Long-Lasting Suppression of Mechanical Allodynia in Mice by Intrathecal Administration of an Anti-Cell Adhesion Molecule 1 Antibody

Abstract

Objective

Cell adhesion molecule 1 (CADM1), an immunoglobulin superfamily member, is abundantly expressed on nerve fibers. Recently, the anti-CADM1 ectodomain antibody 3E1 has been shown to be potentially useful as a local anesthetic due to its high affinity for subcutaneous nerve fibers. When injected intrathecally into mice, where 3E1 accumulates, and whether it serves as an analgesic was examined.

Methods

The 3E1 injection was detected using indocyanine green-based imaging and immunohistochemistry. As a neuropathic pain model, mechanical allodynia-affected mice were created by spinal nerve ligation (SNL). These mice were administered intrathecally with 3E1, control antibody, or pregabalin, and then the pain threshold of the hind paw was measured using von Frey monofilaments. After SNL, followed by intrathecal injection of 3E1, real-time PCR was conducted to examine the expression of pain-related genes, interleukin (IL)-1β, IL-2, IL-6, IL-10, tumor necrosis factor-α, chemokine (C-C motif) ligand (CCL) 2, CCL7, and CCL11.

Results

Intrathecally, injected 3E1 was localized mainly on the dorsal root ganglion cell bodies and spinal dorsal horn nerve fibers, with no motor side effects. In the von Frey test, 3E1-injected mice exhibited higher pain thresholds for up to 2 weeks compared to control mice. Pregabalin also raised the threshold, but the effect dissipated within 1 day. While SNL increased mRNA expression of all pain-related genes examined in spinal dorsal horn neurons, 3E1 injection suppressed the increased expression of IL-6.

Conclusion

3E1 was suggested to be a potential ganglion-blocking analgesic and spinal anesthetic that acts ten or more times longer than pregabalin.

Highlights of the Study

• Anti-cell adhesion molecule 1 ectodomain antibody 3E1, when administered intrathecally into mice, was localized on dorsal root ganglia and spinal dorsal horns.

• Intrathecal 3E1 reversed mechanical allodynia caused by spinal nerve ligation and suppressed the elevation of interleukin-6 expression in the spinal dorsal horn.

• 3E1 provided much longer analgesia than pregabalin, the first-line drug for chronic pain.

Introduction

Cell adhesion molecule 1 (CADM1) is a member of the immunoglobulin superfamily and is expressed in peripheral nerves on the entire length of nerve fibers and on the cell body plasma membrane [1–3]. We originally produced an antibody, named 3E1, recognizing the extracellular region of CADM1 and injected it subcutaneously into mice, and found that it was widely localized in peripheral nerves and suppressed pain induced by subcutaneous injection of formalin [3]. Although the mechanism of action is not yet fully understood, this antibody is expected to be a new analgesic drug.

In animal experiments to evaluate analgesic effects, a neuropathic pain model is known in which mechanical allodynia is produced in the hind paw of mice by ligation of the spinal nerves [4, 5]. The severity of mechanical allodynia is measured by the plantar pain sensitivity threshold using the von Frey filament (called von Frey test). In these allodynia-affected mice, various genes such as interleukin (IL)-1β, IL-2, IL-6, IL-10, tumor necrosis factor (TNF) α, chemokine (C-C motif) ligand (CCL) 2, CCL7, and CCL11 have increased expression levels in the spinal cord and dorsal root ganglion (DRG) [6–9]. These factors are thought to contribute to the reduction of the plantar pain sensitivity threshold in the von Frey test. While inhibitors against these factors are expected to become analgesic agents, attempts have been made to develop anti-TNFα antibodies into drugs, but these have yet to be used in actual clinical practice [10, 11].

Intrathecal administration of anesthetics and ganglion blockade with anesthetics are widely used techniques in clinical practice. Most agents relieve pain but are associated with motor paralysis [12, 13]. In addition, the analgesic effect lasts at most 1 day [13, 14]. Pregabalin is an α₂δ ligand and a standard drug prescribed for neuropathic pain, but has side effects on the central nervous system, such as drowsiness and dizziness [15, 16]. It has long been desired to develop analgesics with a longer duration of action and without motor paralysis or central nervous system side effects.

