Clinical Research

Urinary Extracellular Vesicles of Podocyte Origin and Renal Injury in Preeclampsia

Sarwat I. Gilani, Ulrik Dolberg Anderson, Muthuvel Jayachandran, Tracey L. Weissgerber, Ladan Zand, Wendy M. White, Natasa Milic, Maria Lourdes Gonzalez Suarez, Rangit Reddy Vallapureddy, Åsa Nääv, Lena Erlandsson, John C. Lieske, Joseph P. Grande, Karl A. Nath, Stefan R. Hansson and Vesna D. Garovic
JASN November 2017, 28 (11) 3363-3372; DOI: https://doi.org/10.1681/ASN.2016111202

Abstract

Renal histologic expression of the podocyte-specific protein, nephrin, but not podocin, is reduced in preeclamptic compared with normotensive pregnancies. We hypothesized that renal expression of podocyte-specific proteins would be reflected in urinary extracellular vesicles (EVs) of podocyte origin and accompanied by increased urinary soluble nephrin levels (nephrinuria) in preeclampsia. We further postulated that podocyte injury and attendant formation of EVs are related mechanistically to cellfree fetal hemoglobin (HbF) in maternal plasma. Our study population included preeclamptic (n=49) and normotensive (n=42) pregnant women recruited at delivery. Plasma measurements included HbF concentrations and concentrations of the endogenous chelators haptoglobin, hemopexin, and α1- microglobulin. We assessed concentrations of urinary EVs containing immunologically detectable podocyte-specific proteins by digital flow cytometry and measured nephrinuria by ELISA. The mechanistic role of HbF in podocyte injury was studied in pregnant rabbits. Compared with urine from women with normotensive pregnancies, urine from women with preeclamptic pregnancies contained a high ratio of podocin-positive to nephrin-positive urinary EVs (podocin+ EVs-to-nephrin+ EVs ratio) and increased nephrinuria, both of which correlated with proteinuria. Plasma levels of hemopexin, which were decreased in women with preeclampsia, negatively correlated with proteinuria, urinary podocin+ EVs-to-nephrin+ EVs ratio, and nephrinuria. Administration of HbF to pregnant rabbits increased the number of urinary EVs of podocyte origin. These findings provide evidence that urinary EVs are reflective of preeclampsia-related altered podocyte protein expression. Furthermore, renal injury in preeclampsia associated with an elevated urinary podocin+ EVs-to-nephrin+ EVs ratio and may be mediated by prolonged exposure to cellfree HbF.

Preeclampsia is a multisystem pregnancy disorder characterized by hypertension and proteinuria,1 in which endothelial dysfunction is a central pathogenic process.2 Glomerular endotheliosis, the characteristic renal lesion in preeclampsia, is characterized by occlusion of capillary lumens, glomerular endothelial swelling, and loss of endothelial fenestrations.3 Over the last decade, evidence has increasingly spotlighted injury to podocytes, the principal determinant of glomerular permselectivity, as a critical contributor to proteinuria in preeclampsia.4 Podocytes are terminally differentiated cells located on the side of the glomerular basement membrane that faces the urinary space.5 Podocytes interdigitate via their foot processes, which connect via specialized cell-to-cell junctions to form glomerular slit diaphragms. The slit diaphragm appears to be a modified adherens junction that provides the main size selective filtration barrier in the kidney.

Several lines of evidence suggest that both podocytes and podocyte-specific proteins are present in urine obtained from women at the time of the diagnosis of preeclampsia, thereby supporting podocyte loss and injury as the mechanism of proteinuria in preeclampsia.6–8 Studies of human tissue have shown that the expression of nephrin in women with preeclampsia is decreased compared with women with either normotensive or chronic hypertensive pregnancies,9,10 as well as in podocytes recovered from the urine obtained from preeclamptic women.7 On the other hand, studies have found elevated soluble urinary nephrin (i.e., nephrinuria) levels in women with preeclamptic pregnancies compared with those with normotensive pregnancies, suggesting that nephrin shedding occurs in preeclampsia.8,11 Increased nephrinuria was found to be positively correlated with the degree of proteinuria, suggesting that nephrin shedding is associated mechanistically with disruption of the slit diaphragm.

We hypothesized that renal injury in preeclampsia, demonstrated by the presence of proteinuria and renal dysfunction, is associated with the presence of urinary extracellular vesicles (EVs) of podocyte origin. We posited further that these EVs reflect podocyte-specific protein dysregulation previously described in renal histologic sections of women with preeclampsia compared with normotensive controls: nephrin shed into the fluid phase urine would account for nephrinuria and decreased nephrin expression in EVs and in the kidney itself. EVs are small (0.03–1 µm) membrane-enclosed sacs shed from activated or injured cells, and have been identified in many body fluids, including urine.12 In this study, podocyte injury was assessed by the concentrations of urinary EVs containing podocyte-specific proteins using digital flow cytometry, a methodology that was adopted from studies detecting EVs in blood,13 and which has been applied recently to urine studies.14

In our effort to delineate a cause for podocyte injury in preeclampsia, our attention was drawn to both clinical and experimental evidence pertaining to fetal hemoglobin (HbF). Clinical studies have demonstrated that plasma cellfree HbF levels are elevated in preeclamptic pregnancies as early as the first trimester.15,16 The pathophysiologic significance of such observations has been suggested by experimental studies demonstrating that renal injury in preeclamptic models in sheep and rabbits may be instigated by starvation-induced hemolysis and species-specific cellfree HbF infusions, respectively.17,18 We thus linked these observations back to podocyte injury in preeclampsia, and postulated that levels of cellfree HbF in maternal plasma, along with derangements in mechanisms that protect against hemoglobin/heme toxicity (haptoglobin, hemopexin, and α1-microglobulin), would correlate with maternal disease, renal dysfunction, and podocyte injury. The mechanistic link between cellfree HbF and podocyte damage/formation of EVs of podocyte origin was studied in HbF-treated pregnant rabbits. This animal model recapitulates several renal manifestations of preeclampsia, including albuminuria, disruption of the glomerular filtration barrier, and podocyte dysregulation.18

Results

Demographic and Clinical Variables

This study was approved by the Lund University Hospital Institutional Review Board and all participants gave written informed consent. A total of 91 pregnant women were included in this study, 49 with preeclampsia and 42 with normotensive pregnancies (Table 1). Women with preeclampsia did not differ from those with normotensive pregnancies with respect to age, body mass index, or nulliparous status, but they delivered at an earlier gestational age and were more likely to undergo induction of delivery. Their infants were more likely to have lower birth weights.

