SEARCH

SEARCH BY CITATION

Keywords:

  • D2-40 immunohistochemistry;
  • head-and-neck;
  • human fetus;
  • peripheral lymphatic vessels

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Using D2-40 immunohistochemistry, we assessed the distribution of peripheral lymphatic vessels (LVs) in the head-and-neck region of four midterm fetuses without nuchal edema, two of 10 weeks and two of 15 weeks’ gestation. We observed abundant LVs in the subcutaneous layer, especially in and along the facial muscles. In the occipital region, only a few LVs were identified perforating the back muscles. The parotid and thyroid glands were surrounded by LVs, but the sublingual and submandibular glands were not. The numbers of submucosal LVs increased from 10 to 15 weeks’ gestation in all of the nasal, oral, pharyngeal, and laryngeal cavities, but not in the palate. The laryngeal submucosa had an extremely high density of LVs. In contrast, we found few LVs along bone and cartilage except for those of the mandible as well as along the pharyngotympanic tube, middle ear, tooth germ, and the cranial nerves and ganglia. Some of these results suggested that cerebrospinal fluid outflow to the head LVs commences after 15 weeks’ gestation. The subcutaneous LVs of the head appear to grow from the neck side, whereas initial submucosal LVs likely develop in situ because no communication was evident with other sites during early developmental stages. In addition, CD68-positive macrophages did not accompany the developing LVs.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Using D2-40 immunohistochemistry, we recently described the lymphatic vessels (LVs) in the body including the lower neck and extremities of an 18-week-old human fetus [crown-rump length (CRL) 155 mm; Jin et al. 2010]. We observed many differences compared with LV distribution in adults. For example, subpleural and subperitoneal LVs were well developed, but bore no relation in space to the arterial or venous courses. Some viscera, including the spleen and the parathyroid and adrenal glands, did not contain D2-40-positive LVs at this stage. Finally, the abdominal alimentary canal wall was rich in LVs, whereas the esophagus was not. These findings indicated that fetal LVs are not a miniature version of adult LVs, findings supported by later detailed examination conducted along and around the fetal lesser sac (Kim et al. in press). However, we did not assess head-and-neck LVs in the 18-week-old fetus because we observed anomalies restricted to the eyes, the pterygoid, and the mandibular coronoid process (Jin et al. 2011a).

Knowledge of the peripheral lymphatic anatomy of human fetuses is very limited; most previous studies focused on the thoracic duct and major lymph trunks of early-stage fetuses (Sabin, 1909, 1912; van der Putte & van Limborgh, 1980; Petrenko & Gashev, 2008). Many injection studies have been performed using full-term fetuses (reviewed by Rouvière, 1981). These works suggested that LVs run along the arteries and veins and connect groups of lymph nodes, in both fetuses and adults. However, as lymph nodes develop later than LVs do (Sabin, 1909, 1912; Bailey & Weiss, 1975), the pattern of fetal lymph follicle distribution is unlikely to determine the basic configuration of LV development. We therefore sought to determine whether fetal LVs develop along the arteries or veins, most of which take different courses in the head-and-neck region. To better understand the principles of LV development, we assessed the fetal morphology of head-and-neck LVs. In addition, we identified macrophages in the areas in which LVs were developing because recent studies have highlighted a crucial role of macrophages in lymphangiogenesis (e.g. Kerjaschki, 2005; Melrose & Little, 2010).

Materials and methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

The study was performed in accordance with the provisions of the Declaration of Helsinki 1995 (as revised; Heidi et al. 2000). We examined four fetuses without nuchal edema, two of which had CRLs of 50 and 55 mm (approximately 10 weeks’ gestation) and two of which had CRLs of 100 and 125 mm (approximately 15 weeks’ gestation). The families donated the fetal specimens to the Department of Anatomy of Chonbuk National University, Korea, and the use thereof for research was approved by the University Ethics Committee. All four fetuses were obtained by induced abortion, after which the mothers were orally informed by an obstetrician (at a single hospital) that the fetuses could be donated for research purposes; no attempt was made to actively encourage such donation. The mothers subsequently agreed, and each fetus was stored in 10% w/w formalin, with an assigned specimen number, for more than 3 months. Because specimen numbers were randomized, it was not possible to link any individual fetus with the mother.

