Method Article
* These authors contributed equally
A lumbar intrathecal injection represents a translationally relevant route of administration for delivering gene therapy to the central nervous system. This comprehensive standardized protocol for lumbar intrathecal injections in neonatal, juvenile, and adult mice and rats aims to guide researchers in adopting this technique for preclinical gene therapy studies.
One method to target the central nervous system for treating neurological diseases involves utilizing the lumbar intrathecal route of administration. This approach bypasses the blood-brain barrier to directly access the cerebrospinal fluid and preferentially target cells within the central nervous system. Multiple published preclinical studies employing the lumbar intrathecal injection route have contributed to the development of gene therapy clinical trials; however, the described protocols are variable and dispersed across multiple resources. Here, a comprehensive set of protocols for lumbar intrathecal injections in neonatal, juvenile, and adult mice and rats for preclinical gene therapy studies is presented. With proper training, this injection technique can be performed quickly and reliably. In addition to detailing the injection protocol at each developmental stage, associated parameters, such as injection volume, that can influence study outcomes are discussed. To demonstrate the application of lumbar intrathecal injections for targeting the central nervous system, the expression of adeno-associated virus serotype 9 in the brain, spinal cord, and peripheral tissues is presented following a successful or unsuccessful injection.
A challenge in treating neurological diseases that require global central nervous system (CNS) delivery but are otherwise good candidates for gene therapy has been largely attributed to inefficient targeting of the CNS and relevant cell types1. A substantial amount of research is ongoing to optimize global CNS cell and tissue targeting by engineering delivery vehicles1,2. However, reasonably widespread vector delivery can still be achieved with current gene therapy vector technology, using certain combinations of viral vectors and routes of administration3,4. The current gold standard for obtaining widespread CNS delivery from a one-time treatment is to use adeno-associated virus serotype 9 (AAV9) together with a direct injection into the cerebrospinal fluid (CSF).
There are three typical routes of administration for direct CSF injections: lumbar intrathecal (IT), intracerebroventricular (ICV), and intracisternal (ICM)5. Each of these routes of administration results in different biodistribution patterns in the CNS and peripheral tissues, but they all have the benefit of bypassing the blood-brain barrier (BBB) to reach the cells in the CNS that contribute to neurological disease pathology and phenotypes6. The lumbar IT injection is the standard for clinical drug delivery use in humans, as the clinical procedure is routine and straightforward, with less invasiveness compared to ICV and ICM injections.
The lumbar IT injection is an established technique that is readily used in the anesthetics and analgesic fields, with the first paper published in 18857. The first protocol for lumbar IT injections in adult mice was published in 19808, and it has since been widely adopted and reviewed9. Slight adjustments or improvements to these protocols have been made10,11,12, including a product-conserving technique13. Protocols for lumbar IT injections in adult rats were also first published in 1976, with catheterization for chronic administration14 and direct injection for one-time treatments15. More recently, groups have published protocols for lumbar IT injections in neonatal or juvenile mice and rats16,17.
Wide adoption and validation of this technique to bypass the BBB and target cells in the CNS has led to multiple successful gene therapy preclinical and clinical studies for the treatment of neurological diseases. Positive efficacy and safety data in mice, rats, and non-human primates modeling neurological diseases have piqued excitement and interest around the potential of clinical benefit for these diseases18,19,20,21,22,23. A handful of these studies are now in clinical trials (for example, clinicaltrials.gov identifiers NCT02362438, NCT04737460, NCT03381729, and NCT05518188)3,6. In this article, a simple protocol for lumbar IT injections in mice and rats of different ages is described, without removing CSF, that can be adopted for translational gene therapy projects. This protocol is similar to already-available protocols that are widely adopted; however, there is value in citing these relevant protocols in one place for easy access and reference, along with the accompanying video visuals. This protocol explains the injection for neonatal mice and rats at postnatal day (P) 0-1 and juvenile mice and rats <P10, as well as for mice and rats >P21, with representative results from a successful and unsuccessful lumbar IT injection at P1 in mice. In the discussion, common missteps and specific details that require careful attention while performing this procedure, as well as recommendations on how to practice these injections before starting a preclinical study, are addressed.
The procedures described herein were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Texas Southwestern Medical Center. Wild-type C57BL6/J male and female mice, aged P1-P28, were used for protocols involving mice. Wild-type Sprague-Dawley male and female rats, aged P1-P56, were used for protocols involving rats. Aside from the survival surgery described in section 3, all other procedures are considered to cause only momentary discomfort and do not require the use of anesthetics or analgesics. Individuals should monitor laboratory animals for more than momentary discomfort and seek guidance from their IACUC and veterinary staff on the necessity of anesthetics and analgesics. Details of the reagents and equipment utilized are provided in the Table of Materials.