View this table:
Table 1.

Characteristics of women with normotensive and preeclamptic pregnancies

Table 2 summarizes the biochemical characteristics of the women with preeclampsia versus those with normotensive pregnancies at the time of delivery. Kidney function, as evaluated by both cystatin C and serum creatinine, was decreased in women with preeclampsia. Uric acid levels were elevated as expected, on the basis of previous reports.19 The concentrations of cellfree HbF and α1-microglobulin were elevated in preeclamptic women, whereas plasma levels of haptoglobin and hemopexin levels were decreased.

View this table:
Table 2.

Biochemical characteristics in plasma, EVs in urine that stain for podocyte-specific proteins and nephrinuria in normotensive versus preeclamptic pregnancies at the time of delivery

Validation of Sample Processing for EV Studies in Normotensive and Preeclamptic Pregnancies

This study was approved by the Mayo Clinic Institutional Review Board, and all participants gave written informed consent before inclusion in the study. The effects of centrifugation and freezing/thawing on EV numbers were studied in five women with preeclampsia and five normotensive controls (Supplemental Table 1). Centrifugation led to significant decreases in both podocin-positive and nephrin-positive EVs obtained from the urine of preeclamptic women, but not from those with normotensive pregnancies (Figure 1). Podocin-positive and nephrin-positive EVs were obtained from freshly collected unspun urine, then subjected to freezing at −80°C and thawing at two time points, and reanalyzed at 1 week (one cycle of freezing and thawing) and 2 weeks (two cycles of freezing and thawing) after collection. No significant change was artifactually induced by the freeze/thaw cycles on podocin-positive and nephrin-positive EVs in the urine samples from either preeclamptic or normotensive women (Figure 2). Thus, we concluded that bio-banked samples that had been stored unspun at −80°C since collection would be adequate for the aims of this study.

Figure 1.
Figure 1.

Centrifugation of urine samples from women with preeclampsia results in decrease in the number of extracellular vesicles. (A) Podocin-positive EVs, expressed as median (25th–75th percentile), in women with normotensive pregnancies: unspun urine, 348.5 (187–2473) versus spun urine, 401.5 (85–919); in women with preeclamptic pregnancies: unspun urine, 3150 (2800–3400) versus spun urine, 274 (256–539). (B) Nephrin-positive EVs, expressed as median (25th–75th percentile), in women with normotensive pregnancies: unspun urine, 185 (61–538) versus spun urine, 114.5 (32–281); in women with preeclamptic pregnancies: unspun urine, 997 (525–1209) versus spun urine, 111 (77–129). PE, preeclampsia.

Figure 2.
Figure 2.

Freeze/thaw cycles of urine samples from women with either normotensive or preeclamptic pregnancies do not affect the number of extracellular vesicles. (A) Podocin-positive EVs, expressed as median (25th–75th percentile), in freshly collected unspun urine and after freezing at −80°C and thawing on two occasions, 1 and 2 weeks after collection. Normotensive pregnancies: 348.5 (187–2473), 357 (189–2390), and 355 (194–2462), respectively; preeclamptic pregnancies: 3150 (2800–400), 3178 (3040–3178), and 3393 (3332–3393), respectively. (B) Nephrin-positive EVs, expressed as median (25th–75th percentile), in freshly collected unspun urine and after freezing at −80°C and thawing on two occasions, 1 and 2 weeks after collection. Normotensive pregnancies: 185 (61–538), 181 (63–539), and 171 (66–477), respectively; preeclamptic pregnancies: 997 (525–1209), 1002 (577–1379), and 1022 (622–1423), respectively. PE, preeclampsia.

Characterization and Quantification of Urinary EVs and Nephrinuria in Preeclamptic and Normotensive Pregnancies

Phosphatidylserine (annexin-V)-positive EVs were increased significantly in the urine of women with preeclampsia compared with women with normotensive pregnancies (Table 2). The ratio of podocin-positive EVs to nephrin-positive EVs (Table 2) was increased significantly in women with preeclampsia at the time of delivery. Median values for podocin and nephrin are different in the study cohort compared with the validation cohort. These differences likely reflect the smaller number of patients in the validation cohort. However, the ranges are similar, particularly when comparing the patient cohort (unspun urine samples) to validation samples that were unspun.

Nephrinuria (expressed either in nanograms per milliliter or nanograms per milligram of creatinine) was significantly higher in preeclamptic compared with normotensive women (P<0.001).

Correlations between markers of renal injury, including proteinuria, podocin-positive to nephrin-positive EVs ratios, and nephrinuria with clinical and biochemical parameters at the time of delivery are presented in Table 3. Most notably, positive correlations were observed with the renal functional indices (cystatin C), with markers of preeclampsia severity, including systolic BP, diastolic BP, and uric acid levels. All three markers of renal injury demonstrated negative correlations with plasma hemopexin levels (Table 3). Finally and notably, there was a positive correlation between proteinuria and the podocin-positive to nephrin-positive EVs ratio (rho=0.404; P<0.01) and between proteinuria and nephrinuria (rho=0.846; P<0.001).

View this table:
Table 3.

Correlation between markers of renal injury (proteinuria, urinary podocin-positive to nephrin-positive EVs ratio, and nephrinuria) with clinical parameters and plasma biochemistriesa

Total hemoglobin may contribute to the overall heme burden, particularly in hemolysis, elevated liver enzymes, and low platelet count syndrome (HELLP). The low number of women with HELLP in the cohort (n=5) did not allow for a meaningful subgroup analysis.

