Targeted Neuronal Injury for the Non-Invasive Disconnection of Brain Circuitry

Summary

The goal of the protocol is to provide a method for producing non-invasive neuronal lesions in the brain. The method utilizes Magnetic Resonance-guided Focused Ultrasound (MRgFUS) to open the Blood Brain Barrier in a transient and focal manner, in order to deliver a circulating neurotoxin to the brain parenchyma.

Abstract

Surgical intervention can be quite effective for treating certain types of medically intractable neurological diseases. This approach is particularly useful for disorders in which identifiable neuronal circuitry plays a key role, such as epilepsy and movement disorders. Currently available surgical modalities, while effective, generally involve an invasive surgical procedure, which can result in surgical injury to non-target tissues. Consequently, it would be of value to expand the range of surgical approaches to include a technique that is both non-invasive and neurotoxic.

Here, a method is presented for producing focal, neuronal lesions in the brain in a non-invasive manner. This approach utilizes low-intensity focused ultrasound together with intravenous microbubbles to transiently and focally open the Blood Brain Barrier (BBB). The period of transient BBB opening is then exploited to focally deliver a systemically administered neurotoxin to a targeted brain area. The neurotoxin quinolinic acid (QA) is normally BBB-impermeable, and is well-tolerated when administered intraperitoneally or intravenously. However, when QA gains direct access to brain tissue, it is toxic to the neurons. This method has been used in rats and mice to target specific brain regions. Immediately after MRgFUS, successful opening of the BBB is confirmed using contrast enhanced T1-weighted imaging. After the procedure, T2 imaging shows injury restricted to the targeted area of the brain and the loss of neurons in the targeted area can be confirmed post-mortem utilizing histological techniques. Notably, animals injected with saline rather than QA do demonstrate opening of the BBB, but dot not exhibit injury or neuronal loss. This method, termed Precise Intracerebral Non-invasive Guided surgery (PING) could provide a non-invasive approach for treating neurological disorders associated with disturbances in neural circuitry.

Introduction

The purpose of this method is to provide a means for producing non-invasive neuronal lesions in a targeted region of the brain. The rationale for developing such an approach is to disconnect neuronal circuitry contributing to neurological disorders. For instance, surgery can be quite effective in treating certain medically intractable neurological disorders, such as drug resistant epilepsy (DRE)¹. However, each of the available surgical modalities possess limitations in terms of producing undesirable collateral damage to the brain. Traditional resective surgery can be highly invasive with the risk of bleeding, infection, blood clots, stroke, seizures, swelling of the brain, and nerve damage². Alternatives to resective surgery that are minimally invasive or non-invasive include laser interstitial thermal therapy and radiosurgery, which have also proved to be effective in suppressing seizures in DRE. More recently, thermal lesions produced by high-intensity focused ultrasound (HIFU) have shown promise in reducing seizures. HIFU is non-invasive; however, its treatment window is currently limited to more central areas of the brain because of the risk of thermal injury to non-target tissue located in the vicinity of the skull. Despite such limitations, the benefits of surgery often outweigh the potential risks. For instance, although surgery for DRE can produce collateral brain damage, its beneficial effects in suppressing seizures and improving quality of life typically prevail over the surgical risks.

The method described herein, Precise Intracerebral Non-invasive Guided surgery (PING), was developed for the purpose of disconnecting neural circuitry, while limiting collateral brain damage. The method utilizes low intensity focused ultrasound combined with intravenous injection of microbubbles to open the BBB, in order to deliver a neurotoxin. This approach does not produce thermal lesions to the brain^3,4,5,6,7, and the period of BBB opening can be exploited to deliver BBB-impermeable compounds to the brain parenchyma. The opening of the BBB is transient, and can be produced in a targeted manner using magnetic resonance imaging guidance. In our studies, the period of BBB opening has been utilized to deliver a circulating neurotoxin to a targeted area of the brain parenchyma in rats and mice^8,9. Quinolinic acid is a neurotoxin that is well tolerated when administered intravenously¹⁰, intraarterially¹⁰, or intraperitoneally^8,9,11. The lack of QA toxicity is due to its poor BBB permeability, which has been reported to be negligible¹⁰. In contrast, direct injection of QA into the brain parenchyma produces neuronal lesions that spare neighboring axons^12,13. Thus, when circulating QA gains access to the brain parenchyma in the targeted area of BBB opening, neuronal death is produced^8,9. The present method thus produces focal neuronal loss in a precisely targeted and non-invasive manner.

