Strahlenther Onkol DOI 10.1007/s00066-014-0797-8
O r i g i n a l A rt i c l e
Expression of transforming growth factor beta 1-related signaling proteins in irradiated vessels Raimund H. M. Preidl · Patrick Möbius · Manuel Weber · Kerstin Amann · Friedrich W. Neukam · Andreas Schlegel · Falk Wehrhan
Received: 21 August 2014 / Accepted: 14 November 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Aim Microvascular free tissue transfer is a standard method in head and neck reconstructive surgery. However, previous radiotherapy of the operative region is associated with an increased incidence in postoperative flap-related complications and complete flap loss. As transforming growth factor beta (TGF-β) 1 and galectin-3 are well known markers in the context of fibrosis and lectin-like oxidized low-density lipoprotein 1 (LOX-1) supports vascular atherosclerosis, the aim of this study was to evaluate the expression of TGF-β1 and related markers as well as LOX-1 in irradiated vessels. Materials and methods To evaluate the expression of galectin-3, Smad 2/3, TGF-β1, and LOX-1, 20 irradiated and 20 nonirradiated arterial vessels were used for immunohistochemical staining. We semiquantitatively assessed the ratio of stained cells/total number of cells (labeling index). Results Expression of galectin-3, Smad 2/3, and TGF-β1 was significantly increased in previously irradiated vessels
Raimund H.M. Preidl and Patrick Möbius shared first authorship. R. H. M. Preidl, M.D. () · P. Möbius, D.M.D. · M. Weber, M.D. · F. W. Neukam, M.D., D.M.D., Ph.D. · A. Schlegel, M.D., D.M.D. · F. Wehrhan, M.D., D.M.D. Department of Oral and Maxillofacial Surgery, University of Erlangen- Nürnberg, Glückstraße 11,91054 Erlangen, Germany e-mail: [email protected] M. Weber, M.D. · P. Möbius, D.M.D. · R. H. M. Preidl, M.D. · K. Amann, M.D., Ph.D. · F. W. Neukam, M.D., D.M.D., Ph.D. · A. Schlegel, M.D., D.M.D. · F. Wehrhan, M.D., D.M.D. University of Erlangen- Nürnberg, Erlangen, Germany
compared with nonirradiated controls. Furthermore, LOX-1 was expressed significantly higher in irradiated compared with nonirradiated vessels. Conclusion Fibrosis-related proteins like galectin-3, Smad 2/3, and TGF-β1 are upregulated after radiotherapy and support histopathological changes leading to vasculopathy of the irradiated vessels. Furthermore, postoperative complications in irradiated patients can be explained by increased endothelial dysfunction caused by LOX-1 in previously irradiated patients. Consequently, not only TGF-β1 but also galectin-3inhibitors may decrease complications after microsurgical tissue transfer. Keywords Radiotherapy · Vessel · Radiation-induced vasculopathy · Transforming growth factor beta · Head and neck surgery Expression TGF-β1-bezogener Signalproteine in bestrahlten Gefäßen Zusammenfassung Einführung Der freie mikrovaskuläre Gewebetransfer gilt heute als fester Standard in der rekonstruktiven Kopf-HalsChirurgie. Es zeigte sich jedoch, dass im Falle einer stattgehabten Bestrahlung im Operationsgebiet mit einer erhöhten Rate an transplantatbezogenen Komplikationen gerechnet werden muss. Sowohl TGF-β1 als auch Galektin-3 sind bekannte Mediatoren in Bezug auf die Fibroseentstehung, wohingegen LOX-1 eine unterstützende Rolle bei der Entwicklung von Atherosklerose einnimmt. Das Ziel dieser Studie war es, zu untersuchen, wie sich die Expression von TGF-β1 und verwandten Fibrosemediatoren sowie LOX-1 im bestrahlten Gefäß verhält.
