Circulating extracellular choline acetyltransferase regulates inflammation

Abstract

Background

Choline acetyltransferase (ChAT) is required for the biosynthesis of acetylcholine, the molecular mediator that inhibits cytokine production in the cholinergic anti-inflammatory pathway of the vagus nerve inflammatory reflex. Abundant work has established the biology of cytoplasmic ChAT in neurons, but much less is known about the potential presence and function of ChAT in the extracellular milieu.

Objectives

We evaluated the hypothesis that extracellular ChAT activity responds to inflammation and serves to inhibit cytokine release and attenuate inflammation.

Methods

After developing novel methods for quantification of ChAT activity in plasma, we determined whether ChAT activity changes in response to inflammatory challenges.

Results

Active ChAT circulates within the plasma compartment of mice and responds to immunological perturbations. Following the administration of bacterial endotoxin, plasma ChAT activity increases for 12–48 hours, a time period that coincides with declining TNF levels. Further, direct activation of the cholinergic anti-inflammatory pathway by vagus nerve stimulation significantly increases plasma ChAT activity, whereas administration of bioactive r-ChAT inhibits endotoxin-stimulated TNF production and anti- ChAT antibodies exacerbates endotoxin-induced TNF levels, results suggesting that ChAT activity regulates endogenous TNF production. Administration of r-ChAT significantly attenuates pro-inflammatory cytokine production and disease activity in the dextran sodium sulfate preclinical model of inflammatory bowel disease. Finally, plasma ChAT levels are also elevated in humans with sepsis, with the highest levels observed in a patient who succumbed to infection.

Conclusion

As a group, these results support further investigation of ChAT as a counter-regulator of inflammation and potential therapeutic agent.

Introduction

Choline acetyltransferase is the rate-limiting enzyme in acetylcholine biosynthesis. Acetylcholine, the first neurotransmitter identified, plays an important role in a wide range of physiological processes, including muscle contraction, cardiovascular function, neural plasticity, attention and memory. Although acetylcholine is a neurotransmitter, other studies identified ChAT expression in a number of non-neuronal cells, including immune cells (1, 2). ChAT-expression in immune cells has been linked to regulation of cytokine production, T cell responses, hematopoiesis and blood pressure (3–5). ChAT-expressing T cells are required for antiviral immunity and vasomodulation during viral infection (6), whereas ChAT-expressing CD4+ T cells provide a cellular mechanism for suppression of TNF production during endotoxemia (7), and for vasorelaxation in the regulation of blood pressure (8). Further, acetylcholine produced by ChAT-expressing B lymphocytes regulates steady-state and emergency hematopoiesis (9). We recently demonstrated that in addition to ChAT-expressing CD4+ T cells, administration of recombinant ChAT to hypertensive animals normalizes blood pressure (10). Based on these observations, particularly our findings that ChAT administration regulates cardiovascular function, and that it product, acetylcholine, inhibits cytokine release, here we reasoned that circulating ChAT may be an anti-inflammatory mediator.

ChAT catalyzes the biosynthesis of acetylcholine in a highly efficient and rapid manner from choline and acetyl-CoA (11, 12). As these substrates are adequately available in the cytoplasm, the level of ChAT is the rate-limiting step in acetylcholine production. Although initially thought to be an exclusively intracellular enzyme, ChAT is released by a variety of cells including immune cells upon activation. Human embryonic cells and primary astrocytes in culture also release ChAT (13) and high levels of functional ChAT have been observed within extracellular spaces such as cerebral spinal fluid and plasma (13). Acetylcholine released by immune cells have been implicated in the regulation of inflammation. Immune cells express molecular components of the cholinergic system, including ChAT and both muscarinic and nicotinic acetylcholine receptors (4, 14, 15), but not the vesicular Ach transporter (VAChT), so it is unclear how acetylcholine is stored and released by non-neuronal cells (16). However, splenocytes can be stimulated to release ChAT into the extracellular milieu (13, 17). Given the important role of acetylcholine in inhibiting inflammation, it is plausible that extracellular ChAT harnesses choline and acetyl-CoA in plasma to synthesize acetylcholine and attenuate inflammatory responses. In support of this hypothesis, previous reports have identified changes in ChAT and acetylcholine levels in disease syndromes, including spontaneous hypertension (18). Another recent study demonstrated that therapeutic administration of recombinant ChAT reduces blood pressure by increasing acetylcholine synthesis in a murine model of hypertension (10).

