Interleukin-6 induces nascent protein synthesis in human dorsal root ganglion nociceptors primarily via MNK-eIF4E signaling

Highlights • IL-6 induces nascent protein synthesis in human dorsal root ganglion neurons.• This effect occurs mostly in TRPV1+ presumptive nociceptors.• IL-6-induced nascent protein synthesis is mostly blocked by the MNK1/2 inhibitor eFT508.• Activity-dependent protein synthesis occurs in human nociceptors.


Introduction
Chronic pain affects the lives of billions across the world with patients suffering from debilitating spontaneous pain, hypersensitivity, and many comorbidities, yet current analgesic treatments remain largely ineffective (Finnerup et al., 2015;Dahlhamer et al., 2018;Mills et al., 2019;Price and Ray, 2019).The development and maintenance of chronic pain physiology is believed to be rooted in neuroplasticity with these changes occurring along the neuraxis.Important sites of neuroplasticity driving chronic pain are the DRG, the spinal dorsal horn, and multiple brain regions (Price and Inyang, 2015).In nociceptors, the sensory neurons responsible for detecting noxious stimuli, a key plasticity mechanism is an activity-dependent translation of new proteins from existing pools of mRNAs expressed by these neurons (Jimenez-Diaz et al., 2008;Geranton et al., 2009;Melemedjian et al., 2010;Melemedjian et al., 2014;Moy et al., 2017;Megat et al., 2019b;Barragan-Iglesias et al., 2020).This process plays a crucial role in shaping the sensitivity and responsiveness of nociceptors in particular in response to growth factors like nerve growth factor (NGF) and cytokines like IL-6 (Melemedjian et al., 2010;Moy et al., 2017).This activity-dependent translation, which can be directly measured using nascent protein synthesis assays (Dieterich et al., 2006;Dieterich et al., 2007), can increase the excitability of nociceptors creating peripheral nociceptive signals that are ultimately perceived as pain (Moy et al., 2017;Megat et al., 2019b;Mihail et al., 2019;Li et al., 2024).
Translation of mRNA is comprised of three steps: initiation, elongation, and termination.Initiation is considered the rate-limiting step for translation and is the step that is regulated most tightly in neurons when activity-dependent translation is stimulated (Kelleher et al., 2004;Costa-Mattioli et al., 2009;Yousuf et al., 2021).Extracellular signals such as NGF and IL-6 in DRG neurons, and neurotransmitters in CNS neurons can activate two major pathways to control activity-dependent translation.The major pathways are the mechanistic target of rapamycin (mTOR) pathways and the MAP kinase-interacting kinase (MNK1/ 2)/eukaryotic initiation factor (eIF) 4E, p-eIF4E pathway.(Costa-Mattioli et al., 2009;Khoutorsky and Price, 2018;Yousuf et al., 2021).Studies in rodent DRG neurons suggest that while both pathways play a role in regulating the excitability of nociceptors, MNK1/2-eIF4E signaling is important for controlling the excitability of these neurons (Moy et al., 2018a;Megat et al., 2019b;Barragan-Iglesias et al., 2020;Jeevakumar et al., 2020;Shiers et al., 2020b), and is likely a more tractable pharmacological target for pain treatment for many reasons that have been described extensively elsewhere (Melemedjian et al., 2013;Khoutorsky et al., 2015;Megat and Price, 2018;Yousuf et al., 2021).Importantly, MNK1/2 is expressed in human nociceptors (Shiers et al., 2023) and its inhibition reduces the excitability of spontaneously active human DRG neurons recovered from patients with neuropathic pain undergoing thoracic vertebrectomy surgery (Li et al., 2024).
This study aimed to fill a gap in knowledge between a decade of studies in rodents linking IL-6-induced activity-dependent translation via MNK1/2-eIF4E signaling to enhanced nociception (Melemedjian et al., 2010;Melemedjian et al., 2014;Moy et al., 2017;Moy et al., 2018a;Jeevakumar et al., 2020;Shiers et al., 2020b) and our lack of knowledge concerning the actions of IL-6 on human DRG neurons.We developed an in vitro explant assay to study nascent protein synthesis in the DRG.Using this assay, we show that IL-6 causes nascent protein synthesis in human DRG neurons.This effect occurs in TRPV1+ neurons, demonstrating that it occurs across human nociceptors (Shiers et al., 2020a;Shiers et al., 2021;Tavares-Ferreira et al., 2022), and is largely attenuated by eFT508, a specific inhibitor of MNK1/2 (Reich et al., 2018).We conclude that this signaling mechanism of IL-6 is conserved in human DRG neurons.Our work supports the targeting of translation regulation mechanisms for pain treatment in humans.

