Fibronectin‑integrin signaling regulates PLVAP localization at endothelial fenestrae by microtubule stabilization

Takashi Nakakura1 · Takeshi Suzuki2 · Kotaro Horiguchi3 · Hideyuki Tanaka1 · Kenjiro Arisawa1 · Toshio Miyashita1 ·
Yoko Nekooki‑Machida1 · Haruo Hagiwara1

Received: 10 September 2020 / Accepted: 22 October 2020
© Springer-Verlag GmbH Germany, part of Springer Nature 2021

Abstract
Endothelial fenestrae are the transcellular pores existing on the capillary walls which are organized in clusters referred to as sieve plates. They are also divided by a diaphragm consisting of plasmalemma vesicle-associated protein (PLVAP). In this study, we examined the involvement of fibronectin signaling in the formation of fenestra and diaphragm in endothelial cells. Results showed that Itga5 and Itgb1 were expressed in PECAM1-positive endothelial cells isolated from the anterior lobe (AL) of the rat pituitary, and integrin α5 was localized at the fenestrated capillaries of the rat pituitary and cultured PECAM1-positive endothelial cells isolated from AL (CECAL). Inhibition of both integrin α5β1 and FAK, a key molecule for integrin-microtubule signaling, respectively, by ATN-161 and FAK inhibitor 14, caused the delocalization of PLVAP at the sieve plates and depolymerization of microtubules in CECAL. Paclitaxel prevented the delocalization of PLVAP by the inhibition of integrin α5β1. Microtubule depolymerization induced by colcemid also caused the delocalization of PLVAP. Treatment of CECAL with ATN-161 and colcemid caused PLVAP localization at the Golgi apparatus. The localization of PLVAP at the sieve plates was inhibited by BFA treatment in a time-dependent manner and spread diffusely to the cytoplasm. These results indicate that a constant supply of PLVAP proteins by the endomembrane system via the Golgi apparatus is essential for the localization of PLVAP at sieve plates. In conclusion, the endomembrane transport pathway from the Golgi apparatus to sieve plates requires microtubule cytoskeletons, which are regulated by fibronectin-integrin α5β1 signaling.

Keywords Fibronectin · Plasmalemma vesicle-associated protein (PLVAP) · Endothelial fenestra · Microtubule · Rat pituitary

Abbreviations ECM Extracellular matrix

AL
BFA
BM
BSA
Anterior lobe Brefeldin A Basement membrane
Bovine serum albumin
ECMs
FBS
GP
IgG
Extracellular matrices Fetal bovine serum Guinea pig Immunoglobulin G

CECAL Cultured endothelial cells isolated from the rat AL
PBS
PCR
Phosphate-buffered saline Polymerase chain reaction

DAPI 4′,6-Diamidino-2-phenylindole
DMSO Dimethyl sulfoxide
PECAM1 Platelet endothelial cell adhesion molecule
PFA Paraformadehyde
PLVAP Plasmalemma vesicle-associated protein

*

[email protected]
RT
SEM
Room temperature
Scanning electron microscopy

1Department of Anatomy and Cell Biology, Teikyo University School of Medicine, Tokyo, Japan
2Department of Biology, Sapporo Medical University, Sapporo, Japan
3Laboratory of Anatomy and Cell Biology, Department of Health Sciences, Kyorin University, Tokyo, Japan
VEGF-A Vascular endothelial growth factor-A

Vol.:(0123456789)

Introduction

In mammals, the pituitary is a central endocrine organ composed of the anterior, intermediate, and posterior lobes. Among them, the anterior lobe (AL) is well vascularized by a hypophyseal portal system, which consists of the hypophyseal portal vessels and the fenestrated capillary plexus distributed in the AL and the median eminence of the hypothalamus (Farquhar 1961; Murakami et al. 1987). Our previous studies have shown that vascular endothelial growth factor-A (VEGF-A), a major secreted angiogenic factor, is involved in formation of the hypophyseal portal system (Nakakura et al. 2006; Tanaka et al. 2013). Fenestrae present in the capillary endothelial cells in endocrine glands are essential for the homeostasis of the endocrine system because they act as a channel for peptide hormones and other substances. Fenestrae are the transcellular pores existing on the capillary wall, which are typically 60–80 nm in diameter, and are organized as a cluster of pores referred to as sieve plate (Farquhar 1961; Nakakura et al. 2020; Rhodin 1962). Each fenestral pore is further divided into 5–6-nm openings by a pinwheel-like diaphragm consisting of plasmalemma vesicle-associated protein (PLVAP), which is a type II transmembrane glycoprotein (Stan et al. 1999). PLVAP is the only known molecular component of the fenestral diaphragm and is essential for diaphragm formation (Herrnberger et al. 2012; Stan et al. 2012).
Capillaries are composed of endothelial cells and mural cells such as pericytes and smooth muscle cells, and the basal surfaces are lined by basement membrane (BM) which consists of extracellular matrices (ECMs) such as type IV collagen, laminin-411, laminin-511, and fibronectin (Marchand et al. 2019). These ECM molecules bind to the heterodimeric transmembrane receptors, which consist of two subunits (α and β) of integrin molecules, and are involved in the proper capillary morphogenesis and maintenance of vascular homeostasis (Hynes 2002). Fibronectin, a 440–560 kDa dimeric glycoprotein produced by numerous cell types such as mural cells and fibroblasts (Xu and Shi 2014), is essential for cell adhesion, migration, growth, and differentiation (Pankov and Yamada 2002). Fibronectin also plays an important role in cardiovascular development by binding to integrin α5β1, a major fibronectin receptor, on the endothelial cell membrane (Astrof and Hynes 2009; Francis et al. 2002; George et al. 1997). We have recently shown that fibronectin is necessary to maintain the fenestrae and sieve plates on cultured endothelial cells that are isolated from rat AL (CECAL) (Nakakura et al. 2020). However, the function of fibronectin and its intracellular signaling targets such as integrins is not well understood.

