- Email: [email protected]
Increased apoptosis and secretion of tryptase by mast cells in infantile haemangioma treated with propranolol
Pathology (October 2014) 46(6), pp. 496–500
ANATOMICAL PATHOLOGY
Increased apoptosis and secretion of tryptase by mast cells in infantile haemangioma treated with propranolol RYAN STEEL
AND
DARREN DAY
School of Biological Sciences, Victoria University of Wellington, Wellington, New Zealand
Summary Propranolol is increasingly used to treat problematic infantile haemangioma (IH), although its molecular mechanisms remain unclear. A key feature of propranolol therapy is the decreased deposition of fibrofatty residuum compared with spontaneously involuting IH. This study investigated the molecular consequences of propranolol treatment for IH in vivo. Immunohistochemical and terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) staining was performed on five age matched patients with proliferative IH. Two patients (A and B) were undergoing propranolol treatment at the time of surgical resection. Propranolol treatment increased apoptosis, and induced mast cells to degranulate and secrete tryptase into the interstitium. The microvessels of patient A were immature [weak von Willibrand Factor (vWF), and strong osteoprotegerin (OPG) staining], comparable to untreated proliferative IH, while those of patient B were mature (strong vWF staining, and no OPG staining). The perivascular CD90þ mesenchymal stem cell population was preserved in both propranolol treated patients. Using rarely obtained biopsies from IH patients treated with propranolol, we show increased apoptosis by propranolol for the first time in vivo. We also suggest that mast cells, through secreted proteases, may contribute to the decreased fibrofatty residuum seen with propranolol treatment. Key words: Apoptosis, endothelial cells, immunohistochemistry, infantile haemangioma, mast cells, mesenchymal stem cell, microvessel maturity, osteoprotegerin (OPG), propranolol, tryptase. Received 8 December 2013, revised 29 April, accepted 1 May 2014
Recently the non-specific b-blocker propranolol has emerged as an effective and non-invasive therapy eliciting a rapid regression of the lesion with a significantly reduced fibrofatty residuum.5,7,8 Despite its therapeutic success, the underlying molecular mechanisms of propranolol action on IH are not fully understood. Proposed mechanisms include pro-apoptotic, antiangiogenic and vasoconstrictive effects,9 as well as modulation of the renin angiotensin system (RAS).10 The elucidation of propranolol’s mechanism remains difficult because of the scarcity of propranolol treated IH tissue for analysis. Additional actions of propranolol treatment may be possible. Acceleration of EC maturation normally seen during IH involution,1,4 in which osteoprotegerin (OPG) is lost from and strong von Willibrand Factor (vWF) staining is gained by ECs,11 represents a possible mechanism of action. A significant benefit of propranolol treatment is the reduced fibrofatty residuum. This residuum is possibly derived from the perivascular mesenchymal stem cell (MSC) population,12–14 the regulation of which has been proposed as a therapeutic target for IH.8 Whether this population is reduced by propranolol treatment is unknown. Additionally, mast cells synthesise and secrete a range of bioactive mediators, including proteases that degrade the extracellular matrix (ECM). Their abundance increases during involution, suggesting a role for mast cells in the resolution of this lesion.15 Whether mast cells play a role in the action of propranolol therapy is unexplored. Increased apoptosis, accelerated microvessel maturation and degradation of the ECM by mast cells may be important components of the mechanism of propranolol action. Here we examine these molecular components in biopsies from two propranolol treated patients whom later also underwent surgical resection.
INTRODUCTION Infantile haemangioma (IH) is a common microvascular tumour of infancy. IH is most common in Caucasian, female, and premature infants, affecting 5–10% of the population. Three phases of growth characterise the progression of IH, lasting up to 12 years of age. In proliferative IH (up to 12 months of age), pericytes and plump endothelial cells (ECs) with scattered mitotic figures form small tightly packed capillaries. These capillaries are immunoreactive for EC markers CD31 and CD34, as well as the definitive IH marker GLUT1. As these lesions involute, there is a thickening and lamination of the capillary basement membrane, a flattening of the ECs, and increased deposition of fibrous stroma between the vessels. Large vessels separated by a sparse fibrofatty residuum characterise the fully involuted lesion.1–4 In cases of IH involving significant morbidity, corticosteroids or surgery have been the preferred treatment options.5,6 Print ISSN 0031-3025/Online ISSN 1465-3931 DOI: 10.1097/PAT.0000000000000143
#
MATERIALS AND METHODS Patients Tissue collection and processing were approved by the Wellington Regional Ethics Committee, and signed consent for the use of tissue was obtained from the patients’ next-of-kin. Solid biopsies were collected from the centre of the lesions, often including subcutaneous tissue, of five patients with proliferative IH aged between 4 and 8 months (mean 6 months) at the time of surgical resection. Biopsies were immediately fixed in 10% formalin overnight then paraffin embedded using standard protocols for long-term storage at room temperature. Samples stored in paraffin are stable for several years, although typically tissue used was less than 1 year old. Of these five patients, two (A and B) were treated with propranolol (mean 5 months at age of biopsy) while the remaining three (mean 6 months) received no treatment. The diagnosis of IH for all patients was made by a trained pathologist based on the histopathology of the lesions, as well as expression of the definitive IH marker, GLUT1.4
2014 Royal College of Pathologists of Australasia
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
STUDY OF PROPRANOLOL ACTION ON IH IN VIVO
