| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Regular Articles |
From the Department of Biochemistry and Molecular
Biology,*
Monash University, and the Department of
Nephrology and Monash University Department of
Medicine,
Monash Medical Centre, Clayton,
Victoria, Australia; and the Laboratory of
Immunopathology,
School of Medical Sciences,
State University of Rio de Janeiro, Rio de Janeiro, Brazil
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-39) had the same fractional
clearance as uncharged Ficoll. Both diseases exhibited similar effects
on fractional clearance measurements suggesting an underlying common
mechanism. In disease, there was good agreement between the
fractional clearance of proteins determined by radioactivity as
compared to those determined by radioimmunoassay. A small increase in
the fractional clearance for IgG was evident in disease as compared to
controls, which mirrored the change in the equivalent size
Ficoll, suggesting that the increase is because of the
development of a small proportion of large pores in the glomerular
capillary wall. There was no increase, however, in the
fractional clearance of Ficoll of equivalent size to albumin in either
disease, yet the fractional clearance of the albumin increased
by 12 to 14 times as determined by radioactivity and 4500 to 6600 times
as determined by radioimmunoassay. This study demonstrates that
glomerulonephritis is not a disease associated with changes in
glomerular permeability to albumin but is because of alterations in
albumin processing by cells distal to the glomerular basement membrane.
It is also apparent that approaches to glomerular pathology and
proteinuria as risk factors in renal disease must be
reassessed.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Recent studies have demonstrated that in addition to a fragmentation
process, an albumin retrieval mechanism operates to return 99% of
filtered albumin back to the blood supply.9,16
A possible
location for this retrieval pathway is the glomerular epithelial cell
or proximal tubular cell, a location that would allow the early
retrieval of the albumin, resulting in only a small fraction of
filtered albumin being delivered to the degradation pathway. An
inhibition of the retrieval pathway will result in an increase in the
fractional clearance (FC) of total albumin without any change to
glomerular permselectivity.
Puromycin aminonucleoside (PAN)1,3,10,17-20 and anti-GBM antibodies2,4,21,22 have both been used to induce massive glomerular injury. Pronounced morphological changes at the GCW are induced by PAN including a loss of foot process, a spreading of epithelial cell cytoplasm, a lifting of focal areas of the epithelial cells,18 the appearance of microvilli,3 and a thin, less compact appearance to the GBM. Anti-GBM glomerulonephritis (GN) also results in substantial glomerular changes including glomerular crescent formation,22 vascular sclerosis,4 and an increasing loss of the endothelial layer.2 These morphological changes have been considered the cause of the development of large pores in the GCW and hence the increased excretion of proteins.
This study set out to examine the FC of albumin, IgG, and transferrin in control rats as compared with rats with PAN or anti-GBM GN. As the albumin RIA has been shown to only detect intact albumin, radiolabeled proteins were used, in addition to the RIA, as a comparative method of determining FC as radiolabeling allows the detection of both intact and fragmented protein. The FC of Ficolls of equivalent radius to albumin, transferrin, and IgG are also measured to account for any changes in the permeability properties of the GCW.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Rabbit anti-rat transferrin, rabbit anti-rat albumin,
125I (10 mCi), and bovine IgG were obtained from
ICN Biomedicals Inc., Aurora, OH. Benzoylated dialysis tubing
(molecular weight cut-off, 2000), rat serum albumin (RSA), rat
transferrin, rabbit anti-rat IgG (whole molecule, adsorbed with human
IgG), rat IgG, sodium azide, bovine apo-transferrin, PAN, Freund’s
complete adjuvant, polydisperse Ficoll type 70, and thimerosal were
obtained from Sigma Chemicals, St. Louis, MO. Alzet osmotic pumps
(model 2001) were obtained from Alza Pharmaceuticals, Palo Alto, CA.
Nembutal (60 mg/ml) was obtained from Rhone Mirieux Australia Pty Ltd.,
Pinkenba, Queensland, Australia.
Sodium-[3H]hydride (132 mCi/mg) and tritiated
water were obtained from NEN Research Products Du Pont, Wilmington, DE.
Q-Sepharose, Sephacryl S-300, Sephadex G-100, Sephadex G-25 in PD-10
columns, and Blue dextran T2000 were purchased from Pharmacia Fine
Chemicals, Uppsala, Sweden. Carboxymethyl Ficoll 40 with 13.8%
carboxyl content (equivalent to a net negative charge of -0.34 per
sucrose unit according to the manufacturer) was supplied by TdB
Consultancy, Uppsala, Sweden (for a Ficoll of molecular weight 40,000
substituted with carboxymethyl cellulose the valence would be -39).
