The Neurotrophic Factor Artemin Promotes Pancreatic Cancer Invasion

Abstract

Objective:

To analyze the role of Artemin in pancreatic ductal adenocarcinoma (PDAC) in terms of expression, influence on cancer cell behavior, and pain correlation.

Summary Background Data:

PDAC is characterized by prominent local nerve alterations and perineural invasion, which frequently affects the extrapancreatic nerve plexus, causing severe pain and precluding curative resection. Artemin, a neurotrophic protein controlling growth, regeneration, and survival of neurons was analyzed to highlight the neuro-cancer interactions in PDAC.

Methods:

Artemin and its receptors (GFRα3/RET) were studied in PDAC tissues and normal pancreas by Western blot analysis and immunohistochemistry. RNA expression was analyzed in pancreatic tissues (normal, cancer) and pancreatic cancer cell lines by QRT-PCR. To evaluate whether Artemin influences cancer cell proliferation and invasion, MTT-growth and Matrigel-invasion assays were used. In addition, the tissue expression of Artemin was correlated with pain in PDAC.

Results:

Artemin and GFRα3/RET were both detected at enhanced levels in PDAC compared with normal pancreas, localizing predominantly in hypertrophic nerves and arterial walls, as well as in cancer cells of primary and metastatic lesions. The levels of Artemin and GFRα3 did not correlate with pain in PDAC patients. However, Artemin promoted pancreatic cancer cell invasion up to 5-fold, without affecting cancer cell proliferation.

Conclusion:

Artemin expression was not associated with pain in PDAC. However by increasing cancer cell invasion, Artemin seems to influence neural invasion and thereby contribute to cancer cell spreading along pancreatic nerves.

In this study, we analyzed the expression of Artemin and its receptors GFRα3/RET in pancreatic ductal adenocarcinoma and evaluated its possible influences on pain and pancreatic cancer cell behavior. Artemin and GFRα3/RET are overexpressed in pancreatic cancer with no correlation to pain. However, Artemin promotes pancreatic cancer cell invasion without increasing proliferation.

Pancreatic cancer is an aggressive malignancy with an extremely poor prognosis. At the time of diagnosis, most patients have locally advanced disease and/or distant metastatic lesions precluding RO resection.^1–3 Pancreatic cancer cells frequently come in intimate contact with intrapancreatic nerves and thereby alter, invade, and damage them.⁴ Perineural invasion extending into the extrapancreatic nerve plexus is a histopathologic characteristic in pancreatic cancer, which leads to abdominal pain and retropancreatic tumor extension.^5–7 It precludes curative resection, promotes local recurrence, and finally negatively influences the prognosis of the patients.4,5,7,8 However, the mechanisms contributing to alteration and invasion of intrapancreatic nerves and the spread of cancer cells along extrapancreatic nerves in pancreatic ductal adenocarcinoma (PDAC) are still poorly understood. Therefore, neurotrophic factors and the glial cell line-derived-neurotrophic factor (GDNF) family of proteins are of interest due to recent experimental data showing their involvement in neuro-cancer interactions and carcinogenesis in PDAC.^9–12 For example, GDNF increases the invasion rate of pancreatic cancer cells by activating matrix metalloproteinase 9 (MMP-9).^9–11

Artemin belongs to the GDNF family of ligands consisting of 4 members: GDNF,^13,14 Neurturin,¹⁵ Persephin,¹⁶ and Artemin.^14,17 The signaling specificity of these ligands is achieved by preferential binding to different isoforms of the GDNF family receptor alpha (GFRα): GFRα1 (GDNF), GFRα2 (Neurturin), GFRα3 (Artemin), and GFRα4 (Persephin).^18–21 Subsequent signaling is mediated by a common intracellular pathway: the receptor tyrosine-kinase RET^18,22 with downstream targets in the Ras/ERK-, PI3K/AKT-, p38/MAPK-, and JNK-pathways affecting cell growth, differentiation, survival, and apoptosis.^18,22 Artemin enhances survival, proliferation, and regeneration of neurons and increases the neuronal density and neurite outgrowth of sympathetic neurons in vitro.^17,19,23 Additionally, Artemin acts as a guidance molecule that induces migration andaxonal projection from sympathetic neurons by signaling through GFRα3/Ret complexes.²⁴ Recently, it has been reported that Artemin also reduces neuropathic pain and is involved in restoring neural damage.²⁵

