| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Nephrology Division, Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin 53226
1To whom requests for reprints should be addressed at Medical College of Wisconsin, Room M4040, MEB/CVC/Nephrology, 8701 Watertown Plank Road, Milwaukee, WI 53226. E-mail: msharma{at}mcw.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Key Words: recurrent FSGS • FSGS factor • FSPF • proteinuria • glomerular albumin permeability • plasmapheresis • protein tyrosine phosphorylation • FAK • paxillin
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A circulating substance is believed to cause FSGS in some patients. Rationale for this interpretation include the following observations. First, FSGS recurs after renal transplantation in 30%–40% of patients in whom FSGS is the primary renal disease (21–23). Recurrence of proteinuria can be observed almost immediately after transplantation or may be delayed for days or weeks. Recurrent disease often results in the loss of function of the allograft (21). Second, plasmapheresis during early stages of the disease or immediately after transplantation may reduce proteinuria and stabilize renal function (24–28). Immunoadsorption of plasma proteins has also been found to be effective in some patients (29, 30). Third, injection of plasma or fractions of plasma from patients with recurrent FSGS may cause proteinuria or albuminuria in experimental animals (31). Finally, FSGS serum, plasma, or discarded plasmapheresis fluid (PPF) can increase glomerular albumin permeability (Palb) in an in vitro functional assay (32), and infusion of patient material results in proteinuria and albuminuria (33, 34). We have demonstrated that FS permeability activity, defined by an increase in Palb caused by the focal segmental glomerulosclerosis permeability factor (FSPF), can be neutralized by normal plasma (35). Certain plasma proteins, such as apolipoproteins, can block or neutralize its activity (36). These results support the hypothesis that circulating FSPF(s) may be responsible for glomerular damage and for initiation of glomerular events that contribute to the development of proteinuria in FSGS.
We have reported a protocol of sequential precipitation of plasma of patients with recurrent FSGS. This protocol results in a 100-fold enrichment of FS permeability activity. Material derived as 70% supernatant increases Palb of isolated glomeruli and carries all the activity of the starting material (33). We have also reported that intravenous injection of the 70% supernatant causes proteinuria and albuminuria in rats and that Palb of glomeruli isolated from these proteinuric rats exhibit increased Palb without further experimental treatment (34). Neither the amount of 70% supernatant that must be injected to increase Palb nor the amount required to cause proteinuria has been defined. The cellular mechanisms that are responsible for increased Palb are also unknown. The physicochemical characteristics of FSPF have not been defined, although understanding these characteristics may be essential to the development of therapeutic strategies in FSGS. The current studies were designed to address these aspects of the chemistry and biological activity of FSPF.
In the present report, we describe certain characteristics of the active component as evidenced by binding to chromatographic materials. We further characterize the in vivo effect of FSPF on Palb, the time course of the in vitro effects, changes in tyrosine phosphorylation of glomerular cellular proteins, and effects of blocking these changes in tyrosine phosphorylation. This information will be useful in understanding the mechanisms that lead to proteinuria in FSGS and will permit development of methodology for purification and characterization of FSPF and for its specific removal from affected patients.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Source of FSGS and Normal Plasma.
Serum or plasma specimens were obtained from patients in whom the diagnosis of FSGS in native kidneys or in transplanted kidneys had been made. Details of the criteria used for the diagnosis of FSGS have been described earlier (24, 32, 33). In the studies described here, serum samples and plasmapheresis fluid (PPF), obtained during the course of therapeutic plasmapheresis from 12 patients, were tested. The decision to perform plasmapheresis was made by the primary physician in each case. The 12 patients included 10 adults and 2 children. Of these, 6 were male and 6 female. A number of these patients were represented in prior reports (33, 34), while others had not been studied in the past. Their demographic characteristics were comparable to those of patients previously reported (33). Each was on immunosuppressive medications as part of post-transplant therapy. Detailed information regarding their renal function and medications at the time of plasmapheresis was not available to us. Additional PPF was obtained from a male patient who experienced a recurrence of FSGS and allograft loss in 1985 and who has since been treated with thrice-weekly hemodialysis. This patient’s serum, plasma, and plasmapheresis fluid have consistently shown high levels of FS activity in vitro. This patient was not nephrotic and had normal serum proteins; he was not on any immunosuppressive medications. Plasmapheresis was performed using citrate anticoagulation, and plasma removed was replaced with 0.9% sodium chloride and 5% human albumin solution. Discarded PPF was shipped to our laboratory on dry ice and was stored at -20°C.
