| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
3 Chain Noncollagen Domain 1 Is Not Associated with Glomerular Basement Membrane Antibody: A Potential T Cell-Mediated Mechanism1
*
Department of Basic Sciences, Dental Branch, University of Texas Houston Health Science Center, Houston, TX 77030;
Department of Pathology, Texas Children’s Hospital, Baylor College of Medicine, Houston, TX 77030; and
Department of Pathology, University of Virginia, Charlottesville, VA 22908
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
3 chain noncollagen domain 1 (rCol43NC1) induced
severe glomerulonephritis in 100% of Wistar Kyoto rats, 33% of which
died of this disease around day 35 postimmunization. The renal
pathology demonstrated widespread glomerular damage and a mononuclear
cell infiltration within the interstitial tissue. T cells from
immunized rats responded not only to rCol43NC1, but also to isolated
rat GBM. Sera Abs to rCol43NC1 were detectable in 100% of the rats,
but only 20% of the rats had low levels of Ab to isolated rat GBM by
Western blot, and none by immunofluorescence. Furthermore, IgG/M
binding to or C3 deposition on endogenous GBM in immunized rats were
not detected in most of the experimental rats, and showed no
statistical correlation with disease severity. Additionally, no
electronic dense deposition in the glomeruli was detected in all rats.
Those data revealed a disassociation between the disease and
anti-GBM Ab. T cell-mediated mechanisms, which are currently under
our investigation, may be responsible for the glomerular
disease. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Historically, the involvement of the immune system in
glomerulonephritis was first shown by the demonstration of deposits of
Ab within glomeruli (3, 4, 5). This influenced early ideas to
such an extent that it was thought that all forms of glomerulonephritis
could be explained by Ab-mediated mechanisms, either by a direct
interaction with intrinsic glomerular Ag, such as anti-GBM Ab, or
by the deposition of circulating immune complexes, such as Ab-DNA
complex in systemic lupus erythematosus (4, 5, 6). Ab binding
or deposition triggers an inflammatory response in the glomerulus
associated with complement and/or Fc
R pathway (7, 8).
However, the participation of Ab and associated pathways alone cannot
fully explain many aspects of the pathways in experimental and human
glomerulonephritis (9, 10). For example, in one subtype of
human glomerulonephritis (pauci-immune form), the glomerular damage is
not related to deposition or binding of Abs in the glomerulus
(9).
T cell-mediated cellular immunity has long been suspected as potentially the most important mediator of glomerulonephritis (10). T cells may participate as helper cells in T-dependent Ab response to renal autoantigens. Some autoantibodies associated with human disease, such as anti-GBM Ab, have characteristics of a T cell-dependent response (11, 12). Contribution of general T cell populations to glomerulonephritis, especially of the proliferative/crescentic type, has been investigated in animal models either lacking T or B cells, or with an interrupted B7/CD28 costimulation pathway (13, 14, 15). One of the most significant questions for T cell involvement in glomerulonephritis has been whether Ag-specific CD4+T cells could initiate glomerular injury. Attempts have been made to detect Ag-specific T cells in human glomerulonephritis. For example, there was found a weak T cell-proliferative response specific to a GBM Ag in the peripheral blood of anti-GBM disease patients (16). Whether such a weak response is a true reflection of a T cell-mediated attack against the glomerulus remains unclear. Experimental models are needed to provide more direct evidence supporting the involvement of Ag-specific CD4+T cells in glomerular injury.
Several rodent models for anti-GBM glomerulonephritis have been
established by active immunization with collagen 4-chains (Col4),
well-defined autoantigens in GBM for Goodpasture’s syndrome (17, 18). Susceptibility to anti-GBM glomerulonephritis has been
linked to MHC class II genes (19), but it is unclear
whether T cells are merely helper cells in Ab responses, or whether
they directly participate in glomerular injury, due to largely
unanalyzed T cell responses. An appropriate animal model in which Ab
and T cell responses are well analyzed may allow us to distinguish
whether the disease is associated with T cells, Abs, or both.
We have recently developed a rat model in which a single immunization
with highly denatured recombinant collagen IV3 chain noncollagen
domain 1 (rCol43NC1) of murine sequence induced severe
glomerulonephritis. We discovered that the denatured recombinant
protein elicited Ab to itself, but not endogenous GBM. In contrast, T
cells from immunized rats responded to isolated rat GBM. These results
imply that the Ab may not be responsible for disease induction. This
model is currently under investigation for direct evidence for a causal
role of anti-GBM T cells in glomerulonephritis.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The cDNA encoding mouse Col43NC1 was a gift from Dr. J. H.