A cytoskeleton consisting of microtubules, actin filaments, and intermediate filaments contributes to the regulation of cell morphology, motility, and many other cellular processes. Among them, the microtubules are stabilized by interactions with cellular structures including the cell cortex (Gundersen et al. 2004). They are also responsible for the transport and distribution of various cargoes, such as intracellular vesicles and organelles, the segregation of chromosomes during cell division, and the maintenance of polarity in migrating and epithelial cells (Alfaro-Aco and Petry 2015). Growth and stabilization of microtubules are regulated by integrin-mediated intracellular signaling in addition to a complex network of microtubule- associated proteins (Etienne-Manneville and Hall 2001; LaFlamme et al. 2018).
This study aimed to clarify the relationship between fibronectin-integrin signaling and microtubules in fenestrated endothelial cells of rat AL. We first immunohistochemically confirmed the localization of fibronectin in the rat pituitary and examined the effect of fibronectin coating on the formation and maintenance of sieve plates in the CECAL. We then examined the involvement of fibronectin-integrin α5β1 signaling and microtubules in the formation of fenestra and diaphragm of CECAL by treatment with various inhibitors. We found that integrin signaling is necessary to maintain the microtubule cytoskeleton and localization of PLVAP at the sieve plate in CECAL. The involvement of microtubule cytoskeletons suggests that continuous transport of PLVAP proteins is important for its functional localization in the cell. Therefore, we investigated the transport pathway of PLVAP proteins by monitoring the endomembrane system through inhibition experiments using brefeldin A (BFA) and visualizing the Golgi apparatus by immunofluorescence staining for GM130. The results showed that constant supply of PLVAP proteins by the endomembrane system through the microtubule cytoskeleton is essential for the localization of PLVAP at the sieve plate in CECAL.

Materials and Methods

Animal

Eight-week-old male Wistar rats were purchased from Japan SLC, Inc. (Shizuoka, Japan). All animals were maintained in a temperature-controlled room (22 ± 2 °C) with automatically controlled lighting (light on from 0600 to 1800 hours, daily) and were supplied with food and water ad libitum. All animal experiments were conducted in compliance with the Guide for Care and Use of Laboratory Animals established by Teikyo University.

Immunofluorescence microscopy

For immunohistochemistry, the paraffin blocks of rat pituitary were prepared according to previously described methods (Nakakura et al. 2017a). Deparaffinized sections were heated in an autoclave in 1 mM EDTA at 121 °C for 5 min and were then incubated overnight at 25 °C with rabbit anti-fibronectin (1:2000; 15613-1-AP; Proteintech, IL, USA), goat anti-type IV collagen (1:100; AB769; Merck Millipore, MA, USA), rabbit anti-integrin α5 (1:250; A19069; ABclonal, Tokyo, Japan), and guinea pig (GP) anti-PLVAP (1:5000; Nakakura et al. 2020) in phosphate- buffered saline (PBS) containing 1% bovine serum albumin (BSA). After washing with PBS, the sections were incubated with Cy3-labeled donkey anti-rabbit immunoglobulin G (IgG; Jackson Immunoresearch, PA, USA) and Alexa Fluor 488-labeled donkey anti-GP IgG (Jackson Immunoresearch), and 4′,6-diamidino-2-phenylindole (DAPI; Dojindo, Kumamoto, Japan) for 2 h at 25 °C.
For immunofluorescence staining, CECAL was fixed with 4% paraformaldehyde (PFA) in PBS for 15 min at 25 °C. Immunostaining for cultured cells was performed as previously described (Nakakura et al. 2017b). The cells were then incubated with GP anti-PLVAP (1:2000), rabbit anti-α-tubulin (1:2000; PM054; MBL, Aichi, Japan), rabbit anti-GM130 (1:2000; PM061; MBL), and acti- stain™ 555 phalloidin (1:100; Cytoskeleton, CO, USA) for counterstaining of actin filaments. The samples were analyzed using a confocal laser scanning microscope system A1 (Nikon, Tokyo, Japan).

Isolation and culture of PECAM1‑immunopositive endothelial cells of rat anterior pituitary

The ALs were dissected from male rats, and their cells were dispersed as described previously (Horiguchi et al. 2016). Briefly, for the binding of antibodies, Dynabeads- labeled sheep-anti mouse IgG (Thermo Fisher Scientific, MA, USA) was reacted with mouse anti-rat platelet

