Protocol to improve gradient-index lens implantation in the mouse medial prefrontal cortex using NeuralGlider-assisted insertion

Summary

Gradient-index (GRIN) lens implantation enables deep brain imaging in animal models. Tissue-aspiration based GRIN lens implantation procedures often suffer from low success rates. Here, we present an improved protocol for NeuralGlider-assisted GRIN lens implantation in the mouse medial prefrontal cortex. We describe steps from viral injection to in vivo calcium imaging, with primary focus on GRIN lens implantation. This NeuralGlider-assisted insertion protocol minimized tissue damage through a controlled insertion speed, gentle lens rotation, axial micro-oscillation, and vacuum-based lens stabilization.

Graphical abstract

Highlights

• •Steps from viral injection to in vivo calcium imaging
• •Instructions on NeuralGlider inserter
• •Procedures for NeuralGlider-assisted GRIN lens implantation
• •Guidance on determining in vivo imaging quality

Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

Gradient-index (GRIN) lens implantation enables deep brain imaging in animal models. Tissue-aspiration based GRIN lens implantation procedures often suffer from low success rates. Here, we present an improved protocol for NeuralGlider-assisted GRIN lens implantation in the mouse medial prefrontal cortex. We describe steps from viral injection to in vivo calcium imaging, with primary focus on GRIN lens implantation. This NeuralGlider-assisted insertion protocol minimized tissue damage through a controlled insertion speed, gentle lens rotation, axial micro-oscillation, and vacuum-based lens stabilization.

Before you begin

This protocol describes gradient index (GRIN) lens implantation in the medial prefrontal cortex (mPFC) of adult mice. The workflow can be adapted to other cortical or subcortical brain regions with appropriate stereotaxic coordinate adjustment.

Calcium (Ca²⁺) is a universal second messenger regulating cellular survival,^1,2,3 synaptic transmission,² neurotransmitter release,^4,5 and plasticity, essential for learning memory and overall brain activity.⁶ Precise measurement of neuronal Ca²⁺ activity in vivo is critical for understanding both normal and diseased brain states. Genetically encoded Ca²⁺ indicators (GECIs) including GCaMP variants,^7,8,9,10,11 allow stable and cell-type-specific labeling and long-term measurement of intracellular Ca²⁺ dynamics.^12,13 Head-mounted miniature microscopes (miniscopes) combined with implanted gradient index (GRIN) lenses provide a powerful approach for deep brain Ca²⁺ imaging of freely behaving animals.¹⁴

GRIN lens implantation remains a critical technical bottleneck for high-quality imaging. Conventional methods including tissue-aspiration, often cause irreversible tissue damage and long term inflammation.^15,16 These procedures are time-consuming,¹⁷ and heavily depend on surgical expertise, contributing to variability in experimental outcomes.^{8,9,17,18,19,20,21} To overcome these challenges, we present the NeuralGlider Inserter (NeuralGlider, Actuated Medical, Inc.), an ultrasound-assisted GRIN lens implantation system designed to minimize tissue damage and reduce user-dependent variability. NeuralGlider offers controlled insertion speed, gentle lens rotation and axial micro-oscillation, and vacuum based stabilization to improve lens alignment and implantation precision, enabling high quality deep brain Ca²⁺ imaging in mice.

Experimental timeline overview

• 1.Perform stereotaxic viral injection of fluorescent calcium indicator (e.g., GCaMP).
• 2.Allow 2 weeks for viral expression.
• 3.Implant GRIN lens at the target brain region.
• 4.Allow 3 weeks for surgical recovery after GRIN lens implantation.
• 5.Verify calcium signal quality and attach miniscope baseplate.
• 6.Conduct in vivo calcium imaging using the miniscope.

Pause point: Viral expression period (2 weeks) and post-implantation recovery (3 weeks) can be adjusted.

This manuscript focuses primarily on Step 3 (GRIN lens implantation at the target brain region) and provides detailed procedures, critical considerations, troubleshooting guidance and representative outcomes for this step. The remaining steps are included to provide experimental context and overview of the complete workflow.

