Experimental Eye Research 125 (2014) 203e209
Contents lists available at ScienceDirect
Experimental Eye Research journal homepage: www.elsevier.com/locate/yexer
Absence of lymphatic vessels in the developing human sclera Simona L. Schlereth a, *, Barbara Neuser a, Martina C. Herwig b, Annette M. Müller c, € dl d, Claus Cursiefen a, Konrad R. Koch a, Herbert A. Reitsamer d, Falk Schro Ludwig M. Heindl a a
Department of Ophthalmology, University of Cologne, Kerpenerstr. 62, 50924 Cologne, Germany Department of Ophthalmology, University of Bonn, Ernst-Abbe-Str. 2, 53127 Bonn, Germany c Center of Pediatric Pathology and Pathology, MVZ Venusberg, University of Bonn, Sigmund-Freud-Str. 25, 53127 Bonn, Germany d Department of Ophthalmology and Anatomy, Paracelsus Medical University, Strubergasse 21, 5020 Salzburg, Austria b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 2 March 2014 Accepted in revised form 13 June 2014 Available online 27 June 2014
The adult sclera is free of lymphatic vessels, but contains a net of blood vessels. Whether and when this selectively lymphangiogenic privilege is achieved during embryologic development is not known yet. Therefore, we investigated the developing human sclera for blood- and lymphatic vessels in 34 abortions/stillborns (12e38 weeks of gestation). The probes were subdivided into three groups (group 1: 12 e18 weeks of gestation, n ¼ 10; group 2: 19e23 weeks of gestation, n ¼ 13; group 3: 24e38 weeks of gestation, n ¼ 11), and prepared for paraffin sections followed by immunohistochemistry against CD31 to detect blood vessels, and against lymphatic vessel endothelial hyaluronan receptor-1 (LYVE1)/podoplanin to detect lymphatic vessels. We could show, that in the human episclera distinct CD31 þ blood vessels are present as early as week of gestation 13. Their amount increased during pregnancy, whereas stromal CD31 þ blood vessels were elevated in early pregnancy and regressed with ongoing pregnancy. In the lamina fusca CD31 þ blood vessels were absent at any time point investigated. Single LYVE1 þ cells were identified primarily in the episclera; their amount decreased significantly with increasing gestational ages (group 1 compared to group 3: p < 0.01). However, LYVE1þ/podoplanin þ lymphatic vessels were not detectable in the sclera at any gestational ages analyzed. In contrast to the conjunctiva where LYVE1þ/podoplanin þ lymphatic vessels were detectable as early as week 17, the amount of LYVE1 þ cells in the sclera was highest in early pregnancy (group 1), with a significant decrease during continuing pregnancy (p < 0.001). These findings are the first evidence for a fetal lymphangiogenic privilege of the sclera and show, that the fetal human sclera contains CD31 þ blood vessels, but is primarily alymphatic. Our findings suggest a strong expression of selectively antilymphangiogenic factors, making the developing sclera a potential model to discern antilymphangiogenic mechanisms. © 2014 Elsevier Ltd. All rights reserved.
Keywords: sclera development neovascularization angiogenesis lymphangiogenesis fetal
1. Introduction The human sclera is the rigid collagen cover of the eye with a specialized vascular architecture. Besides the episcleral vessel net and some perforating blood vessels, this tissue is relatively avascular. In adults, most of the scleral blood vessels are surrounded by lymphatic vessel endothelial hyaluronan receptor 1 (LYVE1)þ/
* Corresponding author. Tel.: þ49 221 478 4313; fax: þ49 221 478 5094. E-mail addresses: [email protected] (S.L. Schlereth), [email protected] (B. Neuser), [email protected] (M.C. Herwig), [email protected] (A.M. Müller), [email protected] (K.R. Koch), [email protected] (H.A. Reitsamer), [email protected] €dl), [email protected] (C. Cursiefen), Ludwig.Heindl@uk-koeln. (F. Schro de (L.M. Heindl). http://dx.doi.org/10.1016/j.exer.2014.06.010 0014-4835/© 2014 Elsevier Ltd. All rights reserved.
CD68 þ macrophages (Schlereth et al., 2014). In contrast to blood vessels, the healthy adult human sclera is free of lymphatic vessels, as our group could show recently (Schlereth et al., 2014). During embryogenesis blood vessel development within the choroid and retina is well described (Saint-Geniez and D'Amore, 2004), but only little attention has been paid to scleral vasculogenesis, which is therefore nearly unknown. This is not only the case for scleral blood vessels, but also for lymphatic vessels. Whether or when the lymphangiogenic privilege, which has been shown for adults, is achieved during development is not known yet. Other structures in the eye, for example the lens or the hyaloid body contain blood vessels during development, and become avascular by regression of these vessels later in development. For example, in mice the vasa hyaloidea propria disappears between
204
S.L. Schlereth et al. / Experimental Eye Research 125 (2014) 203e209
day 12 and 16 (Ito and Yoshioka, 1999) and lately it has been shown, that this regression is regulated by a melanopsin-dependent light response (Rao et al., 2013). In contrast, other structures of the eye are primarily alymphatic, e.g. the cornea (Cursiefen et al., 2006b). For the developing fetus, it is not known, whether the fetal sclera is primarily alymphatic or if there is a regression of lymphatic vessels during embryogenesis. Time point and mode of development, as well as possible topography of lymphatic vessels are unknown. This question has important implications, as the adult sclera is selectively alymphatic, but not avascular. During pathologic conditions such as trauma (Wessel et al., 2012) or tumor (Heindl et al., 2009), lymphatic ingrowth can occur, which in case of ocular melanoma is associated with higher recurrence rate, higher metastasis rate and lower survival rate (Heindl et al., 2011b, 2010c). Therefore, we wanted to study the fetal human sclera and investigate blood and lymphatic vessel existence and location as a basis for future experiments, investigating molecular mechanisms, which are selectively inhibiting lymphangiogenesis. The identification of responsible factors might have clinical implications for other diseases involving lymphangiogenesis. 2. Material and methods 2.1. Human fetal bulbus donors Human eyes from stillborn fetuses and abortions between 12 and 38 weeks of gestation (WoG) (n ¼ 34; mean: 21.7 weeks, of both sex: 19 male, 15 female) were obtained from the Division of Ophthalmic Pathology, University Eye Clinic Bonn, according to the Declaration of Helsinki, with informed consent. Both, the University of Bonn and the University of Cologne had approval of the local Ethics committees. 2.2. Formalin fixated paraffin embedded scleral probes Eyes were fixed in phosphate buffered saline (PBS) containing 4% formaldehyde and prepared for paraffin embedding. Subsequently pupil-optic disc sections of 4 mm thickness were obtained and analyzed at anterior, equatorial, and posterior location. Following deparaffinization via series of graded alcohols, immunohistochemistry was performed as described previously (Schlereth et al., 2014). Briefly, sections were incubated with CD31 (monoclonal mouse anti human antibody, Dako, Hamburg, Germany, diluted 1:50 in PBS), lymphatic vessel endothelial hyaluronan receptor 1 (LYVE1 e polyclonal rabbit anti human antibody, Zytomed, Berlin, Germany, diluted 1:50 in PBS), podoplanin (monoclonal mouse anti human D2-40 antibody, Dako, ready-touse) or CD68 (monoclonal mouse anti human antibody, Dako, diluted 1:100 in PBS) for 60 min for LYVE1 and 30 min for CD31, podoplanin and CD 68 at 20 C. For CD31, LYVE1, and podoplanin labeling, antigen retrieval was required (20 min at 98 C). CD68 was pretreated with Fast Enzyme (DCS diagnostics, Hamburg, Germany) for 10 min. DCS Detection line (DCS diagnostics) was performed according to manufacturer's instructions. After incubation with the primary antibody, the secondary antibody was added on the sections for 10 min at 20 C and was marked with a peroxidase-Label (HRP). AEC (3-Amino-9-ethylcarbazole) was used as a chromogen. The sections were counterstained with hemalaun. To confirm the normal anatomy of the eye, the sections were stained with hematoxylin & eosin (H&E). To exclude an unspecific secondary antibody binding, controls were performed by omission of the primary antibodies and resulted in no staining. Conjunctiva and placenta (for CD68) specimens were used as positive controls and showed appropriate results (i.e.