13
2
Material und Methoden 20 bestrahlte und 20 nichtbestrahlte Arterien wurden mittels immunhistochemischer Färbungen auf die Expression von Galektin-3, Smad2/3, TGF-β1 sowie LOX-1 hin untersucht. Semiquantitativ wurde das Verhältnis gefärbte Zellen pro Gesamtzellzahl (Färbungsindex) bestimmt. Ergebnisse Die Expression von Galektin-3, Smad2/3 sowie TGF-β1 war in bestrahlten Arterien signifikant höher als in nichtbestrahlten Gefäßen. LOX-1 wurde in vorbestrahlten Gefäßen ebenso signifikant höher exprimiert. Konklusion Mediatoren der Fibroseentstehung wie Galektin-3, Smad2/3 und TGF-β1 sind in bestrahlten Gefäßen erhöht exprimiert und tragen mutmaßlich zu der Entstehung der radiogen induzierten Vaskulopathie bei. Des Weiteren können postoperative Komplikationen bei bestrahlen Patienten durch die verstärkte endotheliale Dysfunktion, unterstützt durch LOX-1, erklärt werden. Folglich könnten nicht nur TGF-β1- sondern auch Galektin-3-Inhibitoren einen minimierenden Effekt auf die Komplikationen nach mikrovaskulärem Gewebetransfer haben. Schlüsselwörter Radiotherapie · Gefäße · Radiogeninduzierte Vaskulopathie · Transforming growth factor β · Kopf-Hals-Chirurgie Introduction Microvascular free tissue transfer in head and neck surgery is an established reconstructive method and is frequently applied in order to restore functionality considering speech and swallowing as well as achieving esthetic rehabilitation [16]. However, there are still numerous cases of postoperative complications regarding infections, thrombosis, and even complete flap loss especially in patients who previously received radiotherapy in the head and neck area [3, 15]. Previous radiotherapy in the head and neck region is linked to an increased risk for neurovascular events associated with an accelerated atherosclerotic plaque formation, a decrease in blood flow, and subsequent arterial wall thickening [25]. We already know that intima dehiscence as well as increased proteoglycan contents and elevated levels of inflammatory cells with concomitant activation of nuclear factor kappa B can be observed in previously irradiated vessels [5, 19, 21]. This leads to an increase in intima-media thickness of irradiated arteries with elevated contents of collagens, comparable to atherosclerotic disease, resulting even in luminal occlusion [4]. Galectin-3 supports the attraction of inflammatory cells in the vessel wall by inducing endothelial adhesion molecules [13]. Furthermore, TGF-β1 secretion is supported by galectin-3, a well-known mediator in the context of organ
13
R. H. M. Preidl et al.
fibrosis. After radiotherapy an increase in TGF-β1 caused by macrophages and fibroblasts can be observed in the surrounding soft tissue, thus playing an important role in the development of local fibrosis [22]. In the context of cardiovascular pathological changes, TGF-β1 is known to stimulate fibroblasts and vascular smooth muscle cells resulting in atherosclerosis and vessel stenosis as wells as hypertension [18]. Regarding the TGF-β receptor 1 pathway, Smad2 and Smad3 are phosphorylated by activated TGF-β1 receptors and influence the transcription of fibronectin, procollagens, as well as plasminogen activator-inhibitor 1 (PAI-1) in the vessel wall [18]. Therefore, Smad2 and Smad3 predominantly transmit TGF-β signals throughout the cytoplasm into the nucleus. Lectin-like oxidized low-density lipoprotein 1 (LOX-1) plays an essential role in the pathogenesis of atherosclerotic lesions by inducing endothelial adhesion molecules and attracting inflammatory cells [20]. Being almost undetectable under physiological conditions, it is upregulated in endothelial cells and smooth muscle cells by proinflammatory cytokines leading to vascular cell dysfunction and advanced plaque formation [11]. The aim of this study was to determine whether the pathogenesis of radiation-induced vasculopathy is mediated by the TGF-β1 pathway or if galectin-3 and LOX-1 independently support the histomorphological changes in previously irradiated human arteries compared with nonirradiated controls. Materials and methods Patients and tissue harvesting In all, 20 irradiated and 20 nonirradiated arterial vessels from arterial microvascular anastomoses of patients treated in the Department of Oral and Maxillofacial Surgery at the University of Erlangen-Nürnberg, Germany, were selected for this study. Written and oral informed consent was obtained from all patients prior to enrolment. The study protocol was approved by the ethics committee of the University of Erlangen-Nuremberg, Germany (Ref.