Innovation

NeuralGlider offers significant advancements to implant GRIN lens for deep-brain in vivo calcium imaging in awake behaving mice, especially benefiting larger-diameter (e.g., 1-mm) GRIN lens implantation. The conventional tissue-aspiration method often causes extensive tissue damage, cortical dimpling, bleeding, prolonged surgery and highly variable experimental outcomes that depend heavily on surgeons’ skill. Generally, experienced surgeons (defined as ones with more than 10 successful surgeries yielding images of score 3 and above) perform noticeably better than novices, with higher success rates and good to excellent imaging. NeuralGlider overcomes these issues through ultrasound-assisted insertion, controlled speed, gentle lens rotation, and vacuum-based lens stabilization. This approach produces significantly better in vivo calcium imaging quality and dramatically reduces user dependence. With less tissue disruption and more consistent lens alignment, it yields clearer signals, higher active neuron counts in the field of view at the mPFC, and no significant difference in quality scores between Novice and Experienced Users. It also improves overall surgical efficiency, lowering mortality during and after surgery by shortening GRIN lens implantation time which reduces exposure to anesthesia and improves postoperative recovery.

Institutional permissions

All animal procedures must be approved by the relevant Institutional Animal Care and Use Committee (IACUC) and conducted in accordance with NIH guidelines. Investigators must obtain appropriate institutional approval prior to initiating experiments.

The experiments described here were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Wyoming and were conducted in accordance with NIH guidelines for the care and use of laboratory animals. Male and female Tardbp flox/flox (Tdp-43^F/F) mice, aged 7–23 months, were injected with GCaMP8m and used for all experiments. Tdp-43^F/F mice contain a flanked exon 3 in the mouse Tardbp gene, serving as a conditional knockout model for investigating neural circuitry mechanisms of TDP-43 loss-of-function. Mice were group housed prior to experimental enrollment and individually housed after viral vector injection under a 12-hour light/dark cycle with ad libitum access to food and water.

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Bacterial and virus strains
AAV9-CaMKII-jGCaMP8M	Addgene	176751-AAV9
Chemicals, peptides, and recombinant proteins
Vetbond Tissue Adhesive	Santa Cruz Animal Health	sc-361913
Fibronectin	Sigma	F1141-1MG
Sterile Saline (0.9% Sodium Chloride)	Hospira	0409-4888-02
Buprenorphine	Midwest Veterinary Supply, Inc	191.26890.3
Meloxicam	Midwest Veterinary Supply, Inc	562.10130.3
Metabond Dental Cement Powder	Parkell Inc.	S396
Metabond Quick Base	Parkell Inc.	S398
Metabond Catalyst	Parkell Inc.	S371
DuraLay Powder	Reliance Dental Mfg	2244-R
DuraLay Liquid	Reliance Dental Mfg	2244-R
Carbon Powder	SIGMA-ALDRICH	484164-10G
Critical commercial assays
NeuralGlider Inserter	Actuated Medical, Inc.	RRID:SCR_023753
Rocker 300 Vacuum Pump	Southern Labware, Inc	167300
GRIN Lens (1-mm diameter)	GO!FOTON	ILW-100-P0460-055-NC
Microliter Syringe	Hamilton Company	7653-01
34 gauge Needle	Hamilton Company	207435
Miniscope hardware	Dr. Da-Ting Lin, IRP/NIDA	Custom designed
Baseplate	PROTOLABS	1438-7472-004
Experimental models: Organisms/strains
Tdp-43F/F mice	Dr. Philip Wong, Johns Hopkins University School of Medicine	Genotype: homozygous Age: 7-23 months of age Gender: males and females
Software and algorithms
NeuralGlider Inserter	Actuated Medical, Inc.	RRID:SCR_023753
NeuView	Dr. Da-Ting Lin, IRP/NIDA	Custom designed
ImageJ	http://imagej.org	1.54f
R programming	https://www.r-project.org/	4.2.3

Step-by-step method details

Stereotaxic viral injection of jGCaMP8m

Timing: ∼90 min per mouse

Here, we briefly describe steps for AAV viral injection into the medial prefrontal cortex (mPFC), to label specific groups of neurons with fluorescent Ca²⁺ indicator.

• 1.
  Perform stereotaxic viral injection following the procedures described in Thapa et al., 2021.²¹

  Note: Procedures outlining AAV viral injection targeting the medial prefrontal cortex (mPFC) of adult mice causing expression of calcium indicator (GCaMP) was not described in detail here. For complete details of viral injection procedure, please refer to Thapa et al., 2021.²¹

  • a.
    Anesthetize mice (2% isoflurane).
  • b.
    Inject 500 nl of AAV1-CaMKIIa-jGCaMP8m-WPRE (1.8 × 10¹³ GC/mL, diluted 1:2 in sterile saline, Addgene, MA, USA) into the mPFC via a 34 gauge needle and microliter syringe at the following coordinates relative to bregma: A/P +1.94 mm, M/L +0.50 mm, D/V -1.75 mm.

