing the second run in each time-point, as it typically yields the most stable walking patten. Proper training without habituation is crucial for reliable results, as outlined in the protocol above. A limitation of both Ladder-rung and Cylinder tests is that mice with large strokes may not move at all during phase A-B (days 3 and 7 in Figure 4c-e), thus providing no data and increasing variance; these mice often show maximum fault steps during phase C. To overcome this limitation and minimize stress of the animals we suggest testing mice from the end of Phase B onwards (e.g., > day 10). Another limitation is the fact that Ladder-rung test seems to lose its sensitivity for forelimb deficits´, but this is compensated by the Cylinder test. Eventually, these tests combined can objectively quantify both fore- and hindlimb deficits and their long-term spontaneous improvement.

In conclusion, we recommend the mSU as standard of care for translational mouse stroke models. At the same time, we recommend the accompanied ESS (fESS/gESS), Ladder-rung test and Cylinder test as an simple, time-efficient, cost-effective, quantitatively sensitive battery of tests to quantify long-term stroke deficits in mice. Eventually, mSU could be applied in the future in any other mouse model incorporating severe brain lesions (e.g. traumatic brain injury, models of brain hemorrhages, etc), where intense, clinically translational, mouse support is needed.

No disclosures to report.

We would like to thank Nikolaos Plakopitis and Ioannis Tatsidis for valuable surgical support during parts of the study.