Dissecting bi-directional neuroimmune crosstalk in an in vitro model relevant to apical periodontitis Authors: Nandita Menon, Tatiana D. Mayorova, Adriano Senatore, Anil Kishen Published: 2026 ABSTRACT Inflammatory response associated with Apical Periodontitis (AP) is shaped by neuroimmune signaling mediated by neuropeptides and cytokines that regulate interactions among sensory neurons, immune cells, and stromal cells in the periapical microenvironment. To investigate these mechanisms, this study employed an in vitro multicellular culture system modeling key neuroimmune features relevant to AP rather than anatomical lesion formation. Neuron-like cells (NLCs) were generated from periodontal ligament cells (PDLCs) using dual-SMAD inhibition, and bi-directional paracrine and juxtacrine signaling between NLCs, PDLCs, and THP-1-derived macrophages (MQs) was examined under lipopolysaccharide (LPS)-induced inflammatory conditions. Neuropeptides released by NLCs, Substance P (SP) and Calcitonin Gene-Related Peptide (CGRP), enhanced LPS-driven inflammatory responses in PDLC-MQ paracrine systems, characterized by increased macrophage expression of CD68 and CD80 and elevated secretion of IL-1β and TGF-β1. In contrast, during PDLC-MQ juxtacrine interactions, NLC-derived neuropeptides supported a regulatory profile, with reduced expression of CD68, CD80, and STAT1 and decreased production of TNF-α and IL-1β. Reciprocally, conditioned media from LPS-treated PDLC-MQ co-cultures modulated NLC signaling by increasing TRPV1 expression, while IL-10 present in the conditioned media suppressed TRPA1 expression, resulting in an overall CGRP-dominant and SP-suppressed neuropeptide profile. These findings identify a dynamic, bi-directional neuroimmune circuitry among neuron-like cells, macrophages, and periodontal ligament cells that differentially regulates inflammatory and regulatory signaling in a system relevant to Apical Periodontitis, providing mechanistic insight into neuroimmune coordination during periapical inflammation.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-026-56016-4. Introduction Apical Periodontitis (AP) is an inflammatory disease caused by bacterial infection of periradicular tissues. The role of sensory neurons in influencing disease progression and severity in AP has been documented; activated primary afferents can participate in neurogenic inflammation by releasing neuropeptides such as Substance P (SP) and Calcitonin Gene-Related Peptide (CGRP), which can promote plasma extravasation, edema, and pain in inflamed tissues1. Mechanistic studies have linked SP and CGRP to transient receptor potential (TRP) channels – vanilloid subtype TRPV1, and ankyrin subtype TRPA1 in sensory neurons, which are activated by inflammatory mediators, and can amplify neuropeptide release2. Consistent with this, neuropeptide signaling has been associated with host pro-inflammatory responses including increased infiltration of immune cells, processes that can contribute to bone resorption in chronic AP2. However, in preclinical models, ablation of neurons accelerated inflammatory responses with increased immune-cell presence and larger osteolytic lesions, suggesting context-dependent regulatory roles of sensory neurons in limiting excessive immune activation and supporting tissue repair in AP3. Thus, bi-directional neuroimmune crosstalk is a key regulator of inflammatory dynamics and repair in AP, mediated by pleiotropic molecules signaling through receptors expressed on neurons, immune cells, and stromal cells2. However, the signaling pathways that specify pro- versus anti-inflammatory outcomes in AP neuroimmune interactions remain incompletely defined. Macrophages (MQs) have been reported to constitute up to ~46% of inflammatory cells in AP4, playing a vital role in transforming the AP landscape from inflammation to wound healing by participating in both the inflammatory phase and resolution phase5. MQs perform such dual functions by secreting cytokines, chemokines, neuropeptides, and growth factors in response to inflammatory cues. However, MQs do not act in isolation. In addition to neuroimmune interactions that drive inflammation and tissue repair, recent studies indicate that human PDL fibroblasts, which comprise a major fraction (up to 60%) of the PDL cellular compartment, influence MQ behaviour in AP6,7. These observations support incorporating immune-stromal interactions when characterizing neuroimmune crosstalk in AP. For this study we used an in vitro multicellular system relevant to AP and generated neuron-like cells (NLCs) from human PDL cells (PDLCs) to enable mechanistic examination of neuroimmune signaling while avoiding anatomical over-claims8. PDLCs, predominantly fibroblasts with a stem-cell subpopulation, are of ectomesenchymal origins, which permits neuronal differentiation under defined conditions9–11. The dual-SMAD inhibition technique12 was used to induce neuronal programs in PDLCs12; subsequent maturation employed media enriched with brain-derived neurotrophic factor (BDNF), glial cell line-derived neurotrophic factor (GDNF), and human beta nerve growth factor (β-NGF). Using these NLCs, we examined the differential effect of LPS-induced neuron-like signaling on PDLC-MQ interactions across autocrine, juxtacrine and paracrine configurations, and assessed reciprocity by testing how conditioned media from LPS-treated PDLC-MQ co-cultures reprogram NLC signaling. We hypothesized that NLC-derived neuropeptides differentially modulate PDLC-macrophage signaling in paracrine versus juxtacrine settings, and that inflammatory cues from PDLC-MQ co-cultures reciprocally tune NLC TRP channel-dependent pathways. Accordingly, our in vitro system is framed as relevant to Apical Periodontitis rather than a reproduction of lesion architecture. Materials and methods Cell lines Human periodontal ligament cells (PDLCs) were generously gifted by Dr. Douglas W. Hamilton (The University of Western Ontario, Ontario, Canada) and previously characterized13. PDLCs were cultured in complete DMEM (Gibco, Grand Island, NY) supplemented with 10% heat-inactivated fetal bovine serum (Sigma-Aldrich, St Louis, MO) and 1% antibiotic/antimycotic (Gibco, Grand Island, NY). THP-1 monocytes (ATCC TIB-202; American Type Culture Collection, Rockville, MD) were differentiated into macrophages (MQs) by incubating in complete RPMI-1640 (Gibco, Grand Island, NY) supplemented with 10% heat-inactivated fetal bovine serum (Sigma-Aldrich, St Louis, MO) and 1% antibiotic/antimycotic (Gibco, Grand Island, NY) containing 100nM phorbol-12-myristate-13-acetate (Sigma Aldrich, St Louis, MO) for 24h and resting in fresh medium overnight before use14. Cells from the passages 3-5 were used in this study. Neuronal differentiation of PDLCs PDLCs were seeded on to petri dishes coated with 25µg/ml poly-D lysine (Sigma Aldrich, St Louis, MO). Media was refreshed daily, using media supplemented with inhibitor molecules to