You are here

Article Text

Article menu
Download PDFPDF
Original article
Infantile haemangioma expresses embryonic stem cell markers
  1. Tinte Itinteang1,2,3,
  2. Swee T Tan1,2,3,4,
  3. Helen D Brasch5,
  4. Ryan Steel1,
  5. Heather A Best1,
  6. Anasuya Vishvanath1,
  7. Jun Jia2,
  8. Darren J Day1,2,3
  1. 1School of Biological Sciences, Victoria University of Wellington, Wellington, New Zealand
  2. 2Gillies McIndoe Research Institute, Wellington, New Zealand
  3. 3Centre for the Study & Treatment of Vascular Birthmarks, Wellington Regional Plastic, Maxillofacial and Burns Unit, Hutt Hospital, Wellington, New Zealand
  4. 4University of Otago, Wellington, New Zealand
  5. 5Department of Pathology, Hutt Hospital, Wellington, New Zealand
  1. Correspondence to Professor Swee T Tan, Wellington Regional Plastic, Maxillofacial and Burns Unit, Hutt Hospital, High Street, Private Bag 31-907, Lower Hutt, New Zealand; swee.tan{at}huttvalleydhb.org.nz

Abstract

Background The origin of infantile haemangioma (IH) remains enigmatic. A primitive mesodermal phenotype origin of IH with the ability to differentiate down erythropoietic and terminal mesenchymal lineages has recently been demonstrated.

Aims To investigate the expression of human embryonic stem cell (hESC) markers in IH and to determine whether IH-derived cells have the functional capacity to form teratoma in vivo.

Methods Immunohistochemical staining and quantitative reverse transcription PCR were used to investigate the expression of hESC markers in IH biopsies. The ability of cells derived from proliferating IH to form teratomas in a mouse xenograft model was investigated.

Results The hESC markers, Oct-4, STAT-3 and stage-specific embryonic antigen 4 were collectively expressed on the endothelium of proliferating IH lesions, whereas Nanog was not. Nanog was expressed by cells in the interstitium and these cells did not express Oct-4, stage-specific embryonic antigen 4 or STAT-3. Proliferating IH-derived cells were unable to form teratomas in severely compromised immunodeficient/non-obese diabetic mice.

Conclusion The novel expression of hESC on two different populations of cells in proliferating IH and their inability to form teratomas in vivo infer the presence of a primitive cellular origin for IH downstream from hESC.

  • Angiogenesis
  • biological sciences
  • cancer research
  • cancer stem cells
  • cell biology
  • embryonic stem cells
  • haemogenic endothelium
  • head and neck cancer
  • infantile haemangioma
  • primitive mesoderm
  • tumour biology
  • tumour markers

https://doi.org/10.1136/jclinpath-2011-200462

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Infantile haemangioma (IH), the most common tumour of infancy, affects 10% of infants with a predilection for white, female and premature infants.1–4 It typically undergoes rapid growth during infancy and is characterised by the proliferation of immature capillaries that spontaneously involute to form a fibro-fatty residuum over the next 10–12 years.5 6

    We have recently shown that proliferating IH expresses brachyury, a marker of the primitive mesoderm, as well as markers associated with a neural crest phenotype.7 We have proposed that this primitive mesoderm gives rise to a haemogenic endothelium phenotype characterised by the collective expression of CD133, ACE, GATA-2, Tal-1 and CD34.8 In addition, in support of being a haemogenic endothelium, we have demonstrated that when IH biopsies are grown in vitro the cells emanating from the explants have the ability to undergo erythropoesis9 and mesenchymal differentiation,10 and that peptides derived from angiotensinogen are regulators of proliferation and differentiation (S.T. Tan, T. Itinteang, D.J. Day, et al.,Treatment of infantile haemangioma with low dose captopril. In preparation).11

