Biochemical and Biophysical Research Communications

journal homepage:
Biochemical and Biophysical Research Communications 574 (2021) 97e103

Intracellular trafficking pathway of albumin in glomerular epithelial cells
Takahito Moriyama*, Fumio Hasegawa, Yoei Miyabe, Kenichi Akiyama, Kazunori Karasawa, Keiko Uchida, Kosaku Nitta
Department of Nephrology, Tokyo Women’s Medical University, Tokyo, Japan

a r t i c l e i n f o

Article history:
Received 23 July 2021
Accepted 18 August 2021
Available online 24 August 2021

Keywords: Podocytes Albumin Endocytosis Transcytosis Exocytosis
a b s t r a c t

The intracellular trafficking pathway of albumin in podocytes remains controversial. We therefore analysed albumin endocytosis through caveolae, subsequent transcytosis, and exocytosis. In Western blot and immunofluorescence analysis in vitro, methyl-beta-cyclodextrin (MBCD) treatment significantly decreased the expression of caveolin-1 and albumin in cultured human podocytes after incubation with albumin; additionally, MBCD interfered with albumin endocytosis through caveolae in the experiment using Transwell plates. In the immunofluorescence analysis, albumin was incubated with cultured hu- man podocytes, and colocalisation analysis with organelles and cytoskeletons in the podocytes showed that albumin particles colocalised with caveolin-1 and Fc-receptor but not clathrin in endocytosis, colocalised with actin cytoskeleton but not microtubules in transcytosis, and colocalised with early endosomes and lysosomes but not proteasome, endoplasmic reticulum, or Golgi apparatus. In the electron microscopic analysis of podocytes in nephrotic syndrome model mice, gold-labelled albumin was shown as endocytosis, transcytosis, and exocytosis with caveolae. These results indicate the intra- cellular trafficking of albumin through podocytes. Albumin enters through caveolae with the Fc-receptor, moves along actin, and reaches the early endosome, where some of them are sorted for lysosomal degradation, and others are directly transported outside the cells through exocytosis. This intracellular pathway may be a new aetiological hypothesis for albuminuria.

© 2021 Elsevier Inc. All rights reserved.
1. Introduction

The glomerular filtration barrier (GFB) comprises three physi- ological layers: glomerular epithelial cells (podocytes) located at the outermost layer to face Bowman’s space, glomerular endothe- lial cells (GEnC) as the innermost layer facing the capillary lumen, and glomerular basement membrane (GBM) located between these two layers. GFB plays a pivotal role in the production of urine by filtration from a large amount of plasma, and results in stabilisation of homeostasis, removal of residual water and toxins from protein metabolites, and adjustment of acid-base and electrolyte distur- bances. In glomerulonephritis, which is inflammation of the

* Corresponding author. Department of Nephrology, Tokyo Women’s Medical University, 8-1 Kawasa-cho, Shinjuku-ku, Tokyo, 162, Japan
E-mail addresses: [email protected] (T. Moriyama), [email protected] (F. Hasegawa), [email protected] (Y. Miyabe), [email protected] (K. Akiyama), [email protected] (K. Karasawa), [email protected] (K. Uchida), [email protected] (K. Nitta).
glomeruli, a large amount of urinary albumin excretion is observed when serum albumin passes through the GFB and saturates the reabsorbed capacity of renal proximal tubular cells. The traditional albumin excretory pathways through podocytes have been recog- nised as the route of desquamated podocytes and gaps with injured slit diaphragms between foot processes; recently, however, the new pathogenesis of albuminuria by albumin passage through the novel route as an intracellular trafficking pathway in podocytes has been reviewed [1,2].

The mechanisms of albumin endocytosis, transcytosis, and exocytosis through podocytes are not fully understood. We re- ported that the amount of urinary albumin excretion in glomeru- lonephritis was significantly correlated with the expression of caveolin-1 (Cav-1) on the glomerular capillary walls in renal bi- opsy specimens [3]. Cav-1 is a 21e24 kDa structural protein of caveolae, which are rich in cholesterol and sphingolipids, flask- shaped invaginations on the cell surface with a diameter of 50e100 nm and are important in endocytosis and transcytosis of macromolecules, cell signalling regulation, vascular permeability,