Anesthetic ether was purchased from Ajax Chemicals, Auburn, Australia.
Sheep anti-rabbit antibody was generously supplied by David Casley of
the Department of Medicine, Austin and Repatriation Medical Center,
Australia.
Experimental Design
Normal or diseased rats were constantly infused with radiolabeled proteins (albumin, transferrin, or IgG) or polydisperse Ficoll for a 7-day period via mini osmotic pumps. Blood and 24-hour urine collections were made on days 2, 5, and 7. Animals were killed 7 days after pump implantation and the radioactive material in blood, urine, and one kidney examined to determine the FC of individual molecules.
Experimental Animals
Eight-week-old male Sprague-Dawley rats were obtained from the Monash University Central Animal House. Throughout the experimental period they were maintained as pairs, housed in rat boxes, under a 12-hour day/night cycle with free access to standard rat chow and water. All IgG and transferrin experimental groups contained 7 rats, all albumin experimental groups contained 14 rats, and the polydisperse Ficoll 70 experimental groups contained 7 PAN rats and 6 control and anti-GBM GN rats. Permission to perform these animal experiments was given by Monash University Animal Ethics Committee.
Induction of PAN Nephritis
Induction of PAN was performed as previously described3,17,19 with the dosage of PAN according to Osicka and colleagues.17 Briefly the rats were immobilized by towel wrapping and an injection of 10 mg/100g body weight of PAN made up as a 3.5% solution in phosphate-buffered saline (PBS), was administered via the tail vein. Osmotic pumps were implanted on day 2 after PAN administration. Animals were killed 7 days later, with one kidney taken for assessment of disease pathology (see below). Age- and weight-matched controls were administered with an equivalent volume of PBS. After killing, one kidney was removed for assessment of disease pathology and the other kidney was analyzed for radioactivity. Controls were age and weight matched.
Tissues for pathological assessment were fixed in 10% neutral-buffered formalin and embedded in paraffin. Sections (2 µm) were stained with hematoxylin and eosin (H&E) or periodic acid-Schiff reagents. Immunohistochemistry staining for ED1+ macrophages and W3/13+ T cells was performed on 4-µm paraffin sections. After hydration, sections were microwaved for 10 minutes in 0.01 mol/L sodium citrate buffer, pH 6.0, to enhance antigen detection. Immunostaining used a three-layer peroxidase anti-peroxidase method with the chromogen 3,3,diaminobenzidine (Sigma Chemical Co.), as previously described.21
Induction of Anti-GBM Glomerulonephritis
Passive accelerated anti-GBM GN was induced in rats as previously described.21 Briefly, the rats were immunized subcutaneously with 5 mg of normal sheep IgG in Freund’s complete adjuvant and 7 days later were injected intravenously with 5 ml/kg of sheep anti-rat GBM serum. Osmotic pumps were implanted 3 hours before anti-GBM serum administration. Animals were killed 7 days after anti-GBM serum injection, with one kidney taken for assessment of disease pathology (as described above), and the other kidney analyzed for radioactivity. Controls were age and weight matched.
Radiolabeling of Proteins
IgG and transferrin were tritiated using the reductive methylation technique.23 The specific activities achieved were: [3H]transferrin 1.921 x 107dpm/mg, and [3H]IgG 1.635 x 107dpm/mg.
Polydisperse Ficoll 70 and carboxymethyl Ficoll 40 were tritiated according to Van Damme and colleagues.24 The specific activity of polydisperse Ficoll 70 was 2.2 x 107 dpm/mg and carboxymethyl Ficoll 40 was 5.9 x 105 dpm/mg.
The RSA was labeled with 14C using a reductive methylation technique modified from Eng25 using [14C]formaldehyde. The specific activity achieved was 2.02 x 106dpm/mg.
Column Chromatography
Plasma and urine samples were analyzed using a Sephadex G-100
column (column dimensions, 1.5 x 75 cm2) or
a Sephacryl S-300 column (column dimensions, 2 x 66
cm2). The columns were run at 4°C with a flow
rate of 20 ml/h and the buffer changed to PBS with 2 mg/ml of bovine
serum albumin and 0.02% sodium azide. The columns were calibrated
using blue dextran (2 mg/ml) and tritiated water to determine the void
volume and the total volume, respectively. Ninety-five fractions of
1.65 ml (G-100) or 1.45 ml (S-300) were collected with recoveries
between 95 and 105% obtained. Kav was determined
by the formula (Ve -
Vo)/(Vt -
Vo), where Vo is the void
volume, Ve the elution volume, and
Vt is the total volume of the column.