    Human embryonic stem cells (hESC) are pluripotent cells derived from the inner cell mass of blastocyst that have the ability to differentiate into all three embryonic germ layers.12 13 hESC have been identified by the collective expression of transcription factors and cell surface receptors including Oct-4,14 STAT-3,15 stage-specific embryonic antigen 4 (SSEA-4)16 and Nanog.17 The presence of tumours derived from embryonic stem cells (ESC) is characterised by their ability to form cells of all three germ layers in vitro and teratoma in vivo.18 As hESC have been proposed as precursors of haemangioblasts19 that give rise to endothelial and haematopoietic stem cells, and we have previously demonstrated that IH is a primitive mesoderm-derived haemogenic endothelium with a neural crest phenotype,7 8 we wished to determine if IH maintained the expression of hESC markers.

    In this study we investigated the presence of hESC-associated markers using immunohistochemical staining and quantitative reverse transcription PCR (qRT–PCR) in IH. In addition, we examined the ability of cells isolated from proliferating IH lesions to form teratomas in a severely compromised immunodeficient/non-obese diabetic (SCID/NOD) mouse xenograft model.

    Materials and methods

    Tissues biopsies

    Biopsies of proliferating IH lesions affecting the skin and/or the underlying subcutaneous tissues in six patients aged 3–10 months (mean 6 months) and involuted lesions from three patients aged 8, 9 and 11 years and normal subcutaneous fat from two adults undergoing elective surgery were obtained according to a protocol approved by the Wellington Regional Ethics Committee.

    Immunolabelling

    Biopsies of proliferating IH lesions from six patients and involuted-IH lesions from three patients were formalin fixed and paraffin embedded using standard procedures. H&E and immunohistochemical analyses were performed on 4 μm serial sections and processed as previously described.7–9 Immunolabelling was performed with primary antibodies specific for anti-human CD34 (1:200), GLUT-1 (1:300), smooth muscle actin (SMA) (1:400), Oct-4 (1:300), SSEA-4 (1:300), STAT-3 (1: 200), PCNA (1:100) and Nanog (1:300). Negative controls included staining of normal adult human subcutaneous fat and omission of the primary antibodies.

    Quantitative RT–PCR

    For the analysis of the relative messenger RNA expression of Oct-4 and Nanog, RNA was extracted from snap-frozen samples of proliferating IH biopsies from five patients aged 3–10 months (mean 6 months) and involuted IH biopsies from three patients aged 8, 9, 11 years. RNA was extracted using a combination of Trizol (Life Technologies, Auckland, New Zealand) and column purification as previously described.9 10 Reverse transcription was performed using Superscript III and oligo-dT. qRT–PCR was performed using primers gene specific for Oct-4 (forward CGACCATCTGCCGCTTTGAG and reverse CCCCCTGTCCCCCATTCCTA)20 and Nanog (forward GATTTGTGGGCCTGAAGAAA, reverse AAGTGGGTTGTTTGCCTTTG). Amplification and detection was performed using SYBR green chemistry as previously reported.9 10 Expression relative to that for glyceraldehyde-3-phosphate dehydrogenase is reported as ΔCT values. Data were collected during the extension step and melt-curve analysis and agarose gel electrophoresis was performed to confirm the specificity of the amplification (data not shown).

    Xenografts

    The ability of IH-derived cells to form teratomas in SCID/NOD mice was investigated using cells obtained from proliferating IH lesions excised from five patients, aged 3–10 months (mean 6 months). Cells were isolated from dissociated IH explant biopsies as previously described.7 11 Eight SCID/NOD mice were injected subcutaneously in the left flank with between 1×105 and 1×107 cells derived from each IH patient. The cells were left to grow in six of the mice for a maximum of 8 weeks. Two animals were killed at 10 days and 3 weeks, respectively, and the tumour tissue was examined histologically for teratoma formation. The tumours were allowed to grow for 8 weeks in the remaining animals before killing and histological analysis. As a control, three mice were injected with 1×105 cells from an endothelioma cell line (ATCC, Sydney, Australia).21 All SCID/NOD mice experiments were performed according to a protocol approved by the Victoria University of Wellington Animal Ethics Committee.