Although neuronal alteration is a sine qua non in PDAC, the role of Artemin in this context is yet unknown. Since neural invasion and spreading of cancer cells along pancreatic nerves limits curative resection and contributes to local recurrence, in the present study the expression of Artemin and its receptors (GFRα3/RET) was studied in pancreatic cancer. Furthermore, associations to pain, cancer cell proliferation, and invasiveness were analyzed.

PATIENTS, MATERIALS, AND METHODS

Patients and Tissues

The PDAC tissue samples (n = 70) were collected from patients following tumor resection in Bern, Switzerland and Heidelberg, Germany. Written informed consent was obtained from all patients. The study was approved by the ethics committees of the University of Bern, Switzerland and the University of Heidelberg, Germany. Normal pancreatic tissue samples were collected from healthy organ donors (n = 19). Resected pancreatic tissues were subdivided into parts and the aliquots were 1) frozen in liquid nitrogen and stored at −80°C for protein extraction, 2) taken into RNA later (Ambion, Huntington, UK) for RNA extraction, and 3) immediately fixed in paraformaldehyde and later embedded in paraffin for immunohistochemistry.

In all PDAC patients, the individual pain score was prospectively registered prior to the operation according to pain intensity and frequency. The intensity of pain was graded by using a short scale: 0 = none, 1 = mild, 2 = moderate, and 3 = strong pain. In addition, the frequency of pain was graded as 3 = daily, 2 = weekly, and 1 = monthly. To calculate the total degree of pain, pain intensity and pain frequency of each individual were multiplied. According to the final pain score, the patients were divided in 3 groups: Pain I (0) representing the group of patients who did not have any pain, Pain II (1–3) representing the group of patients who suffered from mild pain, and Pain III (4–9) the group that mirrored the patients with moderate to severe pain.

Reagents

Reagents purchased: RPMI-1640, DMEM, Trypsin-EDTA, and penicillin-streptomycin from Invitrogen (Karlsruhe, Germany); FBS from PAN Biotech (Aidenbach, Germany); mouse recombinant Artemin from R&D Systems (Minneapolis, MN); Artemin-, GFRα3- and RET-rabbit polyclonal antibodies from Abcam (Cambridge, UK). Anti-rabbit IgG HRPO-linked antibodies and ECL immunoblotting detection reagents from Amersham Biosciences (Amersham, UK). DAKO Envision system (Hamburg, Germany) was used for immunohistochemistry. All other reagents were from Sigma Chemical Company (Taufkirchen, Germany).

Real-Time Light Cycler Quantitative Polymerase Chain Reaction (QRT-PCR)

All reagents and equipment for mRNA/cDNA preparation were purchased from Roche Applied Science (Mannheim, Germany). Extraction of mRNA from human pancreatic cell lines and normal and pancreatic cancer tissues were prepared by automated isolation using the MagNA Pure LC instrument and isolation kits I (for cells) and II (for tissue). cDNA was prepared using 1st strand cDNA synthesis kit for RT-PCR according to the manufacturer's instructions. Real-time PCR was performed with the Light Cycler Fast Start DNA SYBR Green kit. All primers were obtained from Search-LC (Heidelberg, Germany). The calculated number of specific transcripts was normalized to the housekeeping genes cyclophilin B and hypoxanthine guanine phosphoribosyltransferase and expressed as an amount per microliter of input cDNA, as described previously.²⁶