Based on these observations, we reasoned that increasing extracellular ChAT activity would inhibit inflammation. Here, we measured plasma ChAT activity in mice subjected to different inflammatory conditions, and in patients diagnosed as having sepsis. To evaluate the potential efficacy of exogenous ChAT administration as an anti-inflammatory therapeutic, we administered pegylated recombinant ChAT (r-ChAT) in preclinical models of sepsis and colitis, which resulted in attenuated inflammation.

Materials and Methods

Recombinant ChAT protein and ChAT PEGylation.

All recombinant ChAT (rChAT) protein used was produced as described previously (10). Briefly, recombinant human ChAT corresponding to residue 119–748 (EC2.3.1.6, UnitProt 28329–3) with a N-histidine tag was expressed in E. coli BL21-Gold (DE3) cells, and purified using high affinity Ni-charged columns. The purified recombinant ChAT was dialyzed at 4°C in buffer containing 0.2 × PBS, 10% glycerol and 0.5 mM TCEP (tris(2-carboxyethyl) phosphine), and subjected to extensive Triton X-114 extraction to remove contaminating endotoxins. After purification, PEGylation was carried out using MS (PEG)12 reagent to extend the plasma half-life of the enzyme (10). Following PEGylation, the protein was dialyzed in buffer and once again extracted with Triton-X-114 to remove contaminating endotoxins. A Bradford protein assay (Bio-Rad) was used to determine rChAT protein concentration. Commercial recombinant ChAT protein was purchased from My BioSource.

Choline Acetyltransferase activity assay.

We developed an assay to measure ChAT activity using an adapted version of the colorimetric assay (Reactions 1–3) (13).

Briefly, biological samples and choline chloride standards are diluted in assay buffer (containing 10mM Tris-HCl, 150mM NaCl, 1mM EDTA, and 0.05% Triton X-100). 100μl of each standard and 40μl of each sample are added in triplicate to a 96-well plate. In wells containing 40μl of samples, 60μl of Cocktail A (250μM Choline chloride, 500μM acetyl coenzyme A, and 10μM physostigmine in assay buffer) are added. After incubation at 37°C for 20 minutes, 50μl of Cocktail B (1U/ml choline oxidase, 50U/ml HRP in Assay Buffer) are added to all wells and the plate is incubated at room temperature for 15 minutes. The assay is developed by addition of 150μL of 3’3’5’5’ Tetramethylbenzidine (TMB) to all wells, and the plate is immediately analyzed spectrophotometrically at 650nm. ChAT Activity is determined from the interpolated amount of choline depleted during the reaction, divided by the reaction time and sample volume, and multiplied by the sample dilution factor (Equation 1). In all biological samples, ChAT activity is normalized to total protein concentration, measured using Bradford protein assay (Bio-Rad) and is represented as nmol/min/mg of total protein.

Equation 1:

ChAT Enzyme Activity Calculation.

$$\text{Choline}\text{Depleted}\left(\frac{\frac{\mathit{\text{nmol}}}{\mathit{\text{min}}}}{\mathit{\text{mL}}}\right)=\frac{\left[\left(\left[\mathit{\text{Choline}}\mathit{\text{Blank}}\right]\right)-\left(\left[\mathit{\text{Plasma}}\mathit{\text{Sample}}\right]\right)\right]}{\mathit{\text{Reaction}}\mathit{\text{Time}}\left(\mathit{\text{mins}}\right)*\mathit{\text{Sample}}\mathit{\text{Volume}}\left(\mathit{\text{mL}}\right)}x\mathit{\text{DF}}$$

For assays with α-NETA (Abcam), the diluted biological sample or recombinant protein was incubated with α-NETA (0–20 μM) at room temperature for 30 minutes prior to the colorimetric assay. For thermal denaturation, the recombinant protein or biological sample were heated in a thermal cycler at 70°C or 95°C respectively for 30 minutes at 1500 x rpm.