facilitate dual-SMAD inhibition12. Differentiation was considered complete on day 14 (Fig. 1). Details of media and inhibitor molecules used is available in SI Table 112.Fig. 1Dual-SMAD inhibition timeline and protocol applied to PDLCs to generate neuron-like cells (NLCs). Dual-SMAD inhibition timeline and protocol applied to PDLCs to generate neuron-like cells (NLCs). Characterization of neuron-like cells derived from PDLCs Gene and protein expression analyses were performed at distinct time points to characterize the neuronal differentiation of PDLCs. Electrophysiological recordings were conducted to compare cell excitability between undifferentiated and differentiated PDLCs. Quantitative polymerase chain reaction (qPCR) Total RNA was extracted from cells on days 0, 5, 8, and 14 using RNeasy Kit for RNA Purification (Qiagen, Germantown, MD). Total RNA (2 μg) was reverse transcribed using SensiFAST™ cDNA Synthesis Kit (Bioline, Cincinnati, OH) and used to perform reverse transcription qPCR with PowerUp™ SYBR® Green Master Mix (Thermo Fisher Scientific, Waltham, MA) using protocol: initial denaturation at 95°C for 2mins, 39 cycles of denaturation at 95°C for 2mins followed by annealing, extension at 60°C for 1min. Samples were normalized to control gene GAPDH. Relative expression is reported using the 2^(ΔCt) method. Primer sequences used are listed in SI Table 2. Immunofluorescence staining Neuron-like cells (NLCs) were stained for Nestin (VWR, Mississauga, ON), PAX-6 (Fisher Scientific, Hampton, NH), RUNX1 (GeneTex, Irvine, CA), SOX10 (Fisher Scientific, Hampton, NH), and TRPV1 (Sigma Aldrich, St Louis, MO) expression, and counterstained with FITC (Santa Cruz Biotechnology, Dallas, TX) using standard staining protocol. Imaging was performed using Nikon Epifluorescence microscope (Nikon Eclipse Ti2-E, Nikon Instruments, Tokyo) at 10X objective magnification. IMARIS 10.0.1 software (Oxford Instruments, Oxfordshire, UK) was used for image quantification. The intensity sum and area sum were recorded after thresholding signal intensities against background. The intensity sum was then standardized with the area sum and plotted for comparisons of different treatment conditions. For all samples, single-channel and merged images were acquired from the same field of view with consistent acquisition settings. Due to the 3D architecture of spheroid NLC cultures, marker expression may not spatially coincide with all DAPI-positive nuclei in 2D projections. Electrophysiological recordings MultiClamp-700B amplifier and Digidata-1440A A/D converter (Axon Instruments, Molecular Devices, San Jose, CA) were used to perform current-clamp recordings. Electrodes pulled using a Sutter P-97 puller (Sutter Instruments, Novato, CA) were filled with intracellular solution containing (in millimolar): 125 KCl, 4 MgCl2, 10 HEPES, 10 EGTA, and 5 MgATP, adjusted to 310 mOsm/L with sucrose, pH 7.2 with KOH12. The external solution contained (in millimolar): 141 NaCl, 4.7 KCl, 1.2 MgCl2, 1.8 CaCl2, 10 glucose, and 10 HEPES, adjusted to 330 mOsm/L with sucrose, pH 7.4 with NaOH15. The resistance of the recording pipette was 2–6 MΩ. Electrophysiological recordings were performed in whole cell current clamp mode. Cells were hyperpolarized between sweeps by injecting −100 pA. The stimulus was applied for 500ms, during which cell excitability was probed by injecting hyperpolarizing or depolarizing currents between −500 and 500 pA. Clampfit 10.7 software (Axon Instruments, Molecular Devices, San Jose, CA) was used for data analysis. The membrane time constant (τ) was determined by fitting a single exponential function to the voltage response following a depolarizing current injection. The fit was applied to the initial portion of the voltage trace following the onset of the current step. Curve fitting was performed using a standard exponential equation of the form:\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f\left(t\right)=\sum_{i=1}^{n} {A}_{i}{e}^{-t/{\tau}_{i }}+C$$\end{document} Where f(t) is the membrane potential at time t, A represents amplitude of exponential decay, e(−t/τi) represents decay of signal over time, and C is the y-offset that represents the value the signal approaches as time approaches infinity. Experimental design PDLCs, MQs, and NLCs cells were seeded in Long-Term Culture Media (LTCM) for all experiments (Fig. 2). Lipopolysaccharide (LPS) from P.gingivalis (Invivogen, San Diego, CA) was used to simulate inflammation (Fig. 3).Fig. 2Schematic showing (A) PDLC, MQ monoculture and co-culture conditions and experimental timeline, and (B) Cellular Pathway experimental conditions and timeline.Fig. 3Schematic showing (A) NLC monoculture conditions and experimental timeline, and (B) Pain Pathway experimental conditions and timeline. Schematic showing (A) PDLC, MQ monoculture and co-culture conditions and experimental timeline, and (B) Cellular Pathway experimental conditions and timeline. Schematic showing (A) NLC monoculture conditions and experimental timeline, and (B) Pain Pathway experimental conditions and timeline. Bi-directional neuroimmune crosstalk was dissected into ‘Cellular pathway’ and ‘Pain pathway’ as described below:Cellular Pathway: To assess paracrine effects of LPS-treated NLC on MQ behaviour and PDLC-MQ interactions.MQ, PDLC Monocultures: MQs (1 × 105 cells/ml), PDLCs (1 × 105 cells/ml) were seeded in 24-well plates, incubated in complete DMEM for 24h, and then treated with 1. 300ng/ml LPS for 3h (ML3, PL3), 2. 300ng/ml LPS for 24h (ML24, PL24), 3. conditioned media from NLCs treated with 300ng/ml LPS for 3h (NL3), and 4. conditioned media from NLCs treated with 300ng/ml LPS for 24h (NL24).Paracrine PDLC-MQ (PM) Interactions: MQs (1 × 105 cells/ml) were seeded in Transwell inserts placed over PDLCs (1 × 105 cells/ml) seeded in 24-well plates. These were incubated in complete DMEM for 24h and then treated with 1. 300ng/ml LPS for 3h (PML3), 2. 300ng/ml LPS for 24h (PML24), 3. NL3, and 4. NL24.Inhibitor Studies: MQs (1 × 105 cells/ml) were seeded in Transwell inserts and PDLCs (1 × 105 cells/ml) were seeded in 24-well plates and incubated in complete DMEM for 24h. The next day, MQs were treated with 1. CGRP-receptor (CGRP-R) heterodimer CLR-RAMP1 inhibitor (CGRP II 8-37, Sigma Aldrich, St Louis, MO, 10−7 M16) for 3h, 2. SP-receptor, NK-1R inhibitor (Aprepitant, Sigma Aldrich, St Louis, MO, 10−7 M17) for 3h. PDLCs, after 24h incubation were treated with 1. CGRP II 8-37 (10−7 M) for 3h, and 2. Aprepitant (10−7 M) for 1h. After respective inhibitor treatments, cells were washed with PBS, before treating MQs monocultures and MQs in Transwell inserts placed over PDLCs (paracrine co-cultures) with NL3.Juxtacrine PDLC-MQ (PMJ) Interactions: MQs (1 × 105 cells/ml) and PDLCs (1 × 105 cells/ml) were co-seeded in 24-well plates. These were incubated in complete DMEM for 24h and then treated with 1. 300ng/ml LPS for 3h (PMJL3), 2. 