Western Blot Analysis

Protein extraction and Western blot analysis of cell culture monolayers or tissues were performed as described previously.²⁷ Artemin, GFRα3 and RET antibodies were used at the dilution of 1:500 at 4°C overnight. Antibody detection was performed with an enhanced chemiluminescence reaction. The equal loading was assured by stripping the blots and reprobing with anti-γ-tubulin antibodies (Santa Cruz Biotechnologies, Heidelberg, Germany). After film scanning, densitometric analysis was performed using the ImageJ software (NIH). Specific signal intensity was calculated and presented as a fold increase over background.

Immunohistochemistry

Paraffin-embedded tissue sections of normal pancreas, pancreatic cancer and liver metastasis were analyzed by immunostaining using the DAKO Envision system as described previously.²⁷ Artemin, GFRα3 and RET antibodies were used at the dilution of 1:100. Rabbit polyclonal IgG (DAKO) was used as negative control instead of primary antibody. Digital imaging was performed using Zeiss AxioCam HR system (Carl Zeiss AG, Oberkochen, Germany).

Cell Culture

Human pancreatic cancer cell lines (Aspc1, BxPc3, Capan1, Colo-357, MiaPaCa2, Panc1, SU86.86, and T3M4) were either purchased from ATCC (Rockville, MD) or received as a gift of Dr. R. S. Metzgar (Durham, NL) and routinely grown in complete medium (RPMI-1640 medium supplemented with 10% FBS, 100 U/mL penicillin and 100 ìg/mL streptomycin) at 37°C saturated with 5% CO₂ in a humid atmosphere.

MTT Assay

To assess cell growth, the MTT (3-(4, 5-methylthiazol-2-yl)-2, 5-diphenyl-tetrazolium bromide) test was used as published before.²⁶ Briefly, cells were seeded at a density of 5000 cells/well in 96-well plates, grown overnight, and exposed or not to Artemin at the concentrations of 10 ng/mL and 100 ng/mL. After 24, 48, or 72 hours of incubation, MTT was added (50 μg/well) for 4 hours. Formazan products were solubilized with acidic isopropanol, and the optical density was measured at 570 nm. All experiments were made in triplicate and repeated 3 times.

Matrigel-Based Invasion Assay

Invasion assay was performed using BD Biocoat Matrigel invasion chambers with 8-μm pore size (BD Biosciences, Heidelberg, Germany). According to the manufacturer's instructions, the Matrigel was hydrated with 0.5 mL DMEM (serum-free) and incubated for 2 hours. Five thousand cells (MiaPaCa2, T3M4, Colo-357, and SU86.86) were seeded into the upper chamber of the invasion chambers and incubated for 24 hours at 37°C, 5% CO₂ atmosphere. To detect the influence on invasion, Artemin was added to the cells into the upper chamber (10 ng/mL or 100 ng/mL) and incubated for 24 hours. For assessment of chemotactic potential, Artemin was put in the lower chamber of the plate, as described previously.^9–11 The noninvading cells were removed from the upper surface of the membrane by wiping with a cotton-tipped swab. Cells adhered to the lower surface were fixed in 75% methanol mixed with 25% acetone and then stained with 1% Toluidine blue. To calculate the total number of invading cells, the membranes were scanned and cell number in every microscopic cutout of the mosaic image of the membrane was counted using the software Zeiss KS300 (Carl Zeiss AG). The assays were performed in duplicate and repeated 3 times.

Statistical Analysis

Results are expressed as mean ± SEM. For statistical analysis, the Student t test or nonparametric Mann-Whitney U test was used. Statistical nonparametric correlation was assessed by the Spearman-Rho test. Significance was set at P< 0.05.