Patient A Patient A presented with an ulcerating IH of the right earlobe at 2.5 months of age and treatment with 1.5 mg/kg/day propranolol was started. The patient experienced on-going ulceration and pain despite slow regression of the haemangioma, so the lesion was debulked at 4 months of age. Following surgery, propranolol treatment was stopped, during which time the haemangioma increased in size. Propranolol treatment was reinstated 3 weeks after surgery at 2 mg/kg/day, which resulted in continued regression of the lesion. Patient B Patient B presented with an ulcerating haemangioma on the buttock at 4 months of age and treatment with 1.5 mg/kg/day propranolol was started. Slower than expected regression and ongoing ulceration was observed. At 6 months of age a biopsy was taken, but the majority of the haemangioma was left in situ. The patient remained on propranolol treatment and the lesion continued to gradually regress. Immunohistochemistry Immunohistochemistry was performed on 4 mm paraffin sections as previously described,11,13,16,17 using primary antibodies raised in either rabbit or mouse against CD34 (1:300; Invitrogen, USA), CD31 (1:300; Abcam, UK), OPG (1:100; R&D Systems, USA), tryptase (1:500; Abcam), von Willibrand Factor (vWF; 1:200; Abcam), CD90 (1:200; Abcam) and smooth muscle actin (SMA; 1:200; Abcam). Species-specific secondary antibodies conjugated to either Alexa Flour 488 or 555 (1:500; Invitrogen) were used to visualise epitope bound primary antibodies. Slides were mounted in ProLong Gold Antifade containing 4’, 6-diamidino-2-phenylindole (DAPI; Invitrogen) to visualise cell nuclei. Apoptosis detection Apoptotic nuclei were detected using an In Situ Cell Death Detection Kit (Roche, Switzerland) following the manufacturer’s instructions. This kit is based on the terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) technique, in which fluorescein-labelled nucleotides are attached at sites of single and double stranded DNA breaks typical of apoptotic cells. The TUNEL reaction was performed after the washing of primary antibodies, and before secondary antibody incubation. Image acquisition and statistical analysis Quantitative analyses were performed on the frequency of TUNEL stained nuclei, and on the maturity of the microvessels within IH on the basis of OPG and vWF staining.11 Multiple fields of view (typically 5 fields for OPG/vWF stained sections, and 25 fields for TUNEL stained sections) were obtained for these analyses to ensure that at least 100–200 events were counted per patient. The number of apoptotic nuclei per field of view was expressed as the percent of all nuclei, whereas the number of immature OPG positive vessels was expressed as the percent of all vessels with an identifiable lumen. All images were captured using an Olympus FV1000 confocal laser-scanning microscope fitted with a krypton/argon laser (Olympus, Japan), and a Student’s t-test was used to compare group means.
497
histologically proliferative containing microvessels that were immunoreactive for the definitive IH marker, GLUT1 (Fig. 1).4 Thus, our analyses report effects of propranolol at the earliest stages of regression. Apoptosis Enhanced apoptosis has been suggested as a possible mechanism for accelerated involution of IH by propranolol.9 Figure 2 shows TUNEL-labelled (green) apoptotic nuclei (counterstained with DAPI, blue) in both propranolol treated and untreated IH. Figure 2A shows a low level of apoptotic nuclei (yellow arrowheads) in untreated patients, while Fig. 2B shows many more TUNEL labelled nuclei in propranolol treated patients. The percentage of TUNEL reactive nuclei in both propranolol treated patients was significantly greater than that observed for untreated patients (Fig. 2C, p < 0.05). Microvessel maturation We have previously shown that OPG expression by immature microvessels is replaced by vWF expression as they mature.11 Strong OPG expression was seen in 89% of the microvessels of untreated patients (green), and these vessels expressed only low levels of vWF (red, Fig. 3A). In propranolol treated tissue, patient A was indistinguishable from the control group (data not shown). In patient B, however, very few of the microvessels were OPG positive and most vessels were also strongly vWF immunoreactive (Fig. 3B). Figure 3C graphically shows the fraction of vessels that were OPG positive for the untreated IH (n ¼ 3), as well as each of the propranolol treated patients (patient A and patient B). Tryptase secretion by mast cells The presence of tryptase positive mast cells in IH has been previously reported.18 Figure 4 identifies mast cells by way of tryptase immunoreactivity (green staining) in untreated (Fig. 4A) and a propranolol treated IH (Fig. 4B). The endothelium is identified by staining for CD31 (red), while nuclei are stained with DAPI (blue). Figure 4B shows that in propranolol treated tissue the mast cells have degranulated and secreted copious amounts of tryptase into the interstitium, while very little interstitial tryptase immunoreactivity was detected in any of the untreated IH patients (Fig. 4A). Increased interstitial tryptase was detected in both propranolol treated patients A and B.
RESULTS Effective therapy had been implemented in both patients treated with propranolol, although the rate of regression was slower than expected. The lesions of both patients were
A
DISCUSSION Propranolol has become the preferred treatment for IH with significant morbidity. Patients treated with propranolol
B
Fig. 1 Confirmation that propranolol treated tissues represent proliferative IH. Histological analysis by H&E and DAB staining showed tightly packed endothelial cells and pericytes forming microvessels expressing GLUT1 (appears brown) in both propranolol treated patient A (A) and patient B (B), confirming the diagnosis of IH. Scale bar ¼ 100 mm.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
498
Pathology (2014), 46(6), October
STEEL and DAY
A
B 3
No treatment
% TUNEL+ Cells
*
C
Propranolol treatment
2
(12,129)
1
(11,320)
0
Fig. 2 Propranolol treatment increased apoptosis in proliferative IH. Apoptotic nuclei identified by TUNEL staining (green, indicated by yellow arrowheads) were observed in both untreated (A) and propranolol treated (B) tissue. (C) Both of the propranolol treated patients (n ¼ 2) had a greater incidence of apoptosis than age matched, untreated controls (n ¼ 3). Cell nuclei are stained with DAPI and appear blue, while CD31þ endothelial cells stain red. Numbers in parentheses indicate the average number of cell nuclei scored for each patient. *p < 0.05 by Student’s t-test. Scale bar ¼ 50 mm.