- No. 83_13B). Each specimen was confirmed to show representative regions of the vessel (intima, media, and adventitia). The radiation dose used during radiotherapy was 50–70 Gy applied to the lower head and neck area where the investigated vessels were situated. The average timespan between end of irradiation and vessel removal was 57.01 ± 8.98 months. Immunohistochemical staining The formalin-fixed, paraffin-embedded tissue samples were sliced in sections of 2-μm thickness by using a rota-
Expression of transforming growth factor beta 1-related signaling proteins in irradiated vessels
3
tion microtome (Leica, Nussloch, Germany). The sections were dewaxed in xylole and then rehydrated in graded alcohol. Staining was performed by applying the APAAP (alkaline phosphatase–anti-alkaline phosphatase) method and an automated staining device (Autostainer plus, Dako Cytomation, Hamburg, Germany). Antigens were detected by incubating tissues in the autostainer (21 °C, 30 min) with specific antibodies. The following primary antibodies were used: anti-galectin-3 (sc-20157, Santa Cruz Biotech., Santa Cruz, Calif.) 1:150, anti-Smad 2/3 (sc-6033, Santa Cruz Biotech.) 1:50, anti-TGF-β1 (sc-146, Santa Cruz Biotech.) 1:100, and anti-LOX-1 (ab85839, abcam. Cambridge, UK) 1:50. Biotinylated immunoglobulins were used for all samples. The samples were incubated in hematoxylin (Dako Cytomation) for nucleus-counterstaining. Negative and positive controls were performed for each immunohistochemical staining. Qualitative and semiquantitative immunohistochemical analysis The samples used in this study were scanned and digitalized completely by using the method of “whole slide imaging.” All scanned samples were virtually microscoped on PC (Pannoramic MIRAX viewer, Zeiss, Jena, Germany). Quality controls were performed with a bright-field microscope (Zeiss Axioskop and Axiocam 5, magnification: ×5–400). As the vessels revealed a homogeneous expression of antigens in the immunohistochemical stainings, three representative visual fields per section at a magnification of ×200 were selected from each sample. The visual fields were imported into Biomas (MSAB, Erlangen, Germany) to perform cell counting by three independent observers. The number of positive-stained cells for galectin-3, Smad 2/3, TGF-β1, and LOX-1 in the intima, media, and adventitia was evaluated for each sample. Statistical analysis To quantify the immunohistochemical stainings, the labeling index (ratio of positive cells to total cell number resembling the expression rate) per visual field, sample, and study group was determined. The results are expressed as median, the interquartile range (IQR), standard deviation (SD), and range. Box plot diagrams illustrate the median, the interquartile range, minimum (Min), and maximum (Max). Two-sided, adjusted p values of ≤ 0.05 were considered statistically significant. All analyses were done with SPSS 21.0 for Mac (IBM Inc., New York, N.Y.).
Fig. 1 Immunohistochemical stainings for galectin-3, Smad2/3, TGFβ1, and LOX-1 in irradiated (right column) and nonirradiated (left column) arterial vessels