Normal serum from healthy volunteers and discarded normal plasma from the Blood Center of Southeastern Wisconsin (Milwaukee, WI) were obtained. Each sample was stored at -20°C until use. Plasma from 35 normal individuals was pooled and designated normal pooled plasma (NPP). NPP was used to provide control material for activity studies and for protein purification.
Enrichment of FSPF from PPF and Identical Fractions from NPP.
As described previously, an enriched preparation of FSPF (FSGS 70% supernatant) was obtained by removing inactive plasma components from the PPF (33). Briefly, cryoprecipitate was removed by centrifugation. Lipid-rich lipoproteins and chylomicrons were precipitated by dextran sulfate and calcium chloride and removed by centrifugation. From the post-lipoprotein supernatant, immunoglobulins and albumin-rich fractions were salted out at 50% and 70% ammonium sulfate saturation, respectively. Precipitates were not tested for activity. The supernatant at 70% ammonium sulfate saturation contained FS activity. FS activity was not present in other fractions (33). NPP was treated identically to obtain the supernatant at 70% ammonium sulfate. The same amount of total protein from FSGS or NPP 70% supernatant was used in studies of Palb and in injection studies.
Experimental Animals and Preparation of Glomeruli.
Animals were cared for according to NIH guidelines. All protocols were reviewed and approved by the Institutional Committee for care and use of animals at the Medical College of Wisconsin (MCW). Rats were maintained at the Animal Resource Center with free access to rat chow and drinking water.
For studies of Palb, glomeruli were isolated from kidneys of normal male Sprague-Dawley rats (175–200 g). Rats were anesthetized with Metofane vapor (methoxyflurane, Mallinckrodt Veterinary Inc., Mundelein, IL), kidneys removed, and rats euthanized by ex-sanguination. Renal medulla was removed, and fine pieces of the cortical slices were passed through stainless steel sieves of decreasing pore sizes. Glomeruli were collected from atop the 200-mesh sieve. Details of the techniques and conditions used have been described (24, 32–35).
Collection of Urine and Determination of Protein and Creatinine.
Pretreatment urine was collected from rats in metabolic cages. For intravenous injection of FSPF, rats (250–275 g) were anesthetized with intraperitoneal Brevital (methohexital sodium) at 50 mg/kg (Eli Lilly and Co., Indianapolis, IN). After injection of FSPF, rats were allowed to recover and placed in metabolic cages for urine collection (24 hrs).
Every fraction of plasma to be tested for FS activity was dialyzed in TBS (10 mM Tris-HCl containing 1% sodium chloride, pH 7.8). Protein concentration in urine or chromatography fractions was determined by Bradford’s method using Coomassie Blue Dye Reagent (Bio-Rad) and spectrophotometric measurement of absorbance at 595 nm (37). Urinary creatinine was determined using a reagent kit from Sigma Chemical.
Detection of FS Activity in FSPF by In Vitro Functional Assay.
We have defined FS activity as the ability of serum, plasma, or plasma fractions to increase albumin permeability (Palb) after in vitro incubations (33). Fractions obtained during purification were tested for their capacity to increase glomerular Palb using methods established in our laboratory. Details of the rationale and methodology for using isolated glomeruli and measurement of the reflection coefficient (
alb) and Palb and its usefulness in the purification of FSPF have been described previously (33, 34, 38). In short, glomeruli from superficial cortex were isolated at room temperature by sieving in isolation/ incubation medium (pH 7.4) containing 4% BSA as an oncotic agent. Extensively dialyzed and diluted normal and FSGS plasma fractions (20 µl aliquot) in TBS were incubated with glomeruli in a final volume of 1 ml at 37°C for 2–10 min. Incubated glomeruli were observed using videomicroscopy before and 1 min after exposure to a transcapillary oncotic gradient. The oncotic gradient resulted in influx of fluid and expansion of glomerular capillaries. The relative volume increase (
V) of each glomerulus was used to calculate albumin convectional permeability (Palb). Palb is a dimensionless variable with a minimum of zero in normal glomeruli and a maximum of 1.0 after injury to the permeability barrier.
For studies of proteinuria in rats after injection of FSPF, we have reported numerical values. For studies of binding and recovery of FSPF from various chromatography media, we have defined Palb 
0.5 as positive. Lower Palb activity in patient sera may be present, but we have selected 0.5 as a cutoff because it exceeds the average value derived using sera of normal people or patients without FSGS by 2 standard deviations and permits us to exclude nonspecific activity with assurance.