The head-and-neck regions of the two fetuses of CRL 55 and 125 mm were cut into sagittal sections, whereas the equivalent regions of the other two fetuses, of CRL 50 and 100 mm, were cut into horizontal sections. Before dehydration, the two larger specimens were decalcified by immersion in EDTA (0.5 m, pH 7.5) for 2 days at 4 °C. After routine paraffin embedding prior to histology, sections 10 μm in thickness were cut at 20-μm (the two 10-week-old fetuses) or 50-μm (the two 15-week-old fetuses) intervals, with at least two adjacent sections being prepared at each slice point. Most sections were stained with hematoxylin and eosin (HE), but adjacent sections (10% in number compared to HE sections) were subjected to immunohistochemistry to identify lymphatic vessels. The primary antibody (monoclonal anti-human podoplanin, 1 : 100 dilution; Nichirei D2-40, Nichirei, Tokyo, Japan) was used after specimen immersion in a ligand activator (contained in the Histofine SAB-PO kit; Nichirei) followed by autoclaving (105 °C, 10 min). This D2-40 antibody reacts with not only LVs but also cartilages, the basal layer of the mucosa and the dura mater (Jin et al. 2010, 2011b). In addition, CD68 antibody (1 : 100 dilution; Dako, Glostrup, Denmark) was applied to identify macrophages in the areas in which LVs were developing. Three fetuses (50, 100 and 125 mm CRL) were used for this part of work. The secondary antibody (contained in the Dako Chem Mate Envison kit; Dako) was tagged with horseradish peroxidase (HRP), and antigen-antibody reaction was detected via HRP-catalyzed color development using diaminobenzidine. Such samples were later counterstained with hematoxylin.

To evaluate numerical density conventionally, the numbers of LVs in the two 15-week-old specimens were counted in the highest-density area in the section at ×200 magnification using a ×20 objective lens. This method was based on Ishikawa et al. (2006). We counted LVs in at least four points in a single area of each of the two specimens after immunohistochemical staining. At this magnification, the two nasal walls, such as the mucosal layers lining the nasal septum and the medial turbinate, were all included in the objective field. Both cross-sections and longitudinal sections were examined and the total numbers of LVs calculated.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

In all four specimens, LVs were D2-40-immunoreactive, and antibody specificity was demonstrated by strong positive staining of subcutaneous LVs (e.g. those shown in Fig. 1). However, the intensity of immunoreactivity depended on both the specimen and the site examined. We found that the numerical density of LVs at 15 weeks was highest in the thyroid gland (range, 75–99) followed by the laryngeal (25–79) and nasal (6–29) submucosal layers (Table 1). Because we previously reported LV density in the lower neck, including the fetal jugular node chain (a chain of deep cervical nodes along the great vessels; Jin et al. 2010), this area was not studied in the present work.

image

Figure 1.  Lymphatic vessels in the orbital, facial, and occipital regions. Horizontal (A,B,F) and sagittal (C,D,E,G) sections were examined. Panel A (from a fetus of CRL 100 mm) shows the upper eyelid near the medial angle of the eye; most lymphatic vessels (arrows) run along the left–right axis of the eyelid. No vessel was observed along the conjunctival space (CS). Panel B (CRL 100 mm) shows the optic nerve (ON), the sheath of which (asterisks) reacted with D2-40. Panel C (CRL 125 mm) shows the subcutaneous tissue of the occipital region, with the occipital subcutaneous vessels (arrows) running along the supero-inferior axis. Panel D (CRL 55 mm) includes the pterygopalatine ganglion (PPG) and a nearby vessel. Panel E (CRL 125 mm) shows the anterior facial region above the angle of the mouth, with abundant vessels (arrows) cutting transversely along the maxilla (MX). Panel F (CRL 100 mm) shows vessels (arrows) running between the deep back muscles near the cervical dura mater (dura). The dura shows a cross-reaction with D2-40. Panel G (CRL 125 mm) shows the posterior fontanelle and the covering lymphatic vessels (arrows). Panels D and G are prepared at a different magnification (scale bar in A). LAO, levator anguli oris muscle; LC, lacrimal canaliculus.