    Note: Wait for 2 weeks before proceeding to the next step.

Preparation for NeuralGlider-assisted GRIN lens implantation

Timing: ∼30 to 40 min per mouse

Here, we describe all preparation steps before the NeuralGlider-assisted GRIN lens implantation, including GRIN lens being sterilized and coated, NeuralGlider system being initiated, and craniotomy being prepared on the targeted mouse skull.

• 2.
  GRIN lens sterilization.

  • a.
    Sterilize 1.0-mm diameter, 0.46 pitch GRIN lenses (ILW-100-P0460-055-NC) purchased from Go!Foton (Somerset, NJ, USA), previously coated with Parylene-C (Broomfield, CO, USA)²² in 70% ethanol for 15 min and rinse in sterile saline for 15 min.
  • b.
    Coat lens with Fibronectin (50 μg/mL, Sigma-Aldrich, GA, USA) under sterile conditions.

CRITICAL: Avoid touching the lens surface to maintain sterility.

• 3.
  NeuralGlider hardware setup.

  • a.
    Perform hardware setup and initialization according to the manufacturer’s operating instruction for the NeuralGlider System (Figure 1).
  • b.
    Set up the NeuralGlider Actuator, Motor Controller, and Control Box and connect to the computer using USB cables before powering on.
  • c.
    Check the system for proper connection and verify the power and fault indicators on the Control Box. Turn on the Control Box only after confirming that all USB connections are secured.
  • d.
    Ensure that the power indicator is ON and the fault indicator is OFF before proceeding. If a fault indicator (yellow light) is present, check and correct all connections before continuing.
• 4.
  Craniotomy preparation.

  • a.
    Anesthetize mouse with 5% isoflurane in O₂ (1 L/min) and remove any regrown hair around the surgical area.
  • b.
    Settle the mouse on a warmed heating pad (35°C) and secure it onto the stereotaxic frame with ear bars and nose clip, continuously supplied with 2% isoflurane in O₂ (1 L/min).
  • c.
    Apply ophthalmic ointment to both eyes.
  • d.
    Clean the scalp with 3 alternating scrubs of 7.5% povidone iodine and 70% ethanol.
  • e.
    Remove skin and underlying connective tissue over the skull using sterile scissors, then adhere any remaining skin to the skull with Vetbond Tissue Adhesive.
  • f.
    Level the skull along the anterior-posterior (A/P) axis by adjusting the nose clip until bregma and lambda are aligned at the same dorsal-ventral (D/V) coordinates.
  • g.
    Lightly mark the targeted location (A/P +1.94 mm, M/L 0.8 mm from bregma) on the skull using a 0.5 mm drill bit.
  • h.
    Drill a full-thickness craniotomy at the marked coordinate using a 1.2 mm drill bit, intermittently irrigating sterile saline to prevent heat buildup.
  • i.
    Puncture and carefully remove the underlying dura mater using a 30-gauge needle tip to expose brain tissue for lens insertion.
  • j.
    Flushing the craniotomy with sterile saline to clean the debris and aspirate excess liquid with a 26G needle connecting to a vacuum canister of the house vacuum (referred to as the surgical vacuum).

CRITICAL: Avoid compressing underlying cortical tissue.

• 5.
  NeuralGlider software launch and communication verification.

  • a.
    Launch NeuralGlider software on the computer and wait for Initialization window to appear.
  • b.
    The initialization window will open and indicate if the communication port for the Motor Controller was detected.
  • c.
    If the port is not found, stop, close the software, check connections and restart the system.
• 6.
  NeuralGlider Motor initialization.

  • a.
    When prompted, a pop_up window will ask if you “Would like to use relative position of zero the Motor?”

    Note: “Current position” to maintain the current Motor location, or “Zero” to move the Motor to the top position and establish absolute positioning.

  • b.
    If “Zero” is selected, allow the Motor to reach the top limit without interruption.
  • c.
    Confirm that the position displays updates correctly.

    

• 7.
  Set NeuralGlider Motor home position.

  Note: The Set Home window will open to set the Motor home position.

  • a.
    Enter a home position value between 0 and 26 mm.

    

  • b.
    Click “Go to Home” and wait the motor stops moving.
• 8.
  Final check for NeuralGlider initializations.