Results Qualitative analysis Galectin-3 expression in irradiated vessels was predominantly detected in the nucleus and only rarely in the cytoplasm throughout the intima, media, and adventitia. Smad2/3 was detected mainly in the media with only a few stained cells in the intima and almost none in the adventitia in both groups. Furthermore, stained cells for Smad2/3 showed a predominant expression in the nucleus rather than in the cytoplasm. By contrast, nonirradiated vessels revealed a cytoplasmic expression of galectin-3 with only a few stained nuclei. TGF-β1 was mostly present in the intima and media and not in the adventitia. Compared with galectin-3, TGF-β1 staining within the cells was localized predominantly in the cytoplasm. By contrast, LOX-1 could be detected in the cytoplasm of cells localized in the intima and media in irradiated as well as nonirradiated vessels (Fig. 1). Quantitative analysis Immunohistochemical stainings revealed a significantly increased labeling index for galectin-3 in irradiated vessels compared with nonirradiated controls [labeling index galectin-3: 3226 ± 20.01 (irradiated group) vs. 10.14 ± 8.39 (nonirradiated group); p
O r i g i n a l A rt i c l e
Expression of transforming growth factor beta 1-related signaling proteins in irradiated vessels Raimund H. M. Preidl · Patrick Möbius · Manuel Weber · Kerstin Amann · Friedrich W. Neukam · Andreas Schlegel · Falk Wehrhan
Received: 21 August 2014 / Accepted: 14 November 2014 © Springer-Verlag Berlin Heidelberg 2014
Abstract Aim Microvascular free tissue transfer is a standard method in head and neck reconstructive surgery. However, previous radiotherapy of the operative region is associated with an increased incidence in postoperative flap-related complications and complete flap loss. As transforming growth factor beta (TGF-β) 1 and galectin-3 are well known markers in the context of fibrosis and lectin-like oxidized low-density lipoprotein 1 (LOX-1) supports vascular atherosclerosis, the aim of this study was to evaluate the expression of TGF-β1 and related markers as well as LOX-1 in irradiated vessels. Materials and methods To evaluate the expression of galectin-3, Smad 2/3, TGF-β1, and LOX-1, 20 irradiated and 20 nonirradiated arterial vessels were used for immunohistochemical staining. We semiquantitatively assessed the ratio of stained cells/total number of cells (labeling index). Results Expression of galectin-3, Smad 2/3, and TGF-β1 was significantly increased in previously irradiated vessels
Raimund H.M. Preidl and Patrick Möbius shared first authorship. R. H. M. Preidl, M.D. () · P. Möbius, D.M.D. · M. Weber, M.D. · F. W. Neukam, M.D., D.M.D., Ph.D. · A. Schlegel, M.D., D.M.D. · F. Wehrhan, M.D., D.M.D. Department of Oral and Maxillofacial Surgery, University of Erlangen- Nürnberg, Glückstraße 11,91054 Erlangen, Germany e-mail: [email protected] M. Weber, M.D. · P. Möbius, D.M.D. · R. H. M. Preidl, M.D. · K. Amann, M.D., Ph.D. · F. W. Neukam, M.D., D.M.D., Ph.D. · A. Schlegel, M.D., D.M.D. · F. Wehrhan, M.D., D.M.D. University of Erlangen- Nürnberg, Erlangen, Germany
compared with nonirradiated controls. Furthermore, LOX-1 was expressed significantly higher in irradiated compared with nonirradiated vessels. Conclusion Fibrosis-related proteins like galectin-3, Smad 2/3, and TGF-β1 are upregulated after radiotherapy and support histopathological changes leading to vasculopathy of the irradiated vessels. Furthermore, postoperative complications in irradiated patients can be explained by increased endothelial dysfunction caused by LOX-1 in previously irradiated patients. Consequently, not only TGF-β1 but also galectin-3inhibitors may decrease complications after microsurgical tissue transfer. Keywords Radiotherapy · Vessel · Radiation-induced vasculopathy · Transforming growth factor beta · Head and neck surgery Expression TGF-β1-bezogener Signalproteine in bestrahlten Gefäßen Zusammenfassung Einführung Der freie mikrovaskuläre Gewebetransfer gilt heute als fester Standard in der rekonstruktiven Kopf-HalsChirurgie. Es zeigte sich jedoch, dass im Falle einer stattgehabten Bestrahlung im Operationsgebiet mit einer erhöhten Rate an transplantatbezogenen Komplikationen gerechnet werden muss. Sowohl TGF-β1 als auch Galektin-3 sind bekannte Mediatoren in Bezug auf die Fibroseentstehung, wohingegen LOX-1 eine unterstützende Rolle bei der Entwicklung von Atherosklerose einnimmt. Das Ziel dieser Studie war es, zu untersuchen, wie sich die Expression von TGF-β1 und verwandten Fibrosemediatoren sowie LOX-1 im bestrahlten Gefäß verhält.