0006-291X/© 2021 Elsevier Inc. All rights reservedand nitric oxide synthesis [4e6]. Caveolae have also been reported to play a role in the endocytosis of albumin into cells and subse- quent transcytosis [7]. In our previous report [3], Cav-1 expression on the capillaries highly colocalised with pathologische anatomie leiden-endothelium (PAL-E), which is an endothelial marker; therefore, we have analysed and confirmed albumin endocytosis through caveolae [8], transcytosis in the cytoplasm, and exocytosis to the outside of cells [9] in GEnCs in vitro. However, in a previous study [3], some Cav-1 expression on the capillaries did not coloc- alise with PAL-E, suggesting that caveolae were also located on the podocytes. In experiments employing human urine derived podocyte-like epithelial cells (HUPECs), albumin endocytosis through caveolae [10] and degradation in lysosomes [11] have been reported, but the features of immortalised podocyte-like cells were found to be different from those of native podocytes [1], and the trafficking pathway should be analysed in real human podocytes.

In the present study, we employed native podocytes in an
in vitro study to confirm albumin endocytosis, subsequent trans- cytosis, and exocytosis, and validate the intracellular trafficking of immunogold albumin through mouse podocytes in vivo.

2. Materials and methods

2.1. Culture of human podocytes

Primary human glomerular epithelial cells (podocytes) were kindly provided by K.K [12]. and grown in Roswell Park Memorial Institute medium supplemented with 10% foetal bovine serum (FBS) and penicillin (10,000 U/mL)/streptomycin (10,000 mg/mL). The cell line was confirmed to be positive for the podocyte markers nephrin (G-20; Santa Cruz Biotechnology, Dallas, TX, USA) and podocin (Merck & Co., Inc., Kenilworth, NJ, USA) and negative for the endothelial cell marker von Willebrand factor (Abcam, Cam- bridge, MA, USA). Prior to the experiments, the podocytes were incubated in serum-free medium for 24 h, followed by incubation with 20 mg/mL Alexa Fluor 488-labelled bovine serum albumin (AF488-BSA; Molecular Probes, Eugene, OR, USA) for immunoflu- orescence (IF) analysis or 100 mg/mL human serum albumin (Sigma-Aldrich, St. Louis, MO, USA) for Western blot (WB) analysis, and analysis using Transwell plates.

2.2. Antibodies and reagents

The primary antibodies, namely rabbit polyclonal anti-Cav-1 (N- 20), mouse monoclonal anti-clathrin heavy chain (A-8), rabbit polyclonal anti-FCRn (H-274), rabbit polyclonal anti-early endo- some antibody 1 (EEA1; H-300), mouse monoclonal anti-20 S proteasome a7/a8 (B-4), and mouse monoclonal anti-microtubule (AE-8), were purchased from Santa Cruz Biotechnology. Mouse monoclonal antibodies against the endoplasmic reticulum (ER), protein disulphide isomerase (PDI) (RL90), lysosome, lysosome- associated membrane protein 1 (LAMP1; H4A3), and human serum albumin were purchased from Abcam. Mouse anti-human Golgi zone monoclonal antibody was acquired from Millipore (Temucula, CA, USA), and rabbit anti-actin polyclonal antibody was obtained from Cytoskeleton Inc. (Denver, CO, USA). The reagent 40,6-diamidino-2-phenylindole dihydrochloride (DAPI) was pur- chased from Life Technologies (Grand Island, NY, USA). AF488-BSA, and Alexa Fluor 546-labelled goat anti-rabbit IgG [H L] and Alexa Fluor 594-labelled donkey anti-mouse IgG [H L] antibodies for secondary antibody were from Molecular Probes.

2.3. Indirect immunofluorescence analysis

Specimens were fixed in 10% neutral-buffered formalin at 37 ◦C
for 20 min and 100% methanol at 20 ◦C for 20 min and then incubated for 30 min in Tris-buffered saline with 0.1% Tween-20 (TBST) containing 3% FBS for blocking. The samples were incu- bated with a primary antibody (Cav-1, proteasome, EEA, FcRn, AE-8, and clathrin diluted 1:50, PDI, LAMP1, and actin diluted 1:100, golzi-zone diluted 1:20) in TBST containing 1% FBS at room tem- perature for 1 h, and then washed three times for 5 min each with TBST. They were subsequently incubated with a secondary antibody (Alexa Fluor 546 or 594 diluted 1:200) at room temperature for 1 h and washed three times for 5 min each with TBST. To stain the nuclei in the human renal glomerular endothelial cells (HRGECs), the cells were incubated with 300 nM DAPI for 5 min and washed three times with phosphate-buffered saline (PBS). The samples were viewed via confocal microscopy (LSM 710; Carl Zeiss, Jena, Germany) using the ZEN 2011 software (blue edition; Carl Zeiss). The intensity of the signal of BSA colocalised with each organelle, receptor, or cytoskeletal component in HRGECs was assessed using ZEN 2011 blue and black versions, as described previously [8,9].