Determination of Michaelis-Menten kinetic constants.

To determine the Km constants for rChAT samples, we utilized a previously published Coenzyme A detection assay (12). Briefly, Coenzyme A standard was diluted in assay buffer. 100μl of each standard and 40μl of each sample was added to the 96-well plate in triplicate. 60μl of Cocktail A was added to all sample wells. Immediately, 50μl of Cocktail B was added to all sample and standard wells. The plate was immediately quantitated spectrophotometrically at 324nm every 30 seconds for 30 minutes. Enzyme velocity was quantified as CoA formed (nmol/min/ml).

Animal Studies.

All animal procedures were approved by the Institutional Animal Care and Use Committee and the Institutional Biosafety Committee of the Feinstein Institutes for Medical Research, Northwell Health, Manhasset, NY in accordance with NIH guidelines. Animals were kept at 25°C, 55–60% humidity, on a 12-h light-dark cycle with ad libitum access to food and water. Wild type, male, C57/Bl6 mice, 8–10 weeks of age, were purchased from Jackson Laboratory (Bar Harbor, ME) or Charles River (Wilmington, MA) and maintained in fully AAALAC accredited facilities at the Feinstein Institutes for Medical Research.

Endotoxemia.

Lipopolysaccharide (LPS, Escherichia coli 0111:B4; Sigma) was sonicated for 30 min, vortexed and administered intraperitoneally (LD50 7 mg/kg). Animals were euthanized by CO₂ asphyxiation 90 min post-LPS administration and blood was collected by cardiac puncture using heparinized syringes . Blood samples were kept on ice and centrifuged for 10 minutes at 5000g. Plasma TNF levels were quantified using a commercially available ELISA (Invitrogen). Plasma total protein concentration was determined using a Bradford Protein Assay (Bio-Rad). illness scores were determined at respective timepoints up to 72 hours following endotoxin administration as described previously (19), where a maximum score of 28 per mouse denotes the greatest clinical sickness score. Mouse plasma ChAT concentration was determined using ELISA (MyBioSource: MBS724080).

Administration of PEG-ChAT and Anti-ChAT Neutralizing Antibody.

Pegylated recombinant ChAT protein or a custom made monoclonal anti-ChAT antibody (GenScript) was administered intraperitoneally at a 1mg/kg dose 30 minutes prior to LPS administration. Acute endotoxemia was induced with a dose of 0.3mg/kg LPS. Animals were returned to their home cages and euthanized 90 minutes after LPS administration.

Cervical Vagus Nerve Stimulation.

Aseptic technique was used for all surgical procedures. Vagus nerve isolation was carried out as previously described (20). Briefly, mice were anesthetized in a supine position with isoflurane (oxygen flow 2L/min; isoflurane 1.5%). Cervical vagus nerve was isolated and placed on a custom-built bipolar cuff electrode (Microprobes) as described previously (20). Electrical pulses were delivered to the nerve using a constant impedance stimulator, MC Stim II using the following stimulation parameters: Four-minute duration, pulse width (16.1ms) amplitude (250μA) and frequency (30Hz). The same surgical procedures were performed on sham operated mice but without electrical stimulation. After stimulation, the skin was stapled closed, and the animals were returned to their home cages for recovery. Two hours following stimulation, blood was collected by cardiac puncture using heparinized syringes.

DSS Colitis Model.