OPG vWF
A
OPG vWF
B
C
% Vessels expressing OPG
100
Control 80 60
Propranolol treated: Patient A Propranolol treated: Patient B
40 20 0
Fig. 3 Propranolol treatment may induce a more mature microvessel phenotype in proliferative IH. (A) Immature microvessels that stain strongly for OPG (green) and weakly for vWF (red) predominated in proliferative IH. In propranolol treated tissue, patient A had an almost exclusively immature microvessel phenotype indistinguishable from age matched controls. (B) Patient B, however, had an almost exclusively mature vessel phenotype that stained strongly for vWF but not OPG. (C) Graphical depiction of the proportion of OPG positive vessels in untreated control (n ¼ 3), as well as each of the propranolol treated patients. Cell nuclei are stained with DAPI and appear blue. Scale bar ¼ 50 mm.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
STUDY OF PROPRANOLOL ACTION ON IH IN VIVO
Tryptase CD31
A
499
Tryptase CD31
B
Fig. 4 Increased tryptase secretion from mast cells in propranolol treated IH. Tryptase immunoreactivity is shown in green, and the endothelium of microvessels is identified by CD31 (red) in both panels. Staining for tryptase is tightly localised to cell nuclei (stained with DAPI, blue) in the representative untreated proliferative IH (A), but is diffusely dispersed in the vicinity of mast cells in the interstitium of propranolol treated tissue, as well as being present on the mast cells themselves (B). Scale bar ¼ 50 mm.
typically respond rapidly with a decolourisation and pronounced regression of the lesion within 2 weeks. Continued regression often completely resolves the lesion with minimal fibrofatty residuum.7,9,19 As such, propranolol treated patients rarely require surgery. We obtained proliferative IH tissue from two patients in whom regression of the lesion with propranolol treatment was evident, although slower than expected. This study thus represents a unique immunohistochemical investigation of the cellular consequences of propranolol treatment of patients with proliferative IH. We investigated whether propranolol increased apoptosis and vessel maturation in IH, and looked for potential explanations for the decreased fibrofatty residuum. Apoptosis in untreated patients was observed at levels comparable to previous reports,20 while both propranolol treated patients in this study had a significantly higher level of apoptosis as indicated by TUNEL staining. Increased apoptosis has previously been proposed as a molecular mechanism of propranolol treatment.9 Indeed, recent studies with haemangioma derived endothelial cells (HemDECs) have shown that they express both b1 and b2 adrenergic receptors, and that treatment with propranolol causes increased levels of HemDEC apoptosis.21,22 Our results do not clearly show a selective induction of apoptosis on CD31þ endothelial cells, suggesting that apoptosis of other cell populations in addition to endothelial cells within IH may contribute to the mechanism of propranolol treatment. Regardless of the specific cell types affected, our data confirm the induction of apoptosis in IH by propranolol for the first time in vivo. We also speculate that propranolol may accelerate the maturation of microvessels, as normally seen during involution of IH.1 We have previously shown that immature ECs in proliferative IH express OPG, and that OPG is lost as ECs mature and increase expression of vWF.11 The microvessels of patient B were almost exclusively of the mature phenotype with minimal OPG and strong vWF immunoreactivity, whereas those of patient A showed weak vWF and strong OPG staining, typical of immature microvessels. Since both lesions were diagnosed as being proliferative, we suggest that the difference observed between the two propranolol treated patients reflects a more advanced stage of response to propranolol treatment in patient B, rather than inherent differences in these samples. It is possible that if the propranolol treatment had been continued
for a longer period prior to surgery, vessels with a mature phenotype would have predominated in both propranolol treated patients. A beneficial effect of propranolol treatment is a reduced fibrofatty residuum. This reduced residuum could result from depletion of the MSC population, inhibition of their differentiation, or enhanced degradation of the basement membrane that normally develops during involution. Staining for CD90 (Supplementary Fig. 1, http://links.lww.com/PAT/A19) to identify the perivascular MSC population did not identify any differences in the number of CD90 staining cells between the samples, suggesting that propranolol does not induce depletion of adipocyte precursors. No attempt was made to determine the effects of propranolol on adipocyte differentiation in this study. Mast cells are known to secrete a range of mediators to degrade the ECM, including tryptase which is important in tissue remodelling.15 A marked increase in degranulation of mast cells was seen in both propranolol treated patients A and B. This is consistent with the idea that propranolol induces mast cell degranulation and tissue remodelling through the action of secreted proteases, contributing to the reduced residuum. Consistent with this idea, increased levels of matrix metalloprotease-2 (MMP2) are found in the urine of propranolol treated IH patients.23 However, it cannot be discounted that the degranulation observed here is a consequence of ulceration of the lesions, rather than a consequence of propranolol treatment. IH represents a multifaceted disease involving not only endothelial cells, but also haematopoietic, mesenchymal, stromal and progenitor cell populations.8,12,13,15,16,24,25 In vitro studies have shown a range of effects of propranolol on HemDECs,21,22 with similar effects also observed in nonhaemangioma endothelial cell lines.22 That propranolol treatment has no discernable effects on the developing vasculature of the infant, but does on the vasculature of IH, highlights the limitations of in vitro cell studies. Our analysis of propranolol treated IH patients shows an increase in apoptosis, and we suggest accelerated microvascular maturation and enhanced tissue remodelling mediated by mast cell degranulation. We expect that these processes would be more pronounced in the vast majority of cases for which the response to propranolol is swifter than the two slowly responding patients analysed in this study.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
