Differential regulation of TRP channel gene and protein expression by intervertebral disc degeneration and back pain

Intervertebral disc (IVD) degeneration and consequent low back pain (LBP) are common and costly pathological processes that require improved treatment strategies. Transient Receptor Potential (TRP) channels constitute a family of multimodal ion channels that have recently emerged as contributors to disc pathologies and were thus proposed as potential therapeutic targets, although limited data on their presence and function in the IVD exist. The purpose of this study was to determine the mRNA and protein expression of TRP channels in non-degenerated and degenerated human IVD tissue (with different pain intensity and chronicity) using gene array, conventional qPCR and immunohistochemistry. We could demonstrate that 26 out of 28 currently known TRP channels are expressed in the IVD on the mRNA level, thereby revealing novel therapeutic candidates from the TRPC, TRPM and TRPML subfamilies. TRPC6, TRPM2 and TRPML1 displayed enhanced gene and protein expression in degenerated IVDs as compared to non-degenerated IVDs. Additionally, the gene expression of TRPC6 and TRPML1 was influenced by the IVD degeneration grade. Pain intensity and/or chronicity influenced the gene and/or protein expression of TRPC6, TRPM2 and TRML1. Interestingly, decreased gene expression of TRPM2 was observed in patients treated with steroids. This study supports the importance of TRP channels in IVD homeostasis and pathology and their possible application as pharmacological targets for the treatment of IVD degeneration and LBP. However, the exact function and activation of the highlighted TRP channels will have to be determined in future studies.

In a subgroup of those affected by IVD degeneration, inflammatory processes take place within the IVD tissue. Inflammation has been described as a major contributor to the development of painful disc degeneration (also known as degenerative disc disease (DDD)), hence presumably distinguishing symptomatic from asymptomatic IVD degeneration 4 . On the molecular level, DDD can be characterized by an up-regulation of pro-inflammatory cytokines such as interleukin (IL)-6, IL-1B and tumor necrosis factor (TNF)-α 3 . Moreover, mechanical loading can also induce inflammation, depending on its type, applied magnitude, duration and frequency 3,5,6 . Similarly, altered IVDs osmolarity can contribute to tissue inflammation by modulating pro-inflammatory mediators and pathways (e.g. mitogen-activated protein (MAP) kinases, T-cells 5/tonicity response element-binding protein (NFAT5/TonEBP)) 7, 8 and molecules (e.g. IL-6, IL-1B) 9 .
Although inflammation as well as mechanical and osmotic stress have been identified as important factors in the development of painful disc degeneration and hence back pain, the exact pathobiological mechanism remain to this day unknown. However, a superfamily of multimodal ion channels, the so-called transient receptor potential (TRP) channels, have recently emerged as potential contributors to disc pathologies 10 . TRP channels are of utmost interest in IVD research as they are regulated by a diverse range of stimuli, including mechanical and osmotic stress, and furthermore modulate inflammatory responses and mediate a variety of sensations, including pain. Stimulation of a TRP channel will cause its activation (i.e. opens channels pore), leading to ion movement and resulting in elevated cytosolic intracellular calcium. Applied stimuli (e.g. shear stress) can change a channel's molecular distribution, hence altering its membrane trafficking and spatial/temporal distribution, which in turn can influence its activity threshold levels 11,12 . Very recent studies highlighted TRPC6 and TRPV4 as possible contributors to the IVD's health and disease 9,[13][14][15] , but numerous other members of the various TRP families (ankyrin TRPA, canonical TRPC, vanilloid TRPV, melastatin TRPM, mucolipin TRPML and polycystic TRPP) have not yet been investigated.
To gain a better insight into the role of TRP channels in the IVD and low back pain, the purpose of this study was to identify the presence of all currently known TRP channels in non-degenerated and degenerated human IVDs with varying pain intensity and chronicity, and to highlight their possible involvement in IVD pathologies.