IL-6-sIL-6R treatment enhances protein synthesis in human DRG neurons by FUNCAT
We hypothesized that the IL-6-sIL-6R complex would induce nascent protein synthesis in human DRG tissue.To test this, we performed AHA labeling and FUNCAT in human DRG explants recovered from organ donors.We first sought to test whether FUNCAT fluorescence intensity reliably reports the presence of newly synthesized proteins.If this were true, then anisomycin treatment should reduce FUNCAT signal compared to vehicle baseline, serving as a negative control for the translation dependence of AHA incorporation (Fig. 1A-G).We found anisomycin (60 µM) inhibited FUNCAT signal in cryosections from human DRG explants when compared to those treated with vehicle (Fig. 1B-C, E-F, H).We next asked if treatment with the IL-6-sIL-6R complex enhanced FUNCAT signal when compared to vehicle baseline.We tested this by administering a 20-minute pulse of the complex (10 ng/mL) to explants.Our analysis revealed that IL-6-sIL-6R induces increased FUNCAT signal (Fig. 1D, G, and H; Table 1).These findings demonstrate that the IL-6-sIL-6R complex drives nascent protein synthesis in human DRG explants.

IL-6-sIL-6R treatment enhances p-eIF4E IF signal in human DRG neurons
We next asked whether IL-6-sIL-6R treatment drives phosphorylation of eIF4E at serine residue 209 (p-eIF4E) (Fig. 3).We tested whether IL-6-sIL-6R enhances this signaling by immunostaining for p-eIF4E in sections from human DRG explants treated as described above.We then analyzed p-eIF4E immunofluorescence (p-eIF4E IF) signal per DRG neuronal soma (Fig. 3A-B).We found that p-eIF4E IF was increased after IL-6-sIL-6R treatment when compared to vehicle baseline (Fig. 3C).This change in signaling was also readily evident in our images where more than 50 % of neurons appeared to exhibit dramatically enhanced p-eIF4E IF intensity (Fig. 3D).The data demonstrate that phosphorylation of eIF4E at serine 209 is induced by IL-6-sIL-6R-signaling in human DRG neurons.This suggests that this signaling pathway may be responsible for increased nascent protein synthesis in IL-6-sIL-6R-treated hDRG neurons, a hypothesis we tested in a subsequent experiment.

IL-6-sIL-6R treatment drives MNK1/2-dependent phosphorylation of eIF4E and protein synthesis in human DRG neurons
We sought to determine whether MNK1/2 activity is required for IL-6-sIL-6R-driven increases in p-eIF4E IF and FUNCAT signals in human DRG neurons (Fig. 4).To test this possibility, we used the specific MNK1/2 inhibitor eFT508 (100 nM) as a co-treatment with 20 min IL-6-sIL-6R pulse (Fig. 4A).If IL-6-sIL-6R-driven enhancements in p-eIF4E and nascent proteins in human DRG neurons depend on MNK1/2, then eFT508 should attenuate increases in p-eIF4E IF and FUNCAT in these neurons.IL-6-sIL-6R treatment increased p-eIF4E IF and FUNCAT signals in human DRG neurons when compared to vehicle (Fig. 4B-F).On average, this equated to an increase in fluorescence intensity of about 54 % and 57 % for p-eIF4E IF and FUNCAT, respectively (Table 3).Cotreatment with eFT508 reduced IL-6-sIL-6R-driven increases in p-eIF4E IF signal below baseline levels by about 29 % on average (Fig. 4E; Table 3).This co-treatment also inhibited IL-6-sIL-6R treatment-induced enhancements in FUNCAT intensity from approximately 57 % to about 11 % above vehicle baseline (Fig. 4F; Table 3).As shown in Fig. 4E and  F, plotting our intensity data as a function of soma surface area size revealed that many individual neurons within the IL6-sIL-6R-treated explants exhibited increased p-eIF4E signals when compared to vehicle, while a large portion of those in the eFT508-cotreated explants displayed intensities below baseline (Fig. 4G).Similarly, analysis of FUNCAT signal in IL6-sIL-6R-treated explants showed notable increases compared to vehicle, whereas eFT508-cotreated explants did not exhibit similar decreases (Fig. 4H).Our quantitative data appear to indicate clustering towards smaller neuronal sizes for both p-eIF4E IF and FUNCAT; however, by using soma surface area, we did not account for well-known effects of intracellular fluorescent dye concentration on intensity detection which is a potential confound of the result (Fink et al., 1998;Waters, 2009;Morikawa et al., 2016).Again, FUNCAT signal was not decreased below baseline with eFT508 treatment.Taken together, the data demonstrates that IL6-sIL-6R treatment increased nascent protein synthesis in human DRG neurons via MNK-eIF4E signaling.Critically, engagement of this pathway is directly correlated with nascent protein synthesis on a per-neuron basis in human DRG.

