Journal Logo

Research Paper

Protective role of natural killer cells in neuropathic pain conditions

Lassen, Josephinea,*; Stürner, Klarissa Hanjab; Gierthmühlen, Jannea; Dargvainiene, Justinac; Kixmüller, Dorthec; Leypoldt, Frankb,c; Baron, Ralfa; Hüllemann, Philippa

Author Information
doi: 10.1097/j.pain.0000000000002274

1. Introduction

Chronic pain is a widespread phenomenon in modern Western society—and a serious problem for healthcare systems. About one-fifth of the European population is affected by chronic pain and is impaired at home or at work.6 However, only 2% of the patients who require treatment by specialized pain therapists are treated adequately.6,40 Results of a representative sample in Germany show that 31.9% of patients with impairing pain were treated by a pain specialist underlining the current lack of therapists.21

Seven to 18% of the population5 remain in pain after the initial disease has healed and reach a stage of chronic neuropathic pain.1 Apart from certain risk factors for chronic pain (such as lower socioeconomic status,39 depression,31,48 and history of abuse or job dissatisfaction37,39), it is well known that the immune system has a major impact on the development of pain after nerve injury. The most important mechanisms of cellular immune response are inflammation and cytotoxicity.13,32,33,49

Nerve injuries often lead to chronic neuropathic pain associated with peripheral and central neuroimmunological activation and inflammatory responses of nerve tissue.1 Inflammatory mediators promote the activation of immunocompetent nerve cells (eg, B and T cells), sensitize afferent neurons, and lead to hyperalgesia,29 which is symptomatic for neuropathic pain. In this process, a maladaptive immune response may promote permanent pain.7 Furthermore, a suppression of the immune response within the context of nerve injury can prevent the development of hyperalgesia.10

Austin and Moalem-Taylor demonstrated that after nerve lesions, both the innate and the adaptive immune systems are decisively involved: Peripheral nerve injuries provoke reactions in the immune system, such as the infiltration of inflammatory cells (eg, T cells) or the activation of resident immune cells (eg, mast cells and microglia).1

Previous animal studies demonstrated an association between specific immune cell patterns within the cerebrospinal fluid (CSF) and (persistent) pain conditions, finding leukocyte trafficking into the spinal cord after peripheral L5 nerve transsection, which correlated with mechanical allodynia.28,35,45

Immunological markers of pain chronicity are of high value to identify patients with a higher risk for pain chronicity at an earlier stage and allow modifying treatment strategies. In humans, the immune cell pattern within the CSF has not been investigated with respect to chronic neuropathic pain conditions so far.

Two of the pain conditions predisposing the development of chronic pain are herpes zoster neuralgia and polyneuropathy (PNP). Specific treatment remains challenging, and the intake of pain medication is often required for years.

As part of the ERA-Net Neuron consortium, we analyzed the immune cell profile within the CSF in correlation with the somatosensory phenotype in these specific neuropathic pain disorders. Because the CSF mainly contains T, B, and natural killer (NK) cells,43,44 we concentrated on these cells in the flow cytometry (FACS) analysis. The aim of the project was to identify immune cell markers for chronic pain.

2. Methods

2.1. Study design

Ten patients suffering from herpes zoster neuralgia and 31 patients with PNP were examined from June 2016 up to January 2019.

During baseline examination, patients received the painDETECT questionnaire (PDQ) as well as quantitative sensory testing (QST) in the most affected area and the corresponding contralateral side (patients with zoster) or on both sides of the most affected area (patients with PNP) to assess the somatosensory profile.

Cerebrospinal fluid was analyzed according to clinical routine diagnostics including cell counts, protein, lactate, and glucose. In addition, a FACS analysis of the CSF was performed. Lymphocyte subsets were analyzed using the following antibodies (with fluorochromes) from BD Biosciences: anti-CD45 (PerCP-Cy 5.5), anti-CD56 (PE), anti-CD3 (FITC), anti-CD19 (APC), anti-CD4 (PE-Cy7), and anti-CD8 (APC-Cy7). All analyses were performed using a BD FACSCanto analyzer (BD Biosciences), and data were analyzed using FlowJo (version 10.6.2). Only samples with at least 500 lymphocytes (defined by CD45+ expression) were included into the analysis. Quality control for doublet exclusion and live/dead cell staining was performed exemplary, which showed neither doublets nor a relevant amount of apoptotic cells (no CSF sample exceeded 100 cells/µL). Most importantly, CSF samples were processed within 30 minutes after a lumbar puncture. For these samples, the frequencies of CD3+CD4+ cells (T-helper cells), CD3+CD8+ cells (cytotoxic T cells), CD19+ cells (B cells), and CD56+ cells (NK cells) were assessed (Fig. 1) and further correlated with somatosensory profiles. Three months later, QST and PDQ were repeated to identify pain chronicity in patients with zoster and patients with PNP. All patients gave their written informed consent to participate in the study. The study is registered at German Clinical Trials Register (Registration trial: DRKS00023537) and was performed in accordance with the Declaration of Helsinki and approved by the local ethics committee (ID: D552/15).

Figure 1.
Figure 1.:
Gating strategy for flow cytometric analysis in the cerebrospinal fluid. Leukocytes were identified by CD45 expression (A) and next, the NK-cell and T-cell population was identified (B). T cells were further subdivided into CD4+ T helper and CD8+ cytotoxic T cells (D), whereas B cells were identified according to CD19 expression from CD3-negative and CD56-negative lymphocytes. NK, natural killer.

2.2. Interview and questionnaire

Demographic data such as age, sex, duration of disease, and affected area were assessed.

2.3. Pain intensity and characteristics

Presence of pain, pain location, and severity (numeric rating scale ranging from 0 [no pain] to 10 [maximum intensity]) were assessed by means of an interview.

2.4. Pain characteristics

PainDETECT15 is a screening tool to identify a neuropathic pain component in patients with chronic pain. It includes questions about general pain intensity, pain development, and possible radiating pain. It also asks for the presence and intensity of typical neuropathic symptoms. Finally, an end score is calculated to quantify whether a neuropathic pain component is unlikely, uncertain, or likely. The PDQ score is already used as a progression parameter and has proven to be valid and reliable.