Miner (Washington University, St. Louis, MO) (20). A pair
of primers with NdeI and BamHI sites at 5' and 3'
ends were designed for PCR based on the published mouse
sequence: 5'-GGG AAT TCC ATA TGC CAG GCT TAA AGG G-3' and 5'-ACG
TGC TGG ATC CTT TGA TTT CGT C-3'. The PCR fragment was isolated,
subcloned into pNoTA/T7 shuttle vector (5 Prime
3 Prime Boulder, CO),
and sequenced by an automatic sequencer. The cDNA fragment was further
released by NdeI and BamHI, and inserted into
pET22b expression vector (Novagen, Madison, WI) (see Fig. 1
A). This procedure resulted in a nonfusion protein with
only an additional N-terminal methionine residue. The cDNA sequences
inserted in pET22b were verified by automatic sequencing. Proteins were
expressed in Escherichia coli strain BL21 (DE3) as inclusion
bodies, purified by a preparative SDS-PAGE (12%), and
electrophoretically eluted from the gel. Purified proteins were
dialyzed against eight changes of PBS. After dialysis, the protein
spontaneously formed aggregates, which could not be dissolved even in 7
M guanidine. Bacterial contamination was essentially removed by
repeated sonication in 100% ethanol and PBS. The purified recombinant
protein, which showed very low activity in stimulating unprimed rat
splenic cells, was stored at -20°C until use. The amino acid
composition of purified protein was determined at the Biochemical Core
Facility, University of Virginia (Charlottesville, VA), and purity was
verified by SDS-PAGE.
|
Female Wistar Kyoto (WKT) rats (4- to 6-wk old) were purchased from Harlan Breeders (Indianapolis, IN). The rats were maintained in the animal facility at the University of Texas Houston Health Science Center, and allowed to acclimate for 3 days. For glomerulonephritis induction, rats were immunized with 300 µg of recombinant protein, emulsified in CFA, in one hind footpad and at the base of the tail. Rats immunized with CFA alone served as controls. Rats immunized with crude GBM (105 glomeruli) in CFA served as positive controls. Random urine samples were monitored daily by Multstix (Bayer, Pittsburgh, PA) or analyzed weekly using a Vitros 250 Chemistry Analyzer (Ortho Diagnostics, Raritan, NJ) in the Clinical Chemistry Laboratory, Texas Children’s Hospital (Houston, TX) starting at day 14. Urine albumin was semiquantitated by 12% SDS-PAGE using BSA as a standard (2 µl urine/lane). Twenty-four-hour urine samples were collected weekly using metabolism cages, and were analyzed as described above.
Blood samples from immunized rats were collected at different time points (pre- or postimmunization) by tail venipuncture, and sera were used for determination of specific Abs. The animals were sacrificed at day 42 postimmunization, and their different tissues were fixed for pathology. Renal tissues fixed in Bouin’s fixative were used for H&E, and fixed in 10% formalin for periodic acid-Schiff, trichrome, or Jones stainings. Part of the renal tissue was snap-frozen in liquid nitrogen for direct immunofluorescence staining or for Ab elution. Selected renal samples were fixed in glutaraldehyde (2%) and processed for transmission electronic microscopy (TEM). TEM examination was conducted in the Department of Pathology, Texas Children’s Hospital.
Isolation of GBM
A previously described method was followed with some modifications for GBM isolation (5, 21, 22). Briefly, the cortex of normal rat kidney was dissected out from the medulla and rinsed with cold HBSS. The cortex was minced and pushed through a stainless steel mesh (no. 150). The fraction through the mesh was washed and collected by centrifugation (1500 x g) to remove soluble proteins, followed by gentle centrifugation (200 x g) to separate the tubules from glomeruli. The gentle centrifugation was repeated until glomeruli exceeded 95% in purity, as determined by microscopic observation. The purified glomeruli were sonicated for 1 h, followed by centrifugation (1500 x g) to remove soluble proteins released from the cells. An aliquot of the sonicated glomeruli, designated as crude GBM, was directly frozen down for SDS-PAGE, Western blot, and immunization. The remaining part was further digested by collagenase (CLSPA) at 37°C for 20 h (Worthington, Woodland, NJ), followed by heating at 80°C for 15 min. The digested GBM was stored at -80°C and used as Ags for lymphocyte proliferative assay (LPA).
