| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Center for Rheumatology/Bloomsbury Rheumatology Unit, Department of Medicine, University College London, London, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The immunopathogenesis of SLE, including the origin of the wide spectrum of autoantibodies that are found in patients, remains elusive (6). However, studies have revealed that autoantibodies to different components of the same nuclear particle often occur in linked sets and in an ordered hierarchy of expansion with time in the sera of individual patients with SLE (7, 8). Epitope spreading is a possible mechanism by which the generation of multiple autoantibodies could arise, the theory being that once immunological tolerance to one component of a particle is broken, the immune response can diversify, allowing the recognition of new epitopes within the complex (9, 10). Experimental proof that this phenomenon occurs in lupus is limited, although there is recent evidence from our laboratory4 and others (11, 12) that epitope spreading may be occurring in patients with SLE.
Recent experiments by James et al. (13, 14) showed that epitope spreading occurred in normal NZW rabbits and certain inbred mouse strains (markedly A/J and AKR/J) that had been immunized with a MAP peptide derived from Sm B/B', PPPGMRPP-MAP. This work was particularly notable because the epitope spreading within the structurally associated snRNP proteins, Sm D, and U1 snRNP polypeptides 70K, A and C, subsequently extended to include production of Abs to dsDNA and was also associated with the development of symptoms reminiscent of SLE. The proline-rich sequence PPPGMRPP, which occurs three times toward the carboxyl-terminal end of Sm B/B', has been demonstrated to be the predominant antigenic epitope in anti-Sm-positive patient sera (15).
To our knowledge, there is no published study to date independently analyzing this model of lupus autoimmunity. Our group has attempted to generate the model of lupus described by James et al. (13, 14). Our findings confirm the ability of PPPGMRPP-MAP to trigger spreading of the autoimmune response in NZW rabbits, but contrary to the findings of James et al. (13, 14), our results do not support the generation of lupus-like autoimmunity as a result of immunization with PPPGMRPP-MAP.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
NZW female rabbits, weighing approximately 2.5 kg, were supplied by Froxfield Farms (Petersfield, U.K.). Accurate identification of the individual NZW rabbits was ensured throughout the experimental procedures by the s.c. injection of microchip code numbers, which could be read with a hand-held scanner. Female A/J mice (H-2a, 5–6 wk old) and C57BL/10/ScSn mice (H-2b, 3–4 wk old) were obtained from Harlan U.K. (Bicester, U.K.). Mice were identified by tail marking or ear punching. All the animals were supplied barrier reared and were acclimatized for 2 wk in the animal facility before immunization. All procedures were conducted under license from the Home Office in accordance with the Animals (Scientific Procedures) Act 1986.
Immunogens
The peptide used for immunization was in a multiple Ag peptide (MAP) format, in this case consisting of eight copies of a synthetic peptide attached to a immunogenically inert core matrix of lysine residues (MAP, Applied Biosystems, Foster City, CA). The MAP format peptide used in this study was PPPGMRPP-MAP, a major antigenic region of Sm B/B' that occurs three times at positions 191–198, 216–223, and 231–238. (PPPGMRPP-MAP was supplied by J. A. James and J. B. Harley, Oklahoma Medical Research Foundation, Oklahoma City, OK). Control groups were included that received the MAP lysine backbone with adjuvant or sterile saline with adjuvant and also a naive unimmunized group of animals.
Immunizations
The protocol for immunization adhered closely to that used by James et al. (13); we used the same peptide doses, adjuvants, and injection routes. The rabbits were immunized on day 0 with peptide diluted in sterile saline emulsified 1/1 (v/v) with CFA (Difco, Detroit, MI) and were boosted on days 26, 53, and 99 with peptide emulsified in IFA. At each immunization the rabbits received 500 µg of peptide; half was injected s.c., and half was injected i.p. A final boost containing 500 µg of peptide in saline without adjuvant was given i.v. on day 152. Immunization groups consisted of six rabbits immunized with PPPGMRPP-MAP, three rabbits received MAP-lysine backbone and three rabbits were given saline and adjuvant alone.