The Rabbit Model of Preeclampsia

A total of eight pregnant rabbits were studied. The animals were euthanized on gestational day (GD) 29 after being injected every other day from GD 20 with either cellfree HbF (20 mg/kg, n=5) or buffer only (n=3).

Urinary albumin, expressed as median (25th–75th percentile), increased significantly from GD 20–29 in the HbF-treated group, from 2.90 µg (1.84–4.44) to 50.45 µg (6.05–132.48; P=0.04), but did not increase in the control group, from 0.85 µg (0.37–3.57) to 6.40 µg (5.27–17.71; P=0.11). Both annexin-V and podocin-positive EVs were increased significantly in the HbF-treated rabbits versus the controls (Figure 3, A and B, respectively). Electron microscopy of kidney sections revealed the presence of endotheliosis in the HbF group, but not in the controls (Figure 4).

Figure 3.
Figure 3.

Administration of fetal hemoglobin to pregnant rabbits results in increase in urinary extracellular vesicles. (A) Annexin-V–positive and (B) podocin-positive EVs, expressed as median (25th–75th percentile), were significantly increased in HbF compared with control animals: 39,094 (21,719–43,537) versus 8808 (5760–10,525); P=0.04, and 1850 (1241–3569) versus 543 (230–719); P=0.04, respectively.

Figure 4.
Figure 4.

Electron microscopy analysis of renal sections from HbF-treated rabbits showing endotheliosis. HbF-treated rabbit (A) demonstrated endotheliosis (arrow), which was absent in the control (B).

Discussion

Our data demonstrate significant differences at the time of delivery in the number of urinary EVs of podocyte origin in women with preeclamptic versus those with normotensive pregnancies. The podocin-positive to nephrin-positive EVs ratio, as well as nephrinuria, were higher in preeclamptic versus normotensive women. Podocyte-specific protein expressions in the urinary EVs from women with preeclampsia were reflective of those described in renal sections, specifically decreased nephrin expression.9 Coupled with the urinary measures of soluble nephrin, these findings suggest that nephrin shedding from podocytes occurs in preeclampsia, with a resultant increase in nephrinuria. Concentrations of plasma cellfree HbF were elevated in preeclampsia as reported previously.20 Furthermore, negative correlations were observed between the podocin-positive to nephrin-positive EVs ratio and nephrinuria, and plasma levels of hemopexin. These findings suggest that the greater consumption of this heme-binding protein may reflect the higher levels of HbF (and the attendant release and increased presence of free heme) to which the podocyte is exposed, with consequent podocyte injury (Figure 5). The mechanistic link between HbF and generation of EVs of podocyte origin was confirmed by demonstrating that the administration of HbF to pregnant rabbits increased numbers of annexin-V and podocin-positive EVs.

Figure 5.
Figure 5.

Renal injury in preeclampsia is associated with the presence of urinary extracellular vesicles of podocyte origin and is related to fetal hemoglobin.

Injury to and loss of podocytes are increasingly implicated as mechanisms underlying kidney disease in preeclampsia. Differential expressions of the podocyte-specific proteins that contribute to the structural and functional integrity of the glomerular slit diaphragm have been reported in studies of renal tissue from women with preeclamptic versus those with normotensive pregnancies.9,10 The expression of nephrin was decreased in preeclampsia, whereas the expression of podocin was comparable. These changes in renal tissue may be reflected in the urine in two ways. First, podocyte detachment and shedding in the urine (i.e., podocyturia) may be best identified by podocin, as its expression is unaffected in renal tissue sections in preeclampsia.9 Consequently, podocin expression is likely to be preserved in detached podocytes. Our initial study of podocyturia in preeclampsia indicated that staining for podocin, compared with other podocyte-specific proteins, was superior in identifying urinary podocytes in preeclampsia.6 Second, the downregulation of nephrin in renal tissue may be, at least in part, due to urinary nephrin losses, i.e., nephrinuria. Studies of urinary supernatants from preeclamptic women using a nephrin ELISA showed that urine nephrin levels were elevated compared with levels in normotensive pregnant women,8,11 and that urine nephrin levels correlated with proteinuria, diastolic BP, and renal dysfunction.11 Our results, obtained by concurrent staining for two podocyte-specific proteins (i.e., nephrin and podocin) expressed by the same EVs of podocyte origin, indicate that EVs of podocyte origin reflect podocyte-specific protein dysregulation that was previously described in the renal sections, i.e., decreased nephrin expression.9 Together with elevated nephrinuria in women with preeclampsia, these findings suggest that nephrin shedding from podocytes is related mechanistically to renal injury in preeclampsia. A trend toward an increased number of podocin-positive EVs in urine from preeclamptic compared with normotensive women (Table 2), although not statistically significant, may be explained by an increased number of urinary podocytes in preeclampsia.

Urinary EVs originate from the cells facing the urinary space and contain cargo (protein, lipids, and microRNA) representative of their cells of origin.21 On the basis of size, content, and biogenesis, EVs can be classified further into exosomes, microvesicles, and apoptotic bodies. The term EV is preferred, however, because of the overlapping physical and biologic properties of these three entities, and the lack of scientific accord regarding how these entities are best defined.22 In addition, no consensus exists regarding rigorous and recommended methods (gold standards) for the isolation and/or purification of EVs from body fluids, including urine.23,24 Although removal of the cells is considered the preferred way for sample preparation, an article from the International Society for Extracellular Vesicles23 has recognized that use of clinical samples containing cells that may subsequently release EVs might be desirable in some instances, depending upon the issue under investigation. This applies to preeclampsia and, indeed, our laboratory has long sought methods that entail less cost and are more reproducible in documenting the presence of urinary podocytes than currently available cytologic tests. We confirmed that centrifugation results in a significant decrease in the number of EVs in preeclamptic urine, whereas freezing/thawing cycles do not affect the number of EVs positive for podocyte-specific proteins. These observations open new venues for studies of EVs and their cargos using historical samples that are adequately collected and stored, including those that were analyzed in this study. Another challenge in the field relates to the optimal, scientifically justified method for expressing and normalizing changes in the number and content of EVs. Resolution of these analytic challenges would facilitate not only adequate comparisons between the study groups, but comparisons across studies.12,25 In this study, we calculated the ratio between podocin-positive EVs and nephrin-positive EVs, which may serve as a marker of disease activity, similar to a previous study of progression of renal disease in an animal model of glomerulosclerosis,26 which indicated that an elevated podocin/nephrin mRNA ratio is associated with renal injury. Our results are consistent with recent findings that urinary nephrin mRNA expression is reduced in podocytes recovered from preeclamptic pregnancies, and that the podocin/nephrin mRNA ratio increases significantly with increasing proteinuria.27