LPA
For the rats immunized with rCol43, draining lymph nodes were
removed and a single-cell suspension was prepared 2 wk after
immunization. T cells were enriched by a rat T cell enrichment column
(R&D Systems, Minneapolis, MN), and the purity of T cell population was
verified by flow cytometry using anti-rat CD4-PE and anti-rat
IgG/M-FITC (PharMingen, San Diego, CA). Thymocytes were isolated
from normal rats and irradiated (2000 rad). T cells and thymocytes (at
a 1:1 ratio) were cultured in 96-well plates at 4 x
105 cells/well in 200 µl of DMEM supplemented
with 10% FCS, 2 mM L-glutamine, nonessential and essential
amino acids, sodium pyruvate, 100 U/ml penicillin, 100 µg/ml
streptomycin (BioWhittaker, Walkersville, MD), and 5 µM
2-mercaptoethanol, with or without Ags in triplicate. The digested GBM
from 
25 glomeruli was added to each well. Recombinant Col43NC1
was added to a final concentration of 10 µg/ml. Purified protein
derivative was used as positive control. The cells were incubated at
37°C in a humidified, 5% CO2 atmosphere for
72 h. The activated cells were pulsed with
[3H]thymidine, 0.5 µCi/well, for 18 h
(ICN, Costa Mesa, CA). The cells were harvested onto glass fiber
filters using a semiautomatic cell harvester (Skatron, Sunnyvale, CA),
and incorporated radioactivity was measured by a liquid scintillation
counter (Beckman, Fullerton, CA). The results were expressed as 
cpm
(mean triplicate cpm with Ag – mean triplicate cpm without
Ag).
In vitro production of Ab
The rats were immunized with 200 µg of rCol43NC1 in CFA and
boosted 2 wk later with rCol43NC1 (20 µg/rat) in Incomplete
Freund’s Adjuvant. The lymph nodes were harvested from these rats, and
a single cell suspension was made. Using similar culture conditions as
that for the LPA, the lymphocytes (1 x 107
cells/ml) were cultured with or without rCol43NC1 (10 µg/ml).
Lymphocytes from CFA-immunized rats were used as controls. Supernatants
were collected 4 days later and kept at -80°C until use. Rat IgG in
the supernatants was determined by ELISA using purified rat IgG as
standard.
Elution of Abs
Kidneys from experimental or control rats were stored at -80°C until use. The cortical portions of the kidneys were dissected out by slicing with a razor blade, mixed with cold PBS (pH 7.2), and homogenized in a 5-ml Dounce homogenizer. This homogenate was spun at 3000 x g for 15 min at 4°C. The sediment thus obtained was washed with cold PBS seven times by centrifugation. After the final wash, the sediment was used for elution of Ab by two published methods (5, 23). In the first method, the sediment was suspended in 0.02 M citrate buffer, pH 3.2 (5 parts buffer:1 part sediment, v/v), and incubated at room temperature with constant shaking for 2 h (5). After incubation, the mixture was spun at 10,000 x g for 30 min at 4°C. The supernatant was removed, immediately brought to pH 7.0 with 0.1N NaOH, and dialyzed against several changes of PBS. To avoid exposure of Ab to low pH, the second method was also applied (23). In the second method, the sediment was mixed with 5 times volume of 2.5 M potassium iodide in Tris-HCl buffer (pH 9.0) and rocked for 50 min at room temperature. After centrifugation at 10,000 x g at 4°C for 30 min, the supernatant was dialyzed against cold PBS. The amount of rat IgG in the eluate was measured with ELISA, and the eluate was concentrated by a centricon device (Millipore, Bedford, MA).
Detection of Abs
Blood samples were taken from the tail vein weekly or at end of the experiments. Sera were isolated and kept at -80°C until use. Ab response was examined at three levels: Abs to 1) recombinant proteins by Western blot and ELISA, 2) crude rat GBM by Western blot, and 3) native GBM by immunofluorescence.
Western blotting was used to detect Ab to the recombinant protein or crude GBM proteins. Briefly, either recombinant protein (12% gel) or crude GBM proteins (7.5% gel) were separated by SDS-PAGE and transferred to nitrocellulose membranes. The membranes were blocked and incubated with diluted sera at 1/5,000 to 1/20,000 for the recombinant protein, and 1/1,000 for crude GBM. After washing with PBS containing 0.05% Tween 20, the blots were incubated with HRP-labeled goat anti-rat IgG (1:10,000). The reactants were visualized by ECL (Pierce, Rockford, IL).
ELISA was also used to measure Ab response to rCol43NC1. Briefly,
rCol43NC1 after elution during purification was dialyzed for 3
h against water. At this point, rCol43NC1 was still in soluble form
and diluted 1/10 in carbonate buffer (pH 9.5). Fifty microliters of
diluted rCol43NC1 was added to each well of a 96-well plate. After
overnight incubation, the wells were washed with a solution (70%
methanol, 30% Tris-HCl, pH 7.2), followed by PBS-Tween 20
buffer. The plate was blocked with 3% BSA for 3 h at room
temperature. Serial dilutions of the sera were added to the wells in
duplicate. After a 1-h incubation, the plate was washed and goat
anti- rat IgG-HRP (1:10,000) was added. The color was developed
using the substrate o-phenylenediamine (Sigma, St.
Louis, MO), and OD values at 490 nm were determined by an ELISA reader
(Molecular Devices, Sunnyvale, CA).