300ng/ml LPS for 24h (PMJL24), 3. NL3, and 4. NL24.Pain Pathway: To assess paracrine effects of LPS-treated MQs and PDLC-MQ co-cultures on NLC behaviour.Effects of Macrophage (M) Monocultures: NLCs (1 × 105 cells/ml) were seeded in 24-well plates and incubated in LTCM for 24h before treatment with 1. 300ng/ml LPS for 3h, 2. 300ng/ml LPS for 24h, 3. conditioned media from MQ treated with 300ng/ml LPS for 3h (ML3), and 4. conditioned media from MQ treated with 300ng/ml LPS for 24h (ML24).Effects of Paracrine PDLC-MQ (PM) Interactions: NLCs (1 × 105 cells/ml) were incubated in LTCM for 24h before treatment with 1. conditioned media from PML3 (combined media from Transwell insert and plate well), and 2. conditioned media from PML24 (combined media from Transwell insert and plate well).Inhibitor Studies: NLCs (1 × 105 cells/ml) were incubated in LTCM for 24h. The next day, NLCs were treated with 1. CGRP II 8-37 (Sigma Aldrich, St Louis, MO, 10−7 M) for 1h, 2. Aprepitant (Sigma Aldrich, St Louis, MO, 10-7 M) for 1h, and 3. IL-10 receptor (IL-10R) inhibitor (recombinant human IL-10 R alpha protein, Biotechne, Etobicoke, ON, 1ng/ml18) for 3h. After respective inhibitor treatments, NLCs were washed with PBS and then treated with PML3.Effects of Juxtacrine PDLC-MQ (PMJ) Interactions: NLCs (1 × 105 cells/ml) were incubated in LTCM for 24h before treatment with 1. conditioned media from PMJL3, and 2. conditioned media from PMJL24. Cellular Pathway: To assess paracrine effects of LPS-treated NLC on MQ behaviour and PDLC-MQ interactions. Cellular Pathway: To assess paracrine effects of LPS-treated NLC on MQ behaviour and PDLC-MQ interactions. MQ, PDLC Monocultures: MQs (1 × 105 cells/ml), PDLCs (1 × 105 cells/ml) were seeded in 24-well plates, incubated in complete DMEM for 24h, and then treated with 1. 300ng/ml LPS for 3h (ML3, PL3), 2. 300ng/ml LPS for 24h (ML24, PL24), 3. conditioned media from NLCs treated with 300ng/ml LPS for 3h (NL3), and 4. conditioned media from NLCs treated with 300ng/ml LPS for 24h (NL24). MQ, PDLC Monocultures: MQs (1 × 105 cells/ml), PDLCs (1 × 105 cells/ml) were seeded in 24-well plates, incubated in complete DMEM for 24h, and then treated with 1. 300ng/ml LPS for 3h (ML3, PL3), 2. 300ng/ml LPS for 24h (ML24, PL24), 3. conditioned media from NLCs treated with 300ng/ml LPS for 3h (NL3), and 4. conditioned media from NLCs treated with 300ng/ml LPS for 24h (NL24). Paracrine PDLC-MQ (PM) Interactions: MQs (1 × 105 cells/ml) were seeded in Transwell inserts placed over PDLCs (1 × 105 cells/ml) seeded in 24-well plates. These were incubated in complete DMEM for 24h and then treated with 1. 300ng/ml LPS for 3h (PML3), 2. 300ng/ml LPS for 24h (PML24), 3. NL3, and 4. NL24. Paracrine PDLC-MQ (PM) Interactions: MQs (1 × 105 cells/ml) were seeded in Transwell inserts placed over PDLCs (1 × 105 cells/ml) seeded in 24-well plates. These were incubated in complete DMEM for 24h and then treated with 1. 300ng/ml LPS for 3h (PML3), 2. 300ng/ml LPS for 24h (PML24), 3. NL3, and 4. NL24. Inhibitor Studies: MQs (1 × 105 cells/ml) were seeded in Transwell inserts and PDLCs (1 × 105 cells/ml) were seeded in 24-well plates and incubated in complete DMEM for 24h. The next day, MQs were treated with 1. CGRP-receptor (CGRP-R) heterodimer CLR-RAMP1 inhibitor (CGRP II 8-37, Sigma Aldrich, St Louis, MO, 10−7 M16) for 3h, 2. SP-receptor, NK-1R inhibitor (Aprepitant, Sigma Aldrich, St Louis, MO, 10−7 M17) for 3h. PDLCs, after 24h incubation were treated with 1. CGRP II 8-37 (10−7 M) for 3h, and 2. Aprepitant (10−7 M) for 1h. After respective inhibitor treatments, cells were washed with PBS, before treating MQs monocultures and MQs in Transwell inserts placed over PDLCs (paracrine co-cultures) with NL3. Inhibitor Studies: MQs (1 × 105 cells/ml) were seeded in Transwell inserts and PDLCs (1 × 105 cells/ml) were seeded in 24-well plates and incubated in complete DMEM for 24h. The next day, MQs were treated with 1. CGRP-receptor (CGRP-R) heterodimer CLR-RAMP1 inhibitor (CGRP II 8-37, Sigma Aldrich, St Louis, MO, 10−7 M16) for 3h, 2. SP-receptor, NK-1R inhibitor (Aprepitant, Sigma Aldrich, St Louis, MO, 10−7 M17) for 3h. PDLCs, after 24h incubation were treated with 1. CGRP II 8-37 (10−7 M) for 3h, and 2. Aprepitant (10−7 M) for 1h. After respective inhibitor treatments, cells were washed with PBS, before treating MQs monocultures and MQs in Transwell inserts placed over PDLCs (paracrine co-cultures) with NL3. Juxtacrine PDLC-MQ (PMJ) Interactions: MQs (1 × 105 cells/ml) and PDLCs (1 × 105 cells/ml) were co-seeded in 24-well plates. These were incubated in complete DMEM for 24h and then treated with 1. 300ng/ml LPS for 3h (PMJL3), 2. 300ng/ml LPS for 24h (PMJL24), 3. NL3, and 4. NL24. Juxtacrine PDLC-MQ (PMJ) Interactions: MQs (1 × 105 cells/ml) and PDLCs (1 × 105 cells/ml) were co-seeded in 24-well plates. These were incubated in complete DMEM for 24h and then treated with 1. 300ng/ml LPS for 3h (PMJL3), 2. 300ng/ml LPS for 24h (PMJL24), 3. NL3, and 4. NL24. Pain Pathway: To assess paracrine effects of LPS-treated MQs and PDLC-MQ co-cultures on NLC behaviour. Pain Pathway: To assess paracrine effects of LPS-treated MQs and PDLC-MQ co-cultures on NLC behaviour. Effects of Macrophage (M) Monocultures: NLCs (1 × 105 cells/ml) were seeded in 24-well plates and incubated in LTCM for 24h before treatment with 1. 300ng/ml LPS for 3h, 2. 300ng/ml LPS for 24h, 3. conditioned media from MQ treated with 300ng/ml LPS for 3h (ML3), and 4. conditioned media from MQ treated with 300ng/ml LPS for 24h (ML24). Effects of Macrophage (M) Monocultures: NLCs (1 × 105 cells/ml) were seeded in 24-well plates and incubated in LTCM for 24h before treatment with 1. 300ng/ml LPS for 3h, 2. 300ng/ml LPS for 24h, 3. conditioned media from MQ treated with 300ng/ml LPS for 3h (ML3), and 4. conditioned media from MQ treated with 300ng/ml LPS for 24h (ML24). Effects of Paracrine PDLC-MQ (PM) Interactions: NLCs (1 × 105 cells/ml) were incubated in LTCM for 24h before treatment with 1. conditioned media from PML3 (combined media from Transwell insert and plate well), and 2. conditioned media from PML24 (combined media from Transwell insert and plate well). Effects of Paracrine PDLC-MQ (PM) Interactions: NLCs (1 × 105 cells/ml) were incubated in LTCM for 24h before treatment with 1. conditioned media from PML3 (combined media from Transwell insert and plate well), and 2. conditioned media from PML24 (combined media from Transwell insert and plate well). Inhibitor Studies: NLCs (1 × 105 cells/ml) were incubated in LTCM for 24h. The next day, NLCs were treated with 1. CGRP II 8-37 (Sigma Aldrich, St Louis, MO, 10−7 M) for 1h, 2. Aprepitant (Sigma Aldrich, St Louis, MO, 10-7 M) for 1h, and 3. IL-10 receptor (IL-10R) inhibitor (recombinant human IL-10 R alpha protein, Biotechne, Etobicoke, ON, 1ng/ml18) for 3h. After respective inhibitor treatments, NLCs were washed with PBS and then treated with PML3. Inhibitor Studies: NLCs (1 × 105 cells/ml) were incubated in LTCM for 24h. The next