500
STEEL and DAY
Conflicts of interest and sources of funding: We thank the Wellington Medical Research Foundation and Victoria University of Wellington for financial support. The authors state that there are no conflicts of interest to disclose. Address for correspondence: Dr D. Day, School of Biological Sciences, Victoria University of Wellington, PO Box 600, Wellington 6140, New Zealand. E-mail: [email protected]
References 1. Boscolo E, Bischoff J. Vasculogenesis in infantile hemangioma. Angiogenesis 2009; 12: 197–207. 2. Kilcline C, Frieden IJ. Infantile hemangiomas: How common are they? A systematic review of the medical literature. Pediatr Dermatol 2008; 25: 168–73. 3. Chang LC, Haggstrom AN, Drolert BA, et al. Growth characteristics of infantile hemangiomas: Implications for management. Pediatrics 2008; 122: 360–7. 4. North PE, Waner M, Mizeracki A, et al. GLUT1: A newly discovered immunohistochemical marker for juvenile hemangiomas. Hum Pathol 2000; 31: 11–22. 5. Zimmermann AP, Wiegand S, Werner JA, et al. Propranolol therapy for infantile haemangiomas: Review of the literature. Int J Pediatr Otorhinolaryngol 2010; 74: 338–42. 6. George ME, Sharma V, Jacobson J, et al. Adverse effects of systemic glucocorticosteroid therapy in infants with hemangiomas. Arch Dermatol 2004; 140: 963–9. 7. Leaute-Labreze C, de la Roque ED, Hubiche T, et al. Propranolol for severe hemangiomas of infancy. N Engl J Med 2008; 358: 2649–51. 8. Yu Y, Fuhr J, Boye E, et al. Mesenchymal stem cells and adipogenesis in hemangioma involution. Stem Cells 2006; 24: 1605–12. 9. Storch CH, Hoeger PH. Propranolol for infantile haemangiomas: insights into the molecular mechanisms of action. Br J Dermatol 2010; 163: 269–74. 10. Itinteang T, Brasch HD, Tan ST, et al. Expression of components of the renin-angiotensin system in proliferating infantile haemangioma may account for the propranolol-induced accelerated involution. J Plast Reconstr Aesthet Surg 2011; 64: 759–65.
Pathology (2014), 46(6), October
11. Vishvanath A, Itinteang T, Tan ST, et al. Infantile haemangioma expresses tumour necrosis factor-related apoptosis-inducing ligand (TRAIL), TRAIL receptors, osteoprotegerin and receptor activator for nuclear factor kappa B ligand (RANKL). Histopathology 2011; 59: 397–406. 12. Itinteang T, Vishvanath A, Day DJ, et al. Mesenchymal stem cells in infantile haemangioma. J Clin Pathol 2011; 64: 232–6. 13. Itinteang T, Tan ST, Brasch H, et al. Primitive mesodermal cells with a neural crest stem cell phenotype predominate proliferating infantile haemangioma. J Clin Pathol 2010; 63: 771–6. 14. Yuan SM, Chen RL, Shen WM, et al. Mesenchymal stem cells in infantile hemangioma reside in the perivascular region. Pediatr Dev Pathol 2012; 15: 5–12. 15. Sun ZJ, Zhao YF, Zhao JH. Mast cells in hemangioma: A double-edged sword. Med Hypotheses 2007; 68: 805–7. 16. Itinteang T, Tan ST, Brasch H, et al. Haemogenic endothelium in infantile haemangioma. J Clin Pathol 2010; 63: 982–6. 17. Itinteang T, Tan ST, Jia J, et al. Mast cells in infantile haemangioma possess a primitive myeloid phenotype. J Clin Pathol 2013; 66: 597–600. 18. Tan ST, Wallis RA, He Y, et al. Mast cells and hemangioma. Plast Reconstruct Surg 2004; 113: 999–1011. 19. Hsu TC, Wang JD, Chen CH, et al. Treatment with propranolol for infantile hemangioma in 13 Taiwanese newborns and young infants. Pediatr Neonatol 2012; 53: 125–32. 20. Iwata J, Sonobe H, Furihata M, et al. High frequency of apoptosis in infantile capillary haemangioma. J Pathol 1996; 179: 403–8. 21. Chim H, Armijo BS, Miller E, et al. Propranolol induces regression of hemangioma cells through HIF-1 alpha-mediated inhibition of VEGF-A. Ann Surg 2012; 256: 146–56. 22. Stiles J, Amaya C, Pham R, et al. Propranolol treatment of infantile hemangioma endothelial cells: A molecular analysis. Exp Ther Med 2012; 4: 594–604. 23. Kleber CJ, Spiess A, Kleber JB, et al. Urinary matrix metalloproteinases-2/ 9 in healthy infants and haemangioma patients prior to and during propranolol therapy. Eur J Pediatr 2012; 171: 941–6. 24. Itinteang T, Tan ST, Brasch HD, et al. Infantile haemangioma expresses embryonic stem cell markers. J Clin Pathol 2012; 65: 394–8. 25. Itinteang T, Tan ST, Brasch HD, et al. Primitive erythropoiesis in infantile haemangioma. Br J Dermatol 2011; 164: 1097–100.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
ANATOMICAL PATHOLOGY
Increased apoptosis and secretion of tryptase by mast cells in infantile haemangioma treated with propranolol RYAN STEEL
AND
DARREN DAY
School of Biological Sciences, Victoria University of Wellington, Wellington, New Zealand