immunoreactivity in conjunctival and placental macrophages (Vinnars et al., 2010); not shown). Our definition of lymphatic vessels required the co-localization of the two established lympathic markers LYVE1 and podoplanin and a visible vessel lumen (Bock et al., 2013). Our definition of blood vessels required immunoreactivity for the endothelial marker CD31 and a visible vessel lumen. 2.3. Morphometric and statistical analyses The fetal human sclera and conjunctiva were analyzed immunhistochemically for the existence of blood (CD31þ) and lymphatic vessels (LYVE1þ/podoplaninþ) at anterior, equatorial, and posterior location in a blinded fashion by two independent persons. The probes were subdivided into three groups: 1) 12e18 WoG, n ¼ 10; 2) 19e23 WoG, n ¼ 13; and 3) 24e38 WoG, n ¼ 11. Sections were analyzed using a Leica DM2500 microscope (Leica GmbH, Wetzlar, Germany) and documented digitally using a JVC digital camera KY-F75U (JVC, Yokohama, Japan). In all gestational groups, CD31 þ vessels and LYVE1 þ immunoreactive cells were quantified at the different locations (episclera-anterior, -equatorial, -posterior and scleral stroma-anterior, -equatorial and posterior and conjunctiva). Therefore, pictures of 0.8 mm2 size were taken from each localization. The amount of cells or vessels detected in the pictures was graded into four levels. The grading score applied as follows: 0 for negative result (no LYVE1 þ cells or CD31 þ vessel in the picture), 1 for 1e2 LYVE1 þ cells/1e2 CD31 þ vessels, 2 for 3e5 LYVE1 þ cells/3e5 CD31 þ vessels and 3 for more than 5 LYVE1 þ cells/>5 CD31 þ vessels) by two independent persons (Table 1). The number of CD31 þ blood vessels or LYVE1 þ cells was evaluated in relation to the gestational age. Statistical analysis was performed using the Mann-Whitney-UTest and the Wilcoxon Test in SPSS software (IBM Corporation, Armonk, NY, USA). P-values< 0.05 were considered statistically significant. 3. Results 3.1. The human scleral stroma contains CD31 þ blood vessels as early as week of gestation 13 To study fetal hemangiogenesis, formalin embedded paraffinfixated eyes of 34 human fetuses were investigated for agedependent and location-specific differences in CD31 þ blood vessels. In general, we detected blood vessels as early as week 13 in the episclera, the limbus and the conjuctiva. At the same time, the stroma displayed a reduced amount of blood vessels and the lamina fusca did not show any blood vessels at all time points. The number of episcleral blood vessels increased during development in all locations investigated from group 1 (anterior episclera score: 2.1 ± 1 equatorial episclera score: 0.5 ± 0.9 and posterior episclera score: 1.3 ± 1) to a higher score in group 3 (anterior episclera score: Table 1 Scoring system used for analysis of LYVE1 þ cells and CD31 þ blood vessels within sclera or conjunctiva.
No LYVE1 þ cells or no CD31 þ vessels Low amount of LYVE1 þ cells or CD31 þ vessels Moderate amount of LYVE1 þ cells or CD31 þ vessels High amounts of LYVE1 þ cells or CD31 þ vessels
Definition
Score
Absent 1e2 cells/1e2 vessels
0 1
3e5 cells/3e5 vessels
2
>5 cells/>5 vessels
3
S.L. Schlereth et al. / Experimental Eye Research 125 (2014) 203e209
2.6 ± 0.6, equatorial episclera score: 1.4 ± 1 and posterior episclera score: 1.8 ± 0.7). The number of anteriorly localized stromal blood vessels, which were slightly more numerous (score: 0.9 ± 1.2) compared to other stromal locations in group 1, regressed during the ongoing development to a score of zero (Fig. 1 and Fig. 2). Meanwhile, in the conjunctiva the number of CD31 þ blood vessels increased steadily from group 1 (score 1.8 ± 1.2) to group 3 (score 2.8 ± 0.4) (p < 0.05) (Fig. 2). In the limbal area the number of blood vessels was high as early as week 13 (score 2 ± 1 in group 1) and stayed at a similar level from this point on (score 2.3 ± 0.9 in group 3). The cornea was free of blood vessels at all time points (data not shown). Although the episcleral CD31 þ blood vessel score was higher than the stromal blood vessel score at all time points, in week of gestation (WoG) 12e18 the scleral stroma, contained blood vessels (score 0.9 ± 1.2), which has not been described in the stroma of the adult sclera. Beginning from week 19, the differences
205
between episclera and stroma increased significantly (group 2: anteriorly p < 0.001, equatorially p < 0.01, posteriorly p < 0.01; group 3: anteriorly p < 0.001, equatorially p < 0.05, posteriorly p < 0.05) (Fig. 2). Summarizing, the fetal sclera contained blood vessels as early as week 13, mainly located in the episclera. The number of blood vessels increased in all areas analyzed, except for the scleral stroma, which contained a few blood vessels in the first group that regressed within the stroma with ongoing pregnancy. 3.2. The human sclera is primarily devoid of LYVE1þ/ podoplanin þ lymphatic vessels LYVE1þ/podoplanin þ lymphatic vessels with a lumen were not detected in any scleral specimens at any stage of gestation (Fig. 3). These were clearly detectable in the conjunctiva as early as week 17 (Fig. 3). In some probes single LYVE1 þ cells were detectable. These
Fig. 1. CD31 þ blood vessel development in the fetal human episclera and stroma Immunohistochemistry of CD31 þ blood vessels (brown diaminobenzidine (DAB) reaction product, arrows) shows age and location dependent differences in scleral blood vessel development. We analyzed three age dependent groups (differences in weeks of gestation (WoG)) and location (episclera, stroma) at anterior, equatorial and posterior anatomy. Scale bar indicates 100 mm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
206
S.L. Schlereth et al. / Experimental Eye Research 125 (2014) 203e209 4
**
3,5
conjunctiva
*
**
episclera stroma
CD31+ blood vessel score
3
2,5
2
1,5
1
0,5
0 anterior
equatorial
posterior
anterior
WoG 12-18
equatorial WoG 19-23
posterior
anterior
equatorial
posterior
WoG 24-38
Fig. 2. Conjunctiva and anterior episclera showed the highest amount of CD31 þ blood vessels during embryologic development Analysis of immunohistochemistry revealed significant differences in the amount of conjunctival, episcleral and scleral stromal blood vessels. The amount of CD31 þ blood vessels in anterior stromal location in WoG 12e18 decreased with ongoing pregnancy to zero in WoG 19e38, whereas the amount of episcleral and conjunctival blood vessels increased anteriorly and equatorially during gestation. (*p < 0.05, **p < 0.01).
slides were additionally stained for CD68 (a macrophage marker) as serial sections. Only very few CD68 þ cells were detectable, mainly located in the conjunctiva (data not shown). Serial sections performed for LYVE1 and CD68, revealed that only a minority of CD68 þ cells were also positive for LYVE1 (4 out of 16 CD68 þ cells in 36 slides e data not shown). 3.3. The amount of LYVE1 þ cells is low in the sclera and high in the conjunctiva in WoG 12e18 Quantification of LYVE1 þ cells in the conjunctiva and sclera revealed only a very small number in the sclera and a higher numbers in the conjunctiva (Figs. 3 and 4). In the episclera the number of these cells decreased from posterior (score 1.4 ± 0.5) towards anterior (score 0.9 ± 0.7), and even more towards the stroma to less than one third (score 0.5 ± 0.9) anteriorly, equatorially and posteriorly in group 1. There were more cells in the anterior episclera (score 0.9 ± 0.7) detectable, compared to the anterior stroma (score 0.2 ± 0.6) (Fig. 4). At later stages of pregnancy the number of cells deceased in all areas analyzed and reached a minimum in WoG 24e38. Significant differences between early pregnancy (group 1) and late pregnancy (group 3) were detected in the episclera (anterior: p < 0.01, equatorial: p ¼ 0.01, posterior: p ¼ 0.001), within the conjunctiva (group 1 vs 3: p < 0.001), and between episclera and stroma in early pregnancy (p < 0.05 in all locations) (Fig. 4). LYVE1 þ cells in the sclera were often seen in close relation to the blood vessels (Fig 3 magnification box). In conclusion, the highest number of LYVE1 þ cells was counted in early pregnancy within the conjunctiva and among scleral locations at the posterior episclera. However, the number of LYVE1 þ cells decreased in all areas analyzed and was nearly absent in WoG 24e38. 4. Discussion The adult sclera is devoid of lymphatic vessels but contains blood vessels mainly in the episclera (Schlereth et al., 2014). This is
in contrast to the adult human cornea, which lacks both blood- and lymphatic vessels (Cursiefen et al., 2002). The healthy avascular cornea belongs to the few angiogenically privileged and thereby immunologically privileged tissues of the human body (Cursiefen, 2007; Hos and Cursiefen, 2014). Whereas several factors preserving avascularity have been identified in the avascular cornea (Bucher et al., 2012; Cursiefen et al., 2006a, 2006b; Hos and Cursiefen, 2014; Regenfuss et al., 2012), the mechanisms maintaining the alymphatic nature of the sclera under physiological conditions are poorly understood. Weather these mechanisms start in early fetal development was not known yet and whether and when the sclera develops blood or lymphatic vessels has not been described either. Our study revealed three major results: i) The human sclera is free of LYVE1þ/podoplanin þ lymphatic vessels during embryologic development. ii) The human sclera contains CD31 þ blood vessels as early as week 13. iii) Singular LYVE1 þ cells are detectable in the developing sclera and their amount decreases during pregnancy. Here, we provide novel evidence that the fetal human sclera is primarily devoid of lymphatic vessels, but not of blood vessels. The fetal scleral stroma contained blood vessels, whereas in the adulthood the stroma is free of blood vessels except for perforating vessels. During development of the sclera, the blood vessels are more spread in the episclera and anterior stroma. Since we could not detect a single blood vessel in the lamina fusca, we therefore conclude that the blood vessels are not growing through the lamina fusca. The function of intrastromal blood vessels can only be speculated. They might be necessary within the stroma for the nutrition of a high amount of cells during the development. One explanatory approach could be that the choroidal or conjunctival vessels are not sufficient yet to nourish the sclera. However, vasculogenesis occurs early in ocular development. We detected established CD31 þ blood vessels with lumen in week 13, indicating that
S.L. Schlereth et al. / Experimental Eye Research 125 (2014) 203e209
207
Fig. 3. LYVE1 þ cells were detected in the sclera and conjunctiva, but only in the conjunctiva lymphangiogenesis occurred, whereas the sclera was primarily alymphatic. Immunohistochemistry of LYVE1 þ cells showed the absence of lumen containing lymphatic vessels at any time point and location investigated within the fetal sclera. LYVE1 þ cells were seen mainly located in the episclera (marked with white arrowheads), adjacent to blood vessels (LYVE1 negative with lumen and eventually erythrocytes, marked with black arrowhead). With ongoing pregnancy less LYVE1 þ cells were seen in the sclera. In the conjunctiva, in WoG 12e18 many singular LYVE1 þ cells were detected (marked with gray arrowhead), and beginning from WoG 19, LYVE1 þ lumen-containing lymphatic vessels were visible within the conjunctiva (marked with black arrows). Note that the number of LYVE1 þ cells decreased with occurring lymphatic vessels. We analyzed three age dependent groups (differences in weeks of gestation (WoG)) and location (episclera, stroma) at anterior, equatorial and posterior anatomy, as well as conjunctiva. Scale bar indicates 100 mm.
