The Rab-Rabphilin system in injured human podocytes stressed by glucose overload and angiotensin II

The Rab-Rabphilin system in injured human podocytes stressed by glucose overload and angiotensin II. injury in hyper- tension and diabetes entails, among in other structures, damage in a key cell of the glomerular ﬁltration barrier, the podocyte. Podocytes are polarized and highly differentiated cells in which vesicular transport, partly driven by Rab GTPases, is a relevant process. The aim of the present study was to analyze Rab GTPases of the Rab-Rabphilin system in human immortalized podocytes and the impact of high glucose and angiotensin II. Furthermore, alterations of the system in urine cell pellets from patients with hypertension and diabetes were studied. Apoptosis was analyzed in podocytes, and mRNA level quantiﬁcation, Western blot analysis, and immunoﬂuorescence were developed to quantify podocyte-speciﬁc molecules and Rab-Rabphilin components (Rab3A, Rab27A, and Rabphilin3A). Quantitative RT- PCR was performed on urinary cell pellet from patients. The results showed that differentiated cells had reduced protein levels of the Rab-rabphillin system compared with undifferentiated cells. After glucose overload and angiotensin II treatment, apoptosis was in- creased and podocyte-speciﬁc proteins were reduced. Rab3A and Rab27A protein levels were increased under glucose overload, and Rabphilin3A decreased. Furthermore, this system exhibited higher levels under stress conditions in a manner of angiotensin II dose and time treatment. Immunoﬂuorescence imaging indicated different expression patterns of podocyte markers and Rab27A under treatments. Finally, Rab3A and Rab27A were increased in patient urine pellets and showed a direct relationship with albuminuria. Collectively, these results suggest that the Rab-Rabphilin system could be involved in the alterations observed in injured podocytes and that a mechanism may be activated to reduce damage through the vesicular transport en-hancement directed by this system. polyclonal polyclonal monoclonal


INTRODUCTION
Kidney injury, characterized by a progressive decrease in glomerular filtration rate and/or an increase in urinary albumin excretion (UAE) (10,40), is a frequent complication associated with diabetes mellitus (DM) and hypertension (HTN) (58,67). Despite efforts to prevent and treat it, the incidence of renal injury is increasing, often associated with alterations in the glomerular filtration barrier, in which podocytes play a central role in preserving structure and function (53,63). Podocytes are well-differentiated and highly polarized cells with an elaborate cytoarchitecture (15). Due to the complexity of their structure and function, the mechanisms involved in their injury are still not fully understood, but their involvement in the pathophysiology of renal diseases has been previously described (41,46,68). In this regard, in vitro studies have revealed that exposure to glucose and angiotensin II (ANG II) induces alteration of specific proteins and activation of intracellular pathways leading to podocyte injury (35,56).
One of the most important processes in podocytes is membrane trafficking, essential for maintaining cell polarization and for close cell-to-cell communication. Altered membrane trafficking may be the cause and/or result of podocyte injury. In this sense, Rab proteins, a group of monomeric small GTPases, are key counterparts in regulating vesicle formation, controlling both the specificity and directionality of intracellular transport (12,47,71). They are critical regulators of secretory vesicle exocytosis and have been widely described in synaptic neurons (13,33). Specifically, recent literature reports that the Rab proteins Rab3A and Rab27A cooperate to achieve vesicle exocytosis by recruiting specific effector proteins such as Rabphilin3A to drive vesicle docking at target membranes (48).
However, despite its important function in vesicular traffic, studies analyzing the presence of the Rab-Rabphilin system (Rab3A, Rab27A, and their effector Rabphilin3A) in the kidney are scarce (2,14,49,50). Previous studies by our group have identified an association between the RPH3A gene polymorphism, codifying for Rabphilin3A, and risk of UAE (39). A previous study (23) showed the association of the same polymorphism with UAE in the Framingham cohort. However, further research is needed on the as-yet unclear implications of this system in podocyte injury in the context of DM and HTN.
In the present study, we sought to investigate the presence of the Rab-Rabphilin system in cultured human podocytes and analyze changes in their expression when challenged with elevated glucose concentrations and ANG II. Additionally, we aimed to detect and quantify Rab GTPases mRNA levels in urinary samples from patients with HTN and DM and their association with UAE levels.