In the degenerated IVDs, the highest gene expression, detected in both NP and AF, was measured for TRPC1, TRPM7, PKD1, TRPP1 (PKD2) and TRPV4. TRPC7 was detected only in the AF (2 out of 2 degenerated AF samples) and TRPM1 only in the NP (2 out of 2 degenerated NP samples). TRPP3 (PKD2L2) could be detected only in one out of two AF sample. Figure 1 presents overall (combined NP and AF) gene expression as measured on the gene array normalized to YWHAZ (measured Ct values for all tested TRP targets in the gene array screening are presented in the Supplementary Material Table S1). Based on the gene   array results, 12 targets (TRPC1, TRPC3, TRPC4, TRPC6, TRPM2, TRPM7, TRPML1, TRPML2, PKD1, TRPP1  (PKD2), TRPV1, TRPV4) were chosen for further analysis on the additional 26 IVD samples (8 non-degenerated IVDs: 4x NP and 4x AF; 18 degenerated: 8x AF, 8x NP, 2x mix tissue) using conventional qPCR (see Table 3). When combining data from all 34 samples, TRPV4 was the most highly expressed target in both non-degenerated and degenerated IVDs, followed by TRPC1, TRPP2, PKD1, and TRPM7 (Fig. 2). None of the investigated targets differed between the NP and AF zones (p > 0.05, data not shown).

TRP channel expression (mRNA): Disease state and degeneration grade.
When comparing the expression of the TRP channels in the non-degenerated and degenerated IVDs, statistically significant differences were found in the mRNA expression levels of TRPC6 (Fig. 3a, p = 0.001), TRPM2 (Fig. 3b, p = 0.0002) and TRPML1 (Fig. 3c, p = 0.03). In all three targets, mRNA levels were higher in the degenerated IVD samples than in the non-degenerated samples.
Within the degenerated IVDs (n = 22), the expression of several TRP channels was significantly affected by the degeneration grade: TRPC4 ( TRP channel expression (mRNA): Modic changes. Modic changes represent a classification of pathological changes to the endplates, where 0 indicates no pathological changes, I is characterized by a vertebral bone edema, and II by fatty replacement of the bone marrow 18 . Modic changes may be associated with the pain development 19 . We hence tested whether the presence of Modic changes affects TRP channel expression and found a significant effect for TRPC4 (Fig. 5a, I > II, p = 0.01) and TRPC6 (Fig. 5b, 0 > I p = 0.02, 0 > II p = 0.007).
TRP channel expression (mRNA): Pain intensity, pain duration and steroid use. Degenerated IVD samples obtained from patients who assessed their pain as moderate were characterized by significantly higher TRPC6 expression levels compared to those obtained from patients with intense pain (p = 0.007, Fig. 6a) or disabling pain (p = 0.03, Fig. 6a). Interestingly, TRPC6 expression tended to increase with the duration of pain symptoms, but no statistical significance was found between the groups (p = 0.058, Fig. 6d). Expression of TRPM2 and TRPML1 was found to be the highest in the IVDs obtained from patients who defined their pain as intense and was found to be statistically different from the IVDs obtained from patients with disabling (TRPM2 p = 0.04 Fig. 6b, TRPML1 p = 0.04 Fig. 6c).
In addition, TRPM2 mRNA expression was significantly decreased in IVD samples obtained from patients who received steroid treatment (p = 0.04, Fig. 7). TRP channels protein expression: Immunohistochemistry. In the next step, TRPC6, TRPM2 and TRPML1 expression changes observed during degeneration progression on the mRNA level (Figs. 2, 3a-c, 4b,d, 5b, 6a-d and 7) were investigated on the protein level by immunohistochemistry. A total of seven surgical symptomatic degenerated IVD tissue samples, two non-degenerated young IVD samples and one aged asymptomatic (non-painful) IVD sample with mild multifocal degenerative changes were examined per target. Details regarding patient information and a complete evaluation, as well as the applied scoring system can be found in Tables 4-7, respectively.
TRPC6 protein expression levels were overall comparable to TRPM2, thus reflecting our gene expression data ( Fig. 2). While no expression was observed in non-degenerated young samples (Fig. 9a), the observed staining intensity was similarly high in symptomatic degenerated IVD samples, showing a moderate signal positivity in up to 20% of the cells, and the asymptomatic aged sample (Fig. 9b-d), presenting moderate to strong positivity in most of the cells. In the examined symptomatic degenerated IVD tissue samples, TRPC6 was highly expressed in 6 out of 7 samples (6.00 ± 2.38), with an expression pattern comparable to the TRPC6 gene expression results on degeneration grade and Modic score (Figs. 3a and 5b). One sample was considered to be negative for the TRCP6 protein expression with the overall score value 2, which was a result of lower samples quality with insufficient amount of preserved tissue structure.