For detecting Abs to GBM, indirect immunofluorescence was conducted. Sera were diluted at 1/50 in 3% BSA-PBS and added onto frozen sections of normal rat or SCID mouse kidney. After a 1-h incubation, the sections were incubated with FITC-labeled goat anti-rat IgG or IgM Abs (1:50, Southern Biotechnology Associates, Birmingham, AL). The sections were viewed by fluorescence microscopy (BH-2; Olympus, New Hyde Park NY).
For detection of GBM-bound IgG/M or C3 deposition in the glomerulus, direct immunofluorescence assays were conducted. The kidneys from experimental rats were snap frozen, and frozen sections were cut. The sections were fixed in cold acetone for 10 min and incubated with FITC-labeled goat anti-rat IgG or IgM Abs (1:50, see above). In parallel, the sections were incubated with anti-rat C3 Ab labeled with FITC (1:100; ICN).
Sera transfer to naive recipients
Blood was collected by heart punch when the animals were
sacrificed, and sera was isolated. Sera Abs to recombinant protein or
peptides were determined, pooled, and stored at -80°C until use.
Three milliliters of sera either from rCol43- or CFA-immunized rats
was passively transferred by i.p. injection into naive WKT rats of 4–6
wk of age. For SCID mice, each animal received 0.5 ml of sera. Some
sera recipients were sacrificed 24 h later for detecting GBM
binding IgG. Others were kept for up to 2 wk, and their proteinuria and
renal pathology were examined.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
3NC1
Recombinant Col43NC1 was chosen as the Ag for the present
study. Keeping in mind that this model was designed to investigate the
Ag-specific T cells in glomerulonephritis, we considered the amino acid
sequence of the Ag to be more important than the correctness of its
three-dimensional structure. Two decisions were made for production of
this recombinant protein as the Ag. First, we decided to use a cDNA
encoding murine Col43NC1, which is highly homologous in amino acid
sequence to rat Col43NC1 (96.5%), for construction of expression
vector. We believed that the recombinant protein of mouse sequence
might be able to elicit CD4+ T cell responses to
endogenous rat Col43NC1. Second, we decided to express rCol43NC1
without a His tag or other fusion partner. This was based on our
previous experience that His tag and/or the leading sequence of a
fusion protein might become a major T cell epitope and thus skew the
immune response away from the core protein (N. Griggs, K. S. K. Tung,
and Y.-H. Lou, unpublished observations).
Murine rCol43NC1 (an initiating methionine plus 246 aa residues)
expressed in E. coli was highly denatured. This protein
aggregated after intensive dialysis following purification, and could
not be dissolved in 7 M guanidine. SDS-PAGE analysis of the purified
protein showed a single band with an expected molecular mass (26.1 kDa)
(Fig. 1
B). The correct
sequence of the protein was further verified by amino acid composition
analysis (Fig. 1C). The amino acid composition of
rCol43NC1 was very close (error <5%) to the expected values,
suggesting minimal bacterial protein contamination in the purified
protein.
Recombinant Col43NC1 induced severe glomerulonephritis in
WKT rats
WKT female rats 4 to 6 wk of age (average 65 g) were
immunized with rCol43NC1 emulsified in CFA. The immunized rats began
to develop proteinuria and hematuria 14–18 days later. Both
proteinuria and hematuria progressed rapidly and plateaued around day
21–26 (protein >2000 mg/dl, hemoglobin >10 mg/dl) (Fig. 2A). The average concentration
of urine proteins in the immunized rats at day 30 was 2790 ± 168
mg/dl with a range from 2016 to 3955 mg/dl. The proteinuria dropped
slightly after 35 days. Severe albuminuria for each sample was
confirmed by semiquantitative SDS-PAGE (Fig. 2B). In the
first series of experiments, 5 of 15 immunized rats had severe ascites
around day 28, probably due to loss of large quantities of serum
albumin. The volumes of urine produced in 24 h in those animals
reduced to <5 ml/24 h. The animals with ascites died around 35 days.
Except for creatinine, which dropped to one-fifth to one-tenth the
normal level, other urine indices in these rats were generally normal.
Blood urea nitrogen and sera creatinine were measured for the rats with
ascites around day 30. Elevated blood urea nitrogen concentration
(179 ± 19 mg/dl) and sera creatinine (4.5 ± 0.6 mg/dl) were
observed.
|
, A and C, insets). A thorough search for the signs
of pathology in various organs was conducted in five rats (including
two that died around day 35) with severe glomerulonephritis.
Histopathology did not reveal any abnormalities in other organs. Renal
tissues from all experimental animals showed extremely severe
glomerular lesions (Fig. 3, C–F) histologically similar to
the human rapidly progressive glomerulonephritis. Glomerular lesions,
ranging from hypercellularity to crescent formation, affected nearly
100% of glomeruli in all cases. Crescent formation, which was the most
common form of the lesion, affected nearly 100% of glomeruli in about
half of the experimental rats (Table I). Increased cellularity in the
interstitial tissue, mainly due to infiltration of mononuclear cells
around glomeruli and blood vessels, was notable. Other pathological
evaluation (Jones, periodic acid-Schiff, trichrome stainings,
and EM) showed thickened irregular GBM and increase in mesangial matrix
(Fig. 3, D and F). Histology revealed frequent
protein accumulation in the lumens of the tubules, suggesting leaking
of large quantities of serum protein from glomeruli, which resulted in
proteinuria (Fig. 3C). The lesion was primarily restricted
to glomeruli. Enlargement of tubules in some areas were also notable.