Incubation of glomeruli with medium containing pooled normal serum (1:50 dilution) served as a negative control, while incubation with known active FSGS plasma (1:50 dilution) served as a positive control (33).
Effect of Lyophilization and Reconstitution of Sera on FS Activity.
Aliquots (50 µl) of serum from each of the 12 patients with recurrent FSGS were frozen at -20°C overnight. These samples then were further cooled on dry ice for 3 hrs and lyophilized. The dry residue was left at room temperature (about 21°C) for 3 days and reconstituted to the original volume with water. Then aliquots (20 µl) of fresh or reconstituted serum were used to determine Palb in a final volume of 1 ml (1:50 dilution). FS activity of lyophilized specimens was compared to activity of serum that had not been lyophilized (20 µl in 1 ml, 1:50 dilution).
Removal of FSPF Using Chromatography Media.
The FSGS 70% supernatant was used to test the binding of FSPF with each chromatography material packed in a gravity-flow column. Established principles and techniques of chromatography were followed for each analysis. We tested Cibacron blue F3GA gel (39, 40), con-A (41), heparin-agarose (42), and silica- and agarose-bound protein A (43) for affinity binding; and Mono-S (44), DEAE-Sephacel (45), and Mono-Q (46) for ionic exchange binding. Table 1
provides an outline of the chromatography materials, equilibration and sample loading buffers, and elution conditions employed for each chromatographic procedure. The FSGS 70% supernatant (10 mg total protein) was dialyzed in the equilibration buffer for each chromatography technique. Each chromatography material was equilibrated in the buffer listed under ‘‘Equilibration and Loading Conditions’’ (Table 1). The FSGS 70% supernatant was chromatographed on each column, and the unbound fractions were collected. Each column was washed three times with equilibration buffer to remove the unbound fraction. Bound proteins were eluted using buffers described under ‘‘Elution Conditions’’ (Table 1). All fractions were dialyzed against TBS and concentrated to 1 ml, and 20-µl aliquots were used for Palb assay. Binding of FSPF with a particular chromatography material was further confirmed using 70% supernatant preparations from several FSGS patients.
|
Time Course of Increase in Palb In Vitro.
To determine the earliest time when increased Palb could be measured, glomeruli were isolated in 5% BSA medium as described previously. An aliquot of glomerular suspension was added to the observation chamber, and glomeruli were allowed to settle and adhere to the polylysine-coated glass surface. Adherent glomeruli were observed after the addition of medium containing FSPF (20 µg protein/ml) for 2–10 mins. Hypo-oncotic medium was then added and Palb measured using changes in glomerular diameter that occurred within 1 min (38). In a limited number of studies, isolated glomeruli were allowed to equilibrate with isoncotic high-molecular-weight dextran and were observed after the addition of FSPF. In these studies, fluid was lost from glomerular capillaries, and glomerular diameter diminished when Palb increased. The finding that isoncotic dextran caused capillary collapse provided definitive evidence that the capillary reflection coefficient for albumin decreased and that an increase in Palb rather than any other change in glomerular characteristics was responsible for the experimental responses we observed (38).
In Vitro Change in Glomerular Protein Tyrosine Phosphorylation.
The rapid increase in Palb is consistent with a signaling pathway such as protein phosphorylation. Since tyrosine phosphorylation of cytoskeleton-associated proteins focal adhesion kinase (FAK) and paxillin alters adhesion of cells to substrate, we postulated that the change in Palb might be associated with changes in tyrosine phosphorylation of these proteins.
Glomeruli were incubated with FSPF for 3 mins in RPMI-1640, followed by homogenization in a lysis buffer (150 mM sodium chloride, 1.5 mM magnesium chloride, 1 mM EGTA, 10% glycerol, 1% Triton X-100, 1 µg/ml aprotinin, 1 µg/ml leupeptin, 1 µg/ml pepstatin, 1 mM PMSF, and 200 µM sodium orthovanadate in 50 mM HEPES, pH 7.4). Protein was determined by modified Lowry’s method using a Bio-Rad reagent kit (47).