2.4.Western blot analysis

According to our protocol [8,9,12], after incubating albumin with or without methyl-beta-cyclodextrin (MBCD), each cell monolayer was washed three times with PBS. The cells were scraped, har- vested, disrupted using an ultrasound homogeniser (Bioruptor®, BM Equipment Co., Ltd., Tokyo, Japan), and dissolved in gel buffer containing sodium dodecyl sulfate (SDS). Samples were loaded onto 5%e20% Super Sep SDS polyacrylamide gels (Wako Pure Chemical Industries, Osaka, Japan) and electrophoresed. Proteins were transferred onto nitrocellulose membranes (Millipore Cor- poration, Billerica, MA, USA) followed by blocking of the mem- branes in Odyssey® blocking buffer (LI-COR Biosciences, Inc.,
Lincoln, NE, USA) at room temperature (15e25 ◦C) for 1 h. Mem-
branes were incubated for about 12 h with overnight at 4 ◦C with primary antibodies against albumin (1:5000), Cav-1 (1:500), and actin (1:2000) with constant rotation. The membranes were washed three times with TBST at room temperature for 5 min with agitation, incubated with secondary antibody conjugated to IRDye 680 or IRDye 800 diluted 1:10,000 (LI-COR Biosciences), and washed three times with TBST at room temperature for 5 min with agitation. The bands were visualised using an Odyssey® infrared imaging system (LI-COR Biosciences), which measured the inte- grated intensity of each band. The signal of each band was nor- malised to that of actin in the same sample.

2.5. Transwell transcytosis study

According to our previous report [9], after growing to full confluence on Transwell membranes, podocytes were incubated with 1500 mg/mL human serum albumin, pipetted inside the wells, for 0, 15, 30, 60, 120, and 240 min, with the outside of the Transwell filled with serum and albumin-free medium. Both media from the inside and outside of the Transwells were collected, and the con- centrations of albumin were measured via absorbance using a spectrophotometer (The Plate Chameleon; Hidex, Turku, Finland). In some experiments, the cells were pre-incubated with MBCD (Sigma-Aldrich), a caveolae-disrupting agent, and then albumin transcytosis were analysed as described above.

2.6. Animals and in vivo experimental protocols

Electron microscopic specimens for in vivo analysis were used in a previous study [9]. Male C5BL/6j mice (B6 mice), 7 weeks of age, were purchased from CLEA Japan, Inc. (Tokyo, Japan). All animals were given free access to standard food and water and were MBCD interferes albumin internalisation into podocytes through caveolae. (a) In WB analysis, the expression of albumin and Cav-1 was dramatically decreased by dose- dependent MBCD treatment, and their relative expression was significantly decreased in comparison to 0 mM MBCD (albumin: p < 0.001 in 0.5, 1, and 2 mM MBCD; and Cav-1: p < 0.05 in 2 mM MBCD). (b) In IF analysis, the expression of albumin (green) and Cav-1 (red) was dramatically decreased by dose-dependent MBCD treatment, and the relative intensity of the expression of them was significantly decreased in comparison to 0 mM MBCD (albumin: p < 0.05 in 0.5 mM, p < 0.01 in 1 mM, and p < 0.01 in 2 mM; and Cav-1: p < 0.001 in 0.5, 1 and 2 mM). *p < 0.001; **p < 0.05; ***p < 0.01. WB, Western blot; IF, immunofluorescence. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)
allowed to acclimatise to the animal facility environment for 1 week prior to the experiments. The puromycin aminonucleoside (PAN; 450 mg/kg; Sigma-Aldrich) was injected intraperitoneally on days 0 and 4, and 150 mL saline was injected daily. On day 9, the mice were sacrificed and histological analysis of the mice by elec- tron microscopy was performed. The animal experimental protocol was approved by the Independent Animal Care and Use Committee of Tokyo Women's Medical University (#AE19-046 and #AE20- 048).