The pathology findings were very consistent in eight independent
experiments (n = 37) (Table I).
|
|
, A and B). Immunization with crude GBM also induced glomerulonephritis in three rats. However, the induced disease was mild in two and minimal in one. Proteinuria never exceeded 1000 mg/dl in all three animals.
Characterization of circulating Abs in rats immunized with
rCol43NC1
GBM Ab has been shown to initiate anti-GMB glomerulonephritis
either in experimental models or human diseases (4, 5).
Therefore, it would be critical to determine the Ab response in the
experimental rats. As described above, rCol43NC1 probably was highly
denatured, and thus was very different in its three-dimensional
structure from the native Col43NC1 in endogenous GBM. The Abs
elicited by the recombinant protein might display complex
specificities, and were analyzed at several levels: 1) the circulating
Ab to rCol43NC1 by Western blot and ELISA; 2) the circulating Ab to
rat GBM by indirect immunofluorescence on normal rat renal tissue, and
Western blot with crude rat GBM proteins; and 3) Ab deposition in or
binding to GBM in the immunized rats by direct immunofluorescence and
electron microscopy. In addition, C3 deposition in the GBM in the
immunized rats was also examined by direct immunofluorescence.
The Ab to rCol43NC1 became detectable after 14 days, and increased
thereafter in >60% of animals (Fig. 4, A and C). High levels of circulating Ab were
observed around or after 21 days in those animals. However, Ab levels
were quite variable among the animals. In some animals (<40%), no
increase in Ab levels was observed. There was no correlation between Ab
level and disease severity (Fig. 4B).
|
3NC1 in >60% of
animals, Western blot analyses showed detectable Abs to crude rat GBM
in only some animals (5 of 25) (Fig. 4C). No correlation
between the Abs to crude rat GBM and rCol43NC1 was seen (Fig. 4C). In contrast, immunization of crude GBM elicited a
relatively high level of Ab to GBM proteins but not rCol43NC1 (Fig. 4C). Interestingly, rats with extremely severe disease
(usually with ascites) produced very low levels of Abs to rCol43NC1,
and had no detectable Ab to isolated rat GBM by Western blot.
Circulating Abs to native protein in GBM were under detectable levels
during the entire experimental period in all animals, as revealed by
indirect immunofluorescence on normal rat kidney (data not shown).
We further transferred a large quantity of sera with high levels of Ab
to rCol43NC1 into SCID mice, which have no endogenous Igs. Again, no
rat IgG binding to GBM was observed (Fig. 5A). As a control, transfer of
antisera from rats immunized with the crude GBM revealed intensive
staining of IgG in the glomerulus of the recipient (Fig. 5B). Sera from very sick animals were transferred into naive
rats (n = 3). However, no sign of proteinuria or
glomerulonephritis was observed. Again, no IgG binding to GBM was
observed in the recipients.
|
3NC1-immunized rats
It is possible that the anti-GBM Ab in the circulation could
have been absorbed by in situ binding to endogenous GBM or other
tissues such as lung in the immunized animals. A search for binding of
IgG/M in glomeruli and lung was conducted by direct immunofluorescence
in all immunized rats. We could not detect any linear binding of IgM to
GBM in renal tissues of all rats. IgG binding to GBM was weakly
positive or positive detected in 19 of 37 (Fig. 6A). The other immunized rats
showed a similar background staining of GBM as seen in normal kidney
(Fig. 6, B and C). No correlation between GBM
bound IgG and disease severity (Table I). No IgG or IgM binding to lung
basement membrane was observed in any rats. TEM analyses also failed to
demonstrate Ig deposition in five rats with extremely severe disease
(data not shown). In addition, weak C3 deposition in glomeruli was
found only found in a few animals (data not shown). Again, no
correlation between C3 deposition and disease severity was seen. As
expected, direct immunofluorescence showed linear staining for GBM in
rats immunized with crude rat GBM.
|
3NC1. Supernatant from the culture was tested
for rat IgG concentration and Ab activities to rCol43NC1 and crude
GBM via Western blot, and endogenous GBM by immunofluorescence assay.