Discussion
Activity-dependent translation of new proteins has long been recognized as a core mechanism for synaptic plasticity in the CNS    ( Kelleher et al., 2004;Costa-Mattioli et al., 2009) and for changes in nociceptor excitability in the PNS (Khoutorsky and Price, 2018;Yousuf et al., 2021).Our foundation of knowledge on this key mechanism for neuronal plasticity is almost entirely built on experiments done in rodents, largely mice.In this body of work, we demonstrate that IL6signaling causes an activity-dependent increase in nascent protein synthesis in human DRG neurons recovered from organ donors.This effect is entirely dependent on MNK signaling because it is blocked by eFT508, a specific inhibitor of both MNK1 and MNK2 (Reich et al., 2018), both of which are expressed by human DRG neurons (Shiers et al., 2023).The increase in nascent protein synthesis is also associated with increased p-eIF4E, a specific target of MNK1/2 signaling (Pyronnet et al., 1999), providing biochemical evidence of engagement of this pathway in human DRG neurons.Finally, our findings show that IL-6-sIL-6R-induced increases in nascent protein synthesis occur in TRPV1+ neurons in human DRG, demonstrating that this signaling pathway is engaged in human nociceptors (Shiers et al., 2020a;Shiers et al., 2021;Tavares-Ferreira et al., 2022).We conclude that this signaling mechanism, which is directly associated with IL-6-induced nociception in mice (Melemedjian et al., 2010;Moy et al., 2017), is similarly engaged in humans supporting the conclusion that IL-6induced pain in humans is also driven by MNK activation in nociceptors.
An important observation in our experiments is that MNK signaling only appears to be required for IL-6-sIL-6R-enhanced protein synthesis and not for constitutive protein synthesis in human DRG neurons.Treatment with eFT508 clearly decreased p-eIF4E levels below baseline in our experiments but it did not suppress baseline AHA-incorporation.In contrast, anisomycin, which blocks peptide bond formation, decreased AHA-incorporation well below baseline levels.The effect of eFT508 treatment in our experiments was specific for reducing the enhanced AHA-incorporation, measured as FUNCAT signal, when slices were co-treated with IL-6-sIL-6R.It has long been recognized that MNK-eIF4E signaling influences the translation of a subset of mRNAs that are involved in neuronal plasticity (Aguilar-Valles et al., 2018;Amorim et al., 2018;Moy et al., 2018b;Megat et al., 2019b), immune response (Joshi et al., 2009), and oncogenesis (Furic et al., 2010;Altman et al., 2013).Some of these mRNAs have been identified in cancer cells, immune cells and in DRG neurons (Furic et al., 2010;Aguilar-Valles et al., 2018;Amorim et al., 2018;Moy et al., 2018b), but the guiding principles for how p-eIF4E regulates the translation of specific subsets of mRNAs have still not been discovered (Scheper and Proud, 2002;Chen et al  2023).Given the importance of this pathway for pain, and our demonstration that it is engaged in human nociceptors, a future priority should be using techniques like ribosome profiling to understand precisely which mRNAs are translated when p-eIF4E levels are increased in human nociceptors.We recognize that the effect of eFT508 on FUNCAT signal was not complete, suggesting additional mechanisms control activity-dependent translation in human DRG neurons.The remaining ~10 % increase in FUNCAT signal likely arises from IL to 6-sIL-6R engagement of mTOR signaling that is independent of direct effects of the MNK-eIF4E signaling axis (Melemedjian et al., 2010).In human tissues, this eFT508-dependent reduction in nascent proteins was more dramatic than expected when compared to findings in rodent tissues (Melemedjian et al., 2010).This suggests the potential contribution of mTOR function to nociceptive plasticity in human nociceptors, a finding that would be consistent with rodent studies (Price et al., 2007;Jimenez-Diaz et al., 2008;Geranton et al., 2009;Liang et al., 2013;Uttam et al., 2018;Megat et al., 2019b;Megat et al., 2019a).However, we did not directly test the contribution of mTOR for two reasons.The first is that our previous studies show that blocking mTOR signaling causes feedback activation of MAPK signaling that exacerbates pain over time, and that this effect is mediated by DRG neurons (Melemedjian et al., 2013).The second is that mTOR inhibitors have potent immune-suppressing activity (Thomson et al., 2009), an effect that would not be advantageous for pain treatment from a side-effect perspective.Given the precious nature of the human DRG samples, we have chosen to focus on therapeutic mechanisms with the greatest opportunity for clinical translation.
We recently demonstrated that blocking MNK signaling in human DRG neurons recovered from patients undergoing thoracic vertebrectomy surgery reverses spontaneous action potential activity that is associated with neuropathic pain symptoms in these patients (Li et al., 2024).Previous RNA sequencing experiments on DRGs from these patients found that cytokines and chemokines are upregulated in DRGs from male and female patients who had neuropathic pain in associated dermatomes (North et al., 2019;Ray et al., 2023).Our work with IL-6-sIL-6R links these electrophysiological findings in patient DRG neurons to induction of this signaling pathway using a system that can be used to screen other cytokines or chemokines or even groups of cytokines and chemokines to examine whether MNK-eIF4E signaling is engaged by these factors.It can also be used in conjunction with proteomic or transcriptomic methods to identify the nascent proteins that are synthesized or the mRNAs that are translated, respectively.We envision that the assay described here can be used as a discovery platform for gaining a deeper understanding of the intricacies of human nociceptor plasticity.