2.7. Electron microscopic examination of immunogold labelling albumin

For electron microscopy with immunogold labelling of albumin, small pieces of renal cortex were fixed with 0.05% glutaraldehyde in 4% paraformaldehyde for 1 h at 4 ◦C and then washed with 0.05 M pH 7.4 PBS. The samples were subsequently fixed with 1% osmium tetroxide and washed with PBS (0.05 M, pH 7.4) at 4 ◦C. Samples were dehydrated in graded ethanol solutions (50%, 70%, 80%, 90%, 95%, and 100% twice for 15 min each), replaced with n-butyl gly- cidyl ether (QY-1; Oken, Tokyo Japan) and alcohol diluted (1:3), (1:1), and (3:1) for 15 min two times, and embedded by inverting Poly/Bed 812-filled Better Equipment for Electron Microscopy (BEEM®) capsules (Polysciences, Warrington, PA, USA) at room temperature. The blocks were cured for 2 d at 60 ◦C, and the ul- trathin sections were cut with a diamond knife on an ultramicro- tome. The samples were incubated in PBS with 1% BSA for 30 min, washed with 0.05 M pH 7.4 PBS 20 times, and incubated overnight with anti-mouse albumin antibody (Abcam). After washing with distilled water 20 times, the samples were incubated with a sec- ondary antibody for immune-gold (Sigma-Aldrich). The samples were washed again with distilled water 20 times, and stained with uranyl acetate for 20 min and subsequently with lead citrate for 5 min. The samples were observed under a JEM-1400 Plus trans- mission electron microscope (JEOL, Peabody, MA, USA) at 80 kV.

2.8. Statistical analysis

All in vitro experiments were performed at least in triplicate to confirm reproducibility. The mean ± standard error for normally distributed data and the interquartile range (IQR) for skewed data were calculated from the combined data. Differences in expression in WB and IF experiments were assessed using unpaired t-tests. Serial changes in the concentration of albumin in experiments with Transwell plates were evaluated using the Friedmann test. Statis- tical significance was set at P < 0.05. Data were analysed using JMP
15.0.1 software (SAS Institute, Cary, NC, USA).

3. Results

3.1. Albumin endocytosis through caveolae

In Fig. 1a, albumin in podocytes were dramatically decreased by the dose-dependent treatments with MBCD, and relative albumin expressions were significantly decreased as MBCD amount was increased from 0.5 to 2 mM, compared to controls (non-treated with MBCD). Relative Cav-1 expression tended to decrease in a dose- dependent manner and was significantly decreased in podocytes treated with 2 mM MBCD. Similarly, in IF analysis (Fig. 1b), abundant AF488-BSA was observed in control podocytes; however, it dramatically decreased in a dose-dependent manner following MBCD treatment, and little AF488-BSA was observed in podocytes treated with 2 mM MBCD. Cav-1 was also highly expressed in con- trol podocytes and was dramatically decreased after treatment Colocalisation of Alexa Fluor 488-labelled BSA incubated with organelles, cytoskeletal components, and receptors in podocytes for 30, 60 and 90 min. Albumin particles colocalised with (a) Cav-1 and (b) FcRn but not with (c) clathrin as endocytosis. Albumin particles colocalised with (d) actin, but not with (e) microtubules as transcytosis. Albumin particles were colocalised with (f) endosomes and (g) lysosomes, but not with (h) proteasome, (i) ER and (j) GA. (Arrows are colocalised particles, and arrowheads are not) (k) The relative intensity of Alexa Fluor 488-labelled BSA with endosomes at 90 min was the highest among the colocalised particles, and it was significantly higher than that of particles with Cav-1 at 30 min, and clathrin, microtubules, proteasome, ER, and GA at any timepoint. *p < 0.05; **p < 0.01with MBCD. The relative expression of both Cav-1 and AF488-BSA was significantly decreased by MBCD treatment in a dose- dependent manne

3.2. Colocalisation of Alexa Fluor 488 BSA with several markers of endocytosis and transcytosis

AF488-BSA was highly colocalised with Cav-1 and FcRn (Fig. 2a and b) as a marker of endocytosis but rarely colocalised with cla- thrin (Fig. 2c). Considering transcytosis, AF488-BSA was highly colocalised with cytoskeletal actins but not microtubules (Fig. 2d and e). As for the organelle markers, AF488-BSA was highly colo- calised with early endosomes and lysosomes (Fig. 2f and g) but rarely colocalised with proteasomes, ER, and GA (Fig. 2hej). shows the relative expression of the intensity of AF488-BSA colo- calised with each marker. The relative expression was observed to be highest with early endosomes at 90 min, and that with Cav-1 (at60 and 90 min), FcRn, actin, and lysosomes were similar. In contrast, the relative expression of AF488-BSA colocalised with clathrin, microtubules, proteasome, ER, or GA was significantly lower.