RESULTS

Protein and mRNA Expression of Artemin and Its Receptors in PDAC

Initially, a quantitative RT-PCR analysis was performed to investigate Artemin-mRNA expression in normal pancreatic tissue and in PDAC. In normal pancreas, the average expression of Artemin was 24 ± 7 transcripts/μL (mean ± SEM), and in PDAC Artemin expression was increased nonsignificantly (34 ± 4 transcripts/μL) (Fig. 1A). Western blot analysis confirmed the accumulation of Artemin and GFRα3/RET in the cancer tissues (Fig. 1B). Interestingly, Artemin protein levels showed a more than 30-fold (P < 0.01) increase in cancer tissues compared with normal tissues. The protein levels of the Artemin binding receptors GFRα3 and RET were also significantly (P < 0.01) increased in pancreatic cancer compared with normal tissue by 20-fold and 3-fold, respectively. To determine whether cancer cells expressed Artemin-mRNA, QRT-PCR of Artemin was performed in the pancreatic cancer cell lines. All 8 pancreatic cancer cell lines tested expressed Artemin, ranging from 15 ± 4 transcripts/μL in Panc1 to 1272 ± 225 transcripts/μL in T3M4 cells (Fig. 1C).

FIGURE 1. Artemin and its receptor complex (GFRα3/RET) in pancreatic tissues. A, Artemin-mRNA expression, analyzed by QRT-PCR in pancreatic tissues of healthy organ donors (n = 19) and in pancreatic cancer patients (n = 70), did not show any significant difference. B, Western blot analysis of Artemin, GFRα3, and RET in normal pancreas and in pancreatic cancer showed a 30-fold increase of Artemin, a 20-fold increase of GFRα3, and a 3-fold increase of RET in PDAC. The comparison was made by the ImageJ software. C, Artemin-mRNA expression analyzed by QRT-PCR in the pancreatic cancer cell lines Aspc1, BxPc3, Capan1, Colo-357, MiaPaCa2, Panc1, SU86.86, and T3M4. B, Western blot analysis of Artemin and GFRα3 demonstrating the ubiquitous protein expression in the pancreatic cell lines Aspc1, BxPc3, Capan1, Colo-357, MiaPaCa2, Panc1, SU86.86, and T3M4.

Immunolocalization of Artemin and Its Receptors in PDAC

In normal pancreas, Artemin immunoreactivity was faintly present in the smooth muscle cells of arterial walls with no immunoreactivity in intrapancreatic nerves or ducts (Fig. 2A, B). This pattern changed completely in pancreatic cancer. Intense Artemin immunoreactivity was detected in arteries and moderate to strong immunoreactivity was observed in nerves and ganglia (Fig. 2C, D). The most impressive feature was the strong immunoreactivity of Artemin in tubular complexes, PDAC-associated PanIN-lesions, and cancer cells (Fig. 2D). Furthermore, in liver metastasis strong Artemin immunoreactivity also was present in intra metastatic arteries and in cancer cells (Fig. 2E, F). To elucidate cellular targets of Artemin in the pancreas, immunohistochemistry of the Artemin receptor GFRα3 and RET was performed using consecutive sections of normal and pancreatic cancer tissues. Weak to moderate GFRα3/RET immunoreactivity was present in arteries, nerves, and ductal cells of normal pancreatic tissue. In contrast, the pattern changed in parallel to Artemin in the cancer tissues with strong GFRα3/RET immunoreactivity in nerves, PanIN lesions, and pancreatic cancer cells (Table 1).

FIGURE 2. Immunoreactivity of Artemin in normal pancreas, pancreatic cancer, and liver metastasis. A and B, Immunostaining of Artemin in normal pancreatic tissue samples was only faintly present in the smooth muscle cells of arteries. In pancreatic cancer (C and D), the immunoreactivity pattern changed dramatically and demonstrated a high accumulation in arteries, nerves, PanIN-lesions, and cancer cells. The same pattern was observed in liver metastasis (E–F) (all images, original magnification ×200).