The A/J and C57BL/10/ScSn mice were divided into the following immunization groups: PPPGMRPP-MAP (n = 10), MAP-lysine backbone (n = 7), saline (n = 7), and naive unimmunized mice (n = 7). All mice were immunized with 100 µg of peptide emulsified with CFA on day 0, half administered s.c. and half given i.p., and were boosted with 100 µg of peptide, half administered s.c. and half given i.p., in IFA on days 14, 28, and 64 (A/J) and on days 14, 28, and 63 (C57BL/10/ScSn).
Clinical assessment
Throughout the experiment, the animals were under constant observation for the development of any clinical disease symptoms. Before the initial immunization, following each boost, and at frequent intervals, the animals were weighed, urine was taken to assess proteinuria using Albustix (Bayer Diagnostics, Basingstoke, U.K.), animals were bled, and serum was stored for analysis.
Anti-peptide Ig ELISAs
Standard solid phase assays were used to assess the IgG Ab
response to the immunizing peptide. Polystyrene 96-well plates
(Maxisorp, Nunc, Roskilde, Denmark) were half-coated with either
PPPGMRPP-MAP or MAP lysine backbone at 8 µg/ml in bicarbonate buffer,
pH 9.6; the other half of each plate was coated with buffer alone.
After incubation overnight at 4°C, plates were washed three times
with PBS (pH 7.2) and blocked with 1% (w/v) BSA (Sigma, Poole, U.K.)
in PBS. Serum samples, at varying dilutions in PBS/1% BSA/0.05%
Tween-20 (Sigma), were applied to coated and uncoated wells and
incubated for 1 h at 37°C. Following three washes, bound Abs
were detected by incubation for 1 h at 37°C with a 1/10,000
dilution of affinity-purified goat anti-mouse or anti-rabbit
-chain-specific IgG conjugated to alkaline phosphatase conjugate
(Sigma) and, following washing, developed with the substrate
p-nitrophenol phosphate (Sigma). OD was measured at 410 nm.
Autoantibody ELISAs
Autoantibodies to Sm, Sm/RNP, and Ro (60 and 52 kDa) were measured using standard solid-phase ELISAs. Ag-precoated wells were obtained from Shield Diagnostics (Dundee, U.K.) along with uncoated control wells, sample diluent, and borate washing buffer. Bound Abs were detected as described for the anti-peptide IgG ELISA.
Autoantibodies to denatured ssDNA and to native dsDNA were detected using a solid phase ELISA. Pure DNA was prepared from salmon sperm DNA (Calbiochem-Novabiochem, Nottingham, U.K.) by phenol/chloroform extraction, and ssDNA was prepared by boiling for 10 min and cooling on ice for 15 min. Microtiter plates (Nunc) were coated for 1 h at 37°C with 50 µg/ml poly-L-lysine (Sigma) in distilled water. After washing, plates were divided into three parts coated with 5 µg/ml dsDNA or 2.5 µg/ml ssDNA or control wells with distilled water for 1 h at 37°C. Plates were washed twice and blocked with 2% casein (BDH, Poole, U.K.) in PBS for 1 h at 37°C. Sera at varying dilutions in PBS/Tween-20 were incubated for 1 h at 37°C, and bound Abs were detected as described above.
Western blotting
Autoantibodies against nuclear Ags were detected using an
ANA-Bioblot kit (Biocode Biotechnology, Sclessin, Belgium). Nuclear Ags
extracted from human HeLa cells were separated by SDS-PAGE and blotted
onto nitrocellulose together with precolored m.w. markers facilitating
accurate interpretation of the results. These ready-to-use strips were
incubated for 15 min with a 1/100 dilution of rabbit or mouse serum in
Tris-buffered saline containing blocking proteins and Tween-20. Bound
Abs were detected using affinity-purified goat anti-rabbit or
anti-mouse -chain-specific IgG conjugated to alkaline
phosphatase diluted 1/1000 and then developed with the substrates,
5-bromo-1-chloro-3-indolyl phosphate and nitroblue tetrazolium
(Promega, Madison, WI). Results were assessed with reference to the
control strips provided, showing the positions of Sm, RNP, Ro, La,
Jo-1, Scl-70, and Ribosomal Po Ag bands.