Protocol

All methods described here have been approved by the University of Virginia Animal Care and Use Committee.

1. Preparation of reagents

1. On the day of surgery, prepare 6.0 mL of injectable quinolinic acid (QA). Dissolve 450 mg of QA in 4.0 mL of 1.0 N NaOH. Add 0.6 mL of 10x PBS, pH to 7.4, and bring to a final volume of 6.0 mL with dH₂O. Filter through 0.22 µm syringe filter. The solution is stable for 2 weeks at 4 °C.
2. Prepare an aqueous dispersion of microbubbles in normal saline by probe sonication from decafluorobutane gas and stabilize with DSPC/PEG stearate monolayer shell¹².
3. Size microbubbles by flotation at normal gravity. Determine microbubble concentration and size by electrozone sensing using Multisizer III counter. Microbubble concentration and size distribution should be 6 x 10⁸/mL and ~2 μm (mean particle diameter), respectively. The largest bubbles are removed by flotation exclusion/separation.
4. Alternatively, purchase commercially available microbubbles.

2. Preparation of animals

1. Acclimatize the animal (rat or mouse) for 3 days after delivery. The experiments described here used Sprague-Dawley rats (5–6 weeks of age) or telencephalic internal structural heterotopia (tish) rats (local colony).
2. House the animals under a 12 hour light: 12 hour dark cycle.
3. Record the animals’ weights. This information is important throughout the procedure.
4. Obtain T2-weighted MR images the day before the FUS procedure in order to establish preoperative baseline images. Use the following parameters for T2 imaging: repetition time/echo time [TR/TE] =3,000/138 milliseconds, 3 averages, field of view=29 x 45 mm², matrix size = 125 x 192, slice thickness = 0.23 mm.
5. Anesthetize the animal with isoflurane (4% induction, 2% maintenance). Confirm adequate depth of anesthesia using a toe pinch. Apply an ophthalmic ointment to the eyes.
6. Shave the animal’s scalp, and remove the remaining hair with depilatory cream.
7. Use a tail vein catheter for the infusion of microbubbles, contrast agent, and QA. The catheters consist of a length of PE10 tubing fitted with a 30 G x ½ inch needle. Leave a 1 mL syringe with heparinized saline in line, to be removed and reattached when the line is used for infusions.