Immortalized Human Podocyte Cell Culture
Conditionally human immortalized podocytes (cell line: AB8/13) were kindly provided by Prof. Moin Saleem (Children's Renal Unit and Academic Renal Unit, University of Bristol, Southmead Hospital, Bristol, UK). Particularly in this cell line, at the "permissive" temperature of 33°C, the SV40 T antigen is active, maintaining cells in an undifferentiated proliferative state. Thermoswitching them to the "nonpermissive" 37°C temperature silences the transgene, and the cells become growth arrested and differentiated (51). Briefly, podocytes were cultured on collagen-coated flasks or plates in RPMI-1640 with 10% FBS (Biowest), 1% penicillin-streptomycin, and 1% insu-lin᎑transferrin᎑selenium (GIBCO, ThermoFisher Scientific). Cells were propagated at 33°C until 80% confluence was reached and were trypsinized and replated at proper density. Next, they were allowed to differentiate at 37°C with a lower FBS concentration (2%) for 10 -14 days. All treatment experiments were performed on differentiated podocytes starting at day 10 of differentiation and with a confluence of 80%. Cells were serum starved and then treated with normal glucose (NG; 5.5 mM D-glucose) and high glucose (HG; 30 mM D-glucose) concentrations for 72 h. Differentiated podocytes were also treated with increasing concentrations (0.1, 1, and 2 M) of ANG II (Sigma-Aldrich) for 6, 24, and 48 h for Western blot analyses, 24 h for quantitative RT-PCR, and 48 h for flow cytometry and immunofluorescence analyses.

Optical and Confocal Microscopy Imaging
A Leica DMi1 optical microscope (Wetzlar) and Leica LAS (version 4.9) camera software were used to follow the podocyte differentiation process. Images were taken at different stages of differentiation and also during podocyte treatments. Moreover, measures and morphological parameters were analyzed in treated podocytes (membrane blebbing and area and circularity of cells) using ImageJ software (version 1.46 r, National Institutes of Health). For this analysis, measurements were taken from 150 cells of each experimental group in random and nonoverlapping fields of three independent replicates.
For immunofluorescence analysis, samples were fixed in 4% paraformaldehyde for 15 min at room temperature, washed in cold PBS, and blocked in PBS ϩ BSA (1%) for 30 min at room temperature. After being blocked, samples were incubated overnight at 4°C with the same primary antibodies as for the Western blot analyses: anti-synaptopodin (1/40), anti-nephrin (1/50), anti-Rab3A (1/80), anti-Rab27A (1/50), and anti-Rabphilin3A (1/50) in PBS-Tween 20 (PBS-T) ϩ BSA (1%) buffer. Next, they were incubated in darkness with Alexa-conjugated secondary antibody (Invitrogen) for 1 h at room temperature in the same buffer. Within the last half an hour of the secondary antibody incubation, PBS-T ϩ BSA (1%) with DAPI (1/1,000) was added to identify nuclei. After incubation, samples were washed for 3 ϫ 5 min with 1% PBS ϩ BSA. Cell apoptosis of treated podocytes was assessed using DAPI and determining chromatin condensation, nuclei fragmentation, and nuclear condensation of podocytes following the protocol described by Cummings et al. (5) and analyzing at least 250 cells/sample. Analysis of the F-actin cytoskeleton was assessed by phalloidin staining on differentiated podocytes treated with HG and ANG II concentrations. Cells were fixed in 4% paraformaldehyde for 15 min at room temperature, washed three times with PBS, permeabilized with 0.1% Triton X-100, washed, and incubated with phalloidin-iFluor 594 Reagent (1/1,000, Abcam) for 1 h at room temperature. Within the last half an hour of the incubation period, cells were stained with DAPI. After incubation, cells were washed three times in PBS. Finally, fluorescence samples were observed with Leica confocal microscope DMi8 (Wetzlar), and images were processed using ImageJ (version 1.46 r, National Institutes of Health) software. F-actin fiber orientation was measured with the FibriTool plugin (ImageJ), which quantifies the orientation and anisotropy of fibrillary structures in raw images. Values close to 0 correlate to isotropic or no ordered fibers, and values close to 1 represent perfectly orientated filaments.