In the present study, we administered 3E1 intrathecally to mice and examined the in vivo localization of 3E1 using live imaging and histologic analyses. Next, as a model of neuropathic pain, we created mice affected with mechanical allodynia by ligating the spinal nerve in one limb. Then we injected 3E1 intrathecally into the mice and conducted the von Frey test at various timepoints to measure pain sensitivity thresholds of the plantar feet. Pregabalin was also injected for comparison. Additionally, the expression of pain-related genes in the spinal dorsal horn was examined. We used two strains of mice, C57BL/6 J for antibody detection and ddY for pain-related behavioral assays.

Materials and Methods

Animals

Male ddY mice (6 weeks old, 30–35 g) and C57BL/6 J female mice (15–20 weeks old, 20–25 g) were purchased from Japan SLC Inc. (Shizuoka, Japan). The mice were habituated to the animal facility for at least 1 week before experimentation. The breeding environment was as follows; temperature, 24 ± 1°C; humidity, 55% ± 10%; light-dark cycle, lights on at 07:00–19:00; food and water, freely available. C57BL/6 J mice were used for in vivo detection of antibodies, and ddY mice were used for pain-related experiments. The use of ddY mice was based on our laboratory’s experience in creating behavioral analysis models [17], while we used C57BL/6 J mice for imaging because we plan to conduct future experiments with CADM1 knockout mice (C57BL/6 J strain).

Computed Tomography

Under anesthesia, mice were administered 5 μL of iopamidol (Bayer Pharma, Osaka, Japan) between the L1 spinal nerve via 26 G cannulation (09-049, NIPRO, Osaka, Japan), and contrast-enhanced images were obtained using the LaTheta LCT-200 X-ray computed tomography (CT) system (Hitachi-Aloka, Tokyo, Japan) (pixel size 96 μm, number of slices 100, slice thickness 96 μm, slice interval 192 μm, 360° rotation).

Antibodies

Two antibodies against CADM1 were used; anti-CADM1 C-terminal (S4945, Sigma-Aldrich, Tokyo, Japan; 1:200 for immunohistochemistry), and anti-CADM1 N-terminal, a chicken-mouse chimeric version of 3E1 (chicken IgY, CM004-3; Medical & Biological Laboratories, Nagoya, Japan) (chicken-mouse chimeric 3E1 [cm3E1]), which we originally developed previously [18, 19]. Anti-chicken IgY Fab antibody (AB_2339283) was purchased from ImmunoResearch Laboratories (West Grove, PA, USA; 1:100 for immunohistochemistry). Control mouse IgG1 (DNP-M1) was purchased from ACROBiosystems (Tokyo, Japan).

Fluorescence Labeling and Imaging

An ICG-Labeling Kit-NH₂ (Dojindo, Kumamoto, Japan) was used to conjugate indocyanine green (ICG) to the antibodies according to the manufacturer’s instructions. ICG-labeled cm3E1 or control mIgG1 (15 μg, each) was administered intrathecally to the L5/L6 intervertebral space of mice whose backs had been shaved, and ICG fluorescence was detected at the indicated timepoints using the Pearl Trilogy Small Animal Fluorescence Imaging system (LI-COR, Tokyo, Japan). Mice were temporarily anesthetized before imaging.

Immunohistochemistry

The immunohistochemical procedures have been described in the supplementary material (for all online suppl. material, see https://doi.org/10.1159/000547340).

Spinal Nerve Ligation

To induce peripheral nerve injury, we used the spinal nerve ligation (SNL) model [4, 20] with a slight modification. Briefly, under anesthesia, the right L5 spinal nerve was exposed by removing a small piece of the paravertebral muscles and a part of the right spinous process of the L5 lumbar vertebra. The L5 spinal nerve was then carefully isolated and tightly ligated with 8-0 silk thread. After nerve ligation, the muscle, the adjacent fascia, and the skin were closed with sutures.

