Earlier onset of chemotherapy-induced neuropathic pain in females by ICAM-1-mediated accumulation of perivascular macrophages.

INTRODUCTION

Severe chemotherapy-induced neuropathic pain (CINP) often hinders the completion of chemotherapy, reducing treatment effectiveness and substantially affecting both patients’ quality of life and overall survival ( 1 ). Bortezomib (BTZ) is a primary chemotherapeutic agent for multiple myeloma (MM), and BTZ-induced peripheral neuropathy is recognized as one of the most severe, unpredictable, and potentially permanent nonhematological side effects of chemotherapy ( 2 ). Large clinical studies report an incidence of BTZ-induced peripheral neuropathy ranging from 31 to 64%, with neuropathic pain occurring in 25 to 80% of cases, potentially compromising antineoplastic outcomes ( 3 ). Studies have noted sex differences in pain ( 4 ). For example, women exhibit a higher prevalence of pain compared to men in most of chronic neuropathic pain conditions ( 5 ). While research on gender disparities in neuropathic pain primarily focuses on the pathogenesis, fewer efforts have been made to investigate the timing of onset. It remains unclear whether a sex disparity exists in the onset timing of BTZ-induced neuropathic pain.

Macrophages are widely present in the body and play a key role in the body’s defense against pathogens and foreign substance. Evidence also indicated that monocytes/macrophages in the circulating blood are recruited to the injury site to participate in neuropathic pain induced by nerve injury ( 6 ), and the infiltration of circulating CX3CR1 + macrophage into the sciatic nerve is involved in chemotherapeutic vincristine-induced neuropathic pain ( 7 ). Moreover, macrophage can infiltrate the nervous system and contribute to neuropathic pain ( 8 ). For example, the infiltrated macrophage in the dorsal root ganglion may release the cytokine/chemokine [e.g., interleukin-6 (IL-6), tumor necrosis factor–α, and CC-chemokine ligand 2 (CCL2)], which results in the peripheral sensitization and mediates the neuropathic pain ( 9 , 10 ). In addition to macrophages in the circulating blood, tissue-resident macrophages also play important roles in both healthy and disease settings ( 11 ). In the central nervous system (CNS), tissue-resident macrophage includes microglia and border-associated macrophages ( 12 ). Brain perivascular macrophages, a type of border-associated macrophages, are located at the interface between the blood vessels and the brain parenchyma. Accumulating evidence suggests that these macrophages are involved in the pathogenesis of various diseases such as neurodegenerative diseases and cerebrovascular dysfunction ( 11 ). Under physiological conditions, peripheral macrophages do not typically enter the nervous system. However, perivascular macrophages allow peripheral replenishment through the infiltrating circulating monocytes during pathological states or following external interventions (chemotherapy or specific ablation) ( 13 ). Whether BTZ application induces the macrophage infiltrating into the space surrounding the spinal cord vessels remains unclear. Moreover, a recent study showed that nerve injury promoted the circulating monocytes/macrophages to infiltrate into the hippocampal perivascular space, which led to the neuroinflammation and the cognitive impairment ( 14 ). In the current study, we found that the application of BTZ promoted the infiltration of circulating macrophage into the spinal perivascular space. However, it remains unclear whether these spinal perivascular macrophages, recruited from circulating monocytes/macrophages, contribute to the sex difference observed in BTZ-induced neuropathic pain. While initial research on pain mechanisms primarily focused on neuronal function, recent attention has shifted toward the interactions between nonneuronal cells and neurons ( 15 , 16 ). Notably, the mechanisms by which infiltrated perivascular macrophages regulate the neuronal activity to mediate the sex difference of neuropathic pain remain unclear.

In the present study, we observed an earlier onset of neuropathic pain in females compared to males in both clinical patients and rodent models treated with BTZ. We hypothesized that BTZ treatment promoted macrophage infiltration into the spinal cord perivascular space, contributing to the earlier onset of neuropathic pain in females. Clarifying the mechanism behind macrophage infiltration is crucial for the precise prevention and treatment of BTZ-induced neuropathic pain in females.

RESULTS

Earlier onset of sensory disorders in females with MM undergoing BTZ treatment is related to monocyte/macrophage proliferation and migration

A retrospective analysis was performed in the cohort of patients with MM at the First Affiliated Hospital of Sun Yat-sen University between 2014 and 2021. Among 681 patients with MM, 211 patients (including 81 females and 130 males), who received BTZ-based combination therapy, were ultimately included in the study according to the predefined inclusion and exclusion criteria. Among 211 patients, 71 showed symptoms of numbness, pain, and other sensory disorders, of which 27 were females and 44 were males ( Fig. 1A and table S1). There were no significant differences in basic characteristics, including demographic factors, immune classification, and Revised International Staging System (R-ISS) stage, between females and males (table S2). Next, we compared the differences in the sensory disorders between males and females by the ordinal multiple logistic regression analysis and found that the risk of late onset of sensory disorders in male patients with MM was 1.315 ( P = 0.009) times higher than that in female patients following BTZ treatment ( Table 1 ). These results suggest that sensory disorder may occur earlier in females after BTZ treatment compared to males. Next, the statistical analysis further confirmed that the incidence of sensory disorder in females was significantly higher than that in males after the first cycle of treatment with BTZ ( Fig. 1B ), which included four sessions of BTZ treatment. Next, we observed the number of BTZ treatment sessions for sensory disorder occurrence in both females and males and found that the number of treatment sessions for BTZ-induced sensory disorders in females was significantly lower than that of males ( Fig. 1C ). To further confirm the earlier onset of sensory disorder in females, an additional retrospective analysis was performed in a separate cohort of patients with MM and BTZ treatment from the Third Affiliated Hospital of Sun Yat-sen University between 2010 and 2020 (table S3). Among the 40 patients (including 23 females and 17 males) who showed symptoms of pain and other sensory disorders, the females showed a higher proportion of sensory disorder after the first cycle of treatment with BTZ ( Fig. 1D ) and the lower number of BTZ treatment sessions for sensory disorder occurrence ( Fig. 1E ). These findings indicated that female patients with MM exhibited the earlier onset of sensory disorders following BTZ treatment.

We further confirm the sex difference in the chemotherapeutic BTZ-induced neuropathic pain in mice. The result showed that BTZ induced a significant mechanical allodynia on days 7, 10, and 14 in female and male mice ( Fig. 2A ). Consistent with findings in patients, the withdrawal threshold was significantly decreased on the third day in female mice rather than male following BTZ treatment ( Fig. 2A ). We further validated that BTZ treatment decreased the mechanical withdrawal threshold on days 3 and 4 in female mice compared with the male group ( Fig. 2B ). Electrophysiological results revealed that the number of action potential was significantly increased on day 3 following BTZ treatment in female, but not in male ( Fig. 2C ), indicating an earlier occurrence of neuronal hyperexcitability in female mice.

To elucidate the mechanisms underlying the earlier onset of neuropathic pain in female mice in contrast to male mice, we extracted the spinal dorsal horn proteins on the third day after BTZ treatment for cytokine quantification with cytokine protein chip detection. The results revealed that there were significant changes in the expression of 29 cytokines in female mice treated with BTZ on the third day compared to the control group of female mice (fig. S1A), and there were significant changes in the expression of 15 cytokines in male mice on the third day following BTZ treatment compared to male control mice (fig. S1B). Then, we performed the Gene Ontology (GO) enrichment analysis for the changed cytokines in female and male mice following BTZ treatment. The results showed that the altered cytokines in the female mouse group were predominantly involved in leukocyte migration and proliferation, whereas no clear pattern was observed in the male mice ( Fig. 2D ). Furthermore, the GO analysis revealed that the changed cytokines in the female group were enriched in the pathway for immune cells migration and chemotaxis compared with the male group following BTZ treatment (fig. S1D). Flow cytometry analysis further revealed an increase in CD45 high /CD11b + monocytes/macrophages in the spinal dorsal horn of female mice, whereas male mice did not exhibit significant changes on the third day following BTZ treatment ( Fig. 2, E and F ). Together, these data indicated that the number of CD45 high /CD11b + cells in the lumbar enlargement of the spinal cord increased in female mice during the early stages, which suggests an increase in macrophages following BTZ treatment.