  • a.
    Click “OK” to enable the main NeuralGlider interface.
  • b.
    Ensure that the LED indicator on the Control Box is green. If not, click “Reconnect” and wait until the LED indicator is green.

Figure 1.

NeuralGlider Inserter components and software interfaces

(A) NeuralGlider Inserter Components, #1–9.

(B) Dashed insert from Panel A, magnification of Vacuum Coupler holding GRIN lens.

(C) NeuralGlider Software Illustration. Left panel: Save Info Tab, Insertion Control- Manual. Right panel: Insertion Information Tab, Insertion Control- Auto.

NeuralGlider-assisted GRIN lens implantation

Timing: ∼30 to 40 min per mouse

Here, we describe essential steps for GRIN lens being inserted into the targeted depth of the mPFC, assisted by the NeuralGlider system.

• 9.
  Mounting the GRIN lens to the Vacuum Coupler.

  • a.
    Turn on the second vacuum connected to the NeuralGlider Vacuum Coupler (i.e., the lens retention vacuum).
  • b.
    Grab the sterilized GRIN lens with forceps, place it below the NeuralGlider Vacuum Coupler, and align it vertically.
  • c.
    Confirm that the GRIN lens is firmly attached to the NeuralGlider Vacuum Coupler with stable vacuum attachment without wobbling or tilting.

Note: Below is a picture of the entire surgical set-up, including the stereotaxic frame, NeuralGlider system, and two vacuums. To verify secure attachment, briefly enable actuation and observe lens. A properly attached lens will remain stably coupled during rotation, whereas an improperly attached lens may detach immediately. Proceed with the insertion procedure only after secure fixation has been confirmed.

• 10.
  Establishing the anatomical zero.

  • a.
    Position the lens above the bregma using the stereotaxic arm.
  • b.
    Slowly lower the lens until it barely touches the skull surface without applying too much pressure.
  • c.
    Record the D/V position as the anatomical zero.
• 11.
  Positioning the GRIN lens over craniotomy.

  • a.
    Lift up the lens slightly (1–2 mm) above the skull surface.
  • b.
    Move the stereotaxic arm to the position where the GRIN lens is directly above the craniotomy center.
  • c.
    Lower the lens slightly into the craniotomy to check the alignment and ensure the GRIN lens is centered and does not touch any edges of the skull.
  • d.
    Retract the GRIN lens and return it to the anatomical zero position.
• 12.
  Irrigation and aspiration of the craniotomy.

  • a.
    Turn on the surgical vacuum to intermittently remove excess saline and blood from the craniotomy.
  • b.
    Ensure continuous irrigation with sterile saline to maintain a clean and moisturized surgical field.
• 13.
  Set the insertion parameters in the NeuralGlider Insertion Control Panel (Software).

  • a.
    Ultrasonic Actuation Power: 0.5–1 W.
  • b.
    Insertion Velocity: 0.03 mm/s.
  • c.
    Target Depth: 1.8 mm.

• 14.
  Lens insertion process assisted by NeuralGlider.

  • a.
    Confirm that all parameters are correct and the GRIN lens is at the anatomical zero.
  • b.
    Verify if the Actuator is functional by clicking the Enable Actuation in the Insertion Control Area to indicate “ON”.
  • c.
    Click “Manual Mode”, then click the down arrow above Move to advance the GRIN lens slowly towards the targeted depth.
  • d.
    Continue insertion process until the GRIN lens reaches the programmed targeted depth.

Note: Ensure that a complete hemostasis (i.e., no active bleeding is observed) is achieved prior to initiating lens insertion. Maintain a continuous saline irrigation during insertion to preserve a clear surgical field. Bleeding during lens advancement due to tissue displacement should be continuously cleared by aspiration ensuring it does not contact the top surface of the GRIN lens which can compromise image quality. Use the surgical vacuum as needed but keep it away from touching the top of the GRIN lens. If encountering any resistance, pause insertion by clicking “Stop”. Clean the surgical site with saline irrigation and aspiration, then resume by pressing the down arrow above “Move” to advance the GRIN lens until the programmed depth is reached.

• 15.
  Post-insertion lens settlement.

  • a.
    Continue irrigation with saline briefly until bleeding is minimal or stops.
  • b.
    Turn off the surgical vacuum while keeping the GRIN lens retention vacuum ON.
  • c.
    Leave the GRIN lens undisturbed for 15–20 s to allow tissue settlement.
• 16.
  Vacuum Coupler removal.