Intrathecal Injection

The cm3E1 and control mIgG1 were dissolved in artificial cerebrospinal fluid (NaCl 138 mm, KCl 3 mm, CaCl₂ 1.25 mm, MgCl₂ 1 mm, d-glucose 1 mm) to a final concentration of 3 μg/μL. Pregabalin (Sigma-Aldrich, St. Louis, MO, USA) was dissolved in the same fluid to a final concentration of 2 and 6 μg/μL. The intrathecal injections were carried out according to previously reported procedures [21]. Briefly, a 5-μL volume of each solution was administered to the intervertebral space at the level of the fifth or sixth lumbar vertebra.

Evaluation of Mechanical Allodynia

Mechanical allodynia of the hind paw was assessed by von Frey test with load-weight-calibrated von Frey filaments (Stoelting Co., Wood Dale, IL, USA). Briefly, tactile stimuli producing a 50% likelihood of hind paw withdrawal response (50% paw withdrawal threshold [50% threshold]) were measured using an up-down paradigm [22]. At each measurement time point, stimulation was performed 3 times for each intensity, with at least 5 s between each stimulus. Measures were blinded to the treatment group. The number of mice per experimental group was set at 6, based on the rationale provided in the online supplementary material.

Extraction of RNA and Real-Time Polymerase Chain Reaction

Mice were deeply anesthetized with sodium pentobarbital (100 mg/kg, intraperitoneal injection), exsanguinated by cardiac saline perfusion, and the lumbar dorsal horn (∼1 cm) was rapidly removed and stored at −80°C. Tissues were homogenized with QIAzol reagent (Qiagen Inc., Valencia, CA, USA), and total RNA was isolated using NucleoSpin RNA (TaKaRa Bio Inc., Shiga, Japan). The reagents used for the reverse transcription and real-time polymerase chain reaction (PCR) included PrimeScript™ RT reagent kit (TaKaRa Bio) and SYBR Premix Ex Taq (Takara Bio), respectively. Real-time PCR was performed on a QuantStudio 3 instrument (Applied Biosystems, Waltham, MA, USA). The thermal profile for two-step shuttle PCR was as follows: 40 cycles of 95°C for 5 s and 60°C for 30 s. The gene-specific primers used for PCR are shown in online supplementary Table 1. The expression of mRNAs was normalized to that of GAPDH mRNA for each sample. Experiments were performed in a blinded fashion for the treatment groups and repeated 3 times independently.

Data Analysis and Statistics

Data are reported as the mean ± standard error of the mean. Graphs were generated and statistical analyses were conducted using SigmaPlot Software (ver. 14.5, Systat Software, Ltd., Chicago, IL, USA). Statistical comparisons between groups were performed using one-way analysis of variance followed by the Dunnett or Tukey test. For the analyses of mechanical thresholds, we employed the Mann-Whitney U test. The p value <0.05 was considered significant.

Results

In vivo Live Imaging and Histological Detection of Intrathecally Injected 3E1

First, to confirm the intrathecal injection technique in mice, a 28-gauge needle was inserted into the L1/L2 intervertebral space, and contrast medium was injected. CT imaging revealed that the contrast agent was correctly administered intrathecally (Fig. 1a).

Fig. 1.

Live imaging of 3E1-injected intrathecally into mice. a Upper panel; X-ray transmission image of a mouse. The contrast agent was injected intrathecally between lumbar vertebrae L1 and L2 (indicated by an arrowhead). The scale on the lower border of the picture indicates the slice number of the CT scan. Lower panel; CT scan image. Three images numbered 64, 68, and 81 are shown. The contrast agent is visible in white. b Chicken-mouse chimeric 3E1 (cm3E1) and control mouse IgG1 (mIgG) labeled with ICG (15 μg each) were administered intrathecally to back-shaved mice at the L5/L6 intervertebral space (indicated by arrowheads in the leftmost panel). ICG fluorescence was detected at the indicated timepoints. d0, the immediately after injection; d1–d7 and d23, days 1 to 7 and 23 (d1 means 24 h after the injection). On d23, the mouse was skinned prior to imaging; the framed area is enlarged in the inset.