RNA Isolation and cDNA Synthesis
RNA was extracted from podocyte and human urine pellets using the miRNeasy mini kit (Qiagen), based on spin column isolation and following the manufacturer's instructions as previously reported (43). After extraction, RNA concentration and purity were confirmed by the 260-to-280-nm absorbance ratio using a NanoDrop 2000 spectrophotometer (ThermoFisher Scientific). Total RNA was reverse transcribed to cDNA using Ready-To-Go You-Prime First-Strand Beads (GE Healthcare) following the protocol indications. cDNA samples were stored at Ϫ80°C until use.

Quantitative RT-PCR
Quantitative RT-PCR was performed using Qiagen Multiplex PCR Master Mix with LC Green reagent (Qiagen) in the LightCycler 480 II real-time PCR system (Roche). A seminested PCR was performed specifically for the Rabphilin3A gene (RPH3A) to improve the sensitivity and specificity of cDNA amplification. Briefly, this process involved the use of two primer sets and two successive PCRs. The first set of primers was designed to anneal to sequences upstream from the second set of primers and was used in the initial PCR. Amplicons resulting from the first PCR were used as a template for a second set of primers (using the same 3= primer as the 3= primer used in the first PCR) and a second amplification step. For RPH3A amplicon identification, we performed high-resolution capillary electrophoresis using QIAxcel (Qiagen), and  Table 1. All assays were run in triplicate, including appropriate controls and the non-template control. Besides these, melting curve analysis was performed to evaluate the specificity of the amplicons generated. ␤-Actin (ACTB) and ␤ 2-microglobulin (B2MG) housekeeping genes were used to normalize podocyte and urine pellet gene expression. Finally, the relative amount for each target gene was calculated by the 2 Ϫ⌬⌬C t comparative method (where Ct is threshold cycle).

Flow Cytometry
After glucose and ANG II treatments, podocyte cell pellets were trypsinized, resuspended, centrifuged for 3 min at 1,300 rpm, and washed with sterile PBS. They were then resuspended in 200 L annexin V binding buffer (Immunostep, Salamanca, Spain) and incubated in darkness for 15 min at room temperature with 5 L annexin V and 5 min with 7 L 7-amino-actinomycin D. Finally, 400 L of annexin binding buffer were added. A total of 10,000 events from each replicate were analyzed by flow cytometer (BD LSRFortessa X-20) with FACSDiva 8.0.1 software (BD Bioscience, Franklin Lakes, NJ).

Human Subjects and Urine Samples
The study population included 64 patients with essential HTN, 30 of them with DM, recruited from the Internal Medicine area of Hospital Clínico Universitario of Valencia, Spain. HTN was defined according to the European Society of Hypertension (systolic blood pressure Ͼ 140 mmHg and/or diastolic blood pressure Ͼ 90 mmHg) (67), and DM was defined according to the World Health Organization (1). Among all patients, 10 patients were diagnosed also with DM (DM no UAE) and 20 patients with DM and increased UAE (DM UAE). Twenty patients were nondiabetic patients (no DM no UAE), and 14 patients nondiabetic with increased UAE (no DM UAE). Table 2 shows clinical characteristics of the patient groups. All patients received antihypertensive treatment at the time of study, and patients with DM were treated with oral antidiabetic agents. The study protocol was approved by the Ethics Committee of the Hospital Clínico Universitario of Valencia in accordance with the Declaration of Helsinki of 1975 as revised in 2013 (69). All patients gave written informed consent.
Patient fresh urine samples were taken from the first morning urine and were processed following our previously reported protocol (44). UAE was measured using a nephelometric immunoassay (Behring Institute). The level of albuminuria for each patient was considered as the mean value obtained from the morning spot urine samples and expressed as the albumin (mg)-to-creatinine (g) ratio (ACR). Increased UAE was defined as an ACR of Ն30 mg/g.