Download figure to PowerPoint

Table 1.   Numbers of lymphatic vessels of 15-week-old fetuses counted in the highest density area in the section under ×20 magnification field of objective lens (× 200).
SitesAverageRange (difference between sections)
Thyroid, capsular, and interfollicular8575–99
Parotid gland52–16
Submandibular or sublingual gland21–5
Nasal submucosal layer176–29
Oral submucosal layer62–10
Pharyngeal submucosal layer83–15
Laryngeal submucosal layer4525–79
Tongue muscles31–18
Other sites including the ear< 10–3
Subcutaneous layer
 Face including eyelids185–46 (dilated)
 Occipital region53–12

LVs were well developed in the subcutaneous tissue of all four specimens. Dilated LVs were evident in the face, including the eyelids (Figs 1 and 2), and in the occipital region (Fig. 1). Subcutaneous LVs ran almost parallel to the supero-inferior axes of the occipital region, and along the left–right axis of the face including the eyelids. Facial muscle bundles were separated by dilated vessels (Figs 1E and 2B) and LVs were observed even at the bony attachment (Fig. 2A). However, LVs were not concentrated along or around hair follicles (Figs 1A,C and 2B). Occipital subcutaneous LVs were irregularly arranged along the posterior fontanelle (Fig. 1G). We could not demonstrate satisfactorily any communicating vessel between occipital subcutaneous LVs and the deep vessels running between the cervical back muscles (Fig. 1F). LV density at 15 weeks was much greater in the face than in the occipital region (Table 1). The parietal and temporal head skin of all four specimens contained few LVs. Consequently, the subcutaneous LVs appeared to grow upward from the neck to the face and occipital region.

image

Figure 2.  Lymphatic vessels and facial muscles. Sagittal sections were assessed. Panels A and B show infra- and supra-orbital areas of the same specimen (CRL 125 mm), respectively. In panel A, the vessels (stars) appear to invade the bony attachment of the orbicularis oculi muscle (arrowheads) to the maxilla (MX). Likewise, in panel B, the vessels appear to separate between muscle bundles (arrowheads) in the upper eyelid. Magnification of this figure is higher than Figs 3–7.

Download figure to PowerPoint

The submucosal layers of the nasal, oral, pharyngeal, and laryngeal cavities contained abundant LVs (Figs 3–6). LVs were particularly rich along the posterior aspects of the pharyngeal and laryngeal walls, in which regions the vessels ran along the supero-inferior axes (Figs 4D and 6D). However, LVs were not concentrated around the developing mucosal gland (Figs 4C and 6B) or the teeth germ (Fig. 4E). Notably, in the 10-week-old specimens, fewer submucosal LVs were evident along and around the nasal cavity, especially in the future maxillary sinus area (Fig. 3A), suggesting that LVs in these regions develop at between 10 and 15 weeks’ gestation. In the 15-week-old specimens, LVs attained the roof of the nasal cavity and the olfactory area, but few LVs were observed in the conjunctiva (Fig. 1A), the posterior half of the hard palate (Fig. 3F,G), the middle ear (Fig. 4A), the pharyngotympanic tube (Fig. 4C) or the cervical esophagus. Although the palatine and pharyngeal tonsils were in the early stages of development, these tissues contained D2-40-positive vessel-like structures (Fig. 4D). Consequently, submucosal LVs developed early in the larynx and pharynx and later in the oral and nasal regions. However, the direction of LVs differed among sites.

image

Figure 3.  Lymphatic vessels around the nasal cavity and in the palate. Horizontal (A,B,C,E) and sagittal (D,F,G) sections were examined. Panels A (CRL 50 mm) and B (CRL 100 mm) display the same lateral area of the nose from which the maxillary sinus later develops. The lymphatic vessels (arrows) appear near the maxilla (MX, panel A) and increase in number over time (panel B). The inferior nasal meatus (INM) is not yet developed in panel A. Panels C (CRL 100 mm) and D (CRL 125 mm) include the inferior turbinate (IT) and nasal septum (NS), respectively. Because of the folding, the nasal septum is cut longitudinally in the sagittal section. Panel E (CRL 100 mm) shows the midline submucosal tissue under the vomer. Cartilages as well as the basal layer of the mucosa (arrowheads in C–E) was positive for D2-40. Panels F (CRL 55 mm) and G (CRL 125 mm) show the mucosal and submucosal layers of the hard palate near the palatine bone. The number of palatine vessels did not increase in number between stages. All panels are prepared at the same magnification (bar in A).