Circulating monocytes contribute to the increased perivascular macrophages in the spinal cord following BTZ treatment in female mice

To further exclude the activated microglia as the phenotype of these CD45 high /CD11b + cells, we stained the spinal cord slices with TMEM119 antibody (a microglial-specific marker) and F4/80 antibody (the mature macrophage marker). Double immunofluorescence revealed a population of the elongated macrophages in the dorsal horn of the spinal cord that highly expressed F4/80 but not TMEM119 ( Fig. 3A ). Immunofluorescence results showed that the number and the soma area of TMEM119 + cells (a ramified morphology cell) did not change, while the number of F4/80 high /TMEM119 − cells significantly increased on the third day following BTZ treatment in the female-treated group compared with the control group ( Fig. 3, B and C ). In male mice, neither microglia activation nor F4/80 up-regulation occurred on the third day following BTZ treatment ( Fig. 3, A to C ). These indicate that the increased CD45 high /CD11b + cells in females, as previously mentioned, may be macrophages marked by F4/80. Notably, F4/80 high cells were predominantly distributed in the perivascular spaces but not in the spinal parenchyma ( Fig. 3D ). Perivascular macrophages, as a type of border-associated macrophages, are involved in the pathogenesis of various diseases. Immunofluorescence analysis revealed a marked up-regulation of F4/80 + signals localized to the spinal cord vasculature on day 3 post-BTZ treatment in female mice, while only a very small amount of F4/80 + signals was detected in the control group ( Fig. 3, D and E ).

To determine whether these macrophages might potentially originate from the peripheral monocytes, the circulating monocytes were depleted by intravenous injection of clodronate liposomes (CLLs; 15 ml/kg). The results showed that the CLL injections inhibited the macrophage up-regulation in the vascular of spinal cord ( Fig. 3, F and G ) and alleviated the increase in spinal cord lamina II neuron firing and mechanical allodynia on the third day after BTZ treatment ( Fig. 3, H and I ). These results indicated that the recruitment of peripheral macrophages contributed to the earlier onset of BTZ-induced neuropathic pain in female mice.

Up-regulation of ICAM-1 in the dorsal horn vascular promotes the circulating monocyte transmigration into perivascular space

To elucidate the mechanism of the earlier onset of neuropathic pain in female mice, we separately analyzed the differences in cytokine changes between the female treatment group and the female control group and between the female and male treatment groups. The results revealed that there were significant changes in the expression of 29 cytokines in female mice treated with BTZ on the third day compared to the control group of female mice (fig. S1A), while there were significant changes in the expression of 34 cytokines in female mice compared to male mice on the third day following BTZ treatment (fig. S1C). The cross-analysis of two expression profiles indicates that nine cytokines may be involved in the earlier onset of BTZ-induced neuropathic pain in female mice ( Fig. 4A ). Furthermore, quantitative polymerase chain reaction (qPCR) assay confirmed a significant increase in intercellular adhesion molecule–1 ( Icam1 ) mRNA expression in the female mice on days 3 and 4 compared to the female control group or the male mice with BTZ treatment ( Fig. 4B and fig. S2A). Cytokine protein chip detection and Western blot showed that the ICAM-1 protein level in the female mice was enhanced on day 3 relative to that of the control female group or the BTZ-treated male group ( Fig. 4C and fig. S3A). Immunofluorescence revealed the abundant expression of ICAM-1 on the spinal cord vascular endothelium, which was labeled by CD31 antibody ( Fig. 4D ). Preoral administration of an ICAM-1 inhibitor A205804 (10 mg/kg per day) attenuated the mechanical allodynia induced by BTZ on days 3 and 4 in female mice, while it had no obvious effect on painful behavior of male mice ( Fig. 4, E and F ).

Next, we investigated the impact of ICAM-1 on macrophage infiltration in the spinal dorsal horn. The results from the high-resolution microscope showed that ICAM-1 was colocalized with macrophage (F4/80-positive signal) near the blood vessels (CD31 + signal) ( Fig. 4G ). To further determine the role of ICAM-1 in monocyte migration, we performed a tail vein injection of AAV-CAG–ICAM-1 to induce targeted overexpression of ICAM-1 specifically in endothelial cells (fig. S3B). Western blot results showed a significant increase in ICAM-1 protein expression in the group that received tail vein injection of AAV-CAG–ICAM-1 compared to the control virus group (fig. S3C). Consecutive image and three-dimensional reconstruction showed that macrophages were penetrating the vascular walls following overexpression of ICAM-1 by intravenous injection of AAV-CAG–ICAM-1 ( Fig. 4H ). The above results suggested that the adhesion molecular ICAM-1 promoted the infiltration of macrophage in the spinal dorsal horn in female mice.

Up-regulation of ICAM-1 in the dorsal horn vascular induced macrophage accumulation and neuropathic pain in female mice

ICAM-1 is involved in the transendothelial migration of leukocytes by triggering endothelial cell contraction and by weakening the bonds of junctional adhesion molecules ( 17 ). To test the contribution of ICAM-1 to BTZ-induced macrophage infiltration and neuropathic pain, we treated BTZ treatment female mice three times with anti–ICAM-1 small interfering RNA (siRNA) or scramble control starting before the injection of BTZ. Immunofluorescence results showed that the intrathecal injection of ICAM-1 siRNA, which obviously decreased the spinal ICAM-1 expression (fig. S3D), significantly inhibited the increase in F4/80 + cells near the vascular endothelial cell induced by BTZ in female mice ( Fig. 5, A and B ). Moreover, consecutive intrathecal injection of ICAM-1 siRNA attenuated the hyperexcitability of spinal cord lamina II neurons on day 3 and the mechanical allodynia induced by BTZ on days 3 and 4 in females ( Fig. 5, C and D ). Furthermore, we performed a tail vein injection of AAV-CAG–ICAM-1 to induce targeted overexpression of ICAM-1 specifically in endothelial cells. Immunofluorescence results showed that intravenous injection of AAV-CAG–ICAM-1 resulted in an increased population of F40/80 + cells in female mice ( Fig. 5, E and F ). Overexpression of ICAM-1 significantly induced the hyperexcitability of spinal cord lamina II neurons and mechanical allodynia in females ( Fig. 5, G and H ). In addition, we explored the role of AAV-CAG–ICAM-1 on the macrophage infiltration and pain behavior in male mice. The CCR2 receptor, which mediates the excretion of monocytes from the bone marrow to the circulation, is also an important marker for distinguishing circulating blood monocytes from microglia. The results showed that the specific overexpression of endothelial cell ICAM-1 did not induce the increases in CCR2 expression and mechanical allodynia in male mice (fig. S4). Together, these results suggest the necessity of ICAM-1 in BTZ-induced macrophage infiltration and neuropathic pain in female mice.

