Importance of whole-body imaging with complete coverage of hands and feet in alveolar rhabdomyosarcoma staging.

BACKGROUND: Alveolar rhabdomyosarcoma commonly arises in the extremities and is characterized by aggressive biology and high frequency of metastases. Whole-body imaging is increasingly employed in pediatric oncology but not recommended as standard in the staging of soft-tissue sarcomas. OBJECTIVE: After observing patients with a large symptomatic alveolar rhabdomyosarcoma lesion and a smaller silent lesion in the more distal part of an extremity we sought to estimate the frequency of this constellation. MATERIALS AND METHODS: We retrospectively evaluated the data of prospectively registered paediatric patients (age <21 years) with alveolar rhabdomyosarcoma in the SoTiSaR (Soft Tissue Sarcoma Registry) of the Cooperative Weichteilsarkom Studiengruppe (CWS) 09/2011-04/2015 with regard to whole-body imaging. RESULTS: Seventy-five patients were eligible. Images of 57 patients had been submitted for reference consultation, including 80 whole-body examinations in 36 patients. Among them were 5 patients (14%, 95% confidence interval 3-25%) who had been diagnosed because of a symptomatic lesion while an additional silent lesion in the distal part of an extremity had remained unnoticed and had only been detected by later whole-body imaging. It is noteworthy that in 42 (53%) of all 80 whole-body examinations, the hands and feet had been only partially covered or completely excluded. CONCLUSION: In alveolar rhabdomyosarcoma silent lesions can be overlooked when the distal parts of the limbs are not thoroughly examined and not completely covered by imaging. Missing them influences treatment decisions and prognosis. Our results should be considered when evaluating the potential role of whole-body imaging in rhabdomyosarcoma.

Treatment and outcome of Ganglioneuroma and Ganglioneuroblastoma intermixed.

BACKGROUND: Ganglioneuroma (GN) and ganglioneuroblastoma intermixed (GNBI) are mature variants of neuroblastic tumors (NT). It is still discussed whether incomplete resection of GN/GNBI impairs the outcome of patients. METHODS: Clinical characteristics and outcome of localized GN/GNBI were retrospectively compared to localized neuroblastoma (NB) and ganglioneuroblastoma-nodular (GNBN) registered in the German neuroblastoma trials between 2000 and 2010. RESULTS: Of 808 consecutive localized NT, 162 (20 %) were classified as GN and 55 (7 %) as GNBI. GN/GNBI patients presented more often with stage 1 disease (68 % vs. 37 %, p < 0.001), less frequently with adrenal tumors (31 % vs. 43 %, p = 0.001) and positive mIBG-uptake (34 % vs. 90 %, p < 0.001), and had less often elevated urine catecholamine metabolites (homovanillic acid 39 % vs. 62 %, p < 0.001, vanillylmandelic acid 27 % vs. 64 %, p < 0.001). Median age at diagnosis increased with grade of differentiation (NB/GNBN: 9; GNBI: 61; GN-maturing: 71; GN-mature: 125 months, p < 0.001). Complete tumor resection was achieved at diagnosis in 70 % of 162 GN and 67 % of 55 GNBI, and after 4 to 32 months of observation in 4 GN (2 %) and 5 GNBI (9 %). Eleven patients received chemotherapy without substantial effect. Fifty-five residual tumors (42 GN, 13 GNBI) are currently under observation (median: 44 months). Five patients (3 GN, 2 GNBI) showed local progression; all had tumor residuals > 2 cm. No progression occurred after subtotal resection. Two patients died of treatment, none of tumor progression. CONCLUSIONS: GN/GNBI account for one quarter of localized NT and differ from immature tumors in their clinical features. Chemotherapy is not effective. Subtotal resection appears to be a sufficient treatment. TRIAL REGISTRATION: ClinicalTrials.gov identifiers - NB97 (NCT00017225; registered June 6, 2001); NB2004 (NCT00410631; registered December 11, 2006).

Genotyping circulating tumor DNA of pediatric Hodgkin lymphoma.