day, NLCs were treated with 1. CGRP II 8-37 (Sigma Aldrich, St Louis, MO, 10−7 M) for 1h, 2. Aprepitant (Sigma Aldrich, St Louis, MO, 10-7 M) for 1h, and 3. IL-10 receptor (IL-10R) inhibitor (recombinant human IL-10 R alpha protein, Biotechne, Etobicoke, ON, 1ng/ml18) for 3h. After respective inhibitor treatments, NLCs were washed with PBS and then treated with PML3. Effects of Juxtacrine PDLC-MQ (PMJ) Interactions: NLCs (1 × 105 cells/ml) were incubated in LTCM for 24h before treatment with 1. conditioned media from PMJL3, and 2. conditioned media from PMJL24. Effects of Juxtacrine PDLC-MQ (PMJ) Interactions: NLCs (1 × 105 cells/ml) were incubated in LTCM for 24h before treatment with 1. conditioned media from PMJL3, and 2. conditioned media from PMJL24. Inhibitor studies: Pharmacological inhibitors were used to evaluate the contribution of NK-1R-, CGRP-R-, and IL-10R-mediated signaling to their respective ligand-induced responses (SI Fig 1). Given the established association between NK1R activation and NF-κB signaling, attenuation of NF-κB gene expression following Aprepitant treatment was used as a functional indicator of NK1R inhibition in this system19. Increased AKT expression following CGRP-R inhibition is consistent with effective modulation of CGRP-mediated signaling20. In NLCs, concurrent upregulation of NF-κB and downregulation of AKT following IL-10R inhibition indicates disruption of IL-10-associated signaling pathways21,22. Cytokine and neuropeptide analysis Cell culture supernatant was collected post-treatment, centrifuged (10,000×g, 5mins), and stored at ˗80°C until analyses. Expressions of cytokines TNF-α, IL-1β, IL-6, IL-10, and TGF-β1 were analysed using multiplex immunoassay (MilliporeSigma, St. Louis, MO). Expressions of neuropeptides were analysed using Substance P Parameter Assay Kit (Bio-Techne, Etobicoke, ON) and CGRP ELISA Kit (MyBioSource, San Diego, CA). Immunofluorescence staining MQs were stained for cluster of differentiation (CD) markers CD68-AF647 (BioLegend, San Diego, CA), CD80-AF594 (Cedarlane, Burlington, ON), CD206-AF488 (BioLegend, San Diego, CA) and signal transducer and activator of transcription (STAT) markers STAT1-AF594 (BioLegend, San Diego, CA), STAT3-AF488 (BioLegend, San Diego, CA), STAT6-AF647 (Santa Cruz Biotechnology, Dallas, TX) antibodies. NLCs were stained for TRPV1-FITC (Sigma-Aldrich, St. Louis, MO) and TRPA1-FITC (Sigma-Aldrich, St. Louis, MO). Standard immunofluorescence staining protocol was applied23. Imaging and quantitative analyses were performed as described in section "Immunofluorescence Staining", using standardized intensity (intensity sum normalized to area sum) for comparisons of different treatment conditions. Statistical analysis All experiments were conducted in triplicates and data is represented as mean ± standard deviation of the mean. One-way and two-way ANOVAs with Tukey’s multiple comparison were used for datasets comparing multiple groups. Mann-Whitney U test was used to compare τ values. Statistical analyses were conducted using GraphPad Prism 10.1.1 software (GraphPad, La Jolla, CA). P value ≤ 0.05 (95% confidence interval) was considered statistically significant. Results Neuronal differentiation of PDLCs PDLCs transitioned from their characteristic spindle shape to compact, clustered cellular aggregates as neuronal differentiation progressed (Fig. 4A). By day 14, gene expression of early neural crest markers PAX6 and SOX10 was reduced by over 50% (not statistically significant), concomitantly with increased expression of neuronal markers NES and TRPV1 (Fig. 4B). RUNX1 expression transiently increased at day 8 before returning towards baseline by day 14, suggesting temporal regulation during differentiation. Immunofluorescence analysis demonstrated reduced protein expressions of PAX6 and SOX10 and increased NES, and TRPV1 protein abundance in NLCs at day 14 relative to day 0 (Fig. 4C). Differentiated NLCs also showed increased expression of the neuron receptor NTRK1 (NGF receptor) and detectable baseline expression of neuropeptides SP and CGRP, with SP release at higher levels than CGRP (SI Fig. 2).Fig. 4Changes in (A) morphology and (B) neuronal development gene expressions (n=3) were observed in PDLCs. Immunolabeling of neuronal development markers (C) are shown. Changes in membrane potential upon electrical stimulation: (D) averaged data collected in 11 NLCs (orange trace, solid and dashed lines for depolarizing and hyperpolarizing currents, respectively) and in 18 PDLCs (green trace, solid and dashed lines for depolarizing and hyperpolarizing currents, respectively); shaded area represents standard deviation (E) Comparison of individual traces of NLCs vs non-responsive and responsive PDLCs (F) Time constants for NLCs and responsive PDLC are shown. All data are mean ± SD (*p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001). PDLCs = undifferentiated cells, NLCs = neuron-like cells, Scale bar 100µm. Changes in (A) morphology and (B) neuronal development gene expressions (n=3) were observed in PDLCs. Immunolabeling of neuronal development markers (C) are shown. Changes in membrane potential upon electrical stimulation: (D) averaged data collected in 11 NLCs (orange trace, solid and dashed lines for depolarizing and hyperpolarizing currents, respectively) and in 18 PDLCs (green trace, solid and dashed lines for depolarizing and hyperpolarizing currents, respectively); shaded area represents standard deviation (E) Comparison of individual traces of NLCs vs non-responsive and responsive PDLCs (F) Time constants for NLCs and responsive PDLC are shown. All data are mean ± SD (*p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001). PDLCs = undifferentiated cells, NLCs = neuron-like cells, Scale bar 100µm. Functionally, all NLCs responded to the applied current clamp protocol, exhibiting stimulus-evoked membrane potential during depolarizing steps, and a depolarization-like response at the stimulus onset (Fig. 4D, E). No electrical activity was observed during the hyperpolarizing steps marked by smooth traces. In contrast, among undifferentiated PDLCs, only 8 of 18 cells responded to the stimulation protocol (Fig. 4D, E). Comparative analysis revealed significantly lower membrane time constants (τ) in NLCs relative to responsive PDLCs (Fig. 4F), indicating faster membrane charging and a narrower peak voltage response in differentiated NLCs. Cellular pathway Effect of LPS-treated NLC on macrophage monocultures LPS stimulation significantly increased expression of macrophage activation markers CD68, CD80, and STAT1 expression in MQ (Fig. 5A, SI SI Fig. 3). At 24h (ML24), STAT1 expression remained elevated, CD68 returned to baseline, and CD80 declined below baseline. while STAT3 was induced and STAT6 expressions remained sustained; CD206 expression was unchanged at both 3h and 24h. LPS treatment further increased secretion of TNF-α and IL-1β while TGF-β1 was transiently reduced at 3h and