IL-1α released by endothelial cells promotes ICAM-1–mediated macrophage infiltration in female mice

The migration of immune cells toward the gradient of chemokines is a key process of infiltration. To elucidate the mechanism of ICAM-1–mediated macrophage infiltration, we reanalyzed the levels of chemokine/cytokine that have chemotactic effects on monocytes and found that levels of IL-1α, CCL5, and CCL1 were significantly elevated in the female treatment group compared with the control group (fig. S5, A and B). qPCR results confirmed that the level of Ccl1 and Il1a , not that of Ccl5 , was significantly increased on day 3 following BTZ treatment in female mice ( Fig. 6A ). In addition, the levels of Il1a , Ccl5 , and Ccl1 remained unchanged on day 3 following BTZ treatment in male mice (fig. S5C). Immunofluorescence staining showed that endothelial cells expressed IL-1α, but not CCL1, in female mice treated with BTZ ( Fig. 6, B and C ), suggesting that the up-regulated IL-1α in endothelial cells might participate in the macrophage infiltration. Furthermore, intrathecal injection of IL-1α siRNA, which decreased the IL-1α expression (fig. S3E), inhibited the up-regulation of the number of F4/80 + cells near the spinal cord blood vessels and neuropathic pain in female mice on day 3 following BTZ treatment ( Fig. 6, D to F ). Although single intraspinal injection of recombinant IL-1α induced mechanical hypersensitivity, it did not increase the number of spinal cord macrophages in female naïve mice (fig. S5, D to F).

To clarify the role of IL-1α in the BTZ-induced macrophage infiltration, we further explored the expression of interleukin-1 receptor type 1 (IL-1R1). The results showed that there was a marked expression of IL-1α receptor IL-1R1 on the macrophages around the blood vessels ( Fig. 6G ), and intrathecal injection of IL-1R1–neutralizing antibody inhibited the increased number of the macrophages around the blood vessels induced by BTZ in female mice ( Fig. 6, H and I ). Moreover, compared with the control group, intraspinal injection of recombinant IL-1α significantly up-regulated the number of spinal cord macrophages and decreased the mechanical withdrawal threshold in female mice with intravenously injected AAV-CAG–ICAM-1 ( Fig. 6, J to L ), and preintrathecal injection of IL-1R1–neutralizing antibody inhibited the increased number of the macrophages around the blood vessels and the mechanical allodynia induced using recombinant IL-1α in AAV-CAG–ICAM-1 mice ( Fig. 6, J to L ). These results suggested that the IL-1α released by endothelial cells promotes ICAM-1–mediated macrophage infiltration in female mice following BTZ.

The infiltrated macrophage-released CCL1 directly enhanced the neuron excitability and contributed to the neuropathic pain in female mice following BTZ treatment

Although CCL1 was not expressed in endothelial cells, its expression was substantially up-regulated on day 3 following BTZ treatment ( Fig. 6, A and C ). Immunostaining showed an obvious CCL1 expression on the infiltrated macrophage on day 3 following BTZ treatment in female mice ( Fig. 7, A and B ). Next, we found that the CCL1 receptor CCR8 was expressed on the astrocytes and neurons in the spinal dorsal horn ( Fig. 7, C and D ). Intrathecal injection of CCL1-neutralizing antibody prevented the enhanced neuronal excitability and the mechanical allodynia on day 3 after BTZ treatment in the female mice ( Fig. 7, E and F ). Furthermore, local application of recombinant CCL1 significantly induced the increase in action potential number in naïve slice and the mechanical allodynia in naïve mice ( Fig. 7, G and H ). Furthermore, preapplication of CCR8 antagonist R243 inhibited the neuronal hyperexcitability induced by CCL1 up-regulation ( Fig. 7I ). R243 alleviated the mechanical allodynia induced by BTZ or CCL1 in female mice ( Fig. 7, J and K ). To exclude the role of astrocyte CCR8 in the up-regulated neuronal excitability, we specifically knocked down the CCR8 expression on the astrocytes by intraspinal injection of AAV2/5-GfaABC1D-CCR8 short hairpin RNA (shRNA) (fig. S6A). The results showed that the application of CCL1 induced an increase in neuronal excitability and the mechanical allodynia in female mice with knockdown of astrocyte CCR8 ( Fig. 7, L and M ). In addition, intraspinal injection of AAV2/5-GfaABC1D-CCR8 shRNA attenuated the BTZ-induced neuropathic pain (fig. S6B). These results suggested that the released CCL1 by the infiltrated macrophage can directly increase the neuron excitability and contribute to the neuropathic pain in female mice.

CCL1 directly activated astrocytes and contributed to the neuron hyperexcitability and mechanical allodynia

Because the CCL1 receptor CCR8 was also expressed in the astrocytes ( Fig. 7D ), immunofluorescence staining showed notable morphological changes in the astrocytes closely adjacent to the perivascular macrophages after 3 days of BTZ treatment ( Fig. 8A ). Furthermore, we also found that BTZ treatment notably activated the astrocyte manifested as cell body enlargement and protrusion thickening, which is colocalized with the CCR8 (fig. S7). To explore the role of CCL1/CCR8 pathway in astrocytes in the earlier onset of neuropathic pain in female mice, we specifically labeled the astrocyte by intraspinal injection of AAV2/5-GfaABC1D-CRE in female mice, and the two-photon astrocyte Ca 2+ imaging results showed that application of CCL1 significantly enhanced intracellular calcium signaling of astrocytes (labeled with SR101) in the spinal cord slices with the genetically encoded intracellular indicator GCaMP6s ( Fig. 8, B to D , and fig. S6C). Furthermore, preincubation of R243 reduced the CCL1-induced increase in intracellular calcium signaling in astrocytes ( Fig. 8, E to G ). To determine whether the activation of astrocytes regulated by CCL1 originates from neuronal hyperactivity, we blocked action potentials by incubation of the voltage-gated sodium channel blocker tetrodotoxin (TTX; 1 μM). The results showed that the increased calcium signaling of astrocytes to CCL1 was not affected by TTX ( Fig. 8, H to J ), suggesting that CCL1-mediated astrocyte activation did not depend on the neuronal activity. Next, we explored whether the astrocyte activation can affect the neuronal activity. In the Cre-dependent chemogenetic virus (AAV-DIO-hM3Dq-mCherry) and AAV2/5-GfaABC1D-CRE into the spinal cord of naïve female mice (fig. S6D), we found that the intensity of a Ca 2+ signal indicator fluo-4 in astrocytes was significantly increased after clozapine N-oxide (CNO) incubation ( Fig. 8, K to M ). Moreover, following the astrocyte activation with CNO, the action potential frequency of neurons and the mechanical allodynia significantly enhance in female mice ( Fig. 8, N and O ). These results suggested that CCL1 may indirectly enhance the neuronal excitability via astrocyte activation to contribute to the earlier onset of neuropathic pain in female mice.

DISCUSSION

Although it is well known that females can respond differently to pain ( 18 , 19 ), research on pain has mainly focused on male animals for decades. A survey shows that women are more likely to experience various types of pain, including lower back pain, neck pain, and migraines ( 20 ). Studies on the sexual dimorphism of neuropathic pain have predominantly concentrated on the intensity and mechanism of pain generation, with few addressing the timing of pain onset. For example, research has shown that chemotherapy with paclitaxel induces more substantial pain hypersensitivity in female rats than in males on days 21 and 28 ( 21 ). The earlier onset of CINP often reduces the likelihood of completing the treatment cycle. In the present study, both clinical and animal model data showed that the females experienced an earlier onset of neuropathic pain following BTZ application compared to males. Moreover, a recent study demonstrated sexually dimorphic responses to sphingosine 1-phosphate receptor 1 antagonists and A(3) adenosine receptor agonists in BTZ-induced neuropathic pain ( 22 ). These results suggest that the mechanism of BTZ-induced neuropathic pain may differ between males and females.