We used hybrid capture-targeted next-generation sequencing of circulating cell-free DNA (ccfDNA) of pediatric Hodgkin lymphoma (PHL) patients to determine pathogenic mechanisms and assess the clinical utility of this method. Hodgkin-Reed/Sternberg (HRS) cell-derived single nucleotide variants, insertions/deletions, translocations and VH-DH-JH rearrangements were detected in pretherapy ccfDNA of 72 of 96 patients. Number of variants per patient ranged from 1 to 21 with allele frequencies from 0.6 to 42%. Nine translocation breakpoints were detected. Genes involved in JAK/STAT, NFkB and PI3K signaling and antigen presentation were most frequently affected. SOCS1 variants, mainly deletions, were found in most circulating tumor (ct) DNAs, and seven of the nine translocation breakpoints involved SOCS1. Analysis of VH-DH-JH rearrangements revealed an origin of PHL HRS cells from partially selected germinal center B cells. Amounts of pretherapy ctDNA were correlated with metabolic tumor volumes. Furthermore, in all ccfDNA samples of 43 patients with early response assessment quantitative qPET < 3, indicative of a favorable clinical course, ctDNA was not detectable. In contrast, in five of six patients with qPET > 3, indicative of an unfavorable clinical course, ctDNA remained detectable. ccfDNA analysis of PHL is thus a suitable approach to determine pathogenic mechanisms and monitor therapy response.

Stanniocalcin 2 promotes invasion and is associated with metastatic stages in neuroblastoma.

Stanniocalcin 2 (STC2) is a secreted glycoprotein of as yet unknown functions. We investigated STC2 in human neuroblastoma, the most common solid extra-cranial tumor of infancy. In primary tumor samples, we found that expression of STC2 is associated with the metastatic Stages 4 and 4s and MYCN expression. In vitro, however, we demonstrate that cell proliferation is reduced by STC2 due to an increase in the basal apoptosis rate of the transfected cells. On the other hand, in vitro assays showed that STC2-transfected neuroblastoma cells have an increased invasive potential and display higher activity of collagen-degrading matrix metalloproteinase 2 (MMP2). Using experimental tumors on the chick chorioallantoic membrane (CAM), we observed that STC2 expressing cells show signs of emigration from the solid tumor and destroy blood vessels of the CAM, giving rise to massively bleeding tumors. Erosion of blood vessels was also seen when purified STC2 protein was applied on the CAM. Taken together, we demonstrate a dual role for STC2 in neuroblastoma. It reduces proliferation of tumor cells in vitro, but increases the invasive potential and induces bleeding, and thereby may facilitate early metastasis. The potential of STC2 as a surrogate marker for metastatic neuroblastoma calls for further investigation.

Keratoepithelin reverts the suppression of tissue factor pathway inhibitor 2 by MYCN in human neuroblastoma: a mechanism to inhibit invasion.

Neuroblastoma is the most frequent solid malignancy of children. The most reliable prognostic factor in neuroblastoma is the amplification status of the MYCN oncogene, but exceptions from this rule have been observed. Recently we have demonstrated that keratoepithelin (BIGH3, TGFBI) expression significantly reduces proliferation and invasion of neuroblastomas in vitro and in vivo. In these experiments, we also observed that tissue factor pathway inhibitor 2 (TFPI2, PP5, MSPI), a potent inhibitor of matrix-metalloproteinases, is most prominently up-regulated. As MYCN-amplified neuroblastomas are highly invasive, we sought to determine the interaction between MYCN, keratoepithelin and TFPI2. In this study we provide initial evidence that i) keratoepithelin expression in neuroblastoma inversely correlates with MYCN expression; ii) TFPI2 expression in neuroblastoma also correlates inversely with MYCN expression but positively with keratoepithelin expression and iii) keratoepithelin induces elevated TFPI2 transcript levels in neuroblastoma cells without alterations of MYCN expression.

Keratoepithelin suppresses the progression of experimental human neuroblastomas.

Neuroblastoma is the most common extracranial childhood tumor. High expression of activin A is associated with a favorable prognosis, but the contributing mechanisms have remained unclear. Our previous demonstration of the activin A-mediated up-regulation of keratoepithelin led to the consideration that keratoepithelin could modulate neuroblastoma growth and/or progression. We report here that enhanced keratoepithelin expression in human neuroblastoma cells suppresses neuroblastoma cell cohesion and adhesion to various extracellular matrix proteins and that it inhibits neuroblastoma cell proliferation and invasion in vitro and in vivo. Using microarray analysis, we identified several keratoepithelin-regulated genes that may contribute to these biological changes. Together with the observation that keratoepithelin is expressed in human neuroblastomas in vivo, our data suggest that keratoepithelin could play a beneficial role in neuroblastoma development and/or progression.