endothelial cell adhesion molecule 1 (PECAM1; clone TLD-3A12; GTX74899; GeneTex, CA, USA) in isolation buffer (PBS, 0.1% BSA, 2 mM EDTA) overnight at 4 °C as described previously (Nakakura et al. 2020). After washing with isolation buffer, the dispersed cells were incubated in isolation buffer containing the Dynabeads- labeled PECAM1 antibody for 40 min at 25 °C. PECAM1- immunopositive cells were separated by magnetic stand, seeded in 24- or 48-well plates (10,000 cells/cm2; Nalge Nunc, NY, USA) coated with human fibronectin (Corning, NY, USA), and then cultured in Endothelial Cell Growth Medium MV 2 (ECGM; PromoCell, Heidelberg, Germany) supplemented with antibiotics. To perform the inhibition experiments with the following antagonists, the cells were preincubated in ECGM containing 0.1% BSA without fetal bovine serum (FBS) and growth factors for 2 h, followed by treatment with ATN-161 (1, 10, or 100 nM; Sigma, MO, USA), cilengitide (10 μM; Cayman, MI, USA), paclitaxel (100 nM; Cayman), FAK inhibitor 14 (10 μM; Cayman), colcemid (10, 100, or 1000 fg/ml; Thermo Fisher Scientific), and brefeldin A (BFA, 10 μM; Cayman) for 4 or 20 h. Subsequently, the cells were treated with a fixative solution for morphological analysis or lysis buffer for RNA extraction.

Real‑time polymerase chain reaction

Total RNA was extracted from the cells using the NucleoSpin RNA kit according to the manufacturer’s instructions (Takara, Shiga, Japan). First-strand cDNA was synthesized with PrimeScript RT Master Mix (Takara) using total RNA. Real-time PCR was performed in the 7500 Fast Real-time PCR system (Applied Biosystems, CA, USA) with Brilliant III Ultra-Fast SYBR Green QPCR Master Mix with Low ROX (Agilent Technologies, CA, USA) using gene-specific primers (see Table 1), as previously described (Nakakura et al. 2017b). Cycle threshold values were converted into relative gene expression levels by using the 2-(ΔCt sample-ΔCt control) method.

Table 1 Primer sequences for real-time PCR with SYBR green.

Gene
Plvap

Primer sequence (5′–3′)
5′-TCTTCGTGTCGCTCATCCAG-3′

Product size 130 bp

Accession number NM_020086

5′-TGGCTGTATAGGTTGTCGGC-3′
Itga5 5′-GCAGCTGCCCCAAAAGAAAC-3′ 146 bp NM_001108118
5′-TGTAGAGGACATAGATGAGCAG-3′
Itgb1 5′-GCGGAAGACAAGTGTGTTGA-3′ 127 bp NM_017022
5′-TCACAATGGCACACAGGTTT-3′
Tuba1b 5′-GCCTTCTAACCCGTAGCTATCA-3′ 75 bp NM_001044270
5′-ATTGCCGATCTGGACACC-3′
Hprt1 5′-CTCATGGACTGATTATGGACAGGAC-3′ 123 bp NM_012583
5′-GCAGGTCAGCAAAGAACTTATAGCC-3′

Observation of the cell membrane by scanning electron microscopy

Cells were cultured on fibronectin-coated cover slips and fixed with 2% PFA and 2.5% glutaraldehyde in 0.1 M cacodylate buffer for 1 h at 4 °C, as previously described (Nakakura et al. 2020). After washing with 0.1 M cacodylate buffer, the cells were postfixed with 1% OsO4 in 0.1 M cacodylate buffer and uranyl acetate, followed by dehydration in a series of increasing concentrations of ethanol and t-butyl alcohol. The cellular surface was coated with platinum at 10-nm thickness and observed using a SEM (SU8010; Hitachi High-Technologies Corp., Tokyo, Japan).

Statistical analysis

All data are presented as the mean ± standard error of the mean (SEM). For statistical analysis, differences groups were evaluated with the Mann–Whitney U test or non- repeated measures ANOVA followed by Dunnett’s test. A value of P < 0.05 was considered statistically significant.

Results

Immunolocalization of fibronectin in rat AL and the effect of fibronectin in CECAL

We first examined the distribution of fibronectin in rat AL by immunohistochemistry with antibodies against PLVAP, a marker for endothelial fenestra, or type IV collagen, a marker of BM (Fig. 1a-–f). The capillary BM in the rat AL contained both fibronectin and type IV collagen (Fig. 1a-–c), and PLVAP-positive capillary endothelial cells were pre- sent on the fibronectin-containing BM (Fig. 1d-–f). We have previously shown that PLVAP signal is a good marker of the sieve plates of the fenestrae in the CECAL (Nakakura et al. 2020). To clarify the effect of fibronectin alone on the formation and maintenance of fenestrae and the sieve plates of the CECAL cultured on fibronectin-coated dishes, cells were starved by removal of FBS and growth factors from ECGM medium and analyzed by immunofluorescence stain- ing against PLVAP and F-actin (Fig. 1g-–i). Results showed that PLVAP-positive sieve plate structures were maintained in the CECAL after both 24 and 48 h in the FBS- and growth factor-starved medium.

Gene expression and localization of integrins in endothelial cells of rat AL

Since integrin α5β1 is known as a major fibronectin receptor, we quantitatively analyzed the expression levels of Itga5 and Itgb1 in newly isolated PECAM1-positive and

PECAM1-negative cells from rat AL (Fig. 2 a and b). The gene expression levels of Itga5 and Itgb1 in the isolated PECAM1-positive cells were significantly higher than that in PECAM1-negative cells. We next observed the localization of integrin α5 in the rat AL (Fig. 2c-–e) and the CECAL (Fig. 2f-–h) by double-immunofluorescence staining with the antibodies against PLVAP. Results showed that integrin α5 was located on the cell membrane of the endothelial cells of the rat AL and the CECAL.