The Kav for radiolabeled albumin, transferrin, and IgG on Sephacryl S-300 was 0.510, 0.429, and 0.367, respectively. The FC of [3H]Ficoll corresponding to these protein radii was then determined by examining the elution of urine and plasma samples containing [3H]Ficoll at these Kav values.
In Vivo FC Studies of Radiolabeled Proteins and Polydisperse Ficoll 70 Using the Osmotic Pump Method
As previously described12
Alzet osmotic pumps were
filled with [14C]albumin and one of
[3H]IgG,
[3H]transferrin or polydisperse
[3H]Ficoll 70, along with 0.002% w/v of
thimerosal to inhibit bacterial growth. The Alzet osmotic pumps model
2001 have a mean filling volume of 227 ± 6 µl, pumping rate
0.96 ± 0.04 µl/hour, length 3 cm, diameter 0.7 cm, and unfilled
weight 1.1 g. Once filled the pumps were then incubated in PBS for
4 hours at 37°C. The concentrations of radioactivity initially in the
pump were [3H]IgG 5.26 x
108
dpm/ml,
[3H]transferrin 6.79 x
108
dpm/ml, [14C]RSA
1.62 x 107
dpm/ml, and polydisperse
[3H]Ficoll 2.54 x
108
dpm/ml.
The rats were anesthetized with ether and the osmotic pumps were
implanted subcutaneously between the scapulae using sterile technique.
The rats were given one pump filled with
[3H]Ficoll 70 or two pumps, one with
[14C]albumin and one with either
[3H]immunoglobulin or
[3H]transferrin. The rats were then maintained
as pairs in rat boxes with free access to food and water at all times,
being placed in metabolic cages on days 2, 5, and 7 for 24 hours where
urine collections and corresponding plasma samples were taken at the
end of the urine collection period. Both blood and urine samples were
centrifuged at 3000 rpm in a KS-5200C Kubota bench top centrifuge
(Kubota Corp., Tokyo, Japan) for 10 minutes and then analyzed for
radioactivity. Samples taken on days 2, 5, and 7 to ensure that steady
state was reached on day 7 as shown previously using the same osmotic
pumps.12
Figure 1
demonstrates that labeled protein (data shown for albumin) remains
intact from day 0 to day 7 within the osmotic pump and in day 7 plasma
for control, PAN, and anti-GBM GN rats. Degradation products were not
present in these samples. On day 7 the animals were anesthetized with
Nembutal and sacrificed by cardiac puncture and their kidneys were
removed for radioactivity analysis and histological assessments. Figure 2
examines the turnover over a bolus
injection of [3H]albumin in various tissues
including the kidney. Throughout the period of 7 days plasma
radioactivity continuously decreases with significant uptake
particularly by the muscle tissue. Uptake by the liver and kidney was
relatively low (<5%). There was no evidence of tissue uptake with
rapid resynthesis and disappearance of labeled material from the tissue
into the plasma. Glomerular filtration rate was determined by the
creatinine assay.26
The urine flow rate was determined by
measuring the volume of the 24-hour urine collection in ml.
|
|
Because the specific activity of the carboxymethyl Ficoll 40 was too low to allow it to be used in the osmotic pumps we examined its clearance as compared to uncharged Ficoll by a short-term steady-state method. Sprague-Dawley rats were injected in the tail vein with 4.0 x 107 dpm of [3H]Ficoll 70 or 2.0 x 107 dpm of [3H]carboxymethyl Ficoll 40 and placed in a metabolic cage. Urine was collected between 4 and 6 hours (around midday to 2 PM) after the injection by the urine container in the metabolic cage and by collection from the bladder at 6 hours. All experiments were performed at the same time of day. There was a small reduction in the plasma radioactivity throughout this period; the plasma concentration for molecules with 36Å radius was taken as the mean of 4- and 6-hour plasma radioactivity as determined by size exclusion chromatography. Although this method is not as accurate as the osmotic pump method, the major issue to be resolved is whether there were significant differences in the FC of Ficoll 70 as compared to carboxymethyl Ficoll 40. Blood was collected via the tail vein into an heparinized syringe at 4 hours and a cardiac puncture was performed at 6 hours with a heparinized syringe. All blood and urine samples were spun at 3,000 rpm for 10 minutes in a Kubota bench centrifuge and the resulting plasma and urine counted for radioactivity. Plasma and urine samples at 6 hours were analyzed for creatinine.26 The plasma and urine samples were fractionated on a Sephacryl S-300 column as described vide supra and FCs of fractions with radii of 36Å determined. The charge integrity of the carboxymethyl Ficoll was checked on Q-Sepharose where it would normally elute at a sodium chloride concentration of 0.45 mol/L under a linear gradient eluting condition of sodium chloride concentration ranging from 0.05 to 1.2 mol/L. The charge concentration of the carboxymethyl Ficoll was not changed after excretion in urine. The salt concentration required to elute the carboxymethyl Ficoll from Q-Sepharose is equivalent to a dextran sulfate with a degree of substitution of 0.4 sulfates per glucose residue.