Three reactive compartments in venous malformations.

Vascular malformations affect 3% of neonates. Venous malformations (VMs) are the largest group representing more than 50% of cases. In hereditary forms of VMs gene mutations have been identified, but for the large group of spontaneous forms the primary cause and downstream dysregulated genes are unknown. We have performed a global comparison of gene expression in slow-flow VMs and normal saphenous veins using human whole genome micro-arrays. Genes of interest were validated with qRT-PCR. Gene expression in the tunica media was studied after laser micro-dissection of small pieces of tissue. Protein expression in endothelial cells (ECs) was studied with antibodies. We detected 511 genes more than four-fold down- and 112 genes more than four-fold up-regulated. Notably, chemokines, growth factors, transcription factors and regulators of extra-cellular matrix (ECM) turnover were regulated. We observed activation and "arterialization" of ECs of the VM proper, whereas ECs of vasa vasorum exhibited up-regulation of inflammation markers. In the tunica media, an altered ECM turnover and composition was found. Our studies demonstrate dysregulated gene expression in tunica interna, media and externa of VMs, and show that each of the three layers represents a reactive compartment. The dysregulated genes may serve as therapeutic targets.

Embryonic development and malformation of lymphatic vessels.

In the human, malformations of lymphatic vessels can be observed as lymphangiectasia, lymphangioma and lymphangiomatosis, with a prevalence of 1.2-2.8 per thousand. Their aetiology is unknown and a causal therapy does not exist. We investigated the origin of lymphatic endothelial cells (LECs) in avian and murine embryos, and compared the molecular profile of LECs from normal and malformed lymphatics of children. In avian embryos, Prox1+ lymphangioblasts are located in the confluence of the cranial and caudal cardinal veins, where the jugular lymph sac (JLS) forms. Cell lineage studies show that the JLS is of venous origin. In contrast, the lymphatics of the dermis are derived from mesenchymal lymphangioblasts located in the dermatomes, suggesting a dual origin of LECs in avian embryos. The same may hold true for murine embryos, where Lyve1+ LEC precursors are found in the cardinal veins, and in the mesenchyme. The mesenchymal cells express the pan-leukocyte marker CD45, indicating a cell type with lymphendothelial and leukocyte characteristics. In the human, such cells might give rise to Kaposi's sarcoma. Microarray analyses of LECs from lymphangiomas of children show a large number of regulated genes, such as VEGFR3. Our studies show that lymphvasculogenesis and lymphangiogenesis occur simultaneously in the embryo, and suggest a function for VEGFR3 in lymphangiomas.

Elevated expression of VEGFR-3 in lymphatic endothelial cells from lymphangiomas.

BACKGROUND: Lymphangiomas are neoplasias of childhood. Their etiology is unknown and a causal therapy does not exist. The recent discovery of highly specific markers for lymphatic endothelial cells (LECs) has permitted their isolation and characterization, but expression levels and stability of molecular markers on LECs from healthy and lymphangioma tissues have not been studied yet. We addressed this problem by profiling LECs from normal dermis and two children suffering from lymphangioma, and also compared them with blood endothelial cells (BECs) from umbilical vein, aorta and myometrial microvessels. METHODS: Lymphangioma tissue samples were obtained from two young patients suffering from lymphangioma in the axillary and upper arm region. Initially isolated with anti-CD31 (PECAM-1) antibodies, the cells were separated by FACS sorting and magnetic beads using anti-podoplanin and/or LYVE-1 antibodies. Characterization was performed by FACS analysis, immunofluorescence staining, ELISA and micro-array gene analysis. RESULTS: LECs from foreskin and lymphangioma had an almost identical pattern of lymphendothelial markers such as podoplanin, Prox1, reelin, cMaf and integrin-alpha1 and -alpha9. However, LYVE-1 was down-regulated and VEGFR-2 and R-3 were up-regulated in lymphangiomas. Prox1 was constantly expressed in LECs but not in any of the BECs. CONCLUSION: LECs from different sources express slightly variable molecular markers, but can always be distinguished from BECs by their Prox1 expression. High levels of VEGFR-3 and -2 seem to contribute to the etiology of lymphangiomas.

High activin A-expression in human neuroblastoma: suppression of malignant potential and correlation with favourable clinical outcome.