Indirect immunofluorescence for detection of ANA
Serum samples were screened for ANA using standard indirect immunofluorescence techniques. Prefixed Hep-2 cells (INOVA Diagnostics, San Diego, CA) and rat liver and kidney cryosections were used as substrates. After incubation with varying dilutions of serum, ANA binding was detected using goat anti-rabbit IgG FITC (Sigma) or goat anti-mouse IgG FITC (Harlan Sera-lab, Loughborough, U.K.).
Solid phase peptide synthesis and pepscan assay
Pepscan assays were used to investigate the extent of any epitope spreading of the immune response within Sm B/B'. The 233 octamer peptides, overlapping by seven amino acids, encompassing the entire length of Sm B/B' were simultaneously synthesized from F-moc-protected amino acids (Calbiochem-Novabiochem) using the solid phase multipin peptide synthesis system (16) developed by Chiron Mimotopes (Victoria, Australia).
Incubation steps of the pepscan assays were performed by lowering the
solid-phase peptides, bound to the pins, into solutions contained in
microtiter plates, and washing steps were conducted by agitation in
PBS/0.05% Tween-20 in plastic boxes. Pins were incubated for 1 h
at 37°C in blocking buffer consisting of 1% OVA (Sigma), 1% BSA,
1% casein, and 0.1% Tween-20 in PBS. Sera was diluted to 1/500 in
blocking buffer and incubated with pins overnight at 4°C. After
washing in three changes of PBS, pins were immersed for 1 h at
37°C in affinity-purified goat anti-mouse or anti-rabbit
-chain-specific IgG conjugated to alkaline phosphatase diluted
1/10,000 or 1/2,000, respectively, in blocking buffer and developed
with the substrate p-nitrophenol phosphate. OD was measured
at 410 nm. Following the assay, peptides were regenerated by sonication
of the pins for 15 min in disruption buffer (0.13 M sodium dihydrogen
phosphate with 1% SDS (BDH) and 0.1% 2-ME (Sigma), pH 7.2) at 60°C
to ensure removal of the reaction products. Pins were then washed twice
for 2 min in distilled water at 60°C and boiled for 3 min in methanol
before being air-dried and stored at 4°C with silica desiccant.
Successful regeneration of the unbound peptides was verified throughout
these experiments by performing the assay with the omission of the
serum incubation stage.
Histology
Formalin-fixed, paraffin-embedded, kidney sections from the rabbits and mice were stained with hematoxylin and eosin. These were then examined by a histopathologist for morphological evidence of kidney disease. Acetone fixed, frozen sections of the rabbits’ kidneys were stained for Ig deposition using horseradish peroxidase-conjugated monoclonal anti-rabbit Igs (clone RG16, Sigma), and bound Abs were detected with 3,3'-diaminobenzidine. Similarly, paraffin wax kidney sections from the A/J mice were stained with rabbit anti-mouse Ig-horseradish peroxidase (Dako, Denmark) and developed with 3,3'-diaminobenzidine.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Both the PPPGMRPP-MAP-immunized NZW rabbits and A/J mice
demonstrated high titer anti-PPPGMRPP-MAP IgG Ab responses,
1/25,000 and 1/10,000, respectively. The anti-PPPGMRPP-MAP response
was seen after the primary immunization and persisted for at least 327
days in the rabbits (Fig. 1
A)
and 217 days in the mice (Fig. 1B). Sera from the control
groups of animals showed no anti-PPPGMRPP-MAP response. No Ab
response to the MAP lysine backbone was observed in sera from the NZW
rabbits and A/J mice. The C57BL/10 mice were nonresponders to
PPPGMRPP-MAP in our hands, having no anti-PPPGMRPP-MAP IgG in their
sera following the full immunization schedule. These mice were
therefore sacrificed on day 125.