Download figure to PowerPoint

image

Figure 4.  Lymphatic vessels in the ear and pharynx and around the oral cavity. Horizontal (A–D) and sagittal (E–G) sections are shown, with arrows indicating lymphatic vessels. Panels A and B (CRL 100 mm) show the middle and external ear regions, respectively. The parotid gland (PG) is surrounded by vessels (panel A). The auricular cartilage (AC) was also positive for D2-40 (panel B). Panels C and D (CRL 100 mm) show the pharyngeal walls, with abundant lymphatic vessels evident along and near the pharyngeal cavity (PX), especially at the posterior site (panel D). In contrast, the pharyngotympanic tube (PTT) had few lymphatic vessels. Panel E (CRL 125 mm) shows the upper tooth germ and oral mucosa. Panels F (CRL 55 mm) and G (CRL 125 mm) show vessels along and near the mandible (M); these increased in number from 10 to 15 weeks. All panels are prepared at the same magnification (bar in panel A). EAM, external auditory meatus; LC, longus colli muscle; MM, masseter muscle; PT, pharyngeal tonsil (not yet developed); TM, tympanic membrane.

Download figure to PowerPoint

image

Figure 5.  Lymphatic vessels in and around the tongue. Sagittal (A,C,E) and horizontal (B,D, F) sections are shown, with arrows indicating lymphatic vessels. Panels A (CRL 55 mm) and B (CRL 100 mm) show vessels in the tongue, with panel B showing vessels in the transverse and vertical (T&V) muscle layer. Panels C (CRL 125 mm) and D (CRL 100 mm) include the sublingual and/or submandibular glands (SLG, SMG). Panel E (CRL 125 mm) shows the soft palate between the oral and pharyngeal cavities (oral, PX), with the basal layers of the oral and pharyngeal mucosa being positive for D2-40 (arrowheads). Panel F (CRL 100 mm) shows vessel-rich connective tissue between the epiglottis (EP) and the root of the tongue. All panels are prepared at the same magnification (bar in panel A). GG, genioglossus muscle; H, hyoid bone; HG, hyoglossus muscle.

Download figure to PowerPoint

image

Figure 6.  Lymphatic vessels in and around the larynx. Horizontal (A,B,D,G) and sagittal (C,E,F) sections are shown, with arrows indicating lymphatic vessels. Panels A (CRL 50 mm) and B (CRL 100 mm) show vessels around the laryngeal cavity (LX), which increased in number until CRL attained 100 mm. Panel C (CRL 125 mm) shows a tangential section of the pharyngeal wall, with the submucosal vessels running transversely. Panel D (CRL 100 mm) includes the posterior part of the larynx, with the submucosal vessels running along the supero-inferior axis, and the basal layer of the laryngeal mucosa is also positive for D2-40 (arrowheads). Panel E (CRL 55 mm) shows the submucosal vessels of the trachea, and panel F (CRL 125 mm) shows vessels around the trachea, with the latter showing D2-40 positivity in the tracheal cartilage. Panel G (CRL 100 mm) shows the thyroid gland; both the capsule and the gland contain large numbers of vessels (arrowheads). All panels are prepared at the same magnification (bar in A). AR, arytenoid cartilage; CR, cricoid cartilage; ES, esophagus; PX, pharyngeal cavity; TH, thyroid cartilage.

Download figure to PowerPoint

The parotid and thyroid glands were enclosed by fascia containing abundant LVs (Figs 4A and 6G), whereas the sublingual and submandibular glands were accompanied by or contained few LVs (Fig. 5C,D). The interfollicular septa of the thyroid gland also contained abundant LVs: this confirmed our previous observation (Jin et al. 2010). The peripheral nerves, including the cranial autonomic and dorsal root ganglia, were accompanied by few LVs (Fig. 1B,D). The jugular node chain attained the level of the primitive hyoid body at 10 weeks and a level above the nodosa ganglion, but below the jugular foramen, at 15 weeks (data not shown). Likewise, no LVs were found along or near the dura mater (Fig. 1F) or in the plica ventralis, a primitive tentorium cerebelli (Cho et al. 2010). In the subcutaneous layer, no LVs accompanied the developing cutaneous nerve. The optic nerve sheath (Fig. 1B), dura mater (Fig. 1F), and choroid plexus of the meninges reacted with D2-40. Apart from that of the mandible, cartilage was not accompanied by surface LVs. Consequently, at 15 weeks, organ-specific configurations of LVs were seen in the parotid and thyroid glands.