TABLE 1. Semiquantitative Analysis of Artemin-, RET-, and GFRα3 Immunoreactivity in Normal Pancreas (NP) and in Human Pancreatic Ductal Adenocarcinoma (PDAC) Tissues: Localization of Artemin, RET, and GFRα3 in Normal Pancreas and Pancreatic Cancer

Influence of Artemin on Pancreatic Cancer Cell Proliferation and Invasion

Western blot analysis revealed the ubiquitous presence of Artemin and GFRα3 in all tested pancreatic cancer cell lines (Fig. 1D), suggesting that Artemin may directly influence the activity of pancreatic cancer cells. Furthermore, RET was present in all pancreatic cancer cell lines (data not shown). To further investigate whether Artemin may function as a growth factor for pancreatic cancer cells, as demonstrated before in human neuroblastoma cells,¹⁷ all 8 human pancreatic cancer cell lines were exposed to 10 ng/mL or 100 ng/mL of recombinant Artemin. Time dependent analysis of the cell growth was assessed using the MTT test. Treatment of all tested pancreatic cancer cell lines with Artemin led to minor nonsignificant variations in cell proliferation (Fig. 3).

FIGURE 3. Effect of Artemin on pancreatic cancer cell proliferation. The pancreatic cancer cell lines, T3M4 A, Colo-357 B, MiaPaCa2 C, and SU86.86 D, were exposed to 10 ng/mL and 100 ng/mL Artemin for 24, 48, and 72 hours, respectively. Artemin did not show any significant changes in the proliferation rate of the pancreatic cancer cell lines.

To determine a possible contribution of Artemin on the invasive potential of pancreatic cancer cells and on the formation of metastasis, a Matrigel-based invasion assay was used. Exposure of cancer cells to Artemin at a dose of 10 ng/mL increased the invasion ratio 2-fold in T3M4, Colo-357, Mia-PaCa-2, and SU86.86 cancer cells, whereas 100 ng/mL of Artemin promoted invasion up to 6-fold in T3M4, and up to 3-fold in the other tested pancreatic cancer cell lines (P < 0.02) (Fig. 4A). Since augmented chemotaxis may contribute to increased invasiveness, the chemoattractant potential of Artemin on pancreatic cancer cells was also investigated. The chemoattraction ratio rose 3- to 4-fold in T3M4 and SU86.86 cells (P < 0.02) when Artemin was applied at concentrations of 10 ng/mL or 100 ng/mL, respectively (Fig. 4B). Thus, Artemin did not influence pancreatic cancer cell proliferation, but rather had the ability to significantly increase the invasive potential of pancreatic cancer cells.

FIGURE 4. Influence of Artemin on invasion and chemoattraction in pancreatic cancer. A, Artemin changed the invasive behavior of the pancreatic cancer cell lines T3M4, Colo-357, MiaPaCa2, and SU86.86 and increased the invasion ratio of all cells in a dose-dependent manner (P < 0.02). B, To analyze the chemotactic potential, Artemin was added in the lower chamber of the Matrigel invasion plate. Hereby, Artemin increased the invasion potential of the pancreatic cancer cells T3M4 and SU86.86 significantly (P < 0.02).

Since the pro-invasive activity of other GDNF-family member is attributed to the activation of metalloproteinases (MMP),¹⁰ we also investigated whether Artemin-induced cancer cell invasion was associated with the up-regulation of MMPs. All 8 pancreatic cancer cell lines were treated for 24 hours with Artemin (10 ng/mL and 100 ng/mL), then the mRNA-expression of MMP-2 and -9 were detected by QRT-PCR analysis. Neither the expression of MMP-2-mRNA nor of MMP-9-mRNA was affected by Artemin (data not shown).