increased by 24h (Fig. 5B). Neuropeptide analysis showed suppression of SP and an approximately 10-fold increase in CGRP expression at 24h (Fig. 5B).Fig. 5Characterizing changes in macrophage polarization (A) and secretome profiles (B) comparing untreated macrophages (M) against macrophages treated with LPS for 3h (ML3), macrophages treated with LPS for 24h (ML24), macrophages treated with conditioned media from LPS-treated NLC for 3h (NL3_M) and 24h (NL24_M). All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001, # below limit of detection). Characterizing changes in macrophage polarization (A) and secretome profiles (B) comparing untreated macrophages (M) against macrophages treated with LPS for 3h (ML3), macrophages treated with LPS for 24h (ML24), macrophages treated with conditioned media from LPS-treated NLC for 3h (NL3_M) and 24h (NL24_M). All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001, # below limit of detection). Conditioned media from LPS-treated NLCs (NL3, NL24) reduced CD80 expression while inducing higher STAT1, CD206, and STAT6 levels relative to LPS alone (Fig. 5A, S.I Fig. 3). NL24 further increased STAT3 expression compared with ML24. Cytokine profiling demonstrated NL3 increased IL-1β, IL-6, and TGF-β1 (Fig. 5B), whereas NL24 further elevated TNF-α, IL-1β while reducing IL-6 relative to NL3. IL-10 secretion increased under NL24 conditions, coinciding with a decline in TGF-β1. Neuropeptide output was also altered, with CGRP markedly increased at NL24, while SP was suppressed at NL3 and increased at NL24, exceeding levels observed in NL3 and ML24. CGRP levels at NL24 were higher than NL3 but comparable to ML24. PDLC-macrophage co-culture response to LPS LPS treatment PDLC monocultures increased secretion of IL-1β, IL-6, IL-10, TGF-β1, SP, and CGRP (Fig. 6). By 24h, IL-1β levels were maintained, IL-6 and SP declined relative to 3h, and IL-10, TGF-β1 and CGRP showed sustained elevation.Fig. 6Characterizing changes in PDLC secretome profile upon LPS treatment for 3h (PL3) and 24h (PL24). All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001). Characterizing changes in PDLC secretome profile upon LPS treatment for 3h (PL3) and 24h (PL24). All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001). In paracrine PDLC-MQ co-cultures, PDLC signaling attenuated LPS-induced expression of CD68, CD80, STAT1, and STAT6 in MQ (Fig. 7A, SI Fig. 3). At 24h, CD68 expression further declined and CD80 was nearly extinguished, while STAT1 remained elevated relative to ML24 and STAT6 levels aligned with ML24. D206 was low across all groups. At 3h, PML increased TNF-α above baseline but below ML3, suppressed IL-10 and SP, and induced IL-6 secretion, which was absent in untreated M and ML3 (Fig. 7C). PDLC interactions restored TGF-β1 levels suppressed by LPS in MQs, and increased CGRP secretion by more than 5-fold. At 24h, TNF-α, IL-1β, IL-6, IL-10 and CGRP were elevated, while TGF-β1 declined relative to ML24, and SP remained suppressed compared with baseline but exceeded ML24 levels.Fig. 7(A) Effects of PDLC paracrine interactions on MQ polarization (PML3, PML24) and (B) effects of PDLC juxtacrine interactions on MQ polarization (PMJL3, PMJL24). Secretome profiles of (C) paracrine PDLC-MQ treated with LPS for 3h (PML3) and 24h (PML24) and (D) juxtacrine PDLC-MQ treated with LPS for 3h (PMJL3) and 24h (PMJL24). All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001, # below limit of detection). (A) Effects of PDLC paracrine interactions on MQ polarization (PML3, PML24) and (B) effects of PDLC juxtacrine interactions on MQ polarization (PMJL3, PMJL24). Secretome profiles of (C) paracrine PDLC-MQ treated with LPS for 3h (PML3) and 24h (PML24) and (D) juxtacrine PDLC-MQ treated with LPS for 3h (PMJL3) and 24h (PMJL24). All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001, # below limit of detection). Juxtacrine PDLC-MQ interactions produced a distinct immunophenotype. At 3h, CD68, CD80, and STAT3 were reduced while STAT6 expression increased (Fig. 7B, SI Fig 4). By 24h, CD68 and CD80 remained unchanged; STAT6 aligned with ML24 levels, and CD206 and STAT3expression increased. Cytokine and neuropeptide analyses showed comparable TNF-α, IL-1β, IL-10, and CGRP release between PM (Fig. 7C) and PMJ (Fig. 7D), whereas PMJL exhibited lower TNF-α and IL-1β at both timepoints, an early increase in IL-6 release, and suppression of TGF-β1 and SP. Effect of LPS-treated NLC on PDLC-macrophage co-cultures Paracrine: NL3 exposure increased MQ expression of CD68 and CD80 while reducing CD206 and STAT1, relative to PML3 (Fig. 8A, SI Fig 5). At 24h, NL24 produced expression profiles comparable to PML24 for CD68, CD80, and CD206, but uniquely induced STAT6 expression in MQ. Cytokine analysis showed comparable TNF-α, IL-1β secretion between PML3 and NL3 groups (Fig. 8C). NL3 significantly increased IL-6 and IL-10 secretion relative to PML3, while TGF-β1 was reduced. Neuropeptide profiling demonstrated lower SP and elevated CGRP release in NL3 compared with PML3.Fig. 8Characterizing macrophage polarization of (A) MQs in paracrine PDLC-MQ co-cultures treated with LPS (PML3, PML24), conditioned media from LPS-treated NLC for 3h (NL3_PM) and 24h (NL24_PM), (B) MQs in juxtacrine PDLC-MQ co-cultures treated with LPS (PMJL3, PMJL24), conditioned media from LPS-treated NLC for 3h (NL3_PMJ) and 24h (NL24_PMJ), secretome profiles of (C) paracrine PDLC-MQ co-cultures treated with LPS (PML3, PML24), conditioned media from LPS-treated NLC for 3h (NL3_PM) and 24h (NL24_PM), (D) juxtacrine PDLC-MQ co-cultures treated with LPS (PMJL3, PMJL24), conditioned media from LPS-treated NLC for 3h (NL3_PMJ) and 24h (NL24_PMJ). All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001, # below limit of detection). Characterizing macrophage polarization of (A) MQs in paracrine PDLC-MQ co-cultures treated with LPS (PML3, PML24), conditioned media from LPS-treated NLC for 3h (NL3_PM) and 24h (NL24_PM), (B) MQs in juxtacrine PDLC-MQ co-cultures treated with LPS (PMJL3, PMJL24), conditioned media from LPS-treated NLC for 3h (NL3_PMJ) and 24h (NL24_PMJ), secretome profiles of (C) paracrine PDLC-MQ co-cultures treated with LPS (PML3, PML24), conditioned media from LPS-treated NLC for 3h (NL3_PM) and 24h (NL24_PM), (D) juxtacrine PDLC-MQ co-cultures treated with LPS (PMJL3, PMJL24), conditioned media from LPS-treated NLC for 3h (NL3_PMJ) and 24h (NL24_PMJ). All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001, # below limit of detection). At 24h, NL24 reduced TNF-α and IL-1β secretion, while IL-6 levels mirrored those observed in the PML24 group. IL-10 and SP secretion were highest in the NL24 group, and although TGF-β1 increased relative to NL3, its level remained below those observed in PML24. CGRP secretion increased at NL24 compared with NL3 but did not reach PML24 levels. Juxtacrine:PMJ cultures exhibited minimal changes in CD and STAT immunoreactivity following LPS3h (L3) and NL3 