Both male (Supplementary Figures 1A–D) and female mice were initially used to develop systemic and neuronal inflammatory models. No significant sex-dependent differences in GlyR expression were observed. This study was conducted on adult C57BL/6J mice (female, 5–8 weeks old, weighing 20–25 grams) after approval from the Institutional Animal Care and Use Committee (IACUC, protocol no. 11026/23002). The mice were housed in plastic cages under a 12-hour light/dark cycle, at 22 ± 2 °C and 50 ± 10% humidity, with free access to a pellet diet and water. All experimental procedures and animal welfare protocols were carried out in accordance with the Ethical Regulations on the Care and Use of Laboratory Animals at Texas Tech University Health Sciences Center, El Paso, and were approved by the IACUC committee for animal experiments.

Systemic inflammation in C57BL/6J mice (n=4–6 mice) was induced by intraperitoneal (i.p.) injection of lipopolysaccharide (LPS, #L2755; Sigma-Aldrich) at a dose of 5 mg/kg body weight, freshly prepared in sterile phosphate-buffered saline (PBS). 2 and 5 days post-LPS injection, animals were euthanized with an i.p. injection of a ketamine-xylazine cocktail (100/16 mg/kg) following the American Veterinary Medical Association (AVMA) guidelines. Adequate anesthesia was confirmed by the absence of a withdrawal reflex to firm hind paw pressure. Blood samples were collected from the heart (cardiac puncture) in sterile EDTA vacutainers (Greiner, USA) to avoid clotting. Post cessation of breathing and heartbeat, cervical dislocation was done to ensure death. The brain, spleen, and bone marrow were rapidly removed, snap-frozen in dry ice, collected in sterile PBS (BM) on ice, and stored at -80 °C until further processing.

To develop a neuroinflammatory mouse model, mice were anesthetized with an i.p. injection of 100 mg/kg ketamine and 16 mg/kg xylazine. The mice’s toe pinch reflex was used to ensure the appropriate anesthesia. Each mice were injected with a stereotactic intracortical injection of LPS (12µg/5µl) into the left hemisphere of the brain (n=4–6 mice) for 2 and 5 days. In both cohorts, the mice had not received any treatment for inflammation. Post-surgery, the mice were monitored for sickness behavior and recovery. At the beginning of the experiment (day 0), following the AVMA guidelines, healthy mice (n=4-6) were sacrificed as controls. At the endpoint, mice were euthanized with an i.p. injection of a ketamine-xylazine cocktail (100/16 mg/kg), and the toe pinch reflex was used to confirm the deep anesthesia. A 25-30g needle on a 1ml syringe was inserted into the heart for blood collection. After blood collection, mice were verified as having no heartbeat and no breathing, and cervical dislocation was then performed to confirm death. The brain, spleen, and bone marrow were collected to study GlyRs and cytokine response.

To prevent blood from clotting, heparin or EDTA-coated tubes were used to collect blood. The blood samples were treated with 10 mL red blood cell (RBC) lysis buffer and incubated for 5 min at room temperature. The cells were centrifuged at 300xg for 5 min., followed by two washes with cold PBS at 500xg for 5 min at 4 °C. The supernatant was aspirated, and then the WBCs were collected as a cell suspension for qRT-PCR and immunostaining.

Human PBMCs and CD14+ monocytes were purchased directly from BioIVT (HUMANCD14-0107684, HUMANPBMC-0002132, Westbury, NY, USA). These PBMCs and CD14+ monocytes from healthy individuals were used to perform RNA extraction for gene expression and immunofluorescence for GlyRs localization studies.

To isolate the bone marrow from CB57L/6 mice, the surrounding musculature was excised to access the mouse femur. The proximal end of the femur, as well as just beyond its distal end, was cut to ensure complete removal. The femur was then placed in cold PBS to reduce and prevent cell death. Within a BSL-2 hood, residual tissue on the femur was removed. Surgical scissors were used to cut the distal and proximal ends of the femur, allowing access to the bone marrow in the central part of the bone. A 20-gauge needle was used to inject cold PBS through the femur to dislodge the bone marrow onto a sterile petri dish. The bone marrow was filtered through a 40 μm nylon mesh filter into a 50 mL Falcon tube, with the syringe’s rubber stopper used to homogenize the marrow through the filter. The filtered bone marrow suspension was centrifuged at 500×g for 5 minutes, and the PBS was then removed. To the sterile bone marrow cell pellet, 200 µL of RBC lysis buffer was added and incubated for 5 minutes. After centrifugation, the RBC lysis buffer was discarded, and the cell pellet was collected.