Processing of Albumin in Vivo
To examine possible degradation products arising from the
extra-renal degradation of albumin, rats were anesthetized and their
kidneys were prevented from filtering by occlusion of both ureters.
After 10 minutes, 2 x 107
dpm of
[3H]albumin was injected into the femoral vein.
The rats were then maintained under anesthesia for 4 hours. The rats
were then sacrificed by cardiac puncture and the plasma analyzed for
radioactivity. The integrity of the labeled tracer in the plasma was
also analyzed by size exclusion chromatography.
Analysis of Kidneys for Radioactivity
One of the kidneys excised from the rat was weighed, minced, and made up to 3 ml with 1.4 mol/L NaOH in glass centrifuge tubes (the other kidney was used for histology). The centrifuge tubes were covered loosely and suspended in boiling water for 15 to 30 minutes to allow complete digestion. Four sample aliquots of 100 µl each were taken, 50 µl of hydrogen peroxide was added to decolorize the samples, and the volume made up to 1 ml with 850 µl of water. Four µl of scintillation fluid was added to the samples and they were rested in the dark overnight to reduce chemiluminescence. The samples were counted for radioactivity and the presence of the tracer in the kidneys determined as dpm/g of tissue.
Albumin RIA
The RIA used 125I-labeled RSA, prepared using the Chloramine T method,27 rabbit antiserum (polyclonal) to rat albumin, and sheep anti-rabbit antibodies. The urinary albumin concentration measured by this double-antibody RIA had an interassay coefficient of variation of 7% at a concentration of 180 ng/ml. The detection limit of the assay was 31.2 ng/ml. The standard curve was prepared using a RSA standard (1 mg/ml) that was diluted to give a range of 4000 to 31.2 ng/ml.
IgG RIA
The RIA used 125I-labeled rat serum IgG, prepared according to the Chloramine T method27 using 10 µl of a 1 mg/ml IgG standard, rabbit antiserum (polyclonal) to rat IgG, sheep anti-rabbit antibodies, and nonimmune rabbit serum. The detection limit of the assay was 31.2 ng/ml. The standard curve was prepared using a rat serum IgG standard (1 mg/ml) that was diluted to give a range of 4000 to 31.2 ng/ml.
Transferrin RIA
The RIA used 125I-labeled rat serum transferrin, prepared according to the Chloramine T method27 using 10 µl of a 1 mg/ml transferrin standard, rabbit antiserum (polyclonal) to rat transferrin, sheep anti-rabbit antibodies, and nonimmune rabbit serum. The detection limit of the assay was 6 ng/ml. The standard curve was prepared using a rat serum IgG standard (1 mg/ml) that was diluted to give a range of 1600 to 6 ng/ml.
Total Protein Assay
Total urinary protein was measured with a bicinchoninic acid protein assay kit (that measures peptide bonds) (Pierce, Rockford, IL).
Counting of Radioactivity
Tritium and [14C] radioactivity was determined using a 1:3 sample to Optiphase scintillation fluid ratio28 and recorded on a Wallac 1409 liquid scintillation counter (Wallac, Finland).
Statistics
All data are expressed as mean ± SD with n representing the number of determinations. Significance of results was determined using the Student’s t-tests.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The induction of PAN or anti-GBM GN led to an increase in urine
flow rate, a reduction in glomerular filtration rate (for PAN), and a
reduction of plasma protein concentration (Table 1)
. Microscopically, animals with
anti-GBM GN exhibited glomerular hypercellularity, focal periodic
acid-Schiff deposits in most glomeruli, occasional early crescents, and
tubular casts. There was a marked glomerular and interstitial
infiltrate of ED1+ macrophages and W3/13+ T cells, with areas of
tubulitis seen. PAN also exhibited glomerular hypercellularity and
tubular casts, with prominent macrophage and T-cell infiltration in the
interstitium and areas of tubulitis.