Effects of integrin inhibitors on PLVAP distribution and gene expression in CECAL

To clarify the effect of fibronectin-integrin α5β1 signaling on fenestral formation, we treated CECAL with ATN-161, an antagonist of integrin α5β1 and αvβ3, and cilengitide, an antagonist for integrin αvβ3 and αvβ5 (Kapp et al. 2017), followed by examining the localization of PLVAP in CECAL by immunofluorescence staining (Fig. 3). The oval- shaped distribution of PLVAP signals in the CECAL that reflected the structure of the sieve plate were broken down dose-dependently with ATN-161 treatment, which led to the accumulation of PLVAP signals at the perinuclear region of the cells at concentrations of 10 and 100 nM (Fig. 3a–c). However, the distribution of PLVAP was not affected by cilengitide (Fig. 3d). We also performed quantitative expression analysis of Plvap, Itga5, and Itgb1 in ATN-161-treated CECAL by real- time PCR and confirmed that the expression levels of these genes were not affected by ATN-161 treatment (Fig. 3e–g).

Involvement of microtubules in the distribution of PLVAP regulated by fibronectin‑integrin α5β1 signaling

Since integrin signaling is known to regulate microtubule stabilization via focal adhesion kinase (FAK) (Palazzo et al. 2004), we examined the distribution of microtubules in CECAL treated with ATN-161 and FAK inhibitor 14 by immunofluorescence staining against α-tubulin (Fig. 4). Microtubules labeled with anti-α-tubulin antibody were observed in the cytoplasm of control cells (Fig. 4a), but disappeared upon treatment with ATN-161 (Fig. 4b) and FAK inhibitor 14 (Fig. 4c). We next conducted an ATN- 161 inhibition experiment using paclitaxel, a microtubule- stabilizing agent and found that paclitaxel disabled the effect of ATN-161 not only on the microtubule morphology but also on the oval-shaped PLVAP distribution (Fig. 4d).

Relationship between PLVAP distribution and microtubules in CECAL

To elucidate the relationship between PLVAP distribution and microtubules, we examined the effect of colcemid, an

Fig. 1 Immunolocalization of fibronectin in the rat anterior pituitary and effect of fibronectin on the CECAL. a–c Double immunofluores- cence images for fibronectin (a, red) and type IV collagen (b, green) in the rat anterior pituitary are shown. d–f Double immunofluores- cence images for fibronectin (d, red) and PLVAP (e, green) in the rat anterior pituitary are shown. Fibronectin signals were seen spe- cifically in collagen IV-positive extracellular matrix in the rat anterior pituitary, but not in PLVAP-positive endothelial cells (c and f). Nuclei

were counterstained with DAPI (blue). Asterisks denote capillaries. g–i Signals of PLVAP (green) and F-actin (red) in CECAL cultured on fibronectin-coated cover slips are shown. g CECAL cultured for 48 h in ECGM with FBS and growth factors are shown. h, i CECAL cultured for 24 h (h) and 48 h (i) in ECGM without their supplements are shown. Nuclei were counterstained with DAPI (blue). All experi- ments were performed with three animals. Bars 20 μm (a–f), 20 μm (g–i).

inhibitor of microtubule polymerization, on the distribution of PLVAP by immunofluorescence staining. Colcemid eliminated PLVAP localization at the sieve plates in addition to the microtubule filaments in a dose-dependent manner (Fig. 5a-–c). We also quantitatively examined the effects of colcemid on the expression levels of Tuba1b, Plvap, Itga5, and Itgb1 in CECAL by real-time PCR (Fig. 5d-–g). Tuba1b levels were increased significantly by colcemid in a dose- dependent manner (Fig. 5d); however, Plvap, Itga5, and Itgb1 levels remained unchanged (Fig. 5e-–g). To confirm the effect of colcemid on the fenestra structure, we also observed the cell surface of colcemid-treated and control

CECAL using SEM (Fig. 6). Although sieve plate structures containing fenestrae were observed in both colcemid-treated and non-treated CECAL, fenestral diaphragms were only observed in the control CECAL (Fig. 6 a and b) but not in the colcemid-treated CECAL (Fig. 6 c and d).

Intracellular transport of PLVAP through the Golgi apparatus in CECAL

We performed immunofluorescence staining for GM130, a marker of the Golgi apparatus, to confirm whether PLVAP localized at the Golgi apparatus when CECAL was treated

Fig. 2 Expression and immunolocalization of integrin α5β1 in endothelial cells of rat anterior pituitary. a, b The expression levels of Itga5 and Itgb1 in the PECAM1-positive (white bars) and PECAM1- negative (black bars) fractions were determined by real-time PCR, followed by normalization with the expression levels of Hprt1, which was used as the control housekeeping gene. Results of three inde- pendent experiments are shown. Statistical analyses were performed using Mann–Whitney U test. **P < 0.01 vs. PECAM1-positive frac-

tion. c–e Double immunofluorescence images for integrin α5 (c, red) and PLVAP (d, green) in the AL are shown. Integrin α5-positive sig- nals were seen specifically in PLVAP-positive endothelial cells (e). Nuclei were counterstained with DAPI (blue). f–h Signals of integrin α5 (f, red) and PLVAP (g, green) in the CECAL cultured on fibronec- tin-coated cover slips are shown. Integrin α5-positive signals were observed in cell membranes and cytoplasmic regions. Bars 20 μm (c–e), 20 μm (f–h).

with ATN-161 or colcemid (Fig. 7). The results clarified that PLVAP was localized at the Golgi apparatus in addition to the sieve plates in the non-treated CECAL (Fig. 7a). It was also proven that PLVAP was localized at the Golgi apparatus in ATN-161- and colcemid-treated CECAL (Fig. 7 b and c). To investigate the role of PLVAP localized at the Golgi apparatus, we used BFA, which specifically inhibits protein transfer from the endoplasmic reticulum to

the Golgi apparatus (Fujiwara et al. 1988). BFA did not affect the orientation of microtubule cytoskeletons in cells (Fig. 8a, c, e) but inhibited PLVAP localization at sieve plates as well as GM130 localization in the perinuclear space (Fig. 8b, d, f). The PLVAP signals localized at the sieve plates were gradually lost by BFA treatment and spread diffusely to the cytoplasm in a time-dependent manner.