Amplification of the MYCN oncogene contributes to the malignant progression of human neuroblastomas, but the mechanisms have remained unclear. We have previously demonstrated that N-Myc facilitates angiogenesis by downregulating an angiogenesis inhibitor identified as the inhibin betaA homodimer activin A. Here, we have sought to define the molecular, biological and clinical consequences of activin A expression in human neuroblastoma. We report that enhanced activin A expression suppresses proliferation and colony formation of human neuroblastoma cells with amplified MYCN in vitro; that it inhibits neuroblastoma growth and angiogenesis in vivo; that it is highly expressed in differentiated, but not undifferentiated human neuroblastomas; and that it correlates with favourable outcome of neuroblastoma patients. Our results indicate that high activin A expression plays an important beneficial role in human neuroblastoma.

Microarray analysis reveals differential gene expression patterns and regulation of single target genes contributing to the opposing phenotype of TrkA- and TrkB-expressing neuroblastomas.

Expression of neurotrophin receptors of the tyrosine kinase receptor (Trk) family is an important prognostic factor in solid tumors including neuroblastoma. High expression of TrkA (NTRK1) is associated with a favorable biology and outcome of neuroblastoma, whereas TrkB (NTRK2) is expressed on aggressive neuroblastomas with unfavorable outcome. To gain new insights into the global gene expression program resulting in these divergent biological phenotypes, we stably expressed either TrkA or TrkB in the human SH-SY5Y neuroblastoma cell line. Gene expression profiles were obtained from parental cells and transfectants activated by their ligands in a time course over 24 h using oligonucleotide microarrays. Basal activation of Trk receptors in the absence of exogenous ligand was sufficient to induce broad and divergent genetic changes. Global gene regulation following external ligand stimulation was surprisingly similar in SY5Y-TrkA and SY5Y-TrkB cells except for the differential expression of distinct novel target genes. Consistent with their divergent biological phenotype, SY5Y-TrkA cells were characterized by upregulation of proapoptotic genes and angiogenesis inhibitors, whereas SY5Y-TrkB cells demonstrated upregulation of genes involved in invasion or therapy resistance. We suggest that the transcriptional program of neuroblastoma cells is modulated by Trk-receptor expression and basal activation rather than by ligand-induced activation. Fine-tuning of the malignant phenotype may be achieved by additional ligand stimulation with subsequent activation of a few specific genes.

Hepatocyte growth factor/c-Met signaling promotes the progression of experimental human neuroblastomas.

Neuroblastoma is the most frequent solid childhood malignancy. Despite aggressive therapy, mortality is high due to rapid tumor progression to advanced stages. The molecules and mechanisms underlying poor prognosis are not well understood. Here, we report that cultured human neuroblastoma cells express the hepatocyte growth factor (HGF) and its receptor c-Met. Binding of HGF to c-Met triggers receptor autophosphorylation, indicating functional relevance of this interaction. HGF activates several downstream effectors of c-Met such as the mitogen-activated protein kinases extracellular signal-regulated kinase 1/extracellular signal-regulated kinase 2 and phospholipase C-gamma, whereas signal transducer and activator of transcription 3 is constitutively activated in neuroblastoma cells expressing c-Met. In addition, HGF is able to stimulate expression and proteolytic activity of matrix metalloproteinase-2 and tissue-type plasminogen activator in neuroblastoma cells, thereby promoting degradation of extracellular matrix components. We show that HGF stimulates invasion of neuroblastoma cells in vitro and in vivo, and it promotes the formation of angiogenic neuroblastomas in vivo. These processes can be blocked by specific inhibitors of the mitogen-activated protein kinase cascade, by inhibitors of phospholipase C-gamma, and also by the expression of a dominant negative signal transducer and activator of transcription 3 mutant. Our data provide the first evidence that the HGF/c-Met pathway is essential for invasiveness and malignant progression of human neuroblastomas. They further suggest that specific inhibitors of this pathway may be suitable as therapeutic agents to improve clinical outcome of neuroblastomas.

N-myc oncogene overexpression down-regulates IL-6; evidence that IL-6 inhibits angiogenesis and suppresses neuroblastoma tumor growth.