|
Table 2
shows the concentration of
radiolabeled material in the plasma as compared to the kidneys on day 7
of the steady-state osmotic pump studies. For all of the tracers there
is no more than 2 to 3 times the plasma levels in the kidney, this
being representative of normal flux through the kidney, rather than
major accumulation of the tracer. In both control and disease states
the percentage of the osmotic infusion rate appearing in the urine is
<1%. The urinary excretion rate for
[14C]albumin was lower than the other labeled
compounds because the [14C]albumin had
considerably lower specific activity. For all radiolabeled proteins the
net amount of protein being infused into the blood supply was
negligible compared to the plasma protein concentration.
|
. The size exclusion
analysis of the control urines (Figure 3)
showed that in the controls, most of the protein was excreted in the
form of fragments, with a small proportion of the radiolabeled proteins
excreted in macromolecular form (we use the term "macromolecular"
to identify the protein as distinct from "intact" to allow for the
possibility that some minor alteration may have occurred to some of the
protein during renal passage even though it elutes at essentially the
same position on the size exclusion column) (Table 3)
. The degradation is not the result of
enzymes present in the urine9,10
or the result of
extrarenal degradation before filtration because the material in the
plasma was intact (Figure 1)
. This was further confirmed in studies of
control rats with nonfiltering kidneys. At 4 hours after the
administration of intact [3H]albumin the plasma
contained 2 x 106
dpm/ml of radioactivity.
When analyzed on Sephadex G-100 (loading quantity, 100,000 dpm) there
was no low-molecular weight material present (not shown). If the
kidneys had been filtering normally we would have expected at least
720,000 dpm of radiolabeled material (mostly fragments) in urine. The
fact that no radiolabeled fragments were observed in the 4-hour plasma
provides direct evidence that albumin fragments in urine are produced
specifically by the kidney and not by extra-renal sources.
|
|
2% to nearly
90% of the total albumin excreted (Figure 3)
.
The FCs of the IgG, transferrin, and albumin were determined using both
radiolabeled proteins and the RIA (Table 4)
. By radioactivity the FC of the IgG,
transferrin, and albumin in the controls were, respectively, 220 times,
420 times, and 380 times greater than the FC determined by the RIA.
Anti-GBM GN and PAN both result in a significant increase in FC of all
three proteins as determined by both RIA and radioactivity. The large
increase in FC by RIA with anti-GBM GN and PAN reflect the large shift
to macromolecular protein excretion caused by the diseases (Table 3)
.
|
. This suggests that the RIA is
specific for macromolecular transferrin and IgG and does not detect
small fragments of these proteins; a similar phenomenon was found for
albumin (see the Introduction).
|
. With the induction of PAN and
anti-GBM GN the excretion rates of the three proteins as determined by
RIA increased significantly from the controls, whereas by
radioactivity, only albumin and PAN IgG increased significantly. The
difference in the excretion rates as determined by the two methods,
loses significance for albumin with disease because fragmentation is
inhibited. Total protein excretion as determined by the bicinchoninic
acid assay gave values for the control of 79.4 ± 13.3 µg/min
(n = 7), for PAN as 327.5 ± 104.3 µg/min
(n = 7), and for anti GBM GN as 454.6 ±
131.9 µg/min (n = 7). These values are
comparable with albumin excretion rates determined by radioactivity. It
is assumed that total protein estimates will only be semiquantitative,
particularly for controls, because of variability of chromogen
formation with low molecular peptides.14,15
|
shows the increase with disease
of the FC and excretion rate of the three proteins by RIA and
radioactivity. Although the increase by RIA is in the range of 200 to
6600 times, the increases by radioactivity are between 2 to 14 times.
|
Experiments to test charge selectivity of Ficoll corresponding to
radius of 36Å gave similar values for both Ficoll 70 and carboxymethyl
Ficoll (Figure 5)
. Ion exchange
chromatographic analysis of the carboxymethyl Ficoll in the urine
demonstrated that the charged Ficoll retained all its charged groups
(not shown). The value obtained for Ficoll 70 (radius 36Å) with these
short-term experiments is also similar to that obtained for Ficoll 70
(36Å) as studied by the osmotic pump method (Figure 6)
to be discussed in more detail below.