Overall, we observed a clear increase in the number of LVs between 10 and 15 weeks in the nasal submucosa (Fig. 3A,B), the intrinsic tongue muscle (Fig. 5A,B), and the periosteum or perichondrium of the mandible (Fig. 4F,G). At 10 weeks, we found no communicating vessel between nasal submucosal LVs and other regional LVs, such as the facial subcutaneous LVs, the pharyngeal submucosal LVs, and/or the upward extension of the jugular node chain. Until 15 weeks, LVs showed little development along the conjunctiva, the hard palate, and the middle ear cavity, including the pharyngotympanic tube. We did not observe valve-like structures in the head-and-neck LVs of any specimen. Fetal head LVs did not accompany any artery or vein. Moreover, the LVs were usually thicker than nearby arteries and veins in diameter, with a few exceptions (i.e. the internal carotid artery and internal jugular vein were thicker than nearby LVs).

In addition, in all three fetuses examined, CD68-positive small macrophages were sparsely distributed along the cartilage and bone, along peripheral nerves and in striated muscles. There were fewer positive cells in the small fetus (50 and 55 mm CRL) than the larger ones (100 and 125 mm CRL). Abundant CD68-positive large cells (possibly osteoclasts) were concentrated in the developing bone. However, the subcutaneous and submucosal layers contained no or few positive cells. Likewise, we found no or few positive cells in the areas in which LVs were densely distributed (Fig. 7).

image

Figure 7.  CD68-positive macrophages in areas in which abundant lymphatic vessels are developing. Panel A displays a section near Fig. 3D (125 mm CRL), panel B exhibits a section near Fig. 6G (100 mm CRL) and panel C shows a section near Fig. 4D (100 mm CRL). CD68-positive macrophages (arrows) are seen along the nasal septum cartilage (A), along the laryngeal aspect of the cricoids cartilage (B) and in the loose tissue distant from the posterior pharyngeal wall. Notably, in (B), the thyroid gland contains no macrophage. The macrophage distribution is quite different from that of lymphatic vessels. All panels are prepared at the same magnification (bar in A). CR, cricoid cartilage; PX, pharyngeal cavity; NS, nasal septum.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

Utilizing D2-40 immunohistochemistry, we assessed the configuration of human fetal LVs in the head-and-neck region; the data extend those of a previous report (Jin et al. 2010). We observed no or few LVs along the peripheral nerves of the head and neck, including the cranial ganglion and nerve root, in contrast to our previous finding of distinct concentrations of LVs at and around the fetal intervertebral foramen including the neck region (Jin et al. 2010). Such LVs are especially rich in the thoracic region. LVs adjacent to the dura mater are most likely to contribute to cerebrospinal fluid outflow (Johnston, 2003; Johnston et al. 2004; Koh et al. 2005; Pollay, 2010). Thus, if any fetal outflow into LVs occurs, this likely commences in the thoracic region and expands into other regions. Although nasal submucosal LVs are thought to contribute greatly to fluid outflow (Kida et al. 1993; Johnston, 2003), such LVs do not develop until 10 weeks’ gestation, despite the early and rapid growth of the brain and associated fluid spaces. In addition, D2-40 expression in the optic nerve sheath may reflect initial development of LVs (Nishimura-Magari, 1953).

In agreement with previous findings (Jin et al. 2010), we observed considerable differences in LV density among sites, with the thyroid gland showing the highest density. Notably, thyroid gland LV density in fetuses aged 10 and 15 weeks (75–99 per objective field) was similar to that of an 18-week-old fetus (80–96 per objective field) (Jin et al. 2010). We also observed an extremely high density of LVs in the laryngeal submucosal layer, although the laryngeal submucosal LVs become restricted in the vocal cord prior to 7–10 months’ gestation (Liu et al. 2006). Fetal laryngeal LVs may provide mechanical conditions suitable for formation of a specific soft tissue, the vocal cord that is rich in regularly arrayed elastic fibers. This hypothetical condition seems to be consistent with the facial subcutaneous layer in which we demonstrated high density of LVs. The other submucosal layers of the head and neck had LV densities similar to or lower than those of the intestine and colon (Jin et al. 2010).