    Results

    Immunohistochemistry

    The endothelium of all proliferating IH was immunoreactive for the endothelial associated marker, CD34 (figure 1A,C, green staining), and the CD34+ cells were surrounded by an outer concentric pericyte layer showing immunoreactivity for SMA (data not shown) as previously described.7 8 The endothelium of all IH lesions also showed immunoreactivity for the IH marker, GLUT-1, as expected8 22 (data not shown).

    Figure 1
    Figure 1

    Immunohistochemical staining of infantile haemangioma (IH) for human embryonic stem cell (hESC) markers. (A–D) Staining of representative sections of proliferating IH. (A) The endothelium of vessels are immunoreactive for CD34 (green). The same CD34+ cells are also immunoreactive for Oct-4 (red). (B) The cells of the endothelium are immunoreactive for stage-specific embryonic antigen-4 (green) but not for Nanog (red). A subpopulation of cells within the interstitium show immunoreactivity for Nanog but do not stain for stage-specific embryonic antigen-4. (C) Staining for CD34 (green) to unequivocally identify the endothelium, and Nanog (red). The staining in (C) corroborates the staining pattern present in (B) as none of the red Nanog+ cells are immunoreactive for CD34. (D) Staining for the transcription factor STAT-3 (red) and the proliferation marker PCNA (green) that identifies the nuclei of replicating and recently replicated cells. STAT-3 is expressed predominantly on the endothelium, with weaker immunoreactivity being detected on cells within the interstitium. The PCNA immunoreactive nuclei are also predominantly found on the cells of the endothelium. Selective doubly stained STAT-3 and PCNA immunoreactive nuclei are indicated with yellow arrow heads. (E and F) Staining of a representative involuted IH lesion. (E) Staining for CD34 (green) and Nanog (red). Larger blood vessels with a CD34+ endothelium are readily detected as expected, but no Nanog immunoreactivity is detected when compared with the Nanog staining of proliferating IH presented in (B and C). Similarly, (F) shows staining for CD34 (green) and Oct-4 (red) in an involuted IH lesion. No Oct-4 immunoreactivity is detected compared with that seen in proliferating lesions as described in (A). (G and H) Staining for CD34 (green) and Nanog (red, G) and Oct-4 (red, H) on normal adult subcutaneous fat. The staining pattern for the adult subcutaneous fat is very similar to the involuted IH samples presented in (E and H), and is markedly different from that seen in the proliferating lesions presented in (A–C). Cell nuclei are counterstained with DAPI (blue) in all panels. The scale bar is 10 μm for all panels except (D) in which it is 20 μm.

    Expression of the hESC-associated markers Oct-4, STAT-3, SSEA-4 and Nanog was investigated in proliferating and involuted IH. Figure 1 shows representative proliferating IH sections demonstrating strong immunoreactivity for Oct-4, STAT-3, SSEA-4 and Nanog. Figure 1A shows that the CD34+ cells (green staining) that form the endothelium also show strong immunoreactivity for Oct-4 (red staining). Immunoreactivity for SSEA-4 (green staining) is also localised to the endothelium and is sparse on the cells in the interstitium (figure 1B). However, only a subpopulation of cells within the interstitium shows strong immunoreactivity for Nanog (figure 1B, red staining). Figure 1C confirms that the cells that stain intensely for Nanog (red staining) are not those of the endothelium that stain for CD34 (green staining). Figure 1D shows that immunoreactivity for STAT-3 (red staining) is predominantly localised in the cells that form the endothelium of the microvessels. Regions that show the strongest immunoreactivity for STAT-3 also demonstrate the strongest immunoreactivity for PCNA (green staining), indicating that STAT-3 is expressed in proliferative regions of IH. Immunoreactivity for PCNA is predominantly located near the endothelium of the microvessels (yellow arrow heads). To determine whether the expression of these ESC-associated markers are lost as the tumour involutes, staining of involuted IH was undertaken using the same parameters for immunohistochemistry as that for the proliferating IH. Figure 1E,F shows that there are fewer CD34+ microvessels (green staining) and minimal immunoreactivity for Nanog (figure 1E, red staining) and Oct-4 (figure 1F, red staining). As with involuted IH, immunoreactivity for CD34 (green staining) was detected in normal adult subcutaneous fat, which, however, showed no immunoreactivity for Nanog (figure 1G, red staining).