Accumulating evidence supports the role of brain immune cells in regulating crucial processes, including development and diseases. Brain immune cells comprise both resident and infiltrated immune cells. Perivascular macrophages, as one of the resident monocytes in CNS, are specialized populations of macrophages that reside in the space around the cerebral vessels ( 23 , 24 ). Under physiological condition, perivascular macrophages are highly stable and are not replenished by peripheral monocytes. However, under pathological conditions, such as following chemotherapy or specific ablation, circulating monocytes can infiltrate into the space around the cerebral vessels. In this study, GO enrichment analysis revealed that cytokines altered in the spinal dorsal horn of female mice treated with BTZ were predominantly associated with monocyte/macrophage proliferation and migration, compared to the males. Flow cytometry and immunofluorescence analyses confirmed a significant increase in F4/80 + cells (the macrophage marker), but not TMEM119 + cells (a microglial-specific marker), on day 3 following BTZ treatment in females. The increased F4/80 + cells were located near the endothelium in perivascular spaces in female mice but not in male mice. Deletion of circulating monocytes prevented the increase in perivascular macrophages and alleviated the enhanced action potential firing and mechanical allodynia on the third day of BTZ treatment in females. These results indicated that the recruitment of peripheral macrophages contributed to the earlier onset of BTZ-induced neuropathic pain in female mice by enhancing the neuronal excitability. However, the exact mechanism by which macrophages infiltrate the spinal cord and participate in the neuropathic pain following BTZ treatment remains unclear. Evidence showed that adhesion and migration across microvasculature are critical for macrophage accumulation at inflammation sites. Moreover, previous studies have reported that chemotherapeutic drugs induce the expression of various cytokines/chemokines and related molecules in CNS ( 25 ). For instance, the IL-23/IL-17A/TRPV1 axis regulated female-specific mechanical pain via neuroimmune interactions ( 26 ). Here, we found that BTZ treatment increased the expression of ICAM-1, IL-1α, and CCL1 on day 3 after BTZ treatment in female mice.

ICAM-1 is a transmembrane glycoprotein of the immunoglobulin superfamily with constitutive expression at a low level in the CNS ( 27 ), which might be up-regulated during neuropathophysiological conditions ( 28 ). On the vascular endothelium, ICAM-1 acts as a ligand for macrophage-1 antigen (Mac-1), a receptor presented on macrophages ( 29 ). The interaction between ICAM-1 and Mac-1 facilitates macrophage adhesion, enhances vascular permeability, and promotes the transmigration of circulating macrophage through the blood-brain barrier ( 30 , 31 ). In the present study, Western blot and immunofluorescence staining showed that the increased ICAM-1 expressed around CD31 + cell (the marker of endothelial cells) following BTZ treatment in female mice. Inhibition of ICAM-1 prevented the macrophage up-regulation around endothelial cells and reduced mechanical allodynia in females. A recent study demonstrated that IL-4 and IL-13 induced the expression of endothelial adhesion molecule vascular cell adhesion molecule–1 (VCAM-1) and also stimulated the production of CC chemokines in structural cells (including epithelial cells). VCAM-1 and CC chemokines synergistically induced immune cell eosinophil migration to the site of tissue inflammation ( 32 ). We found an obvious increase in IL-1α in spinal cord vascular endothelial cells and prominent expression of its receptor, IL-1R1, on perivascular macrophages. While modulation of IL-1α function did not alter the number of perivascular macrophages, it affected the pain behavior. The combined application of ICAM-1 and IL-1α synergistically promoted macrophage infiltration into perivascular space in females.

Studies have shown that macrophages play important roles in various diseases by releasing inflammatory mediators, such as cytokines or chemokines, into tissue sites ( 33 ). Here, the increased CCL1 expression was observed in perivascular macrophages, while its receptor CCR8 was detected on the astrocytes and neurons in the spinal dorsal horn. Blockage of CCL1/CCR8 pathway inhibited the enhanced neuronal excitability and mechanical allodynia observed on day 3 following BTZ treatment in females. Moreover, activation of CCL1/CCR8 pathway through intrathecal injection of recombinant protein induced the pain behavior in naïve mice. On the basis of the previous report that CCL1 up-regulation only occurs in the early stage of neuropathic pain by diabetes in mice ( 34 ), we propose that the released CCL1 from perivascular macrophages may contribute to the earlier onset of BTZ-induced neuropathic pain in females. We further found that recombinant CCL1 significantly increased the calcium signaling in astrocytes and action potential frequency in neurons, suggesting that CCL1 simultaneously activates both astrocytes and neurons. To clarify the relationship between the activated astrocytes and neurons, we specifically knocked down the expression CCR8 in astrocytes and found that CCL1 still increased the neuronal excitability in these CCR8 knockdown mice, indicating that CCL1 can directly activate neurons. Furthermore, CCL1 incubation significantly increased the Ca 2+ activity in the specifically labeled astrocytes in female mice, which was prevented by preincubation with CCR8 antagonist R243. Incubation with the voltage-gated sodium channel blocker TTX did not affect the increase in CCL1-induced Ca 2+ activity, suggesting that CCL1-mediated astrocyte activation is independent of neuronal activity. We found that CNO-induced Ca 2+ signal in astrocytes led to an increase in neuronal action potential frequency and mechanical allodynia in female mice. These results suggested that the released CCL1 from the infiltrated macrophages activated both astrocytes and neurons simultaneously, and the excited astrocytes further sensitized neuronal activity, thereby mediating the earlier onset of neuropathic pain induced by BTZ in females.

In summary, our study revealed a sex disparity in the onset timing of BTZ-induced neuropathic pain in females and males. The earlier elevation of vascular endothelial cells ICAM-1 synergizing with IL-1α promoted macrophage passage through the spinal cord vessel wall. The CCL1 released by infiltrated macrophage enhanced neuron excitability, either directly or indirectly through astrocyte activation ( Fig. 9 ).

MATERIALS AND METHODS

Experimental design

This study investigates BTZ-induced chemotherapy pain, aiming to clarify the neuroimmune mechanisms underlying sex differences in pain occurrence. We demonstrated that during BTZ treatment, pain onset occurs earlier in females than in males in both humans and mice. In a mouse model, monocyte-macrophage migration and infiltration into the perivascular area of the spinal cord occurred early in female mice following BTZ treatment. Last, we investigated how these perivascular macrophages induce neuronal sensitization, ultimately leading to the onset of pain. Mice were randomly assigned to groups in all experiments without investigator blinding, sex as a biological variable. Mixed sex groups were used for most experiments, and sex-dimorphic effects are reported. All n values reflect biological replicates unless otherwise stated, and the numbers of mice per group are included, with the statistical test performed, in the legend for each figure.

Patients and methods

This retrospective cohort study was approved by the ethics committee of ZhongShan School of Medicine of Sun Yat-sen University [approval number: (2021) no. 001]. From the medical records in the First Affiliated Hospital of Sun Yat-sen University, we screened 618 patients with MM who received the BTZ treatment from January 2014 to June 2021. The inclusion criteria for the current study included the following: (i) the patient was diagnosed as MM according to the Guidelines for the Diagnosis and Treatment of Multiple Myeloma in China; (ii) age ≥ 18 years; (iii) had complete information available about their chemotherapy regimen, adverse reactions, examination results, and medical orders; and (iv) had completed at least 5 cycles of chemotherapy. Exclusion criteria included the following: (i) had been diagnosed and treated at other hospitals; (ii) BTZ was not used; (iii) accompanied by sensory impairment before admission for treatment; and (iv) had severe cardiac dysfunction, hepatic insufficiency, and other malignant solid tumors or hematological tumors. A sensory impairment was determined on the basis of the patient case records. Next, sensory impairment was characterized as a dependent variable (it was classified into grades 1–5 based on the treatment phase at the time of sensory disturbance onset), while age, gender, immune classification, and R-ISS stage were characterized as the independent variables. According to the same criteria, we also obtained a separate cohort of 40 patients from the Third Affiliated Hospital of Sun Yat-sen University from January 2010 to December 2020.