3.3. Albumin passage through podocytes on Transwell plates

In the control podocytes, albumin concentration significantly decreased from 1500 mg/mL at 0 min to 1072.2 mg/mL at 240 min in the inside medium (p < 0.001) but remained constant in the podocytes treated with MBCD at 240 min (p ¼ 0.481); however, the

Serial changes of albumin concentration in the each medium inside and outside the Transwell plate. (a) After podocytes were grown to full confluence, 1500 mg/mL of human serum albumin was added inside the Transwell, while the outside was fully filled with medium without albumin. (b) The serial changes of albumin concentration significantly decreased (p < 0.001) in the inside medium of control podocytes from 0 to 240 min but did not in the inside medium of podocytes treated with MBCD (p ¼ 0.481). The serial changes were significantly increased during the 240 min in the outside medium of both control podocytes and podocytes with MBCD (p < 0.001). *: p < 0.001.
Electron microscopy shows the immunogold-labelled albumin particles observed in caveolae of the podocytes in nephrotic syndrome modelled mice (a-f). Arrows: immunogold-labelled albumin. Pod: podocytes, GEnC: glomerular endothelial cells, GBM: glomerular basement membrane, CL: capillary lumen, BC: Bowman’s capsule. The scale bars indicate 200 nm without C right panel (100 nm)concentration in the outside medium in the control and MBCD- treated podocytes were both significantly elevated (Fig. 3).

3.4. Electron microscopic analysis of intracellular trafficking pathway of immunogold-labelled albumin particle with caveolae vesicles

Immunogold-labelled albumin particles were captured in cav- eolae (Fig. 4a and b), the albumin-containing vesicles of which were pinched off from the cell membrane and underwent endocytosis and transcytosis (Fig. 4c and d). Exocytosis of albumin particles from caveolae vesicles to Bowman's capsule lumen vesicles was observed (Fig. 4e and f).

4. Discussion

In this study, we demonstrated that albumin endocytosis through caveolae with FcRn but not with clathrin, transcytosis with actin but not with microtubules, reaching to early endosomes and lysosomes, but not proteasome, and exocytosis to the other side of cells by bypassing the ER and GA. These results indicate that al- bumin entered podocytes through caveolae, moved along caveolae vesicles, and reached the early endosome, where albumin may be sorted for lysosomal degradation or to bypass ER and GA and be excreted outside the cells as the new aetiology of albuminuria.
Previously, in in vitro experiments using HUPECs, albumin colocalised with Cav-1 and FcRn but not with clathrin. Moreover, nystatin, an inhibitor of caveolae-mediated endocytosis, interfered with albumin internalisation into HUPECs, but Pitstop2, an inhibitor of clathrin-mediated endocytosis, did not [10]. Albumin also colocalises with actin, early endosome, and lysosome [10], is degraded by the lysosome [11], and induces apoptosis of podocytes, which increases nuclear factor-kB, interleukin (IL)-1b, tumour ne- crosis factor, and IL-6 [13]. In an in vivo study with PAN-induced nephrotic rats, gold-labelled albumin particles were observed in the cell body of podocytes and glomerular endothelium [14]. FcRn functioned as an albumin receptor in podocytes [15], and vesicle transportation of albumin was observed [16]. These reports support our results, which demonstrated the endocytosis, transcytosis, and exocytosis of gold-labelled albumin in caveolar vesicles through the experiment with human podocytes. However, the pathogenesis of albuminuria remains unclear.

In addition, the degree of albumin level through GFB is uncer- tain. Traditionally, a small amount of albumin passes the normal GFB consisting of GEnC with glycocalyx as a charge barrier, nega- tively charged GBM with laminin, type IV collagen, and fibronectin as a charge and size barrier, and, podocytes with slit membranes as size barriers. However, recent research proposed the leaky glomerular barrier theory, which states that a large amount of al- bumin passes through the normal GFB and is reabsorbed by tubular cells [17e19]. Moreover, foot process effacement occurred, and the gaps between the podocytes were dramatically reduced in the glomerulonephritis, swelling occurred, and fenestrae were nar- rowed in the GEnC; thus, it seemed adequate to have a new path- ogenesis of urinary albumin excretion. Consequently, the intracellular trafficking pathways of albumin through podocytes and glomerular endothelium were assumed as a new pathogenesis of albuminuria [1,2,20] and may become a novel therapeutic target to reduce albuminuria.