    Figure 1A–C shows expression of CD34 (green staining) that identifies the endothelium of capillaries of proliferating IH. Figure 1A,B shows that the CD34+ endothelium also expresses hESC markers Oct-4 (red staining) and STAT-3, respectively (red staining). Figure 1C demonstrates that cells showing immunoreactivity for Nanog (red staining) are located in the interstitium but not the endothelium of the microvessels. Figure 1D confirms dual expression of Oct-4 (green staining) and SSEA-4 (red staining) by cells of the capillary endothelium.

    Figure 2A,B confirms that the expression of Oct-4 and SSEA are localised to the endothelium by counterstaining with SMA to identify the pericyte layer. Figure 2A shows that the immunoreactivity of Oct-4 (red staining) and SMA (green staining) does not overlap, similarly figure 2B shows that the endothelium stains for SSEA-4 (green staining) while the pericyte layer stains for SMA (red staining).

    Figure 2
    Figure 2

    The pericyte layer in proliferating infantile haemangioma (IH) does not express Oct-4 or stage-specific embryonic antigen-4. (A and B) Representative immunohistochemical staining of proliferating IH in which the pericyte layer is identified by immunoreactivity for smooth muscle actin (SMA, green). (A) The SMA+ pericyte layer (green) shows minimal immunoreactivity for Oct-4 (red) but the endothelium shows strong immunoreactivity for Oct-4 consistent with the data presented in figure 1(A). Similarly, the immunoreactivity for stage-specific embryonic antigen-4 is greatest on the endothelium, with only weak staining being detected on the SMA+ (red staining) pericyte layer. Cell nuclei are counterstained with DAPI (blue) in both panels. Scale bars are 10 μm.

    Quantitative RT–PCR

    To corroborate the staining presented in figures 1 and 2, qRT–PCR of mRNA isolated from proliferating and involuted IH tissues was undertaken. Expression levels for Oct-4 and Nanog were determined relative to those for glyceraldehyde dehydrogenase as a reference gene.10 Quantitative analysis confirmed the abundant expression of gene transcripts for Oct-4 and Nanog in proliferating IH, whereas the transcripts were only just detectable for the involuted lesions (table 1).

    View this table:
    Table 1

    Relative gene expression profile for proliferating and involuted IH tissues

    Xenografts

    All SCID/NOD mice injected with the endothelioma cell line grew tumours as expected (data not shown). All SCID/NOD mice showed obvious tumour formation 1 week following the injection of cells derived from proliferating IH lesions, and these tumours regressed spontaneously after 3 weeks in all six animals. H&E staining of the harvested tumour (figure 3A) in the two mice killed at day 10 and week 3, respectively, did not reveal teratoma formation (figure 3B,C), but the tumour tissues showed immunoreactivity for GLUT-1 (figure 3D) using a human-specific antibody confirming the presence of human IH-derived cells organised into vessel-like structures within the tumour. For the remaining six mice that were left for 8 weeks, H&E staining showed fat deposition but no identifiable tumour (data not shown).

    Figure 3
    Figure 3

    Cells derived from proliferating infantile haemangioma (IH) form tumour in severely compromised immunodeficient/non-obese diabetic (SCID/NOD) mice. (A) A freshly excised tumour obtained after 10 days after injection of proliferating IH-derived cells into a SCID/NOD mouse. (B) H&E and the enlargement in (C) show that the tumour is a composed of adipocytes (left-hand side) and proliferating endothelial cells (right-hand side). Immunohistochemical staining with a human-specific antibody for GLUT-1, a marker for IH, shows that a small region of the tumour consists of capillary-like structures expressing GLUT-1 (D, red). (C and D) Correspond to the box region of (B). Scale bars are 1 mm for (B) and 125 μm for (C and D).