Mice

In the present study, male and female mice (C57BL/6J) with age of more than 8 weeks were purchased from the Institute of Experimental Animals of Sun Yat-sen University. The animals were housed on sawdust bedding in plastic cages and provided with food and water ad libitum. The room was maintained at 23° to 25°C and 50 to 60% humidity with a 12-hour light-dark cycle. All of the experimental protocols were approved by the Sun Yat-sen University Animal Care and Use Committee and were performed in accordance with the guidelines of the National Institutes of Health (NIH) on the care and ethical treatment of animals. The accreditation number of the ethical approval for animal experiments is SYSU-IACUC-2023-002128. All efforts were made to minimize animal suffering and to reduce the number of mice used.

Drug administration and virus injection

BTZ solution (40 μg/ml GlpBio, GC17644) was diluted with normal saline before each use and was injected (3.2 mg/kg per day, ip) into the mice for five consecutive days. A-205804 [10 mg/kg; MedChemExpress (MCE), HY-100226], a potent and selective inhibitor of ICAM-1 expression ( 35 ), was orally administered to mice 30 min before BTZ treatment. For siRNA or nontargeting control siRNA, mice were administered intrathecally (3 nmol and 10 μl) 30 min before the injection of BTZ. For neutralizing antibodies CCL1 (20 μg/kg; R&D Systems, MAB845) and IL-1R1 (20 μg/kg; Bio X Cell, BE0256), mice were administered intrathecally 30 min before the injection of BTZ or the IL-1R1 ligand, IL-1α. For recombinant proteins IL-1α (10 μg/kg; R&D Systems, 400-ML-025/CF) and CCL1 (5 μg/kg; MCE, HY- P73913 ), mice were administered intraspinally (bilateral administration, the lumbar enlargement of the spinal cord). R243 (20 μg/kg; MCE, HY-122219), a potent and selective CCR8 antagonist, was intrathecally administered to mice 30 min before the injection of BTZ or the CCR8 ligand, CCL1. For AAV viruses (AAV2/5-GfaABC1D-CCR8 shRNA, BrainVTA, Wuhan, China), mice were administered intraspinally (200 nl per side, bilateral administration, the lumbar enlargement of the spinal cord). For AAV viruses (AAV2/BI30-CAG-ICAM-1 and control viruses AAV2/BI30-CAG-3xflag, BrainVTA, Wuhan, China), mice were administered intravenously (100 μl per mouse).

Behavioral testing

The 50% paw withdrawal threshold was assessed using von Frey hairs. Briefly, each animal was loosely restrained in a plastic box and allowed to adapt for at least 30 min/day for three consecutive days before testing. Then, a series of von Frey filaments with different bending forces was presented to the midplantar surface of each hind paw. If no paw withdrawal reaction occurred, then the next stronger fiber was selected for stimulation; if a paw withdrawal reaction occurred, then the weaker stimulus was chosen. The “up and down” method was used to define the mechanical sensitivity that produced a 50% likelihood of paw withdrawal. Optimal threshold calculation by this method required five responses in the immediate vicinity of the 50% threshold. The values obtained from both hind paws were averaged. Baseline behavioral measurements were taken to assess mechanical allodynia before the administration of test substances or their vehicles.

RNA extraction and qPCR

Mice were anesthetized with sodium pentobarbital, and cardiac perfusion was performed using 0.9% physiological saline. The L4-L6 segment of the spinal dorsal horn was quickly isolated, and the total RNA was extracted using TRIzol from the tissues. Reverse transcription was performed following the protocol of a PCR kit. The PCR primer sequences for the investigated mRNAs are shown in table S4. The PCR cycle conditions were as follows: initial denaturation at 95°C for 3 min, followed by 40 cycles of 10 s at 95°C, 20 s at 58°C, and 10 s at 72°C. The mRNA expression ratio in the spinal dorsal horn was normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and analyzed using the 2 −ΔΔCt method.

Cytokine antibody array assay

The levels of various cytokines of spinal cord dorsal horn samples were analyzed using a Quantibody Mouse Cytokine Array 4000 Kit (RayBiotech, QAM-CAA-4000) with the assistance of Changyu Medical Technology. Briefly, protein samples (100 μl of sample dilution and final protein concentration of 500 μg/ml) were diluted and added to wells, followed by overnight incubation at 4°C. The slides were then washed and incubated with a biotinylated antibody cocktail at room temperature for 2 hours, followed by a 1-hour incubation with cyanine 3 (Cy3) dye–streptavidin reagent. After five washes, the slides were scanned using a laser scanner imaging system (Axon Instruments, GenePix) on the Cy3 channel. The signal values of the spots were processed with RayBio analysis software, and each array included internal controls for data normalization. The data analysis was performed using QAM-CAA-4000 data analysis software.

Western blotting

Mice were anesthetized with sodium pentobarbital, and cardiac perfusion was performed using normal saline. The L4-L6 segment of the spinal dorsal horn was quickly isolated. Dorsal horn proteins were extracted with lysis buffer (Beyotime, P0013, Shanghai, China), a protease cocktail, and phosphatase inhibitors. The protein samples were separated by SDS–polyacrylamide gel electrophoresis and then transferred onto a polyvinylidene difluoride (PVDF) membrane. After blocking in block buffer for 1 hour at room temperature, the PVDF membrane was incubated overnight at 4°C with primary antibodies against ICAM-1 (1:1000; R&D Systems, AF796) and GAPDH (1:1000; Cell Signaling Technology, 2118). The blots were then incubated for 1 hour at room temperature with the appropriate secondary antibodies. The antigen-antibody complexes were visualized by enhanced chemiluminescence (Pierce). The immunostained bands were quantified using a computer-assisted imaging analysis system (ImageJ).

Enzyme-linked immunosorbent assay

The L4-L6 segment of the mouse spinal dorsal horn was quickly isolated and homogenized. ICAM-1 (Proteintech, KE10063) and IL-1α (Elabscience, E-EL-M3059) in the supernatant were detected using enzyme-linked immunosorbent assay kits according to the manufacturer’s instructions.

Cell preparation and flow cytometry

Anesthetized mice were subjected to perfusion with ice-cold phosphate-buffered saline (PBS) to remove blood from the tissues. The lumbar enlargement of the spinal cord was then isolated and chopped into small pieces. The tissue pieces were digested in Dulbecco’s modified Eagle’s medium/F12 medium containing papain (1 mg/ml), dispase II (1.2 U/ml), deoxyribonuclease I (20 U/ml), and collagenase IV (100 U/ml) at 37°C for 30 min. Homogenization was achieved by gently pipetting the solution up and down. The homogenates were washed and passed through a 70-μm cell strainer to obtain a single-cell suspension. The cells from the lumbar enlargement were centrifuged at 500 g for 5 min at 4°C and resuspended in PBS. To prevent nonspecific binding, the cells were incubated with CD16/CD32 Fc receptor blocking antibody for 10 min at 4°C. After one wash, the cells were resuspended in 200 μl of PBS containing CD11b-allophycocyanin and CD45–fluorescein isothiocyanate antibodies and incubated for 1 hour at 4°C. Following another wash, the cells were resuspended in 2% fetal bovine serum/PBS for fluorescence-activated cell sorting analysis. The frequencies of CD11b + CD45 mid microglia and CD11b + CD45 high macrophages were measured using a Beckman CytoFlex S flow cytometer and analyzed using FlowJo software (Tree Star).