Statistical Analysis
Results are expressed as fold changes and means Ϯ SE for n independent observations, as indicated in the figures. A Kolmogorov-Smirnov test was used to analyze the normal distribution of the variables. Comparisons between two groups were performed with Student's t test or Mann-Whitney U test depending on data distribution, and Pearson's correlation coefficient was calculated to analyze   the association between variables. The level of significance was set at P Ͻ 0.05. Statistical analyses were performed with SPSS software (version 20, SPSS), and graphical plots were created with SigmaPlot (Systat Software).

Identification of Differentiated Human Immortalized Podocytes and Podocyte-Associated Molecules
At the morphological level, proliferating podocytes showed small cell bodies and an epithelial-like cobblestone shape when cultured at 33°C (undifferentiated state). After thermoswitching to 37°C for 10 -14 days, they were differentiated into arborized cells, characterized by enlarged cell bodies with foot processes and spindle-like projections and junctions (Fig. 1A).
Additionally, we analyzed mRNA and protein levels of differentiation and podocyte-specific markers. We observed that synaptopodin and nephrin mRNA levels were higher in differentiated podocytes (4.96-fold increase, P Ͻ 0.001, and 2.12-fold increase, P Ͻ 0.05, respectively), WT-1 levels were lower (2.58-fold decrease, P Ͻ 0.001), and CD2AP showed no significant changes compared with the undifferentiated state (Fig. 1B).

Expression of the Rab-Rabphilin System in Undifferentiated and Differentiated Podocytes
The presence and differential expression of the Rab-Rabphilin subfamily was analyzed in human immortalized podocytes. RAB3A and RAB27A mRNA levels were higher in differentiated podocytes (1.75-fold increase, P Ͻ 0.01, and 1.66-fold increase, P Ͻ 0.05, respectively; Fig. 2A). Focusing on the RPH3A gene, we had to perform a seminested PCR technique with two primer sets to detect its expression in human podocytes, as previously reported (50) (Fig. 2B). We were able to detect RPH3A amplicons in both undifferentiated and differentiated podocytes, as shown by capillary electrophoresis captures performed to identify and measure fragment length (Fig. 2C).