vasculogenesis of scleral and conjunctival blood vessels takes place before week 13 and therefore earlier than the induction of lymphatic vessels. According to our findings and analogically to other anatomical sites of the human body, the development of blood and lymphatic vessel is not necessarily connected to each other. In the anterior eye three tissues with diverging vascularization are neighboring each other: 1. The cornea, primarily blood and lymphatic vessel free (Cursiefen et al., 2006b), 2. the conjunctiva with both blood and
lymphatic vessels and 3. the sclera with only blood but no lymphatic vessels. Because we were not able to analyze eyes earlier than WoG 12, we cannot rule out the theoretical possibility of very early lymphatic vessels in the sclera that already regressed until week 12. Interestingly, the number of LYVE1þ/CD68 þ cells is very low in the fetal sclera, compared to the adult sclera, where the blood vessels are surrounded by LYVE1þ/CD68 þ macrophages (Schlereth et al., 2014). But even in the fetal sclera LYVE1þ cells are located
208
S.L. Schlereth et al. / Experimental Eye Research 125 (2014) 203e209
*** **
3,5 3
**
***
*
*
LYVE1 + cell score
2,5
conjunctiva episclera
2
stroma
1,5 1 0,5 0 anterior equatorial posterior
anterior equatorial posterior
anterior equatorial posterior
WoG 12-18
WoG 19-23
WoG 24-38
Fig. 4. Regression of LYVE1 þ cells during fetal development in conjunctiva and sclera Analysis of immunohistochemistry revealed higher numbers of singular LYVE1 þ cells in early pregnancy compared to late pregnancy in conjunctiva and different scleral locations. (*p < 0.05, **p < 0.01, ***p < 0.001).
adjacent to blood vessels. This may indicate, that these cells migrate out of blood vessels and then are regulated by tissue specific factors, to further create de novo lymphatic vessels in the conjunctiva or being inhibited in the sclera. It may also indicate, that the colocalization of CD68 and LYVE1 is specific for the adult sclera, as we could barely detect LYVE1þ/CD68 þ cells in fetal tissue. Therefore fetal LYVE1þ/CD68 negative cells should be distinguished from LYVE1þ/CD68 þ cells seen in adulthood. It can be speculated, that LYVE1þ/CD68- cells are lymphatic endothelial cells, migrating through the tissue and if stimulated by tissue factors are able to induce lymphangiogenesis. On the other hand, LYVE1þ CD68 þ macrophages and LYVE1-/CD68 þ macrophages might be involved in regulating and promoting lymphangiogenesis (Nucera et al., 2011) or to regulate lymphatic vessel caliber during development, as it has been demonstrated in the skin (Gordon et al., 2010). This is in line with our observations, that most of the CD68 þ cells were located in the lymphatic-containing conjunctiva. In tumors and wound healing LYVE1 expressing macrophages have been detected (Schledzewski et al., 2006) and in many tumor entities a significant negative correlation has been shown between LYVE1 and the prognosis quoad vitam (Heindl et al., 2010a, 2010b, 2011a, 2009, 2010b, 2011a; Miyahara et al., 2007). Interestingly, Nucera et al. report several similarities of tumor associated macrophages and embryonic macrophages (Nucera et al., 2011), for example an overlapping gene expression profile (Pucci et al., 2009) and direct interaction with endothelial cells to facilitate blood vessel growth (De Palma et al., 2005). In the developing conjunctiva higher numbers of LYVE1 þ cells can be detected in WoG 12e18 than in later stages (WoG 19e38). The number of these cells decreases while an increase of LYVE1þ or podoplanin þ lymphatic vessels occurs at the same time, indicating, that these cells are potentially involved in the lymphvasculogenesis of the conjunctiva. A similar observation had been made in the cornea for CD11b þ macrophages, which can become integral components of newly formed lymphatic vessels while producing high amounts of angiogenic factors (Maruyama et al., 2005). In the sclera however, the respective LYVE1þ cells are rare and their lymphvasculogenic potential is eventually inhibited in the scleral microenvironment. In another collagen rich tissue, such as the knee cartilage, the absence of lymphatic vessels in 12 and 14 week old human knee
joint (Melrose and Little, 2010) has been described, similar to our results. This group also detected CD68 þ macrophages and the authors concluded, that the macrophages are involved in the clearance of metabolic waste (Melrose and Little, 2010), which could also be the case for scleral tissue. The macrophage marker CD68 might be necessary for immune modulating functions of these cells. The fetal immune system differs from the adult, which has been studied e.g. in corneal allograft rejection (Schwartzkopff et al., 2010a, 2010b). Here a not fully established immune privilege has been ascribed, due to the infant immune system, leading to higher rejection rates of corneal transplants in newborns. If LYVE1þ/ CD68 þ macrophages are involved in this process is not known and needs further investigation, but might be a hint towards a better understanding of the reduced immune privilege of the fetal and infant eye. As others have shown, there is a significant number of micro-RNA sequences or gene expression levels, which show agerelated differences in expression levels in fetal and adult eyes, e.g. microRNA involved in collagen gene regulation or transforming growth factor beta1 (TGF-b1) (Metlapally et al., 2013; Zhou et al., 2006). 5. Conclusion In summary we showed for the first time, that the fetal human sclera is free of lymphatic vessels, but not of blood vessels. Scleral, limbal and conjunctival hemangiogenesis begins before WoG 13, and therefore earlier than conjunctival lymphangiogenesis. Scleral hemangiogenesis is increased in early pregnancy and regresses with ongoing development in the scleral stroma. In contrast to CD31 þ blood vessels, the scleral lymphangiogenic privilege therefore already exists in utero. This suggests either a lack of prolymphangiogenic factors or a strong expression of selectively antilymphangiogenic factors. If the latter applied, the developing sclera would be a good model (Regenfuss et al., 2010) to identify antilymphangiogenic mechanisms. Conflict of interest disclosures The authors declare that they have no conflict of interest.