Summary Propranolol is increasingly used to treat problematic infantile haemangioma (IH), although its molecular mechanisms remain unclear. A key feature of propranolol therapy is the decreased deposition of fibrofatty residuum compared with spontaneously involuting IH. This study investigated the molecular consequences of propranolol treatment for IH in vivo. Immunohistochemical and terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) staining was performed on five age matched patients with proliferative IH. Two patients (A and B) were undergoing propranolol treatment at the time of surgical resection. Propranolol treatment increased apoptosis, and induced mast cells to degranulate and secrete tryptase into the interstitium. The microvessels of patient A were immature [weak von Willibrand Factor (vWF), and strong osteoprotegerin (OPG) staining], comparable to untreated proliferative IH, while those of patient B were mature (strong vWF staining, and no OPG staining). The perivascular CD90þ mesenchymal stem cell population was preserved in both propranolol treated patients. Using rarely obtained biopsies from IH patients treated with propranolol, we show increased apoptosis by propranolol for the first time in vivo. We also suggest that mast cells, through secreted proteases, may contribute to the decreased fibrofatty residuum seen with propranolol treatment. Key words: Apoptosis, endothelial cells, immunohistochemistry, infantile haemangioma, mast cells, mesenchymal stem cell, microvessel maturity, osteoprotegerin (OPG), propranolol, tryptase. Received 8 December 2013, revised 29 April, accepted 1 May 2014
Recently the non-specific b-blocker propranolol has emerged as an effective and non-invasive therapy eliciting a rapid regression of the lesion with a significantly reduced fibrofatty residuum.5,7,8 Despite its therapeutic success, the underlying molecular mechanisms of propranolol action on IH are not fully understood. Proposed mechanisms include pro-apoptotic, antiangiogenic and vasoconstrictive effects,9 as well as modulation of the renin angiotensin system (RAS).10 The elucidation of propranolol’s mechanism remains difficult because of the scarcity of propranolol treated IH tissue for analysis. Additional actions of propranolol treatment may be possible. Acceleration of EC maturation normally seen during IH involution,1,4 in which osteoprotegerin (OPG) is lost from and strong von Willibrand Factor (vWF) staining is gained by ECs,11 represents a possible mechanism of action. A significant benefit of propranolol treatment is the reduced fibrofatty residuum. This residuum is possibly derived from the perivascular mesenchymal stem cell (MSC) population,12–14 the regulation of which has been proposed as a therapeutic target for IH.8 Whether this population is reduced by propranolol treatment is unknown. Additionally, mast cells synthesise and secrete a range of bioactive mediators, including proteases that degrade the extracellular matrix (ECM). Their abundance increases during involution, suggesting a role for mast cells in the resolution of this lesion.15 Whether mast cells play a role in the action of propranolol therapy is unexplored. Increased apoptosis, accelerated microvessel maturation and degradation of the ECM by mast cells may be important components of the mechanism of propranolol action. Here we examine these molecular components in biopsies from two propranolol treated patients whom later also underwent surgical resection.
INTRODUCTION Infantile haemangioma (IH) is a common microvascular tumour of infancy. IH is most common in Caucasian, female, and premature infants, affecting 5–10% of the population. Three phases of growth characterise the progression of IH, lasting up to 12 years of age. In proliferative IH (up to 12 months of age), pericytes and plump endothelial cells (ECs) with scattered mitotic figures form small tightly packed capillaries. These capillaries are immunoreactive for EC markers CD31 and CD34, as well as the definitive IH marker GLUT1. As these lesions involute, there is a thickening and lamination of the capillary basement membrane, a flattening of the ECs, and increased deposition of fibrous stroma between the vessels. Large vessels separated by a sparse fibrofatty residuum characterise the fully involuted lesion.1–4 In cases of IH involving significant morbidity, corticosteroids or surgery have been the preferred treatment options.5,6 Print ISSN 0031-3025/Online ISSN 1465-3931 DOI: 10.1097/PAT.0000000000000143
#
MATERIALS AND METHODS Patients Tissue collection and processing were approved by the Wellington Regional Ethics Committee, and signed consent for the use of tissue was obtained from the patients’ next-of-kin. Solid biopsies were collected from the centre of the lesions, often including subcutaneous tissue, of five patients with proliferative IH aged between 4 and 8 months (mean 6 months) at the time of surgical resection. Biopsies were immediately fixed in 10% formalin overnight then paraffin embedded using standard protocols for long-term storage at room temperature. Samples stored in paraffin are stable for several years, although typically tissue used was less than 1 year old. Of these five patients, two (A and B) were treated with propranolol (mean 5 months at age of biopsy) while the remaining three (mean 6 months) received no treatment. The diagnosis of IH for all patients was made by a trained pathologist based on the histopathology of the lesions, as well as expression of the definitive IH marker, GLUT1.4
2014 Royal College of Pathologists of Australasia
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
STUDY OF PROPRANOLOL ACTION ON IH IN VIVO