Fig. 3 Immunofluorescence images showing PLVAP localization and changes in
the expression levels of Plvap, Itga5, and Itgb1 in CECAL treated with integrin inhibitors. Signals of PLVAP (a–e, green) and F-actin (a’–e’, red) in CECAL treated with water as control (a–a’’), 1 nM ATN-
161 (b–b’’), 10 nM ATN-161 (c–c’’), 100 nM ATN-161 (d–d’’), and 10 μM cilengitide (e–e’’) were observed. Nuclei were counterstained with DAPI (blue). Bars 20 μm. The expression levels of Plvap (f), Itga5 (g), and Itgb1 (h) were quantitatively estimated by
real-time PCR and normalized by the expression of Hprt1. Results of three independent experiments are shown. Statistical differences were not detected by the non-repeated measures ANOVA followed by Dunnett’s test.

Fig. 4 Immunolocalization of PLVAP and distribution of micro- tubules in CECAL treated with ATN-161, FAK inhibitor 14, and paclitaxel. Signals of PLVAP (a–d, green) and microtubules (a’–d’, red) in CECAL treated with water as control (a–a’’), 10 nM ATN-

161 (b–b’’), 10 μM FAK inhibitor 14 (c–c’’), and 10 nM ATN- 161 + 100 nM paclitaxel (d–d’’) are shown. Nuclei were counter- stained with DAPI (blue). Bars 20 μm.

Discussion

In this study, we showed that fibronectin was localized in the vascular BM of rat AL and was involved in the maintenance of PLVAP-positive sieve plate structures in starved CECAL. We also found that integrin α5β1 was expressed on the cell membrane of rat AL endothelial cells and CECAL, and that the inhibition of integrin α5β1 affects the distribution of PLVAP at sieve plates in CECAL. Genetic studies have reported that the absence

of fibronectin (George et al. 1997, 1993) or integrin α5 (Francis et al. 2002) leads to embryonic lethality by causing the abnormalities of the cardiovascular system. Similarly, endothelial-specific deletion of integrin β1 causes severe vascular defects and lethality (Lei et al. 2008; Tanjore et al. 2008). Therefore, since normal expression of fibronectin and integrin α5β1 is essential for vascular homeostasis, our findings suggest that fibronectin-integrin α5β1 interaction plays an important role in homeostasis of fenestrated endothelial cells in rat AL.

The polymerization and accumulation of α/β-tubulin heterodimers to form microtubules are controlled by various intracellular signals (Akhmanova and Steinmetz 2015). Among them, integrins are known to regulate microtubule nucleating activity (Colello et al. 2012), growth, and stabilization (Etienne-Manneville and Hall 2001). FAK is a cytoplasmic tyrosine kinase and a key molecule in integrin-mediated signal transduction (Zhao and Guan 2011). Integrin-activated FAK regulates microtubule stabilization by facilitating small GTPase Rho signaling (Palazzo et al. 2004). Genetic approaches using endothelial cell-specific FAK knockout and overexpressing mice also indicated that FAK is required for angiogenesis (Peng et al. 2004; Shen et al. 2005). In addition, fibronectin activates FAK by inducing autophosphorylation via binding to integrin (Kim et al. 2001). FAK inhibitor 14 and ATN-161, a specific inhibitor of integrin α5β1, induced microtubule disassembly in CECAL, similar to colcemid, a microtubule-depolymerizing agent. Co-treatment with ATN-161 and paclitaxel, a microtubule-stabilizing drug, did not induce microtubule disassembly. The results indicated that fibronectin-integrin α5β1 signaling regulates microtubule stabilization via the FAK-mediated pathway in the fenestrated endothelial cells in rat AL.
Microtubules are polarized filaments that normally act as a rail for intracellular transport of membrane-bound vesicles containing secretory and membrane proteins (Akhmanova and Steinmetz 2015; Alfaro-Aco and Petry 2015). The intracellular transport of vesicles along microtubules is regulated by motor proteins such as kinesins (Hirokawa et al. 2009). The proteins are synthesized and processed in the rough endoplasmic reticulum and the Golgi apparatus, which are further transported to intracellular places by kinesins. In this study, we showed that the oval-shaped distribution of PLVAP signals in CECAL disappeared after microtubule disassembly. Treatment with colcemid also eliminated the diaphragmed fenestrae in CECAL. Several independent studies on animals with PLVAP deletion have also reported that PLVAP is necessary for the formation of fenestral diaphragms in endothelial cells (Herrnberger et al. 2012; Stan et al. 2012). Thus, it is considered that loss of microtubules induced by inhibition of fibronectin-integrin α5β1 signaling in CECAL reduced the transport of PLVAP to fenestrae, resulting in the disappearance of diaphragms in fenestrae of CECAL. In addition, since PLVAP signals in the GM130-positive Golgi apparatus diffused throughout the cytoplasm by BFA treatment time-dependently, it is considered that the constant supply of PLVAP from the Golgi apparatus by the endomembrane system is essential for its stable localization to the sieve plates. Therefore, we conclude that microtubules are involved in the intracellular transport of PLVAP from the Golgi apparatus to the fenestrae in endothelial cells of rat AL.