|
The A/J mice from all immunization groups had no detectable
anti-Sm IgG Abs when tested by ELISA on days 13, 27, 41, 63, 84,
106, 273, and 429. The A/J sera were also tested up until day 429 for
anti-Sm/RNP, anti-Ro, anti-ssDNA, and anti-dsDNA IgG
Abs by ELISA and were negative for all these autoantibodies. The NZW
rabbit sera were negative in all immunization groups when tested by
ELISA for anti-Sm and anti-Sm/RNP on days 26, 76, 327, and 365
and were also negative for anti-Ro on days 327 and 365 and for
anti-ssDNA and anti-dsDNA IgG Abs on days 26, 76, 327, and 365.
Sera from MRL/Mp-lpr/lpr mice, a spontaneous model of SLE,
were used as a positive control in these assays, with the expected
20–30% of animals giving high titers of anti-Sm and virtually all
having anti-DNA Abs (data not shown). However, over time, the
development of ANAs in the sera from three of the rabbits immunized
with PPPGMRPP-MAP was revealed by Western blotting of nuclear Ags
extracted from HeLa cells (Fig. 2). After
immunization with PPPGMRPP-MAP, serum from rabbit 1 showed a faint Sm
B/B' band by Western blotting on day 26; this band was stronger by day
76 and still persisted on day 230. We also observed in sera from rabbit
1 faint bands of binding to other extractable nuclear Ags (ENA). Serum
from rabbit 6, also immunized with PPPGMRPP-MAP, showed no binding to
Sm B/B' in the Western blot. However, bands suggesting Ab binding to
RNP C and ribosomal Po were observed for rabbit 6 on day 26. By day 76
serum from rabbit 6 gave strong bands of binding to Sm B/B', which
still persisted on day 230. Sera from a third rabbit (rabbit 5)
immunized with PPPGMRPP-MAP did not bind to Sm B/B' at any time point
tested, but the bands suggested Ab binding to RNP A on days 26 and 76
and to Ro (SS-A) on day 230. The other three rabbits immunized with
PPPGMRPP-MAP (rabbits 2, 3, and 4) showed no binding to Sm B/B'. It is
possible that Abs might arise spontaneously during the lifetime of a
rabbit or as a result of the immunization process, which could
cross-react with nuclear Ags in the Western blot. We tested control
sera from MAP backbone and saline-immunized rabbits to exclude the
possibility that bands seen on the Western blots arose from such
cross-reacting Abs. No ANA binding was observed by Western blotting of
sera from these control rabbits. The A/J mice immunized with
PPPGMRPP-MAP showed no binding to Sm B/B' or the other ENAs by Western
blotting (data shown for day 84; Fig. 2). Furthermore, these ANAs were
undetectable when the sera were incubated at dilutions ranging from
1/10 to 1/80 on slides containing Hep2 cells or rat liver and kidney
sections.
|
Pepscan assays were used to investigate B cell epitope
spreading of the immune response away from the original epitope,
PPPGMRPP. The 233 octamer peptides, overlapping by seven amino acid
residues, of Sm B/B' were synthesized according to the sequence of 240
amino acids published by Rokeach et al. (17). Each peptide is referred
to by the position of the first amino acid residue of the octamer in
the sequence of Sm B/B'. A pool of sera from three unimmunized
MRL/Mp-lpr/lpr mice was used as a positive control. As
expected, autoantibodies in this serum pool bound to many of the
peptides along the whole sequence of Sm B/B' (Fig. 3A). The highest peaks of
binding were found at the sites of the octapeptides corresponding to
and adjacent to PPPGMRPP itself. Sera taken from nine A/J mice,
immunized with PPPGMRPP-MAP, at four time points (days 27, 84, 161, and
300) during the experiment showed strong peptide binding to PPPGMRPP
and adjacent epitopes, but no spreading to other epitopes. The pepscans
obtained using a pool of sera from four of the A/J mice immunized with
PPPGMRPP-MAP are shown for days 27, 84, and 300 (Fig. 3, B–D). Sera from the A/J mice immunized with MAP lysine
backbone, saline alone, and the unimmunized mice showed no reactivity
with the octapeptides of Sm B/B'. Pepscans that omitted the serum
incubation step and where the peptide pins were immersed in the
anti-mouse -chain-specific IgG conjugate and then developed with
the substrate showed no reactivity.