1. Lumbar IT injection of mice >P21
Figure 1: Schematic of finger and syringe placement for lumbar intrathecal injection in mice and rats. (A) Lateral view of a mouse >P21 showing needle placement and syringe angle transition during a lumbar IT injection. The dashed red oval indicates finger positioning over the mouse's iliac crest. An enlarged view of the spine shows the intrathecal space (blue) with approximate needle placement (green arrow) and the spinal cord (pink). Dorsal view of (B) a mouse >P21, (C) a mouse <P10, and (D) a rat >P21, with reference points for needle placement (green circle), pelvic girdle (dashed yellow oval), and finger placement (dashed red oval). The incision site and retractors are also depicted in (D). Please click here to view a larger version of this figure.
2. Lumbar IT injection of mice and rats <P10
3. Lumbar IT injection of rats >P21
NOTE: There are multiple IT injection procedures described in the literature, ranging from unanesthetized techniques to more extensive surgical approaches14,15. A procedure for direct injection using a minimally invasive technique is described using light anesthesia and a small skin incision. The use of anesthesia, such as isoflurane gas, can help with restraint, relax musculature, and prevent movement during the injection. Making a small incision in the skin over the injection site improves the accuracy of the injection by allowing visualization of the intervertebral space and eliminating the need to puncture through thick skin. Due to the incision, the use of anesthetics and analgesics is required. With practice, it is possible to perform lumbar IT injections in rats older than P21 without anesthesia or an incision at the user's discretion and pending institutional requirements15. Follow institutional guidelines and considerations regarding anesthesia, appropriate analgesics, and survival surgery for laboratory animals.
Although many factors can influence gene therapy vector transduction, histological staining of tissue remains the most accurate method for determining the success of lumbar intrathecal (IT) injections. Broad and even distribution of the gene therapy vector within the central nervous system (CNS) following injection is indicative of a successful procedure. Figure 2C represents a successful injection of a self-complementary AAV9-mediated gene therapy driving weak ubiquitous transgene expression under the JeT promoter at a dose of 1.3 × 1011 vg/mouse in neonatal (P1) mice, 4 months post-injection. RNAscope analysis, using a probe targeting the transgene delivered by the gene therapy, reveals broad distribution in the lumbar spinal cord, cervical spinal cord, and brain. A section of the liver and heart is included to highlight that, even with direct cerebrospinal fluid (CSF) injection, the gene therapy vector can still be distributed to peripheral tissues.
Figure 2: Stained 5-micron tissue sections of mouse CNS and peripheral tissue 4 months post-injection at P1. Red staining indicates transgene expression via RNAscope, and blue counterstaining of nuclei is done via hematoxylin. (A) Injection with a control formulation buffer. (B) Unsuccessful intraparenchymal injection with an AAV9 vector (scAAV9_JeT-hDDX3Xopt-SpA). (C) Successful lumbar IT injection with an AAV9 vector (scAAV9_JeT-hDDX3Xopt-SpA). The gene therapy vector used in (B) and (C) was given at a translationally relevant dose of 1.3E11 vg/mouse. Scale bar: top panel (5 mm); bottom panel (1 mm). Please click here to view a larger version of this figure.
Concentrated expression in the lumbar spinal cord paired with a lack of expression in the brain, as shown in Figure 2B, may indicate an intraparenchymal injection of the spinal cord and should be considered a failed injection. This happens when the needle is inserted too far into the spinal column, past the intrathecal space, and into the spinal cord. Additionally, very low or no expression in the spinal cord and brain (not shown) should also be considered a failed off-target injection, assuming that a vector and dose are being used where broad CNS distribution is expected. This may be due to not inserting the needle far enough or from being lateral to the midline.
The expression patterns observed from successful injections can differ due to the following eight factors as examples: (1) age at injection, (2) pre-existing immunity, (3) infusion rate, (4) gene therapy vector, (5) dose of gene therapy, (6) cell surface proteins, (7) tropism, and (8) if applicable, an appropriate promoter choice to drive transgene expression6. While expression patterns can differ, the broad, even, widespread distribution will be universal as long as the dose is sufficiently high with an effective vector, such as AAV9.
Major limitations of using histological analysis to confirm lumbar IT injection success are the extensive waiting times -- waiting until the end of a study, after necropsies are performed and tissue is collected -- and extensive resources needed if processing tissue from all mice in a large gene therapy study. Unfortunately, our experience has been that immediate and direct indicators of a positive or negative injection can be unreliable; however, the tail flick reflex as the needle enters the intrathecal space is a good indicator of successful positioning in real-time and likely indicates a successful injection. Do not confuse a twitch as the needle pierces the skin in an unanesthetized mouse/rat with a tail flick response to the needle entering the intrathecal space. The use of pharmacological agents such as NMDA, substance P, and lidocaine, either in training or mixed with the experimental injection solution, has been reported to provide a more immediate indication of injection success9,11,24. If considering these agents, it is important to evaluate their compatibility with the gene therapy vector.