High levels of Ab activity to rCol43NC1 were detected by Western
blot or ELISA, as well as high quantities of rat IgG (1.02 ± 0.14
µg/ml). However, Abs specific to GBM in the supernatant were
undetectable, either by immunofluorescence on the normal renal tissue,
or by Western blot with crude GBM. T cell response elicited by rCol4a3NC1
As we mentioned earlier, rCol43NC1 was extremely denatured and
was unglycosylated. This protein probably possesses no B cell epitopes,
which may be identical or similar to those of native protein.
Therefore, immunization with rCol43NCl probably induced an Ab
response only to itself, but not to native Col43 chain in GBM. In
contrast, T cell response is mainly determined by the linear sequence
of a protein or a peptide. It is highly possible that the T cell
response elicited by rCol43NC1 would react with self-Col43 chain
in the host.
T cell responses in rCol43NC1-immunized rats were determined by
LPAs. T cells were isolated from lymph nodes of either rCol43NC1-
(n = 5) or CFA-immunized rats (n = 3)
(Fig. 7B). As shown in Fig. 7A, T cells from rCol43NC1-immunized rats responded
strongly to rCol43NC1, and a persistent response to the digested GBM
was also detected. In contrast, T cells from CFA-immunized controls
also slightly responded to rCol43NC1, probably due to a minute
amount of bacterial contamination in the rCol43NC1, but did
not react with digested GBM. The result demonstrated that rCol43NC1
elicited a T cell response to the endogenous Ag Col43.
|
3NC1 may not be associated with anti-GBM
Ab. Other components of the immune system, presumably
CD4+ T cells, might play a causal role in
glomerulonephritis in this model. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
3NC1 was capable
of inducing extremely severe glomerulonephritis in WKT rats. Several
unique results from this glomerulonephritis model should be emphasized.
First, circulating Ab levels to rCol43NC1 in the immunized rats were
highly variable despite similar severity of the disease in those
animals. Second, IgG/M binding to and/or C3 deposition on endogenous
GBM was statistically not correlated with severity of the disease.
Third, Ag-specific T cell responses to isolated rat GBM were detected
in rats immunized with rCol43NC1. These results suggest that
anti-GBM Abs may not be associated with glomerular injury in this
particular model. Traditionally, Ab and its associated pathways have
been implicated in causing glomerulonephritis. However, an Ab-mediated
mechanism alone does not explain many cases of human
glomerulonephritis, or some aspects of pathogenesis in this disease.
Our model may allow us to explore other mechanisms, especially T
cell-mediated mechanisms, in mediating glomerulonephritis.
Many animal models have been developed by active immunization with GBM
collagen in the past several years (18, 19, 22, 24, 25, 26).
In most models, the Ags were either purified native GBM proteins from
other species or recombinant proteins expressed in mammalian cells that
also closely resembled the native protein. In those models, Abs to
endogenous GBM were uniformly induced (18, 19, 22, 27).
Although some models used peptides as Ags (22, 25, 26),
only one group was successful in induction of glomerulonephritis using
keyhole limpet hemocyanin-linked peptides (25). As
the anti-GBM Ab was also detectable in the immunized rats, the
authors considered the induced Ab to be a mediator of the disease. In
contrast, T cell responses in those models were largely not analyzed.
Hence it was difficult to determine which mechanism, Ab, T cells, or
both, was most important in the process of glomerular injury. In
comparison to those models, we have used a nonfusion, highly denatured
rCol43NC1 as the Ag with both T cell and Ab responses being
carefully analyzed in the rats immunized with this recombinant protein.
Thus, we were able to determine whether the elicited Ab and T cell
responses were against endogenous GBM protein. As we described above,
we did detect a T cell, but not Ab, response to endogenous GBM. This
led us to suspect that T cells may play a crucial role in the process
of glomerular injury. In fact, we have generated several Ag-specific
CD4+ T cell lines from rCol43NC1-immunized
rats, and those T cell lines are currently being tested for their roles
in glomerulonephritis.
Abs specific to glomerular Ags have been repeatedly shown to be major
players in glomerulonephritis, probably inflammation in the lung
(3). Several methods have been used to determine
anti-GBM Abs in our model. The results from serial experiments,
aiming to analyze Ab specificity, led to our conclusion that
rCol43NC1 may elicit no or undetectable Abs against native rat
Col43 chain in GBM. First, we analyzed sera Ab to rCol43NC1.
There was no positive correlation between disease severity and Ab
levels to rCol43NC1. Rats with the most severe disease usually had
the lowest Ab levels against rCol43NC1. Furthermore, only a few
animals showed sera Abs to GBM by Western blot, and none by
immunofluorescence. We realized that the sera Abs could not represent
the original repertoire of Ab specificities, as anti-GBM Abs might
have been absorbed by endogenous GBM. Therefore, we tested in
vitro-produced Abs, which are not absorbed by endogenous Col43
chains. In vitro-produced Abs reacted strongly with rCol43NC1, but
not with native rat GBM by any methods. Finally, the results were
further confirmed by direct detection of IgG binding to GBM and C3
deposition in the glomeruli. Among rats with severe glomerulonephritis,
Abs binding to native GBM or C3 deposition in glomeruli were either not
detectable or weakly detectable. Clearly, rCol43NC1 may only share
few, if any, B cell epitopes with native rCol43 chain in
GBM.