Acute human DRG explants
Human tissue recovery was performed in accordance with preapproved guidelines set by the Institutional Review Boards at the University of Texas at Dallas (UTD).Procedures were consistent with those outlined in the Declarations of Helsinki.In collaboration with the Southwestern Transplant Alliance (STA), lumbar human DRGs were recovered from neurologic determination of dead organ donors as described previously (Shiers et al., 2021;Tavares-Ferreira et al., 2022).Donor information is provided in Table 4. Human DRGs were transported to UTD in ice-cold, oxygenated artificial cerebrospinal fluid containing N-methyl-D-glucamine (NMDG-aCSF) as described previously (Valtcheva et al., 2016).After arrival, whole DRGs were immediately moved to ice cold, actively oxygenating NMDG-aCSF, cleaned, and then rapidly sliced into 1 mm explants.Slices were then distributed into treatment groups and transferred to actively oxygenating recording/labeling-aCSF (pH 7.4; 32.5 • C) (R/L-aCSF) for 3.5 hrs or 4 hrs pre-treatment recovery.Formulas for aCSF solutions are provided in Table 5. Number of DRG Neurons analyzed with average somatic mean grey and percent difference from vehicle baseline per treatment condition for p-eIF4E IF and FUNCAT for each of five Organ Donor Replicates in Fig. 4. Parameters for laser power and HV were as shown.Italics denote parameters for bleed through testing where p-eIF4E IF detected using goat anti-rabiit-Alexa Fluor 488.Imaging settings are further described in Methods. 100 µ m serial cryosections were obtained every 400 µm, and all reactions and imaging were performed at the same time, respectively.Abbreviations: Fluorescent Non-Canonical Amino Acid Tagging (FUNCAT or FUN), Fractional difference from vehicle baseline (% Difference), Average Mean Grey of all vehicle-treated DRG neuronal somata from vehicle-treated explants analyzed (Vehicle Mean), Detector Voltage (HV).