A recent study comprehensively identified a variety of urinary EVs positive for cell-specific markers from different nephron segments, including podocytes, parietal cells, proximal tubule, thin and thick loop of Henle, distal tubule, and collecting duct.14 Urinary EVs, to date, have been studied as markers of renal disease in several diseases, including glomerular diseases such as FSGS28 and diabetic nephropathy.29 Although preeclampsia has been associated with both glomerular and tubular injury, we did not study urinary EVs that may originate from other cell types/nephron segments because analysis of EVs from other renal cells and from specific nephron segments is an involved and extended undertaking beyond the purview of this study, which specifically focused on the involvement of the podocyte. Analyses of EVs from other renal cells are of clear and relevant interest and will be pursued in future studies. An increase in podocyte-specific EVs is viewed as a marker of direct glomerular injury, whereas a reduction may herald podocyte loss and related chronic disease. Our study extends previous findings to acute renal injury in preeclampsia by characterizing podocyte-specific EVs and by implicating HbF as the proximate cause for such injury.

We found that the concentrations of cellfree HbF and α1-microglobulin were elevated in preeclamptic women, whereas haptoglobin and hemopexin were decreased. This is consistent with findings reported recently for the original study group consisting of 98 patients with preeclampsia and 47 normotensive controls20; from this study population, a subgroup with urine samples was identified (49 with preeclampsia and 42 with normotensive pregnancies) and served as the study population for this investigation. An acute increase in glomerular permeability through oxidative stress has been reported in response to HbF in a rat kidney perfusion model.30 Additional evidence for the injurious effects of HbF was provided from a model of preeclampsia in rabbits, induced by species-specific, cellfree HbF infusions.18 Podocytes exhibited ultrastructural evidence of mitochondrial and endoplasmic reticulum swelling, along with apoptosis. Using the same model, this study provides evidence that HbF infusion leads to quantifiable changes in urinary EVs of podocyte origin, thus mechanistically linking HbF with podocyte damage in preeclampsia. Constituents of the hemoglobin molecule, such as its tetrapyrrole heme prosthetic group and the iron it contains, are cytotoxic because of their oxidant, inflammatory, and proapoptotic effects.17,18,31 The potential for HbF to inflict cell injury needs to be viewed within the context of the endogenous elimination pathways that bind hemoglobin (haptoglobin), sequester its free heme moiety (hemopexin and α1-microglobulin), and those that degrade heme, heme oxygenase 1, and heme oxygenase 2. The potential roles of these pathways in pregnancy are generating increasing interest. For example, elevated heme levels, achieved by either the administration of exogenous heme in wild-type mice or by using heme oxygenase 1 knockout mice, recapitulated several critical features of preeclampsia, including suboptimal placentation followed by intrauterine growth restriction and fetal lethality.32 The classic preeclamptic renal lesion of endotheliosis was present in the pregnant ewe preeclampsia model, whereby starvation led to the signs and symptoms of preeclampsia via hemolysis.17 Renal endotheliosis was absent in rescue experiments in which starved animals were treated with the heme scavenger, α1-microglobulin. This further supports the role of free heme in inducing renal injury in preeclampsia,17 and raises the exciting possibility of α1-microglobulin as a therapeutic agent in preeclampsia.

Our study has several important strengths. It utilized well defined clinical samples, and all urinary analyses were performed in a blind manner. These results, together with mechanistic data from HbF-treated pregnant rabbits, provide preliminary evidence that heme-induced renal toxicity may be a mechanism of renal injury in preeclampsia. It sets the stage for future studies that will aim to characterize the content of EVs in preeclampsia, to determine whether urinary EVs expressing podocyte proteins predate proteinuria in preeclampsia, and to study the dynamics of EVs after preeclamptic pregnancies as a marker of ongoing renal injury and ensuing CKD.

Concise Methods

Patient Recruitment Strategy and Study Sample Collection

The study was approved by the Mayo Clinic and Lund University Hospital institutional review boards and all participants gave written informed consent. Participants were identified retrospectively from an ongoing prospective Swedish study of women diagnosed with preeclampsia and normotensive pregnancies.20 The study began in 2006 and has recruited close to 900 participants to date. A random sample of 98 women with preeclampsia and 47 women with normotensive pregnancies (recruited between 2006 and 2011) initially was identified for a study of heme-related metabolites as potential biomarkers of preeclampsia.20 This study included 91 of these 145 study participants for whom urine samples were collected up to 24 hours before delivery and were available for EVs analyses: 42 women with normotensive pregnancies and 49 with preeclampsia.

Normotensive, preeclamptic, eclamptic, and HELLP pregnancies were defined on the bases of published criteria.1 Exclusion criteria included gestational hypertension, essential hypertension, and/or gestational diabetes. Maternal venous blood and random urine samples were collected before delivery. Random urine samples were collected in 10-ml Sarstedt tubes. For protein stabilization, 300 μl of protease inhibitor cocktail was added to all tubes, as previously described.33,34 Urine samples were immediately stored at −80°C. Urine samples were sent on dry ice to the research laboratory at Mayo Clinic for the analyses of urinary EVs positive for podocyte-specific proteins, which were performed in a blind fashion.

Characterization and Validation of Sample Processing for EV Studies in Normotensive and Preeclamptic Pregnancies

Before analyzing historical unspun samples, we conducted a validation study of the effects of (1) centrifugation and (2) freezing/thawing cycles on EV numbers. To that end, five women with preeclampsia and five normotensive controls who delivered at Mayo Clinic during the study period were recruited (Supplemental Table 1). Random spot urine samples of approximately 100 ml each were collected and immediately separated into two 50-ml aliquots, one of which was centrifuged and the other was left unspun. One milliliter aliquots were taken from each, and immediately processed for podocin- and nephrin-positive EVs. The rest of the spun and unspun urines were frozen at −80°C and thawed and reanalyzed on two occasions, 1 and 2 weeks after collection. Podocin- and nephrin-positive EVs were characterized as described below.