    Discussion

    We have previously demonstrated the expression of primitive mesodermal (brachyury),7 primitive haematopoietic-associated (CD133, ACE, GATA-2, Tal-1 and EPO-R)8 9 markers and embryonic-associated haemoglobin (ζ chain)9 on the haemogenic endothelium of capillaries of proliferating IH and that the mature haematopoietic-associated marker, CD45, is only expressed by cells in the interstitium.8 We have also shown that this haemogenic endothelium has the ability to undergo primitive erythropoiesis in vitro9 and mesenchymal differentiation.10

    The isolation of hESC from the inner cell mass of the blastocyst has enabled the identification and study of the pluripotent nature of hESC.12 Oct-4, also known as POU5F1, is a mammalian POU transcription factor, which is expressed by pluripotent early embryonic cells.14 STAT-3, a member of the signal transducer and activator transcription factor family, has been shown to be crucial in ESC renewal and blockage of differentiation.15 SSEA-4 is a cell surface marker expressed by hESC.16 Nanog, a homeobox-containing transcription factor, is essential for the maintenance of pluripotency.17 The hallmark of ESC-derived tumours is the ability to form teratoma both in vivo and all three germ layers in vitro.18

    To the best of our knowledge this is the first report demonstrating the expression pattern of hESC-associated markers in IH. It is interesting that the expression of Oct-4, STAT-3 and SSEA-4 is localised to the endothelium of the capillaries in proliferating IH, while the expression of Nanog is limited only to cells in the interstitium, away from the endothelium. These data suggest that IH does not contain hESC, but is consistent with the notion that the haemogenic endothelium in IH being derived from hESC.

    The expression of Oct-4 by the endothelium and Nanog by cells in the interstitium is intriguing. Possible explanations are that the expression of the more primitive ESC-associated marker, Oct-4, represents the most primitive cell population within proliferating IH that gives rise to cells expressing the more downstream transcription factor, Nanog.17 Alternatively, there are two different primitive populations within proliferating IH lesions, one giving rise to the haemogenic endothelium and the other to the Nanog+ population in the interstitium. It is equally intriguing that proliferating IH-derived cells do not possess the ability to form teratoma in SCID/NOD mice, as shown in the study by Khan et al,23 in which CD133+ proliferating IH-derived cells were used.

    We have previously shown that the CD133+ proliferating cells in IH are localised to the endothelium of the capillaries of proliferating IH,7 8 11 and this study confirms the expression of the hESC markers Oct-4, SSEA-4 and STAT-3, but not Nanog, by the same population of cells. Results of these experiments are further supported by the clinical observation of the absence of teratoma formation in patients affected by IH.24 We therefore conclude that the expression of some, but not all, of the hESC markers used in the study by a single cell population in proliferating IH does not indicate the presence of ESC. It appears that this population of cells resembles a phenotype at a differentiation state downstream of hESC. These cells may reflect the true origin of IH, with more recent data inferring a placental chorionic villous mesenchymal core cell origin for IH.25

    The novel identification of hESC markers within proliferating IH provides further evidence that IH is a developmental anomaly. The 10% incidence of IH in the Caucasian population implies a potential source of ex-utero cells downstream of hESC, and IH as a unique model for studying human development.

    Take-home messages

    We have recently demonstrated a primitive mesodermal phenotype origin of IH with the ability to differentiate down erythropoietic and terminal mesenchymal lineages. The unique expression of hESC-associated transcription factors on two different populations of cells in proliferating IH and their inability to form teratomas in vivo infer the presence of a primitive cellular origin for IH downstream from hESC.

    This JCP review article has an accompanying set of multiple choice questions (MCQs)

    To access the questions, click on BMJ Learning: Take this module on BMJ Learning from the content box at the top right and bottom left of the online article. For more information please go to: http://jcp.bmj.com/education. Please note: The MCQs are hosted on BMJ Learning—the best available learning website for medical professionals from the BMJ Group. If prompted, subscribers must sign into JCP with their journal's username and password. All users must also complete a one-time registration on BMJ Learning and subsequently log in (with a BMJ Learning username and password) on every visit.