DSS (DSS salt, colitis grade (MW: 36000–50000), MP Biomedicals, CA) was added to the drinking water in a final concentration of 3% (wt/vol) for 7 days. The water was changed to regular water on day 8, and maintained till the end of the experiment. Survival and the disease activity index (DAI) were evaluated daily in a blinded manner. On day 14, mice were euthanized under CO₂ asphyxiation and blood and tissue were collected. Colons were removed, cleaned, and sectioned longitudinally. The luminal contents were removed with cold phosphate buffered saline (PBS). Starting from the most distal end (rectum) the colon was rolled with luminal side facing upward, resulting in a “Swiss roll” with the distal end in the center and the proximal colon in the outer side of the roll. The Swiss roll was held together with a small pin and placed in 10% buffered formalin overnight at 4°C. The fixed tissue was then embedded in paraffin, sectioned at 10 μm and stained with hematoxylin and eosin. Histological analysis was carried out in a blinded manner by three independent researchers. The degree of colon inflammation was scored under the light microscope (Zeiss, Germany) using the following criteria: inflammation severity (0 – none; 1 – mild; 2 – moderate; 3 – severe), inflammation extent (0 – none; 1 – mucosal; 2 – mucosal and submucosal; 3 – transmural), Crypt damage (0 – none; 1 – basal, 1/3 damaged; 2 – basal, 2/3/damaged; 3 – crypt loss), damage distribution (0 – 0%; 1 – 1–25%; 2 – 26–50%; 3 – 51–75%; 4 – 76–100%). Each mouse tissue was scored over 6 different sites of the colon. Serum cytokine levels (IL-6, TNFα and MCP-1) were analyzed using mouse V-Plex Biomarker Group 1 array, and the assay was performed according to the manufacturer’s instructions.

Human Plasma Samples.

Blood samples were obtained from normal healthy controls and patients with sepsis or septic shock recruited to the Northwell Health System between 2018–2019 as described previously (21). All participants provided informed consent, and the study was approved by the institutional review board (IRB) of the Feinstein Institutes for Medical Research (Clinicaltrials.gov # NCT03389555). Patients were diagnosed with sepsis or septic shock by the Sepsis 3 criteria (22) and blood samples were obtained within 24 hours of the sepsis diagnosis (defined as “time 0”). Sample size was determined by availability, and no blinding or randomization was applied for these non-interventional observations. Sample cohort consisted of (n=11) septic patients with an age range of 62 – 94, and (n=7) control patients with an age range of 60 – 88 years old. ChAT activity was assessed in human plasma samples using our established assay and ChAT concentration determined by ELISA (MyBioSource: MBS766113).

Data analysis and statistics.

Data were analyzed using Graphpad Prism 9.0 software using two-tailed unpaired Mann Whitney t-tests for comparisons between 2 groups or two-way ANOVA or mixed-effects model followed with Šídák’s multiple comparisons test for comparisons of 3 or more groups. Michaelis Menten kinetic constants were determined using the Michaelis-Menten approximation to calculate Km and Vmax from a velocity vs. substrate relationship. Correlation coefficients were calculated using a nonparametric Spearman Correlation. For all analyses, P ≤ 0.05 was considered statistically significant.

Results

Active ChAT is present in normal mouse plasma.

Using ELISA, we first confirmed that ChAT protein was detectable in the plasma of normal C57BL6 mice 8–12 weeks old (65.9 ±12.8 ng/mL; n=5). We then established a novel enzyme assay to measure the activity of ChAT in biological samples. Recombinant ChAT (r-ChAT) was expressed in Escherichia coli, purified to homogeneity (10), and used to establish the ChAT enzyme assay (Supplementary Fig. 1). A concentration-dependent increase in ChAT activity (Fig. 1A) is observed with a quasi linear activity curve between dilutions of 1:200 and 1:50. Consistent with r-ChAT, plasma ChAT activity is significantly reduced after thermal denaturation at 95°C (Fig. 1B).

Figure 1: Active ChAT is present in normal serum and increases following an acute endotoxin challenge.

(A) Concentration dependent detection of ChAT enzymatic activity in murine plasma samples obtained from normal mice. (B) Thermal denaturing of plasma at 95°C for 30 minutes significantly attenuates ChAT activity (**** P < 0.0001). (C-E) Male C57BL/6 mice were given IP injection of lipopolysaccharide (LPS) or saline, and monitored and euthanized at various timepoints. Plasma TNF levels (C) were significantly elevated at 1.5 hours post LPS administration in comparison to saline (**** P < 0.0001). Illness score (D) was significantly elevated in mice that received LPS at all timepoints in comparison to saline administration. (E) ChAT activity (nmol/min/mg) increased significantly in mice given as compared to saline at 12 and 24 hours (* P < 0.05) and 48 hours (**** P < 0.0001). Data were analyzed using a two-way ANOVA followed by Šídák’s multiple comparisons test. (mean ± SEM, n = 5 to 10 per group).

Plasma ChAT activity increases following endotoxin challenge.