S.L. Schlereth et al. / Experimental Eye Research 125 (2014) 203e209
Funding/support This work was supported by the German Research Foundation (HE 6743/2-1 to L.M.H.; Priority Research Project SFB 643: B10, CU 47/6-1, CU 47/4-1 to C.C.), the European Union (FP7 STRONG to C.C.), the University of Cologne (GEROK program to S.L.S., K.R.K. and L.M.H.), and the Austrian Research Promotion Agency (FFG-830770 to F.S.). Acknowledgments We thank Martina Becker for great support in the field of immunohistochemistry. References Bock, F., Maruyama, K., Regenfuss, B., Hos, D., Steven, P., Heindl, L.M., Cursiefen, C., 2013. Novel anti(lymph)angiogenic treatment strategies for corneal and ocular surface diseases. Prog. Retin. Eye Res. 34, 89e124. Bucher, F., Parthasarathy, A., Bergua, A., Onderka, J., Regenfuss, B., Cursiefen, C., Bock, F., 2012. Topical ranibizumab inhibits inflammatory corneal hem- and lymphangiogenesis. Acta Ophthalmol. 92, 143e148. Cursiefen, C., 2007. Immune privilege and angiogenic privilege of the cornea. Chem. Immunol. Allergy 92, 50e57. Cursiefen, C., Chen, L., Saint-Geniez, M., Hamrah, P., Jin, Y., Rashid, S., Pytowski, B., Persaud, K., Wu, Y., Streilein, J.W., Dana, R., 2006a. Nonvascular VEGF receptor 3 expression by corneal epithelium maintains avascularity and vision. Proc. Natl. Acad. Sci. U. S. A. 103, 11405e11410. Cursiefen, C., Rummelt, C., Junemann, A., Vorwerk, C., Neuhuber, W., Kruse, F.E., Schroedl, F., 2006b. Absence of blood and lymphatic vessels in the developing human cornea. Cornea 25, 722e726. Cursiefen, C., Schlotzer-Schrehardt, U., Kuchle, M., Sorokin, L., Breiteneder-Geleff, S., Alitalo, K., Jackson, D., 2002. Lymphatic vessels in vascularized human corneas: immunohistochemical investigation using LYVE-1 and podoplanin. Invest. Ophthalmol. Vis. Sci. 43, 2127e2135. De Palma, M., Venneri, M.A., Galli, R., Sergi Sergi, L., Politi, L.S., Sampaolesi, M., Naldini, L., 2005. Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell 8, 211e226. Gordon, E.J., Rao, S., Pollard, J.W., Nutt, S.L., Lang, R.A., Harvey, N.L., 2010. Macrophages define dermal lymphatic vessel calibre during development by regulating lymphatic endothelial cell proliferation. Development 137, 3899e3910. Heindl, L.M., Hofmann, T.N., Adler, W., Knorr, H.L., Holbach, L.M., Naumann, G.O., Kruse, F.E., Cursiefen, C., 2010a. Intraocular tumor-associated lymphangiogenesis a novel prognostic factor for ciliary body melanomas with extraocular extension? Ophthalmology 117, 334e342. Heindl, L.M., Hofmann, T.N., Knorr, H.L., Rummelt, C., Schrodl, F., SchlotzerSchrehardt, U., Holbach, L.M., Naumann, G.O., Kruse, F.E., Cursiefen, C., 2009. Intraocular lymphangiogenesis in malignant melanomas of the ciliary body with extraocular extension. Investig. Ophthalmol. Vis. Sci. 50, 1988e1995. Heindl, L.M., Hofmann, T.N., Schrodl, F., Holbach, L.M., Kruse, F.E., Cursiefen, C., 2010b. Intraocular lymphatics in ciliary body melanomas with extraocular extension: functional for lymphatic spread? Arch. Ophthalmol. 128, 1001e1008. Heindl, L.M., Hofmann-Rummelt, C., Adler, W., Bosch, J.J., Holbach, L.M., Naumann, G.O., Kruse, F.E., Cursiefen, C., 2011a. Prognostic significance of tumor-associated lymphangiogenesis in malignant melanomas of the conjunctiva. Ophthalmology 118, 2351e2360. Heindl, L.M., Hofmann-Rummelt, C., Adler, W., Bosch, J.J., Holbach, L.M., Naumann, G.O., Kruse, F.E., Cursiefen, C., 2011b. Tumor-associated lymphangiogenesis in the development of conjunctival melanoma. Investig. Ophthalmol. Vis. Sci. 52, 7074e7083.
209
Heindl, L.M., Hofmann-Rummelt, C., Adler, W., Holbach, L.M., Naumann, G.O., Kruse, F.E., Cursiefen, C., 2010c. Tumor-associated lymphangiogenesis in the development of conjunctival squamous cell carcinoma. Ophthalmology 117, 649e658. Hos, D., Cursiefen, C., 2014. Lymphatic vessels in the development of tissue and organ rejection. Adv. Anat. Embryol., Cell. Biol. 214, 119e141. Ito, M., Yoshioka, M., 1999. Regression of the hyaloid vessels and pupillary membrane of the mouse. Anat. Embryol. 200, 403e411. Maruyama, K., Ii, M., Cursiefen, C., Jackson, D.G., Keino, H., Tomita, M., Van Rooijen, N., Takenaka, H., D'Amore, P.A., Stein-Streilein, J., Losordo, D.W., Streilein, J.W., 2005. Inflammation-induced lymphangiogenesis in the cornea arises from CD11b-positive macrophages. J. Clin. Invest. 115, 2363e2372. Melrose, J., Little, C.B., 2010. Immunolocalization of lymphatic vessels in human fetal knee joint tissues. Connect. tissue Res. 51, 289e305. Metlapally, R., Gonzalez, P., Hawthorne, F.A., Tran-Viet, K.N., Wildsoet, C.F., Young, T.L., 2013. Scleral micro-RNA signatures in adult and fetal eyes. PloS one 8, e78984. Miyahara, M., Tanuma, J., Sugihara, K., Semba, I., 2007. Tumor lymphangiogenesis correlates with lymph node metastasis and clinicopathologic parameters in oral squamous cell carcinoma. Cancer 110, 1287e1294. Nucera, S., Biziato, D., De Palma, M., 2011. The interplay between macrophages and angiogenesis in development, tissue injury and regeneration. Int. J. Dev. Biol. 55, 495e503. Pucci, F., Venneri, M.A., Biziato, D., Nonis, A., Moi, D., Sica, A., Di Serio, C., Naldini, L., De Palma, M., 2009. A distinguishing gene signature shared by tumorinfiltrating Tie2-expressing monocytes, blood “resident” monocytes, and embryonic macrophages suggests common functions and developmental relationships. Blood 114, 901e914. Rao, S., Chun, C., Fan, J., Kofron, J.M., Yang, M.B., Hegde, R.S., Ferrara, N., Copenhagen, D.R., Lang, R.A., 2013. A direct and melanopsin-dependent fetal light response regulates mouse eye development. Nature 494, 243e246. Regenfuss, B., Bock, F., Cursiefen, C., 2012. Corneal angiogenesis and lymphangiogenesis. Curr. Opin. allergy Clin. Immunol. 12, 548e554. Regenfuss, B., Onderka, J., Bock, F., Hos, D., Maruyama, K., Cursiefen, C., 2010. Genetic heterogeneity of lymphangiogenesis in different mouse strains. Am. J. Pathol. 177, 501e510. Saint-Geniez, M., D'Amore, P.A., 2004. Development and pathology of the hyaloid, choroidal and retinal vasculature. Int. J. Dev. Biol. 48, 1045e1058. Schledzewski, K., Falkowski, M., Moldenhauer, G., Metharom, P., Kzhyshkowska, J., Ganss, R., Demory, A., Falkowska-Hansen, B., Kurzen, H., Ugurel, S., Geginat, G., Arnold, B., Goerdt, S., 2006. Lymphatic endothelium-specific hyaluronan receptor LYVE-1 is expressed by stabilin-1þ, F4/80þ, CD11bþ macrophages in malignant tumours and wound healing tissue in vivo and in bone marrow cultures in vitro: implications for the assessment of lymphangiogenesis. J. Pathol. 209, 67e77. Schlereth, S.L., Neuser, B., Caramoy, A., Grajewski, R.S., Koch, K.R., Schrodl, F., Cursiefen, C., Heindl, L.M., 2014. Enrichment of lymphatic vessel endothelial hyaluronan receptor 1 (LYVE1)-Positive macrophages around blood vessels in the Normal human sclera. Invest. Ophthalmol. Vis. Sci. 55, 865e872. Schwartzkopff, J., Berger, M., Birnbaum, F., Bohringer, D., Reinhard, T., 2010a. Accelerated corneal graft rejection in baby rats. Br. J. Ophthalmol. 94, 1062e1066. Schwartzkopff, J., Schlereth, S.L., Berger, M., Bredow, L., Birnbaum, F., Bohringer, D., Reinhard, T., 2010b. NK cell depletion delays corneal allograft rejection in baby rats. Mol. Vis. 16, 1928e1935. Vinnars, M.T., Rindsjo, E., Ghazi, S., Sundberg, A., Papadogiannakis, N., 2010. The number of CD68(þ) (Hofbauer) cells is decreased in placentas with chorioamnionitis and with advancing gestational age. Pediatr. Dev. Pathol. : Official J. Soc. Pediatr. Pathol. Paediatr. Pathol. Soc. 13, 300e304. Wessel, J.M., Hofmann-Rummelt, C., Kruse, F.E., Cursiefen, C., Heindl, L.M., 2012. Invasion of lymphatic vessels into the eye after open globe injuries. Invest. Ophthalmol. Vis. Sci. 53, 3717e3725. Zhou, J., Rappaport, E.F., Tobias, J.W., Young, T.L., 2006. Differential gene expression in mouse sclera during ocular development. Invest. Ophthalmol. Vis. Sci. 47, 1794e1802.