  • a.
    Release the GRIN lens retention vacuum and wait for 10–15 s.
  • b.
    Slowly raise the stereotaxic arm to lift the Vacuum Coupler away.
  • c.
    Confirm that the lens remains stable and does not move.

CRITICAL: Maintain continuous irrigation to minimize tissue damage and prevent bleeding; precise alignment and controlled insertion speed are essential for high-quality imaging

GRIN lens fixation on the skull and post-surgery care

Timing: ∼15 min

Here, we describe steps for GRIN lens being fixed onto the mouse skull with dental cement and the subsequent post-surgery care.

• 17.Fill the space surrounding the implanted GRIN lens with 1% melted agarose to stabilize the GRIN lens.
• 18.Clean the skull thoroughly with sterile saline and cotton swabs.
• 19.Apply dental cement with a low-to-medium viscosity to allow smooth, even spreading around the GRIN lens, covering the entire exposed skull while minimizing mechanical pressure on the implant.
• 20.After the cement hardens, apply a thin top layer of dental cement mixed with charcoal.
• 21.Affix a protective cap (e.g., a customized PCR tube bottom) with cyanoacrylate to protect the lens.
• 22.Trim the mouse’s hind limb nails to reduce the risk of scratching the surgical site during post-operative recovery, which may damage the implant and administer buprenorphine (0.2 mg/kg, subcutaneous) for analgesia.
• 23.Transfer the mouse to a clean cage and incubate in a 37°C isothermal chamber for 3 days, provide meloxicam (5 mg/kg, subcutaneous) treatment twice-daily for three days, and monitor wound healing and general activity.

Note: Recovery is monitored daily via food and water intake, general locomotory activity, posture and grooming and nestling behavior.

Pause point: Allow a 3-weeks recovery period before proceeding to baseplate attachment and in vivo imaging.

CRITICAL: Ensure the GRIN lens is fully stabilized and the cement is secure to prevent movement; continue monitoring mice closely for signs of distress or infection.

Miniscope baseplate affixation and in vivo calcium imaging

Timing: ∼30 min for affixing baseplate (step 24)

Timing: ∼15 min per imaging session (step 25)

Here, we describe steps for miniscope baseplate being attached to the mouse skull and the subsequent in vivo calcium imaging through the implanted GRIN lens.

• 24.
  Affixing miniscope baseplate.

  • a.
    Anesthetize the mouse with 5% isoflurane. Place the mouse on a warmed heating pad (35°C) and secure it onto the stereotaxic frame with ear bars and nose clip.
  • b.
    Maintain anesthesia with 2% isoflurane in O₂ (1 L/min) and remove the protective cap from the GRIN lens implanted region.
  • c.
    Clean GRIN lens surface with sterile saline.
  • d.
    Temporarily assemble a miniscope with its baseplate and identify the optimal focal plane using live Ca²⁺ imaging.
  • e.
    Secure the baseplate onto the skull using dental cement.
  • f.
    Detach the miniscope from its baseplate and cover lens with the protective cap.

CRITICAL: Corrected baseplate alignment is essential for stable, high-quality imaging across sessions.

• 25.
  In vivo calcium imaging.

  • a.
    Anesthetize the mouse with 5% isoflurane in an induction chamber.
  • b.
    Take out the mouse from the induction chamber, remove the protective cap, and mount a miniscope onto the baseplate.
  • c.
    Allow the mouse to wake up and adjust focal plane if necessary.
  • d.
    Adjust LED intensity to optimize signal while minimizing photobleaching.
  • e.
    Record imaging session (typically 15 min; 3 × 5 min segments).

Expected outcomes

We performed stereotaxic viral injection into the mPFC to express the calcium indicator jGCaMP8m predominantly in excitatory neurons. Two weeks later, GRIN lenses were implanted using either the NeuralGlider or the Conventional tissue aspiration-based method, and in vivo calcium imaging was subsequently performed using a head-mounted miniscope. The NeuralGlider Inserter (NeuralGlider, Figure 1) uses ultrasonic micro-vibration to reduce forces during insertions. It assists with insertions of neural implants and optics in preclinical models. NeuralGlider consists of a reusable Actuator connected to a reusable Motor (Figure 1A). The Actuator is powered by a reusable Control Box, and the Motor is powered by a reusable Motor Controller. The NeuralGlider Actuator ultrasonically vibrates a limited-use Vacuum Coupler (Figure 1B) connected to neural implant or GRIN lens. The Coupler designs convert a portion of the axial oscillation into a wave that slowly rotates the GRIN lens, further improving force reduction and placement. The Actuator and Motor are controlled using the NeuralGlider Software, which allows the User to specify position, velocity, actuation on/off, and power (Figure 1C).