Impact of HG and ANG II in Human Podocytes
Apoptosis-induced mRNA and protein levels of podocytespecific molecules and the Rab-Rabphilin system were analyzed after glucose overload and addition of ANG II in cultured human podocytes.
Apoptosis and actin cytoskeleton rearrangements induced by podocyte treatments. Apoptosis was analyzed in human immortalized podocytes subjected to glucose overload and ANG II treatments by different approaches. Analyses by flow cytom-  etry showed an increase of apoptosis in the HG group compared with the NG group (22.4% vs. 13.9%, P Ͻ 0.05) and a trend towards this under the highest ANG II treatment (2 M) (12.3% vs 9.6%, P ϭ 0.33; Fig. 3A).
Podocyte apoptosis studied by optical microscopy revealed that both treatments, HG and ANG II, induced typical apoptosis morphological alterations, such as membrane blebbing, cell volume reduction, and rounding (Fig. 3, B and C). Additionally, cell apoptosis analyses by DAPI staining of nuclei showed different parameters indicative of apoptosis increased in HG and ANG II-treated podocytes (Fig. 3, D and E).
F-actin fiber staining by rhodamine phalloidin staining was also performed in podocytes treated with elevated glucose and ANG II concentrations. In control podocytes, F-actin was distributed homogenously in bundles along the cell. Exposure to treatments induced cytoskeleton rearrangements that showed a disorganization of fibers evidenced by the anisotropy measurements performed, with reduced values in treated cells compared with control cells. We also observed more grouped F-actin fibers in peripheral regions of treated podocytes (see Supplemental Fig. S1 in the Supplemental Material, available online at https://doi.org/10.6084/m9.figshare.12443282).
Podocyte-specific molecule and Rab-Rabphilin system levels under glucose overload. When we analyzed mRNA levels under HG, the podocyte-associated molecules synaptopodin, WT-1, nephrin, and CD2AP did not reach statistically significant differences (data not shown); however, interestingly, the Rab-Rabphilin system was altered. RAB3A showed lower levels (1.51-fold decrease, P Ͻ 0.05) while RAB27A showed higher levels (1.27-fold increase, P Ͻ 0.05) in the HG group (Fig. 4A). A nested PCR of the RPH3A gene in podocytes subjected to glucose overload demonstrated its expression in treated podocytes as well. We detected the correct amplicon by capillary electrophoresis, but no quantification was performed (Fig. 4B).
Regarding protein changes, reduced levels of the podocyte markers WT-1 (1.20-fold decrease, P Ͻ 0.05), CD2AP (1.79fold decrease, P Ͻ 0.01), and nephrin (1.99-fold decrease, P Ͻ 0.05) as well as a trend in synaptopodin (1.19-fold decrease, P Ͼ 0.05) were observed (Fig. 4C). These results were also confirmed by immunofluorescence in which under NG conditions, synaptopodin displayed a cytoplasmic pattern, extending its expression through podocyte foot processes, and nephrin showed a perinuclear expression pattern. Under HG conditions, synaptopodin revealed a rearranged pattern, not observed under NG conditions, and nephrin showed increased cytoplasmic expression and reduced at the perinuclear region; overall, its protein levels were reduced compared with NG conditions (Fig. 4D).
Interestingly, Rab GTPase proteins also showed protein changes under HG conditions. Rab3A and Rab27A revealed increased protein levels (1.39-fold increase and 1.27-fold increase, P Ͻ 0.05, respectively), whereas Rabphilin3A had reduced levels (1.54-fold decrease, P Ͻ 0.01; Fig. 4E). In immunofluorescence experiments, Rab3A, Rab27A, and Rabphilin3A displayed cytoplasm and membrane dotted expression patterns as expected in vesicular proteins, reaching also foot process extensions, and their levels were similar to those observed by Western blot analysis. Interestingly, under HG conditions, Rab27A revealed a change in its expression pattern compared with NG conditions, showing increased levels close to the podocyte plasma membrane (Fig. 4, F and G).
Podocyte-specific molecules and Rab-Rabphilin system levels under ANG II treatment. With the addition of ANG II, mRNA quantification revealed higher levels of WT-1 (5.40fold increase, P Ͻ 0.05) and CD2AP (1.41-fold increase, P Ͻ 0.05) only at the lowest ANG II concentration (Fig. 5A). Likewise, RAB3A and RAB27A mRNA levels were increased (Fig. 5B).
At the protein level, the podocyte-associated molecules synaptopodin, WT-1, nephrin, and CD2AP showed generally lower levels with ANG II-increased concentrations (Fig. 5, C  and D). These observations were also reflected in immunofluorescence experiments, showing reduced synaptopodin expression and a rearranged pattern and lower levels of nephrin, with less perinuclear staining (Fig. 5E).
The Rab proteins (Rab3A, Rab27A, and Rabphilin3A) showed mainly increased expression during treatments and also showed differences according to ANG II doses and time treatments (Fig. 5, F and G). After ANG II treatment, cells were immunostained, and fluorescence analyses showed similar alterations in concordance with changes at the protein level, with increased Rab3A, Rab27A, and Rabphilin3A signals under ANG II stimulation. Interestingly, as was observed with glucose treatment, Rab27A showed increased membrane expression under ANG II treatment (Fig. 5, H and I).

Levels of the Rab-Rabphilin System in Human Urine Samples From Patients With HTN and Relationship With UAE
Additionally, we analyzed the presence of mRNA levels of these Rab GTPases in urine pellets of patients with and without DM. Table 2 shows clinical characteristics of the patient groups. Higher RAB3A (2.17-fold increase, P Ͻ 0.01; Fig. 6A) and RAB27A levels (1.53-fold increase, P Ͻ 0.05; Fig. 6, C and D) were observed in the elevated UAE patient group than in the nonalbuminuric group. RAB3A was elevated in both patients without and with DM (2.56-and 2.34-fold increases, P Ͻ 0.01 and P Ͻ 0.05, respectively; Fig. 6B). Moreover, UAE levels were positively correlated with both RAB3A (r ϭ 0.515, P Ͻ 0.05) and RAB27A (r ϭ 0.475, P Ͻ 0.05).