Equal amounts of protein from all samples were loaded (1:1) in a sample buffer (4% SDS, 20% glycerol, 0.005% bromophenol blue, 5% mercaptoethanol, 125 mM Tris-HCl, pH 6.8) and resolved by SDS-PAGE on a Tris-glycine (1-mm–thick minigel) at 4°C. Proteins were electrotransferred to nitrocellulose membrane (30 V, overnight at 4°C) using a transfer buffer (0.3% Tris base, 1.44% glycine, 20% methanol). Blots were incubated with 5% nonfat milk in TTBS (10 mM Tris base, 100 mM sodium chloride, 0.05% Tween-20, pH 9.7) and washed with TTBS as needed. Tyrosine phosphorylation of glomerular proteins was determined by Western blotting using a mouse antiphosphotyrosine monoclonal antibody (1:2000 in 5% milk) as the primary antibody (1 hr, room temperature, 3 x 10-min washes with TTBS) and an HRP-conjugated rabbit anti-mouse IgG (1:50,000 in 5% fat-free milk powder) as the secondary antibody (1 hr, room temperature, 3 x 10-min washes with TTBS). Images of the HRP-conjugated protein-antibody complexes were obtained by incubation with ECL reagent (Amersham-Pharmacia) followed by exposure to photography film.
Glomerular lysate was subjected to immunoprecipitation, SDS-PAGE, and immunoblotting to analyze the effect of FSPF on tyrosine phosphorylation of glomerular FAK and paxillin. Glomerular lysate (1 mg protein) was mixed with anti-FAK (10 µg/mg protein) or antipaxillin (10 µg/mg protein) antibody and incubated overnight at 4°C on a low-speed gyratory shaker. The immune complex was captured mixing with 100 µl of protein G agarose bead slurry for 2 hrs at 4°C. The beads were centrifuged (14,000 g, 1 min), washed with ice-cold lysis buffer (3x), mixed with the sample buffer (50 µl), and heated at 95°C for 5 mins, and the immunocomplex in the supernatant was used for SDS-PAGE (4%–20% gradient for FAK and a 12% gel for paxillin). Transfer of proteins from the gel to a nitrocellulose membrane, blocking, and washes were carried out as outlined previously. Mouse anti-antiphosphotyrosine antibody and HRP-conjugated rabbit anti-mouse IgG were used as the primary and secondary antibodies, respectively, to carry out Western blotting as described previously. Chemiluminescence reaction and detection of immune complexes were carried out following the manufacturer’s instructions (Amersham-Pharmacia). Immunoblotting of actin in the glomerular lysate was carried out using mouse anti-actin antibody as the primary and goat anti-mouse IgM (Calbiochem) as the secondary antibody, respectively (48).
Effect of Inhibitor of Tyrosine Phosphatase on Altered Palb.
Glomeruli were isolated in 4% BSA medium. Glomeruli were incubated under standard conditions with FSGS 70% supernatant alone or with FSGS 70% supernatant in the presence of sodium orthovanadate (200 µM), a phosphatase inhibitor. Palb was measured after 10 mins of incubation.
Statistical Analysis.
Values of Palb are expressed as mean ± SEM. N represents the number of glomeruli studied. Results were compared using ANOVA, and significance was defined as P < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
).
|
Chromatography with Affinity Media.
The FSGS 70% supernatant was passed through affinity media, including (i) Cibacron blue F3GA gel (blue gel), (ii) con A–sepharose gel (con-A), (iii) heparin-agarose, and (iv) agarose- or silica-bound protein A. Bound and unbound fractions were collected and analyzed for FS activity. Buffers used for chromatography are briefly described in Table 1
.
FSPF was retained by blue gel and the bound fraction eluted between 0.4 and 1.5 M of sodium chloride in 0.025 M sodium phosphate buffer (Fig. 2a). FSPF was not retained by con-A (Fig. 2b) and was detected in the unbound fraction. FSPF did not bind with heparin-agarose (Fig. 2c) and was largely present in the unbound fraction.
|
) and partially eluted with 0.025 M citrate (pH 2.4), whereas washing with 1 M sodium thiocyanate completely eluted FSPF. Identical results were obtained using silica-bound protein A and are not presented. Only about 4% of the total proteins in the 70% ammonium sulfate supernatant were retained by protein A column. Protein A affinity chromatography resulted in a 10,000-fold enrichment of FSPF (Table 2).