To determine whether endogenous ChAT activity is altered in the setting of acute inflammation, we utilized an established model of endotoxemia. Animals were exposed to a sublethal dose of endotoxin (LD₅₀ dose, 7mg/kg administered intraperitoneally, and plasma ChAT activity analyzed at varying time points over the following 72 hours. LPS administration significantly increased plasma TNF levels at 90 minutes (Fig. 1C) with a significant increase in illness score over time (Fig. 1D). In mice exposed to endotoxin, a significant increase in ChAT activity was observed beginning after 12 hours peaking by 48 hours before returning to baseline by 72 hours (Fig. 1E).

Circulating ChAT activity increases following vagus nerve stimulation.

Activation of the cholinergic anti-inflammatory pathway by vagus nerve stimulation increases acetylcholine levels and requires ChAT-expressing T cells to attenuate TNF in endotoxemia (7). Activation of lymphocytes also induces the release of both ChAT and acetylcholine (5, 7, 13). Because vagus nerve stimulation increases acetylcholine levels (7), we reasoned that it may also produce increased circulating ChAT levels. To evaluate this possibility, we measured plasma ChAT activity in mice receiving sham versus electrical stimulation of the left cervical vagus nerve trunk (n=24 per group). Plasma ChAT activity levels are significantly elevated in plasma 2 hours after vagus nerve stimulation (Fig. 2), suggesting that vagus nerve signalling enhances ChAT activity in circulation.

Figure 2: ChAT activity increases following cervical vagus nerve stimulation.

Cervical Vagus Nerve stimulation (VNS) was performed on C57Bl6 mice for 4mins (250 μA; 30 Hz; 250 ms pulse width). Mice recovered in home cage and 2 hours post stimulation mice were euthanized via exsanguination. ChAT activity increased significantly in VNS group compared to a sham control (*P < 0.05). Data were analyzed using an unpaired t-test. (Data are represented as individual mouse data points with mean ± SEM, n = 5 to 10 per group).

Circulating ChAT regulates inflammatory responses.

To evaluate the role of circulating ChAT in regulating inflammatory responses, we administered highly purified pegylated recombinant ChAT (PEG-ChAT, 1mg/kg) or vehicle 30 minutes prior to LPS administration. Within 90 min, animals receiving vehicle exhibit increased levels of serum TNF, whereas administration of PEG-ChAT significantly attenuates serum TNF levels as compared to the vehicle group (Fig. 3A). Administration of PEG-ChAT also induces a decrease in splenic TNF levels (Fig. 3B). To determine whether neutralization of circulating ChAT would exacerbate the disease, we neutralized ChAT activity using an antibody to ChAT. Passive immunization of un-anesthetized mice with a single dose of an anti-ChAT antibody (1 mg/kg) 30 min before a sublethal dose of LPS (0.3mg/kg) significantly increases endotoxin-induced TNF levels in both the serum and spleen (Fig. 3 C–D).

Figure 3: ChAT administration attenuates inflammation, whereas Anti-ChAT administration exacerbates acute inflammation.

C57/BL6 male mice were administered vehicle or 1mg/kg PEG-ChAT or 1 mg/kg anti-ChAT antibody 30 minutes prior to 0.3mg/kg lipopolysaccharide (LPS) injection. Ninety minutes later, mice were euthanized, and tissue was collected and assayed for TNF by ELISA. Administration of PEG-ChAT significantly ameliorated serum TNF (A) compared to the vehicle group (** P < 0.01). In contrast, administration of anti-ChAT antibody significantly increases serum (C) and spleen (D) TNF in comparison to vehicle treated (* P < 0.05). Analyses are shown as mean ± SEM. Data were analyzed using a Mann-Whitney t-test.

PEG-ChAT administration attenuates experimental colitis.

Next, we evaluated whether administration of PEG-ChAT ameliorates inflammation in active disease using the well-established DSS-colitis model. Mice received 3% DSS in the drinking water for 7 days, when body weight and disease activity monitored daily (Fig 4A). DSS-exposed mice significantly lose weight from the 5th day after disease initiation (Fig. 4B). Intraperitoneal administration of PEG-ChAT (1mg/kg) twice daily from day 7 to day 14 after disease induction significantly reduces severity of colitis. Animals receiving PEG-ChAT with colitis show a significant attenuation of weight loss (Fig. 4B) and significant improvement in disease activity index (DAI) on day 9 as compared to vehicle-treated animals (Fig. 4C).