Monocyte depletion

According to our previous study, CLLs were intravenously injected to deplete the monocyte. Control mice received liposome-encapsulated PBS at the same time points. The dose of CLLs was 10 μl/g of body weight. Because the depletion of circulating monocytes lasted for 72 hours after each CLL injection, CLLs were injected into the tail vein 72 hours before the BTZ treatment.

Immunostaining and confocal imaging

Mice were anesthetized with sodium pentobarbital, and cardiac perfusion was performed using normal saline, followed by 4% paraformaldehyde in PBS. Next, L4-L6 spinal cord tissues were removed, post-fixed for 30 min in the same fixative, and then dehydrated with 30% sucrose. Cryostat sections (40 μm thick) were cut and blocked with QuickBlock Blocking Buffer for Immunol Staining (Beyotime, P0260) for 1 hour at room temperature. Sections were then incubated overnight at 4°C with primary antibodies against goat anti–ICAM-1 (1:100; R&D Systems, AF796), rat anti-F4/80 (1:200; Abcam, ab6640), rabbit anti-F4/80 (1:200 Abcam, ab300421), rat anti-CD31 (1:50; BD Biosciences, 550274), rabbit anti-TMEM119 (1:500; Abcam, ab209064), rabbit anti–IL-1α (1:200; Abcam, ab300500), rat anti-CCL1 (1:200; R&D Systems, MAB846), mouse anti–IL-1RI (1:50; R&D Systems, AF771), rabbit anti-CCR8 (1:50 Boster, M03518), mouse anti–glial fibrillary acidic protein (1:500; Abcam, ab4648), and mouse anti-neuron-specific nuclear protein (NeuN) (1:500; Merck, MAB377). The following day, sections were incubated with Cy3-, Alexa Fluor 488–, and Alexa Fluor 647–conjugated secondary antibodies at room temperature for 1 hour. After staining, images were captured using a Carl Zeiss LSM 980 confocal microscope (Zeiss, Oberkochen, Germany) and analyzed with ImageJ software (NIH).

Spinal cord slice preparation and patch-clamp recordings

Mice were deeply anesthetized with sodium pentobarbital. The lumbosacral spinal cord was quickly removed and transferred to oxygenated (95% O 2 and 5% CO 2 ) ice-cold slice solution containing 240 mM sucrose, 25 mM NaHCO 3 , 2.5 mM KCl, 1.25 mM NaH 2 PO 4 , 0.5 mM CaCl 2 , and 3.5 mM MgCl 2 (pH 7.4). Four hundred–micrometer–thick spinal cord slices were cut on a vibratome (Leica, VT-1000 S). The slices were incubated at 32°C for at least 30 min in artificial cerebrospinal fluid (ACSF) containing 126 mM NaCl, 3 mM KCl, 1.3 mM MgCl 2 , 2.5 mM CaCl 2 , 26 mM NaHCO 3 , 1.25 mM NaH 2 PO 4 , and 11 mM glucose, equilibrated with 95% O 2 and 5% CO 2 . Following the incubation, the slices were transferred to the recording chamber, which was continuously perfused with preheated ACSF at a rate of 2 ml/min. Superficial dorsal horn neurons were visualized using a 40× water immersion objective on an upright infrared Nikon microscope (Nikon, Tokyo, Japan).

Whole-cell recordings were made from lamina II neurons using patch pipettes fabricated from thin-walled, fibre-filled capillaries. The patch pipettes had a resistance of (6 to 8 megohm). Action potential or drug-induced potential recordings were made under current-clamp conditions using a pipette filled with an internal solution containing 130 mM K gluconate, 5 mM KCl, 2 mM MgCl 2 , 0.6 mM EGTA, 10 mM Hepes, 2 mM Mg–adenosine 5′-triphosphate, and 0.3 mM Na–guanosine 5′-triphosphate (pH 7.3, osmolality of 300 mosmol/liter). An EPC 10 amplifier (HEKA Elektronik, Germany) was used to record data. Patchmaster software (HEKA Elektronik) was used to deliver stimuli and acquire data. Action potentials were evoked by injecting current every 10 s using step intervals ranging from 0 to 200 pA and lasting for a period of 500 ms. The relationship between firing frequency and injected current was analyzed using Clampfit (Axon Instruments Inc., USA).

Ex vivo two-photon Ca 2+ imaging

The lumbar enlargement of the spinal cord from female mice with preinjection of AAV2/5-GfaABC1D-CRE and AAV-DIO-GCaMP6s–enhanced green fluorescent protein (or AAV-DIO-hM3Dq-mCherry) was extracted and sectioned into 600-μm-thin slices. Following incubation in ACSF at 32°C for 30 min, the slices were transferred to a 60-mm dish for imaging. The ventral side of the slice was fixed to the bottom of the dish using glue and the dish filled with ACSF. Following a 1-min baseline recorded to monitor spontaneous activity, the CCL1 recombinant protein, CCR8 antagonist R243, and TTX (1 μM) were added into the solution. The fluorescence intensity of astrocytes expressing GCaMP6s was recorded. In spinal cord slices from hM3Dq/GfaABC1D-Cre mice, the slices were loaded with the Ca 2+ indicator Fluo-4 AM (10 μM), and astrocytes were separately labeled with SR101 (10 μM) at 32° to 34°C for 20 min in the dark in standard ACSF. The overall cellular fluorescence intensity of astrocytes expressing Fluo-4 was collected, followed by a 240-s bath application of CNO (10 μM). Two-photon fluorescence measurements were obtained with a dual-scanhead two-photon microscope (Olympus, FVMPE-RS). Fluorescence signals were collected using a 25× 1.05 numerical aperture water immersion objective (Olympus). Images were acquired at a 1.42-Hz frame rate and at a 512 × 512 pixel resolution. Fluorescence intensity measurements were made using FV30S-SW software. The changes in Ca 2+ fluorescence intensity were expressed as the percentage of pretreatment fluorescence levels (Δ F / F ).

Statistical analysis

Statistical analysis was performed using IBM SPSS 29.0 and GraphPad Prism 8.0.1 software. All data are expressed as the means ± SEM. The Shapiro-Wilk test was used to check the normality of data. The normally distributed data were analyzed using the Student’s t test (two independent samples or paired samples) or one-way analysis of variance (ANOVA). Post hoc analyses were performed using Tukey’s post hoc or Dunn’s multiple comparisons test, as appropriate. In addition, the electrophysiological data of action potential were analyzed using the two-way ANOVA, and the data of behavioral tests were analyzed using repeated-measures ANOVA. A nonparametric test (Mann-Whitney test or Wilcoxon matched-pairs signed-rank test) was used instead when the normality test was not satisfied. The chi-square test or Fisher’s exact test was used for statistical analysis of count data. Predictors of early peripheral sensory disturbance after BTZ treatment were analyzed by ordinal multiple logistic regression. The criterion for statistical significance was P < 0.05. While no power analysis was performed, the sample size was based on previous studies.

Acknowledgments

Funding: This work was supported by the Key-Area Research and Development Program of Guangdong Province (2023B0303010002 to W.-J.X.); the National Natural Science Foundation of China (82071224 to X.F., 82371189 to X.F., 82471238 to J.-D.X., and 82472592 to C.-C.L.); the Guangdong Basic and Applied Basic Research Foundation (2023A1515030020 to T.X., 2022A1515012124 to X.-Z.Z., and 2022B1515120026 to J.-D.X.; the Science and Technology Planning Project of Guangdong Province (2023B1212060018 and 2021B1212040017 to W.-J.X.); the Key Field R&D Plan Project of Guangzhou (2023B03J0172 to X.-Z.Z.); the Guangzhou Science and Technology Plan Project (20226060004 to W.-J.X.); and the Guangzhou Clinical High-Tech Project (2023P-GX10 to X.F.).