Patient A Patient A presented with an ulcerating IH of the right earlobe at 2.5 months of age and treatment with 1.5 mg/kg/day propranolol was started. The patient experienced on-going ulceration and pain despite slow regression of the haemangioma, so the lesion was debulked at 4 months of age. Following surgery, propranolol treatment was stopped, during which time the haemangioma increased in size. Propranolol treatment was reinstated 3 weeks after surgery at 2 mg/kg/day, which resulted in continued regression of the lesion. Patient B Patient B presented with an ulcerating haemangioma on the buttock at 4 months of age and treatment with 1.5 mg/kg/day propranolol was started. Slower than expected regression and ongoing ulceration was observed. At 6 months of age a biopsy was taken, but the majority of the haemangioma was left in situ. The patient remained on propranolol treatment and the lesion continued to gradually regress. Immunohistochemistry Immunohistochemistry was performed on 4 mm paraffin sections as previously described,11,13,16,17 using primary antibodies raised in either rabbit or mouse against CD34 (1:300; Invitrogen, USA), CD31 (1:300; Abcam, UK), OPG (1:100; R&D Systems, USA), tryptase (1:500; Abcam), von Willibrand Factor (vWF; 1:200; Abcam), CD90 (1:200; Abcam) and smooth muscle actin (SMA; 1:200; Abcam). Species-specific secondary antibodies conjugated to either Alexa Flour 488 or 555 (1:500; Invitrogen) were used to visualise epitope bound primary antibodies. Slides were mounted in ProLong Gold Antifade containing 4’, 6-diamidino-2-phenylindole (DAPI; Invitrogen) to visualise cell nuclei. Apoptosis detection Apoptotic nuclei were detected using an In Situ Cell Death Detection Kit (Roche, Switzerland) following the manufacturer’s instructions. This kit is based on the terminal deoxynucleotidyl transferase dUTP nick end labelling (TUNEL) technique, in which fluorescein-labelled nucleotides are attached at sites of single and double stranded DNA breaks typical of apoptotic cells. The TUNEL reaction was performed after the washing of primary antibodies, and before secondary antibody incubation. Image acquisition and statistical analysis Quantitative analyses were performed on the frequency of TUNEL stained nuclei, and on the maturity of the microvessels within IH on the basis of OPG and vWF staining.11 Multiple fields of view (typically 5 fields for OPG/vWF stained sections, and 25 fields for TUNEL stained sections) were obtained for these analyses to ensure that at least 100–200 events were counted per patient. The number of apoptotic nuclei per field of view was expressed as the percent of all nuclei, whereas the number of immature OPG positive vessels was expressed as the percent of all vessels with an identifiable lumen. All images were captured using an Olympus FV1000 confocal laser-scanning microscope fitted with a krypton/argon laser (Olympus, Japan), and a Student’s t-test was used to compare group means.
497
histologically proliferative containing microvessels that were immunoreactive for the definitive IH marker, GLUT1 (Fig. 1).4 Thus, our analyses report effects of propranolol at the earliest stages of regression. Apoptosis Enhanced apoptosis has been suggested as a possible mechanism for accelerated involution of IH by propranolol.9 Figure 2 shows TUNEL-labelled (green) apoptotic nuclei (counterstained with DAPI, blue) in both propranolol treated and untreated IH. Figure 2A shows a low level of apoptotic nuclei (yellow arrowheads) in untreated patients, while Fig. 2B shows many more TUNEL labelled nuclei in propranolol treated patients. The percentage of TUNEL reactive nuclei in both propranolol treated patients was significantly greater than that observed for untreated patients (Fig. 2C, p < 0.05). Microvessel maturation We have previously shown that OPG expression by immature microvessels is replaced by vWF expression as they mature.11 Strong OPG expression was seen in 89% of the microvessels of untreated patients (green), and these vessels expressed only low levels of vWF (red, Fig. 3A). In propranolol treated tissue, patient A was indistinguishable from the control group (data not shown). In patient B, however, very few of the microvessels were OPG positive and most vessels were also strongly vWF immunoreactive (Fig. 3B). Figure 3C graphically shows the fraction of vessels that were OPG positive for the untreated IH (n ¼ 3), as well as each of the propranolol treated patients (patient A and patient B). Tryptase secretion by mast cells The presence of tryptase positive mast cells in IH has been previously reported.18 Figure 4 identifies mast cells by way of tryptase immunoreactivity (green staining) in untreated (Fig. 4A) and a propranolol treated IH (Fig. 4B). The endothelium is identified by staining for CD31 (red), while nuclei are stained with DAPI (blue). Figure 4B shows that in propranolol treated tissue the mast cells have degranulated and secreted copious amounts of tryptase into the interstitium, while very little interstitial tryptase immunoreactivity was detected in any of the untreated IH patients (Fig. 4A). Increased interstitial tryptase was detected in both propranolol treated patients A and B.
RESULTS Effective therapy had been implemented in both patients treated with propranolol, although the rate of regression was slower than expected. The lesions of both patients were
A
DISCUSSION Propranolol has become the preferred treatment for IH with significant morbidity. Patients treated with propranolol
B
Fig. 1 Confirmation that propranolol treated tissues represent proliferative IH. Histological analysis by H&E and DAB staining showed tightly packed endothelial cells and pericytes forming microvessels expressing GLUT1 (appears brown) in both propranolol treated patient A (A) and patient B (B), confirming the diagnosis of IH. Scale bar ¼ 100 mm.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
498
Pathology (2014), 46(6), October
STEEL and DAY
A
B 3
No treatment
% TUNEL+ Cells
*
C
Propranolol treatment
2
(12,129)
1
(11,320)
0
Fig. 2 Propranolol treatment increased apoptosis in proliferative IH. Apoptotic nuclei identified by TUNEL staining (green, indicated by yellow arrowheads) were observed in both untreated (A) and propranolol treated (B) tissue. (C) Both of the propranolol treated patients (n ¼ 2) had a greater incidence of apoptosis than age matched, untreated controls (n ¼ 3). Cell nuclei are stained with DAPI and appear blue, while CD31þ endothelial cells stain red. Numbers in parentheses indicate the average number of cell nuclei scored for each patient. *p < 0.05 by Student’s t-test. Scale bar ¼ 50 mm.