2.5. Quantitativ sensory testing

Quantitative sensory testing was used to assess the patient's somatosensory profile. It is a standardized measurement that tests the somatosensory function of primary afferent nerve fibers (Aβ, Aδ, and C fibers) and their central pathways.

Quantitative sensory testing was performed in accordance to the protocol of the German Research Network on Neuropathic Pain (Deutscher Forschungsverbund Neuropathischer Schmerz, DFNS).17,34,52 Testing was performed in the affected dermatome and the corresponding contralateral side in patients with zoster and on both sides in the most affected area in patients with PNP (ie, dorsum of the feet [n = 29] and dorsum of the hand [n = 2]).

The following 13 parameters were recorded: cold detection threshold (CDT), warm detection threshold (WDT), thermal sensory limen (TSL), cold pain threshold (CPT), heat pain threshold (HPT), pressure pain threshold (PPT), mechanical pain threshold (MPT), mechanical pain sensitivity (MPS), wind-up ratio (WUR), mechanical detection threshold (MDT), vibration detection threshold (VDT), dynamic mechanical allodynia (DMA), and paradoxical heat sensation (PHS).

2.6. Statistical analysis

The analysis of the collected data was performed using IBM SPSS statistics for Mac (version 26.0).

The QST results were analyzed according to the current guidelines and compared with a reference database of healthy controls.27,51

Individual QST parameters of the 2 groups were compared using the Mann–Whitney U test. The intraindividual comparisons from the first and second examinations were analyzed using the Wilcoxon test.

Aiming to find associations between immune cell distribution within the CSF and the sensory phenotype, we performed the Spearman correlation analysis. In a first step, we used an exploratory approach without correction for multiple testing. In a second step, we identified robust results using Bonferroni correction, including FACS data, PDQ scores, and QST markers for pain chronicity (ie, HPT, as marker for peripheral sensitization,25 and MPS, MPT, and WUR, as markers for central sensitization3); P values <0.05 were considered as statistically significant.

3. Results

The epidemiological data of patients with herpes zoster and patients with PNP are shown in Table 1.

Table 1 - Epidemiological data of patients with herpes zoster and patients with polyneuropathy.
Patients with herpes zoster Patients with PNP
Number [n](%) 10 (24) 31 (76)
Age 63.2 y (±14.34) 65.19 y (±14.5)
Gender
 Female [n](%) 4 (40) 12 (39)
 Male [n](%) 6 (60) 19 (61)
Duration of disease [± SD] 1.7 mo (±1.14) 48.98 mo (±79.62)
Affected area
 Face [n](%) 2 (20) 0 (0)
 Arm [n](%) 2 (20) 0 (0)
 Hand [n](%) 1 (10) 2* (6)
 Trunk [n](%) 4 (40) 0 (0)
 Foot [n](%) 0 (0) 29 (94)
 No affected area [n](%) 1 (10) 0 (0)
*Both patients suffer from a chronic inflammatory demyelinating polyneuropathy (CIDP), in both cases the hands are much more affected than the feet.
PNP, polyneuropathy.

3.1. Comparison of pain intensity and painDETECT questionnaire scores of the 2 patient groups at baseline and follow-up

The pain ratings and PDQ scores are shown in Table 2. The 2 patient groups did not differ significantly in pain parameters (minimum and maximum pain; pain intensity within the previous 72 hours) or PDQ scores (P > 0.1 for all comparisons). There were no differences between baseline and follow-up assessment of pain and PDQ scores (P > 0.5 for all comparisons).

Table 2 - Pain ratings and PDQ scores of patients with herpes zoster and patients with polyneuropathy.
Patients with herpes zoster Patients with PNP
Baseline Follow-up Baseline Follow-up
Pain intensity within the previous 72 h [mean ± SD] (range) 4.44 ± 3.36 [0-9], N 9 2.33 ± 2.07 [0-5], N 6 3.16 ±3.53 [0-10], N 31 2.9 ± 2.7 [0-8], N 30
Minimal pain
 [Mean ± SD] (range) 3.8 ± 3.9 [0-9], N 5 2.0 ± 2.45 [0-5], N 6 1.0 ±1.79 [0-7], N 28 1.31 ± 2.04 [0-7], N 26
Maximum pain
 [Mean ± SD] (range) 6.4 ± 4.34 [0-10], N 5 3.67 ± 3.39 [0-10], N 6 4.21 ±3.99 [0-10], N 28 4.67 ± 3.79 [0-10], N 30
PDQ total score
 [Mean ± SD] (range) 14.89 ± 6.75 [4-22], N 9 9.2 ± 4.09 [2-12], N 5 10.52 ± 8.03 [0-26], N 31 10.57 ± 7.5 [0-25], N28
PDQ evaluation: neuropathic pain component unlikely/uncertain/likely [n](%) 3 (33)/2 (22)/4 (44) 5 (100)/0 (0)/0 (0) 19 (61)/6 (19.5)/6 (19.5) 17 (61)/6 (21)/5 (18)
All values are depicted as mean ± SD [minimum–maximum], number of patients.
PNP, polyneuropathy; PDQ, painDETECT questionnaire.

3.2. Comparison of somatosensory profiles of the 2 patient groups at baseline and follow-up

Regarding the QST parameters, differences were found between the patients with zoster and the patients with PNP regarding WDT (WDT Z-values zoster: −0.23 ±1.72 vs WDT Z-values PNP: −1.63 ±1.34, P = 0.03), TSL (TSL Z-values zoster: −0.7 ±1.51 vs TSL Z-values PNP: −1.75 ±1.15, P = 0.025), CPT (CPT Z-values zoster: 0.6 ±1.04 vs CPT Z -values PNP: −0.36 ± 1.02, P = 0.012), HPT (HPT Z-values zoster: 0.83 ±1.98 vs HPT Z-values PNP: −0.89 ±1.17, P = 0.004), and VDT (VDT Z-values zoster: −1.07 ±2.33 vs VDT Z-values PNP: −3.43 ±2.54, P = 0.021) (Fig. 2), indicating a pronounced function loss of fibers mediating temperature, touch, and vibration in patients with PNP.

