Wound-Healing Factors Can Prime Head and Neck Cancer Cells to Increase Their Tumor-Forming Capacity
Gustaf Lindgren, MD; Elisabeth Kjell´en, MD, PhD; Johan Wennerberg, MD, PhD; Lars Ekblad, PhD
Abstract
Objectives/Hypothesis: We investigated whether exposing a wound-healing-sensitive cell line to human wound fluid (HWF) could prime the cells to increase their tumor-forming ability in nude mice and, if so, whether this ability can be inhib- ited by pharmacological substances.
Study Design: Experimental animal model.
Methods: Take rate was measured in BALB/c nude mice after pretreatment of the cells with HWF using human serum and fetal bovine serum as controls. Inhibition of signal transducer and activator of transcription 3 (STAT3) with S3I-201 toci- lizumab, and of interleukin 6 receptor (IL6R) with tocilizumab was performed.
Results: Preincubation with HWF resulted in a significant increase in take rate compared to controls. The increase in take rate could be decreased by both STAT3 and IL6R inhibition.
Conclusions: The results indicate that head and neck squamous cell cancer cells might be stimulated to increase their tumor-forming ability both close to a surgical wound and at more distant locations, as a consequence of the wound-healing response. The work also suggests new treatment modalities aimed at decreasing these stimulatory effects.
Key Words: Head and neck squamous cell carcinoma, wound healing, metastasis, interleukin 6 receptor, signal trans- ducer and activator of transcription 3, tocilizumab.
INTRODUCTION
Surgery remains the primary curative treatment for head and neck squamous cell cancer (HNSCC) at most sites, especially in early-stage disease. However, local and regional recurrences are frequent and pose major problems in clinical care. These recurrences also seriously impair the final treatment outcomes, and as a result, HNSCC causes more than 350,000 deaths per year worldwide.1 Addressing mechanisms that increase the risk of recurrence could increase the treatment response for a large number of people.
The origin of recurrences may be remaining precur- sor lesions that undergo additional genetic alterations. These may result in a second invasive cancer or residual cancer cells remaining in the surgical wound in undetectable numbers (minimal residual cancer).2 Another possible source is circulating tumor cells,3–13 introduced to the vascular system before and during surgery. From there, they are supposedly able to enter, and establish growth in, the environment provided by a healing wound, rich in epidermal growth stimuli, in analogy with the tumor self-seeding hypothesis.14
It has been shown that wound-healing factors can stimulate the proliferative capacity of tumor cells.15–25 Determining the effect of wound healing on tumor growth and aggressiveness, and pinpointing the molecu- lar factors that are involved, and potentially could be exploited as drug targets, is a really challenging task. We have adopted a simple model using collected human wound fluid (HWF) to mimic the effect of wound healing in vitro.26 Using this approach, we hope to collect enough information to be able to suggest new therapeu- tic regimens that might be successful in decreasing the tumor stimulatory effect of wound healing.
Previously, we were able to show a cell line–depend- ent effect of early wound-healing factors (i.e., collected during the first day after surgery) on several aggressive characteristics of HNSCC cell lines. In the experiment, one out of four cell lines displayed a considerable increase in aggressiveness, as measured by proliferation, migration, cell spread, and invasive growth, when exposed to HWF. These effects could be largely counter- acted by signal transducer and activator of transcription 3 (STAT3) and interleukin 6 receptor (IL6R) inhibition, suggesting the possibility that the effects might be tar- getable by pharmacological intervention.
In the present work, we investigated the effects of early wound-healing factors in vivo on the cell line that was shown to be sensitive in vitro. Specifically, we were interested to determine if the wound-healing factors could increase the tumor-forming ability in live tissue and also, whether the cells could be primed in one loca- tion for a higher tumor formation capacity at other sites. Inhibition by pharmacological treatment, in analogy to in vitro, was also attempted.
MATERIALS AND METHODS
Cell Line and Growth Conditions
We used the cell line LU-HNSCC-7, which was established in our laboratory from a recurrence of a moderately differenti- ated squamous cell carcinoma of the bucca (rT2, N0, M0).27 It was maintained at 378C under a humidified atmosphere with 5% CO2 in Dulbecco’s modified Eagle’s medium (DMEM) supple- mented with 10% fetal bovine serum gold (FBS) from PAA Laboratories (Pasching, Austria), 100 units/mL penicillin, and 100 units/mL streptomycin sulfate (complete medium). Single tan- dem repeat analysis was performed, showing no cross- contamination with common contaminants or other cell lines handled in the laboratory. The morphology of the cells was checked regularly, showing no visible changes. Tests for myco- plasma infection were negative.