TRPML1 protein expression was weak to absent in asymptomatic and non-degenerated IVD samples (Fig. 10e,f). In symptomatic degenerated IVD samples, TRPML1 protein expression was observed in 6 out of 7 cases ( Fig. 10a-d), with an average final score of 4.57 ± 1.72, and was characterized by weak to moderate signal positivity and mostly in areas, where clusters of cells were present. One sample was negative for TRPML1 presenting only single cells with a faint signal (score 2). In general, the TRPML1 signal intensity was weaker compared to TRPC6 and TRPM2, and the positivity was limited to a lower number of cells. In contrast, on the gene level TRPML1 expression was slightly higher than TRPC6 and TRPM2 expression (Fig. 2). Within the symptomatic degenerated IVD samples, TRPML1 tended to be expressed overall higher in samples collected from patients experiencing intense pain (Fig. 10a,b) in comparison to disabling pain (Fig. 10c,d), and asymptomatic and non-degenerated samples (Fig. 10e,f), hence illustrating a similar pattern as observed on the gene level (Fig. 6c).

Discussion
Degeneration of the intervertebral disc is one of the leading causes of low back pain, which not only lessens the patients' quality of life, but also constitutes a high financial burden on the society. With the aging population, the occurrence as well as costs of IVD degeneration and LBP are predicted to grow. Hence, it is crucial to gain a better insight into the molecular processes of IVD degeneration along with a potential to develop new and targeted  www.nature.com/scientificreports www.nature.com/scientificreports/ treatment strategies. Recently, TRP channels, which constitute a family of ion channels that were shown to play an important role in various diseases 20 , have emerged as potential targets 10 , but their exact function in the IVD remains unknown. This study is the first one to report a complete screening of TRP channels in non-degenerated and degenerated IVDs. Our results confirm the important role of the TRPC subfamily in the IVD, but also highlight previously unexplored targets stemming from e.g. the TRPM or TRPML subfamily.
According to our results and in line with previously reported findings 13 , the highest observed expressed TRP channel overall in the IVD is TRPV4. However, no significant differences were found for any of the analyzed patient/tissue factors for TRPV4. Similarly, TRPV1 was well expressed in both non-degenerated and degenerated IVD tissue, but no significant differences were found when evaluated against patient-or tissue-related factors. TRPV1 is responsive to pro-inflammatory agents (e.g. TNF-α), which are known to be higher expressed in painful IVD tissue. Therefore, the inflammatory environment associated with DDD likely increases the probability for channel opening 21 . Other members of the TRPV channel family, but TRPV1 and TRPV4, where not detected or were expressed at the detection limit during the initial screening and were hence excluded from the further analysis in this study. The TRP vanilloid (V) subfamily constitutes of six members (V1-6) in humans, with a main role in thermo-sensation. Additionally, TRPV4 was shown to also play a role in mechano-and osmo-sensation in the IVD and cartilage 22 and TRPV1 may be gated by low pH (≤5.9) 20 . Upon channel activation (via e.g. mechanical or osmotic stress), TRPV4 causes transient increases of intracellular calcium, which acts as a secondary messenger that activates NFAT5/TonEBP, MAP kinases and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathways 23 , which are involved in cell hemostasis as well as inflammation 24 . It was demonstrated that TRPV4 expression increases with decreasing IVD osmolarity as observed during IVD degeneration 9 and may correlate with pro-inflammatory molecules such as TNF-α 9 and interferons (IFN) A1 and B1 13 in the degenerated IVD. Interestingly, TRPV1 was shown to function as an osmosensor in the brain 25,26 , although no supporting data exists on such function in the IVD. Taken together, since TRPV1 and TRPV4 are expressed in both non-degenerated and degenerated IVD tissue, it may indicate that both channels are important for the IVD cells fundamental functions. However, future studies should confirm their exact function and further analyze the previously reported involvement of TRPV4 in IVD degeneration 9 .