VEGF-A is one of the growth factors found in ECGM that is used as a culture medium for CECAL and is important for inducing fenestra formation on the capillary walls of several tissues, such as pancreatic islet and renal glomerulus (Eremina et al. 2003; Lammert et al. 2003; Roberts and Palade 1995). However, in the present study, we showed that fenestrae and sieve plate structures were maintained in the CECAL cultured on fibronectin-coated dishes in the VEGF-A-free medium. This result indicated that fibronectin regulates PLVAP localization by integrin α5β1-mediated microtubule stabilization. Since PLVAP is not involved in fenestrae formation of endothelial cells (Herrnberger et al. 2012; Stan et al. 2012), VEGF-A and fibronectin are unlikely to be directly involved in the fenestra formation of CECAL.
It has been reported that phorbol myristate acetate, an activator of protein kinase C (PKC), induces the formation of diaphragmed fenestra in cultured endothelial cells (Lombardi et al. 1986). PKC regulates various signal transductions that are activated by a variety of external factors including hormones and growth factors (Rosse et al. 2010) and controls the assembly of cytoskeleton networks such as actin remodeling (Larsson 2006). On the other hand, the formation and number of fenestrae in brain capillary endothelial cells and liver sinusoidal endothelial cells are increased by the treatment with actin polymerization inhibitors such as latrunculin A and cytochalasin B (Braet and Wisse 2002; Ioannidou et al. 2006). The ultra-structure observation by SEM revealed that when the microtubules in the CECAL were depolymerized by colcemid treatment, the individual fenestrae were reduced in size but remained, and forming the sieve plates similar to the control cells (Fig. 6). Therefore, there is a possibility that the unknown intrinsic mechanisms via PKC signaling and actin remodeling may contribute to endothelial cell fenestration and sieve plate formation in rat AL.
In this study, we demonstrated that the intracellular signaling activated by fibronectin-integrin α5β1 interaction is essential for the transportation of PLVAP via microtubules in fenestrated endothelial cells of rat AL. Since the loss of PLVAP leads to the reduction of fenestrae in size, it is a possible that PLVAP may be involved in the maintenance of permeability of fenestrae. We also clarified that a constant supply of PLVAP proteins by the endomembrane system via the Golgi apparatus is essential for the localization of PLVAP at sieve plates. The endomembrane transport pathway from the Golgi apparatus to sieve plates requires microtubule cytoskeletons, which are regulated by fibronectin-integrin α5β1 signaling. In the future, we will further clarify the physiological function of PLVAP localized at the fenestrae by measuring the permeability of fenestrae in endothelial cells of rat pituitary anterior lobe.

◂Fig. 5 Changes of PLVAP distribution in CECAL by treatment with colcemid. Signals of PLVAP (a–c, green), F-actin (a’–c’, red), and α-tubulin (a’–d’, magenta) in CECAL treated with water as control (a–a’’), 10 fg/ml colcemid (b–b’’), and 100 fg/ml colcemid (c–c’’) are shown. Nuclei were counterstained with DAPI (blue). Bars 20 μm. The expression levels of Tuba1b (d), Plvap (e), Itga5 (f), and Itgb1 (g) were quantitatively estimated by real-time PCR and normalized by the expression of Hprt1. Results of three independ- ent experiments are shown. Statistical differences were determined using the non-repeated measures ANOVA followed by Dunnett’s test. *P < 0.05, **P < 0.01 vs. control.

Fig. 6 SEM images showing the cell surface of CECAL. (a, b) The cell surface of CECAL treated with water as control are shown. (b) Enlarged photograph of the area surrounded by a black line in the photograph in (a). (c, d) The cell surface of CECAL treated with

100 fg/ml colcemid was observed. (d) Enlarged photograph of the area surrounded by a black line in the photograph in (c). In (a) and (c), the dot lines indicated the sieve plates. Bars 1 μm (a, c), 500 nm (b, d).

Fig. 7 Immunolocalization of PLVAP and GM130 in CECAL treated with ATN-161 and colcemid. Signals of PLVAP (a–c, green) and GM130 (a’–c’, red) in CECAL treated with water as control (a–a’’),

10 nM ATN-161 (b–b’’), and 100 fg/ml colcemid (c–c’’) are shown. Arrowheads indicate the same parts. Nuclei were counterstained with DAPI (blue). Bars 10 μm.

Fig. 8 Changes of PLVAP distribution in CECAL by treat- ment with brefeldin A. Signals of PLVAP (a–f, green),α-tubulin (a’, c’, e’, red), and GM130
(b’, d’, f’, red) in CECAL treated with DMSO as control (a–a’’, b–b’’), 10 μM brefeldin A (c–c’’, d–d’’) for 4 h, and
10 μM brefeldin A (e–e’’, f–f’’) for 20 h are shown. Nuclei
were counterstained with DAPI (blue). Bars 10 μm.