Inhibitory Substances
The STAT3 inhibitor S3I-201 was from Merck (Darmstadt, Germany); tocilizumab (RoActemra) was from Roche AB (Stock- holm, Sweden).
Wound Fluids
HWFs were collected, in accordance with the procedure in clinical routine, in sterile drainage bags during the first 24 hours after operation from patients operated on for benign head and neck disease. The collection was approved by the Lund Eth- ical Review Board (decision ref. DNR 512/2008). All samples were collected with the patients’ informed consent, in compli- ance with the Helsinki Declaration.28 Prior to use in cell cul- tures, the HWFs were centrifuged at 100,000g for 60 minutes at 48C to remove particulate matter, filtered through a 0.2 lm sterile filter, and then stored at 2808C until use. In the reported experiments we used HWFs from one patient. This HWF displayed similar effects in variables measured in vitro compared to the HWFs used in a previous publication.26 Ali- quots were stored at 2808C. Human serum (HS) “off the clot” was obtained from PAA Laboratories.
Tumor Take Rate
The cells were grown until approximately 50% confluent. One day before the inoculations, the growth medium was changed to DMEM with antibiotics and 10% admixture of serum or HWF and inhibitors, as indicated. The cells were tryp- sinized and resuspended in DMEM with 10% FBS, HS, or HWF (the same supplement used at the inoculation), and then incubated for 30 minutes at 378C. After this, the cells were pelleted and resuspended as previously described, and the cell concen- trations determined by manual counting in a hemocytometer (N 5 4 for each sample, to ensure low intra- and interexperimental variations in injected cell numbers). The cells were diluted to appropriate concentrations, and 2 3 0.1 mL was injected subcutaneously on each side of the lower back of 6 to 8-week-old BALB/c nude (nu/nu) mice. The time from trypsinization to injection was approximately 2.5 hours and was kept constant between the cell batches.
After injection, the mice were inspected every day for 1 week, and then 2 to 3 times per week for more than 4 weeks, as specified for each experiment, in a blinded fashion. The pres- ence of visible tumors close to the injection sites was noted, and when they reached a measurable size, the length and width of the tumors were measured for the remaining duration of the experiment to establish if the tumor was growing. Take was defined as the first day a growing tumor, having a doubling time of <15 days and an end volume (last day of the experi- ment) of >50 mm2, was detected. These limitations were set to ensure that only viable, established tumors were studied.
Statistics
The take rate curves were compared using the log-rank (Mantel-Cox) test in the GraphPad Prism 5.0 software package (GraphPad Software Inc., San Diego, CA).
RESULTS
As discussed in an earlier article,26 the control in this type of experiment is a construction, as cells in a tumor do not normally experience several of the soluble factors present in serum, and definitely not those pres- ent in bovine serum. In fact, many molecules released during blood clotting, and therefore present in serum, are themselves part of the wound-healing response. In the previous study,26 we showed a major, cell line–spe- cific effect in vitro of HWF, compared to both FBS and HS, on the LU-HNSCC-7 cell line. In the present experi- ments, we were interested in the effect of the wound- healing factors present in HWF, as compared to those present in serum, on the tumor-forming capacity in vivo. Therefore, we used both FBS and HS as controls in this work too. The incubation time with HWF was set at 24 hours, as this matched the collection time of the HWF, thus imitating the time frame at which tumor cells in a surgical wound could be exposed to the wound-healing factors.
In an initial experiment, the minimum number of LU-HNSCC-7 cells grown under ordinary conditions with 10% FBS, resulting in measurable take during a 30-day inspection period, was 5 3 105 per injection site (not shown). This cell number was chosen for the experi- ments described below, mainly to allow for measurement of increases in take rate.