OPG vWF
A
OPG vWF
B
C
% Vessels expressing OPG
100
Control 80 60
Propranolol treated: Patient A Propranolol treated: Patient B
40 20 0
Fig. 3 Propranolol treatment may induce a more mature microvessel phenotype in proliferative IH. (A) Immature microvessels that stain strongly for OPG (green) and weakly for vWF (red) predominated in proliferative IH. In propranolol treated tissue, patient A had an almost exclusively immature microvessel phenotype indistinguishable from age matched controls. (B) Patient B, however, had an almost exclusively mature vessel phenotype that stained strongly for vWF but not OPG. (C) Graphical depiction of the proportion of OPG positive vessels in untreated control (n ¼ 3), as well as each of the propranolol treated patients. Cell nuclei are stained with DAPI and appear blue. Scale bar ¼ 50 mm.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
STUDY OF PROPRANOLOL ACTION ON IH IN VIVO
Tryptase CD31
A
499
Tryptase CD31
B
Fig. 4 Increased tryptase secretion from mast cells in propranolol treated IH. Tryptase immunoreactivity is shown in green, and the endothelium of microvessels is identified by CD31 (red) in both panels. Staining for tryptase is tightly localised to cell nuclei (stained with DAPI, blue) in the representative untreated proliferative IH (A), but is diffusely dispersed in the vicinity of mast cells in the interstitium of propranolol treated tissue, as well as being present on the mast cells themselves (B). Scale bar ¼ 50 mm.
typically respond rapidly with a decolourisation and pronounced regression of the lesion within 2 weeks. Continued regression often completely resolves the lesion with minimal fibrofatty residuum.7,9,19 As such, propranolol treated patients rarely require surgery. We obtained proliferative IH tissue from two patients in whom regression of the lesion with propranolol treatment was evident, although slower than expected. This study thus represents a unique immunohistochemical investigation of the cellular consequences of propranolol treatment of patients with proliferative IH. We investigated whether propranolol increased apoptosis and vessel maturation in IH, and looked for potential explanations for the decreased fibrofatty residuum. Apoptosis in untreated patients was observed at levels comparable to previous reports,20 while both propranolol treated patients in this study had a significantly higher level of apoptosis as indicated by TUNEL staining. Increased apoptosis has previously been proposed as a molecular mechanism of propranolol treatment.9 Indeed, recent studies with haemangioma derived endothelial cells (HemDECs) have shown that they express both b1 and b2 adrenergic receptors, and that treatment with propranolol causes increased levels of HemDEC apoptosis.21,22 Our results do not clearly show a selective induction of apoptosis on CD31þ endothelial cells, suggesting that apoptosis of other cell populations in addition to endothelial cells within IH may contribute to the mechanism of propranolol treatment. Regardless of the specific cell types affected, our data confirm the induction of apoptosis in IH by propranolol for the first time in vivo. We also speculate that propranolol may accelerate the maturation of microvessels, as normally seen during involution of IH.1 We have previously shown that immature ECs in proliferative IH express OPG, and that OPG is lost as ECs mature and increase expression of vWF.11 The microvessels of patient B were almost exclusively of the mature phenotype with minimal OPG and strong vWF immunoreactivity, whereas those of patient A showed weak vWF and strong OPG staining, typical of immature microvessels. Since both lesions were diagnosed as being proliferative, we suggest that the difference observed between the two propranolol treated patients reflects a more advanced stage of response to propranolol treatment in patient B, rather than inherent differences in these samples. It is possible that if the propranolol treatment had been continued
for a longer period prior to surgery, vessels with a mature phenotype would have predominated in both propranolol treated patients. A beneficial effect of propranolol treatment is a reduced fibrofatty residuum. This reduced residuum could result from depletion of the MSC population, inhibition of their differentiation, or enhanced degradation of the basement membrane that normally develops during involution. Staining for CD90 (Supplementary Fig. 1, http://links.lww.com/PAT/A19) to identify the perivascular MSC population did not identify any differences in the number of CD90 staining cells between the samples, suggesting that propranolol does not induce depletion of adipocyte precursors. No attempt was made to determine the effects of propranolol on adipocyte differentiation in this study. Mast cells are known to secrete a range of mediators to degrade the ECM, including tryptase which is important in tissue remodelling.15 A marked increase in degranulation of mast cells was seen in both propranolol treated patients A and B. This is consistent with the idea that propranolol induces mast cell degranulation and tissue remodelling through the action of secreted proteases, contributing to the reduced residuum. Consistent with this idea, increased levels of matrix metalloprotease-2 (MMP2) are found in the urine of propranolol treated IH patients.23 However, it cannot be discounted that the degranulation observed here is a consequence of ulceration of the lesions, rather than a consequence of propranolol treatment. IH represents a multifaceted disease involving not only endothelial cells, but also haematopoietic, mesenchymal, stromal and progenitor cell populations.8,12,13,15,16,24,25 In vitro studies have shown a range of effects of propranolol on HemDECs,21,22 with similar effects also observed in nonhaemangioma endothelial cell lines.22 That propranolol treatment has no discernable effects on the developing vasculature of the infant, but does on the vasculature of IH, highlights the limitations of in vitro cell studies. Our analysis of propranolol treated IH patients shows an increase in apoptosis, and we suggest accelerated microvascular maturation and enhanced tissue remodelling mediated by mast cell degranulation. We expect that these processes would be more pronounced in the vast majority of cases for which the response to propranolol is swifter than the two slowly responding patients analysed in this study.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.