Correlation of Artemin With Pain in PDAC

There were 31 patients without any pain (Pain I). Nine patients had mild pain (Pain II) and 30 patients had moderate to severe pain (Pain III). To evaluate a possible relationship between Artemin and pain, the groups Pain I and III were compared by both QRT-PCR and Western blot analyses. Artemin-mRNA expression levels were not significantly different between patients in the Pain I group (39.8 ± 6.1 transcripts/μL) and in the Pain III group (45.3 ± 13.3 transcripts/μL) (Fig. 5A). Western blot analysis of Artemin and GFRα3 mirrored the results of the QRT-PCR-analysis, confirming the finding that Artemin and GFRα3 protein levels were not correlated with pain in PDAC patients (Fig. 5B).

FIGURE 5. Correlation of Artemin and GFRα3 with pain in pancreatic cancer. A, Artemin-mRNA expression analyzed by QRT-PCR in pancreatic tissues of patients suffering from moderate to severe pain (n = 30/Pain III) and of patients without any pain (n = 31/Pain I) revealed no statistically significant differences. B, Western blot analysis of Artemin and GFRα3 from patients in the Pain I group with no pain (n = 8) comparing the Pain III group with moderate to severe pain (n = 8) showed no significant changes.

DISCUSSION

Proteins belonging to the GDNF family of neurotrophic factors (GDNF, Neurturin, Persephin, and Artemin) and their receptors GFRα1–4/RET have been proposed as therapeutic agents for neurodegenerative diseases on the basis of their ability to promote survival of various neurons including peripheral autonomic, sensory, central motor, and dopaminergic neurons.^13,28,29 On the other hand, mutations of the RET receptor cause several human diseases such as papillary thyroid carcinoma, multiple endocrine neoplasia (types 2A and 2B), and Hirschsprung's disease.³⁰ Although few reports advocate that GDNF ligands control cancer cell invasion and growth, their exact roles in the pathogenesis of malignant disorders are poorly understood. Pancreatic cancer represents a good model to analyze the implications of such neuro-cancer interaction since perineural invasion ultimately dictates the local recurrence of this disease.^9–11 In this study, we demonstrated for the first time that Artemin and its receptor complex (GFRα3/RET) were overexpressed in PDAC compared with the normal pancreas.

In adult and fetal mice, Artemin expression is present in the nervous system, concentrated in dorsal root ganglia (DRG), immature Schwann cells, and weakly in the brain.¹⁷ Under physiologic conditions, Artemin is localized in vascular smooth muscle cells, in the adventitia of the dorsal aorta and arteries entering the gut.^17,19,24 These data are in concordance with our immunohistochemical findings showing that in the normal pancreas, Artemin was faintly present in smooth muscle cells of arteries but absent in intrapancreatic nerves or ducts. Surprisingly, while the mRNA level of Artemin in pancreatic cancer samples was only slightly increased, the protein level increased up to 30-fold in PDAC. This divergence was also present in the pancreatic cancer cell lines, where the highest protein levels did not always correspond to the level of mRNA expression. The discrepancy between the Artemin-mRNA and protein levels suggest that the excessive increase of Artemin protein may be due to intrapancreatic accumulation of the protein produced by an extrapancreatic secondary source. As shown in the literature, one of the most prominent sites of Artemin production is the DRG.¹⁷ It is possible that continuing damage of intrapancreatic nerves by cancer cell invasion leads to the activation of neurons in the dorsal root ganglia, which produce Artemin. To restore the intrapancreatic neural integrity, Artemin then might be transported to the pancreas via retrograde axonal transport in sensory afferents, which has been described for the GDNF family of ligands like GDNF, Neurturin, and Persephin.³¹