Figure 2.
Figure 2.:
Comparison of baseline QST findings of patients with herpes zoster and patients with PNP. Z -values of the 13 QST parameters are given here. The gray area indicates the range of normative values according to the database of the German Research Network on Neuropathic Pain (DFNS). Error bars indicate the standard error of the mean. Z-value = each individual parameter is related to its region-specific, age-specific, and sex-specific reference range and is displayed as the number of SDs above or below the normal mean value. CDT, cold detection threshold; CPT, cold pain threshold; DMA, dynamic mechanical allodynia; HPT, heat pain threshold; MDT, mechanical detection threshold; MPS, mechanical pain sensitivity; MPT, mechanical pain threshold; PHS, paradoxical heat sensation; PNP, polyneuropathy; PPT, pressure pain threshold; QST, quantitative sensory testing; TSL, temperature sensory limen; VDT, vibration detection threshold; WDT, warm detection threshold; WUR, wind-up ratio.

Comparing follow-up data with baseline, no significant changes in QST findings were found, neither in patients with zoster (P > 0.05 for all comparisons) nor in patients with PNP (P > 0.1 for all comparisons) (Fig. 3).

Figure 3.
Figure 3.:
Comparison of baseline and follow-up QST findings of (A) patients with PNP and (B) patients with herpes zoster. Z-values of the 13 QST parameters are given here. The gray area indicates the range of normative values according to the database of the German Research Network on Neuropathic Pain (DFNS). Error bars indicate the standard error of the mean. Z-value = each individual parameter is related to its region-specific, age-specific, and sex-specific reference range and is displayed as the number of SDs above or below the normal mean value. CDT, cold detection threshold; CPT, cold pain threshold; DMA, dynamic-mechanical allodynia; HPT, heat pain threshold; MDT, mechanical detection threshold; MPS, mechanical pain sensitivity; MPT, mechanical pain threshold; PHS, paradoxical heat sensation; PNP, polyneuropathy; PPT, pressure pain threshold; TSL, temperature sensory limen; VDT, vibration detection threshold; WDT, warm detection threshold; WUR, wind-up ratio.

3.3. Clinical routine cerebrospinal fluid analysis of patients with herpes zoster and patients with polyneuropathy

The CSF findings of all patients are shown in Table 3. Apart from the leukocyte count (entire cohort: 18.11/μL ± 26.6/μL at zoster vs 1.9/μL ± 2.4/μL at PNP; P = 0.000436; FACS cohort: 23.33/μL ± 32.15/μL at zoster vs 2.5/μL ± 3.2/μL at PNP; P = 0.04; Fig. 4), the CSF findings did not differ significantly between the 2 groups of patients with zoster and patients with PNP (P > 0.05 for all comparisons).

Table 3 - Cerebrospinal fluid findings of all patients.
Unit Reference range Herpes Zoster Polyneuropathy
Leukocyte count /μL <5 5.46 ± 13.69 [1-76], N 10 3.61 ± 8.19 [0-51], N 31
Glucose mmol/L 2.8-4.4 3.47 ±0.32 [3.04-3.89], N 10 3.79 ±0.56 [2.96-5.72], N 31
Glucose quote L/S >0.5 0.66 ± 0.06 [0.52-0.76], N 9 3.85 ± 17.84 [0.4-100], N 31
Protein (total) mg/L 150-450 485 ± 237.98 [255-963], N 10 500.87 ± 182.74 [229-936], N 31
Albumin mg/L 35-53 297.1 ± 207.76 [39.1-679], N 10 308 ± 118.19 [105-628], N 31
Lactate mmol/L 1.3-2.4 1.8 ±0.26 [1.46-2.25], N 10 1.78 ± 0.3 [1.34-2.71], N 31
IgA liquor mg/L 0.7- 4.0 4.6 ±2.68 [0.84-9.82], N 10 5.32 ±7.14 [0.84-39.7], N 31
IgG liquor mg/L 7-16 31.14 ± 14.82 [16.7-57.9], N 10 49.74 ± 63.17 [10.1-294], N 31
IgM liquor mg/L 0.4 – 2.3 0.87 ± 1.73 [0-5.73], N 10 0.94 ± 0.45 [0.25-2.25], N 31
Q albumin *10−3 7.68 ± 4.28 [3.22-15.9], N 10 7.76 ± 3.37 [2.95-16.7], N 31
Q IgG *10−3 3.8 ± 1.81 [1.76-7.2], N 10 4.74 ± 5 [1.26-23], N 31
Q IgA *10−3 2.27 ± 1.27 [0.9-4.36], N 10 2.04 ± 1.2 [0.7-6.29], N 31
Q IgM *10−3 0.68 ± 0.5 [0.16-1.41], N 10 0.55 ± 0.62 [0.09-2.81], N 30
All values are depicted as mean ± SD [minimum–maximum], number of patients.
CSF, cerebrospinal fluid.

Figure 4.
Figure 4.:
Comparison of the leukocyte counts of patients with zoster and patients with PNP. The CSF results of the entire cohort as well as the FACS cohort are shown. (A) Patients with herpes zoster (B) patients with PNP. Entire cohort: P = 0.000436; FACS cohort: P = 0.04; * P < 0.05; ** P < 0.01; *** P < 0.001. CSF, cerebrospinal fluid. PNP, polyneuropathy; QST, quantitative sensory testing.

3.4. Flow cytometry analysis

Following strict inclusion criteria, each CSF sample had to contain at least 500 lymphocytes to undergo further analysis, resulted in the inclusion of 22 eligible FACS data.

Epidemiological parameters as well as pain intensity within the previous 72 hours, minimal pain and maximum pain, of patients with eligible FACS data were comparable with the entire study cohort and are shown in Tables 4 and 5.

Table 4 - Epidemiological data of patients with eligible FACS data.
Patients with herpes zoster Patients with PNP
Number [n](%) 6 (27) 16 (73)
Age 58.5 y (±12.74) 58.63 y (±14.46)
Sex
 Female [n](%) 2 (33) 6 (38)
 Male [n](%) 4 (67) 10 (62)
Duration of disease [± SD] 1.63 mo (±1.28) 45.02 mo (±70.06)
Affected area
 Arm and shoulder [n](%) 2 (33) 0 (0)
 Trunk [n](%) 4 (67) 0 (0)
 Foot [n](%) 0 (0) 16 (100)
PNP, polyneuropathy.