|
); the bound fraction eluted with 1 M sodium chloride in MOPS buffer (pH 7.4) did not increase Palb. FSPF did bind with DEAE-Sephacel and Mono-Q. The bound factor eluted with 0.1 M sodium chloride in 0.01 M Bis-Tris, pH 6.5, from DEAE-Sephacel (Fig. 3b) and with 0.5 M sodium chloride in 0.01 M Bis-Tris, pH 6.5, from Mono-Q column (Fig. 3c). As shown in Table 2, about 80% of the total protein in the 70% ammonium sulfate supernatant was retained by the Mono-Q column and eluted with 0.5 M sodium chloride. Thus, the use of low-pressure Mono-Q anion exchange chromatography did not increase the purity of the protein preparation that contains FSPF. However, it demonstrated an important characteristic of the active agent that interacts with glomerular membrane(s).
|
).
|
).
|
). Tyrosine phosphorylation of FAK and paxillin were decreased after 3 mins of incubation with FSPF. Tyrosine phosphorylation of additional unidentified proteins was also decreased at this time. Representative gels from six analyses are shown in Figure 7. These findings are consistent with the hypothesis that FSPF increases Palb via pathways involving activation of tyrosine phosphatase or inactivation of tyrosine kinases in glomerular cells.
|
|
. These results indicate that phosphorylation/dephosphorylation may play a role in FSPF-induced increase in glomerular permeability.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To date, we have identified FSPF as a protein associated with a low-molecular-weight fraction from plasma (33, 34) and have shown that activity in our in vitro assay is correlated with the severity of disease and risk of recurrence after transplantation (32, 49). Permeability activity is blocked by components of normal plasma (35, 36), cyclosporine A (50), indomethacin (51), or specific glycosides derived from Trypterygium wilfordii (52).
Despite this correlation and our finding that FS permeability activity is greater in patients with severe and rapidly progressive or recurrent disease, we have found instances in which proteinuria was unaffected by removal of activity with plasmapheresis (53). We interpret this persistence of proteinuria as evidence of causes of proteinuria that are independent of activity; these may include glomerulosclerosis with fixed anatomical changes in the glomerular capillaries. In another independent series of patients with FSGS and steroid-resistant proteinuria, full or partial remission occurred during treatment with cyclosporine A despite continuing FS permeability activity (54). We interpret this as support of our prior observations that cyclosporine A has a direct protective effect of the capillary permeability barrier. Our work regarding the FSGS factor has been recently summarized (55).
Others have employed methods derived from our initial protocols for measuring Palb (29, 36, 56–58) and have confirmed the association between Palb and recurrence following transplantation in children (56) and the presence of blocking substances in normal plasma (36) and have described several components of fractions comparable to the FSGS 70% supernatant employed in the current studies (57–60). Additional reports regarding permeability activity in sera from patients with genetic abnormalities in podocin (61) and the presence of a blocking activity in the urine of patients with idiopathic FSGS are of interest (62). However, it is not currently possible to determine whether the substance(s) responsible for these actions correspond to those that we are studying. Recently, increased interest in this area and the complexity of the problem emphasize the need to study each variant of FSGS individually and to the common characteristics among them.
Previously, we demonstrated that serum/plasma or partially purified FSPF in 70% ammonium sulfate supernatant from 15 individuals increased Palb and was susceptible to heat and proteolytic enzyme (33). In the present studies, we observed that lyophilization followed by storage at room temperature and reconstitution 3 days later did not alter the glomerular effect of FSGS serum (Fig. 1). Palb of glomeruli treated with serum before or after lyophilization was not significantly different (P > 0.3, n = 12). Lyophilization may alter secondary and tertiary structures of many macromolecules and result in altered biological properties (63–67). Preservation of biological activity and antigenicity of many macromolecules following dehydration may require addition of stabilizers (68, 69). Our experiment using sera from 12 individuals indicates that FSPF withstands lyophilization and storage up to 3 days without significant change in biological activity in terms of its effect on Palb. This observation also suggests that FSPF may not have a complex secondary structure and that FSPF may be a low-molecular-weight molecule(s) (33).
Further information on the physicochemical nature of FSPF was obtained using affinity binding materials as well as anion and cation exchangers. Plasma from each patient was treated separately to avoid confusion based on potential differences among patients. We have followed the active factor by testing Palb under standard conditions and have used the criterion of Palb 0.5 to determine the presence of active material, as in prior studies.