The lumbar IT injection is a quick and minimally invasive procedure that reliably delivers a gene therapy vector into the CSF for the treatment of CNS diseases5,6. The procedure is translationally relevant, and the protocol described here details how to perform this route of administration in mice and rats across all ages, from neonates to adults. It is important to define this protocol for mice and rats of all ages, along with providing supporting videos, to help investigators in the adoption of this method for gene therapy administration. Our laboratory's experience is that this protocol can be implemented consistently across multiple users and studies over time18,25,26,27,28,29,30.
There are important differences when performing the lumbar IT injection in younger mice/rats compared to older mice/rats, most notably the angle at which the needle is inserted into the spine and the recommended volume that is injected. Reported lumbar IT injection volumes vary considerably across studies and across species31. Consideration of the injection volume is important to avoid long-lasting elevations in intracranial pressure (ICP), which can compromise the flow of CSF and cerebral blood, cause discomfort, and lead to chronic neurological complications, including hydrocephalus, ischemia, cellular injury, and death32,33. ICP is determined by the volume of the CSF, cerebral blood, and CNS tissue, which cannot be directly correlated to body weight. At normal functioning, the ICP is autoregulated by many factors, including CSF volume, cerebral blood volume, respiration, body position, CSF production rate, and the rate of CSF drainage into the blood33,34. IT injection volumes should, therefore, be determined based on CSF properties (Table 1) instead of body weight25,27,28,30. Recommended volumes to inject at each age in each species are indicated in bold.
IT Injection Volume | Adult CSF Values | |||||||
P0-1 (µL) | P5-7 (µL) | P10 (µL) | >P21 (µL) | Total Volume (µL) | Rate of Production (µL/min) | Turnover (h) | Intracranial Pressure (mm Hg) | |
Mice | 3 | 5 | 5-10 | 5-20 | 30-4025,30 | 0.32-0.3525,30 | 1.7-225,30 | 5.0 +/- 0.528 |
Rats | 5 | 5-10 | 10-30 | 10-200 (20-75) | 15025 | 1.7-2.825 | 2-2.6625 | 8.6 +/- 1.7,27 |
Table 1: Summary of lumbar IT injection volumes for mice and rats at different ages. Bold values are recommended and have been delivered safely. Maximum possible volumes have not been formally evaluated. Additional information about known CSF parameters -- total volume, rate of production, turnover, and intracranial pressure -- in mice and rats is included for reference.
There is a lack of knowledge across the field regarding the upper threshold for one-time bolus IT injection volumes. In adult humans, rats, and mice where CSF volume is known, a 30% increase in total CSF volume does not seem to cause chronic injury or maladies31,33,35,36. The lack of known CSF volumes in juvenile or neonatal mice makes a similar extrapolation impossible. Some groups are beginning to look at CSF volume and production in younger animals37. Until additional investigations in these areas are substantiated, injection volume will continue to be subject to investigator-reported values.
Mice and rats, especially when treated at younger ages or with high injection volumes, may elicit flexing of muscles, extension of limbs, rapid breathing, or temporary hindlimb paralysis that should self-resolve within a few minutes. In extreme cases, acute ICP elevations can cause cardiovascular and respiratory abnormalities, which can be deadly32,33. If any post-procedural abnormalities persist after 24 h, the mice/rats should be removed from the study and humanely euthanized. Persistent hindlimb paralysis can occur if the needle is inserted too far, impacting the spinal cord. This can be due to a common mistake while performing the lumbar IT injection: movement of the syringe after the tail flick while depressing the plunger. Syringe and needle movement should be avoided. If the correct position cannot be acquired with the initial puncture, a second attempt at the same location can be made. If the second attempt is also unsuccessful, try altering the needle position to target the next intervertebral space. Note that multiple needle sticks may lead to leakage of a subsequently successful injection.
Becoming proficient at the lumbar IT injection can take time. To practice the injections as a terminal procedure, follow the above protocol using an approved dye solution, such as Evans blue or 0.2 microns filtered McCormick green dye (Figure 3) or with the pharmacological agents addressed in the Representative Results section. Using dye is recommended to troubleshoot and master the injection because it is easy to tell if an injection was a success or failure within 1 min. Practice with dye is for non-survival procedures only, as animals can develop a reaction to the dye when directly administered to the CNS. This reaction can occur within a minute of a successful injection and is characterized by rapid itching and squirming movements. The animals should be immediately euthanized once this reaction is observed to minimize discomfort. After a successful practice dye injection, dye will stay localized in the spine (no dye in nearby peripheral tissue), and move up the spinal column to the cerebellum, cerebrum, and olfactory bulbs. At P1, the skin is transparent enough that the dye can be seen moving down the spinal column in the tail. If the dye does not reach the brain within a few minutes, the injection fails.