Table 5 - Questionnaire results (baseline and follow-up).
Patients with herpes zoster Patients with PNP
Baseline Follow-up Baseline Follow-up
Pain intensity in the previous 72 hours 4.83 ± 3.97 [0-9], N 6 3 ± 2.65 [0-5], N 3 2.56 ± 3.41 [0-10], N 16 2.38 ± 2.55 [0-8], N 16
Minimal pain 3.80 ± 3.90 [0-9], N 5 2.33 ± 2.52 [0-5], N 3 0.86 ± 1.56 [0-5], N 14 0.94 ± 1.69 [0-4], N 16
Maximum pain 6.40 ± 4.34 [0-10], N 5 4.67 ± 5.03 [0-10], N 3 4.2 ± 2.09 [0-10], N 15 4.25 ± 3.82 [0-10], N 16
PDQ total score 17.2 ± 6.02 [7-22] N 9 6.5 ± 6.36 [2-11] N 5 8.63 ± 7.75 [0-24], N 16 9.79 ± 8.66 [0-25], N 14
PDQ evaluation: neuropathic pain component unlikely/uncertain/likely [n](%) 3 (33)/2 (22)/4 (44) 5 (100)/0 (0)/0 (0) 19 (62)/6 (19)/6 (19) 17 (61)/6 (21)/5 (18)
All values are depicted as mean ± SD [minimum–maximum], number of patients.
PNP, polyneuropathy; PDQ, painDETECT questionnaire.

The exact distribution of the different cells in the CSF is shown in Table 6. There were no significant differences between patients with zoster and patients with PNP regarding cell distribution within the CSF (P > 0.5 for all comparisons).

Table 6 - Distribution of the different cells in the CSF
Patients with Zoster at baseline Patients with PNP at baseline
CD4 in % 58.56 ± 12.74 [36.72-71.82], N 6 51.65 ± 18.92 [15.77-77.79], N 16
CD8 in % 14.4 ± 11.06 [1.07-31.16], N 6 13.3 ± 9.05 [1.64-32.23], N 16
CD19 in % 3.54 ± 5.72 [0.32-15.14], N 6 4.01 ± 7.2 [0.43-30.06], N 16
CD56 in % 7.11 ± 6.48 [2.2-19.4], N 6 6.06 ± 4.56 [0.16-17.93], N 16
All values are depicted as mean ± SD [minimum–maximum], number of patients.
CSF, cerebrospinal fluid; PNP, polyneuropathy.

The PDQ score indicated a significant difference between patients with zoster and patients with PNP (PDQ score zoster: 17.2 ± 6.02 vs PDQ score PNP: 8.63 ± 7.75; P = 0.04); ie, patients with zoster reported symptoms that made a neuropathic pain component more likely more often than patients with PNP.

Apart from that, the 2 groups did not differ in pain characteristics (P > 0.1 for all comparisons). In addition, there were no significant differences between baseline and follow-up assessment of pain and PDQ scores (P > 0.5 for all comparisons).

3.4.1. Correlation analysis of immune cell frequencies and painDETECT questionnaire scores

There were no significant correlations between the FACS data and PDQ scores for all patients or in any of the subgroups (P > 0.05 for all comparisons).

3.4.2. Correlation analysis of immune cell frequencies and somatosensory parameters

3.4.2.1. All patients

Cytotoxic T cells (CD8+) correlated positively (r = 0.482, P = 0.023, n = 22) with HPT, ie, increased CD8+-cell frequency was associated with heat hyperalgesia.

There was a negative correlation (r = −0.596, P = 0.004, n = 21) between MPS and the frequency of NK cells: Increased NK-cell frequency correlated with a reduced MPS. This finding remained robust after Bonferroni correction (adjusted P = 0.036) (Fig. 5).

Figure 5.
Figure 5.:
Significant correlations of QST parameters with immune cells. (A) All patients, (B) patients with PNP, and (C) patients with herpes zoster. The figures with black outline remained significant after Bonferroni correction. *One patient with zoster wished to terminate the examination prematurely. HPT, heat pain threshold; MPS, mechanical pain sensitivity; n.s., not significant; PNP, polyneuropathy; QST, quantitative sensory testing.
3.4.2.2. Patients with herpes zoster

At baseline, a positive correlation was found between CD8+-cell frequencies and MPS (r = 0.990, P = 0.001, n = 5). This finding remained robust after Bonferroni correction (adjusted P = 0.009) (Fig. 5).

3.4.2.3. Patients with polyneuropathy

A high CD8+-cell frequency was associated with a reduced MPS (r= −0.516, P = 0.041, n = 16) as well as increased heat hyperalgesia (r = 0.551, P = 0.027, n = 16).

There was a significant correlation (r = −0.545, P = 0.029, n = 16) between MPS and the frequency of NK cells. An increased NK-cell frequency correlated with a reduced MPS.

3.5. Correlation analysis of immune cell frequencies and pain outcomes

We found no significant correlations between FACS data and PDQ scores in the follow-up (P > 0.05), and there was neither any significant change in PDQ scores between baseline and follow-up (P > 0.1).

In 2 patients, the PDQ revealed that a neuropathic pain component was uncertain at baseline but had become likely at the follow-up examination. Both patients showed low NK-cell frequencies compared with the mean of all eligible NK frequency data.

Two patients developed an abnormal HPT at follow-up assessment; no valid CSF results are available for either patient.

There were 4 patients whose initial normal MPS values became abnormal over time. Two patients had valid FACS findings and showed a reduced NK-cell frequency compared with the mean.

One patient developed an abnormal WUR over time and showed reduced NK-cell frequency compared with the mean.

4. Discussion

This study aimed to find specific immune cell patterns within the CSF, which can contribute to pain chronicity.

We found a significant association of NK-cell frequency and MPS, an important QST marker for central sensitization and therefore a potential marker for chronic pain. These results indicate a protective effect of NK cells regarding pain chronicity. We also found significant correlations regarding cytotoxic T cells, which at first glance seemed contradictory. In patients with zoster, CD8+-cell frequency correlated with MPS (high levels of CD8+ cells were associated with pronounced central sensitization). However, in patients with PNP, CD8+-cell frequency was associated with a reduced MPS, indicating less signs of central sensitization. These findings are discussed as follows:

4.1. Protective effect of natural killer cells on central sensitization

The most interesting result was a significant inverse correlation between MPS (as one of the most relevant markers for central sensitization in the QST22) and the frequency of NK cells. This indicates that NK cells might have a protective effect on pain sensitization.