exposure, with marked expression remaining near baseline levels (Fig. 8B, SI Fig 5). STAT6 showed a modest increase in PMJL3 and NL3, which did not reach statistically significance. At 24h, NL24 increased CD68, CD80, and STAT1 expression, while CD206, STAT3, and STAT6 expression was highest in PMJ cultures treated with NL24. Cytokine profiling revealed that NL3 significantly increased IL-6, IL-10, and SP expression at 3h relative to baseline and PMJL3 (Fig. 8D). At 24h, TNF-α, IL-6, and CGRP exceeded baseline levels but remained below those observed in LPS24h (L24) controls. IL-10 secretion was highest in NL24, while SP levels were higher than LPS24h but lower than NL3, a difference that did not reach statistical significance. IL-1β levels remained at baseline in both NL3 and NL24 groups, whereas TGF-β1 increased in NL24, reaching levels comparable to LPS24h. Pain pathway Effect of LPS-treated macrophage and PDLC monocultures on NLC signaling Direct LPS stimulation of NLCs increased TRPV1 and TRPA1 expressions at 3h (Fig. 9A, SI Fig 6). Neuropeptide analysis revealed suppressed SP release, while CGRP levels remained at baseline (Fig. 9B). By 24h, TRPV1 expression further increased, TRPA1 returned to baseline, SP remained suppressed, and CGRP secretion increased, reaching approximately 1.5-fold above baseline.Fig. 9Comparing neuropeptide release (A) and changes in TRP channel expression (B) in Neuron-Like Cells (NLCs) treated with LPS (NL3, NL24), conditioned media from LPS-treated macrophages (ML3, ML24) and LPS-treated PDLCs (PL3, PL24) for 3h and 24h. All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001, # below limit of detection). Comparing neuropeptide release (A) and changes in TRP channel expression (B) in Neuron-Like Cells (NLCs) treated with LPS (NL3, NL24), conditioned media from LPS-treated macrophages (ML3, ML24) and LPS-treated PDLCs (PL3, PL24) for 3h and 24h. All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001, # below limit of detection). Conditioned media from LPS-treated MQ (ML3) reduced CGRP secretion, suppressed TRPV1 expression, and extinguished TRPA1 expression compared with NL3 (Fig. 9A, B). Similar effects were observed with conditioned media from LPS-treated PDLC (PL3), which reduced neuropeptide release and TRPV1 expression while fully suppressing TRPA1. At 24h, ML24 further suppressed SP release, while markedly increasing CGRP secretion, paralleling trends observed in NL24. ML24 also increased TRPV1 expression (not statistically significant) and partially restored TRPA1 expression relative to ML3, although both remained lower than NL24 levels. In contrast, PL24 reduced SP relative to PL3 without inducing CGRP, and continue to suppress TRPV1 and TRPA1 expression, indicating sustained inhibition of nociceptive signaling under PDLC-derived conditions. Effect of LPS-treated PDLC-macrophage co-cultures on NLC signaling At 3h, conditioned media from both PML and PMJL cultures maintained low TRPV1 expression in NLCs, however, by 24h, a significant increase in TRPV1 expression was observed specifically in PML group (Fig. 10A, SI Fig 6). TRPA1 expression in NLCs was comparable to NL3 under PML group exposure at 3h but declined by 24h, whereas PMJL conditioned media induced sustained suppression of TRPA1 at 24h. Neuropeptide analysis demonstrated that CGRP secretion from NLCs increased over time in both PML and PMJL groups, with higher levels observed in PMJL group (Fig. 10B), while PMJL exposure led to a progressive, time-dependent reduction in SP secretion.Fig. 10Comparing neuropeptide release (A) and changes in TRP channel expression (B) in Neuron-Like Cells (NLCs) treated with conditioned media from LPS-treated paracrine PDLC-MQ (PML3, PML24) and LPS-treated juxtacrine PDLC-MQ (PMJL3, PMJL24) for 3h and 24h. All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001). Comparing neuropeptide release (A) and changes in TRP channel expression (B) in Neuron-Like Cells (NLCs) treated with conditioned media from LPS-treated paracrine PDLC-MQ (PML3, PML24) and LPS-treated juxtacrine PDLC-MQ (PMJL3, PMJL24) for 3h and 24h. All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001). Effects of NLC-derived SP (NSP) and CGRP (NCGRP) on PDLC-macrophage interaction Pharmacologic blockade of NSP and NCGRP reversed MQ expression of CD206 and STAT1 expression and significantly reduced secretion of IL-1β and TGF-β1 release (Fig. 11A, SI Fig 7). Blockade of either neuropeptide increased IL-10 secretion, whereas NSP blockade selectively promoted IL-6 release and NCGRP blockade attenuated it.Fig. 11Characterizing effects of neuropeptides on Cellular and Pain pathways. (A) Conditioned media from LPS-treated NLC was applied to MQs for 3h after inhibition of NK-1R (NL3_M+Si), and inhibition of CGRP-receptor (NL3_M+Ci), (B) conditioned media from LPS-treated NLC was applied to PDLC-MQ paracrine co-cultures for 3h after inhibition of NK-1R (NL3_PM+Si), and inhibition of CGRP-receptor (NL3_PM+Ci), (C) conditioned media from LPS-treated PDLC-MQ co-cultures was applied to NLCs for 3h after inhibition of NK-1R (PML3_N+Si), inhibition of CGRP-receptor (PML3_N+Ci), and inhibition of IL-10 receptor (PML3_N+10i). All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001, # below limit of detection). Characterizing effects of neuropeptides on Cellular and Pain pathways. (A) Conditioned media from LPS-treated NLC was applied to MQs for 3h after inhibition of NK-1R (NL3_M+Si), and inhibition of CGRP-receptor (NL3_M+Ci), (B) conditioned media from LPS-treated NLC was applied to PDLC-MQ paracrine co-cultures for 3h after inhibition of NK-1R (NL3_PM+Si), and inhibition of CGRP-receptor (NL3_PM+Ci), (C) conditioned media from LPS-treated PDLC-MQ co-cultures was applied to NLCs for 3h after inhibition of NK-1R (PML3_N+Si), inhibition of CGRP-receptor (PML3_N+Ci), and inhibition of IL-10 receptor (PML3_N+10i). All data are mean ± SD (n=3, *p≤0.05, ** p≤0.01, *** p≤0.001, ****p≤0.0001, # below limit of detection). In PDLC-MQ co-cultures, NLC-derived NSP and NCGRP supported MQ expression of CD68 and CD80 (Fig. 11B, SI Fig 7). Blockade of SP or CGRP increased IL-1β and TNF-α, secretion, further elevated IL-10, and contributed equally to IL-6 production. TGF-β1 secretion in NL3-treated PM cultures was not significantly altered by SP or CGRP blockade. Effects of PDLC-macrophage-derived SP (PMSP), CGRP (PMCGRP), and IL-10 (PMIL10) on NLC signaling Neutralization experiments demonstrated that PMSP, PMCGRP, and PMIL10 contributed to NLC TRPV1 expression in NLCs, with IL-10 exerting the weakest effect (Fig. 11C, SI Fig 8). PMSP and PMCGRP also contributed to TRPA1 expression in NLC treated under PML3 conditions, although the magnitude of change was smaller than that observed for TRPV1. Blockade of IL – 10 signaling increased TRPA1 expression, indicating that PDLC-MQ-derived IL – 10 actively suppressed TRPA1 activity in NLCs. Discussion In vitro neuronal differentiation provides a