Characterization and Quantification of Urinary EVs and Nephrinuria

Urine Sample Preparation for EVs Analysis by Digital Flow Cytometry

Mayo Clinic investigators received urine samples from 91 Swedish study participants: 42 women with normotensive pregnancies and 49 women with preeclampsia. Two-milliliter aliquots of unspun urine samples from each participant were received on dry ice and immediately were stored at −80°C until the time of analysis. All clinical identifiers were removed to ensure that samples were analyzed in a blinded fashion.

Frozen urine samples were thawed in a 37°C water bath for 5 minutes at the time of processing. Each sample was thawed twice. After the first thaw, a concentration check was performed and the respective samples were immediately frozen at −80°C. After the second thaw, samples were stained with a predetermined set of antibodies for flow cytometric analysis using our standardized methodology.14 Each urine sample was stained for annexin-V, an EV marker,35 and for podocyte-specific proteins, including nephrin and podocin. While nephrin is a transmembrane protein,36 podocin forms a membrane-associated hairpin-like structure with the N and C terminal domains facing the cytosolic side of the slit diaphragm.37 Previous studies have shown that the podocyte membrane surface stains for podocin without previous permeabilization.38

Digital Flow Cytometry Analysis of Urinary EVs

A digital flow cytometer (FACSCanto) was used to analyze urinary EVs (Figure 6). The flow cytometer settings and gates for recording events and analysis were recently published.14 The absolute counts of single and double fluorescence-labeled urinary EVs were expressed as urinary EV per microliter of urine, using the standardized method previously described.13,14,39 The number of podocin-positive EVs was normalized to nephrin-positive EVs and expressed as a podocin-positive to nephrin-positive EVs ratio, as previous studies have indicated that the nephrin mRNA/podocin mRNA ratio may serve as a marker of disease activity/progression.26 Furthermore, using the podocin-positive to nephrin-positive EVs ratio reduces test variability, which is commonly encountered with urine analytes because of changes in osmolality, pH, and contaminants.

Figure 6.
Figure 6.

Digital flow cytometry analysis of urinary EVs indicating an elevated podocin-positive to nephrin-postive EVs ratio in preeclampsia. Normotensive pregnant woman on the left, preeclamptic pregnant woman on the right, with a calculated podocin-positive to nephrin-positive EVs ratio of 1.56 and 17.57, respectively.

Measurements of Urinary Nephrin Concentrations (Nephrinuria)

Nephrinuria was measured using ELISA from Exocell (Philadelphia, PA), according to the manufacturer’s instructions. Briefly, urinary samples were diluted to 1:10 by the buffer provided with the ELISA kit. The range for the standard curve was 2–2000 ng/ml. All samples were measured in duplicate, with a within assay variation of <10%.

Plasma Analytes

Plasma uric acid, creatinine, and cystatin C concentrations were measured using standard methods on a Cobas 6000 (Roche Diagnostics Limited, Rotkreuz, Switzerland) in the Clinical Chemistry Laboratory at the Skåne University Hospital in Lund, Sweden. Cellfree HbF, haptoglobin, α1-microglobulin, and hemopexin plasma concentrations were determined as described previously.20

For a more detailed description of the methods, see the Supplemental Material.

Animal Experiments

The study was approved by the Ethical Committee for Animal Studies at Lund University, Sweden (permission no: M58–12). Nineteen pregnant rabbits were used for the experiments, which showed that α1-microglobulin treatment improves preeclampsia-like symptoms induced by cellfree HbF.18 This study included eight of these animals: five HbF-treated rabbits and three controls for which we had complete biochemical data and stored urine and kidney tissue specimens. Details of the model are described in the original article.18 Briefly, pregnant rabbits were injected every other day from GD 20 until GD 28 with either cellfree HbF (20 mg/kg) or buffer only. Animals were anesthetized and euthanized on GD 29. Organs, including kidneys, were collected. The remaining kidney samples were sent to the research laboratory at Mayo Clinic for this study after the completion of the primary experiments.18 Urine samples were initially analyzed for albuminuria using a species-specific albumin ELISA (Biosite; Nordic BioSite, Täby, Sweden), then stored at −80°C, and sent on dry ice to the research laboratory at Mayo Clinic for the analyses of urinary EVs positive for podocyte-specific proteins.

Statistical Analyses

Categorical data are presented as absolute numbers with percentages. The normal distribution of each continuous variable was tested using the Kolmogorov–Smirnov test. Quantitative variables are expressed as mean values with SDs, or as medians with interquartile ranges (for data without Gaussian distributions). The paired t test or Mann–Whitney U test were used to assess differences in quantitative variables between the pregnancy groups (preeclamptic versus normotensive pregnancies) for independent samples. Categorical variables were analyzed using the chi-squared test. Correlations among the various parameters in the studied population were analyzed using Pearson correlation or the Spearman correlation coefficient according to data distribution. Graphics for paired data were created using an interactive graph tool.40 Statistical analysis was performed using SPSS (SPSS for Windows, version 21.0; IBM SPSS, Chicago, IL). P<0.05 was considered to be statistically significant.

Disclosures

V.D.G. is the inventor of technology used in this research; the technology has been licensed to a commercial entity and V.D.G. and Mayo Clinic have contractual rights to receive royalties from the licensing of this technology. S.R.H. holds patents for the diagnosis and treatment of preeclampsia and is one of the founders of the companies Preelumina Diagnostics AB and A1M Pharma AB. All other authors report no conflict of interest.