FUNCAT and IF
Human DRG explants were fixed by rapid submersion into paraformaldehyde (PFA) (4 %) and fixed for 15 min in pH 6.8 PFA and then to pH 9-10 PFA for an additional 20 min and then to pH 7.4 PFA for a final 25 min.Slices were then flash frozen in powdered dry ice and stored at − 80 • C until serial cryosectioning (80 µm: Donors 1-4; 100 µm: Donors 5-6) into free-floating sections for click chemistry.Incorporated AHA was detected using Fluorescent Non-canonical Amino Acid Tagging (FUNCAT) via cycloaddition of alkyne-conjugated Alexa Fluor 647 (Thermo Scientific A10278) (Dieterich et al., 2006;Dieterich et al., 2007).All click chemistry and other reagents used in this study are listed in Table 6.Click chemistry reactions were carried out at room temperature while mixing for 20 to 24 hrs.Sections were then processed for immunofluorescence (IF) to detect either TRPV1 (4 8 8) or p-S209-eIF4E (488 or 555) and NeuN (488 or 555) (Shiers et al., 2021;Mitchell et al., 2023).All antibodies used in this study are listed in Table 6.Samples were treated with primary antibodies overnight at room temperature and incubated with secondary antibodies the next day for 1 hr before mounting.All FUNCAT and IF steps were done in a blinded fashion (the experimenter did not know the treatment condition).

Fluorescence intensity imaging and quantification of FUNCAT and p-eIF4E IF signals
An Olympus FV3000RS laser scanning confocal microscope was used for imaging of human DRG sections.Four-color 10x images were acquired of entire human DRG sections for fluorescence intensity analyses.Two to 18 images were acquired for 7 to 20 sections per treatment condition for 6 Donor replicates.Acquisition parameters for FUNCAT and/or p-S209-eIF4E IF were kept consistent across experiments.Gain (1.000) and offset (4 %) parameters were held constant for every section.Lipofuscin, a structure inherent to human nervous tissues that is highly autofluorescent, was detected as described previously for subtraction (Mitchell et al., 2023).For explant fluorescence intensity approximations, FUNCAT signal was measured across the field of view for the entire section surface area for 4 donors.For DRG neuronal somata, ROIs were identified in 12-bit images of TRPV1 IF (nociceptors) or NeuN IF (DRG neurons) channels.Somata were defined as having a minimum surface area of 350 µm 2 (Mitchell et al., 2023).Then, raw FUNCAT and/ or p-eIF4E IF signals were measured per ROI, respectively.Image analysis was not done in a blinded fashion.

Subtraction and transformation of autofluorescent Lipofuscin signal
Lipofuscin auto-signal was subtracted from each channel in 10x images using FIJI (NIH, Bethesda, MD).Auto-signal ROIs were identified using automated detection or manual tracing as described previously (Mitchell et al., 2023).Our automated macro parameters and codes are available upon request.Confocal images were then transformed by uniformly setting the lipofuscin signal (AU) to 1 AU.This permitted subtraction to mitigate its effect on FUNCAT and p-eIF4E IF intensity measurements per soma.This was done for all FUNCAT and p-eIF4E IF fluorescence intensities presented in this study using the following formula:

Statistical analyses
Graphs and Statistical analyses were generated using GraphPad Prism version 9.4 (GraphPad Software, Inc.San Diego, CA USA).All statistical tests used with associated biological replicate sizes, parameters, and p values are described in figure legends.Data in graphs is shown as dot plots for all data points and also represented as mean ± SEM.For both explant and somata analyses, transformed intensities were compared between treatment conditions either pairwise using ttests where N = Donor Replicates, or for 3 or more conditions using Kruskal-Wallis with Dunn's post hoc test with multiple comparisons where N = serial cryosections or neuronal somata.P values from multiple comparisons results are represented in all associated figures.All p values are expressed as *p < 0.05; **p < 0.005; ***p < 0.0005; ****p < 0.0001 in figures.