When the cells were incubated with 10% HWF 24 hours before and during the injection, the take rate increased to 100% (P < .001), as compared to 25%, when the cells were incubated with 10% FBS (Fig. 1). The take rate also increased by exchanging the FBS with HS, but only to 50% (P < .05), indicating a significant difference from the HWF injections (P < .001) (Fig. 1). To investigate if the presence of HWF in the injection volume was important for the increase in take rate com- pared to the effect of preincubation, cells were incubated with 10% HWF either 24 hours before the injection or during the injection. Pretreatment with HWF resulted in an increase in take rate to 65% (P < .001). Addition of HWF during the injection, on the other hand, did not significantly affect the take rate (30% compared to 25% in the presence of FBS, P 5 .77) (Fig. 2). In the above- mentioned experiments, the average doubling time for tumor growth was 5.1 days, with no significant differen- ces between the groups.
We previously reported that STAT3 and IL6R are involved in the response of the LU-HNSCC-7 cell line to HWF in vitro. In the present study, we therefore con- ducted experiments with the STAT3 inhibitor S3I-201 and the IL6R inhibitor tocilizumab to investigate if this signaling pathway was involved also in the increased formation of tumors elicited by HWF in vivo. As the stimulatory effect of HWF was manifested during the preincubation with HWF, the inhibitors were applied before injecting the cells. In this way, we also avoided introducing confounding effects due to interactions between the substances and the mouse tissue. Incuba- tion with 100 nmol/L S3I-201 resulted in a decrease in tumor take rate. The experiment was repeated twice, with similar decrease, but was statistically significant (P 5 .018) in only one experiment (Fig. 3) and not in the other (P 5 .19). Similarly, incubation with increasing concentrations of tocilizumab resulted in a qualitatively dose-dependent decrease in take rate (Fig. 4). The effect of S3I-201 was weaker compared to that of tocilizumab in vitro.26 This fact, in combination with small varia- tions in injected cell numbers between the experimental arms, was probably the reason for the differences in sta- tistical significance between the experiments.
DISCUSSION
As detailed in the introduction, we have previously shown that wound-healing factors can increase the aggressive behavior, in terms of proliferation, migration, cell scattering, and invasive growth, of HNSCC cells in vitro in a cell line–dependent manner.26 The advantage of this methodology was that we were able to study the effect of authentic human wound-healing solutes on human tumor cells, and also the pharmacological inhibi- tion of the effects after isolating important molecular mechanisms. For a more precise mechanistic description of what happens with tumor cells left in a surgical wound, much more elaborate studies would of course be needed, but we think that the methodology was adequate to generate hypotheses for future clinical studies.
However, it is also essential to gain information on the effect of early wound-healing factors in a biological context. Therefore, we set up the current study using the previously identified, HWF-sensitive cell line LU- HNSCC-7 in a nude mouse model. The aim was to inves- tigate if the cell line would behave more aggressively also in biological tissue, if the effect in that context would be more long lasting than previously shown in vitro, and if STAT3 and IL6R inhibition would have a similar effect.
More precisely, we investigated if wound-healing fac- tors can stimulate HNSCC cells to increase their ability to grow after implantation in a new site and also, if this stimulation remains after withdrawal of the wound- healing factors, thus priming the cells to increased tumor-forming ability. This is important, as such effects on tumor cells during surgical extirpation of HNSCC might increase the risk of either reimplantation in the surgical wound after resection, or reseeding of circulating tumor cells in the wound. Also, it might increase the risk of distant metastases as a side effect of the treatment.
In the first experiment, the cells were incubated with HWF prior to, and during, the inoculation. In this setup, the take rate increased radically when comparing HWF to FBS and HS (Fig. 1), demonstrating an increased aggressive potential also in vivo. In the second experiment, we incubated the cells with HWF either only before or only during the inoculation to investigate if the effect on tumor take rate of wound-healing factors had a longer-lasting effect than seen in vitro.26 The cells that were incubated with HWF before the inoculation had a significantly higher take rate compared to those incubated with FBS, but the addition of HWF during the inoculation did not affect the resulting number of tumors (Fig. 2). The results therefore suggest that HNSCC cells can be stimulated to a more aggressive phenotype in the wound and also to be primed by wound-healing factors to increase their tumor-forming ability outside the wound environment.
As a consequence of the low take rate in some treatment groups, we were not able to compare the growth rate of the tumors between all groups. Where a compari- son was possible (shown for the first experiment), there was no consistent effect on the growth rate of the tumors, as measured by the doubling time, indicating that the effect of wound-healing factors was not persis- tent over a longer time. Therefore, it seems that the wound fluid factors influence the initial steps of tumor establishment.
