(an inhibitor of clathrin-mediated endocytosis) did not [24]. The cholesterol-lowering agents, statins, also interfered with the endocytosis of albumin into glomerular epithelial cells during in vitro and in vivo assays [25]. Additionally, caveolae are enriched in cholesterol and sphingolipids, and statins have been reported to interfere with caveolae-dependent BK virus internalization into renal proximal tubular epithelial cells [26], further supporting the hypothesis of the endocytosis of albumin through caveolae. Taken together, these results suggest that FcRn-bound albumin particles enter glomerular epithelial cells through caveolae and are transported to early endosomes. In another study, it was demonstrated that, after the endocytosis of albumin through caveolae, albumin particles were transported to early endosomes that move along actin; further, several albumin particles were degraded in lysosomes [27]. Albumin exposure also induced cell apoptosis through combined activation of caspase 3/7 and nuclear factor kappa B and release of interleukin-1β, tumor necrosis factor, and interleukin-6 [28]. In a recent report, Schießl et al. [29] used an intravital multiphoton microscope to show transcellular transport of albumin – through caveolae – across the glomerular epithelial cells, and noted that this transcytosis increased upon stimulation with angiotensin II [30]. These data imply that there may be a transcellular pathway for albumin transport across glomerular epithelial cells.
New Intracellular Trafficking Pathway of Albumin through Glomerular Endothelial and Epithelial Cells
The glomerular capillary wall is composed of glomerular epithelial cells, glomerular basement membrane (GBM), and glomerular endothelial cells. Glomerular endothelial cells comprise the innermost layer and connect to capillary lumen; glomerular epithelial cells comprise the outermost layer and connect to Bowman’s space; and GBM is located between the glomerular endothelial and epithelial cells. These three components serve as the glomerular filtration barrier to form a crude urine filtrate from blood. Traditionally, in glomerular endothelial cells, albumin particles pass through fenestrae that are 100 nm in size and lack a diaphragm, while in glomerular epithelial cells, albumin passes through a slit diaphragm located in the gap between podocyte foot processes (Fig. 2a). However, in several kinds of primary and secondary glomerulonephritis, glomerular endothelial cells become swollen, causing the fenestrae to narrow and the foot processes of glomerular epithelial cells to fuse; this fusion results in the loss of the gap between foot process (Fig. 2b). In this situation, the caveolae pathway is repurposed to serve as an alternative albumin pathway across glomerular endothelial and epithelial cells. In glomerular endothelial cells, albumin particles enter through caveolae using an unknown receptor, move freely and not in conjunction with cytoskeletal structures (e.g., actin and microtubules), and arrive at early endosomes. In early endosomes, these albumin particles are able to bypass typical endosome-associated organelles (e.g., GA and ER) without undergoing modifications, such as glycosylation, proteolytic processing, sulfation or phosphorylation; the albumin particles also bypass the lysosome and proteasome without undergoing degradation and are transported directly to the other side of the cells for ultimate excretion (Fig. 3).
After passage through GBM, albumin particles enter glomerular epithelial cells through FcRn present in caveolae, then move along cytoskeletal components (e.g., actin) and arrive at early endosomes. In early endosomes, some albumin particles are sorted for degradation by lysosomes or may induce cell apoptosis through cytokine activation; however, other albumin particles are transported to Bowman’s space. In the experimental diabetic nephritis mouse model, MBCD treatment decreased albuminuria, without effects on blood glucose, body weight, or kidney weight. Moreover, MBCD treatment did not affect glomerular deterioration, demonstrated by the measurement of glomerular and mesangial surface area [31]. Combined with our reports that MBCD treatment decreases Cav-1 expression on glomerular capillaries and decreases albuminuria in the PAN-induced nephrotic syndrome mouse model, these data indicate that the caveolae pathway in both glomerular endothelial and epithelial cells may provide an additional etiology for albuminuria along with the classical fenestral pathway in glomerular endothelial cells and the slit membrane pathway in glomerular epithelial cells.
Fig. 2. Passage of albumin particles through glomerular capillary. a In the normal kidney, few albumin particles pass through fenestra in the glomerular endothelial cells, through glomerular basement membrane, then through the slit membrane in the gap between foot processes of glomerular epithelial cells. b In glomerulonephritis, fenestrae in the glomerular endothelial cells are narrowed and gaps between foot processes in glomerular epithelial cells are fused. In this situation, albumin may pass through glomerular endothelial and epithelial cells by caveolae-dependent pathways.
Fig. 3. Albumin endocytosis, transcytosis, and exocytosis through glomerular endothelial cells. After albumin particles (yellow) are captured in caveolae, caveolae are pinched off from cell membrane. Albumin particles in the caveolae-coated pit move independently from the cytoskeleton and arrive at early endosomes. Albumin particles are then sorted to the other side of cell membrane for exocytosis and avoid degradation by any of the following: proteasome, lysosome, Golgi apparatus, or endoplasmic reticulum.
Conclusions
In this report, we reviewed the new pathway by which albumin travels through caveolae in both glomerular endothelial and epithelial cells. This caveolae-dependent pathway appears to match several aspects of pathophysiology in glomerulonephritis, including the narrowing of fenestra in glomerular endothelial cells and the fusion of foot processes in podocytes. We suspect that this new theory may become part of the general etiology of albuminuria along with the fenestral pathway and the slit membrane pathway.
Acknowledgements
We thank Ryan Chastain-Gross, PhD, from Edanz Group (www.edanzediting.com/ac) for editing a draft of this manuscript.
References
1Yamada E: The fine structure of the gall bladder epithelium of the mouse. J Biophys Biochem Cytol 1955;1:445–458.
2Stan RV: Structure of caveolae. Biochim Biophys Acta 2005;1746:334–348.
3Lamaze C, Tardif N, Dewulf M, Vassilopoulos S, Blouin CM: The caveolae dress code: structure and signaling. Curr Opin Cell Biol 2017;47:117–125.
4Busija AR, Patel HH, Insel PA: Caveolins and cavins in the trafficking, maturation, and degradation of caveolae: implications for cell physiology. Am J Physiol Cell Physiol 2017;312:C459–C477.