Natural killer cells belong to the innate immune system and their primary function is destroying tumor-infected or virus-infected cells. However, it has become more and more clear that their function comprises much more. There are few studies that demonstrate their recruitment in the periphery after nerve lesions. Rats with injured nerves showed a significant upregulation of NK cells in these nerves compared with sham-operated controls.11 In another experiment in mice, it could be shown that the injury of peripheral nerves led to the recruitment of NK cells into these nerves. The function of NK cells correlated with a reduced incidence of hypersensitivity. The authors consider this selective NK-cell-mediated destruction of damaged axons as a supplement to Wallerian degeneration and point to the therapeutic potential of NK cells in painful neuropathy by clearing partially damaged nerves.12 Moreover, Gao et al. showed that NK cells were increased both in their activity and their quantity in the spleen and peripheral blood after electrical stimulation of ligatured sciatic nerves in rats. Because of repetitive electroacupuncture, IL-2 and β-EP, 2 efficacious activators of NK cells, were increased.16

The infiltration of immune cells into the central nervous system during peripheral neuroinflammation is discussed controversially.28 After peripheral nerve transection, increased spinal activity of CD4+ and MHC-II cells accompanied by persistent mechanical allodynia was observed,46 which points to a consecutive central neuroinflammatory response. Another study showed the trafficking of CD3+ T lymphocytes and MHC-II cells into the spinal cord correlating with mechanical allodynia.45 Rutkowski et al. demonstrated extravasation of microglia in the central nervous system of rats suffering from neuropathic pain due to L5 peripheral spinal nerve transection. The authors suggest a specific role of infiltrating cells with neuroprotective or antihyperalgesic effects.35 Our demonstrated reduction of MPS in combination with an increased NK-cell frequency suites these findings and provides the first data in humans. By contrast, recent studies could not find any evidence for an infiltration of peripheral immune cells into the spinal cord parenchyma after peripheral injury.19,47 These studies transected peripheral nerve tissue within an animal model, without finding an immune cell migration into the central nervous system. The latter studies investigated the spinal cord parenchyma, whereas in our study the CSF was analyzed. Notably, immune cell infiltration into the CSF does not necessarily equal immune cell infiltration into the spinal cord. The next scientific step should be the comparison of immune cell activity within the CSF and spinal cord parenchyma during neuroinflammation to enlighten this topic.

Known QST markers that indicate (peripheral or central) sensitization are HPT, MPS, WUR, and MPT. There were 4 patients who showed sensitization as their initial normal MPS values became abnormal over time. Two of them had valid FACS findings; both showed reduced NK-cell frequencies in the FACS analysis. In addition, there was one patient who developed an abnormal WUR. This patient also showed a reduced NK-cell frequency compared with the mean of NK-cell frequencies of the entire study cohort. These findings also support the assumption that NK cells could have a protective effect on pain sensitization. These observations should of course be reevaluated in a larger cohort of subjects.

There is ample evidence that NK cells are involved in the pathogenesis of herpes zoster.4,14,30,53 NK cells circulate in the blood and migrate into inflammated tissue as part of the innate immune response and thus react early to infection.50 NK cells are reported to contain herpes zoster infection. On the contrary, Campbell et al. showed that the varicella-zoster virus specifically penetrates healthy NK cells and thereby contributes to the spread of the infection.8

4.2. Cytotoxic T cells and sensitization

In patients with zoster, high CD8+-cell levels were associated with pronounced central sensitization (MPS), whereas in patients with PNP, CD8+-cell frequency was associated with less signs of central sensitization.

These findings suggest that in the acute stage of a neuropathic pain condition (within our study cohort herpes zoster neuralgia occurred within the first 3 months) cytotoxic T cells promote central sensitization, most probably as a relevant motor within the neuroinflammatory cascade.

Recent studies were able to identify CD8+ cells as one of the major cell groups infiltrating ganglia after or during zoster infection.18,41,42 Today, it is known that VZV antigens induce T-cell-mediated immune responses.24,36 In a murine spinal model, Cao et al. reported that T-lymphocyte-deficient mice presented a reduction in injury-induced hypersensitivity, suggesting T lymphocytes as an important factor for the maintenance of neuropathic pain.9

On the contrary, in a chronic neuropathic pain condition (our cohort of patients with chronic painful PNP exceeded pain onset by years), CD8+-cell levels were associated with reduced central sensitization. These findings suggest that CD8+ cells develop protective features when neuropathic pain becomes chronic. What underlines this assumption is the result of a murine study concerning chronic arthritis: Their results indicate that CD8+ cells have a protective and analgesic effect on inflammatory pain by the release of endogenous opioids in the chronic stage of disease.2 In addition, another murine experiment showed that CD8+ cells are necessary for the recovery of paclitaxel-induced or cisplatin-induced mechanical allodynia and they abate spontaneous pain and numbness.26 In chronic pain conditions such as complex regional pain syndrome and fibromyalgia, a reduction in circulating CD8+ T cells within the blood has been found.23

This switch from promoter to protector is known for some cytokines and immune cells: Within the immune response they can have both a promoting and a protective character. However, we do not know when and why this switch occurs and what causes it. Improved knowledge of this could be of therapeutic importance because it could give an indication of the progress of a disease and might help to adapt therapeutic strategies.

4.3. Leukocyte trafficking into the central nervous system

The leukocyte cell count of the patients with zoster was significantly higher than in patients with PNP. This result indicates possible leukocyte migration into the central nervous system during peripheral inflammation within the spinal nerve root. An increase of the leukocyte cell count within the spinal cord due to varicella-zoster infection has been previously reported.38 The authors explain these CSF changes by anatomical proximity of the affected ganglia to the central nervous system. Haanpää et al. also suggested direct spread of the varicella virus from the dorsal root ganglion into the central nervous system.20 Leukocyte trafficking into the central nervous system remains a controversial topic on which little is known so far.

5. Limitations

As a pilot study on human CSF-FACS-analysis in patients with zoster and patients with PNP, the study's main limitation is the relatively low number of patients. Because we had to define strict quality criteria for inclusion into the FACS analysis, the number of cases with valid FACS results was markedly smaller than the original group size. Furthermore, drawing CSF from patients is an invasive procedure that needs a clear neurological indication, which is not always the case in acute zoster or PNP. Thus, the extraction of CSF was limited because of ethical reasons and only possible when indicated.