These results demonstrate that charge selectivity is negligible and
that the FC of Ficoll is considerably higher than that measured for
albumin by RIA or radioactivity as studied in control rats (Figure 5)
.
|
|
To determine whether the FC increases of the proteins were the
result of size-selective changes in the GCW, polydisperse
[3H]Ficoll 70 was introduced into the
circulation of control, PAN, and anti-GBM GN rats using osmotic pumps
and the FC of 36Å, 48Å, and 55Å Ficoll was determined (because these
radii correspond with the radii of albumin, transferrin, and IgG,
respectively) using size exclusion analysis on Sephacryl S-300. Figure 6
illustrates the changes in the FC of albumin, transferrin, and IgG
(in order) as determined by RIA and radioactivity and the FC of the
relevant-sized [3H]Ficoll in both controls,
PAN, and anti-GBM GN. It can be seen that whereas there was an increase
in the FC of albumin by RIA and radioactivity in PAN and anti-GBM GN,
there was no corresponding increase in the FC of the 36Å Ficoll in the
disease states. For the IgG, an increase in FC was observed by both RIA
and radioactivity with the diseases. For both PAN and anti-GBM GN a
change in the FC of the 55Å Ficoll was observed, with this change
being not significantly different from the change observed for IgG
using radioactivity. For transferrin, an increase in FC was observed by
RIA and radioactivity. The FC of the 48Å Ficoll was greater than
either the control or the disease radioactivity results, indicating
that like the case of albumin, something other than the GCW is
influencing transferrin’s passage through the kidney.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
that the RIA only detects macromolecular IgG and
transferrin.11,13
As the RIA fails to detect the protein
fragments that are excreted, its use results in a severe
underestimation of the FC of proteins. It is apparent that normal
functioning of the kidney entails almost complete degradation of
filtered protein before excretion.
All efforts to demonstrate that some or all of the fragments may
possibly have been derived from extra-renal sources have proved
negative. Studies presented in Figures 1 and 2
demonstrate that no
fragments were apparent in the plasma at any time during the course of
the study nor was there any significant uptake and resynthesis of
radiolabeled material by extra-renal tissues. The method used to detect
fragments using radiolabeled material and chromatography represents an
extremely sensitive method to detect fragments. We have previously
demonstrated that production of extra-renal-labeled fragments is
negligible10
during the simultaneous appearance of the
fragments in urine. The studies of rats with nonfiltering kidneys
conclusively demonstrated that fragments in urine are the result of
renal-mediated degradation and not from degradation associated with
extra-renal sources.
The results in Tables 3 and 5
are also an important confirmation that
in severe proteinuric states the RIA analysis for macromolecular
protein is similar to that from radioactivity/chromatography. For
control albumin excretion the RIA would predict 0.26% macromolecular
material whereas the amount measured was 2.3% (Table 3)
, which
reflects the presence of immuno-unreactive material being excreted as
established previously.31
In disease states, however, in
which there are relatively low quantities of immuno-unreactive material
but comparable quantities of macromolecular material and fragments,
particularly for IgG and transferrin, the radioactivity/chromatography
method for estimating macromolecular protein excretion rate is in good
agreement with that measured by RIA. For example, for IgG in anti-GBM
GN the RIA would predict the percentage of macromolecular material to
be 64% (Table 5)
where a value of 53.2 ± 13.6% was obtained
from chromatography (Table 3)
. The excretion rate of IgG fragments was
significant in disease at 1.89 µg/min (as compared to the control
value of 4.46 µg/min). Similarly, for albumin in PAN the excretion of
RIA-sensitive macromolecular material was 93.7% of the total material
excreted as determined by radioactivity (Table 5)
that compares
favorably with the 88.1 ± 5.8% as determined by chromatography
(Table 3)
. The fragment excretion rate in disease was 20.22 µg/min
(as compared to control of 92.04 µg/min). It is clear that the RIA
provides a more accurate value of the excretion of filtered protein in
proteinuric states in which the fragmentation process has been
inhibited as compared to the control.
The massive increases in the FC of all proteins as measured by conventional RIA with PAN and anti-GBM are now recognized to be in large part because of the decrease in the lysosomal degradation process. This could explain similar data in which the large increases in albumin FC in disease were found as compared to changes in dextran clearance.3-7 Previously, these results have been interpreted as the result of changes in the permeability of the GCW possibly involving the development of large pores and the loss of the charge selectivity. This increase in the FC of the proteins cannot be explained by a decrease in charge selectivity as we demonstrate in this study that there is no charge selectivity between charged Ficoll and uncharged Ficoll. This result can be understood relative to previous studies of charge selectivity in which the charged transport probe used has subsequently been shown to be biochemically altered during filtration and renal passage and that this will have a profound effect on clearance measurements; before excretion dextran sulfate is desulfated32 and charged proteins are degraded to peptides.8,9,33 The lack of charge selectivity confirms previous studies showing that the GBM34 and the GCW8,9,32,33,35,36 do not exhibit any charge selectivity and that transglomerular passage is determined solely by size selectivity. Albumin interaction with charged polysaccharides and nonrenal extracellular matrix tissues has also been demonstrated to be independent of charge effects.37-38 The lack of change in glomerular permselectivity for molecules such as albumin would also eliminate the possibility that an increase in the ratio of macromolecular form to fragment form was the result of saturation of the degradation pathway.