Acknowledgments

This project was supported by award number P-50AG44170 (to V.D.G. and M.J.) from the National Institute on Aging; award number R01 DK47060 (to K.A.N) from the National Institute of Diabetes and Digestive and Kidney Diseases; the Building Interdisciplinary Careers in Women’s Health award number K12HD065987 (to T.L.W.) from the Office of Women’s Health Research; and a generous gift from Mrs. Cynthia L. Rosenbloom and Mr. David S. Rosenbloom. The project was further supported by the Swedish Research Council and The Wallenberg and Maggie Stephens Foundations.

Footnotes

References

  1. ↵
    1. American College of Obstetricians and Gynecologists
    2. Task Force on Hypertension in Pregnancy
    : Hypertension in pregnancy. Report of the American college of obstetricians and gynecologists’ task force on hypertension in pregnancy. Obstet Gynecol 122: 1122–1131, 2013pmid:24150027
    OpenUrlCrossRefPubMed
  2. ↵
    1. Roberts JM
    : Endothelial dysfunction in preeclampsia. Semin Reprod Endocrinol 16: 5–15, 1998pmid:9654603
    OpenUrlCrossRefPubMed
  3. ↵
    1. Stillman IE,
    2. Karumanchi SA
    : The glomerular injury of preeclampsia. J Am Soc Nephrol 18: 2281–2284, 2007pmid:17634433
    OpenUrlAbstract/FREE Full Text
  4. ↵
    1. Craici IM,
    2. Wagner SJ,
    3. Weissgerber TL,
    4. Grande JP,
    5. Garovic VD
    : Advances in the pathophysiology of pre-eclampsia and related podocyte injury. Kidney Int 86: 275–285, 2014pmid:24573315
    OpenUrlCrossRefPubMed
  5. ↵
    1. Mundel P,
    2. Shankland SJ
    : Podocyte biology and response to injury. J Am Soc Nephrol 13: 3005–3015, 2002pmid:12444221
    OpenUrlFREE Full Text
  6. ↵
    1. Garovic VD,
    2. Wagner SJ,
    3. Turner ST,
    4. Rosenthal DW,
    5. Watson WJ,
    6. Brost BC,
    7. Rose CH,
    8. Gavrilova L,
    9. Craigo P,
    10. Bailey KR,
    11. Achenbach J,
    12. Schiffer M,
    13. Grande JP
    : Urinary podocyte excretion as a marker for preeclampsia. Am J Obstet Gynecol 196: 320.e1–320.e7, 2007pmid:17403404
    OpenUrlCrossRefPubMed
  7. ↵
    1. Zhao S,
    2. Gu Y,
    3. Coates G,
    4. Groome LJ,
    5. Saleem MA,
    6. Mathieson PW,
    7. Wang Y
    : Altered nephrin and podoplanin distribution is associated with disturbed polarity protein PARD-3 and PARD-6 expressions in podocytes from preeclampsia. Reprod Sci 18: 772–780, 2011pmid:21422051
    OpenUrlCrossRefPubMed
  8. ↵
    1. Wang Y,
    2. Zhao S,
    3. Loyd S,
    4. Groome LJ
    : Increased urinary excretion of nephrin, podocalyxin, and βig-h3 in women with preeclampsia. Am J Physiol Renal Physiol 302: F1084–F1089, 2012pmid:22301621
    OpenUrlCrossRefPubMed
  9. ↵
    1. Garovic VD,
    2. Wagner SJ,
    3. Petrovic LM,
    4. Gray CE,
    5. Hall P,
    6. Sugimoto H,
    7. Kalluri R,
    8. Grande JP
    : Glomerular expression of nephrin and synaptopodin, but not podocin, is decreased in kidney sections from women with preeclampsia. Nephrol Dial Transplant 22: 1136–1143, 2007pmid:17255128
    OpenUrlCrossRefPubMed
  10. ↵
    1. Zhao S,
    2. Gu X,
    3. Groome LJ,
    4. Wang Y
    : Decreased nephrin and GLEPP-1, but increased VEGF, Flt-1, and nitrotyrosine, expressions in kidney tissue sections from women with preeclampsia. Reprod Sci 16: 970–979, 2009pmid:19528353
    OpenUrlCrossRefPubMed
  11. ↵
    1. Son GH,
    2. Kwon JY,
    3. Lee S,
    4. Park J,
    5. Kim YJ,
    6. Yun B,
    7. Park JH
    : Comparison of serum and urinary nephrin levels between normal pregnancies and severe preeclampsia. Eur J Obstet Gynecol Reprod Biol 166: 139–144, 2013pmid:23116596
    OpenUrlCrossRefPubMed
  12. ↵
    1. Ranghino A,
    2. Dimuccio V,
    3. Papadimitriou E,
    4. Bussolati B
    : Extracellular vesicles in the urine: Markers and mediators of tissue damage and regeneration. Clin Kidney J 8: 23–30, 2015pmid:25713706
    OpenUrlCrossRefPubMed
  13. ↵
    1. Jayachandran M,
    2. Miller VM,
    3. Heit JA,
    4. Owen WG
    : Methodology for isolation, identification and characterization of microvesicles in peripheral blood. J Immunol Methods 375: 207–214, 2012pmid:22075275
    OpenUrlCrossRefPubMed
  14. ↵
    1. Jayachandran M,
    2. Lugo G,
    3. Heiling H,
    4. Miller VM,
    5. Rule AD,
    6. Lieske JC
    : Extracellular vesicles in urine of women with but not without kidney stones manifest patterns similar to men: A case control study. Biol Sex Differ 6: 2, 2015pmid:25729563
    OpenUrlCrossRefPubMed
  15. ↵
    1. Anderson UD,
    2. Olsson MG,
    3. Rutardottir S,
    4. Centlow M,
    5. Kristensen KH,
    6. Isberg PE,
    7. Thilaganathan B,
    8. Akerström B,
    9. Hansson SR
    : Fetal hemoglobin and alpha1-microglobulin as first- and early second-trimester predictive biomarkers for preeclampsia. Am J Obstet Gynecol 204: 520.e1–520.e5, 2011pmid:21439542