500
STEEL and DAY
Conflicts of interest and sources of funding: We thank the Wellington Medical Research Foundation and Victoria University of Wellington for financial support. The authors state that there are no conflicts of interest to disclose. Address for correspondence: Dr D. Day, School of Biological Sciences, Victoria University of Wellington, PO Box 600, Wellington 6140, New Zealand. E-mail: [email protected]
References 1. Boscolo E, Bischoff J. Vasculogenesis in infantile hemangioma. Angiogenesis 2009; 12: 197–207. 2. Kilcline C, Frieden IJ. Infantile hemangiomas: How common are they? A systematic review of the medical literature. Pediatr Dermatol 2008; 25: 168–73. 3. Chang LC, Haggstrom AN, Drolert BA, et al. Growth characteristics of infantile hemangiomas: Implications for management. Pediatrics 2008; 122: 360–7. 4. North PE, Waner M, Mizeracki A, et al. GLUT1: A newly discovered immunohistochemical marker for juvenile hemangiomas. Hum Pathol 2000; 31: 11–22. 5. Zimmermann AP, Wiegand S, Werner JA, et al. Propranolol therapy for infantile haemangiomas: Review of the literature. Int J Pediatr Otorhinolaryngol 2010; 74: 338–42. 6. George ME, Sharma V, Jacobson J, et al. Adverse effects of systemic glucocorticosteroid therapy in infants with hemangiomas. Arch Dermatol 2004; 140: 963–9. 7. Leaute-Labreze C, de la Roque ED, Hubiche T, et al. Propranolol for severe hemangiomas of infancy. N Engl J Med 2008; 358: 2649–51. 8. Yu Y, Fuhr J, Boye E, et al. Mesenchymal stem cells and adipogenesis in hemangioma involution. Stem Cells 2006; 24: 1605–12. 9. Storch CH, Hoeger PH. Propranolol for infantile haemangiomas: insights into the molecular mechanisms of action. Br J Dermatol 2010; 163: 269–74. 10. Itinteang T, Brasch HD, Tan ST, et al. Expression of components of the renin-angiotensin system in proliferating infantile haemangioma may account for the propranolol-induced accelerated involution. J Plast Reconstr Aesthet Surg 2011; 64: 759–65.
Pathology (2014), 46(6), October
11. Vishvanath A, Itinteang T, Tan ST, et al. Infantile haemangioma expresses tumour necrosis factor-related apoptosis-inducing ligand (TRAIL), TRAIL receptors, osteoprotegerin and receptor activator for nuclear factor kappa B ligand (RANKL). Histopathology 2011; 59: 397–406. 12. Itinteang T, Vishvanath A, Day DJ, et al. Mesenchymal stem cells in infantile haemangioma. J Clin Pathol 2011; 64: 232–6. 13. Itinteang T, Tan ST, Brasch H, et al. Primitive mesodermal cells with a neural crest stem cell phenotype predominate proliferating infantile haemangioma. J Clin Pathol 2010; 63: 771–6. 14. Yuan SM, Chen RL, Shen WM, et al. Mesenchymal stem cells in infantile hemangioma reside in the perivascular region. Pediatr Dev Pathol 2012; 15: 5–12. 15. Sun ZJ, Zhao YF, Zhao JH. Mast cells in hemangioma: A double-edged sword. Med Hypotheses 2007; 68: 805–7. 16. Itinteang T, Tan ST, Brasch H, et al. Haemogenic endothelium in infantile haemangioma. J Clin Pathol 2010; 63: 982–6. 17. Itinteang T, Tan ST, Jia J, et al. Mast cells in infantile haemangioma possess a primitive myeloid phenotype. J Clin Pathol 2013; 66: 597–600. 18. Tan ST, Wallis RA, He Y, et al. Mast cells and hemangioma. Plast Reconstruct Surg 2004; 113: 999–1011. 19. Hsu TC, Wang JD, Chen CH, et al. Treatment with propranolol for infantile hemangioma in 13 Taiwanese newborns and young infants. Pediatr Neonatol 2012; 53: 125–32. 20. Iwata J, Sonobe H, Furihata M, et al. High frequency of apoptosis in infantile capillary haemangioma. J Pathol 1996; 179: 403–8. 21. Chim H, Armijo BS, Miller E, et al. Propranolol induces regression of hemangioma cells through HIF-1 alpha-mediated inhibition of VEGF-A. Ann Surg 2012; 256: 146–56. 22. Stiles J, Amaya C, Pham R, et al. Propranolol treatment of infantile hemangioma endothelial cells: A molecular analysis. Exp Ther Med 2012; 4: 594–604. 23. Kleber CJ, Spiess A, Kleber JB, et al. Urinary matrix metalloproteinases-2/ 9 in healthy infants and haemangioma patients prior to and during propranolol therapy. Eur J Pediatr 2012; 171: 941–6. 24. Itinteang T, Tan ST, Brasch HD, et al. Infantile haemangioma expresses embryonic stem cell markers. J Clin Pathol 2012; 65: 394–8. 25. Itinteang T, Tan ST, Brasch HD, et al. Primitive erythropoiesis in infantile haemangioma. Br J Dermatol 2011; 164: 1097–100.
Copyright © Royal College of pathologists of Australasia. Unauthorized reproduction of this article is prohibited.