In our previous work, STAT3 and IL6R were seen to be important intracellular mediators of the wound- healing effect. Also in vivo, we were able to inhibit the effect of HWF using the inhibitors S3I-201 and tocilizu- mab (Figs. 3 and 4), the latter at a clinically relevant concentration,29 suggesting that the increase in take rate was, at least partly, dependent on IL-6/STAT3 sig- naling and that inhibition of these pathways may be used for pharmaceutical intervention.
A similar effect of IL6R inhibition has been described before using cancer stem cell–like cells.30 In that study, mice injected with stem cell-like cells were treated with tocilizumab for 4 weeks, leading to a decreased take rate compared to controls. The authors concluded that IL6R signaling might be targeted in HNSCC. Our results also implicate IL6R inhibition as a possible treatment strategy. However, we suggest that the main benefit of this treat- ment may be seen in combination with surgical proce- dures and, possibly, during the first days after surgery. As we previously showed that all tested cell lines expressed IL6R, but only LU-HNSCC-7 the active form,26 the expression of active IL6R could be suggested as a predic- tive marker for IL6R-directed treatment.
As mentioned, locoregional recurrence is a major cause of failure after combined treatment of head and neck cancer. An almost unique feature for head and neck cancer surgery is that the operation is performed in a field contaminated with tumor cells. Consequently, cells remain- ing in the surgical wound are one possible explanation for recurrences; another is seeding of circulating tumor cells in the healing wound area. This latter possibility is rele- vant, not only for head and neck cancer, but also for surgi- cal removal of other solid tumors, and our findings may have implications for other cancer types as well.
CONCLUSION
Our results indicate that HNSCC cells may be stimulated to increase their tumor-initiating ability, both close to a surgical wound and at more distant locations, as a consequence of the wound-healing response. Impor- tantly, this work suggests new treatment modalities aimed at decreasing these stimulatory effects.
BIBLIOGRAPHY
1. International Agency for Research on Cancer. GLOBOCAN 2008. Available at: http://www.iarc.fr/en/media-centre/iarcnews/2010/globocan2008.php. Accessed March 16, 2012.
2. Tabor MP, Brakenhoff RH, Ruijter-Schippers HJ, Kummer JA, Leemans CR, Braakhuis BJ. Genetically altered fields as origin of locally recur- rent head and neck cancer: a retrospective study. Clin Cancer Res 2004; 10:3607–3613.
3. Kulasinghe A, Perry C, Jovanovic L, Nelson C, Punyadeera C. Circulating tumour cells in metastatic head and neck cancers. Int J Cancer 2015; 136:2515–2523.
4. Balasubramanian P, Lang JC, Jatana KR, et al. Multiparameter analysis, including EMT markers, on negatively enriched blood samples from patients with squamous cell carcinoma of the head and neck. PLoS One 2012;7:e42048.
5. Bozec A, Ilie M, Dassonville O, et al. Significance of circulating tumor cell detection using the CellSearch system in patients with locally advanced head and neck squamous cell carcinoma. Eur Arch Otorhinolaryngol 2013;270:2745–2749.
6. Chaubal S, Wollenberg B, Kastenbauer E, Zeidler R. Ep-CAM—a marker for the detection of disseminated tumor cells in patients suffering from SCCHN. Anticancer Res 1999;19:2237–2242.
7. Hristozova T, Konschak R, Stromberger C, et al. The presence NSC 74859 of circulat- ing tumor cells (CTCs) correlates with lymph node metastasis in nonre- sectable squamous cell carcinoma of the head and neck region (SCCHN). Ann Oncol 2011;22:1878–1885.
8. Hsieh JC, Lin HC, Huang CY, et al. Prognostic value of circulating tumor cells with podoplanin expression in patients with locally advanced or metastatic head and neck squamous cell carcinoma. Head Neck 2015;37: 1448–1455.
9. Jatana KR, Balasubramanian P, Lang JC, et al. Significance of circulating tumor cells in patients with squamous cell carcinoma of the head and neck: initial results. Arch Otolaryngol Head Neck Surg 2010;136:1274– 1279.
10. Nichols AC, Lowes LE, Szeto CC, et al. Detection of circulating tumor cells in advanced head and neck cancer using the CellSearch system. Head Neck 2012;34:1440–1444.