    OpenUrlCrossRefPubMed
  16. ↵
    1. Olsson MG,
    2. Centlow M,
    3. Rutardóttir S,
    4. Stenfors I,
    5. Larsson J,
    6. Hosseini-Maaf B,
    7. Olsson ML,
    8. Hansson SR,
    9. Akerström B
    : Increased levels of cell-free hemoglobin, oxidation markers, and the antioxidative heme scavenger α(1)-microglobulin in preeclampsia. Free Radic Biol Med 48: 284–291, 2010pmid:19879940
    OpenUrlCrossRefPubMed
  17. ↵
    1. Wester-Rosenlöf L,
    2. Casslén V,
    3. Axelsson J,
    4. Edström-Hägerwall A,
    5. Gram M,
    6. Holmqvist M,
    7. Johansson ME,
    8. Larsson I,
    9. Ley D,
    10. Marsal K,
    11. Mörgelin M,
    12. Rippe B,
    13. Rutardottir S,
    14. Shohani B,
    15. Akerström B,
    16. Hansson SR
    : A1M/α1-microglobulin protects from heme-induced placental and renal damage in a pregnant sheep model of preeclampsia. PLoS One 9: e86353, 2014pmid:24489717
    OpenUrlCrossRefPubMed
  18. ↵
    1. Nääv Å,
    2. Erlandsson L,
    3. Axelsson J,
    4. Larsson I,
    5. Johansson M,
    6. Wester-Rosenlöf L,
    7. Mörgelin M,
    8. Casslén V,
    9. Gram M,
    10. Åkerström B,
    11. Hansson SR
    : A1M ameliorates preeclampsia-like symptoms in placenta and kidney induced by cell-free fetal hemoglobin in rabbit. PLoS One 10: e0125499, 2015pmid:25955715
    OpenUrlCrossRefPubMed
  19. ↵
    1. Martin AC,
    2. Brown MA
    : Could uric acid have a pathogenic role in pre-eclampsia? Nat Rev Nephrol 6: 744–748, 2010pmid:20956991
    OpenUrlCrossRefPubMed
  20. ↵
    1. Gram M,
    2. Anderson UD,
    3. Johansson ME,
    4. Edström-Hägerwall A,
    5. Larsson I,
    6. Jälmby M,
    7. Hansson SR,
    8. Åkerström B
    : The human endogenous protection system against cell-free hemoglobin and heme is overwhelmed in preeclampsia and provides potential biomarkers and clinical indicators. PLoS One 10: e0138111, 2015pmid:26368565
    OpenUrlCrossRefPubMed
  21. ↵
    1. Salih M,
    2. Zietse R,
    3. Hoorn EJ
    : Urinary extracellular vesicles and the kidney: Biomarkers and beyond. Am J Physiol Renal Physiol 306: F1251–F1259, 2014pmid:24694589
    OpenUrlCrossRefPubMed
  22. ↵
    1. Gould SJ,
    2. Raposo G
    : As we wait: Coping with an imperfect nomenclature for extracellular vesicles. J Extracell Vesicles 2013. Available at http://www.tandfonline.com/doi/full/10.3402/jev.v2i0.20389. Accessed June 7, 2017pmid:24009890
    OpenUrlPubMed
  23. ↵
    1. Witwer KW,
    2. Buzás EI,
    3. Bemis LT,
    4. Bora A,
    5. Lässer C,
    6. Lötvall J,
    7. Nolte-’t Hoen EN,
    8. Piper MG,
    9. Sivaraman S,
    10. Skog J,
    11. Théry C,
    12. Wauben MH,
    13. Hochberg F
    : Standardization of sample collection, isolation and analysis methods in extracellular vesicle research. J Extracell Vesicles. Available at http://www.tandfonline.com/doi/full/10.3402/jev.v2i0.20360. Accessed June 7, 2017pmid:24009894
    OpenUrlCrossRefPubMed
  24. ↵
    1. Lötvall J,
    2. Hill AF,
    3. Hochberg F,
    4. Buzás EI,
    5. Di Vizio D,
    6. Gardiner C,
    7. Gho YS,
    8. Kurochkin IV,
    9. Mathivanan S,
    10. Quesenberry P,
    11. Sahoo S,
    12. Tahara H,
    13. Wauben MH,
    14. Witwer KW,
    15. Théry C
    : Minimal experimental requirements for definition of extracellular vesicles and their functions: A position statement from the International Society for Extracellular Vesicles. J Extracell Vesicles 3: 26913, 2014pmid:25536934
    OpenUrlCrossRefPubMed
  25. ↵
    1. Dear JW,
    2. Street JM,
    3. Bailey MA
    : Urinary exosomes: A reservoir for biomarker discovery and potential mediators of intrarenal signalling. Proteomics 13: 1572–1580, 2013pmid:23129434
    OpenUrlCrossRefPubMed
  26. ↵
    1. Fukuda A,
    2. Wickman LT,
    3. Venkatareddy MP,
    4. Wang SQ,
    5. Chowdhury MA,
    6. Wiggins JE,
    7. Shedden KA,
    8. Wiggins RC
    : Urine podocin: Nephrin mRNA ratio (PNR) as a podocyte stress biomarker. Nephrol Dial Transplant 27: 4079–4087, 2012pmid:22863839
    OpenUrlCrossRefPubMed
  27. ↵
    1. Zhai T,
    2. Furuta I,
    3. Akaishi R,
    4. Ishikawa S,
    5. Morikawa M,
    6. Yamada T,
    7. Koyama T,
    8. Minakami H
    : Alteration of podocyte phenotype in the urine of women with preeclampsia. Sci Rep 6: 24258, 2016pmid:27052160
    OpenUrlCrossRefPubMed
  28. ↵
    1. Zhou H,
    2. Cheruvanky A,
    3. Hu X,
    4. Matsumoto T,
    5. Hiramatsu N,
    6. Cho ME,
    7. Berger A,
    8. Leelahavanichkul A,
    9. Doi K,
    10. Chawla LS,
    11. Illei GG,
    12. Kopp JB,
    13. Balow JE,
    14. Austin HA 3rd,
    15. Yuen PS,
    16. Star RA
    : Urinary exosomal transcription factors, a new class of biomarkers for renal disease. Kidney Int 74: 613–621, 2008pmid:18509321