    References

    1. ↵
      1. Tan ST,
      2. Velickovic M,
      3. Ruger BM,
      4. et al
      . Cellular and extracellular markers of hemangioma. Plast Reconstr Surg 2000;106:529–38.
      OpenUrlCrossRefPubMedWeb of Science
    2. ↵
      1. Frieden IJ,
      2. Eichenfield LF,
      3. Esterly NB,
      4. et al
      . Guidelines of care for hemangiomas of infancy. American Academy of Dermatology Guidelines/Outcomes Committee. J Am Acad Dermatol 1997;37:631–7.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Takahashi K,
      2. Mulliken JB,
      3. Kozakewich HP,
      4. et al
      . Cellular markers that distinguish the phases of hemangioma during infancy and childhood. J Clin Invest 1994;93:357–64.
      OpenUrl
    4. ↵
      1. Tan ST,
      2. Itinteang T,
      3. Leadbitter P
      . Low-dose propranolol for infantile haemangioma. J Plast Reconstr Aesthet Surg 2011;64:292–9.
      OpenUrlCrossRefPubMed
    5. ↵
      1. Itinteang T,
      2. Withers AH,
      3. Leadbitter P,
      4. et al
      . Pharmacologic therapies for infantile hemangioma: is there rational basis? Plast Reconstr Surg 2011;128:499–507.
      OpenUrlCrossRefPubMed
    6. ↵
      1. Tan ST,
      2. Itinteang T,
      3. Leadbitter P
      . Low-dose propranolol for multiple hepatic and cutaneous hemangiomas with deranged liver function. Pediatrics 2011;127:e772–6.
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. Itinteang T,
      2. Tan ST,
      3. Brasch H,
      4. et al
      . Primitive mesodermal cells with a neural crest stem cell phenotype predominate proliferating infantile haemangioma. J Clin Pathol 2010;63:771–6.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Itinteang T,
      2. Tan ST,
      3. Brasch H,
      4. et al
      . Haemogenic endothelium in infantile haemangioma. J Clin Pathol 2010;63:982–6.
      OpenUrlAbstract/FREE Full Text
    9. ↵
      1. Itinteang T,
      2. Tan ST,
      3. Brasch HD,
      4. et al
      . Primitive erythropoiesis in infantile haemangioma. Br J Dermatol 2011;164:1097–100.
      OpenUrlCrossRefPubMed
    10. ↵
      1. Itinteang T,
      2. Vishvanath A,
      3. Day DJ,
      4. et al
      . Mesenchymal stem cells in infantile haemangioma. J Clin Pathol 2011;64:232–6.
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Itinteang T,
      2. Brasch HD,
      3. Tan ST,
      4. 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;64:759–5.
      OpenUrlCrossRefPubMed
    12. ↵
      1. Thomson JA,
      2. Itskovitz-Eldor J,
      3. Shapiro SS,
      4. et al
      . Embryonic stem cells lines derived from human blastocysts. Science 1998;282:1145–7.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Laslett AL,
      2. Filipczyk AA,
      3. Pera MF
      . Characterization and culture of human embryonic stem cells. Trends Cardiovasc Med 2003;13:295–301.
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Nichols J,
      2. Zevnik B,
      3. Anastassiadis K,
      4. et al
      . Formation of pluripotent stem cells in the mammalian embryo depends on the pou transcription factor Oct4. Cell 1998;95:379–91.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Burdon T,
      2. Smith A,
      3. Savatier P
      . Signalling, cell cycle and pluripotency in embryonic stem cells. Trends Cell Biol 2002;12:432–8.
      OpenUrlCrossRefPubMedWeb of Science
    16. ↵
      1. Brimble SN,
      2. Sherrer ES,
      3. Uhl EW,
      4. et al
      . The cell surface glycosphingolipids SSEA-3 and SSEA-4 are not essential for human ESC pluripotency. Stem Cells 2007;25:54–62.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Rodda DJ,
      2. Chew JL,
      3. Lim LH,
      4. et al
      . Transcriptional regulation of nanog by OCT4 and SOX2. J Biol Chem 2005;280:24731–7.
      OpenUrlAbstract/FREE Full Text
    18. ↵
      1. Xu C,
      2. Inokuma MS,
      3. Denham J,
      4. et al
      . Feeder-free growth of undifferentiated human embryonic stem cells. Nat Biotech 2001;19:971–4.
      OpenUrlCrossRefPubMedWeb of Science
    19. ↵
      1. Kennedy M,
      2. D'Souza SL,
      3. Lynch-Kattman M,
      4. et al
      . Development of the hemangioblast defines the onset of hematopoiesis in human ES cell differentiation cultures. Blood 2007;109:2679–87.
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Avery S,
      2. Zafarana G,
      3. Gokhale PJ,
      4. et al
      . The role of SMAD4 in human embryonic stem cell self-renewal and stem cell fate. Stem Cells 2010;28:863–73.
      OpenUrlPubMedWeb of Science
    21. ↵
      1. Williams RL,
      2. Risau W,
      3. Zerwes HG,
      4. et al
      . Endothelioma cells expressing the polyoma middle T oncogene induce hemangiomas by host cell recruitment. Cell 1989;57:1053–63.
      OpenUrlCrossRefPubMedWeb of Science
    22. ↵
      1. North PE,
      2. Milton W,
      3. Mizeracki A,
      4. et al
      . GLUT1: a newly discovered immunohistochemical marker for juvenile hemangiomas. Hum Pathol 2000;31:11–22.
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      1. Khan ZA,
      2. Boscolo E,
      3. Picard A,
      4. et al
      . Multipotential stem cells recapitulate human infantile hemangioma in immunodeficient mice. J Clin Invest 2008;118:2592–9.
      OpenUrlPubMedWeb of Science
    24. ↵
      1. Mulliken JB,
      2. Glowacki J
      . Hemangiomas and vascular malformations in infants and children: a classification based on endothelial characteristics. Plast Reconstr Surg 1982;69:412–19.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Itinteang T,
      2. Tan ST,
      3. Guthrie G,
      4. et al
      . A placental chorionic villous mesenchymal core cellular origin for infantile haemangioma. J Clin Pathol 2011;64:870–4.
      OpenUrlAbstract/FREE Full Text