A previous report showed that albumin endocytoses into
podocytes in PAN and an antibody against FcRn decreased albu- minuria in PAN rats by preventing albumin internalisation into podocytes via inhibition of the key receptor of albumin endocytosis [15]. In an in vitro study, albumin endocytosis was also observed, and statins decreased the albumin uptake of podocytes [21]. Cav- eolae is well known to be abundant in cholesterol and sphingoli- pids, and the cholesterol-lowering agent pravastatin was reported to decrease Cav-1 levels and interfere with the endocytosis of BK virus into renal proximal tubular epithelial cells [22], indicating that statins might interfere with macromolecule endocytosis through caveolae into podocytes by decreasing caveolae levels. In the actual clinical situation, several statins were reported to decrease proteinuria in diabetic nephropathy [23] and chronic kidney diseases [24]. Our previous report showed that sertraline interfered with albumin internalisation through caveolae and decreased albuminuria in PAN mice by inhibiting the activity of dynamin I guanosine triphosphatases, which play an important role in pinching off the caveolae from the cell membrane [12]. It was also previously reported that salidroside decreased Cav-1 phosphorylation by activating adenosine monophosphate- activated protein kinase, inhibited albumin transcytosis across GEnC, and resulted in decreased albuminuria in diabetic ne- phropathy model mice [25]. Another study showed that glomerular matrix accumulation and albuminuria were decreased in Cav-1 knockout mice with diabetic nephropathy, although the trans- cytosis of albumin through GFB has not been reported [26].
These studies support our results of albumin endocytosis, transcytosis, and exocytosis through the podocytes via the caveolae-mediated pathway, although some of these reports also suggested the intracellular trafficking pathway of albumin with vesicles other than caveolae [14,16,27], and it also remains unknown how much albumin passes through the podocytes via the caveolae-mediated pathway.

In conclusion, we showed albumin endocytosis, transcytosis,
and exocytosis through podocytes via a caveolae-mediated pathway. It may be one of the aetiologies of urinary albumin excretion and may be a new therapeutic target.

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Funding sources

This work was funded by JSPS KAKENHI through a Grant-in-Aid for Scientific Research (C) [grant number 18K08223].


This study was supported by the Medical Research Institute (MRI) of the Tokyo Women's Medical University. We would like to thank Editage ( for English language editing.