NeuralGlider was evaluated initially by using agarose as an in vitro tissue-mimicking model (Figure 2). We tested the insertion of optical lenses of varying diameters (400 μm to 1800 μm) and found that applying vibration significantly reduced the insertion force, with reductions greater than 50% across all lens sizes tested. The effect was more pronounced for larger-diameter lenses, where vibration significantly decreased puncture force and minimized agarose compression compared to Control insertions (p < 0.01). We then performed larger-diameter GRIN lens (i.e., 1-mm in diameter) implantations with NeuralGlider assistance and compared the in vivo calcium imaging quality to that of conventional tissue-aspiration method for both Novice and Experienced Users (Figure 3; Table 1). To properly evaluate the quality of in vivo calcium imaging, fields of view were assigned a Score from 0 to 4 based on neuronal visibility, where Score 0 indicates no fluorescence, Score 1 indicates visible fluorescence but no distinguishable neurons, Score 2 indicates fewer than 15 neurons, Score 3 indicates 15 − 50 neurons, and Score 4 indicates > 50 neurons per field (Figure 3A). As indicated in Figures 3B and Table 1, for the Conventional tissue-aspiration method, Novice User demonstrated a higher failure rate, with 29% of surgeries resulting in no fluorescence (Score 0), approximately 28.3% showing non-analyzable fluorescence (Score 1 and Score 2) and approximately 13% achieving good to excellent quality imaging (Score 3 and Score 4). In contrast, an Experienced User using the Conventional method has 0% of no fluorescence (score 0), approximately 37.3% of non-analyzable fluorescence (Score 1 and score 2) and approximately 41.3% of good to excellent imaging (score 3 and score 4). For the NeuralGlider-assisted method, Novice User produced only 3% of failure surgeries resulting in no fluorescence (score 0), approximately 39% showing non-analyzable fluorescence (score 1 and score 2) and approximately 45% achieving good to excellent quality imaging (score 3 and score 4). An Experienced User using the NeuralGlider produced 4% of no fluorescence (Score 0), approx. 16% of non-analyzable fluorescence (Score 1 and Score 2) and approx. 72% of good to excellent imaging (Score 3 and Score 4). When Novice and Experienced Users were combined, the NeuralGlider assisted method yielded a significantly higher survival rate during surgery (90% vs. 73%, p = 0.03), a significantly lower no fluorescence (i.e., Score 0) imaging (3.4% vs. 21%, p = 0.008), and a significantly higher quality (i.e., Score 3) imaging (48% vs. 18%, p = 0.0003), when compared to the Conventional tissue-aspiration method.

Figure 2.

NeuralGlider Inserter Mode of Action

(A) Peak insertion force (mN) of in vitro insertions of various sized optics completed with NeuralGlider Inserter into agarose model. A greater than 50% reduction of force (ROF) was detected following insertions using Vibration (VIB) vs Control (CON) for all lenses tested.

(B) Vibration reduces puncture force of large diameter (400 μm and 1000 μm) optical lenses resulting in decreased tissue compression compared to Control insertions into agarose bench model (p < 0.01, Student’s t test). n = 6–10 insertions/treatment/optics size; Bars represent Means ± SEM.

Figure 3.

Calcium imaging quality scoring system

(A) Representative images for scoring system. Score 0: no fluorescence. Score 1: fluorescence but no distinguishable neurons. Score 2: < 15 neurons. Score 3: 15 − 50 neurons. Score 4: > 50 neurons. Scale bar, 200 μm.

(B) Percentage comparisons of mortality and image quality between Conventional Method and NeuralGlider-assistant method. Left: Comparisons of Novice and Experienced Operator completed procedures. Right: Comparisons with all operators. NeuralGlider-assisted implantation significantly reduced failed imaging (3.4% vs. 21%, ∗∗, p = 0.008), increased high-quality imaging (48% vs. 18%, ∗∗∗, p = 0.0003), and decreased mortality during surgery (9% vs. 27%, ∗, p = 0.012) compared with the conventional method. Analysis with 2-sample test for equality of proportions. Sample sizes were n = 78 for the conventional method and n = 58 for the NeuralGlider-assisted method.