Contents lists available at ScienceDirect
Experimental Eye Research journal homepage: www.elsevier.com/locate/yexer
Absence of lymphatic vessels in the developing human sclera Simona L. Schlereth a, *, Barbara Neuser a, Martina C. Herwig b, Annette M. Müller c, € dl d, Claus Cursiefen a, Konrad R. Koch a, Herbert A. Reitsamer d, Falk Schro Ludwig M. Heindl a a
Department of Ophthalmology, University of Cologne, Kerpenerstr. 62, 50924 Cologne, Germany Department of Ophthalmology, University of Bonn, Ernst-Abbe-Str. 2, 53127 Bonn, Germany c Center of Pediatric Pathology and Pathology, MVZ Venusberg, University of Bonn, Sigmund-Freud-Str. 25, 53127 Bonn, Germany d Department of Ophthalmology and Anatomy, Paracelsus Medical University, Strubergasse 21, 5020 Salzburg, Austria b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 2 March 2014 Accepted in revised form 13 June 2014 Available online 27 June 2014
The adult sclera is free of lymphatic vessels, but contains a net of blood vessels. Whether and when this selectively lymphangiogenic privilege is achieved during embryologic development is not known yet. Therefore, we investigated the developing human sclera for blood- and lymphatic vessels in 34 abortions/stillborns (12e38 weeks of gestation). The probes were subdivided into three groups (group 1: 12 e18 weeks of gestation, n ¼ 10; group 2: 19e23 weeks of gestation, n ¼ 13; group 3: 24e38 weeks of gestation, n ¼ 11), and prepared for paraffin sections followed by immunohistochemistry against CD31 to detect blood vessels, and against lymphatic vessel endothelial hyaluronan receptor-1 (LYVE1)/podoplanin to detect lymphatic vessels. We could show, that in the human episclera distinct CD31 þ blood vessels are present as early as week of gestation 13. Their amount increased during pregnancy, whereas stromal CD31 þ blood vessels were elevated in early pregnancy and regressed with ongoing pregnancy. In the lamina fusca CD31 þ blood vessels were absent at any time point investigated. Single LYVE1 þ cells were identified primarily in the episclera; their amount decreased significantly with increasing gestational ages (group 1 compared to group 3: p < 0.01). However, LYVE1þ/podoplanin þ lymphatic vessels were not detectable in the sclera at any gestational ages analyzed. In contrast to the conjunctiva where LYVE1þ/podoplanin þ lymphatic vessels were detectable as early as week 17, the amount of LYVE1 þ cells in the sclera was highest in early pregnancy (group 1), with a significant decrease during continuing pregnancy (p < 0.001). These findings are the first evidence for a fetal lymphangiogenic privilege of the sclera and show, that the fetal human sclera contains CD31 þ blood vessels, but is primarily alymphatic. Our findings suggest a strong expression of selectively antilymphangiogenic factors, making the developing sclera a potential model to discern antilymphangiogenic mechanisms. © 2014 Elsevier Ltd. All rights reserved.
Keywords: sclera development neovascularization angiogenesis lymphangiogenesis fetal
1. Introduction The human sclera is the rigid collagen cover of the eye with a specialized vascular architecture. Besides the episcleral vessel net and some perforating blood vessels, this tissue is relatively avascular. In adults, most of the scleral blood vessels are surrounded by lymphatic vessel endothelial hyaluronan receptor 1 (LYVE1)þ/
* Corresponding author. Tel.: þ49 221 478 4313; fax: þ49 221 478 5094. E-mail addresses: [email protected] (S.L. Schlereth), [email protected] (B. Neuser), [email protected] (M.C. Herwig), [email protected] (A.M. Müller), [email protected] (K.R. Koch), [email protected] (H.A. Reitsamer), [email protected] €dl), [email protected] (C. Cursiefen), Ludwig.Heindl@uk-koeln. (F. Schro de (L.M. Heindl). http://dx.doi.org/10.1016/j.exer.2014.06.010 0014-4835/© 2014 Elsevier Ltd. All rights reserved.
CD68 þ macrophages (Schlereth et al., 2014). In contrast to blood vessels, the healthy adult human sclera is free of lymphatic vessels, as our group could show recently (Schlereth et al., 2014). During embryogenesis blood vessel development within the choroid and retina is well described (Saint-Geniez and D'Amore, 2004), but only little attention has been paid to scleral vasculogenesis, which is therefore nearly unknown. This is not only the case for scleral blood vessels, but also for lymphatic vessels. Whether or when the lymphangiogenic privilege, which has been shown for adults, is achieved during development is not known yet. Other structures in the eye, for example the lens or the hyaloid body contain blood vessels during development, and become avascular by regression of these vessels later in development. For example, in mice the vasa hyaloidea propria disappears between
204
S.L. Schlereth et al. / Experimental Eye Research 125 (2014) 203e209
day 12 and 16 (Ito and Yoshioka, 1999) and lately it has been shown, that this regression is regulated by a melanopsin-dependent light response (Rao et al., 2013). In contrast, other structures of the eye are primarily alymphatic, e.g. the cornea (Cursiefen et al., 2006b). For the developing fetus, it is not known, whether the fetal sclera is primarily alymphatic or if there is a regression of lymphatic vessels during embryogenesis. Time point and mode of development, as well as possible topography of lymphatic vessels are unknown. This question has important implications, as the adult sclera is selectively alymphatic, but not avascular. During pathologic conditions such as trauma (Wessel et al., 2012) or tumor (Heindl et al., 2009), lymphatic ingrowth can occur, which in case of ocular melanoma is associated with higher recurrence rate, higher metastasis rate and lower survival rate (Heindl et al., 2011b, 2010c). Therefore, we wanted to study the fetal human sclera and investigate blood and lymphatic vessel existence and location as a basis for future experiments, investigating molecular mechanisms, which are selectively inhibiting lymphangiogenesis. The identification of responsible factors might have clinical implications for other diseases involving lymphangiogenesis. 2. Material and methods 2.1. Human fetal bulbus donors Human eyes from stillborn fetuses and abortions between 12 and 38 weeks of gestation (WoG) (n ¼ 34; mean: 21.7 weeks, of both sex: 19 male, 15 female) were obtained from the Division of Ophthalmic Pathology, University Eye Clinic Bonn, according to the Declaration of Helsinki, with informed consent. Both, the University of Bonn and the University of Cologne had approval of the local Ethics committees. 2.2. Formalin fixated paraffin embedded scleral probes Eyes were fixed in phosphate buffered saline (PBS) containing 4% formaldehyde and prepared for paraffin embedding. Subsequently pupil-optic disc sections of 4 mm thickness were obtained and analyzed at anterior, equatorial, and posterior location. Following deparaffinization via series of graded alcohols, immunohistochemistry was performed as described previously (Schlereth et al., 2014). Briefly, sections were incubated with CD31 (monoclonal mouse anti human antibody, Dako, Hamburg, Germany, diluted 1:50 in PBS), lymphatic vessel endothelial hyaluronan receptor 1 (LYVE1 e polyclonal rabbit anti human antibody, Zytomed, Berlin, Germany, diluted 1:50 in PBS), podoplanin (monoclonal mouse anti human D2-40 antibody, Dako, ready-touse) or CD68 (monoclonal mouse anti human antibody, Dako, diluted 1:100 in PBS) for 60 min for LYVE1 and 30 min for CD31, podoplanin and CD 68 at 20 C. For CD31, LYVE1, and podoplanin labeling, antigen retrieval was required (20 min at 98 C). CD68 was pretreated with Fast Enzyme (DCS diagnostics, Hamburg, Germany) for 10 min. DCS Detection line (DCS diagnostics) was performed according to manufacturer's instructions. After incubation with the primary antibody, the secondary antibody was added on the sections for 10 min at 20 C and was marked with a peroxidase-Label (HRP). AEC (3-Amino-9-ethylcarbazole) was used as a chromogen. The sections were counterstained with hemalaun. To confirm the normal anatomy of the eye, the sections were stained with hematoxylin & eosin (H&E). To exclude an unspecific secondary antibody binding, controls were performed by omission of the primary antibodies and resulted in no staining. Conjunctiva and placenta (for CD68) specimens were used as positive controls and showed appropriate results (i.e.