    Footnotes

    • Aspects of this work were presented at the Australian and New Zealand Head and Neck Cancer Society, Sydney, Australia, 2–4 September 2010; the American Society of Plastic Surgeons' Meeting, Toronto, Canada, 1–5 October 2010; the 16th World Congress of the International Confederation for Plastic, Reconstructive and Aesthetic Surgery, Vancouver, Canada, 21–27 May 2011; and the 9th Annual Meeting of the International Society for Stem Cell Research, Toronto, Canada, 15–18 June 2011. This paper was awarded the prize for the best basic science presentation at the New Zealand Association of Plastic Surgeons' Annual Scientific Meeting, Auckland, New Zealand, 27–28 November 2011.

    • Funding This work was supported by grants from the Wellington Regional Plastic Surgery Unit Research and Education Trust, the Wellington Medical Research Foundation, the Surgical Research Trust, Pub Charities and the Cancer Society of New Zealand. AV was supported in part by the Cancer Society of New Zealand. TI is supported by the Royal Australasian College of Surgeons' foundation for surgery scholarship. None of these funders was involved in the research and its publication or for the publication.

    • Competing interests None.

    • Ethics approval All tissues used in this study were collected from patients according to a protocol approved by the Wellington Regional Ethics Committee. All SCID/NOD mice experiments were performed according to a protocol approved by the Victoria University of Wellington Animal Ethics Committee.

    • Provenance and peer review Not commissioned; externally peer reviewed.