Turning to the subcutaneous layers, the face had many more LVs than did other parts of the body. Developing facial muscles, which co-exist with and interrupt growing LVs in the subcutaneous layers of the face, are likely to provide mechanical induction facilitating LV development. A similar situation may exist in the diaphragm, which contains abundant LVs in both fetuses (Jin et al. 2010) and adults (Masada et al. 1992; Li et al. 1996). However, the contribution of mechanical stress to the development of LVs does not appear to have been examined in recent molecular biology literature (e.g. Egorova et al. 2011).

By examining fetuses at 10 and 15 weeks’ gestation, we were able to compare LV morphology at different stages. Between 10 and 15 weeks, we found a clear increase in the number of LVs in the nasal submucosa, the intrinsic tongue muscle, and the periosteum or perichondrium of the mandible. Our findings indicate that LVs first develop in the laryngeal submucosa before 10 weeks’ gestation, followed by the nasal submucosa between 10 and 15 weeks, and the oral submucosa including the palate epithelium after 15 weeks. LVs may begin to develop earlier in the tongue muscles than in other striated muscles, except for the diaphragm, as few LVs were observed in the muscles of the girdle and extremities of a fetus of CRL 155 mm (Jin et al. 2010).

Notably, in the small fetuses, we found no communicating vessel between nasal submucosal LVs and other regional LVs. Likewise, initial LVs developing near the pterygopalatine ganglion were probably isolated from other LVs. Thus, simple sprouting of initial LVs from veins (van der Putte & van Limborgh, 1980; Karkkainen et al. 2003) may not adequately explain fetal development of deep head LVs. The nasal LVs, at least, are likely to develop in situ. According to one hypothesis, LVs in the fetal diaphragm may be of dual origin (Ohtani & Ohtani, 2008). Prior to formation of tubular vessels, numerous lymphatic endothelial cells independently migrate, subsequently coming together and connecting with pre-existing LVs, thereby developing a lymphatic network. In addition, we did not find any pathology in occipital LVs, although this is a major cause of nuchal translucency (Brand-Saberi et al. 1994; Bekker et al. 2006). However, only a few occipital LVs perforated the back muscles.

In contrast to abdominal and thoracic LVs, which develop along the serous membranes, our results suggest that fetal head LVs grow upward from the base of the neck, in subcutaneous and submucosal tissues. However, some initial LVs are likely to develop in situ, especially along the nasal cavities. The arteries and veins of the head seem to provide no inductive support for LV development. Although Melrose & Little (2010) reported a crucial role of macrophages in lymphangiogenesis in the human fetus at 12 and 14 weeks’ gestation, we found no or few macrophages in the areas in which abundant LVs were developing. Thus it seems to be unlikely that, in the head and neck of human fetuses, macrophages transdifferentiate into endothelium of LVs (reviewed by Kerjaschki, 2005).