The binding and elution pattern of FSPF to Cibacron blue F3GA (blue gel) is similar to that for albumin, which also elutes over a wide range of salt concentration. The blue gel is used for affinity-based removal of albumin (39, 40). These observations are consistent with potential association between FSPF and albumin as a carrier molecule. These data demonstrate that we are not able to discriminate between albumin and FSPF and suggest the possibility that FSPF is a small molecule attached to albumin or other larger molecule(s) through hydrophobic interactions. Albumin is known for its role as a carrier molecule (70–72). Although the blue gel may not be useful in purification of the FSGS factor, it has provided helpful clues to the nature of this molecule.
The absence of binding to con A (Fig. 2b) suggests that the factor lacks mannose moieties or that those present are not available to bind with con-A. Results of studies with heparin-agarose matrix (Fig. 2c), cation exchanger Mono-S (Fig. 3a) and anion exchangers, DEAE-Sephacel (Fig. 3b), and Mono-Q (Fig. 3c) are consistent with the presence of anionic rather than cationic charge. Further, anionic nature of FSPF indicates that its potential binding with glomerular cell surface does not depend on charge neutralization. However, Mono-Q appears to retain most of the proteins present in the 70% ammonium sulfate supernatant and shows poor selectivity (Table 2). Elution of bound proteins from Mono-Q matrix by a salt or pH gradient may prove useful in separating the active component from inactive proteins.
Prior studies in our laboratory and those of others indicate that FSPF binds to and can be eluted from protein A (54, 73, 74). Additional findings indicate removal of the permeability factor by immunoadsorption with anti-human immunoglobulins and with low-density lipoprotein apheresis (75, 76). Although protein A immunoadsorption may result in remission in a variety of proteinuric diseases (77), the positive response in patients with FSGS is consistent with speculations that FSPF may have structural similarities to immunoglobulin fragments (33, 78). Several proteins in the fraction retained by and eluted from protein A have been identified, but none of these appears to represent the active molecule (58–60). Our results show that protein A may serve as a useful medium to remove the majority of inactive proteins from the 70% ammonium sulfate preparation (Table 2).
Injection of plasma fractions results in consistent proteinuria in rats and the proteinuric activity is correlated with in vitro activity of the preparation used. We have reported a rapid clearance of injected FSPF from plasma (34), neutralization of FS activity by normal plasma (35), transient albuminuria, and proteinuria several hours after injection of FSPS (33, 34). These observations led us to consider the increase in glomerular permeability and the onset of proteinuria as two distinguishable events and to investigate the possibility that FSPF may alter the glomerular permeability characteristics and cause proteinuria in a dose-related manner.
The current report expands on prior studies by defining the relationship between the dose of FSPF administered and the proteinuric response after injection and by documenting the time course of the permeability response in vitro. Glomerular Palb was increased in isolated from rat kidneys 24 hrs after injection of FSGS 70% ammonium sulfate supernatant, confirming the results of our prior studies (34). The specificity of this response is shown by the absence of increased Palb after injection of an identically prepared fraction from normal pooled plasma. The Palb was increased in response to as little as 12 mg/kg body weight FSGS 70% supernatant, while a larger dose (36–48 mg/kg) was required to produce proteinuria (Fig. 4). The difference between the dose required to increase Palb and that which causes proteinuria is consistent with the fact that proteinuria represents the net effect of protein filtration and tubular reabsorption. This difference in sensitivity provides a potential explanation for the difficulty others have had in using induction of proteinuria as an assay for a putative FSPF and confirms the potential limitations of injection and measurement of proteinuria to guide purification protocols (74). We postulate that the reabsorptive capacity of the tubule must be exceeded before protein is found in the urine. This may arise from saturation of receptors or endocytotic mechanisms or from tubular injury that impairs these mechanisms. Further studies will be required to distinguish between these possibilities. In related studies, we are attempting to determine if increased Palb and proteinuria are chronologically distinct events. The idea that glomerular injury may precede proteinuria is supported by our preliminary results that demonstrate increased Palb at 3 hrs after injection of FSPF. We have previously shown that serum from FSPF-injected rats does not increase glomerular protein permeability at 3 hrs postinjection and that proteinuria is evident only at 6 hrs postinjection (34). Thus, increased Palb and proteinuria appear to be two distinct but related effects of FSPF that depend on the amount of and the interval after injection of FSPF. The association between increased glomerular permeability, transient proteinuria, and changes in glomerular structure are not defined at the present time. Since the clinical features and histological changes associated with FSGS develop over very extended periods of time, microscopic changes in the rat glomerulus may appear many weeks after repeated administration of FSPF. Therefore, we speculate that increased permeability and transient proteinuria in rats after a single dose of FSPF indicate changes in the functional states of the filtration barrier proteins. Long-term effects of single or multiple doses of FSPF on glomerular histology and the underlying molecular mechanisms have not been studied.