Author contributions: Conceptualization: J.-D.X., T.X., X.F., X.-Z.Z., W.-J.X., L.C., X.Z., C.-C.L., C.W., and X.-Q.T. Methodology: J.-D.X., T.X., X.F., W.-J.X., L.C., X.Z., C.-C.L., J.D., C.W., M.-Y.C., and X.-Q.T. Software: J.-D.X., L.C., X.Z., J.D., and M.-Y.C. Validation: J.-D.X., T.X., X.F., W.-J.X., L.C., X.Z., C.-C.L., and C.W. Formal analysis: J.-D.X., T.X., X.F., L.C., X.Z., C.-C.L., P.Y., C.W., M.-Y.C., and J.-M.S. Investigation: J.-D.X., T.X., X.F., L.C., X.Z., C.-C.L., J.D., and C.W. Resources: J.-D.X., T.X., X.F., L.C., X.Z., C.-C.L., P.Y., C.W., X.-Q.T., and J.-M.S. Data curation: J.-D.X., T.X., L.C., X.Z., C.-C.L., and P.Y. Writing—original draft: J.-D.X., T.X., X.F., W.-J.X., L.C., X.Z., C.-C.L., and C.W. Writing—review and editing: J.-D.X., T.X., X.F., X.-Z.Z., W.-J.X., L.C., X.Z., C.-C.L., P.Y., C.W., X.-Q.T., and J.-M.S. Visualization: J.-D.X., T.X., X.F., L.C., X.Z., C.-C.L., and C.W. Supervision: J.-D.X., T.X., X.F., X.-Z.Z., W.-J.X., L.C., X.Z., and C.-C.L. Project administration: J.-D.X., T.X., X.F., W.-J.X., L.C., X.Z., C.-C.L., and J.D. Funding acquisition: J.-D.X., T.X., X.F., X.-Z.Z., W.-J.X., and C.-C.L.

Competing interests: The authors declare that they have no competing interests.

Data and materials availability: All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials.

Supplementary Materials

This PDF file includes:

Figs. S1 to S7

Tables S1 to S4

REFERENCES AND NOTES

1 A. J. Beijers , S. Oerlemans , F. Mols , M. Eurelings , M. C. Minnema , A. Vreugdenhil , L. V. van de Poll-Franse , The magnitude of neurotoxicity in patients with multiple myeloma and the impact of dose modifications: Results from the population-based PROFILES registry . Ann. Hematol. 96 , 653 – 663 ( 2017 ). 28116479 10.1007/s00277-017-2927-8

2 A. J. Cowan , D. J. Green , M. Kwok , S. Lee , D. G. Coffey , L. A. Holmberg , S. Tuazon , A. K. Gopal , E. N. Libby , Diagnosis and management of multiple myeloma: A review . JAMA 327 , 464 – 477 ( 2022 ). 35103762 10.1001/jama.2022.0003

3 S. V. Rajkumar , Multiple myeloma: 2022 update on diagnosis, risk stratification, and management . Am. J. Hematol. 97 , 1086 – 1107 ( 2022 ). 35560063 10.1002/ajh.26590 PMC9387011

4 É. Midavaine , J. Côté , S. Marchand , P. Sarret , Glial and neuroimmune cell choreography in sexually dimorphic pain signaling . Neurosci. Biobehav. Rev. 125 , 168 – 192 ( 2021 ). 33582232 10.1016/j.neubiorev.2021.01.023

5 J. S. Mogil , Sex differences in pain and pain inhibition: Multiple explanations of a controversial phenomenon . Nat. Rev. Neurosci. 13 , 859 – 866 ( 2012 ). 23165262 10.1038/nrn3360

6 K. Ren , R. Dubner , Interactions between the immune and nervous systems in pain . Nat. Med. 16 , 1267 – 1276 ( 2010 ). 20948535 10.1038/nm.2234 PMC3077564

7 E. A. Old , S. Nadkarni , J. Grist , C. Gentry , S. Bevan , K. W. Kim , A. J. Mogg , M. Perretti , M. Malcangio , Monocytes expressing CX3CR1 orchestrate the development of vincristine-induced pain . J. Clin. Invest. 124 , 2023 – 2036 ( 2014 ). 24743146 10.1172/JCI71389 PMC4001538

8 Y. Ye , H. Cheng , Y. Wang , Y. Sun , L.-D. Zhang , J. Tang , Macrophage: A key player in neuropathic pain . Int. Rev. Immunol. 43 , 326 – 339 ( 2024 ). 38661566 10.1080/08830185.2024.2344170

9 X. B. Yu , H. J. Liu , K. A. Hamel , M. G. Morvan , S. Yu , J. Leff , Z. Guan , J. M. Braz , A. I. Basbaum , Dorsal root ganglion macrophages contribute to both the initiation and persistence of neuropathic pain . Nat. Commun. 11 , 264 ( 2020 ). 31937758 10.1038/s41467-019-13839-2 PMC6959328

10 F. A. White , J. Sun , S. M. Waters , C. Ma , D. Ren , M. Ripsch , J. Steflik , D. N. Cortright , R. H. LaMotte , R. J. Miller , Excitatory monocyte chemoattractant protein-1 signaling is up-regulated in sensory neurons after chronic compression of the dorsal root ganglion . Proc. Natl. Acad. Sci. U.S.A. 102 , 14092 – 14097 ( 2005 ). 16174730 10.1073/pnas.0503496102 PMC1236537

11 W. Wen , J. Cheng , Y. Tang , Brain perivascular macrophages: Current understanding and future prospects . Brain 147 , 39 – 55 ( 2024 ). 37691438 10.1093/brain/awad304 PMC10766266

12 T. Goldmann , P. Wieghofer , M. J. C. Jordão , F. Prutek , N. Hagemeyer , K. Frenzel , L. Amann , O. Staszewski , K. Kierdorf , M. Krueger , G. Locatelli , H. Hochgerner , R. Zeiser , S. Epelman , F. Geissmann , J. Priller , F. M. V. Rossi , I. Bechmann , M. Kerschensteiner , S. Linnarsson , S. Jung , M. Prinz , Origin, fate and dynamics of macrophages at central nervous system interfaces . Nat. Immunol. 17 , 797 – 805 ( 2016 ). 27135602 10.1038/ni.3423 PMC4968048

13 K. A. Sailor , G. Agoranos , S. López-Manzaneda , S. Tada , B. Gillet-Legrand , C. Guerinot , J.-B. Masson , C. L. Vestergaard , M. Bonner , K. Gagnidze , G. Veres , P.-M. Lledo , N. Cartier , Hematopoietic stem cell transplantation chemotherapy causes microglia senescence and peripheral macrophage engraftment in the brain . Nat. Med. 28 , 517 – 527 ( 2022 ). 35190726 10.1038/s41591-022-01691-9

14 C.-L. Mai , Z. Tan , Y.-N. Xu , J.-J. Zhang , Z.-H. Huang , D. Wang , H. Zhang , W.-S. Gui , J. Zhang , Z.-J. Lin , Y.-T. Meng , X. Wei , Y.-T. Jie , P. M. Grace , L.-J. Wu , L.-J. Zhou , X.-G. Liu , CXCL12-mediated monocyte transmigration into brain perivascular space leads to neuroinflammation and memory deficit in neuropathic pain . Theranostics 11 , 1059 – 1078 ( 2021 ). 33391521 10.7150/thno.44364 PMC7738876

15 M. Malcangio , Role of the immune system in neuropathic pain . Scand J Pain 20 , 33 – 37 ( 2019 ). 31730538 10.1515/sjpain-2019-0138