Mouse bone marrow cells were cultured in a medium containing DMEM with glutamine, a mixture of penicillin and streptomycin antibiotics, 10% fetal bovine serum (FBS), and 10% macrophage-colony-stimulating factor (M-CSF) to differentiate into bone marrow macrophages (BMMs). These cells were grown in a 6-well plate, on coverslips with or without LPS treatment.

Total RNA was extracted from blood, bone marrow cells, brain, and spleen tissues using the TRIzol (#15596018, Thermo Scientific, USA) method following the manufacturer’s instructions. The extracted RNA was quantified and checked for purity with a NanoDrop 2000 spectrophotometer (Thermo Scientific, USA). An equal amount, specifically 2 µg of RNA from each sample, was used for cDNA synthesis with the high-capacity cDNA reverse transcription kit (#4368814, Thermo Scientific, USA). Mice and human-specific primers targeting different GlyRs, including GlyRα-1, GlyRα-2, GlyRα-3, and housekeeping genes β-actin and beta2-microglobulin (B2M) (Sigma Aldrich, USA), were used (Table 1).

Gene expression was determined in a 10µl reaction volume using 5µl of SYBR Green Master mix (1X) (#KR0389, Kapa Biosystems, South Africa), 0.5µl each of forward and reverse primer (10 µM), 3µl nuclease-free water, and 1µl cDNA. Roche Light Cycler 96 (Roche, USA) was used to run the reaction with the following thermal profile: pre-incubation at 95°C for 3 sec., followed by 43 cycles of amplification (denaturation at 95°C for 10 sec., primer annealing at 60°C for 1 min., and extension at 72°C for 2 sec.) with default acquisition at 72°C. Finally, the gene expression was calculated using the 2^{^-ΔΔCt} as relative fold change.

The freshly isolated WBCs, BM, splenic cells, and macrophages were fixed in 4% paraformaldehyde (PFA) on a slide and left to dry at room temperature. The snap-frozen brain and spleen tissues were embedded in cryomolds (Tissue-Tek, USA) using optimal cutting temperature (OCT) compound (Tissue-Tek, USA) and stored at -80°C. For each brain and spleen tissue, 5µm-thick sections were cut using a cryostat (Leica CM3050S, USA) set at -20°C. The tissue sections on slides were fixed with a 1:1 ratio of acetone and ethanol (-20°C).

The slides were rinsed with PBS, followed by permeabilization using 0.1% Triton. After blocking with 5% normal goat serum (NGS) for 1 hour, the slides and coverslips were stained with primary antibodies targeting human and mouse GlyRα1 (1:500, #NB300-113, Novus Bio, Littleton, CO), GlyRα3 (1:100, #75-417, Neuromab, UC Davis/NIH), CD11b (1:100, #Sc-1186, Santa Cruz Biotechnology), CD-68 (1:100, #Sc-20060, Santa Cruz Biotechnology), and Iba1 (1:400, #MABN92, Millipore Sigma) (Table 2) overnight at 4°C. After incubation, the slides were washed three times with PBS for 5 minutes each, then incubated with appropriate secondary antibodies, i.e., Alexa Fluor 488 (1:200) and/or 594 (1:200) (Thermo-Fisher Scientific) for 1 hour at room temperature. A DAPI mounting agent (#AB104139, Santa Cruz Biotechnology) was used for nuclear staining and mounting. Finally, immunofluorescence images were captured using a fluorescent microscope.

Changes in the expression and localization of GlyRα subunits in WBCs, BMMs, spleen, and brain tissues were observed by imaging with a Nikon Ti-Eclipse microscope (Nikon Instruments Inc., Melville, NY). The 60x magnification lens with oil immersion was set for higher resolving power in the microscope. The NS Elements AR program was utilized to image the slides. Images were analyzed in ImageJ, processed in Adobe Photoshop, and compiled in Illustrator for publication.