13
2
Material und Methoden 20 bestrahlte und 20 nichtbestrahlte Arterien wurden mittels immunhistochemischer Färbungen auf die Expression von Galektin-3, Smad2/3, TGF-β1 sowie LOX-1 hin untersucht. Semiquantitativ wurde das Verhältnis gefärbte Zellen pro Gesamtzellzahl (Färbungsindex) bestimmt. Ergebnisse Die Expression von Galektin-3, Smad2/3 sowie TGF-β1 war in bestrahlten Arterien signifikant höher als in nichtbestrahlten Gefäßen. LOX-1 wurde in vorbestrahlten Gefäßen ebenso signifikant höher exprimiert. Konklusion Mediatoren der Fibroseentstehung wie Galektin-3, Smad2/3 und TGF-β1 sind in bestrahlten Gefäßen erhöht exprimiert und tragen mutmaßlich zu der Entstehung der radiogen induzierten Vaskulopathie bei. Des Weiteren können postoperative Komplikationen bei bestrahlen Patienten durch die verstärkte endotheliale Dysfunktion, unterstützt durch LOX-1, erklärt werden. Folglich könnten nicht nur TGF-β1- sondern auch Galektin-3-Inhibitoren einen minimierenden Effekt auf die Komplikationen nach mikrovaskulärem Gewebetransfer haben. Schlüsselwörter Radiotherapie · Gefäße · Radiogeninduzierte Vaskulopathie · Transforming growth factor β · Kopf-Hals-Chirurgie Introduction Microvascular free tissue transfer in head and neck surgery is an established reconstructive method and is frequently applied in order to restore functionality considering speech and swallowing as well as achieving esthetic rehabilitation [16]. However, there are still numerous cases of postoperative complications regarding infections, thrombosis, and even complete flap loss especially in patients who previously received radiotherapy in the head and neck area [3, 15]. Previous radiotherapy in the head and neck region is linked to an increased risk for neurovascular events associated with an accelerated atherosclerotic plaque formation, a decrease in blood flow, and subsequent arterial wall thickening [25]. We already know that intima dehiscence as well as increased proteoglycan contents and elevated levels of inflammatory cells with concomitant activation of nuclear factor kappa B can be observed in previously irradiated vessels [5, 19, 21]. This leads to an increase in intima-media thickness of irradiated arteries with elevated contents of collagens, comparable to atherosclerotic disease, resulting even in luminal occlusion [4]. Galectin-3 supports the attraction of inflammatory cells in the vessel wall by inducing endothelial adhesion molecules [13]. Furthermore, TGF-β1 secretion is supported by galectin-3, a well-known mediator in the context of organ
13
R. H. M. Preidl et al.
fibrosis. After radiotherapy an increase in TGF-β1 caused by macrophages and fibroblasts can be observed in the surrounding soft tissue, thus playing an important role in the development of local fibrosis [22]. In the context of cardiovascular pathological changes, TGF-β1 is known to stimulate fibroblasts and vascular smooth muscle cells resulting in atherosclerosis and vessel stenosis as wells as hypertension [18]. Regarding the TGF-β receptor 1 pathway, Smad2 and Smad3 are phosphorylated by activated TGF-β1 receptors and influence the transcription of fibronectin, procollagens, as well as plasminogen activator-inhibitor 1 (PAI-1) in the vessel wall [18]. Therefore, Smad2 and Smad3 predominantly transmit TGF-β signals throughout the cytoplasm into the nucleus. Lectin-like oxidized low-density lipoprotein 1 (LOX-1) plays an essential role in the pathogenesis of atherosclerotic lesions by inducing endothelial adhesion molecules and attracting inflammatory cells [20]. Being almost undetectable under physiological conditions, it is upregulated in endothelial cells and smooth muscle cells by proinflammatory cytokines leading to vascular cell dysfunction and advanced plaque formation [11]. The aim of this study was to determine whether the pathogenesis of radiation-induced vasculopathy is mediated by the TGF-β1 pathway or if galectin-3 and LOX-1 independently support the histomorphological changes in previously irradiated human arteries compared with nonirradiated controls. Materials and methods Patients and tissue harvesting In all, 20 irradiated and 20 nonirradiated arterial vessels from arterial microvascular anastomoses of patients treated in the Department of Oral and Maxillofacial Surgery at the University of Erlangen-Nürnberg, Germany, were selected for this study. Written and oral informed consent was obtained from all patients prior to enrolment. The study protocol was approved by the ethics committee of the University of Erlangen-Nuremberg, Germany (Ref.- No. 83_13B). Each specimen was confirmed to show representative regions of the vessel (intima, media, and adventitia). The radiation dose used during radiotherapy was 50–70 Gy applied to the lower head and neck area where the investigated vessels were situated. The average timespan between end of irradiation and vessel removal was 57.01 ± 8.98 months. Immunohistochemical staining The formalin-fixed, paraffin-embedded tissue samples were sliced in sections of 2-μm thickness by using a rota-