Table 1.

Imaging quality analysis

GRIN lens insertion method	Operator	Operator experience	Completed surgeries (N)	Imaged subjects (N)	Calcium image quality score (score proportion/Surgeries)
					Score 0	Score 1	Score 2	Score 3	Score 4
Traditional Method	R	Novice	40	25	10	3	5	6	1
	N	Novice	14	13	6	7	0	0	0
	R+ N	Novice	54	38	16	10	5	6	1
				38/54 = 70%	16/54 = 29%	10/54= 19%	5/54 = 9.3%	6/54 = 11%	1/54 = 1.9%
	R	Experienced	24	19	0	2	7	8	2
				19/24 = 79%	0/24 = 0%	2/24 = 8.3%	7/24 = 29%	8/24 = 33%	2/24 = 8.3%
	Novice vs Experienced Proportion Analysis p-value			0.6	a0.007	0.4	0.056	a0.04	0.5
NeuralGlider Assisted	S	Novice	33	29	1	7	6	14	1
				29/33 = 88%	1/33 = 3%	7/33 = 21%	6/33 = 18%	14/33 = 42%	1/33 = 3%
	S	Experienced	25	23	1	1	3	14	4
				23/25 = 92%	1/25 = 4%	1/25 = 4%	3/25 = 12%	14/25 = 56%	4/25 = 16%
	Novice vs Experienced Proportion Analysis p-value			0.9	1.0	0.1	0.8	0.4	0.2
Traditional Method	All		78	57	16	12	12	14	3
				57/78 = 73%	16/78 = 21%	12/78 = 15%	12/78 = 15%	14/78 =18%	3/78 = 3.8%
NeuralGlider Assisted	All		58	52	2	8	9	28	5
				52/58 =90%	2/58 = 3.4%	8/58 = 14%	9/58 = 16%	28/58 = 48%	5/58 =8.6%
	Insertion Method Proportion Analysis p-value			b0.03	b0.008	0.9	1.0	b0.0003	0.4

Quality Score, higher score represents superior imaging.

^aSignificant difference (p < 0.05) Novice vs. Experienced via 2-sample test for equality of proportions calculated using R.

^bSignificant difference (p < 0.05) Traditional vs. NeuralGlider Assisted via 2-sample test for equality of proportions calculated using R.

Total surgery time required for GRIN lens implantation was also compared between Conventional tissue aspiration method and NeuralGlider assisted implantation (Figure 4). Using the Conventional method, the Novice User required an average of 185.9 ± 30.2 min, whereas the same User as an Experienced User completed the procedure in 170.2 ± 36.4 min. In contrast, using the NeuralGlider assisted system, the Novice User required only 80.16 ± 17.7 min and 64.6 ± 9.5 min as an Experienced User. A Kruskal-Wallis’s test revealed a significant difference in surgery duration among the 4 groups (p < 0.05). Dunn’s multiple comparisons test demonstrated that NeuralGlider Novice and Experienced groups showed significantly shorter surgery time compared to their respective conventional method (adj. p < 0.0001) and no significant difference was observed between Novice and Experienced performance within the same implantation method. These results indicate that NeuralGlider substantially reduces surgery time even for less experienced users.

Figure 4.

Comparison of total GRIN lens implantation time between conventional tissue aspiration and NeuralGlider-assisted insertions

NeuralGlider Inserter significantly reduced surgery duration for both Novice (Green bars) and Experienced (Blue bars) Operators compared to the Conventional Method (Kruskal–Wallis, p < 0.05; Dunn’s multiple comparisons, adj. ∗∗∗∗, p < 0.0001), with no significant difference between experience levels within each method. Bars represent Means ± SD.

As young and old mice could yield different survival rates during GRIN lens implantation surgery, we compared surgical survival rates for mice with different ages comparing the Conventional tissue aspiration method and the NeuralGlider assisted method (Table 2). In the Conventional group (n = 78 mice), 64 were older (12–18 months) with 44 surviving (69%) and 14 were younger (10–13 months) with 13 surviving (93%). In the NeuralGlider group (n = 57 mice), 32 were older (>18 months) with 29 surviving (91%) and 25 were younger (7 − 12 months) with 23 surviving (92%). Using a 2-sample test for equality of proportion, we found that the overall survival was higher in the NeuralGlider group compared to the Conventional group (p < 0.05). Survival among older mice was also significantly improved with the NeuralGlider assisted method (p < 0.05).