|
,
A–D). Another rabbit (rabbit 5) showed moderate epitope
spreading for approximately 200 days (Fig. 4, E and
F), and the remaining three rabbits (rabbits 2–4) showed
very limited epitope spreading (representative rabbit 4 is shown in
Fig. 4G). Sera from the control rabbits immunized with MAP
lysine backbone or saline alone showed negligible reactivity with the
octapeptides of Sm B/B' (a representative rabbit immunized with lysine
MAP backbone is shown in Fig. 4H), as did the sera from a
rabbit bled before immunization with PPPGMRPP-MAP. The peptide pins
incubated with anti-rabbit -chain-specific IgG conjugate and
substrate showed no reactivity.
|
Clinical observations
Despite extensive monitoring, no evidence of clinical disease, including proteinuria, alopecia, or loss of weight was observed in any of the normal animals. The formalin-fixed, paraffin-embedded kidney sections stained with hematoxylin and eosin showed no evidence of disease in the A/J mice. Of the six rabbits immunized with PPPGMRPP-MAP, only rabbit 1 showed evidence of any histological abnormality. This presented as a mild focal interstitial lymphocytic nephritis, which in the absence of proteinuria in this rabbit was unlikely to represent a pathological result of the immunization. Rabbits 5 and 6, which also showed epitope spreading, showed no such histological abnormalities. Examination of the slides stained by immunohistochemistry revealed no evidence of Ig deposition in any of the NZW rabbits or A/J mice immunized with PPPGMRPP-MAP.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
It has been suggested that the phenomenon of epitope spreading may be involved in the development of the autoimmune response in lupus. Epitope spreading can occur at both the B and T cell levels, as demonstrated in models of other autoimmune diseases, for example experimental autoimmune encephalomyelitis (18, 19) and the insulin-dependent NOD mouse (20). The expansion of the immune response is thought to be facilitated by the close proximity or association of Ags (21). In SLE this could explain the occurrence of linked sets of autoantibodies (7, 8). Murine experiments demonstrate that immunization with snRNP A protein also results in the production of Abs to Sm (22), and similarly, immunization with La Ag leads to the generation of anti-Ro Abs (23). The presence of whole intact Ag has been shown to be important in the ordered expression of the autoantibody response (24). In a recent model James et al. (13, 14) suggest that an epitope, PPPGMRPP, derived from Sm B/B' Ag may initiate the autoimmune response in normal animals that ultimately leads to the expression of clinical features of lupus.
We have investigated this model and present our findings herein. We
immunized genetically identical NZW rabbits and A/J and C57BL/10 mice
with PPPGMRPP-MAP peptide obtained from James and colleagues. We have
shown evidence of epitope spreading in three of the six NZW rabbits
that we immunized with PPPGMRPP-MAP, confirming in part the
data reported by James et al. (13). Intriguingly, although we observed
an Ab response to the immunogen, PPPGMRPP, from the beginning of the
experiment (as shown by ELISA in Fig. 1
A), detailed
examination of the anti-peptide response in the pepscans (on day
26; data not shown) revealed that in two of the three rabbits (rabbits
1 and 6) in which epitope spreading was observed, the highest responses
were initially to octamers other than PPPGMRPP. Rabbit 5 was the
exception, with predominant epitopes at positions 191, 215, (not 216),
217, and 231 by day 26. By day 76 after the primary immunization,
autoantibodies from the sera of all six of these rabbits did show
strong binding in the pepscan assays to PPPGMRPP and its adjacent
epitopes.