3. MRI and PING procedures

1. Perform the MRI on a 7 T MR unit with a gradient strength of 600 mT/m/ms (Figure 1 and Figure 2). Perform MRI acquisitions using a surface coil incorporated in the FUS system.
   NOTE: The FUS system used for the experiments comprises three parts: (i) the sonication system is a MR-compatible pre-focused, 8-element annular array, 1.5 MHz transducer (spherical radius = 20 mm ± 2 mm, active diameter = 25 mm (f-number = 0.8), with 80% electric-acoustic efficiency, which is connected to a phased array generator and RF power amplifier; (ii) a MR-compatible motorized positioning stage to move the transducer in the anterior-posterior direction and medio-lateral direction; (iii) a Thermoguide workstation to control the delivery of sonication, including electronic focusing through phase modulation to adjust the focal depth (Figure 2).
2. Place the anesthetized animal on the coil sled assembly (Figure 1) of the MR-compatible FUS system in the prone position. Immobilize the animal using the incisor bar and ear bars incorporated in the cradle of the sled.
3. Apply acoustic gel to the water-wetted scalp; ensuring that no bubbles exist, place the membrane barrier of the water circulator portion of the transducer assembly above the animal’s skull, and lower the transducer assembly as far as possible in a parallel planar orientation relative to the skull plate. Place the diaphragm of the transducer firmly against the animal’s shaven scalp directly over the skull plate.
4. Attach a pneumatic sensor to the body with surgical tape, to monitor respiration. Position the pneumatic sensor on the left lower rib cage.
5. Move the FUS arm assembly with coil, sled, and animal into the 7T MRI unit (Figure 1).
6. Run a T2-scout sequence to determine the general physical position of the transducer assembly relative to the animal’s head, and make mechanical adjustments as necessary (Figure 2). The parameters for T2 imaging are: repetition time/echo time [TR/TE] = 3,000/138 milliseconds, 3 averages, field of view = 29 x 45 mm², matrix size = 125 x 192, slice thickness = 0.23 mm. Thermometry is typically not used in this protocol.
7. Obtain T2 images to refine the transducer positioning. Precisely, define the transducer location and specify the focal point(s) of sonication using the targeting function of the software. The parameters for T2 imaging are: repetition time/echo time [TR/TE] = 3,000/138 milliseconds, 3 averages, field of view = 29 x 45 mm², matrix size = 125 x 192, slice thickness = 0.23 mm.
8. Just prior to sonication, inject 300 μL/kg of microbubbles¹⁴ via the tail vein.
9. Use a 1.5 MHz transducer to produce sonications (30 ms wave packet, 3% duty cycle, 1 Hz burst repetition frequency, 240 s duration/sonication.
10. Immediately after sonication, inject gadodiamide contrast agent via the tail vein, and then perform T1-weighted plus contrast scans to confirm opening of the BBB and the accuracy of targeting. The parameters for T1-weighted imaging: TR/TE = 900/12 milliseconds, 2 averages, field of view = 24 x 30 mm², matrix size = 208 x 256, slice thickness = 0.7 mm. Typically, a single T1 scan is performed.
11. Remove the FUS arm and sled from the MRI, and place the animal on a heating pad set to 40 °C, while maintaining 2% isoflurane anesthesia.
12. Starting 30 min after sonication, use a syringe pump to infuse QA (75 mg/mL stock solution) via the tail vein for 1 h at a rate of 16.8 µL/min to achieve a final dosage of 225 mg/kg (q.s. to 1.0 mL saline).
13. When the infusion is complete, discontinue anesthesia, keeping the animal on a heating pad until alert. Place the animal in a cage and make routine checks for its activity every 15 min, for several hours after the procedure.
14. Return the animal to the vivarium and check every 6 h for the first day for distress or irregular activity.
15. One day post-sonication, perform T2-weighted MR imaging to assess any damages in the area of sonication. The parameters for T2 imaging: repetition time/echo time [TR/TE] = 3,000/138 milliseconds, 3 averages, field of view = 29 x 45 mm², matrix size = 125 x 192, slice thickness = 0.23 mm. Images are evaluated for areas of hyperintensity to identify possible tissue damage/edema.

4. Post-mortem analysis of neuronal loss

1. Allow a post-sonication, survival period of 4–5 days for assessing neuronal loss with Fluoro-Jade histochemistry.
2. Deeply anesthetize the animal with isoflurane, and euthanize via intracardial perfusion with 0.1 M phosphate buffer (pH 7.4) followed by 4% paraformaldehyde in phosphate buffer.
3. Remove the brain from the skull and post-fix for 2 days in 4% paraformaldehyde.
4. Immerse the brain in 30% sucrose for cryoprotection, and cut sections at a thickness of 20–30 μm with a cryostat.
5. Mount cryostat sections onto gelatinized slides and air-dry overnight.
6. Rehydrate slides in distilled water, and then dehydrate in ascending graded ethanols. After dehydration, rehydrate slides in descending graded ethanols.
7. Transfer slides to a solution of 0.06% potassium permanganate for 15 min on an orbital shaker.
8. Rinse slides for 1 min in distilled water and transfer to a 0.001% solution of Fluoro-Jade B in 0.1% acetic acid. Incubate under gentle agitation for 30 min at room temperature. Rinse slides three times for 1 min in distilled water.
9. Dry the slides on a slide warmer, equilibrate in xylenes for 3 min, and coverslip using DPX mounting media.