Figure 3: McCormick green dye in the brain after successful practice lumbar IT injections. All brains are from P21 mice injected with 5 µL dye and are shown in the ventral view. Please click here to view a larger version of this figure.
For additional information on relevant associated parameters for preclinical trial design, like viral titer and doses, see previously published reviews3,6,31.
The authors declare no conflicts of interest.
We would like to thank the UT Southwestern AAV Viral Vector Core Facility for manufacturing the AAV9 vector and Yuhui Hu, Research Scientist in the Gray Lab, for processing and staining the tissue presented in Figure 2.
Name | Company | Catalog Number | Comments |
0.2 micron filter | Electron Microscopy Sciences | 67005 | Used to filter dye solution |
0.5 to 10 µL Pipette | Eppendorf | TI13690026 | Used to measure injection solution |
1.5 mL Microtube | Eppendorf | 22364111 | Used to store injection solutions |
10 µL Syringe | Hamilton | 7635-01 | Injection volume should not be less than 10% of syringe capacity |
10 to 100 µL Pipette | Eppendorf | TI13690029 | Used to measure injection solution |
10µl Pipette Tips | USA Scientific Inc | 11203810 | Used to measure injection solution |
100 µL Syringe | Hamilton | 7638-01 | For rat >21 only. Injection volume should not be less than 10% of syringe capacity |
100 µL Pipette Tips | USA Scientific Inc | 11231840 | Used to measure injection solution |
25 µL Syringe | Hamilton | 7636-01 | Ideal for 5-10 µL injections. Injection volume should not be less than 10% of syringe capacity |
27 Gauge Needle(s) | Hamilton | 7803-01 | For rat >21 only. 27 gauge, Small Hub RN Needle, 1 in, point style 4 at 12°, 6/PK |
30 Gauge Needle(s) | Hamilton | 7803-17 | 30 gauge, Small Hub RN Needle, 0.5 in, point style 4 at 12°, 6/PK |
50 µL Syringe | Hamilton | 7637-01 | For rat >21 only. Injection volume should not be less than 10% of syringe capacity |
70% Ethanol | Pharmco | 111000140 | Used to sanitize workspace and equipment |
70% Isopropyl Alcohol Prep Pads | PDI | B60307 | Used to prepare injection site |
Analgesic | For rat >21 only. | ||
Anesthetic (Isoflurane) | Piramal Critical Care | 66794001725 | For rat >21 only. |
Betadine | Purdue Products | 6906606 | For rat >21 only. Used for skin prep |
Control Solution | Injection solution | ||
Dye Solution (green) | McCormick | For practice, non-survival only | |
Gloves | Kimberly-Clark | 19-149-863B | PPE |
Ice bucket with ice | Fisher Scientific | 03-395-150 | Maintain viral vector solution on ice |
Mosquito Forceps (curved or straight) | Fine Science Tools | 13009-12 | For rat >21 only. Used to palpate intervertebral space. |
Needle Holders | Fine Science Tools | 12002-12 | For rat >21 only. Used for skin closure with suture |
Paper Towel | Berkshire | 18-998-123 | Used to restrain adult mice during injection |
Parafilm | StatLab | PM996 | Used to draw solution into syringe |
Retractors | Stoelting | 52124P | For rat >21 only. Used to hold skin incision open |
Scalpel Blade | Fine Science Tools | 10015-00 | For rat >21 only. Used for incision |
Scalpel Blade Handle | Fine Science Tools | 10003-12 | For rat >21 only. Used for incision |
Sterile Syringe | Fisher Scientific | 14-955-459 | Used to filter dye solution |
Surgical Scrub (Skin Prep) | Medline Industries Inc. | MDS098720 | For rat >21 only. Used for skin prep |
Suture or Wound Clips | Stoelting | 50483 | For rat >21 only. Used for skin closure. |
Syringe / Needle Cleaning Solution | Hamilton | 18311 | Can use alternative cleaning solution |
Thumb Forceps | Fine Science Tools | 11019-12 | For rat >21 only. Used throughout surgical approach and closure |
Vector Solution | Injection solution |
Request permission to reuse the text or figures of this JoVE article
Request PermissionThis article has been published
Video Coming Soon
Copyright © 2025 MyJoVE Corporation. All rights reserved