controllable human-cell-based approach to examine neuroimmune mechanisms while avoiding the translational limitations inherent to animal models. Using dual-SMAD inhibition, we observed a staged transition of PDLCs towards a neuron-like lineage, characterized by early downregulation of progenitor markers PAX6 and SOX10 and subsequent induction of RUNX1 and Nestin (NES), consistent with the emergence of peptidergic neuronal phenotypes12,24. Concurrent upregulation of NTRK1 and TRPV1 further supports acquisition of neuron-like properties associated with nociceptive signaling25 and neurogenic inflammation2. Basal expression of SP and CGRP in differentiated NLCs indicates functional peptidergic capacity, while the differential levels of SP and CGRP secretion likely reflect distinct vesicular release kinetics described for these neuropeptides26. Electrophysiological profiling revealed functional divergence between undifferentiated PDLCs and NLCs. While 44% of undifferentiated PDLCs exhibited electrical responsiveness, likely attributable to non-excitable ion channels supporting proliferation, migration, and homeostasis27, all NLCs responded to depolarizing stimulus and exhibited significantly lower membrane time constants (τ). The faster membrane charging is characteristic of neuron-like behavior and supports the ability to track rapid electrical inputs, a hallmark of excitable cells involved in synaptic transmission28. We acknowledge that in vitro neuronal differentiation is sensitive to protocol specific variables, including cell source, substrate composition, and seeding density29. To mitigate these limitations, we employed well-characterized human cell lines, standardized differentiation protocols, and internal quality control benchmarks to minimize experimental variability. The use of LPS as an inflammatory stimulus represents a well-established approach to model key features of Apical Periodontitis (AP) in vitro30. In particular, LPS derived from Porphyromonas gingivalis was selected given its recognized role in activating host inflammatory pathways and contributing to periodontal tissue breakdown, including bone resorption30. The concentration employed (300ng/ml) was chosen based on evidence demonstrating its sufficiency to induce neuronal activity and subsequent neuropeptide release in neuronal models in vitro31, while maintaining a controlled inflammatory response32. Additionally, the selected stimulation timepoints (3h and 24h) were intended to capture distinct phases of the response, the 3h timepoint reflects early signaling events and neuronal activity33, whereas 24h enables assessment of sustained responses, including cytokine production34. LPS stimulation of macrophage monoculture robustly increased CD80 and STAT1 expression, accompanied by elevated secretion of proinflammatory cytokines TNF-α, IL-1β, consistent with the activation of NF-κβ and TRIF-dependent signaling pathways35. SP signaling via Gαq-coupled NK-1R may further amplify this response through positive feedback-driven SP release, thereby contributing to enhanced TNF-α, IL-1β expression36–38. Sustained or repeated LPS exposure is known to drive macrophages towards a mixed inflammatory/regulatory phenotype, characterized by attenuated TNF-α and SP and enhanced anti-inflammatory mediator production39. Our observation of increased anti-inflammatory TGF-β1 in parallel with stable IL-10 secretion are consistent with activation of JAK/STAT3 axis, enabling autocrine upregulation of IL-10 and dampening of excessive inflammation40. PDLCs mounted a canonical inflammatory response to LPS, likely mediated through TLR4-dependent NF-κβ activation41. Elevated TGF-β1 release aligns with reported IL-6 - TGF-β1 coupling via Mapk/Erk-dependent mechanisms, supporting a coordinated regulatory program in stromal cells42. The sustained increase in CGRP secretion at 24h may suppress NF-κβ signaling, thereby containing IL-1β, reducing IL-6 and promoting IL-10 and TGF-β1 production43. LPS directly increased NLC expression of TRPV1 and TRPA1, channels implicated in inflammatory pain and periodontal tissue breakdown44,45. While TRPV1 expression continued to rise, TRPA1 expression declined over time, consistent with inverse functional coupling between these channels46. Low CGRP release despite high TRP expression suggests enhanced channel trafficking preceding functional activation, as reported previously47,48. Paracrine PDLC-MQ signaling increased STAT1 expression in MQs, albeit to a lesser extent than direct LPS stimulation, indicating partial modulation rather than complete suppression of an M1-like phenotype. Simultaneous attenuation of M2-associated markers and IL-10 secretion suggests that PDLC-derived paracrine cues impose regulatory constraints while preserving macrophage responsiveness to inflammatory stimuli. These findings are consistent with reported immunomodulatory function of PDL fibroblasts49,50. Sustained STAT1 activation likely promoted TNF-α51, IL-1β52, and IL-653 production, with autocrine feedback from TNF-α and IL-6 reinforcing STAT1 signaling54,55. In contrast, juxtacrine PDLC-MQ interactions increased CD206 expression, supporting contact-dependent induction of M2-like polarization state, as previously described56. STAT3 activation through IL6/JAK2 signaling likely enhanced TGF-β1 release57,58, which can suppress IL-1β via mitochondrial dysfunction-dependent mechanisms57–59. Reduced SP levels in contact-dependent cultures may reflect negative feedback regulation, with partial recovery at 24h suggesting attenuation of inhibitory signaling over time60. Collectively, our findings indicate that PDLC-MQ paracrine signaling favors inflammatory persistence, whereas juxtacrine interactions bias towards resolution and tissue repair, underscoring context-dependent stromal-immune regulation23. Neuron like cell-derived SP (NSP) and CGRP (NCGRP) further altered macrophage responses to LPS; blockade of SP abolished IL-1β secretion, consistent with its known role in activating NK-1R on macrophages and promoting pro-inflammatory cytokine release, including IL-1β, which is closely associated with nociceptive sensitization61. Both SP and CGRP can also promote IL-10 release via distinct pathways, including MAPK/ERK and HO-1 signalling for SP62, and cAMP/PKA signalling for CGRP63can promote IL-10 release from macrophages. However, SP blockade resulted in a more pronounced increase in IL-10, suggesting a stronger baseline pro-inflammatory drive compared with NCGRP, consistent with CGRP’s role more commonly associated with immunomodulatory or anti-inflammatory effects2. Similarly, both neuropeptides can enhance IL-6 release from macrophages via activation of PI3K, p42/44 and p38 MAP kinase pathways (SP)64 and the cAMP signaling pathway (CGRP)65. However, our findings indicate that NSP exerts a more pronounced influence on macrophage-derived IL-6 production compared with NCGRP. This more pronounced effect of SP may reflect its relatively