Figure 4: Administration of exogenous PEG-ChAT improved clinical disease outcomes in experimental colitis.

(A) Experimental colitis was induced in C57Bl/6 male mice using 3% Dextran Sodium Sulfate solution for 7 days. Mice were monitored daily for 14 days and were administered vehicle or PEG-ChAT intraperitoneally twice daily beginning on day 7. Administration of PEG-ChAT significantly reduces (B) weight loss (C) and attenuates disease activity in comparison to vehicle treated animals (B). (D) PEG-ChAT improved the macroscopic appearance of harvested colons. (E) Effect of PEG-ChAT on macroscopic signs of colonic inflammation. Representative images of H&E stained colon swiss roll sections show that inflammation severity, crypt damage, and extent of inflammation is reduced in ChAT and PEG-ChAT treated mice. (F) Blinded histological scoring demonstrates reduced severity in the colons of mice receiving PEG-ChAT in comparison to vehicle. Data is expressed as mean ± SEM. * P < 0.05; ** P < 0.01.

Shorter colon length and histological damage are a hallmark of experimental colitis (23), and DSS-exposed mice have significantly shortened colons (Fig. 4D). PEG-ChAT administration significantly improves colon length (Fig. 4D). To evaluate the effect of PEG-ChAT on histological damage, we carried out histopathological analysis in a blinded manner. Damage to the epithelium and crypt as well as the magnitude of edema in the mucosa and submucosa, were all significantly enhanced in vehicle-treated mice, but this was attenuated by administration of PEG-ChAT (Fig. 4 E–F). In addition, the circulating levels of pro-inflammatory cytokines (IL-6, TNF and MCP-1) are significantly reduced in animals receiving PEG-ChAT as compared to vehicle controls (Fig. 5 A–C).

Figure 5: Administration of PEG-ChAT attenuates serum pro-inflammatory cytokine levels in colitis.

Blood was collected through cardiac puncture on day 14 of the model. Serum levels of IL-6 (A), TNF (B), and MCP-1 (C) levels were measured at day 14 of the model using MSD multiplex assay. Data is expressed as individual mouse data with mean ± SEM. *P < 0.05; **P < 0.01.

Circulating ChAT activity is increased in patients with sepsis.

Animal models of human diseases, including the murine endotoxemia and colitis models used here, have inherent limitations (24–26). As an initial step in determining whether ChAT participates in the pathogenesis of human inflammatory diseases, we studied 7 normal subjects and 9 critically ill septic patients. ChAT concentration determined by ELISA show that patients had a significant three-fold increase compared to healthy controls (98.6±33.0 pg/mL versus 30.3±17.9 pg/mL respectively; p=0.023). This finding is reflected by low levels of ChAT activity are detectable in the serum of normal subjects, which are significantly increased in critically ill patients with sepsis (Fig. 6). The highest level was observed in a sepsis patient who succumbed.

Figure 6: ChAT activity is increased in human serum during sepsis.

Serum samples were obtained from (n=11) septic patients at the time of sepsis diagnosis, and (n=7) age-matched control patients. Serum ChAT activity (nmol/min/mg) is significantly increased in sepsis patients compared to controls. All data are represented as individual patient data points (a=patient who died from sepsis) with mean ± SEM. *** P = 0.0003.

Discussion

Physiological anti-inflammatory mechanisms, which are conserved through evolution, are efficient processes to regulate inflammatory conditions. Understanding these mechanisms may provide insight into novel strategies which can be exploited for the treatment of inflammatory disorders. Here, we report circulating ChAT as an anti-inflammatory mediator of inflammation. ChAT activity increases in the setting of acute inflammation in proportion to illness severity, and this increase provides a therapeutic physiological response, because administration of a neutralizing ChAT antibody amplifies the inflammatory response to endotoxin. Conversely, increasing ChAT activity by systemic administration of r-ChAT suppresses inflammation acutely, as well as in a model of inflammatory bowel disease. Vagus nerve stimulation, which activates the cholinergic anti-inflammatory pathway, also induces an increase in plasma ChAT activity. Finally, the observation that ChAT serum activity appears to be increased in patients with sepsis suggests that increased circulating ChAT may be a response to cytokine storm.