immunoreactivity in conjunctival and placental macrophages (Vinnars et al., 2010); not shown). Our definition of lymphatic vessels required the co-localization of the two established lympathic markers LYVE1 and podoplanin and a visible vessel lumen (Bock et al., 2013). Our definition of blood vessels required immunoreactivity for the endothelial marker CD31 and a visible vessel lumen. 2.3. Morphometric and statistical analyses The fetal human sclera and conjunctiva were analyzed immunhistochemically for the existence of blood (CD31þ) and lymphatic vessels (LYVE1þ/podoplaninþ) at anterior, equatorial, and posterior location in a blinded fashion by two independent persons. The probes were subdivided into three groups: 1) 12e18 WoG, n ¼ 10; 2) 19e23 WoG, n ¼ 13; and 3) 24e38 WoG, n ¼ 11. Sections were analyzed using a Leica DM2500 microscope (Leica GmbH, Wetzlar, Germany) and documented digitally using a JVC digital camera KY-F75U (JVC, Yokohama, Japan). In all gestational groups, CD31 þ vessels and LYVE1 þ immunoreactive cells were quantified at the different locations (episclera-anterior, -equatorial, -posterior and scleral stroma-anterior, -equatorial and posterior and conjunctiva). Therefore, pictures of 0.8 mm2 size were taken from each localization. The amount of cells or vessels detected in the pictures was graded into four levels. The grading score applied as follows: 0 for negative result (no LYVE1 þ cells or CD31 þ vessel in the picture), 1 for 1e2 LYVE1 þ cells/1e2 CD31 þ vessels, 2 for 3e5 LYVE1 þ cells/3e5 CD31 þ vessels and 3 for more than 5 LYVE1 þ cells/>5 CD31 þ vessels) by two independent persons (Table 1). The number of CD31 þ blood vessels or LYVE1 þ cells was evaluated in relation to the gestational age. Statistical analysis was performed using the Mann-Whitney-UTest and the Wilcoxon Test in SPSS software (IBM Corporation, Armonk, NY, USA). P-values< 0.05 were considered statistically significant. 3. Results 3.1. The human scleral stroma contains CD31 þ blood vessels as early as week of gestation 13 To study fetal hemangiogenesis, formalin embedded paraffinfixated eyes of 34 human fetuses were investigated for agedependent and location-specific differences in CD31 þ blood vessels. In general, we detected blood vessels as early as week 13 in the episclera, the limbus and the conjuctiva. At the same time, the stroma displayed a reduced amount of blood vessels and the lamina fusca did not show any blood vessels at all time points. The number of episcleral blood vessels increased during development in all locations investigated from group 1 (anterior episclera score: 2.1 ± 1 equatorial episclera score: 0.5 ± 0.9 and posterior episclera score: 1.3 ± 1) to a higher score in group 3 (anterior episclera score: Table 1 Scoring system used for analysis of LYVE1 þ cells and CD31 þ blood vessels within sclera or conjunctiva.
No LYVE1 þ cells or no CD31 þ vessels Low amount of LYVE1 þ cells or CD31 þ vessels Moderate amount of LYVE1 þ cells or CD31 þ vessels High amounts of LYVE1 þ cells or CD31 þ vessels
Definition
Score
Absent 1e2 cells/1e2 vessels
0 1
3e5 cells/3e5 vessels
2
>5 cells/>5 vessels
3
S.L. Schlereth et al. / Experimental Eye Research 125 (2014) 203e209
2.6 ± 0.6, equatorial episclera score: 1.4 ± 1 and posterior episclera score: 1.8 ± 0.7). The number of anteriorly localized stromal blood vessels, which were slightly more numerous (score: 0.9 ± 1.2) compared to other stromal locations in group 1, regressed during the ongoing development to a score of zero (Fig. 1 and Fig. 2). Meanwhile, in the conjunctiva the number of CD31 þ blood vessels increased steadily from group 1 (score 1.8 ± 1.2) to group 3 (score 2.8 ± 0.4) (p < 0.05) (Fig. 2). In the limbal area the number of blood vessels was high as early as week 13 (score 2 ± 1 in group 1) and stayed at a similar level from this point on (score 2.3 ± 0.9 in group 3). The cornea was free of blood vessels at all time points (data not shown). Although the episcleral CD31 þ blood vessel score was higher than the stromal blood vessel score at all time points, in week of gestation (WoG) 12e18 the scleral stroma, contained blood vessels (score 0.9 ± 1.2), which has not been described in the stroma of the adult sclera. Beginning from week 19, the differences
205
between episclera and stroma increased significantly (group 2: anteriorly p < 0.001, equatorially p < 0.01, posteriorly p < 0.01; group 3: anteriorly p < 0.001, equatorially p < 0.05, posteriorly p < 0.05) (Fig. 2). Summarizing, the fetal sclera contained blood vessels as early as week 13, mainly located in the episclera. The number of blood vessels increased in all areas analyzed, except for the scleral stroma, which contained a few blood vessels in the first group that regressed within the stroma with ongoing pregnancy. 3.2. The human sclera is primarily devoid of LYVE1þ/ podoplanin þ lymphatic vessels LYVE1þ/podoplanin þ lymphatic vessels with a lumen were not detected in any scleral specimens at any stage of gestation (Fig. 3). These were clearly detectable in the conjunctiva as early as week 17 (Fig. 3). In some probes single LYVE1 þ cells were detectable. These
Fig. 1. CD31 þ blood vessel development in the fetal human episclera and stroma Immunohistochemistry of CD31 þ blood vessels (brown diaminobenzidine (DAB) reaction product, arrows) shows age and location dependent differences in scleral blood vessel development. We analyzed three age dependent groups (differences in weeks of gestation (WoG)) and location (episclera, stroma) at anterior, equatorial and posterior anatomy. Scale bar indicates 100 mm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
206
S.L. Schlereth et al. / Experimental Eye Research 125 (2014) 203e209 4
**
3,5
conjunctiva
*
**
episclera stroma
CD31+ blood vessel score
3
2,5
2
1,5
1
0,5
0 anterior
equatorial
posterior
anterior
WoG 12-18
equatorial WoG 19-23
posterior
anterior
equatorial
posterior
WoG 24-38
Fig. 2. Conjunctiva and anterior episclera showed the highest amount of CD31 þ blood vessels during embryologic development Analysis of immunohistochemistry revealed significant differences in the amount of conjunctival, episcleral and scleral stromal blood vessels. The amount of CD31 þ blood vessels in anterior stromal location in WoG 12e18 decreased with ongoing pregnancy to zero in WoG 19e38, whereas the amount of episcleral and conjunctival blood vessels increased anteriorly and equatorially during gestation. (*p < 0.05, **p < 0.01).
slides were additionally stained for CD68 (a macrophage marker) as serial sections. Only very few CD68 þ cells were detectable, mainly located in the conjunctiva (data not shown). Serial sections performed for LYVE1 and CD68, revealed that only a minority of CD68 þ cells were also positive for LYVE1 (4 out of 16 CD68 þ cells in 36 slides e data not shown). 3.3. The amount of LYVE1 þ cells is low in the sclera and high in the conjunctiva in WoG 12e18 Quantification of LYVE1 þ cells in the conjunctiva and sclera revealed only a very small number in the sclera and a higher numbers in the conjunctiva (Figs. 3 and 4). In the episclera the number of these cells decreased from posterior (score 1.4 ± 0.5) towards anterior (score 0.9 ± 0.7), and even more towards the stroma to less than one third (score 0.5 ± 0.9) anteriorly, equatorially and posteriorly in group 1. There were more cells in the anterior episclera (score 0.9 ± 0.7) detectable, compared to the anterior stroma (score 0.2 ± 0.6) (Fig. 4). At later stages of pregnancy the number of cells deceased in all areas analyzed and reached a minimum in WoG 24e38. Significant differences between early pregnancy (group 1) and late pregnancy (group 3) were detected in the episclera (anterior: p < 0.01, equatorial: p ¼ 0.01, posterior: p ¼ 0.001), within the conjunctiva (group 1 vs 3: p < 0.001), and between episclera and stroma in early pregnancy (p < 0.05 in all locations) (Fig. 4). LYVE1 þ cells in the sclera were often seen in close relation to the blood vessels (Fig 3 magnification box). In conclusion, the highest number of LYVE1 þ cells was counted in early pregnancy within the conjunctiva and among scleral locations at the posterior episclera. However, the number of LYVE1 þ cells decreased in all areas analyzed and was nearly absent in WoG 24e38. 4. Discussion The adult sclera is devoid of lymphatic vessels but contains blood vessels mainly in the episclera (Schlereth et al., 2014). This is
in contrast to the adult human cornea, which lacks both blood- and lymphatic vessels (Cursiefen et al., 2002). The healthy avascular cornea belongs to the few angiogenically privileged and thereby immunologically privileged tissues of the human body (Cursiefen, 2007; Hos and Cursiefen, 2014). Whereas several factors preserving avascularity have been identified in the avascular cornea (Bucher et al., 2012; Cursiefen et al., 2006a, 2006b; Hos and Cursiefen, 2014; Regenfuss et al., 2012), the mechanisms maintaining the alymphatic nature of the sclera under physiological conditions are poorly understood. Weather these mechanisms start in early fetal development was not known yet and whether and when the sclera develops blood or lymphatic vessels has not been described either. Our study revealed three major results: i) The human sclera is free of LYVE1þ/podoplanin þ lymphatic vessels during embryologic development. ii) The human sclera contains CD31 þ blood vessels as early as week 13. iii) Singular LYVE1 þ cells are detectable in the developing sclera and their amount decreases during pregnancy. Here, we provide novel evidence that the fetal human sclera is primarily devoid of lymphatic vessels, but not of blood vessels. The fetal scleral stroma contained blood vessels, whereas in the adulthood the stroma is free of blood vessels except for perforating vessels. During development of the sclera, the blood vessels are more spread in the episclera and anterior stroma. Since we could not detect a single blood vessel in the lamina fusca, we therefore conclude that the blood vessels are not growing through the lamina fusca. The function of intrastromal blood vessels can only be speculated. They might be necessary within the stroma for the nutrition of a high amount of cells during the development. One explanatory approach could be that the choroidal or conjunctival vessels are not sufficient yet to nourish the sclera. However, vasculogenesis occurs early in ocular development. We detected established CD31 þ blood vessels with lumen in week 13, indicating that
S.L. Schlereth et al. / Experimental Eye Research 125 (2014) 203e209
207
Fig. 3. LYVE1 þ cells were detected in the sclera and conjunctiva, but only in the conjunctiva lymphangiogenesis occurred, whereas the sclera was primarily alymphatic. Immunohistochemistry of LYVE1 þ cells showed the absence of lumen containing lymphatic vessels at any time point and location investigated within the fetal sclera. LYVE1 þ cells were seen mainly located in the episclera (marked with white arrowheads), adjacent to blood vessels (LYVE1 negative with lumen and eventually erythrocytes, marked with black arrowhead). With ongoing pregnancy less LYVE1 þ cells were seen in the sclera. In the conjunctiva, in WoG 12e18 many singular LYVE1 þ cells were detected (marked with gray arrowhead), and beginning from WoG 19, LYVE1 þ lumen-containing lymphatic vessels were visible within the conjunctiva (marked with black arrows). Note that the number of LYVE1 þ cells decreased with occurring lymphatic vessels. We analyzed three age dependent groups (differences in weeks of gestation (WoG)) and location (episclera, stroma) at anterior, equatorial and posterior anatomy, as well as conjunctiva. Scale bar indicates 100 mm.