Expression of transforming growth factor beta 1-related signaling proteins in irradiated vessels
3
tion microtome (Leica, Nussloch, Germany). The sections were dewaxed in xylole and then rehydrated in graded alcohol. Staining was performed by applying the APAAP (alkaline phosphatase–anti-alkaline phosphatase) method and an automated staining device (Autostainer plus, Dako Cytomation, Hamburg, Germany). Antigens were detected by incubating tissues in the autostainer (21 °C, 30 min) with specific antibodies. The following primary antibodies were used: anti-galectin-3 (sc-20157, Santa Cruz Biotech., Santa Cruz, Calif.) 1:150, anti-Smad 2/3 (sc-6033, Santa Cruz Biotech.) 1:50, anti-TGF-β1 (sc-146, Santa Cruz Biotech.) 1:100, and anti-LOX-1 (ab85839, abcam. Cambridge, UK) 1:50. Biotinylated immunoglobulins were used for all samples. The samples were incubated in hematoxylin (Dako Cytomation) for nucleus-counterstaining. Negative and positive controls were performed for each immunohistochemical staining. Qualitative and semiquantitative immunohistochemical analysis The samples used in this study were scanned and digitalized completely by using the method of “whole slide imaging.” All scanned samples were virtually microscoped on PC (Pannoramic MIRAX viewer, Zeiss, Jena, Germany). Quality controls were performed with a bright-field microscope (Zeiss Axioskop and Axiocam 5, magnification: ×5–400). As the vessels revealed a homogeneous expression of antigens in the immunohistochemical stainings, three representative visual fields per section at a magnification of ×200 were selected from each sample. The visual fields were imported into Biomas (MSAB, Erlangen, Germany) to perform cell counting by three independent observers. The number of positive-stained cells for galectin-3, Smad 2/3, TGF-β1, and LOX-1 in the intima, media, and adventitia was evaluated for each sample. Statistical analysis To quantify the immunohistochemical stainings, the labeling index (ratio of positive cells to total cell number resembling the expression rate) per visual field, sample, and study group was determined. The results are expressed as median, the interquartile range (IQR), standard deviation (SD), and range. Box plot diagrams illustrate the median, the interquartile range, minimum (Min), and maximum (Max). Two-sided, adjusted p values of ≤ 0.05 were considered statistically significant. All analyses were done with SPSS 21.0 for Mac (IBM Inc., New York, N.Y.).
Fig. 1 Immunohistochemical stainings for galectin-3, Smad2/3, TGFβ1, and LOX-1 in irradiated (right column) and nonirradiated (left column) arterial vessels
Results Qualitative analysis Galectin-3 expression in irradiated vessels was predominantly detected in the nucleus and only rarely in the cytoplasm throughout the intima, media, and adventitia. Smad2/3 was detected mainly in the media with only a few stained cells in the intima and almost none in the adventitia in both groups. Furthermore, stained cells for Smad2/3 showed a predominant expression in the nucleus rather than in the cytoplasm. By contrast, nonirradiated vessels revealed a cytoplasmic expression of galectin-3 with only a few stained nuclei. TGF-β1 was mostly present in the intima and media and not in the adventitia. Compared with galectin-3, TGF-β1 staining within the cells was localized predominantly in the cytoplasm. By contrast, LOX-1 could be detected in the cytoplasm of cells localized in the intima and media in irradiated as well as nonirradiated vessels (Fig. 1). Quantitative analysis Immunohistochemical stainings revealed a significantly increased labeling index for galectin-3 in irradiated vessels compared with nonirradiated controls [labeling index galectin-3: 3226 ± 20.01 (irradiated group) vs. 10.14 ± 8.39 (nonirradiated group); p