Acknowledgement

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References

This paper was supported by Wonkwang University in 2010.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References
  • Bailey RP, Weiss L (1975) Ontogeny of the human fetal lymph nodes. Am J Anat 142, 1527.
  • Bekker MN, van der Akker NMM, Bartelings MM, et al. (2006) Nuchal edema and venous-lymphatic phenotype distribution in human fetuses and mouse embryos with aneuploidy. J Soc Gynecol Investig 13, 209216.
  • Brand-Saberi B, Flöel H, Christ B, et al. (1994) Alterations of the fetal extracellular matrix in the nuchal oedema of Down’s syndrome. Ann Anat 176, 539547.
  • Cho KH, Rodríguez-Vázquez JF, Han EH, et al. (2010) Human primitive meninges in and around the mesencephalic flexure particularly in their topographical relation to cranial nerves. Ann Anat 192, 322328.
  • Egorova AD, van der Heiden K, van de Pas S, et al. (2011) Tgfbeta/Alk5 signaling is required for shear stress induced Klf2 expression in embryonic endothelial cells. Dev Dyn 240, 16701680.
  • Heidi PF, Ezekiel E, Christine G (2000) The 2000 revision of the Declaration of Helsinki: a step forward or more confusion? The Lancet 358, 14491453.
  • Ishikawa Y, Akasaka Y, Kiguchi H, et al. (2006) The human renal lymphatics under normal and pathological conditions. Histopathology 49, 265273.
  • Jin ZW, Nakamura T, Yu HC, et al. (2010) Fetal anatomy of peripheral lymphatic vessels: a D2-40 immunohistochemical study using an 18-week human fetus (CRL 155 mm). J Anat 216, 671682.
  • Jin ZW, Li CA, Kim JH, et al. (2011a) Fetal head anomaly restricted to the eye, the mandible and the pterygoid process of the sphenoid: a histological study. Clin Anat 24, 599606.
  • Jin ZW, Song KJ, Lee NH, et al. (2011b) Contribution of the anterior longitudinal ligament to ossification and growth of the vertebral body: an immunohistochemical study using the human fetal lumbar vertebrae. Surg Radiol Anat 33, 1118.
  • Johnston M (2003) The importance of lymphatics in cerebrospinal fluid transport. Lymphat Res Biol 1, 4145.
  • Johnston M, Zakharov A, Papaiconomou C, et al. (2004) Evidence of connections between cerebrospinal fluid and nasal lymphatic vessels in humans, non-human primates and other mammalian species. Cerebrospinal Fluid Res 1, 2.
  • Karkkainen MJ, Haaiko P, Sainio K, et al. (2003) Vascular endothelial growth factor C is required for sprouting of the first lymphatic vessels from embryonic veins. Nat Immunol 5, 7480.
  • Kerjaschki D (2005) The crucial role of macrophages in lymphangiogenesis. J Clin Invest 115, 23162319.
  • Kida S, Pantazis A, Weller RO (1993) CSF drains directly from the subarachnoid space into nasal lymphatics in the rat. Anatomy, histology and immunological significance. Neuropathol Appl Neurobiol 19, 480488.
  • Kim JH, Han EH, Jin ZW, et al. (in press) Fetal topographical anatomy of the upper abdominal lymphatics: its specific features in comparison with other abdominopelvic regions. Anat Rec, in press.
  • Koh L, Zakharov A, Johnston M (2005) Integration of the subarachnoid space and lymphatics: is it time to embrace a new concept of cerebrospinal fluid absorption? Cerebrospinal Fluid Res 2, 6.
  • Li J, Zhao Z, Zhou J, et al. (1996) A study of the three-dimensional organization of the human diaphragmatic lymphatic lacunae and lymphatic drainage units. Ann Anat 178, 537544.
  • Liu YH, Xu SC, Tu LL, et al. (2006) A rich lymphatic network exists in the inferior surface of the vocal cord. Surg Radiol Anat 28, 125128.
  • Masada S, Ichikawa S, Nakamura Y, et al. (1992) Structure and distribution of the lymphatic vessels in the parietal pleura of the monkey as studied by enzyme-histochemistry and by light and electron microscopy. Arch Histol Cytol 55, 525538.
  • Melrose J, Little CB (2010) Immunolocalization of lymphatic vessels in human fetal knee joint tissues. Connect Tissue Res 51, 289305.
  • Nishimura-Magari S (1953) Extravasculare Saftbahnen des Augapfels und Sehnerven und ihre Bezieung zu den orbitalen Lymphgefaessen. Acta Schole Med Univ Kioto Jpn 31, 47102.
  • Ohtani O, Ohtani Y (2008) Organization and developmental aspects of lymphatic vessels. Arch Histol Cytol 71, 122.
  • Petrenko VM, Gashev AA (2008) Observations on the prenatal development of human lymphatic vessels with focus on basic structural elements of lymph flow. Lymphat Res Biol 6, 8995.
  • Pollay M (2010) The function and structure of the cerebrospinal fluid outflow system. Cerebrospinal Fluid Res 7, 9.
  • van der Putte SCJ, van Limborgh J (1980) The embryonic development of the main lymphatics in man. Acta Morphol Neerl-Scand 18, 323335.
  • Rouvière H (1981) Anatomie des lymphatiques de l’Homme. Paris: Masson.
  • Sabin FR (1909) The lymphatic system in human embryos, with a consideration of the morphology of the system as a whole. Am J Anat 9, 4391.
  • Sabin FR (1912) The development of the lymphatic system. In: Manual of Human Embryology, Vol. 2. (eds Keubel F, Mall FP), pp. 709745. Philadelphia: Lippincott.