6. Conclusion

Several studies in mice reported a connection between NK cells and pain sensitization. No evidence has yet been found that frequency and activity level of NK cells correlate with loss of sensitivity in human CSF. Our study was the first to show that a high NK-cell frequency in the human CSF is associated with reduced central sensitization (MPS) in neuropathic pain disorders. Thus, NK cells could be a marker for central pain sensitization and therefore a possible marker for chronic pain. Despite the fact that NK cells represent only a small fraction of immune cells, they seem to play a unique role regarding the immune response after nerve injury.

Although our observations were drawn from a limited number of patients, we found promising markers for pain chronicity within the CSF, which withstood correction for multiple testing and which are worthwhile to be considered in future CSF studies in a larger cohort (eg, multicenter trials).

Conflict of interest statement

J. Lassen received financial support from Pfizer OFG Germany GmbH, outside the submitted work. K.H. Stürner reports personal fees from Biogen GmbH, personal fees from Hoffmann La Roche AG, personal fees from Sanofi Genzyme, personal fees from Bayer, and personal fees from Merk KGaA, outside the submitted work. J. Gierthmühlen reports personal fees from TAD Pharma, Lilly, Novartis, and Grünenthal, outside the submitted work. F. Leypoldt reports grants from Germany Ministry of Research and Education, grants from ERA-NET through German Research Council, personal fees from Grifols, personal fees from Roche, personal fees from Biogen, personal fees from Alexion, and personal fees from Novartis, outside the submitted work. R. Baron reports grants from German Federal Ministry of Education and Research (BMBF): Member of the ERA-NET NEURON/IM-PAIN project (01EW1503), during the conduct of the study; grants from EU projects: “Europain” (115007), DOLORisk (633491), IMI-Paincare (777500), German Federal Ministry of Education and Research (BMBF): Verbundprojekt: Frühdetektion von Schmerzchronifizierung (NoChro) (13GW0338C), German Research Network on Neuropathic Pain (01EM0903), Pfizer Pharma GmbH, Genzyme GmbH, Grünenthal GmbH, Mundipharma Research GmbH und Co KG., Novartis Pharma GmbH, Alnylam Pharmaceuticals Inc, and Zambon GmbH. He reports personal fees from Pfizer Pharma GmbH, Genzyme GmbH, Grünenthal GmbH, Mundipharma, Sanofi Pasteur, Medtronic Inc. Neuromodulation, Eisai Co Ltd, Lilly GmbH, Boehringer Ingelheim Pharma GmbH & Co KG, Astellas Pharma GmbH, Desitin Arzneimittel GmbH, Teva GmbH, Bayer-Schering, MSD GmbH, Seqirus Australia Pty. Ltd, Novartis Pharma GmbH, TAD Pharma GmbH, Grünenthal SA Portugal, Sanofi-Aventis Deutschland GmbH, Agentur Brigitte Süss, Grünenthal Pharma AG Schweiz, Grünenthal B.V. Niederlande, personal fees from Pfizer Pharma GmbH, Genzyme GmbH, Grünenthal GmbH, Mundipharma Research GmbH und Co. KG, Allergan, Sanofi Pasteur, Medtronic, Eisai, Lilly GmbH, Boehringer Ingelheim Pharma GmbH&Co.KG, Astellas Pharma GmbH, Novartis Pharma GmbH, Bristol-Myers Squibb, Biogenidec, AstraZeneca GmbH, Merck, Abbvie, Daiichi Sankyo, Glenmark Pharmaceuticals S.A., Seqirus Australia Pty. Ltd, Teva Pharmaceuticals Europe Niederlande, Teva GmbH, Genentech, Mundipharma International Ltd. United Kingdom, Astellas Pharma Ltd. United Kingdom, Galapagos NV, Kyowa Kirin GmbH, Vertex Pharmaceuticals Inc., Biotest AG, Celgene GmbH, Desitin Arzneimittel GmbH, Regeneron Pharmaceuticals Inc. USA, Theranexus DSV CEA Frankreich, Abbott Products Operations AG Schweiz, Bayer AG, Grünenthal Pharma AG Schweiz, Mundipharma Research Ltd. United Kingdom, Akcea Therapeutics Germany GmbH, Asahi Kasei Pharma Corporation, AbbVie Deutschland GmbH & Co. KG, Air Liquide Sante International Frankreich, Alnylam Germany GmbH, Lateral Pharma Pty Ltd, Hexal AG, and Angelini, Janssen, outside the submitted work. P. Hüllemann reports grants from German Federal Ministry of Education and Research (BMBF) and grants from Zambon GmbH, outside the submitted work. The remaining authors have no conflicts of interest to declare.

Supplemental video content

A video abstract associated with this article can be found at http://links.lww.com/PAIN/B347.

Acknowledgments

This study was supported by the German Federal Ministry of Education and Research (BMBF): Member of the ERA_NET NEURON/IM-PAIN Project (grant number: 01EW1503). We thank Stephen McMahon and Franziska Denk for their valuable scientific input.