Funding This work was supported in part by JSPS KAKENHI (C) Grant Number JP19K07257, a research grant from Takeda Science Foundation, Hokuto Foundation for Bioscience, and Teikyo University.

Compliance with ethical standards

Conflict of interest The authors declare that they have no conflicts of interest.

Ethical approval This study was approved by the Laboratory Animal Ethics Committee of Teikyo University (Tokyo, Japan) and conducted according to its guidelines. The document ID of the approval is 17-008. This article does not contain any studies with human participants.

References

Akhmanova A, Steinmetz MO (2015) Control of microtubule organization and dynamics: two ends in the limelight. Nat Rev Mol Cell Biol 16:711–726
Alfaro-Aco R, Petry S (2015) Building the microtubule cytoskeleton piece by piece. J Biol Chem 290:17154–17162
Astrof S, Hynes RO (2009) Fibronectins in vascular morphogenesis. Angiogenesis 12:165–175
Braet F, Wisse E (2002) Structural and functional aspects of liver sinusoidal endothelial cell fenestrae: a review. Comp Hepatol 1:1–1
Colello D, Mathew S, Ward R, Pumiglia K, LaFlamme SE (2012) Integrins regulate microtubule nucleating activity of centrosome through mitogen-activated protein kinase/extracellular signal- regulated kinase kinase/extracellular signal-regulated kinase (MEK/ERK) signaling. J Biol Chem 287:2520–2530
Eremina V, Sood M, Haigh J, Nagy A, Lajoie G, Ferrara N, Gerber H-P, Kikkawa Y, Miner JH, Quaggin SE (2003) Glomerular-specific alterations of VEGF-A expression lead to distinct congenital and acquired renal diseases. J Clin Invest 111:707–716
Etienne-Manneville S, Hall A (2001) Integrin-mediated activation of Cdc42 controls cell polarity in migrating astrocytes through PKCζ. Cell 106:489–498
Farquhar MG (1961) Fine structure and function in capillaries of the anterior pituitary gland. Angiology 12:270–292
Francis SE, Goh KL, Hodivala-Dilke K, Bader BL, Stark M, Davidson D, Hynes RO (2002) Central roles of α5β1 integrin and fibronectin in vascular development in mouse embryos and embryoid bodies. Arterioscler Thromb Vasc Biol 22:927–933
Fujiwara T, Oda K, Yokota S, Takatsuki A, Ikehara Y (1988) Brefeldin
Acauses disassembly of the Golgi complex and accumulation of secretory proteins in the endoplasmic reticulum. J Biol Chem 263:18545–18552
George EL, Baldwin HS, Hynes RO (1997) Fibronectins are essential for heart and blood vessel morphogenesis but are dispensable for initial specification of precursor cells. Blood 90:3073–3081
George EL, Georges-Labouesse EN, Patel-King RS, Rayburn H, Hynes RO (1993) Defects in mesoderm, neural tube and vascular development in mouse embryos lacking fibronectin. Development 119:1079–1091
Gundersen GG, Gomes ER, Wen Y (2004) Cortical control of microtubule stability and polarization. Curr Opin Cell Biol 16:106–112
Herrnberger L, Seitz R, Kuespert S, Bösl MR, Fuchshofer R, Tamm ER (2012) Lack of endothelial diaphragms in fenestrae and caveolae of mutant Plvap-deficient mice. Histochem Cell Biol 138:709–724