    OpenUrlCrossRefPubMed
  29. ↵
    1. Kalani A,
    2. Mohan A,
    3. Godbole MM,
    4. Bhatia E,
    5. Gupta A,
    6. Sharma RK,
    7. Tiwari S
    : Wilm’s tumor-1 protein levels in urinary exosomes from diabetic patients with or without proteinuria. PLoS One 8: e60177, 2013pmid:23544132
    OpenUrlCrossRefPubMed
  30. ↵
    1. Sverrisson K,
    2. Axelsson J,
    3. Rippe A,
    4. Gram M,
    5. Åkerström B,
    6. Hansson SR,
    7. Rippe B
    : Extracellular fetal hemoglobin induces increases in glomerular permeability: Inhibition with α1-microglobulin and tempol. Am J Physiol Renal Physiol 306: F442–F448, 2014pmid:24338823
    OpenUrlCrossRefPubMed
  31. ↵
    1. Tracz MJ,
    2. Alam J,
    3. Nath KA
    : Physiology and pathophysiology of heme: Implications for kidney disease. J Am Soc Nephrol 18: 414–420, 2007pmid:17229906
    OpenUrlAbstract/FREE Full Text
  32. ↵
    1. Zenclussen ML,
    2. Casalis PA,
    3. El-Mousleh T,
    4. Rebelo S,
    5. Langwisch S,
    6. Linzke N,
    7. Volk HD,
    8. Fest S,
    9. Soares MP,
    10. Zenclussen AC
    : Haem oxygenase-1 dictates intrauterine fetal survival in mice via carbon monoxide. J Pathol 225: 293–304, 2011pmid:21744344
    OpenUrlCrossRefPubMed
  33. ↵
    1. Tencer J,
    2. Thysell H,
    3. Andersson K,
    4. Grubb A
    : Stability of albumin, protein HC, immunoglobulin G, kappa- and lambda-chain immunoreactivity, orosomucoid and alpha 1-antitrypsin in urine stored at various conditions. Scand J Clin Lab Invest 54: 199–206, 1994pmid:7518610
    OpenUrlPubMed
  34. ↵
    1. Tencer J,
    2. Thysell H,
    3. Andersson K,
    4. Grubb A
    : Long-term stability of albumin, protein HC, immunoglobulin G, κ- and λ-chain-immunoreactivity, orosomucoid and α 1-antitrypsin in urine stored at -20°C. Scand J Urol Nephrol 31: 67–71, 1997pmid:9060087
    OpenUrlPubMed
  35. ↵
    1. Turco AE,
    2. Lam W,
    3. Rule AD,
    4. Denic A,
    5. Lieske JC,
    6. Miller VM,
    7. Larson JJ,
    8. Kremers WK,
    9. Jayachandran M
    : Specific renal parenchymal-derived urinary extracellular vesicles identify age-associated structural changes in living donor kidneys. J Extracell Vesicles 5: 29642, 2016pmid:26837814
    OpenUrlPubMed
  36. ↵
    1. Donoviel DB,
    2. Freed DD,
    3. Vogel H,
    4. Potter DG,
    5. Hawkins E,
    6. Barrish JP,
    7. Mathur BN,
    8. Turner CA,
    9. Geske R,
    10. Montgomery CA,
    11. Starbuck M,
    12. Brandt M,
    13. Gupta A,
    14. Ramirez-Solis R,
    15. Zambrowicz BP,
    16. Powell DR
    : Proteinuria and perinatal lethality in mice lacking NEPH1, a novel protein with homology to NEPHRIN. Mol Cell Biol 21: 4829–4836, 2001pmid:11416156
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Boute N,
    2. Gribouval O,
    3. Roselli S,
    4. Benessy F,
    5. Lee H,
    6. Fuchshuber A,
    7. Dahan K,
    8. Gubler MC,
    9. Niaudet P,
    10. Antignac C
    : NPHS2, encoding the glomerular protein podocin, is mutated in autosomal recessive steroid-resistant nephrotic syndrome. Nat Genet 24: 349–354, 2000pmid:10742096
    OpenUrlCrossRefPubMed
  38. ↵
    1. Fan Q,
    2. Xing Y,
    3. Ding J,
    4. Guan N,
    5. Zhang J
    : The relationship among nephrin, podocin, CD2AP, and alpha-actinin might not be a true ‘interaction’ in podocyte. Kidney Int 69: 1207–1215, 2006pmid:16501493
    OpenUrlCrossRefPubMed
  39. ↵
    1. Jayachandran M,
    2. Litwiller RD,
    3. Owen WG,
    4. Heit JA,
    5. Behrenbeck T,
    6. Mulvagh SL,
    7. Araoz PA,
    8. Budoff MJ,
    9. Harman SM,
    10. Miller VM
    : Characterization of blood borne microparticles as markers of premature coronary calcification in newly menopausal women. Am J Physiol Heart Circ Physiol 295: H931–H938, 2008pmid:18621859
    OpenUrlCrossRefPubMed
  40. ↵
    1. Weissgerber TL,
    2. Garovic VD,
    3. Savic M,
    4. Winham SJ,
    5. Milic NM
    : From static to interactive: Transforming data visualization to improve transparency. PLoS Biol 14: e1002484, 2016pmid:27332507
    OpenUrlPubMed
View Abstract
Back to top

In this issue

View Selected Citations (0)
Print
Download PDF
Email Article
Citation Tools
Request Permissions
Urinary Extracellular Vesicles of Podocyte Origin and Renal Injury in Preeclampsia
Sarwat I. Gilani, Ulrik Dolberg Anderson, Muthuvel Jayachandran, Tracey L. Weissgerber, Ladan Zand, Wendy M. White, Natasa Milic, Maria Lourdes Gonzalez Suarez, Rangit Reddy Vallapureddy, Åsa Nääv, Lena Erlandsson, John C. Lieske, Joseph P. Grande, Karl A. Nath, Stefan R. Hansson, Vesna D. Garovic
JASN Nov 2017, 28 (11) 3363-3372; DOI: 10.1681/ASN.2016111202
del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Jump to section

Keywords