DISCUSSION
This study is the first to demonstrate the presence of Rab GTPases of the Rab-Rabphilin secretory system in human podocyte cell cultures and in human urinary samples. Aside from the changes observed in these molecules during podocyte differentiation, exposing them to cellular stresses such as HG and ANG II revealed notable alterations in Rab3A, Rab27A, and Rabphilin3A at gene and protein levels. Likewise, pellets from urine samples of patients with HTN and with and without incipient renal injury and/or DM showed increased expression of Rab mRNAs and association with incipient renal damage.
It is widely recognized that HTN and DM are primary causes of renal damage and that podocyte injury plays a major role in glomerular filtration barrier alterations, contributing to the progression of glomerular diseases (41,68). Consequently, reducing podocyte damage has become a priority aim. The study of associated alterations subjecting podocyte cultures to    Consistent with previous reports (6,31,51), protein levels of podocyte-specific markers (synaptopodin, WT-1, nephrin, and CD2AP) increased in differentiated human podocytes, and changes under different treatments were also observed. Concerning mRNA levels, we detected an increase in all markers except for CD2AP and WT-1, showing reduced levels of WT-1 in differentiated podocytes. Other authors have described no WT-1 changes or slight increases during differentiation (6, 51), so our finding could be due to different compensatory regulation mechanisms for the increased protein levels at the end of the differentiation process. These discrepancies have been previously reported as being attributed to different half-lifes and production rates between mRNA and proteins and to a strong regulatory role for processes downstream transcription that can produce, among others, alterations in degradation rates of proteins (8,34,38,66). Also important in these regulation mechanisms is the influence of small RNAs such as microRNAs, key regulators of gene expression (17).
Previous studies have highlighted the involvement of important podocyte metabolic pathways and alterations produced in cytoskeleton components in the onset and progression of renal diseases (22,41,45). In this study, we contribute to this research by showing reduced levels of podocyte-associated structural proteins under HG and ANG II treatments. Our  group has previously demonstrated alterations in podocytespecific markers in renal damage and urine from patients with HTN, revealing them as potential biomarkers of the disease (43,44). Moreover, this study has brought novel insights to renal disease progression by deciphering new implications of the trafficking machinery encrypted by the Rab GTPase subfamily.
Most Rab GTPases studies are focused on the nervous system, and multiple research has associated these molecules with neurodegenerative disorders such as Alzheimer disease (3,24,27). In epithelia, Rab GTPases contribute toward generating polarity by regulating the trafficking of junctional proteins and by defining epithelial transport circuits and recycling endosomes (55). Rab GTPases are recognized by proteins called effectors in their GTP-bound state (29). Specifically, Rab3A plays a central role in neuron synaptic vesicle exocytosis (54,61), a function carried out in cooperation with its effector Rabphilin3A, which interacts with cytoskeletal proteins (4,7,26). Additionally, Rabphilin3A also acts as an effector of Rab27A, whose main function has been described in the exosome secretion pathway (11,28,42). The secretory system enclosed by these three Rabs has previously been elucidated, establishing a relation between these proteins in the context of regulated secretion in exocytosis pathways (48,65).
In podocytes, vesicular trafficking directed by Rab GTPases represents a key process to maintain their polarized cell bodies, for foot processes formation, and also for the recycling and cell communication pathways (9,59,70). A study carried out by Giardino et al. (14) showed that Rab3A knockout produced exocytosis dysregulation and podocyte cytoskeleton alterations, supporting the importance of the Rab-Rabphilin system in vesicular transport of podocytes.
In differentiated podocytes, we showed a decrease of protein levels of Rab3A, Rab27A, and their effector Rabphilin3A and an increase in mRNA levels compared with the proliferative stage, suggesting a compensatory mechanism for the reduction observed, as explained above (8,17,34,38,66). Moreover, under HG and ANG II conditions, the decreased protein levels of podocyte differentiation markers together with the increased Rab3A and Rab27A protein levels (contrary to levels at differentiation) evidence that dedifferentiation may be occurring and that the Rab-Rabphilin system may be influencing on this process.