In contrast to TRPV4 and TRPV1, the expression of TRPC6 was broadly affected by patient-related characteristics. TRPC6 belongs to the TRP canonical (C) subfamily, together with five other channels (C1-6). The TRPC subfamily may be involved in mechanosensing 14,22 , as well as in pain and inflammation in joint tissues 10 . We were able to show that the mRNA expression of TRPC6 was significantly higher in degenerated IVDs than in non-degenerated IVDs and that its expression tended to increase with the IVD degeneration grade as previously reported 13 . In line with it, on the protein level, TRPC6 expression was consistent in the asymptomatic and symptomatic degenerated IVD samples and it was absent in non-degenerated IVD samples. Moreover, we showed that higher TRPC6 gene expression was associated with specific Modic types, moderate pain and tended to increase with the duration of disease symptoms. Due to limited sample size, we were not able to confirm these findings on the protein level. Interestingly, no zonal (AF, NP) differences were found (data not shown) in contrary to a previous publication 13 . However, this may be due to the fact that a clear separation of the zones within IVD tissue (especially in higher stages of degeneration) is difficult, as well as due to a smaller sample size. TRPC6 was previously shown to correlate with inflammatory markers such as TNF-α, IL-6, IL-8 and IL-15 in the IVD 13 and potentially plays a significant role in modulating immune response, as well as mediate inflammatory processes in other tissue types such as lung, kidney and neurons 27 . These data indicate that TRPC6 could be an interesting pharmaceutical target for pain treatment of patients with high degree of disc degeneration.
Besides TRPC6, three other members of the canonical subfamily, TRPC1, TRPC3 and TRPC4, were screened in this study. TRPC1 and TRPC3 were detected at similar levels in non-degenerated and degenerated IVDs and they showed no statistically significant relation to any of investigated factors. Both channels were previously detected on the mRNA, but not on the protein level in human chondrocytes, where elevated TRPC3 mRNA expression levels, but not TRPC1 were observed with higher passaging 25 . TRPC1 and TRPC3 are also hypothesized to have mechanosensitive function, but their exact role in the IVD remains thus far unknown 10 . For the last investigated canonical member, TRPC4, we could show differences in its expression with the degree of IVD www.nature.com/scientificreports www.nature.com/scientificreports/ degeneration and the type of Modic change. Although previously unexplored in the IVD, TRPC4 is known to be expressed in human chondrocytes 25 as well as in the central nervous system (CNS), smooth muscle cells, kidney and endothelium 22 . In the CNS, it contributes to the axonal regeneration after the dorsal root ganglion injury 28 and hence may be involved in the regulation of IVD innervation.
The TRPM2 and TRPM7 channels belong to the TRP melastatin (M) subfamily that consists of eight members (M1-8) in humans and is most known for its function in thermo-sensation 22 . Importantly, TRPM7 was also detected in human chondrocytes, where it showed the overall strongest mRNA expression out of all tested target 25 . Similarly as in the human chondrocytes 25 , TRPM7 was overall consistently expressed in non-degenerated and degenerated human IVDs, albeit with higher expression in IVDs with higher degeneration grades. Although no other data on TRPM7 in the IVD exist, it was shown that TRPM7 may act as a mechano-sensor and mediate osmolarity-induced cell volume changes in the kidney and salivary glands 29,30 . A non-degenerated IVD tissue is characterized by diurnal osmolarity changes (~400-500 mOsm) while reduced tissue osmolarity (~300 mOsm) is a hallmark of IVD degeneration 31 . Since an osmotic balance is necessary for an IVD to maintain its function, it could be hypothesized that TRPM7 may be involved in osmo-sensing and osmo-adaptation in the IVD.
TRPM2 is hypothesized to play an important role in pain development by acting as an oxidative stress and reactive oxygen species (ROS) sensor in immune cells 32 and regulator of pro-inflammatory cytokine release in the CNS 33 . In this study, TRPM2 was expressed at higher mRNA levels in degenerated IVDs compared to non-degenerated IVDs. Additionally, in the non-degenerated IVD samples, TRPM2 was detected only in the AF tissue, whereas in the degenerated IVDs, TRPM2 was detected both in the AF and NP tissue. These data imply that TRPM2 exists in a native human AF tissue, however its elevated levels, as well as its occurrence in both AF and NP tissue may be due to the nerve and blood vessels ingrowth, which are commonly observed during IVD degeneration 34 , as well as to a possible contamination during tissue separation. On the protein level, TRPM2 was consistently well expressed in symptomatic and asymptomatic degenerated IVD samples and to the lesser extend or not at all in the non-degenerated samples, hence pointing towards degeneration-driven TRPM2 expression. Interestingly, we could show that TRPM2 mRNA expression was strongly linked to pain. We could also demonstrate that the TRPM2 mRNA expression was significantly decreased in patients, who underwent steroid treatment and was higher in patients who described their pain as intense. However, it should be noted that the majority of patients who described their pain as disabling were also the ones who received steroid treatment. These data indicate a pivotal role of TRPM2 in pain mediation in the IVD and suggest that its modulation may create new opportunities for therapeutic strategies of the low back pain treatment.