Angiogenesis is an indispensable prerequisite for the progression and metastasis of solid malignancies. Tumor angiogenesis appears to be governed by alterations of tumor suppressor or oncogenes operant in a broad range of tumors. We have addressed this issue in neuroblastoma, a malignancy characterized by the near-exclusive amplification and overexpression of the N-Myc oncogene. Here, we report that N-Myc overexpression results in down-regulation of interleukin-6 (IL-6) and that IL-6 is an inhibitor of endothelial cell proliferation and VEGF-induced rabbit corneal angiogenesis. STAT3 is instrumental for IL-6 activity as infection with adenoviruses expressing a phosphorylation deficient STAT3 mutant renders endothelial cells insensitive to the antiproliferative action of IL-6. Finally, though IL-6 does not influence neuroblastoma cell growth, IL-6-expressing xenograft tumors in mice exhibit reduced neovascularization and suppressed growth. Our data shed new light on the mechanisms by which N-myc oncogene amplification enhances the malignant phenotype in neuroblastomas.

Development and engineering of lymphatic endothelial cells: clinical implications.

Studies on the lymphatic endothelium have been hampered by the difficulty to identify lymphatic endothelial cells (LECs) and to distinguish them from blood vascular endothelial cells (BECs). The situation was greatly improved by the identification of molecules with high specificity for LECs. A great deal of progress in the field of lymphangiogenesis research has been due to the detection of lymphangiogenic growth factors and their receptors, and there is growing evidence that these molecules are also involved in tumor-induced lymphangiogenesis and lymphatic dissemination of tumor cells. There is a considerable spectrum of congenital and acquired lymphedema-lymphangiodysplasia syndromes ranging from primary aplasia, hypoplasia and hyperplasia to secondary (acquired) obstructive, obliterative and surgical hindrance of lymph drainage. Consequently, there are a number of clinical applications for therapeutics that either inhibit or induce lymphangiogenesis. Although natural lymphatic regeneration is mostly very efficient, engineering of LECs may be useful in cases of lymphatic aplasia or hypoplasia. To achieve these goals, studies on the embryonic development and differentiation of LECs will reveal the key regulatory factors that need to be targeted.

Expression analysis of pediatric solid tumor cell lines using oligonucleotide microarrays.

We identified patterns of differentially-expressed genes in cell lines derived from several pediatric solid tumors. Affymetrix Human Cancer G110 Arrays, carrying 1,700 cancer-associated genes, were applied to a panel of 11 cell lines originating from Ewing tumors (ETs), neuroblastomas, and malignant melanoma of soft parts. Hierarchical clustering clearly differentiated these 3 entities and revealed groups of 75, 102, and 36 gene probe-sets exhibiting tumor-type specific up-regulation in these cell lines, respectively. Whereas ET lines demonstrated increased expression of microtubule-associated protein tau (MAPT), protein phosphatase 1 regulatory subunit 1A (PPP1R1A), NIMA (never in mitosis gene a)-related kinase 2 (NEK2), and cyclin D1 (CCND1), neuroblastoma samples exhibited high expression of wingless-type mouse mammary tumor virus integration site family member 11 (WNT11), Drosophila frizzled homolog 2 (FZD2), and adenomatous polyposis coli (APC) which are involved in regulating free beta-catenin levels. These genes likely maintain tumor-specific characteristics and participate in key downstream regulatory mechanisms. We also correlated the expression levels of up-regulated genes in ETs with their chromosomal localization and compared these data to the comparative genomic hybridization profiles of the cell lines. We demonstrate that gains of genetic material contribute essentially to differential gene expression.

Lymphangioblasts in embryonic lymphangiogenesis.

BACKGROUND: The origin of the lymphatic endothelium, either from the venous system or mesenchymal lymphangioblasts, presents as a persistent controversy. Recently, highly specific markers of the lymphatic endothelium have been found, enabling us to reinvestigate the embryonic origin of the lymphatics. METHODS AND RESULTS: The homeobox transcription factor, Prox1, is expressed in lymphatic, but not in blood vascular, endothelial cells throughout murine and avian development and in adult human tissues. Here we show expression of scattered Prox1-positive cells in the dermatome of 4-day-old chick embryos by in situ hybridization and immunostaining. These cells obviously form the Prox1-positive lymphatic networks in the skin of the body wall and the limbs during further embryonic development. The Prox1 protein is localized in the nuclei of the lymphatic endothelial cells (LECs). CONCLUSIONS: The results strongly suggest that the superficial lymphatics develop independently from the deep ones, and are derived from mesenchymal lymphangioblasts rather than veins. Our results argue against the unique origin of lymphatics from veins and suggest a heterogenous origin of LECs. The results are discussed in the context of historical data.

The proepicardium delivers hemangioblasts but not lymphangioblasts to the developing heart.