strong capacity to activate proinflammatory signaling through the NK-1R receptor. In the co-culture system, the effects were more nuanced, suggesting modulation by PDLC-derived paracrine signals (Fig. 12A). PDLCs may contribute anti-inflammatory mediators such as IL-1066, which can shift macrophage phenotypes toward a more regulatory state, attenuating TNF-α and IL-1β release62,63. Given that both macrophages and PDLCs express NK-1R and CGRP receptors, parallel signalling pathways may be activated simultaneously. SP-mediated NF-κB/MAPK activation may drive IL-6 production65, while CGRP signalling via cAMP pathways may further modulate cytokine expression, including effects on mRNA stability in PDLCs67. This may explain the stronger IL-6 response observed under combined stimulation compared with individual treatments.Fig. 12Schematic showing influence of LPS-treated NLC conditioned media on paracrine (A) and juxtacrine (B) PDLC-macrophage signaling. NLC-derived SP preferentially induced IL-1β and limited TGF-β1 to sustain a mixed MQ phenotype; together with CGRP, it modulated inflammation in PDLC–MQ paracrine interactions, while both neuropeptides reduced inflammatory mediators under juxtacrine conditions. Schematic showing influence of LPS-treated NLC conditioned media on paracrine (A) and juxtacrine (B) PDLC-macrophage signaling. NLC-derived SP preferentially induced IL-1β and limited TGF-β1 to sustain a mixed MQ phenotype; together with CGRP, it modulated inflammation in PDLC–MQ paracrine interactions, while both neuropeptides reduced inflammatory mediators under juxtacrine conditions. NLC conditioned media increased SP in PDLC-MQ juxtacrine interactions while suppressing inflammatory cytokines (Fig. 12B), consistent with a CGRP-rich, regulatory microenvironment50,66. In this context, elevated SP may contribute to immune modulation and tissue repair, reflecting its pleiotropic functions beyond classical pro-inflammatory signaling68–70. Conditioned media from LPS-treated macrophage and PDLC monocultures suppressed TRPV1 and TRPA1, likely due to reduced IL-1β and TNF-α, and increased IL-10, all known regulators of TRP channel expression68,71. These data indicate cell-type-specific neuronal reprogramming in which macrophages-derived signals favour delayed TRPV1⁺/TRPA1⁺, CGRP-dominant states, while PDLC signals maintain TRPV1 without restoring TRPA1. Consistent with prior reports72, PDLC-MQ paracrine secretomes elevated TRPV1 expression in NLCs (Fig. 13A). Blockade of PDLC-MQ-derived SP (PMSP), CGRP (PMCGRP) reduced but did not abolish TRPV1 or TRPA1, indicating co-regulation by neuropeptides signaling, whereas IL-10 selectively suppressed TRPA1, as IL-10 receptor blockade restored TRPA1 expression, aligning with reports from dorsal root ganglia models68. These findings identify that IL-10-TRPA1 axis as a potential therapeutic target in AP-associated neuroinflammation.Fig. 13Schematic showing influence of LPS-treated PDLC-macrophage conditioned media on NLC signaling. Paracrine PDLC-macrophage-derived signals delayed TRPV1 induction without restoring TRPA1 (A). PDLC-macrophage juxtacrine signaling yielded a CGRP-dominant, SP-suppressed phenotype with SP/CGRP co-regulating TRPV1 and IL-10 suppressing TRPA1 (B). Schematic showing influence of LPS-treated PDLC-macrophage conditioned media on NLC signaling. Paracrine PDLC-macrophage-derived signals delayed TRPV1 induction without restoring TRPA1 (A). PDLC-macrophage juxtacrine signaling yielded a CGRP-dominant, SP-suppressed phenotype with SP/CGRP co-regulating TRPV1 and IL-10 suppressing TRPA1 (B). Juxtacrine PDLC-MQ secretomes maintained suppression of TRPA1 despite late increases in TRPV1, likely due to the persistent IL-10 signaling (Fig. 13B). Overall, stromal-myeloid crosstalk promoted a CGRP-dominant, SP-suppressed neuronal phenotype, reflecting chronic inflammatory conditioning of neuron-like cells by multicellular periodontal interactions73,74. This study employed a conditioned media-based approach to decouple cell-specific contributions to neuroimmune signaling. While this reductionist design enables identification of source-dependent effects, it does not capture the dynamic, bidirectional interactions present in vivo. Additional limitations include the use of discrete 3h and 24h timepoints, which do not fully capture the continuous temporal complexity of in vivo responses, and the use of equal seeding densities across cell types. This design was intentionally chosen to standardise per-cell contributions to the secretome and avoid confounding effects arising from differences in cell abundance-driven signal strength. While relative cell proportions differ in vivo, controlling cell number in this reductionist system enables isolation of distinct cell type-derived soluble cues and their downstream effects. This approach is consistent with established fibroblast-macrophage co-culture models, where cell ratios are treated as an experimental parameter rather than a strict anatomical reconstruction75,76. Clinical studies77 and in vivo studies78,79 have established roles for neuroimmune mediators in periodontal inflammatory conditions, however, direct evidence of cell-specific, bidirectional neuroimmune crosstalk remains limited. Accordingly, these findings provide controlled mechanistic insight and highlight the need for future in vivo validation. Conclusion This study establishes a physiologically relevant in vitro neuroimmune platform by deriving neuron-like cells (NLCs) from periodontal ligament cells (PDLCs) and applying it to dissect mechanisms underlying LPS-driven inflammation. We demonstrate that neuron-derived Substance P (SP) and Calcitonin Gene-Related Peptide (CGRP) critically shape macrophage responses, amplifying inflammatory signaling in PDLC-MQ paracrine interactions while modulating macrophage polarization and mediator release (IL-1β and TGF-β1) under juxtacrine conditions. By isolating neuronal inputs, this work defines the distinct contribution of neuron-derived neuropeptides to immune–stromal crosstalk. Reciprocal signaling from PDLCs and macrophages initially suppresses NLC activity but progressively promotes a CGRP-dominant profile with dynamic regulation of TRP channels, underscoring the integration of inflammatory cues by neuronal TRPV1 and TRPA1. IL-10 derived from PDLC-MQ interactions selectively suppress TRPA1, whereas immune-stromal neuropeptides enhance TRPV1, revealing how neuronal sensory pathways are differentially tuned within an inflammatory microenvironment. Collectively, these findings identify multicellular neuroimmune interactions as key regulators of inflammatory signaling and transient nociceptive programming in a system relevant to Apical Periodontitis, providing a mechanistic framework to guide future studies aimed at directing inflammation resolution and tissue repair through neuroimmune modulation. Supplementary Information Supplementary Information. Supplementary Information.