Table 2.

Comparison of survival rates between implantation methods

GRIN lens insertion method	Age (months)	Surgery total (N)	Survived (N)	Method proportion Analysis p-value
Conventional Aspiration Method	12–18 (Old)	64	44 44/64 = 69%	a0.035
	10–13 (Young)	14	13 13/14 = 93%	1.0
	–	Total: 78	57 57/ 78 = 73%	a0.012
NeuralGlider Assisted	> 18 (Old)	32	29 29/32 = 91%	a0.035
	7–12 (Young)	25	23 23/25 = 92%	1.0
	–	Total: 57	52 52/57 = 91%	a0.012

^aSignificant difference (p < 0.05) Conventional Aspiration vs. NeuralGlider Assisted via 2-sample test for equality of proportions calculated using R.

In summary, NeuralGlider not only eliminates much of the dependence on User experience but also improves imaging quality, safety, efficiency, and reliability in vivo calcium imaging in mice.

Quantification and statistical analysis

All experiments were independently replicated. Fluorescence imaging quality was evaluated using a graded scoring scale (0 − 4), with higher scores indicating higher quality imaging. Fluorescence Score 0: no fluorescence, Score 1: fluorescence but no distinguishable neurons. Score 2: < 15 neurons. Score 3: 15 − 50 neurons, and Score 4: > 50 neurons. Statistical analyses were performed using GraphPad Prism and R. Proportions were compared using two-sample tests for equality of proportions to evaluate differences between Novice and Experienced Users as well as between Conventional and NeuralGlider assisted methods. Surgery duration was compared across groups using a Kruskal Wallis test with Dunn’s post hoc comparisons. Statistical significance was set as p < 0.05.

Limitations

limitations include equipment needs associated with the NeuralGlider system and the necessity of testing across additional brain regions as our current data are limited to the medial prefrontal cortex.

Troubleshooting

Problem 1

Excessive bleeding during craniotomy (related to Step 4).

Potential solution

This typically arises from damage to the surface vasculature or dura because of overly aggressive or continuous drilling.

Perform drilling in short bursts with frequent irrigation to prevent heat buildup. Stop drilling immediately once resistance changes. Rupture the dura layer carefully using a 30-gauge needle to avoid vascular injury.

Problem 2

Communication error during NeuralGlider system set up (related to Step 5).

Potential solution

This typically arises from unstable USB connection or software failing to detect the COM port.

Restart the software to reinitialize communication to the COM port. Disconnect and reconnect USB and hardware compartments if the issue persists.

Problem 3

Encounter resistance during insertion (related to Step 14).

Potential solution

This typically arises from accumulation of tissue debris or small bone pieces, or misalignment of GRIN lens with craniotomy.

Do not force insertion, as this would cause significant damage to tissue. Instead, retract the GRIN lens slightly, irrigate the surgical site thoroughly and remove fluid and debris via aspiration (i.e., removal of irrigation fluid and suspended debris only, not brain tissue). Check the alignment carefully before resuming the insertion process.

Problem 4

Weak fluorescent signal during in vivo calcium imaging (related to Step 25).

Potential solution

This typically arises from low viral expression, and/or insufficient LED intensity of the miniscope.

Ensure the LED power gradually increased to improve signal detection. Avoid a sudden increase of LED power to high intensity, as this would lead to photobleaching and reduce signal quality over time.

Resource availability

Lead contact

Further inquiries and resources requests can be addressed to Yun Li (yli30@uwyo.edu).

Technical contact

Technical questions regarding this protocol should be directed to Sishir Gautam (sgautam3@uwyo.edu).

Materials availability

This study did not generate new, unique reagents.

Data and code availability

Raw data can be obtained from the lead contact upon reasonable request. No code was used for the analysis in this study.

Acknowledgments

This work was supported by National Institutes of Health (4R01NS129878–02, R21AG087909, and 1R43AG084801-01).

Author contributions

Y.L. designed and supervised the studies. S.G. and Y.L. wrote the manuscript with inputs from O.M.O.-G. and J.K.G. S.G. performed NeuralGlider-assisted GRIN lens implantation. R.T. and N.A. performed tissue-aspiration based GRIN lens implantation. J.K.G. performed in vitro evaluation in agarose. O.M.O.-G. performed statistics.

Declaration of interests

O.M.O.-G. and J.K.G. are employed by Actuated Medical, Inc.