The impressive presence of Artemin in PDAC tissue and pancreatic cancer cells prompts the question, whether in addition to the presumed neurotrophic activity, a direct effect of Artemin on pancreatic cancer cells exists in PDAC or not. Artemin and its receptors GFRα3/RET were abundantly expressed in the pancreatic cancer cell lines. These data provided the basis for considering Artemin as a factor involved in the control of cancer cell behavior, even though the cancer cells were not of neuronal origin. Andres et al²³ and Baloh etal^14,17 demonstrated the increase of neuroblastoma cell proliferation induced by Artemin. However, this effect was not observed in pancreatic cancer cell lines. The molecular basis for this discrepancy may be due to the different nature of the cells (neuronal vs. epithelial) or due to the state of their receptor systems. Most intriguingly, however, was the ability of Artemin to increase the invasive potential of pancreatic cancer cells without affecting their proliferation. This finding may imply its contribution to the perineural invasion as well as to the dismal prognosis of pancreatic cancer patients. With the increased digestive activity of cancer cells toward the extracellular matrix, Artemin might promote their invasive behavior and direct them to intrapancreatic/extrapancreatic nerves. Similar results were presented by Okada et al, who showed that GDNF had chemokinetic effects on pancreatic cancer cells and proposed GDNF as a major mediator of celiac ganglionotropic invasion of pancreatic cancer cells.^10,12 Activation of MMP-9 and integrin expression in tumor cells mediated the pro-invasive activity of GDNF.¹⁰ Since exposure of pancreatic cancer cell lines to Artemin did not alter MMP-2 and MMP-9 expression as quantified by QRT-PCR, we concluded that some other pathway might be involved in the downstream signaling.

One of the most common reasons of local tumor recurrence after curative pancreatic resection is the undetected local cancer cell invasion of the peripancreatic nerves.^6,32,33 In pancreatic adenocarcinoma, cancer cells infiltrate the perineurium and associate with Schwann cells and axons in the endoneurium of intrapancreatic nerves.⁴ This neural destruction leads to a distorted neural homeostasis.⁴ Artemin is an important regulator of the induction of neuronal proliferation and regeneration under physiologic conditions^17,19,24 and GFRα3 is up-regulated in the distal nerve segment after sciatic transsection.³⁴ Therefore, we hypothesize that, following an initial insult to the nerves by cancer cell invasion and inflammation, neurons and/or Schwann cells produce Artemin/GFRα3 in an effort to repair the intrapancreatic nerves. However, the abundance of Artemin attracts further cancer cells to the site of injury, creating a vicious cycle. Ultimately, increased Artemin and GFRα3 levels, due to the continuous nerve injury, may promote cancer invasion and propagation along the neural structures. Perineural invasion extending into the extrapancreatic nerve plexus is a histopathologic characteristic in pancreatic cancer that leads to abdominal pain and retropancreatic tumor extension.^5–7 However, Gardell et al recently showed that systemic application of Artemin reversed the pain syndrome and normalized the morphologic and neurochemical changes in rats after spinal nerve ligation.²⁵ Artemin application in these animals caused no sensory or motor abnormalities and was therefore proposed as a new treatment option for neuropathic pain.²⁵ In our study, there was no obvious correlation between the expression levels of Artemin and pain scores of pancreatic cancer patients. Since perception of pain is one of the most complicated issues a direct correlation between Artemin overexpression and analgesia in PDAC patients would be an oversimplistic explanation. Nevertheless, it is tempting to speculate that the analgesic effect of Artemin could mask the pain incurred by the cancer cell invasion of intrapancreatic nerves. Therefore, the 2 functionally antagonistic effects of Artemin (antinociceptive vs. pro-invasive) may neutralize each other.

CONCLUSION

We demonstrated for the first time that Artemin and its receptors are up-regulated in PDAC and do not correlate with the pain status of the PDAC patients. Because of its potential to increase pancreatic cancer cell invasion, Artemin has to be further evaluated in this context prior to its application as a novel therapeutic agent in neuropathic pain in malignant disorders.

ACKNOWLEDGMENTS

The authors thank Mrs. Kathrin Schneider for her tremendous technical support and Verna and Frank Loosli for the native English correction. The authors are also grateful to Prof. K.-H. Schäfer for reviewing the manuscript.