vasculogenesis of scleral and conjunctival blood vessels takes place before week 13 and therefore earlier than the induction of lymphatic vessels. According to our findings and analogically to other anatomical sites of the human body, the development of blood and lymphatic vessel is not necessarily connected to each other. In the anterior eye three tissues with diverging vascularization are neighboring each other: 1. The cornea, primarily blood and lymphatic vessel free (Cursiefen et al., 2006b), 2. the conjunctiva with both blood and
lymphatic vessels and 3. the sclera with only blood but no lymphatic vessels. Because we were not able to analyze eyes earlier than WoG 12, we cannot rule out the theoretical possibility of very early lymphatic vessels in the sclera that already regressed until week 12. Interestingly, the number of LYVE1þ/CD68 þ cells is very low in the fetal sclera, compared to the adult sclera, where the blood vessels are surrounded by LYVE1þ/CD68 þ macrophages (Schlereth et al., 2014). But even in the fetal sclera LYVE1þ cells are located
208
S.L. Schlereth et al. / Experimental Eye Research 125 (2014) 203e209
*** **
3,5 3
**
***
*
*
LYVE1 + cell score
2,5
conjunctiva episclera
2
stroma
1,5 1 0,5 0 anterior equatorial posterior
anterior equatorial posterior
anterior equatorial posterior
WoG 12-18
WoG 19-23
WoG 24-38
Fig. 4. Regression of LYVE1 þ cells during fetal development in conjunctiva and sclera Analysis of immunohistochemistry revealed higher numbers of singular LYVE1 þ cells in early pregnancy compared to late pregnancy in conjunctiva and different scleral locations. (*p < 0.05, **p < 0.01, ***p < 0.001).
adjacent to blood vessels. This may indicate, that these cells migrate out of blood vessels and then are regulated by tissue specific factors, to further create de novo lymphatic vessels in the conjunctiva or being inhibited in the sclera. It may also indicate, that the colocalization of CD68 and LYVE1 is specific for the adult sclera, as we could barely detect LYVE1þ/CD68 þ cells in fetal tissue. Therefore fetal LYVE1þ/CD68 negative cells should be distinguished from LYVE1þ/CD68 þ cells seen in adulthood. It can be speculated, that LYVE1þ/CD68- cells are lymphatic endothelial cells, migrating through the tissue and if stimulated by tissue factors are able to induce lymphangiogenesis. On the other hand, LYVE1þ CD68 þ macrophages and LYVE1-/CD68 þ macrophages might be involved in regulating and promoting lymphangiogenesis (Nucera et al., 2011) or to regulate lymphatic vessel caliber during development, as it has been demonstrated in the skin (Gordon et al., 2010). This is in line with our observations, that most of the CD68 þ cells were located in the lymphatic-containing conjunctiva. In tumors and wound healing LYVE1 expressing macrophages have been detected (Schledzewski et al., 2006) and in many tumor entities a significant negative correlation has been shown between LYVE1 and the prognosis quoad vitam (Heindl et al., 2010a, 2010b, 2011a, 2009, 2010b, 2011a; Miyahara et al., 2007). Interestingly, Nucera et al. report several similarities of tumor associated macrophages and embryonic macrophages (Nucera et al., 2011), for example an overlapping gene expression profile (Pucci et al., 2009) and direct interaction with endothelial cells to facilitate blood vessel growth (De Palma et al., 2005). In the developing conjunctiva higher numbers of LYVE1 þ cells can be detected in WoG 12e18 than in later stages (WoG 19e38). The number of these cells decreases while an increase of LYVE1þ or podoplanin þ lymphatic vessels occurs at the same time, indicating, that these cells are potentially involved in the lymphvasculogenesis of the conjunctiva. A similar observation had been made in the cornea for CD11b þ macrophages, which can become integral components of newly formed lymphatic vessels while producing high amounts of angiogenic factors (Maruyama et al., 2005). In the sclera however, the respective LYVE1þ cells are rare and their lymphvasculogenic potential is eventually inhibited in the scleral microenvironment. In another collagen rich tissue, such as the knee cartilage, the absence of lymphatic vessels in 12 and 14 week old human knee
joint (Melrose and Little, 2010) has been described, similar to our results. This group also detected CD68 þ macrophages and the authors concluded, that the macrophages are involved in the clearance of metabolic waste (Melrose and Little, 2010), which could also be the case for scleral tissue. The macrophage marker CD68 might be necessary for immune modulating functions of these cells. The fetal immune system differs from the adult, which has been studied e.g. in corneal allograft rejection (Schwartzkopff et al., 2010a, 2010b). Here a not fully established immune privilege has been ascribed, due to the infant immune system, leading to higher rejection rates of corneal transplants in newborns. If LYVE1þ/ CD68 þ macrophages are involved in this process is not known and needs further investigation, but might be a hint towards a better understanding of the reduced immune privilege of the fetal and infant eye. As others have shown, there is a significant number of micro-RNA sequences or gene expression levels, which show agerelated differences in expression levels in fetal and adult eyes, e.g. microRNA involved in collagen gene regulation or transforming growth factor beta1 (TGF-b1) (Metlapally et al., 2013; Zhou et al., 2006). 5. Conclusion In summary we showed for the first time, that the fetal human sclera is free of lymphatic vessels, but not of blood vessels. Scleral, limbal and conjunctival hemangiogenesis begins before WoG 13, and therefore earlier than conjunctival lymphangiogenesis. Scleral hemangiogenesis is increased in early pregnancy and regresses with ongoing development in the scleral stroma. In contrast to CD31 þ blood vessels, the scleral lymphangiogenic privilege therefore already exists in utero. This suggests either a lack of prolymphangiogenic factors or a strong expression of selectively antilymphangiogenic factors. If the latter applied, the developing sclera would be a good model (Regenfuss et al., 2010) to identify antilymphangiogenic mechanisms. Conflict of interest disclosures The authors declare that they have no conflict of interest.