GraphPad Prism (version 7.0, GraphPad Software Inc., USA) was used for data analysis. For comparisons between two groups, an unpaired two-tailed Student’s t-test was applied. Multiple group comparisons were performed using one-way or two-way analysis of variance (ANOVA), followed by Tukey’s post hoc analysis. The data were presented as mean ± S.E.M. Fold change in gene expression was calculated using the 2^{^-ΔΔCt} method with control samples normalized to 1. All the experiments were repeated 2–3 times in duplicates. Statistical significance was defined as p<0.05.

Although GlyRs are extensively characterized and studied in the brain and spinal cord, their expression and functional significance in circulating immune cells remain largely unexplored. We investigated and provided evidence that human circulating white blood cells (WBCs), specifically peripheral blood mononuclear cells (PBMCs) and CD14+ monocytes, express transcripts and proteins corresponding to different GlyRα1, α2, and α3 subunits. A qRT-PCR assay was used to detect the gene expression of GlyR α1, α2, and α3 subunits in PBMCs and CD14+ monocytes collected from healthy individuals. Among these, GlyRα1 (0.77 ± 0.01 and 0.84 ± 0.01) and GlyRα2 (0.56 ± 0.01 and 0.56 ± 0.01) subunits showed the highest expression in PBMCs and CD14+ monocytes, respectively, whereas GlyRα3 was expressed at lower levels (Figure 1A).

Next, protein expression of GlyR was detected in human PBMCs, immunostaining with antibodies against GlyRα1/2 (red) and GlyRα3 (green). Nuclear DNA was stained with DAPI (blue). Quantitative images revealed 55.21 ± 2.8% of GlyRα1/2+ cells and 46.87 ± 3.4% of GlyRα3+ cells with 43.12 ± 2.2% of GlyRα1/2+/GlyRα3+ CD14+ monocytes (Figure 1B). Immunofluorescence results indicated that GlyRα1/2 (Figure 1C, red) and GlyRα3 (Figure 1C, green) are present both in the cytoplasm (16.45% and 7.59%) and on the membrane (73.41% and 49.36%) of PBMCs, respectively. Additionally, the merged image exhibited the co-expression of GlyRα1/2 and GlyRα3. The cellular distribution of GlyRα1/2 and GlyRα3 suggested possibilities of receptor assembly and localization, similar to neuronal GlyRs.

After examining GlyR expression in human WBCs, we were interested in assessing the gene expression levels of GlyRα1, α2, and α3 in the mouse brain (Figure 2A) using qRT-PCR. GlyRα2 showed a statistically significant higher expression (0.019 ± 0.003, p<0.01**), followed by GlyRα3 (0.003 ± 0.0002, p<0.01**) and GlyRα1 (0.0005 ± 5.446e-005, p<0.01**). Immunostaining of brain sections showed co-expression of GlyRα1/2 (Figure 2B, red) and GlyRα3 (Figure 2B, green) mainly within neurons of the hippocampus (Figure 2B).

Next, we detected the expression of GlyRs in the peripheral organ spleen, where GlyRα2 remained dominant (0.05 ± 0.006, p<0.001**), followed by GlyRα3 (0.001 ± 0.0003) and GlyRα1 (8e-005 ± 2.069e-005) expression (Figure 2C). Immunostaining of spleen tissue showed strong GlyRα1/2 expression (Figure 2D, red) in both red (arrow) and white pulps (arrow head), whereas GlyRα3 localized to the macrophage zone of red pulp (Figure 2D, green, arrow).

Circulating WBCs and bone marrow cells (BM) were also analyzed for identifying GlyRα expression. BM cells and WBCs showed dominance of GlyRα2 expression, followed by GlyRα3 and GlyRα1 (Figures 2E, G). Immunostaining revealed that GlyRα1/2+ (red) and GlyRα3+ (green) were more dominantly expressed on the membrane, with less expression in the cytoplasm of BM and WBCs (Figures 2F, H). This is in accordance with the study of Eynden et al. (24). The results show that BM and WBCs express the GlyRα subunits, albeit at different baseline levels.