References

[1]. Austin PJ, Moalem-Taylor G. The neuro-immune balance in neuropathic pain: involvement of inflammatory immune cells, immune-like glial cells and cytokines. J Neuroimmunol 2010;229:26–50.
[2]. Baddack-Werncke U, Busch-Dienstfertig M, Gonzalez-Rodriguez S, Maddila SC, Grobe J, Lipp M, Stein C, Muller G. Cytotoxic T cells modulate inflammation and endogenous opioid analgesia in chronic arthritis. J Neuroinflammation 2017;14:30.
[3]. Baron R, Hans G, Dickenson AH. Peripheral input and its importance for central sensitization. Ann Neurol 2013;74:630–6.
[4]. Biron CA, Byron KS, Sullivan JL. Severe herpesvirus infections in an adolescent without natural killer cells. N Engl J Med 1989;320:1731–5.
[5]. Bouhassira D, Lanteri-Minet M, Attal N, Laurent B, Touboul C. Prevalence of chronic pain with neuropathic characteristics in the general population. PAIN 2008;136:380–7.
[6]. Breivik H, Collett B, Ventafridda V, Cohen R, Gallacher D. Survey of chronic pain in Europe: prevalence, impact on daily life, and treatment. Eur J Pain 2006;10:287–333.
[7]. Calvo M, Dawes JM, Bennett DL. The role of the immune system in the generation of neuropathic pain. Lancet Neurol 2012;11:629–42.
[8]. Campbell TM, McSharry BP, Steain M, Ashhurst TM, Slobedman B, Abendroth A. Varicella zoster virus productively infects human natural killer cells and manipulates phenotype. Plos Pathog 2018;14:e1006999.
[9]. Cao L, DeLeo JA. CNS-infiltrating CD4+ T lymphocytes contribute to murine spinal nerve transection-induced neuropathic pain. Eur J Immunol 2008;38:448–58.
[10]. Clatworthy AL, Illich PA, Castro GA, Walters ET. Role of peri-axonal inflammation in the development of thermal hyperalgesia and guarding behavior in a rat model of neuropathic pain. Neurosci Lett 1995;184:5–8.
[11]. Cui JG, Holmin S, Mathiesen T, Meyerson BA, Linderoth B. Possible role of inflammatory mediators in tactile hypersensitivity in rat models of mononeuropathy. PAIN 2000;88:239–48.
[12]. Davies AJ, Kim HW, Gonzalez-Cano R, Choi J, Back SK, Roh SE, Johnson E, Gabriac M, Kim MS, Lee J, Lee JE, Kim YS, Bae YC, Kim SJ, Lee KM, Na HS, Riva P, Latremoliere A, Rinaldi S, Ugolini S, Costigan M, Oh SB. Natural killer cells degenerate intact sensory afferents following nerve injury. Cell 2019;176:716–28 e718.
[13]. Davies AJ, Rinaldi S, Costigan M, Oh SB. Cytotoxic immunity in peripheral nerve injury and pain. Front Neurosci 2020;14:142.
[14]. Etzioni A, Eidenschenk C, Katz R, Beck R, Casanova JL, Pollack S. Fatal varicella associated with selective natural killer cell deficiency. J Pediatr 2005;146:423–5.
[15]. Freynhagen R, Baron R, Gockel U, Tolle TR. painDETECT: a new screening questionnaire to identify neuropathic components in patients with back pain. Curr Med Res Opin 2006;22:1911–20.
[16]. Gao YH, Wang JY, Qiao LN, Chen SP, Tan LH, Xu QL, Liu JL. NK cells mediate the cumulative analgesic effect of electroacupuncture in a rat model of neuropathic pain. BMC Complement Altern Med 2014;14:316.
[17]. Geber C, Scherens A, Pfau D, Nestler N, Zenz M, Tolle T, Baron R, Treede RD, Maier C. [Procedure for certification of QST laboratories]. Schmerz 2009;23:65–9.
[18]. Gowrishankar K, Steain M, Cunningham AL, Rodriguez M, Blumbergs P, Slobedman B, Abendroth A. Characterization of the host immune response in human Ganglia after herpes zoster. J Virol 2010;84:8861–70.
[19]. Gu N, Peng J, Murugan M, Wang X, Eyo UB, Sun D, Ren Y, DiCicco-Bloom E, Young W, Dong H, Wu LJ. Spinal microgliosis due to resident microglial proliferation is required for pain hypersensitivity after peripheral nerve injury. Cell Rep 2016;16:605–14.
[20]. Haanpaa M, Dastidar P, Weinberg A, Levin M, Miettinen A, Lapinlampi A, Laippala P, Nurmikko T. CSF and MRI findings in patients with acute herpes zoster. Neurology 1998;51:1405–11.
[21]. Hauser W, Schmutzer G, Henningsen P, Brahler E. [Chronic pain, pain disease, and satisfaction of patients with pain treatment in Germany. Results of a representative population survey]. Schmerz 2014;28:483–92.
[22]. Hullemann P, Watfeh R, Shao YQ, Nerdal A, Binder A, Baron R. Peripheral sensitization reduces laser-evoked potential habituation. Neurophysiol Clin 2015;45:457–67.
[23]. Kaufmann I, Eisner C, Richter P, Huge V, Beyer A, Chouker A, Schelling G, Thiel M. Lymphocyte subsets and the role of TH1/TH2 balance in stressed chronic pain patients. Neuroimmunomodulation 2007;14:272–80.
[24]. Kleemann P, Distler E, Wagner EM, Thomas S, Klobuch S, Aue S, Schnurer E, Schild H, Theobald M, Plachter B, Tenzer S, Meyer RG, Herr W. Varicella-zoster virus glycoproteins B and E are major targets of CD4+ and CD8+ T cells reconstituting during zoster after allogeneic transplantation. Haematologica 2012;97:874–82.
[25]. LaMotte RH, Thalhammer JG, Torebjork HE, Robinson CJ. Peripheral neural mechanisms of cutaneous hyperalgesia following mild injury by heat. J Neurosci 1982;2:765–81.
[26]. Laumet G, Edralin JD, Dantzer R, Heijnen CJ, Kavelaars A. Cisplatin educates CD8+ T cells to prevent and resolve chemotherapy-induced peripheral neuropathy in mice. PAIN 2019;160:1459–68.
[27]. Maier C, Baron R, Tolle TR, Binder A, Birbaumer N, Birklein F, Gierthmuhlen J, Flor H, Geber C, Huge V, Krumova EK, Landwehrmeyer GB, Magerl W, Maihofner C, Richter H, Rolke R, Scherens A, Schwarz A, Sommer C, Tronnier V, Uceyler N, Valet M, Wasner G, Treede RD. Quantitative sensory testing in the German Research Network on Neuropathic Pain (DFNS): somatosensory abnormalities in 1236 patients with different neuropathic pain syndromes. PAIN 2010;150:439–50.
[28]. Marchand F, Perretti M, McMahon SB. Role of the immune system in chronic pain. Nat Rev Neurosci 2005;6:521–32.