Next, cm3E1 and control mouse IgG1 were labeled with ICG and were administered intrathecally to mice at the L5/L6 intervertebral space, with the allodynia model experiment to be performed later. The localization of antibodies was examined by detecting ICG with the Pearl system in live mice with shaved back hair. cm3E1 was widely distributed in the thoracic and lumbar spinal cord cavities the day after injection and remained in the spinal cord cavity 1 week later (Fig. 1b). On the other hand, control mIgG1 was weakly found just around the injection site on the day after administration, and was not detected anywhere after the second day (Fig. 1b). On day 23, the mice were skinned and observed. cm3E1 was still detected within the spinal cord (Fig. 1b, d23). In cm3E1-injected mice, no obvious behavioral abnormalities (increase or decrease in spontaneous locomotion, abnormal gait, etc.) or weight loss were observed during the observation period.

At 5 h after antibody administration, immunohistochemistry was conducted on the mouse trunk section and revealed that intrathecally injected cm3E1 was not diffusely distributed throughout the spinal cord but was localized and accumulated at specific sites. The main sites were DRG and the superficial part of the spinal cord (Fig. 2a, b). In the DRG, CADM1 was detected on the surface of ganglion cell body and the nerve fibers, whereas cm3E1 was colocalized to the ganglion cell body surface but not to the nerve fibers (Fig. 2b). In the spinal cord, 3E1 was localized mainly in the superficial layers of the dorsal horn, where CADM1 was abundantly expressed (Fig. 2b). cm3E1 accumulation was also found in the DRG at the levels of L4–L6 and in some fiber bundles in the cauda equina (Fig. 2a).

Fig. 2.

Histological detection of 3E1 injected intrathecally into mice and comparison with CADM1 localization. Mice were injected intrathecally with chicken-mouse chimeric 3E1 (cm3E1; 15 μg) into the L5/L6 intervertebral space, and immunohistochemistry was performed 5 h later. a Transverse plane sections of lumbar vertebrae at various positions (L1–L6) were immune-stained reddish brown with anti-chicken IgY Fab antibody. Scale bar = 500 μm. b Three serial sections of lumbar vertebra L2 were cut; two were immune-stained with either anti-chicken IgY Fab for cm3E1 detection (center) or anti-CADM1 (right) antibody. The remaining one was H&E stained (left). An overall view of the dorsal root ganglion (DRG) is presented in the upper panel, and the framed area within the upper photograph is enlarged in the middle panel. The inset shows staining of CADM1 on nerve fibers (some of which are indicated by arrowheads). The lower panel shows the lumbar spine, and the framed area is enlarged in the inset. The dotted line in each inset indicates the boundary of the substantia gelatinosa (g) and nucleus proprius (N) of the spinal dorsal horn. Scale bar = 100 μm. C, central canal; SAS, subarachnoid space; VB, vertebral body; SP, spinous process; TP, transverse process.

Long-Lasting Suppression of Mechanical Allodynia by cm3E1

In a mouse model of neuropathic pain in which unilateral spinal nerves at the L5 level were ligated, mechanical allodynia, i.e., a reduction in the paw withdrawal threshold to mechanical stimuli, was observed on the ipsilateral hind paw and persisted for at least 4 weeks, whereas no change was observed in the response threshold in the contralateral (non-ligated) hind paw (Fig. 3a). There were no complications related to the surgical procedure itself.

Fig. 3.

von Frey test of mechanical allodynia-affected mice after a single intrathecal injection of 3E1. a After ligation of unilateral spinal nerves at the L5 level in mice (n = 6), the paw withdrawal threshold (y-axis) was measured in the ipsilateral (closed circle) and contralateral (open circle) hind paw over 4 weeks (x-axis) using von Frey filaments. On day 0, measurements were taken before SNL. b, c Two weeks after SNL, mice (n = 6 for each group) were injected intrathecally either with chicken-mouse chimeric 3E1 (cm3E1; 15 μg) or control mouse IgG1 (mIgG; 15 μg) (b), or with pregabalin (10 or 30 μg) or vehicle alone (c) into the L5/L6 intervertebral space, and then the paw withdrawal threshold (Y axis) was measured in the ipsilateral hind paw over the time shown on the X axis using von Frey filaments. At 0 h, measurements were taken before intrathecal injection. Each point represents the mean value, and the standard error is indicated by the vertical line. The dotted horizontal lines in B and C indicate the threshold of the contralateral hind paw. *, p < 0.05 when compared to contralateral (a), mIgG (b), and vehicle (c) values by the Mann-Whitney U test. i.t., intrathecal.