Inflammation-regulated dynamic changes in GlyR expression were analyzed using the LPS-induced inflammatory macrophages model. Firstly, monocytes/macrophages were treated with LPS, and then they were examined using an immunofluorescence assay (Figure 3A). Monocytes showed LPS-dose-dependent significant fold increase in IL-1β and TNFα with a decrease in IL-2 expression (Figure 3B). In CD11b+ monocytes and neutrophils, stronger co-expression with GlyRα1/2 (Figure 3C) and GlyRα3 (Figure 3D) was observed in LPS-treated monocytes compared to the control. LPS-activated monocytes exhibited greater expression of CD68 in parallel to an increase of GlyRα1/2 (Figure 3E) and GlyRα-3 (Figure 3F). The modification of GlyR expression at protein levels suggests that GlyRs may play a functional role in innate immunity.

GlyRα2 is a well-studied GlyR, especially in the brain. To determine whether brain GlyRα1, α2, and α3 respond differently to systemic and neuronal inflammation, mice were administered LPS intraperitoneally and intracortically (neuroinflammation), and samples were obtained 48 hours post-treatment. A group of untreated mice was enrolled as healthy controls. A significant increase in GlyRα3 (2.23 ± 0.2-fold, p<0.001**) expression in neuroinflammatory mice, along with the highest expression of GlyRα1 (5.91 ± 0.73-fold, p<0.001**) in neuro-inflammation compared to systemic inflammation and control was observed in brain tissues (Figure 4A).GlyRα2 expression was increased in response to systemic inflammation (1.4 ± 0.14, p=0.04*), with no change in neuroinflammation (Figure 4A). A heat map revealed up-regulation of brain GlyRα1 and α3 in response to neuroinflammation compared to systemic inflammation (Figure 4B). In contrast, GlyRα2 responds more to systemic inflammation than to neuroinflammation. The cytokines IL-1β, IL-2, and TNFα expression was also studied under inflammation conditions in the brain (Supplementary Figure 2A).

The spleen and BM from mouse models of systemic and neuronal inflammation were used to identify changes in the expression of GlyRα1, α2, and α3. Notably, neuroinflammation significantly increased GlyRα1 (2.22 ± 0.62, p=0.026*), GlyRα2 (1.41 ± 0.18, p=0.014*), and GlyRα3 (2.83 ± 0.76, p=0.004**) expression in spleen (Figure 5A) compared to the control. Systemic inflammation showed no significant changes in the spleen (Figure 5A). The association between systemic or neuronal inflammation and the changes in GlyR expression is represented as a heat map for the spleen (Figure 5C). During both the inflammations, spleen tissue showed increased IL-1β, IL-2, and TNFα expression (Supplementary Figure 2B). A significant splenic GlyR modulation was detected in neuroinflammation mice.

During neuroinflammation, the BM showed a significant increase in the expression of GlyRα1, α2, and α3 (Figure 5B, p<0.0001, p<0.01, p<0.05). In contrast, systemic inflammation reduced the expression of all three GlyRα subunits in the BM (Figure 5B). The BM showed reduced pro-inflammatory cytokine expression in systemic and neuroinflammation (Supplementary Figure 5C). The heat map represented the association of RNA overexpression in GlyRα1, α2, and α3 specifically in response to neuroinflammation (Figure 5D). Systemic inflammation-induced decreases of GlyR expression indicated a suppressive regulatory mechanism in the BM.

Immunofluorescence staining showed that inflammatory conditions influence the expression of GlyR subunits in circulating WBCs. In control WBCs, GlyRα1/2 and α3 signals were mainly expressed in the membrane. In contrast, a significant increase in GlyRα1/2 and α3 staining was observed in the cytoplasm and membrane in neuronal and systemic inflammation models (Figure 6A). The increased co-expression of GlyRα1/2 and α3 was observed in WBCs under inflammatory conditions.