Hirokawa N, Noda Y, Tanaka Y, Niwa S (2009) Kinesin superfamily motor proteins and intracellular transport. Nat Rev Mol Cell Biol 10:682–696
Horiguchi K, Nakakura T, Yoshida S, Tsukada T, Kanno N, Hasegawa R, Takigami S, Ohsako S, Kato T, Kato Y (2016) Identification of THY1 as a novel thyrotrope marker and THY1 antibody-mediated thyrotrope isolation in the rat anterior pituitary gland. Biochem Biophys Res Commun 480:273–279
Hynes RO (2002) Integrins: bidirectional, allosteric signaling machines. Cell 110:673–687
Ioannidou S, Deinhardt K, Miotla J, Bradley J, Cheung E, Samuelsson S, Ng Y-S, Shima DT (2006) An in vitro assay reveals a role for the diaphragm protein PV-1 in endothelial fenestra morphogenesis. Proc Natl Acad Sci U S A 103:16770–16775
Kapp TG, Rechenmacher F, Neubauer S, Maltsev OV, Cavalcanti-Adam EA, Zarka R, Reuning U, Notni J, Wester H-J, Mas-Moruno C, Spatz J, Geiger B, Kessler H (2017) A comprehensive evaluation of the activity and selectivity profile of ligands for RGD-binding integrins. Sci Rep 7:39805
Kim J, Han I, Kim Y, Kim S, Oh ES (2001) C-terminal heparin-binding domain of fibronectin regulates integrin-mediated cell spreading but not the activation of mitogen-activated protein kinase. Biochem J 360:239–245
LaFlamme SE, Mathew-Steiner S, Singh N, Colello-Borges D, Nieves
B(2018) Integrin and microtubule crosstalk in the regulation of cellular processes. Cell Mol Life Sci 75:4177–4185
Lammert E, Gu G, McLaughlin M, Brown D, Brekken R, Murtaugh LC, Gerber HP, Ferrara N, Melton DA (2003) Role of VEGF-A in vascularization of pancreatic islets. Curr Biol 13:1070–1074
Larsson C (2006) Protein kinase C and the regulation of the actin cytoskeleton. Cell Signal 18:276–284
Lei L, Liu D, Huang Y, Jovin I, Shai S-Y, Kyriakides T, Ross RS, Giordano FJ (2008) Endothelial expression of β1 integrin is required for embryonic vascular patterning and postnatal vascular remodeling. Mol Cell Biol 28:794–802
Lombardi T, Montesano R, Furie MB, Silverstein SC, Orci L (1986) Endothelial diaphragmed fenestrae: in vitro modulation by phorbol myristate acetate. J Cell Biol 102:1965–1970
Marchand M, Monnot C, Muller L, Germain S (2019) Extracellular matrix scaffolding in angiogenesis and capillary homeostasis. Semin Cell Dev Biol 89:147–156
Murakami T, Kikuta A, Taguchi T, Ohtsuka A, Ohtani O (1987) Blood vascular architecture of the rat cerebral hypophysis and hypothalamus. A dissection/scanning electron microscopy of vascular casts. Arch Histol Jpn 50:133–176
Nakakura T, Suzuki T, Horiguchi K, Fujiwara K, Tsukada T, Asano- Hoshino A, Tanaka H, Arisawa K, Nishijima Y, Nekooki-Machida Y, Kiuchi Y, Hagiwara H (2017) Expression and localization of forkhead box protein FOXJ1 in S100β-positive multiciliated cells of the rat pituitary. Med Mol Morphol 50:59–67
Nakakura T, Suzuki T, Tanaka H, Arisawa K, Miyashita T, Nekooki- Machida Y, Kurosawa T, Tega Y, Deguchi Y, Hagiwara H (2020) Fibronectin is essential for formation of fenestrae in endothelial cells of the fenestrated capillary. Cell Tissue Res in press. https://
doi.org/10.1007/s00441-00020-03273-y
Nakakura T, Suzuki T, Torii S, Asano-Hoshino A, Nekooki-Machida Y, Tanaka H, Arisawa K, Nishijima Y, Susa T, Okazaki T, Kiuchi Y, Hagiwara H (2017) ATAT1 is essential for regulation of homeostasis-retaining cellular responses in corticotrophs along hypothalamic-pituitary-adrenal axis. Cell Tissue Res 370:169–178
Nakakura T, Yoshida M, Dohra H, Suzuki M, Tanaka S (2006) Gene expression of vascular endothelial growth factor-A in the pituitary during formation of the vascular system in the hypothalamic-pituitary axis of the rat. Cell Tissue Res 324:87–95

Palazzo AF, Eng CH, Schlaepfer DD, Marcantonio EE, Gundersen GG (2004) Localized stabilization of microtubules by integrin- and FAK-facilitated Rho signaling. Science 303:836–839
Pankov R, Yamada KM (2002) Fibronectin at a glance. J Cell Sci 115:3861–3863
Peng X, Ueda H, Zhou H, Stokol T, Shen T-L, Alcaraz A, Nagy T, Vassalli J-D, Guan J-L (2004) Overexpression of focal adhesion kinase in vascular endothelial cells promotes angiogenesis in transgenic mice. Cardiovasc Res 64:421–430
Rhodin JA (1962) The diaphragm of capillary endothelial fenestrations. J Ultrastruct Res 6:171–185
Roberts WG, Palade GE (1995) Increased microvascular permeability and endothelial fenestration induced by vascular endothelial growth factor. J Cell Sci 108:2369–2379
Rosse C, Linch M, Kermorgant S, Cameron AJM, Boeckeler K, Parker PJ (2010) PKC and the control of localized signal dynamics. Nat Rev Mol Cell Biol 11:103–112
Shen T-L, Park AY-J, Alcaraz A, Peng X, Jang I, Koni P, Flavell RA, Gu H, Guan J-L (2005) Conditional knockout of focal adhesion kinase in endothelial cells reveals its role in angiogenesis and vascular development in late embryogenesis. J Cell Biol 169:941–952
Stan RV, Kubitza M, Palade GE (1999) PV-1 is a component of the fenestral and stomatal diaphragms in fenestrated endothelia. Proc Natl Acad Sci U S A 96:13203–13207
ATN-161
Stan RV, Tse D, Deharvengt SJ, Smits NC, Xu Y, Luciano MR, McGarry CL, Buitendijk M, Nemani KV, Elgueta R, Kobayashi T, Shipman SL, Moodie KL, Daghlian CP, Ernst PA, Lee HK, Suriawinata AA, Schned AR, Longnecker DS, Fiering SN, Noelle RJ, Gimi B, Shworak NW, Carriere C (2012) The diaphragms of fenestrated endothelia: gatekeepers of vascular permeability and blood composition. Dev Cell 23:1203–1218
Tanaka S, Nakakura T, Jansen EJR, Unno K, Okada R, Suzuki M, Martens GJM, Kikuyama S (2013) Angiogenesis in the intermediate lobe of the pituitary gland alters its structure and function. Gen Comp Endocrinol 185:10–18
Tanjore H, Zeisberg EM, Gerami-Naini B, Kalluri R (2008) β1 integrin expression on endothelial cells is required for angiogenesis but not for vasculogenesis. Dev Dyn 237:75–82
Xu J, Shi G-P (2014) Vascular wall extracellular matrix proteins and vascular diseases. Biochim Biophys Acta 1842:2106–2119
Zhao X, Guan J-L (2011) Focal adhesion kinase and its signaling pathways in cell migration and angiogenesis. Adv Drug Deliv Rev 63:610–615
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.