To further elucidate possible changes in size selectivity at the GCW
and the role of large pores, polydisperse
[3H]Ficoll 70 was studied in vivo
and within this polydisperse population the FC of Ficoll with radii
36Å, 48Å, and 55Å Ficoll was specifically examined in controls, PAN,
and anti-GBM. Ficoll is a spherical polysucrose, which is filtered
according to the size-selective properties of the GCW and is not
actively reabsorbed or secreted by the tubules. Ficoll is commonly used
to study the sieving properties of the GCW and is becoming the molecule
of choice over dextrans as they are a better representation of globular
proteins. Our results demonstrate that whereas the FC of albumin by
radioactivity increased significantly with PAN and anti-GBM (>0.02 FC
units), there was no significant change detected in the FC of the 36Å
Ficoll (0.1) (Figure 6)
. If glomerular permeability was alone
responsible for the increase in albumin clearance (given all post-GBM
cellular processing is the same) then an increase in the FC of Ficoll
of at least 20% would have been expected but this was not observed;
the trend was that it decreased in disease states. The large pores
associated with the transport of larger molecules such as Ficoll radius
55Å generated increases in FC in disease of 0.004 to 0.011 units.
These increases were not enough to significantly alter the clearance of
Ficoll radius 36Å. Ultimately, the FC of Ficoll 36Å is a balance of
clearance through these larger pores and smaller radius pores. Similar
experimental findings have been consistently made previously in which
the FC of dextran of radius 36Å or Ficoll 36Å has been demonstrated
to be the same or even less in diseased states as compared to the
control3-7,39
even when large pores have been observed to
develop. These results demonstrate that changes in glomerular
permselectivity to albumin are not responsible for the changes in
excretion of albumin.
The results obtained for IgG support the concept that large pores may
develop in the GCW in disease. In control rats the FC of
[3H]IgG is not significantly different to that
of Ficoll with the same radius. This provides further evidence that
this protein is filtered intact and that the degradation occurs distal
to the GBM. Tubular re-absorption is not a significant factor as the
amount of [3H]IgG found in the kidney reaches a
relatively low steady-state level at 7 days (Table 2)
. In disease in
which there is a small but significant increase in the FC of the 55Å
Ficoll (Figure 6)
there is a corresponding increase in the FC of IgG as
determined by radioactivity data. Again there is no accumulation of the
labeled material in the kidney (Table 2)
and the amount present is not
significantly different to that in the control kidney. This shows that
the increase in FC of the IgG by radioactivity in PAN and anti-GBM is
probably the result of small size selectivity changes at the GCW yet
these large pores do not change the FC of albumin-like Ficoll
(irrespective that this Ficoll can be transported through these large
pores). The large increases in FC by RIA in these diseases for IgG will
be the result of an inhibition in the fragmentation process, which is
increasing the excretion of RIA detectable intact IgG.
A further issue related to glomerular structural changes and permeability is one of heterogeneity of glomerular permeability in disease.40 The conclusion from the studies of Yoshioka and colleagues,40 that a direct relationship exists between structural abnormalities and permeability function, was not seen. The increase in the FC seen only for dextrans >36Å was because of hyperfiltering glomeruli that lacked discernable damage. They found no significant change for dextran radius 36Å. These permeability results are consistent with the present studies in vivo in which there was no change in Ficoll 36Å or any change in glomerular filtration rate in disease states. The increase in large pores observed with Ficoll 55Å are consistent with glomerular heterogeneity. The relevant parameter is the fact that at least for anti-GBM GN there was no significant change in glomerular filtration rate as compared to the control. Therefore, although heterogeneity cannot be eliminated it is unlikely to be a major factor to account for albumin clearance.
The passage of transferrin through the kidney appears more like albumin
than IgG. Unlike IgG the FC of transferrin by radioactivity with the
diseases does not follow that of the 48Å Ficoll. Rather we see the
control value of transferrin being 6 times lower than that of the
corresponding Ficoll, and even with PAN and anti-GBM the FC of
transferrin never exceeds the FC of the control Ficoll. The changes in
48Å Ficoll FC in control and disease are not significant. This
indicates that something other than GCW permeability is influencing the
passage of transferrin.