Podocyte dedifferentiation has been previously reported by our group in patients with HTN and lupus nephritis, in which we found viable podocytes secreted in urine and podocyte marker changes that suggested a dedifferentiation rather than apoptosis process (43,44  observed this process in stressed human podocytes subjected to transforming growth factor-␤ and ANG II treatments, but, interestingly, our research points out novel players in this dedifferentiation: Rab3A, Rab27A, and Rabphilin3A membrane-trafficking GTPases. The apoptosis assays performed on podocytes under HG and ANG II treatments also reinforced observations previously confirmed by other authors (16,19,20). Complementing these results, morphological analyses of treated podocytes showed an increase in apoptosis characteristic features in cells exposed to both treatments. We also revealed the presence of apoptotic nuclei in DAPI-stained micrographs, with nuclei and chromatin condensations. However, the lack of statistically significant changes in apoptosis assays under ANG II treatment surprisingly suggests that this might not be a question of marked apoptotic cell death but of initial dedifferentiation followed by cell death, as previously reported by other authors in human podocyte cultures (32,52).
Additionally, F-actin fluorescence staining showed a cytoskeleton disorganization in podocytes subjected to HG and ANG II treatments and a grouping of fibers at the cell periphery and more diffuse in the cytoplasm. Alterations of podocyte structure have been widely observed in stress situations as glucose and ANG II overload and evidence the importance of these disruptions in podocyte injury (21, 25, 36, 37, 57, 64).
Immunofluorescence analyses revealed altered levels and distinct subcellular distributions of podocyte markers and Rab proteins under stressed conditions. Synaptopodin expression was rearranged under HG conditions, as previously observed (30). Nephrin expression changed from a perinuclear to more cytoplasmic pattern and was reduced under HG and ANG II exposure, similar to a previous study (62). These protein pattern changes and reduced expression also support that dedifferentiation occurred in these cells. Strikingly, the Rab GTPases images showed a particular Rab27A expression pattern. In contrast to the pattern observed in normal conditions with cytoplasmic distribution, a higher expression near the podocyte membrane was observed under HG and ANG II treatments. Taking into account Rab27A function in vesicle exocytosis, these observations suggest accentuated exosomal traffic under these stresses, an interesting area to explore in future research.
Consistent with our observations in podocyte cultures, we quantified RAB3A and RAB27A mRNAs in urine samples from patients with HTN and with and without DM and/or renal injury. Our findings showed an increase in both mRNA expression levels in patients with kidney disease and also a direct relationship between UAE levels and Rab mRNAs. In the case of RAB3A, this increase was significant in both albuminuric groups (with and without associated DM), which suggests that  Rab3A could be involved or at least indicate the presence of incipient renal damage. This finding is consistent with the observation made by Rastaldi et al. (50), showing increased Rab3A in the glomerular tissue of a few patients with proteinuric diseases. Related to this, Rab3A knockout showed enhanced UAE excretion and glomerular and podocyte injury under HG (2,14). Moreover, in Rab27A levels, this increase was statistically significant when analyzing both albuminuric groups together, suggesting that Rab27A plays a less potent but still important role in renal injury. Rab27A has also been studied at the renal level but has not been associated with albuminuria (60,70).
In summary, in the present study, we demonstrate, for the first time, the presence of the Rab-Rabphilin system in both human urine samples and cultured human podocytes. The increase and different distribution patterns of the Rab-Rabphilin system in glucose-and ANG II treated podocytes indicates that enhanced vesicular trafficking through the Rab-Rabphilin system could be taking place in podocyte injury. In addition, we found important expression changes of this system in patients with HTN and DM with renal injury as well as an association between these Rab GTPases and albuminuria. Further experimental studies will be required to elucidate if modifying the Rab-Rabphilin system could result in risk reduction of kidney damage. Finally, our results suggest that Rab system GTPases may be orchestrating the alterations observed in podocytes during insults, conditioning the development and progression of renal damage.