The TRP mucolipin (ML) subfamily consists of three, thus far poorly characterized, members (ML1-3) in humans. Recent evidence suggest that TRPML channels may be involved in the lysosomal function and regulation www.nature.com/scientificreports www.nature.com/scientificreports/ of autophagy 22,35 , which can be also present in the IVD. Increased TRPML1 activity was observed during stress conditions and can be triggered by catabolic signaling 36 . Shen et al. 37 showed serum deprivation and inflammation induced autophagy of rat AF cells. In contrast, compression and ROS induced generation of autophagy in rat NP cells 38 and reduced the catabolic effects of IL-1B and TNF-α 39 . However, whether and how TRPML channels are involved in autophagy in the IVD is unknown. Here, we could show for the first time that the TRPML subfamily is present in the IVD, with TRPML1 being significantly higher expressed on the gene level in the degenerated IVDs, in an inverse, grade-depended manner. On the protein level, TRPML1 was detected in symptomatic degenerated IVDs, but not in non-degenerated and asymptomatic (non-painful) IVD samples. Additionally, we could show that the expression of TRPML1 on gene and protein level was higher in patients who described their pain as intense.  www.nature.com/scientificreports www.nature.com/scientificreports/ The TRP polycystin (P) subfamily compromises three channel members (TRPP1 known as PKD2 (previously TRPP2), TRPP2 known as PKD2L1 (previously TRPP3) and TRPP3 known as PKD2L2 (previously TRPP5) 40 . Additional information on the TRPP channel subfamily nomenclature is provided in the Supplementary Material. Importantly, TRPP1 (PKD2) does not produce cation currents until assembled with PKD1 41 . Although we were able to detect PKD1 and TRPP1 (PKD2) in non-degenerated and degenerated specimens at relatively high levels (yet unaffected by patient-or tissue-related characteristics), the lack of general information on their function hinders interpretation of their relevance in the IVD. It was proposed that TRPP1 (PKD2) might either be a Figure 6. The mRNA expression of (a) TRPC6 was significantly higher in the moderate pain group compared to the intense (p = 0.007) and disabling (p = 0.03) pain groups, whereas the mRNA expression of (b) TRPM2 and (c) TRPML1 was statistically different between the intense and disabling pain groups (p = 0.04). Although no statistical differences were found between the duration of symptoms (acute: >2 months, sub-acute: 2-12 months, and chronic: >1 year), (d) the TRPC6 mRNA expression tended to increase with the duration of pain symptoms. Data presented as 2 −ΔCt values (relative to YWHAZ). Asterisks indicates a statistically significant difference between the groups indicated (*p < 0.05). www.nature.com/scientificreports www.nature.com/scientificreports/ mechanosensitive channel or regulate mechanosensitive channels by interacting with the TRPV4 in the cilium 40,42 and rat endothelial cells 43 .
Although TRP channels constitute potential pharmaceutical targets for the treatment of disc-related back pain and IVD degeneration, their modulation has to be approached with caution as several TRP channels are also present in non-degenerated IVDs and therefore are likely to be of relevance not only in tissue pathophysiology, but also homeostasis. Therefore, the channel activity in non-degenerated and degenerated IVDs should also be considered and determined in order to design adequate treatment strategies such as pharmacological inhibition of TRP channels. In recent years, several potent and selective TRP channels agonists and antagonists have been identified (e.g. GSK1016790A activates and GSK2193874 blocks TRPV4, Waixenicin A inhibits TRPM7 etc.), however many compounds commonly used in TRP channel research also activate/inhibit a broad range of other channels (e.g. SKF96365 can inhibit TRPC subfamily, as well as TRPV4) [44][45][46] . Future developments of specific modulators of those TRP channels potentially involved in back pain may constitute a viable alternative to intradiscal steroid injections (with high side effects) or spinal surgery (with high complication rates). Furthermore,  Modic  type  TRPC6  TRPM2 TRPML1   35  1  AF  76  m  DDD  3  2  x  x   36  1  NP  47  m  DH  3  1  x  x   37  1  mix  59  f  DH  3  1  x  x   38  1  AF  56  f  DDD  3  n/a  x  x   39  1  AF  43  f  DDD  3  2  x  x   40  1  mix  32  f  DDD  3  1  x   41  1  mix  50  f  DDD  3  2  x   42  1  mix  47  m  DDD  3  1  x   43  1  mix  74  m  DDD  3  2       www.nature.com/scientificreports www.nature.com/scientificreports/ future studies should determine whether enhanced mRNA and protein expression in degenerated samples (e.g. for TRPC6, TRPM2 and TRPML1) is a protective mechanism or de facto contributes to the progression of IVD degeneration. Clear technical limitations of this study include a small sample size, especially of non-degenerated samples that are rare to obtain. Likewise, a bigger sample size could allow for a more comprehensive screening of selected targets on the protein level.