[29]. Moalem G, Tracey DJ. Immune and inflammatory mechanisms in neuropathic pain. Brain Res Rev 2006;51:240–64.
[30]. Notarangelo LD, Mazzolari E. Natural killer cell deficiencies and severe varicella infection. J Pediatr 2006;148:563–4.
[31]. Ramond A, Bouton C, Richard I, Roquelaure Y, Baufreton C, Legrand E, Huez JF. Psychosocial risk factors for chronic low back pain in primary care--a systematic review. Fam Pract 2011;28:12–21.
[32]. Raoof R, Willemen H, Eijkelkamp N. Divergent roles of immune cells and their mediators in pain. Rheumatology (Oxford) 2018;57:429–40.
[33]. Ren K, Dubner R. Interactions between the immune and nervous systems in pain. Nat Med 2010;16:1267–76.
[34]. Rolke R, Baron R, Maier C, Tolle TR, Treede RD, Beyer A, Binder A, Birbaumer N, Birklein F, Botefur IC, Braune S, Flor H, Huge V, Klug R, Landwehrmeyer GB, Magerl W, Maihofner C, Rolko C, Schaub C, Scherens A, Sprenger T, Valet M, Wasserka B. Quantitative sensory testing in the German Research Network on Neuropathic Pain (DFNS): standardized protocol and reference values. PAIN 2006;123:231–43.
[35]. Rutkowski MD, Lambert F, Raghavendra V, DeLeo JA. Presence of spinal B7.2 (CD86) but not B7.1 (CD80) co-stimulatory molecules following peripheral nerve injury: role of nondestructive immunity in neuropathic pain. J Neuroimmunol 2004;146:94–8.
[36]. Sei JJ, Cox KS, Dubey SA, Antonello JM, Krah DL, Casimiro DR, Vora KA. Effector and central memory poly-functional CD4(+) and CD8(+) T cells are boosted upon ZOSTAVAX((R)) vaccination. Front Immunol 2015;6:553.
[37]. Shaw WS, van der Windt DA, Main CJ, Loisel P, Linton SJ. Decade of the Flags Working G. Early patient screening and intervention to address individual-level occupational factors (“blue flags”) in back disability. J Occup Rehabil 2009;19:64–80.
[38]. Skripuletz T, Pars K, Schulte A, Schwenkenbecher P, Yildiz O, Ganzenmueller T, Kuhn M, Spreer A, Wurster U, Pul R, Stangel M, Suhs KW, Trebst C. Varicella zoster virus infections in neurological patients: a clinical study. BMC Infect Dis 2018;18:238.
[39]. Smith BH, Macfarlane GJ, Torrance N. Epidemiology of chronic pain, from the laboratory to the bus stop: time to add understanding of biological mechanisms to the study of risk factors in population-based research? PAIN 2007;127:5–10.
[40]. Smith BH, Torrance N. Management of chronic pain in primary care. Curr Opin Support Palliat Care 2011;5:137–42.
[41]. Steain M, Sutherland JP, Rodriguez M, Cunningham AL, Slobedman B, Abendroth A. Analysis of T cell responses during active varicella-zoster virus reactivation in human ganglia. J Virol 2014;88:2704–16.
[42]. Sutherland JP, Steain M, Buckland ME, Rodriguez M, Cunningham AL, Slobedman B, Abendroth A. Persistence of a T cell infiltrate in human ganglia years after herpes zoster and during post-herpetic neuralgia. Front Microbiol 2019;10:2117.
[43]. Svenningsson A, Andersen O, Edsbagge M, Stemme S. Lymphocyte phenotype and subset distribution in normal cerebrospinal fluid. J Neuroimmunol 1995;63:39–46.
[44]. Svenningsson A, Hansson GK, Andersen O, Andersson R, Patarroyo M, Stemme S. Adhesion molecule expression on cerebrospinal fluid T lymphocytes: evidence for common recruitment mechanisms in multiple sclerosis, aseptic meningitis, and normal controls. Ann Neurol 1993;34:155–61.
[45]. Sweitzer SM, Hickey WF, Rutkowski MD, Pahl JL, DeLeo JA. Focal peripheral nerve injury induces leukocyte trafficking into the central nervous system: potential relationship to neuropathic pain. PAIN 2002;100:163–70.
[46]. Sweitzer SM, White KA, Dutta C, DeLeo JA. The differential role of spinal MHC class II and cellular adhesion molecules in peripheral inflammatory versus neuropathic pain in rodents. J Neuroimmunol 2002;125:82–93.
[47]. Tashima R, Mikuriya S, Tomiyama D, Shiratori-Hayashi M, Yamashita T, Kohro Y, Tozaki-Saitoh H, Inoue K, Tsuda M. Bone marrow-derived cells in the population of spinal microglia after peripheral nerve injury. Sci Rep 2016;6:23701.
[48]. Taylor JB, Goode AP, George SZ, Cook CE. Incidence and risk factors for first-time incident low back pain: a systematic review and meta-analysis. Spine J 2014;14:2299–319.
[49]. Totsch SK, Sorge RE. Immune system involvement in specific pain conditions. Mol Pain 2017;13:1744806917724559.
[50]. Vivier E, Tomasello E, Baratin M, Walzer T, Ugolini S. Functions of natural killer cells. Nat Immunol 2008;9:503–10.
[51]. Vollert J, Attal N, Baron R, Freynhagen R, Haanpaa M, Hansson P, Jensen TS, Rice AS, Segerdahl M, Serra J, Sindrup SH, Tolle TR, Treede RD, Maier C. Quantitative sensory testing using DFNS protocol in Europe: an evaluation of heterogeneity across multiple centers in patients with peripheral neuropathic pain and healthy subjects. PAIN 2016;157:750–8.
[52]. Vollert J, Maier C, Attal N, Bennett DLH, Bouhassira D, Enax-Krumova EK, Finnerup NB, Freynhagen R, Gierthmuhlen J, Haanpaa M, Hansson P, Hullemann P, Jensen TS, Magerl W, Ramirez JD, Rice ASC, Schuh-Hofer S, Segerdahl M, Serra J, Shillo PR, Sindrup S, Tesfaye S, Themistocleous AC, Tolle TR, Treede RD, Baron R. Stratifying patients with peripheral neuropathic pain based on sensory profiles: algorithm and sample size recommendations. PAIN 2017;158:1446–55.
[53]. Wendland T, Herren S, Yawalkar N, Cerny A, Pichler WJ. Strong alpha beta and gamma delta TCR response in a patient with disseminated Mycobacterium avium infection and lack of NK cells and monocytopenia. Immunol Lett 2000;72:75–82.
Keywords:

Cerebrospinal fluid; Chronicity; Sensitization; Zoster; Polyneuropathy

Supplemental Digital Content

© 2021 International Association for the Study of Pain