Immediately after confirming the establishment of mechanical allodynia (on day 14), the effects of intrathecal injection of cm3E1 and its control (mIgG1) were examined. As a positive control, we also examined the effects of pregabalin, an α₂δ ligand and a standard drug prescribed for neuropathic pain. In cm3E1-injected mice, the paw withdrawal threshold was nearly normal at 5 and 24 h, remained relatively high for 10 days, and then gradually declined at 2 and 3 weeks (Fig. 3b). In contrast, the thresholds remained depressed throughout the 3 weeks in mice injected with control mIgG1 (Fig. 3b). There was no change in the threshold of the contralateral hind paw in either group (data not shown). Intrathecal administration of pregabalin also showed a dose-dependent relieving effect on mechanical allodynia, which peaked 2 h after administration and then gradually weakened (Fig. 3c). No analgesic effect was observed 24 h after administration. Normal observation found no toxicity associated with pregabalin.

cm3E1 Suppresses IL-6 mRNA Elevation in the Spinal Dorsal Horn of Mechanical Allodynia Mice

We examined the expression of pain-related genes in the dorsal horn of the spinal cord 5 h after intrathecal administration of cm3E1 or control mIgG1 to mice with mechanical allodynia caused by SNL. There were no differences in mRNA levels of the housekeeping gene GAPDH between experimental groups, and expression levels of pain-related genes were relative to GAPDH. In comparison with sham-operated mice, the expression of all eight genes examined in this study (IL-1β, IL-2, IL-6, IL-10, TNFα, CCL2, CCL7, and CCL11) was increased in mice treated with control mIgG1 after SNL (Fig. 4). The expression of IL-1β, IL-2, IL-10, TNFα, CCL2, CCL7, and CCL11 mRNA was increased also by cm3E1 treatment after SNL, while the increase in IL-6 mRNA expression was significantly suppressed by cm3E1 treatment (Fig. 4).

Fig. 4.

Effects of intrathecal injection of 3E1 on pain-related gene expression in the spine of mice with mechanical allodynia. RNA was extracted from the dorsal horn of the spinal cord 5 h after intrathecal injection of cm3E1 (gray) or control mIgG1 (black) into mice that had undergone spinal nerve ligation (SNL) and was also from the dorsal horn of mice that did not receive intrathecal injection after sham operation (white), and then real-time RT-PCR was used to quantify the expression of the eight genes indicated here using specific primers listed in online supplementary Table 1. Each expression level is expressed relative to GAPDH levels. Each bar represents the mean value, and the standard error is indicated by the vertical line. * and ^#, p < 0.05 when compared to sham and SNL + mIgG values, respectively, using ANOVA followed by the Dunnett or Tukey test. ANOVA, analysis of variance.

Discussion

In the present study, we found that intrathecally injected cm3E1 accumulated in the DRG cells and the dorsal horn of the spinal cord, consistent with the high expression of CADM1 in these cells. We previously reported that subcutaneously injected 3E1 appeared to accumulate preferentially in unmyelinated dermal nerve fibers because of the physical accessibility of the antibody to CADM1 [3]. Since DRG cells and neurons of the superficial layers of the spinal dorsal horn are not accompanied by myelin sheaths [23], the same mechanism may contribute to the accumulation of intrathecal cm3E1 in these cells.

Intrathecal injection of cm3E1 had a dramatic therapeutic effect on mechanical allodynia in mice. As we discussed previously [3], this effect may be due primarily to two actions of cm3E1: elevation of firing thresholds in neuronal cell bodies and delayed conduction of stimuli on nerve fibers. Importantly, 3E1 accumulation is fairly well confined to the DRG and the superficial layers of the spinal dorsal horn, probably because CADM1 is preferentially expressed at these sites (Fig. 2). Since neurons at these sites are sensory [23, 24], it is unlikely that intrathecal 3E1 would induce motor paralysis. Indeed, in mechanical allodynia-affected mice treated with cm3E1, no motor paralysis or other behavioral abnormalities were observed in the leg that did not undergo nerve ligation.