A very rapid permeability response during in vitro studies, with maximal increase in Palb after only 2 mins (Fig. 5), points to signal transduction in glomerular cells as the mediator of increased permeability. Signal transduction through pathways including activation of phospholipase A2, eicosanoid synthesis, phosphorylation of intracellular proteins, signaling by cyclic nucleotides, or changes in ionized calcium have each been described in glomerular epithelial cells (50, 79, 80). Each is a candidate for mediating increased permeability. In the current studies, we have shown significant changes in tyrosine phosphorylation of FAK, paxillin, and a number of other proteins after exposure to FSPF (Figs. 6–8). These changes in phosphorylation are analogous to those shown by others during increased Palb in response to nitric oxide (57). A dynamic equilibrium in the phosphorylated and dephosphorylated forms of cellular proteins appears to be critical for a large number of biological phenomena, including the control of tight junctions in epithelial cells (81–84) and focal adhesion and barrier function of endothelial cells (85–87). Cellular response to environmental changes is initiated by biochemical signals between and within cells. Cell-matrix interactions affect gene expression, ion movement, and activation of signaling pathways and are indicated by early changes in tyrosine phosphorylation of proteins. Phosphorylation of FAK, a nonreceptor kinase, represents an early event in integrin-dependent organization of adhesion structures (88). Phosphorylated FAK interacts with Src kinases that contain SH2 domains, phosphatidylinositol (PI3) kinase, phospholipase 3 (PLC
) adapter protein Grb7, FAK phosphorylates paxillin, and p130cas, which in turn recruit additional signaling molecules resulting in activation of downstream pathways (89–92).
We postulate that diminished cell-cell interactions via the slit-pore junction or impaired cell-matrix attachment are integral to increased glomerular protein permeability. Since we have previously shown that alteration of eicosanoid synthesis by cyclooxygenase inhibition prevents FSPF-induced increase in Palb (51), we propose that more than one signaling pathway may be involved in increased permeability.
In summary, the current studies provide new information concerning the nature of FSPF and potential mechanisms of its effect on glomerular permeability. FSPF appears to be bound to albumin and other plasma proteins and carries an anionic rather than a cationic charge. The resistance of FS activity to lyophilization suggests that it is a small molecule. Binding of FSPF with protein A is consistent with the postulate that it may share structural features with immunoglobulins. We have confirmed the sensitivity of the in vitro Palb assay and its utility in developing protocols for purification. Insights regarding potential cellular mechanisms of action arise from observations that the increase in Palb during in vitro studies is maximal within 2 mins, and is associated with alterations in tyrosine phosphorylation of cellular proteins, including FAK and paxillin, and is prevented by inhibition of tyrosine phosphatases. These findings strongly implicate intracellular signaling pathways as mediating the effect of FSPF, particularly pathways that involve phosphorylation cascade(s). These data and the binding with anion exchangers indicate that FSPF does not cause glomerular changes by charge neutralization. Our results point the way for future studies of the interaction between the permeability responses and cell detachment with podocyte loss, altered matrix synthesis and turnover, and apoptosis. The fact that increased Palb after systemic infusion occurs at doses lower than those required for proteinuria confirms the sensitivity of the assay for Palb and the potential advantage of its measurement to guide strategies for purification of FSPF.
Taken together, the current observations support the theses that development of chronic proteinuria is a complex process that involves the entire nephron and that changes in albumin permeability represent a glomerular event that can be observed by lower doses of FSPF compared to those required to cause proteinuria and involves changes in the cytoskeleton and cell-matrix or cell-cell interactions. The combined use of biochemical techniques for protein purification and measurement of Palb is useful in defining the characteristics and potential mechanisms of action of FSPF. The 70% ammonium sulfate fraction and its derivatives will provide material for additional proteomic analyses. Understanding the affinity properties of FSPF will provide the basis for design of interventions to remove it from the circulation of patients at risk for recurrent FSGS. Identification of cell signaling pathways that affect increased permeability opens the way for studies of mechanisms of action and for therapy to prevent or reverse proteinuria.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received for publication January 21, 2003. Accepted for publication August 7, 2003.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