This study is the first to present a complete screening of all currently known TRP channels in non-degenerated and degenerated IVD tissue. In summary, our results indicate that TRP channels may play an important role in IVD degeneration, as well as in low back pain. We were able to confirm the presence and relevance of previously described TRP channels (e.g. TPC6) in degenerated IVDs, but also reveal novel TRP candidates originating e.g. from the TRPM, TRPML and TRPP subfamilies. These findings may be of importance in designing future anti-pain treatment strategies, as well as for the better understanding of the molecular mechanisms underlying disc degeneration and degenerative disc disease.

Materials and Methods
Sample collection. A total of 34 IVD samples were obtained from 27 individuals (mean age = 49.7 [age range 17-80 years]): • Twenty individuals underwent elective spinal surgery due to IVD degeneration/herniation (=22 degenerated IVD samples). Informed consent for sample collection was obtained from each patient and the study was approved through the local ethics committee (Ethics Committee of the Canton Lucerne/Switzerland, #1007). • Seven individuals were organ donors without any signs of IVD degeneration and no history of back pain (=12 non-degenerated IVD samples). Informed consent for sample collection was obtained from family members and the study was approved through the local ethics committee (A04-M53-08B). Spines were harvested within 3 h of the aortic clamping. www.nature.com/scientificreports www.nature.com/scientificreports/ Following macroscopic tissue evaluation, each sample was characterized as NP (degenerated n = 10, non-degeneratedd n = 7) and/or AF (degenerated n = 10, non-degeneratedd n = 5) whenever possible. Samples in which no distinction was possible were termed mix (degenerated n = 2). Assessment of the disease state was performed using Pfirrmann grading (IVD degeneration) and Modic grading (endplate changes) for the degenerated samples, as well as Thompson grading for the organ donor samples. Detailed patient information is given in Tables 1 and 2. All methods were performed in accordance with institutional guidelines and regulations.
RNA extraction from tissue. IVD tissue samples were saved in RNAlater (Thermo Fisher, Switzerland) and transported to the laboratory. Upon processing, RNAlater was aspirated and tissue samples were shock frozen in liquid nitrogen, and pulverized using custom-made grinders. The obtained tissue powder was transferred into TRIzol (1 ml per 200 mg tissue, Thermo Scientific, Switzerland) and the sample was further homogenized with a polytron three times for 30 seconds (POLYTRON ® PT 10/35 GT), with cooling on ice in between. Next, samples were incubated for 5 min at RT, vortexed and centrifuged (4 °C, 12000 g, 15 min) to remove tissue debris. Supernatants were supplemented with chloroform (1 part chloroform to 5 parts sample), vortexed and incubated for 5 min at RT. Next, samples were centrifuged (4 °C, 12000 g, 10 min), the aqueous phase was mixed with 70% ethanol (1:1 ratio) and RNA was purified by the RNeasy Mini Kit (Qiagen, Switzerland), according to the manufacturer's recommendation. The quality and quantity of RNA was quantified using a Nanodrop (Thermo Fisher, Switzerland), specifically controlling the 260/280 and 260/230 ratio.