The next question is why rCol43NC1 used in our study shows little or
no similarity of B cell epitopes with native GBM. First, collagen
fibers in GBM are a very complicated, highly organized multimolecular
structure (28). NC1 domains from three collagen IV
-chains together compose a global head of the collagen fiber, and
these global heads further bind to each other to form a special
membrane network (28, 29). Thus, native B cell epitopes of
GBM may not only depend on the three-dimensional structure of a single
collagen chain, but also on the structures formed between molecules. It
was not surprising that some peptides derived from NC of GBM collagen
IV chains elicited Abs only to denatured, but not native GBM proteins
(30). Another report showed that the monomer of isolated
Col43NC1 had much less activity in inducing glomerulonephritis than
the dimer or hexamer (22). Interestingly, the pattern of
IgG binding to GBM was also different in the animals immunized with NC1
monomer from those with dimer or hexamer (22). In one
study, recombinant NC domains of different collagen chains were used to
probe epitopes for autoantibodies from Goodpasture’s patients
(31). That study showed that activity of the autoantibody
to native Col43NC1 was 4-fold greater in comparison to rCol43NC1,
again suggesting a significant difference in B cell epitopes between
native and recombinant proteins. It is worthwhile to mention that a
35-mer peptide derived from Col43NC1, which had been mapped to be a
Goodpasture’s B cell epitope, induced only Abs to itself but not
endogenous GBM (22, 26). This observation suggests that
even a linear B cell epitope may require a certain three-dimensional
arrangement of amino acid residues. In addition to the
three-dimensional structure of GBM, glycosylation of collagen proteins
may also have a marked influence on the native B cell epitopes of GBM
(28). The recombinant protein used in this study was
highly denatured, and unglycosylated. This may lead to significant
differences in structures for B cell epitopes between rCol43NC1 and
native GBM protein. Therefore, it is not surprising that our
rCol43NC1 failed to induce anti-GBM Ab despite its correct
linear amino acid residual sequence. It is interesting to mention a
unique result from our model that lung was free of any inflammation
despite extremely severe glomerulonephritis in this model. It suggests
that lung inflammation in Goodpasture’s syndrome may be mediated by Ab
mechanism.
We have used rCol43NC1 to immunize SJ/L and BalB/c mice, which are
susceptible to glomerulonephritis (18). Although high Ab
levels to recombinant protein were induced, neither anti-GBM Ab nor
glomerulonephritis were observed (J. Wu, J. Borillo, and Y.-H. Lou,
unpublished observation). This further confirms that recombinant
protein may not be able to elicit Ab response to endogenous GBM
proteins, but also suggests that Ab could play a more important role in
mediating glomerulonephritis in mice.
If Ab to GBM is not responsible for induction of glomerulonephritis in
our rat model, what mechanism may cause the disease? Some of our
results point toward T cell-mediated mechanisms. First, in contrast to
Abs, T cells from immunized rats responded not only to rCol43NC1,
but also to isolated GBM. Second, the onset of the disease in this
model was quite rapid (14 days). In many well-known T cell-mediated
autoimmune diseases such as experimental autoimmune encephalomyelitis,
the onset of the disease (days 10–14) is coincident with a detectable
T cell response (32, 33). In fact, the Ab level is usually
very low or under a detectable limit at that time point. However, all
of our results so far have not provided direct evidence. It is
absolutely required to test whether Col43NC1-specific T cells alone
can transfer glomerulonephritis in naive recipients.
T cells have emerged as potentially the most important players in glomerulonephritis in recent years (10). Early evidence has shown that glomerulonephritis could be induced by active immunization of GBM in B cell-deficient chicks, or by transfer of mononuclear cells (34, 35). There is ample evidence for the direct participation of T cells in glomerular injury in human disease. For example, T cells or T cell-mediated inflammation have been histologically observed in both experimental and human glomerulonephritis (36, 37, 38, 39). The T cells infiltrating glomeruli were probably activated (40). Using animal models, it has been shown that blocking of the B7-CD28 costimulation pathway abrogated or reduced glomerulonephritis severity (15). A few studies have demonstrated the requirement of a general T cell population in induction of glomerulonephritis (13, 14). Very recently, a study showed that deficiency in Mgat5 N-glycosylation, which lowers T cell activation thresholds, led to the occurrence of kidney autoimmune disease (41). This suggests that T cells may be the decisive mediators for autoimmune kidney disease. However, it still remains unknown whether T cells infiltrate glomeruli through Ag-specific recognition, or are recruited passively in an Ag-nonspecific manner. An important question has been whether Ag-specific T cells alone can cause glomerular injury. Animal models probably are required to address this critical issue. We believe that our model is suitable for this purpose.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Ya-Huan Lou, Department of Basic Science, Dental Branch, Houston Health Science Center, University of Texas, Houston, TX 77030. E-mail address: ylou{at}mail.db.uth.tmc.edu
3 Abbreviations used in this paper: GBM, glomerular basement membrane; Col4, collagen 4-chain; Col43NC1, mouse collagen IV3 chain noncollagen domain 1; WKT, Wistar Kyoto; TEM, transmission electronic microscopy; LPA, lymphocyte proliferative assay.