The mass of the myocardium and endocardium of the vertebrate heart derive from the heart-forming fields of the lateral plate mesoderm. Further components of the mature heart such as the epicardium, cardiac interstitium and coronary blood vessels originate from a primarily extracardiac progenitor cell population: the proepicardium (PE). The coronary blood vessels are accompanied by lymph vessels, suggesting a common origin of the two vessel types. However, the origin of cardiac lymphatics has not been studied yet. We have grafted PE of HH-stage 17 (day 3) quail embryos hetero- and homotopically into chick embryos, which were re-incubated until day 15. Double staining with the quail endothelial cell (EC) marker QH1 and the lymphendothelial marker Prox1 shows that the PE of avian embryos delivers hemangioblasts but not lymphangioblasts. We have never observed quail ECs in lymphatics of the chick host. However, one exception was a large lymphatic trunk at the base of the chick heart, indicating a lympho-venous anastomosis and a 'homing' mechanism of venous ECs into the lymphatic trunk. Cardiac lymphatics grow from the base toward the apex of the heart. In murine embryos, we observed a basal to apical gradient of scattered Lyve-1+/CD31+/CD45+ cells in the subepicardium at embryonic day 12.5, indicating a contribution of immigrating lymphangioblasts to the cardiac lymphatic system. Our studies show that coronary blood and lymph vessels are derived from different sources, but grow in close association with each other.

Mesenchymal cells with leukocyte and lymphendothelial characteristics in murine embryos.

The development of lymphatic endothelial cells (LECs) from deep embryonic veins or mesenchymal lymphangioblasts is controversially discussed. Studies employing quail-chick grafting experiments have shown that various mesodermal compartments of the embryo possess lymphangiogenic potential, whereas studies on murine embryos have been in favor of a venous origin of LECs. We have investigated NMRI mice from embryonic day (ED) 9.5 to 13.5 with antibodies against the leukocyte marker CD45, the pan-endothelial marker CD31, and the lymphendothelial markers Prox1 and Lyve-1. Early signs of the development of lymphatics are the Lyve-1- and Prox1-positive segments of the jugular and vitelline veins. Then, lymph sacs, which are found in the jugular region of ED 11.5 mice, express Prox1, Lyve-1, and CD31. Furthermore, scattered cells positive for all of the four markers are present in the mesenchyme of the dermatomes and the mediastinum before lymphatic vessels are present in these regions. Their number increases during development. A gradient of increasing CD31 expression can be seen the closer the cells are located to the lymph sacs. Our studies provide evidence for the existence of scattered mesenchymal cells, which up-regulate lymphendothelial and down-regulate leukocyte characteristics when they integrate into growing murine lymphatics. Such stem cells may also be present in the human and may be the cell of origin in post-transplantation Kaposi sarcoma.

Dual origin of avian lymphatics.

The earliest signs of the lymphatic vascular system are the lymph sacs, which develop adjacent to specific embryonic veins. It has been suggested that sprouts from the lymph sacs form the complete lymphatic vascular system. We have studied the origin of the jugular lymph sacs (JLS), the dermal lymphatics and the lymph hearts of avian embryos. In day 6.5 embryos, the JLS is an endothelial-lined sinusoidal structure. The lymphatic endothelial cells (LECs) stain (in the quail) positive for QH1 antibody and soybean agglutinin. As early as day 4, the anlagen of the JLS can be recognized by their Prox1 expression. Prox1 is found in the jugular section of the cardinal veins, and in scattered cells located in the dermatomes along the cranio-caudal axis and in the splanchnopleura. In the quail, such cells are positive for Prox1 and QH1. In the jugular region, the veins co-express the angiopoietin receptor Tie2. Quail-chick-chimera studies show that the peripheral parts of the JLS form by integration of cells from the paraxial mesoderm. Intra-venous application of DiI-conjugated acetylated low-density lipoprotein into day 4 embryos suggests a venous origin of the deep parts of the JLS. Superficial lymphatics are directly derived from the dermatomes, as shown by dermatome grafting. The lymph hearts in the lumbo-sacral region develop from a plexus of Prox1-positive lymphatic capillaries. Both LECs and muscle cells of the lymph hearts are of somitic origin. In sum, avian lymphatics are of dual origin. The deep parts of the lymph sacs are derived from adjacent veins, the superficial parts of the JLS and the dermal lymphatics from local lymphangioblasts.