11. Tinhofer I, Hristozova T, Stromberger C, Keilhoiz U, Budach V. Monitoring of circulating tumor cells and their expression of EGFR/phospho-EGFR during combined radiotherapy regimens in locally advanced squamous cell carcinoma of the head and neck. Int J Radiat Oncol Biol Phys 2012; 83:e685–e690.
12. Wang Z, Cui K, Xue Y, Tong F, Li S. Prognostic value of circulating tumor cells in patients with squamous cell carcinoma of the head and neck: a systematic review and meta-analysis. Med Oncol 2015;32:164.
13. Weller P, Nel I, Hassenkamp P, et al. Detection of circulating tumor cell subpopulations in patients with head and neck squamous cell carcinoma (HNSCC). PLoS One 2014;9:e113706.
14. Kim MY, Oskarsson T, Acharyya S, et al. Tumor self-seeding by circulating cancer cells. Cell 2009;139:1315–1326.
15. Abramovitch R, Marikovsky M, Meir G, Neeman M. Stimulation of tumour growth by wound-derived growth factors. Br J Cancer 1999;79:1392– 1398.
16. Baker DG, Masterson TM, Pace R, Constable WC, Wanebo H. The influ- ence of the surgical wound on local tumor recurrence. Surgery 1989;106: 525–532.
17. Bogden AE, Moreau JP, Eden PA. Proliferative response of human and animal tumours to surgical wounding of normal tissues: onset, duration and inhibition. Br J Cancer 1997;75:1021–1027.
18. Cohen HJ, Laszlo J. Influence of trauma on the unusual distribution of metastases from carcinoma of the larynx. Cancer 1972;29:466–471.
19. Hofer SOP, Shrayer D, Reichner JS, Hoekstra HJ, Wanebo HJ. Wound- induced tumor progression: a probable role in recurrence after tumor resection. Arch Surg 1998;133:383–389.
20. Lee JY, Murphy SM, Scanlon EF. Effect of trauma on implantation of met- astatic tumor in bone in mice. J Surg Oncol 1994;56:178–184.
21. Licitra L, Perrone F, Tamborini E, et al. Role of EGFR family receptors in proliferation of squamous carcinoma cells induced by wound healing flu- ids of head and neck cancer patients. Ann Oncol 2011;22:1886–1893.
22. Murthy SM, Goldschmidt RA, Rao LN, Ammirati M, Buchmann T, Scanlon E F. The influence of surgical trauma on experimental metasta- sis. Cancer 1989;64:2035–2044.
23. Rho O, Kim DJ, Kiguchi K, Digiovanni J. Growth factor signaling path- ways as targets for prevention of epithelial carcinogenesis. Mol Carcinog 2011;50:264–279.
24. Roh JL, Sung MW, Kim KH. Suppression of accelerated tumor growth in surgical wounds by celecoxib and indomethacin. Head Neck 2005;27: 326–332.
25. Tagliabue E, Agresti R, Carcangiu ML, et al. Role of HER2 in wound- induced breast carcinoma proliferation. Lancet 2003;362:527–533.
26. Ekblad L, Lindgren G, Persson E, Kjellen E, Wennerberg J. Cell-line-spe- cific stimulation of tumor cell aggressiveness by wound healing fac- tors—a central role for STAT3. BMC Cancer 2013;13:33.
27. Henriksson E, Baldetorp B, Borg A˚ , et al. p53 mutation and cyclin D1 amplification correlate with cisplatin sensitivity in xenografted human squamous cell carcinomas from head and neck. Acta Oncol 2006;45:300– 305.
28. World Medical Association, Inc. Declaration of Helsinki—ethical principles for medical research involving human subjects. Available at: http://www. wma.net/en/30publications/10policies/b3/index.html. Accessed September 13, 2012.
29. Schmitt C, Kuhn B, Zhang X, Kivitz AJ, Grange S. Disease-drug-drug interaction involving tocilizumab and simvastatin in patients with rheu- matoid arthritis. Clin Pharmacol Ther 2011;89:735–740.
30. Krishnamurthy S, Warner KA, Dong Z, et al. Endothelial interleukin-6 defines the tumorigenic potential of primary human cancer stem cells. Stem Cells 2014;32:2845–2857.