16 N. T. Fiore , S. R. Debs , J. P. Hayes , S. S. Duffy , G. Moalem-Taylor , Pain-resolving immune mechanisms in neuropathic pain . Nat. Rev. Neurol. 19 , 199 – 220 ( 2023 ). 36859719 10.1038/s41582-023-00777-3

17 C. Lawson , S. Wolf , ICAM-1 signaling in endothelial cells . Pharmacol. Rep. 61 , 22 – 32 ( 2009 ). 19307690 10.1016/s1734-1140(09)70004-0

18 G. G. Page , M. J. Hayat , S. L. Kozachik , Sex differences in pain responses at maturity following neonatal repeated minor pain exposure in rats . Biol. Res. Nurs. 15 , 96 – 104 ( 2013 ). 21900309 10.1177/1099800411419493

19 E. J. Bartley , R. B. Fillingim , Sex differences in pain: A brief review of clinical and experimental findings . Br. J. Anaesth. 111 , 52 – 58 ( 2013 ). 23794645 10.1093/bja/aet127 PMC3690315

20 D. L. Blackwell , J. W. Lucas , T. C. Clarke , Summary health statistics for U.S. adults: National health interview survey, 2012 . Vital Health Stat.10 10 , 1 – 161 ( 2014 ). 24819891

21 L. F. Ferrari , D. Araldi , P. G. Green , J. D. Levine , Marked sexual dimorphism in neuroendocrine mechanisms for the exacerbation of paclitaxel-induced painful peripheral neuropathy by stress . Pain 161 , 865 – 874 ( 2020 ). 31917777 10.1097/j.pain.0000000000001798 PMC7085433

22 K. Stockstill , C. Wahlman , K. Braden , Z. Chen , G. L. Yosten , D. K. Tosh , K. A. Jacobson , T. M. Doyle , W. K. Samson , D. Salvemini , Sexually dimorphic therapeutic response in bortezomib-induced neuropathic pain reveals altered pain physiology in female rodents . Pain 161 , 177 – 184 ( 2020 ). 31490328 10.1097/j.pain.0000000000001697 PMC6923586

23 T. Masuda , L. Amann , G. Monaco , R. Sankowski , O. Staszewski , M. Krueger , F. D. Gaudio , L. He , N. Paterson , E. Nent , F. Fernández-Klett , A. Yamasaki , M. Frosch , M. Fliegauf , L. F. P. Bosch , H. Ulupinar , N. Hagemeyer , D. Schreiner , C. Dorrier , M. Tsuda , C. Grothe , A. Joutel , R. Daneman , C. Betsholtz , U. Lendahl , K.-P. Knobeloch , T. Lämmermann , J. Priller , K. Kierdorf , M. Prinz , Specification of CNS macrophage subsets occurs postnatally in defined niches . Nature 604 , 740 – 748 ( 2022 ). 35444273 10.1038/s41586-022-04596-2

24 K. De Vlaminck , H. Van Hove , D. Kancheva , I. Scheyltjens , A. R. P. Antunes , J. Bastos , M. Vara-Perez , L. Ali , M. Mampay , L. Deneyer , J. F. Miranda , R. Cai , L. Bouwens , D. De Bundel , G. Caljon , B. Stijlemans , A. Massie , J. A. Van Ginderachter , R. E. Vandenbroucke , K. Movahedi , Differential plasticity and fate of brain-resident and recruited macrophages during the onset and resolution of neuroinflammation . Immunity 55 , 2085 – 2102.e9 ( 2022 ). 36228615 10.1016/j.immuni.2022.09.005

25 P. Śliwa-Tytko , A. Kaczmarska , M. Lejman , J. Zawitkowska , Neurotoxicity associated with treatment of acute lymphoblastic leukemia chemotherapy and immunotherapy . Int. J. Mol. Sci. 23 , 5515 ( 2022 ). 35628334 10.3390/ijms23105515 PMC9146746

26 X. Luo , O. Chen , Z. Wang , S. Bang , J. Ji , S. H. Lee , Y. Huh , K. Furutani , Q. He , X. Tao , M.-C. Ko , A. Bortsov , C. R. Donnelly , Y. Chen , A. Nackley , T. Berta , R.-R. Ji , IL-23/IL-17A/TRPV1 axis produces mechanical pain via macrophage-sensory neuron crosstalk in female mice . Neuron 109 , 2691 – 2706.e5 ( 2021 ). 34473953 10.1016/j.neuron.2021.06.015 PMC8425601

27 J. B. Dietrich , The adhesion molecule ICAM-1 and its regulation in relation with the blood-brain barrier . J. Neuroimmunol. 128 , 58 – 68 ( 2002 ). 12098511 10.1016/s0165-5728(02)00114-5

28 T. M. Bui , H. L. Wiesolek , R. Sumagin , ICAM-1: A master regulator of cellular responses in inflammation, injury resolution, and tumorigenesis . J. Leukoc. Biol. 108 , 787 – 799 ( 2020 ). 32182390 10.1002/JLB.2MR0220-549R PMC7977775

29 R. B. Henderson , L. H. Lim , P. A. Tessier , F. N. Gavins , M. Mathies , M. Perretti , N. Hogg , The use of lymphocyte function-associated antigen (LFA)-1-deficient mice to determine the role of LFA-1, Mac-1, and α4 integrin in the inflammatory response of neutrophils . J. Exp. Med. 194 , 219 – 226 ( 2001 ). 11457896 10.1084/jem.194.2.219 PMC2193453

30 R. Sumagin , J. M. Kuebel , I. H. Sarelius , Leukocyte rolling and adhesion both contribute to regulation of microvascular permeability to albumin via ligation of ICAM-1 . Am. J. Physiol.Cell Physiol. 301 , C804 – C813 ( 2011 ). 21653902 10.1152/ajpcell.00135.2011 PMC3191564

31 C. Anbarasan , M. Bavanilatha , K. Latchumanadhas , S. A. Mullasari , ICAM-1 molecular mechanism and genome wide SNP’s association studies . Indian Heart J. 67 , 282 – 287 ( 2015 ). 26138191 10.1016/j.ihj.2015.03.005 PMC4495607

32 M. Nagata , H. Yamamoto , K. Tabe , Y. Sakamoto , Eosinophil transmigration across VCAM-1-expressing endothelial cells is upregulated by antigen-stimulated mononuclear cells . Int. Arch. Allergy Immunol. 125 , 7 – 11 ( 2001 ). 11408764 10.1159/000053844

33 G. A. Duque , A. Descoteaux , Macrophage cytokines: Involvement in immunity and infectious diseases . Front. Immunol. 5 , 491 ( 2014 ). 25339958 10.3389/fimmu.2014.00491 PMC4188125

34 M. Zychowska , E. Rojewska , A. Piotrowska , G. Kreiner , I. Nalepa , J. Mika , Spinal CCL1/CCR8 signaling interplay as a potential therapeutic target evidence from a mouse diabetic neuropathy model . Int. Immunopharmacol. 52 , 261 – 271 ( 2017 ). 28961489 10.1016/j.intimp.2017.09.021

35 A. O. Stewart , P. A. Bhatia , C. M. M. Carty , M. V. Patel , M. A. Staeger , D. L. Arendsen , I. W. Gunawardana , L. M. Melcher , G. D. Zhu , S. A. Boyd , D. G. Fry , B. L. Cool , L. Kifle , K. Lartey , K. C. Marsh , A. J. Kempf-Grote , P. Kilgannon , W. Wisdom , J. Meyer , W. M. Gallatin , G. F. Okasinski , Discovery of inhibitors of cell adhesion molecule expression in human endothelial cells. 1. Selective inhibition of ICAM-1 and E-selectin expression . J. Med. Chem. 44 , 988 – 1002 ( 2001 ). 11300880 10.1021/jm000452m