Received for publication March 5, 2001. Accepted for publication June 20, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
3(IV) chain of the basement membrane collagen. J. Biol. Chem. 266:24018.3 chain of type IV collagen induces autoimmune Goodpasture syndrome. Proc. Natl. Acad. Sci. USA 91:6201.3, 4, and 5 chains in rodent basal laminae: sequence, distribution, association with laminins, and developmental switches. J. Cell Biol. 127:879.3 chain of type IV collagen. Kidney Int. 54:1550.[Medline]
3(IV)NC1 and 4(IV)NC1 of type IV collagen. Kidney Int. 53:664.[Medline]
chains of human type IV collagen. J. Pathol. 178:352.[Medline]
This article has been cited by other articles:
|
|
 |
|
  J. Reynolds, D. S. Abbott, J. Karegli, D. J. Evans, and C. D. Pusey Mucosal Tolerance Induced by an Immunodominant Peptide from Rat {alpha}3(IV)NC1 in Established Experimental Autoimmune Glomerulonephritis Am. J. Pathol., June 1, 2009; 174(6): 2202 - 2210. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Shah, A. Tramontano, and S. P. Makker Intramolecular Epitope Spreading in Heymann Nephritis J. Am. Soc. Nephrol., December 1, 2007; 18(12): 3060 - 3066. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Reynolds, A. Albouainain, M. A. Duda, D. J. Evans, and C. D. Pusey Strain susceptibility to active induction and passive transfer of experimental autoimmune glomerulonephritis in the rat Nephrol. Dial. Transplant., December 1, 2006; 21(12): 3398 - 3408. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Chen, T. Hellmark, V. Pedchenko, B. G. Hudson, C. D. Pusey, J. W. Fox, and W. K. Bolton A Nephritogenic Peptide Induces Intermolecular Epitope Spreading on Collagen IV in Experimental Autoimmune Glomerulonephritis J. Am. Soc. Nephrol., November 1, 2006; 17(11): 3076 - 3081. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. G. Tipping and S. R. Holdsworth T Cells in Crescentic Glomerulonephritis J. Am. Soc. Nephrol., May 1, 2006; 17(5): 1253 - 1263. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. K. Bolton, L. Chen, T. Hellmark, J. Wieslander, and J. W. Fox Epitope Spreading and Autoimmune Glomerulonephritis in Rats Induced by a T Cell Epitope of Goodpasture's Antigen J. Am. Soc. Nephrol., September 1, 2005; 16(9): 2657 - 2666. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Le Hir Histopathology of humorally mediated anti-glomerular basement membrane (GBM) glomerulonephritis in mice Nephrol. Dial. Transplant., July 1, 2004; 19(7): 1875 - 1880. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Wu, J. Arends, J. Borillo, C. Zhou, J. Merszei, J. McMahon, and Y.-H. Lou A Self T Cell Epitope Induces Autoantibody Response: Mechanism for Production of Antibodies to Diverse Glomerular Basement Membrane Antigens J. Immunol., April 1, 2004; 172(7): 4567 - 4574. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Hellmark, L. Chen, S. Ohlsson, J. Wieslander, and W. K. Bolton Point Mutations of Single Amino Acids Abolish Ability of {alpha}3 NC1 Domain to Elicit Experimental Autoimmune Glomerulonephritis in Rats J. Biol. Chem., November 21, 2003; 278(47): 46516 - 46522. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. G. Hudson, K. Tryggvason, M. Sundaramoorthy, and E. G. Neilson Alport's Syndrome, Goodpasture's Syndrome, and Type IV Collagen N. Engl. J. Med., June 19, 2003; 348(25): 2543 - 2556. [Full Text] [PDF]
|
|
|
|
 |
|
 H. HOPFER, R. MARON, U. BUTZMANN, U. HELMCHEN, H. L. WEINER, and R. KALLURI The importance of cell-mediated immunity in the course and severity of autoimmune anti-glomerular basement membrane disease in mice FASEB J, May 1, 2003; 17(8): 860 - 868. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
,
12) or CFA- (•, 6) immunized rats as detected by ELISA (sera dilution
1/4000); B, correlation between circulating
anti-rCol4
, n = 3) and rCol4
,
n = 5). T cell responses are expressed as
[3H]thymidine uptake (