S.L. Schlereth et al. / Experimental Eye Research 125 (2014) 203e209
Funding/support This work was supported by the German Research Foundation (HE 6743/2-1 to L.M.H.; Priority Research Project SFB 643: B10, CU 47/6-1, CU 47/4-1 to C.C.), the European Union (FP7 STRONG to C.C.), the University of Cologne (GEROK program to S.L.S., K.R.K. and L.M.H.), and the Austrian Research Promotion Agency (FFG-830770 to F.S.). Acknowledgments We thank Martina Becker for great support in the field of immunohistochemistry. References Bock, F., Maruyama, K., Regenfuss, B., Hos, D., Steven, P., Heindl, L.M., Cursiefen, C., 2013. Novel anti(lymph)angiogenic treatment strategies for corneal and ocular surface diseases. Prog. Retin. Eye Res. 34, 89e124. Bucher, F., Parthasarathy, A., Bergua, A., Onderka, J., Regenfuss, B., Cursiefen, C., Bock, F., 2012. Topical ranibizumab inhibits inflammatory corneal hem- and lymphangiogenesis. Acta Ophthalmol. 92, 143e148. Cursiefen, C., 2007. Immune privilege and angiogenic privilege of the cornea. Chem. Immunol. Allergy 92, 50e57. Cursiefen, C., Chen, L., Saint-Geniez, M., Hamrah, P., Jin, Y., Rashid, S., Pytowski, B., Persaud, K., Wu, Y., Streilein, J.W., Dana, R., 2006a. Nonvascular VEGF receptor 3 expression by corneal epithelium maintains avascularity and vision. Proc. Natl. Acad. Sci. U. S. A. 103, 11405e11410. Cursiefen, C., Rummelt, C., Junemann, A., Vorwerk, C., Neuhuber, W., Kruse, F.E., Schroedl, F., 2006b. Absence of blood and lymphatic vessels in the developing human cornea. Cornea 25, 722e726. Cursiefen, C., Schlotzer-Schrehardt, U., Kuchle, M., Sorokin, L., Breiteneder-Geleff, S., Alitalo, K., Jackson, D., 2002. Lymphatic vessels in vascularized human corneas: immunohistochemical investigation using LYVE-1 and podoplanin. Invest. Ophthalmol. Vis. Sci. 43, 2127e2135. De Palma, M., Venneri, M.A., Galli, R., Sergi Sergi, L., Politi, L.S., Sampaolesi, M., Naldini, L., 2005. Tie2 identifies a hematopoietic lineage of proangiogenic monocytes required for tumor vessel formation and a mesenchymal population of pericyte progenitors. Cancer Cell 8, 211e226. Gordon, E.J., Rao, S., Pollard, J.W., Nutt, S.L., Lang, R.A., Harvey, N.L., 2010. Macrophages define dermal lymphatic vessel calibre during development by regulating lymphatic endothelial cell proliferation. Development 137, 3899e3910. Heindl, L.M., Hofmann, T.N., Adler, W., Knorr, H.L., Holbach, L.M., Naumann, G.O., Kruse, F.E., Cursiefen, C., 2010a. Intraocular tumor-associated lymphangiogenesis a novel prognostic factor for ciliary body melanomas with extraocular extension? Ophthalmology 117, 334e342. Heindl, L.M., Hofmann, T.N., Knorr, H.L., Rummelt, C., Schrodl, F., SchlotzerSchrehardt, U., Holbach, L.M., Naumann, G.O., Kruse, F.E., Cursiefen, C., 2009. Intraocular lymphangiogenesis in malignant melanomas of the ciliary body with extraocular extension. Investig. Ophthalmol. Vis. Sci. 50, 1988e1995. Heindl, L.M., Hofmann, T.N., Schrodl, F., Holbach, L.M., Kruse, F.E., Cursiefen, C., 2010b. Intraocular lymphatics in ciliary body melanomas with extraocular extension: functional for lymphatic spread? Arch. Ophthalmol. 128, 1001e1008. Heindl, L.M., Hofmann-Rummelt, C., Adler, W., Bosch, J.J., Holbach, L.M., Naumann, G.O., Kruse, F.E., Cursiefen, C., 2011a. Prognostic significance of tumor-associated lymphangiogenesis in malignant melanomas of the conjunctiva. Ophthalmology 118, 2351e2360. Heindl, L.M., Hofmann-Rummelt, C., Adler, W., Bosch, J.J., Holbach, L.M., Naumann, G.O., Kruse, F.E., Cursiefen, C., 2011b. Tumor-associated lymphangiogenesis in the development of conjunctival melanoma. Investig. Ophthalmol. Vis. Sci. 52, 7074e7083.
209
Heindl, L.M., Hofmann-Rummelt, C., Adler, W., Holbach, L.M., Naumann, G.O., Kruse, F.E., Cursiefen, C., 2010c. Tumor-associated lymphangiogenesis in the development of conjunctival squamous cell carcinoma. Ophthalmology 117, 649e658. Hos, D., Cursiefen, C., 2014. Lymphatic vessels in the development of tissue and organ rejection. Adv. Anat. Embryol., Cell. Biol. 214, 119e141. Ito, M., Yoshioka, M., 1999. Regression of the hyaloid vessels and pupillary membrane of the mouse. Anat. Embryol. 200, 403e411. Maruyama, K., Ii, M., Cursiefen, C., Jackson, D.G., Keino, H., Tomita, M., Van Rooijen, N., Takenaka, H., D'Amore, P.A., Stein-Streilein, J., Losordo, D.W., Streilein, J.W., 2005. Inflammation-induced lymphangiogenesis in the cornea arises from CD11b-positive macrophages. J. Clin. Invest. 115, 2363e2372. Melrose, J., Little, C.B., 2010. Immunolocalization of lymphatic vessels in human fetal knee joint tissues. Connect. tissue Res. 51, 289e305. Metlapally, R., Gonzalez, P., Hawthorne, F.A., Tran-Viet, K.N., Wildsoet, C.F., Young, T.L., 2013. Scleral micro-RNA signatures in adult and fetal eyes. PloS one 8, e78984. Miyahara, M., Tanuma, J., Sugihara, K., Semba, I., 2007. Tumor lymphangiogenesis correlates with lymph node metastasis and clinicopathologic parameters in oral squamous cell carcinoma. Cancer 110, 1287e1294. Nucera, S., Biziato, D., De Palma, M., 2011. The interplay between macrophages and angiogenesis in development, tissue injury and regeneration. Int. J. Dev. Biol. 55, 495e503. Pucci, F., Venneri, M.A., Biziato, D., Nonis, A., Moi, D., Sica, A., Di Serio, C., Naldini, L., De Palma, M., 2009. A distinguishing gene signature shared by tumorinfiltrating Tie2-expressing monocytes, blood “resident” monocytes, and embryonic macrophages suggests common functions and developmental relationships. Blood 114, 901e914. Rao, S., Chun, C., Fan, J., Kofron, J.M., Yang, M.B., Hegde, R.S., Ferrara, N., Copenhagen, D.R., Lang, R.A., 2013. A direct and melanopsin-dependent fetal light response regulates mouse eye development. Nature 494, 243e246. Regenfuss, B., Bock, F., Cursiefen, C., 2012. Corneal angiogenesis and lymphangiogenesis. Curr. Opin. allergy Clin. Immunol. 12, 548e554. Regenfuss, B., Onderka, J., Bock, F., Hos, D., Maruyama, K., Cursiefen, C., 2010. Genetic heterogeneity of lymphangiogenesis in different mouse strains. Am. J. Pathol. 177, 501e510. Saint-Geniez, M., D'Amore, P.A., 2004. Development and pathology of the hyaloid, choroidal and retinal vasculature. Int. J. Dev. Biol. 48, 1045e1058. Schledzewski, K., Falkowski, M., Moldenhauer, G., Metharom, P., Kzhyshkowska, J., Ganss, R., Demory, A., Falkowska-Hansen, B., Kurzen, H., Ugurel, S., Geginat, G., Arnold, B., Goerdt, S., 2006. Lymphatic endothelium-specific hyaluronan receptor LYVE-1 is expressed by stabilin-1þ, F4/80þ, CD11bþ macrophages in malignant tumours and wound healing tissue in vivo and in bone marrow cultures in vitro: implications for the assessment of lymphangiogenesis. J. Pathol. 209, 67e77. Schlereth, S.L., Neuser, B., Caramoy, A., Grajewski, R.S., Koch, K.R., Schrodl, F., Cursiefen, C., Heindl, L.M., 2014. Enrichment of lymphatic vessel endothelial hyaluronan receptor 1 (LYVE1)-Positive macrophages around blood vessels in the Normal human sclera. Invest. Ophthalmol. Vis. Sci. 55, 865e872. Schwartzkopff, J., Berger, M., Birnbaum, F., Bohringer, D., Reinhard, T., 2010a. Accelerated corneal graft rejection in baby rats. Br. J. Ophthalmol. 94, 1062e1066. Schwartzkopff, J., Schlereth, S.L., Berger, M., Bredow, L., Birnbaum, F., Bohringer, D., Reinhard, T., 2010b. NK cell depletion delays corneal allograft rejection in baby rats. Mol. Vis. 16, 1928e1935. Vinnars, M.T., Rindsjo, E., Ghazi, S., Sundberg, A., Papadogiannakis, N., 2010. The number of CD68(þ) (Hofbauer) cells is decreased in placentas with chorioamnionitis and with advancing gestational age. Pediatr. Dev. Pathol. : Official J. Soc. Pediatr. Pathol. Paediatr. Pathol. Soc. 13, 300e304. Wessel, J.M., Hofmann-Rummelt, C., Kruse, F.E., Cursiefen, C., Heindl, L.M., 2012. Invasion of lymphatic vessels into the eye after open globe injuries. Invest. Ophthalmol. Vis. Sci. 53, 3717e3725. Zhou, J., Rappaport, E.F., Tobias, J.W., Young, T.L., 2006. Differential gene expression in mouse sclera during ocular development. Invest. Ophthalmol. Vis. Sci. 47, 1794e1802.