Next, WBCs were collected from mouse models of neuronal and systemic inflammation. The naïve mice served as the controls. Both systemic and neuronal inflammatory conditions altered the expression of GlyRα1, α2, and α3 in WBCs (Figure 6B). The gene modification of circulatory WBC GlyR by neuroinflammation showed a significant increase in GlyRα1 (3.4 ± 0.72-fold, p=0.002**), GlyRα2 (2.03 ± 0.17-fold, p<0.0001**), and GlyRα3 (3.44 ± 0.61-fold, p=0.0003**) compared to the control (Figure 6B). GlyRα3 in WBCs revealed a higher correlation to neuroinflammation. The responses of GlyR to systemic and neuronal inflammation were compared to the control (Figure 6C). During both inflammations, WBCs showed decreased expression of IL-1β, IL-2, and TNF-α (Supplementary Figure 2D).

The neuropathological process of neuroinflammation at days 2 and 5 altered the expression of GlyRs and pro-inflammatory cytokines in the opposite direction. A significant reduction in IL-1β, IL-2, and TNFα cytokines was observed in the brain (Figure 7A) and blood (Figure 7B) compared with the naive controls (day 0) following neuroinflammation at days 2 and 5. Pathological changes of neuroinflammation coincided with transient cytokine suppression in both the brain and blood. In contrast, a significant increase in WBC GlyRα1, a2, and a3 was observed on day 2, while a reduction of all three subunits of GlyR was seen following day 5. These results indicated that the increase in WBC GlyRs was not correlated with IL-1β, IL-1, and TNFα signaling in the brain and WBCs (Figure 7C). This demonstrated a reciprocal modulation of WBC GlyRs by neuroinflammation that is independent of brain and WBC cytokine signaling.

Next, we investigated how neuropathological changes affected the WBC GlyRs under neuroinflammatory conditions. Pathological evaluation was performed using an immunofluorescence assay in the brain slides with an antibody against Iba1. The naïve brain sections were used as the controls (day 0). Immunofluorescence imaging of the LPS-injected left hemisphere cerebral cortex showed a significantly increased activation of Iba1+ microglia (Figure 8A, red) at days 2 and 5 of post-LPS injection compared to the controls. The neuropathological change in Iba1+ microglia reflects neuroinflammation induced by LPS, with DAPI-counterstained nuclei (Figure 8A, blue). A noticeable increase in Iba1+ microglial density evidenced the neuropathological process caused by LPS on days 2 and 5. Quantification of Iba1+ cell density (percentage index) in brain sections indicated that robust acute microglial activation occurred on day 2 and persisted through to day 5 (Figure 8B).

The mRNA expression of GlyRα1 (3.82 ± 1.08-fold, p=0.025*), GlyRα2 (1.95 ± 0.24-fold, p=0.022**), and GlyRα3 (4.15 ± 0.84-fold, p=0.003**) in WBCs showed a statistically significant increase at day 2, following a decreased WBC GlyRs at day 5 (Figures 8C–E). This coincided with elevated microglial activation. The early peak in GlyR expression in WBCs reflects a dynamic peripheral response mediated by brain microglia activation. These results provide evidence that WBC GlyR signaling occurs in response to neuropathological gliosis through unexplored or glycinergic signaling mechanisms.

Finally, we analyzed whether circulating WBC GlyR signaling correlated with activation of Iba1+ microglia to identify the brain-GlyR-WBCs communication axis. The communication was reflected by pathological changes in Iba1+ microglia (neuroinflammation) and the expression of GlyRα1, α2, and α3 in circulating WBCs (Figures 8F–H). With a transient rise in activation of Iba1+ microglial cells, increases in GlyR expression tend to be high on day 2, suggesting a crosstalk of brain-GlyR-blood during neuroinflammation. Brain Iba1 and blood GlyRα1 and α3 show an acute surge on day 2, indicating synchronized central-peripheral immune activation. While brain Iba1 remains moderately elevated at day 5, blood GlyRα1 and α3 expression sharply decline, suggesting rapid systemic normalization following peak inflammation (Figures 8F, H). GlyRα2 transcript exhibits a slight early upregulation on day 2, inversely correlating with the decline in brain Iba1 expression by day 5 (Figure 8G). These findings demonstrate that LPS-induced neuroinflammation leads to rapid microglia activation accompanied by a synchronized, transient upregulation of circulating WBC GlyRα1 and α3. The reciprocal temporal association suggests that circulating WBC GlyR expression may serve as a potential biomarker for neuroinflammation.