Another point worth noting is that the analgesic effects are very potent and long-lasting. Various reagents have been investigated for the treatment of allodynia by intrathecal administration in rodents. These include existing anesthetics, novel compounds, peptides, nucleic acids, and plant extracts [8, 25–27]. To our knowledge, no other drug is as doubly potent as 3E1, which produces an almost complete remission of allodynia with a single intrathecal dose, and the effect lasts for more than a week. Actually, pregabalin was effective within 1 day, meaning that the duration of action of 3E1 was ten or more times longer than that of pregabalin. The fact that 3E1 reaches the intracranial ventricles (Fig. 1) deserves attention in terms of side effects: CADM1 is expressed in the choroid plexus (unpublished data), and 3E1 binding may cause abnormalities in the physiology of the choroid plexus. This is an issue for further investigation.

Although the mechanism of why cm3E1 produces analgesia remains to be elucidated, we have now found that cm3E1 is involved in IL-6 gene expression in the spinal cord dorsal horn neurons. It is well known that elevated IL-6 expression contributes to the lowering of the paw withdrawal threshold in mechanical allodynia-affected mice [6, 28]. In the present study, we found that intrathecal cm3E1 completely suppressed this increase in IL-6 expression in the spinal dorsal horn cells. Although the mechanism is still unclear, the latest report by Koster et al. [29] is noteworthy. They found that CADM1 is critical for maintaining the function of the mechanosensitive ion channel PIEZO1. It is possible that when cm3E1 induces dysfunction and decreased expression of CADM1, intracellular influx of calcium ions via PIEZO1 is reduced, failing to induce IL-6 gene expression, as Sun et al. [30] showed in their model of cardiac ventricular remodeling. This notion is only speculation and requires further molecular-level analysis.

Limitations of this study are described here. There is insufficient data about the safety of this antibody, particularly with regard to adverse effects in the central and peripheral nervous systems. There is limited information about sex differences, strain differences, or dose-dependence in the efficacy of the antibody. Regarding sex differences, we examined them in our previous formalin test in mice and found no sex differences in the analgesic effect of subcutaneously injected 3E1 [3].

Conclusion

The cm3E1 is a promising long-acting analgesic seed. In comparison with pregabalin, cm3E1 was 10-fold or longer in duration of action. Although its safety needs to be fully confirmed, this antibody is expected to become widely used in clinical practice as a spinal anesthetic and ganglion blocker.

Acknowledgments

The authors thank Pharma Foods International.

Statement of Ethics

The animal experiments were carried out according to the International Association for the Study of Pain guidelines. The Animal Care Committee of Kindai University and the University of Toyama approved the animal experiments (approval Nos. KAME-2022-084, A2017ENG-4, and A2020ENG-3).

Conflict of Interest Statement

Akihiko Ito received financial support from Pharma Foods International Co., Ltd., Kyoto, Japan. The other authors declare no conflict of interest.

Funding Sources

This study was supported by the Japan Society for the Promotion of Science Kakenhi (23K06494 to M.H. and 21K06978 to A.I.), 2024 Kindai University Research Enhancement Grant (KD2401 to A.I.), and Advanced Research and Development Programs for Medical Innovation under grant No. JP23ym0126809 (to A.I.).

Author Contributions

Ichiro Takasaki: writing – original draft, investigation, formal analysis, supervision, methodology, and conceptualization. Fuka Takeuchi: investigation, formal analysis, writing – review and editing, and visualization. Man Hagiyama: investigation and funding acquisition. Ryohei Miyamae, Yuto Mochizuki, Hayate Kinoshita, Azusa Yoneshige, and Takao Inoue: investigation. Tetsuo Narumi: investigation, methodology, and conceptualization, and Akihiko Ito: writing – original draft, supervision, methodology, funding acquisition, and conceptualization. All authors approved the final manuscript.

Funding Statement

This study was supported by the Japan Society for the Promotion of Science Kakenhi (23K06494 to M.H. and 21K06978 to A.I.), 2024 Kindai University Research Enhancement Grant (KD2401 to A.I.), and Advanced Research and Development Programs for Medical Innovation under grant No. JP23ym0126809 (to A.I.).

Data Availability Statement

The data that support the findings of this study are available from the corresponding author on reasonable request.