Fig. 1 .
Fig. 1.FUNCAT fluorescence intensity depends on protein synthesis and is increased after 20 min pulse with the IL-6-sIL6R complex in human DRG explants.(A) Pulse-Chase Treatment Strategy 1 for AHA-labeling of 1 mm human DRG explants for FUNCAT.Three parallel treatments per Donor: vehicle (<0.5 % DMSO), baseline anisomycin (60 µM), or 20 min IL-6-sIL-6R pulse (10 ng/mL) for N = 4 Donors.(B-G) Nascent proteins by FUNCAT in cryosections (80 µm) from human DRG explants treated with vehicle (B, E), anisomycin (C, F), or IL-6-sIL-6R (D, G) (Alkyne-Alexa-647, green).(B-D) 1.25x images of longitudinal DRG sections (scale bar, 2 mm).(E-G) Associated 10x images of DRG neuronal somata (scale bar, 500 µm).(H) Quantitation of FUNCAT intensity fold difference from vehicle baseline in cryosections from two technical replicates for four biological replicates.Scatter plot, values expressed as mean ± SEM. 80 μm thick serial sections were collected from 1 mm thick explants treated with vehicle, N=11; anisomycin, N = 14; IL-6-sIL-6R, N = 20, where N equals the number of sections.Violin plot, Kruskal-Wallis test with post hoc Dunn's test: vehicle vs anisomycin, p = 0.0464; vehicle vs IL-6-sIL-6R, p = 0.0086.*p < 0.05, ** p < 0.01.(For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) DRG explant replicates (1 mm), serial cryosection sample sizes (360 µm separation), and confocal imaging parameters used in approximation of mean FUNCAT signal per treatment condition for each Donor replicate in Fig. 1.Settings for Gain and Offset were 1.000 and 4.0 % and images acquired at 10x were used for all fluorescence intensity quantifications.For each Donor Replicate, click reactions were performed at the same time and 10x images were acquired in the same session.Abbreviations: Number (No.), Fluorescent Non-Canonical Amino Acid Tagging (FUNCAT), Fractional difference from vehicle baseline (% Difference), Average Mean Grey of all vehicle-treated sections analyzed (Vehicle Mean), Detector Voltage (HV).

Fig. 3 .
Fig. 3. IL-6 stimulation enhances p-eIF4E immunofluorescence intensity in a subset of human DRG neurons.(A-B) Immunostaining for p-eIF4E (phosphor-S209) in cryosections (80 µm) from human DRG explants treated with vehicle (A) or IL-6-sIL-6R (B) (p-eIF4E, red).1.25ximages of longitudinal DRG sections (scale bar, 2 mm).Insets show 10x images of human DRG neurons (scale bar, 500 µm).(C) Quantitation of somatic p-eIF4E IF intensity fold difference from vehicle baseline in DRG neurons in cryosections from two technical replicates for three biological replicates.N = 3 organ donor comparison of mean intensity difference after IL-6-sIL-6R-treatment.Scatter plot, paired t-test, two-tailed, p = 0.0249.(D) Spread of fold IF intensity difference from vehicle baseline per DRG neuron.N = DRG neuronal somata from vehicle, N = 677; IL-6-sIL-6R, N = 921.Violin plot.Kruskal-Wallis test with post hoc Dunn's test, p < 0.0001.(For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Table 1
FUNCAT: Technical Replicates for Approximated Explant Means per Organ Donor.
Number of TRPV1 + Nociceptors analyzed with average somatic FUNCAT mean grey and percent difference from vehicle baseline per treatment condition for each of four Organ Donor Replicates in Fig.2.Imaging parameters and histological methods are exactly as described in Table4for 10x micrographs.Abbreviations: Number (No.), Fluorescent Non-Canonical Amino Acid Tagging (FUNCAT), Fractional difference from vehicle baseline (% Difference), Average Mean Grey of all nociceptor somata from vehicle-treated explants analyzed (Vehicle Mean).

Table 3 P
-eif4e if and funcat: drg neuron totals and somata means per donor replicate.

Table 5
Artificial cerebrospinal fluid formulas used in this study.

Table 6
Reagents and resources used in this study including their source and identifier.