Results

This section describes the effect of PING on neurons located in a neocortical dysplasia. Tissue dysplasias are a common feature in the brains of patients with drug resistant epilepsy, and surgical removal of seizure-genic dysplasias can provide excellent control of seizures¹⁵. Defining the effect of PING on dysplastic brain tissue is therefore an important priority. A rat model of genetic cortical dysplasia, the tish rat, was selected for studying this issue because the tish brain exhibits dysplas...

Discussion

The PING method is designed to produce non-invasive, targeted neuronal lesions. The method derives from a strong and growing foundation of research in the field of focused ultrasound^3,4,5,6,7. The ability to provide focal access to specific areas of the brain parenchyma via transient opening of the BBB has created an avenue for delivering a wide variety of age...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
7T-ClinScan MRI System	Bruker Biospin, Ettinglen, Germany		MR Image Acquisition
Acoustic Gel	Litho CLEAR	11-601	High Viscosity Accoustic Transmission Gel
DPX Mounting Medium	Electron Microscopy Sciences	13512	Resin Based Cover Glass Mountant
Fluoro-Jade B	EDM Millipore	AG310	High Affinity Stain For Degenerating Neurons
Fluovac anesthetic adsorber	Harvard Apparatus	34-0388	Organic Anaesthesia Scavenger
FUS System	Image Guided Therapy, Pessac, France	LabFUS	MR Compatible Small Animal Focused Ultrasound System
Gadodiamide	GE Healthcare AS, Oslo, Norway	Omniscan	MR Contrast Agent
Heparin	SAGENT	NDC2502140010	Anti-Coagulant
Hypodermic needle 30G x 1/2	Becton-Dickinson	26027	Tail Vein Catheterization
Insulin syringe 28G1/2 (1ml)	EXEL	26027	Administration of Injectables to Tail Vein Catheter
Isofluorane atomizer	SurgiVet	VCT302	Anaesthesia Administration
Isoflurane	Henry Schein	NDC1169567762	Anaesthesia
KMnO4	Sigma	223468	Reagent Used in Fluoro-Jade B Staining
Microbubbles	Produced internally: A. Klibanov	305106	Blood Brain Barrier Disrupting Agent
Microbubbles (commercial source)	Lantheus Medical Imaging, North Billerica, MA	Definity microbubbles	Blood Brain Barrier Disrupting Agent
Monitoring & Gating System	Small Animal Instruments	Model 1030	Respiration Monitoring
Multisizer 3 Coulter counter	Beckman-Coulter, Hialeah, FL	Multisizer 3	Used to Determine Average Size of Microbubbles
Optixcare EYE LUBE	CLC MEDICA, Ontario, Canada	11611	Corneal Protectant-Eye Lube
PE10 tubing	Becton-Dickinson	427401	Tail Vein Catheter Component
Quinolinic Acid	Santa Cruz Biotechnology, Dallas, TX	CAS 89-00-9	Neurotoxin
Sprague-Dawley Rats	Taconic Biosciences	SD-M	Rat Model
Syringe Pump	Carnegie Medicin	CMA 100	Controlled Delivery of Quinolinic Acid
Thermoguide Software	Image Guided Therapy, Pessac, France	Thermoguide	Drives Lab FUS System
Tish Rats	In-house colony		Rat Model
Veet depilatory cream	Reckitt Benckiser		Removal of Scalp Hair