[1] H. Castrop, I.M. Scheißl, Novel routes of albumin passage across the glomer- ular filtration barrier, Acta Physiol. 219 (2017) 546e555, 10.1111/apha.12760.
[2] F.F. He, Y. Gong, Z.Q. Li, , et al.Wang, A new pathogenesis of albuminuria: role of transcytosis, Cell. Physiol. Biochem. 47 (2018) 1274e1286, 10.1159/000490223.
[3] T. Moriyama, Y. Tsuruta, A. Shimizu, et al., The significance of caveolae in the glomeruli in glomerular disease, J. Clin. Pathol. 64 (2011) 504e509, https://
[4] R.G. Parton, K. Simon, The multiple faces of caveolae, Nat. Rev. Mol. Cell Biol. 8 (2007) 185e194,
[5] M. Simionescu, D. Popov D, A. Sima, Endothelial transcytosis in health and disease, Cell Tissue Res. 335 (2009) 27e40, 008-0688-3.
[6] T.A. John, S.M. Vogel, C. Tiruppathi, et al., Quantitative analysis of albumin uptake and transport in the rat microvessel endothelial monolayer, Am. J. Physiol. Lung Cell Mol. Physiol. 284 (2003) L187eL196, 10.1152/ajplung.00152.2002.
[7] M. Pascariu, M. Bendayan, L. Ghitescu, Correlated endothelial caveolin over- expression and increased transcytosis in experimental diabetes, J. Histochem. Cytochem. 52 (2004) 65e76,
[8] T. Moriyama, T. Takei, M. Itabashi, et al., Caveolae may enable albumin to enter human renal glomerular endothelial cells, J. Cell. Biochem. 116 (2015) 1060e1069,
[9] T. Moriyama, K. Sasaki, K. Karasawa, et al., Intracellular transcytosis of albumin in glomerular endothelial cells after endocytosis through caveolae, J. Cell. Physiol. 232 (2017) 3565e3573,
[10] E. Dobrinskikh, K. Okamura, J.B. Kopp, et al., Human podocytes perform polarized, caveolae-dependent albumin endocytosis, Am. J. Physiol. Ren. Physiol. 306 (2014) F941eF951,
[11] J.M. Carson, K. Okamura, H. Wakashin, et al., Podocytes degrade endocytosed albumin primarily in lysosomes, PloS One 9 (2014), e99771, 10.1371/journal.pone.0099771.
[12] T. Moriyama, K. Karasawa, F. Hasegawa, et al., Sertraline reduces albuminuria by interfering with caveolae-mediated endocytosis through glomerular endothelial and epithelial cells, Am. J. Nephrol. 50 (2019) 444e453, https://
[13] K. Okamura, P. Dummer, J. Kopp, et al., Endocytosis of albumin by podocytes elicits an inflammatory response and induces apoptotic cell death, PloS One 8 (2013), e54817,
[14] A. Tojo, M.I. Onozato, C. Kitiyakara, et al., Glomerular albumin filtration through podocytes cell body in puromycin aminonucleoside nephrotic rat, Med. Mol. Morphol. 41 (2008) 92e98, 0397-8.
[15] S. Kinugasa, A. Tojo, T. Sakai, et al., Selective albuminuria via podocyte albu- min transport in puromycin nephrotic rats is attenuated by an inhibitor of NADPH oxidase, Kidney Int. 80 (2011) 1328e1338, ki.2011.282.
[16] A. Tojo, S. Hatakeyama, S. Kinugasa, et al., Enhanced podocyte vesicle trans- port in the nephrotic rat, Med. Mol. Morphol. 50 (2017) 86e93, https://
[17] L.K. Russo, R.M. Sandoval, M. McKee, et al., The normal kidney filters nephrotic levels of albumin retrieved by proximal tubule cells: retrieval is disrupted in nephrotic states, Kidney Int. 71 (2007) 504e513,

[18] R.M. Sandoval, B.A. Molitoris, Quantifying glomerular permeability of fluo- rescent macromolecules using 2-photon microscopy in Munich Wister rats, J. vis. Exp. 17 (2013), 50052,
[19] R.M. Sandoval, M.C. Wagner, M. Patel, et al., Multiple factors influence glomerular albumin permeability in rats, J. Am. Soc. Nephrol. 23 (2012) 447e457,
[20] T. Moriyama, K. Karasawa, K. Nitta, The role of caveolin on albumin passage through glomerular endothelial and epithelial cells: the new etiology of uri- nary albumin excretion, Contrib. Nephrol. 195 (2018) 1e11, 10.1159/000486929.
[21] J. Eyre, K. Ioannou, B.D. Grubb, et al., Statin-sensitive endocytosis of albumin by glomerular podocytes, Am. J. Physiol. Ren. Physiol. 292 (2007) 674e681,
[22] T. Moriyama, A. Sorokin, Repression of BK virus infection of renal proximal tubular epithelial cells by pravastatin, TransplantLinc 85 (2008) 1311e1317,
[23] X. Shen, Z. Zhang, X. Zhang, et al., Efficacy of statins in patients with diabetic
nephropathy: a meta-analysis of randomized controlled trials, Lipids Health Dis. 15 (2016) 179,
[24] X. Su, L. Zhang, J. Lv, et al., Effect of statins on kidney disease outcomes: a systematic Methyl-β-cyclodextrin review and meta-analysis, Am. J. Kidney Dis. 67 (2016) 881e892,
[25] D. Wu, X. Yang, T. Zheng, et al., A novel mechanism of action for salidroside to alleviate diabetic albuminuria: effects on albumin transcytosis across glomerular endothelial cells, Am. J. Physiol. Endcrinol. Metab. 31 (2016) 225e237,
[26] T.H. Guan, C.B. Gao, M.R. Janssen, et al., Caveolin-1 deficiency protects against mesangial matrix expansion in a mouse model of type 1 diabetic nephropa- thy, Diabetologia 56 (2013) 2068e2077, 2968-z.
[27] I.M. Scheißl, A. Hammer A, V. Kattler, et al., Intravital imaging reveals angiotensin II-induced transcytosis of albumin by podocytes, J. Am. Soc. Nephrol. 27 (2016) 731e734,