RNA extraction from isolated IVD cells. Five out of 22 degenerated IVD samples with sufficient size
were not only used for direct RNA isolation as described above, but also for IVD cell isolation in order to test whether enzymatic digestion affects the TRP channel expression profile (see Supplementary Material). cDNA synthesis and pre-amplification. Extracted RNA was reverse transcribed into a cDNA in a total volume of 60 µL, using the reverse transcription kit (Thermo Fisher, Switzerland). Subsequently, a preamplification step with the TaqMan PreAmp Master Mix (2×) (Thermo Fisher, Switzerland) and Custom TaqMan PreAmp Pools (Thermo Fisher, Switzerland) was conducted according to the manufacturer's protocol. Briefly, 100 ng of cDNA in 5 µL total volume was mixed with 5 µL of Custom TaqMan PreAmp Pool and 10 µL of TaqMan PreAmp Master Mix (2×), followed by a pre-amplification through 14 cycles at 60 °C/90 °C.  TRPM2 TRPML1 TRPC6 TRPM2 TRPML1 TRPC6 TRPM2 TRPML1   35  1  AF  1  1  5  1  6  2   36  1  NP  2  1  4  3  6  4   37  1  mix  1  3  4  3  5  6   38  1  AF  2  1  5  2  7  3   39  1  AF  2  2  5  3  7  5   40  1  mix  3  5  8   41  1  mix  2  5  7   42  1  mix  1  1  2   43  1  mix  2  3  5   44  1  NP  2  2 Table 6. Immunohistochemistry (IHC): Evaluation. 1 = degenerated; 0 = non-degenerated; x = sample on which IHC was performed with a listed antibody; scoring system is described in Table 5.  Immunohistochemical (IHC) analysis of TRPC6, TRPM2 and TRPML1 expression in human IVD tissue. Immunohistochemistry specific for TRPC6, TRPM2 and TRPML1 was performed on 3 μm histologic sections cut from paraffin embedded IVD tissue samples that were excised from patients undergoing spinal surgery due to back pain (n = 7 for each marker). Additional asymptomatic IVD sections (n = 3) were added for each marker (detailed patient information given in Tables 4 and 5). All sections were deparaffinized. For TRPM2, heat-mediated antigen retrieval (0.1 M citrate puffer, ph 6.0) was used for 20 min (97 °C, Agilent Dako PT-Link Pre-Treatment Module), while all other antibodies were "ready-to-use". The endogenous peroxidase was blocked with 0.2% sodium azide and 3% hydrogen peroxide; the non-specific protein was bound by "Protein Block Serum-Free" (Agilent Dako X0909 Lot#1014188). Thereafter, sections were incubated with rabbit polyclonal anti-TRPC6 (1:50, Alomone, #ACC-120, lot: ACC120AN0350), rabbit polyclonal anti-TRPM2 (1:50, Abcam, ab11168, lot: GR30307-24) or rabbit polyclonal anti-Mucolipin 1 (TRPML1, 1:300, Osenses, OSM00039W, lot: Rb0064-180807-WS) overnight at room temperature. Secondary detection was performed with goat anti-mouse and anti-rabbit immunoglobulins IgG (Dako, EnVisionTM + Dual Link System-HRP) for 30 min at room temperature. The amino-9-ethyl-carbazole substrate kit (Dako) was employed as a chromogen. Finally, the sections were counter-stained with Gill's hematoxylin for 3 min, and cover-slipped with an aqueous mounting media (Glicerine, Sigma-Aldrich). The evaluation was performed on the Leica DMR System with Leica DFC320 Camera (Leica Microsystems) and compared with image scans. The images were scanned using NanoZoomer S360 Digital slide scanner: C13220-01 and extracted with NDP.view2 Viewing software. All slides were examined and semi-quantitatively evaluated by two investigators for percentage of positively stained cells and the intensity of the reaction product (visible staining). The two scores (A,B) were added together for a final score with eight possible values. Scores of 0 to 2 were considered negative and scores of 3 to 8 were considered positive ( Table 7). The complete IHC evaluation is presented in Table 6.
Statistical analysis. Data consistency was checked and data were screened for outliers. Continuous variables were also tested for normality by using skewness, kurtosis,and omnibus test. Variance homogeneity between both groups were tested by using variance ratio test and modified Levene test. In case of deviation of normality, randomization tests were computed based on Monte Carlo simulations (due to the small sample sizes). In case of normality, Student t-tests for equal or Aspin-Welch unequal variance tests were applied. All reported tests were two-sided, and p-values < 0.05 were considered as statistically significant. All statistical analyses in this report were performed by use of NCSS ( Ethical approval. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.

Data availability
The datasets generated and analyzed during this study are available from the corresponding author on a reasonable request.