In addition to the observations presented here, significant increases in ChAT activity have been observed in the cerebral spinal fluid (CSF) of patients diagnosed with multiple sclerosis and dementias of different etiologies, including Alzheimer’s disease (13, 27). The cellular source of extracellular ChAT is currently unclear and requires future study. Numerous studies have demonstrated the presence of ChAT in human and rodent lymphoid cells (28–30). Splenic lymphocytes stimulated by LPS in vitro release ChAT (but not ACh) into the medium (13), so potentially circulating lymphocytes which are subject to inflammatory stimulation may release ChAT into the circulation. B cells have also recently been documented to synthesize acetylcholine for use in controlling hematopoiesis in the bone marrow which is a compartment of the systemic circulation (9, 31). In addition to lymphocytes, monocytes also express ChAT which is upregulated in renal allograft rejection, and a small fraction of NK cells at baseline express ChAT which is upregulated by inflammation and has been documented to modulate the activity of monocytes and macrophages in brain inflammation (32). Therefore, given the plethora of immune cells expressing ChAT, it is plausible that circulating leukocytes are a source of plasma ChAT activity. Alternatively, or in addition to these components of the immune system, endothelial cells also express ChAT which given their continuous exposure to the blood may release ChAT when subjected to inflammatory stressors (33). Finally, mesenchymal cells have been documented to secrete acetylcholine subsequently modulating inflammation (34). Given the widespread non-neuronal expression of ChAT, it is likely that other non-neuronal cell types not yet studied may also express ChAT which could be secreted to participate in protective anti-inflammatory responses.

The identification here of circulating extracellular ChAT contributing to regulation of inflammation reveals a previously unknown mechanism in immune homeostasis. The observation that neutralization of circulating ChAT results in an increased inflammatory phenotype in mice indicates that circulating ChAT plays a homeostatic role in regulation of inflammation. From this perspective, circulating ChAT may make acetylcholine available to cells devoid of direct cholinergic innervation by catalyzing its production directly at target sites using plasma choline and acetyl-CoA, providing a biochemical mechanism for acetylcholine-mediated immune regulation.

In sum, these observations establish a role of for circulating ChAT in the regulation of inflammatory processes and suggest that it may be possible to enhance acetylcholine release in the circulation to therapeutically regulate inflammation, insights that may have implications for developing a novel therapy for inflammatory diseases. Moreover, our findings that circulating ChAT activity is regulated by signals transmitted in the vagus nerve may offer another strategy to control the activity of ChAT using neuromodulation strategies which prevent or regulate inflammation.

Supplementary Material

1

Supplementary Figure 1: Detection of recombinant ChAT activity. Representative images of the ChAT enzymatic assay, choline standard and the measured activity in a dose response using in-house prepared recombinant ChAT protein (rChAT) (A). A significant increase in enzymatic activity of the rChAT is observed compared to a commercially available ChAT (B) (**P < 0.01). The Michaelis-Menten approximation (C) and Lineweaver Burke (D) double-reciprocal relationships of rChAT (2μg/mL) show that the Vmax = 0.05 Δ CoA (nmol/min) and the Km = 247.4 (nmol) for choline respectively. Preincubation of 0.4mg/mL rChAT with a competitive inhibitor alpha-NETA demonstrated an IC₅₀ of 0.1485 μM (E). Thermal denaturing of rChAT at 70°C for 30 minutes significantly reduced the detected activity in comparison to native preparation (F) (** P < 0.01).

Abbreviations:

ChAT

choline acetyltransferase

CSF

cerebral spinal fluid

DAI

disease activity index

LPS

lipopolysaccharide

PEG-ChAT

pegylated recombinant choline acetyltransferase

rChAT

recombinant choline acetyltransferase

TNF

tumor necrosis factor

VAChT

the vesicular ACh transporter

Data availability statement

All data supporting the findings of this study are available within the paper and its supplementary materials.