In the case of albumin, it is not necessary for large pores to develop in the GCW for an increase in the FC by radioactivity to occur. A partial inhibition of the retrieval pathway in disease would result in an increased flux of intact albumin, from the glomerular epithelial or tubular epithelial cells where the retrieval pathway may operate to the degradation pathway before excretion in urine. Such an occurrence would result in an increase in FC and excretion rate by both RIA and radioactivity. Although a retrieval pathway has so far been demonstrated for albumin, the possibility exists that similar, if less efficient, retrieval mechanisms may occur to retrieve other highly filtered proteins such as transferrin. With little or no change in the GCW affecting the 48Å Ficoll the much greater change in the FC of transferrin by radioactivity could be attributed to the inhibition of a second retrieval mechanism. Further studies are required to understand the effects of these proteinuric states on the renal handling of proteins and Ficoll of lower molecular weight and size to albumin.
|
Footnotes |
|---|
Dedicated to Prof. Eric Glasgow, a valued friend and highly respected colleague who passed away after a short illness on January 3, 2001. Eric was instrumental in laying the ground work for this study a number of years ago.
Supported by a postdoctoral fellowship from CAPES/Brazil (to M. A. M. G.).
Accepted for publication June 11, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  W. D. Comper, B. Haraldsson, and W. M. Deen Resolved: Normal Glomeruli Filter Nephrotic Levels of Albumin J. Am. Soc. Nephrol., March 1, 2008; 19(3): 427 - 432. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Asgeirsson, D. Venturoli, B. Rippe, and C. Rippe Increased glomerular permeability to negatively charged Ficoll relative to neutral Ficoll in rats Am J Physiol Renal Physiol, November 1, 2006; 291(5): F1083 - F1089. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. Hjalmarsson, M. E. Lidell, and B. Haraldsson Beneficial effects of orosomucoid on the glomerular barrier in puromycin aminonucleoside-induced nephrosis Nephrol. Dial. Transplant., May 1, 2006; 21(5): 1223 - 1230. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Yu, R. M. Sandoval, and B. A. Molitoris Quantitative intravital microscopy using a Generalized Polarity concept for kidney studies Am J Physiol Cell Physiol, November 1, 2005; 289(5): C1197 - C1208. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. L. Hunt, M. R. Pollak, and B. M. Denker Cultured Podocytes Establish a Size-Selective Barrier Regulated by Specific Signaling Pathways and Demonstrate Synchronized Barrier Assembly in a Calcium Switch Model of Junction Formation J. Am. Soc. Nephrol., June 1, 2005; 16(6): 1593 - 1602. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Koltun, J. Nikolovski, K. Strong, D. Nikolic-Paterson, and W. D. Comper Mechanism of hypoalbuminemia in rodents Am J Physiol Heart Circ Physiol, April 1, 2005; 288(4): H1604 - H1610. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Osicka, K. J. Strong, D. J. Nikolic-Paterson, R. C. Atkins, G. Jerums, and W. D. Comper Renal processing of serum proteins in an albumin-deficient environment: an in vivo study of glomerulonephritis in the Nagase analbuminaemic rat Nephrol. Dial. Transplant., February 1, 2004; 19(2): 320 - 328. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Osicka, K. J. Strong, D. J. Nikolic-Paterson, R. C. Atkins, and W. D. Comper Albumin and glomerular permselectivity Nephrol. Dial. Transplant., February 1, 2004; 19(2): 511 - 512. [Full Text] [PDF]
|
|
|
|
 |
|
 G. C. Hard and Kanwar Nasir Khan Invited Review: A Contemporary Overview of Chronic Progressive Nephropathy in the Laboratory Rat, and Its Significance for Human Risk Assessment Toxicol Pathol, February 1, 2004; 32(2): 171 - 180. [Abstract] [PDF]
|
|
|
|
|
|
W. A. Wilmer, B. H. Rovin, C. J. Hebert, S. V. Rao, K. Kumor, and L. A. Hebert Management of Glomerular Proteinuria: A Commentary J. Am. Soc. Nephrol., December 1, 2003; 14(12): 3217 - 3232. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. Wolf and E. Ritz Diabetic Nephropathy in Type 2 Diabetes Prevention and Patient Management J. Am. Soc. Nephrol., May 1, 2003; 14(5): 1396 - 1405. [Full Text] [PDF]
|
|