Not unexpectedly, the observed epitope spreading was often directed toward other proline-rich epitopes in the Sm B/B' sequence, such as 183–186 (TPPPGIMA-PGIMAPPP) and 222–223 (PPPPGIRG-PPPGIRGP), which both occurred in four of the six rabbits immunized with PPPGMRPP-MAP. PPPGIRGP (223) was indeed one of the MAP peptides used by James et al. in their rabbit model (13), as it has been shown to be another early epitope mapped in anti-Sm/RNP-positive patients with SLE (15). The two rabbits that showed the greatest and most persistent epitope spreading, rabbits 1 and 6, did exhibit other epitopes that had fewer amino acids in common with the immunizing peptide. These included epitopes 28 (IGTFKAFD), 29 (GTFKAFDK), 42 (LCDCDEFR), 75 (ENLVSMTV), and 77 (LVSMTVEG). Some of these epitopes, 28/29 and 42, have been shown to be antigenic regions in anti-Sm and anti-nRNP-positive patients (15). Sera from our MRL/Mp-lpr/lpr mice also gave a peak in the region of epitope 42, and the data presented by James et al. (14) show that their A/J mice exhibit a peak in this region only 28 days after immunization with PPPGMRPP-MAP. Peaks in the region of epitopes 75–77 also appear in pepscans from our MRL/Mp-lpr/lpr mice and in pepscan data from a NZW rabbit after immunization with PPPGMRPP-MAP reported by James et al. (13). James et al. have shown an epitope in region 168–170 (YPPGRGTP-PGRGTPPP) to be important in patients (15), rabbits (13), and A/J mice (14). We found no peaks corresponding to this region in pepscans of our NZW rabbits. However, sera from rabbit 5 did bind strongly to a proline-rich epitope at position 174 (TPPPPVGR). Many of the epitopes bound by our NZW rabbits contain proline-rich sequences, but others often feature arginine (R) and lysine (K). This adds weight to the hypothesis put forward by James et al. (15) that the presence of such positively charged amino acids seems to be an important factor determining the antigenicity of a peptide sequence. Three of our six NZW rabbits immunized with PPPGMRPP-MAP have shown epitope spreading within Sm B/B'. However, although the Western blotting data for these three rabbits suggests epitope spreading to Ags beyond Sm B/B', we have been unable to confirm this by ELISA or immunofluorescence techniques. In addition, we have found no evidence of diversification of this response to include Abs to dsDNA or the development of clinical disease in any of our rabbits.
The data published by James et al. (14) indicate that of all the strains of mice they tested, the A/J strain exhibited the greatest degree of epitope spreading to other regions of Sm B/B' and Sm D. This spreading was already marked only 28 days after immunization with PPPGMRPP-MAP. Sera from their mice also showed high autoantibody binding to Sm and U1 snRNP, and 80% of the mice had Abs to dsDNA. Some of these mice also showed clinical symptoms reminiscent of lupus (14). Intriguingly, despite sustaining high titers of anti-PPPGMRPP Abs we saw no epitope spreading within Sm B/B' and no autoantibody response to Sm, U1 snRNP, or dsDNA. We also observed no evidence of clinical disease in any of the A/J mice we immunized. The C57BL/10 mice that we immunized with PPPGMRPP-MAP did not show any anti-PPPGMRPP IgG response by ELISA, whereas James et al. (14) found that C57BL/10 mice gave a high titer antipeptide response, although this strain did not subsequently exhibit epitope spreading or snRNP binding.
The discrepancies between our findings and those of James et al. (13, 14) could be due to several factors, including batch-to-batch variation between different preparations of PPPGMRPP-MAP peptide, environmental factors, and any differences in immunization procedures. We suspect that the exact reproducibility of the MAP peptide synthesis, even within the same facility, may be the most important of these factors. It is possible that any structural variation in the Ag may affect the presentation of Ag to the immune system. The peptides we used were obtained from the same biochemical facility to try and avoid this source of variation; however, it was not possible to use exactly the same batch of PPPGMRPP-MAP as that employed by James et al. (13, 14) due to the limited quantity of peptide produced at each synthesis. The environmental conditions under which the animals were housed appear to have been comparable, although we cannot fully exclude the possibility of an unknown infectious agent having affected the development of autoimmunity (25). The immunization protocol reported in the original paper by James et al. (13) was adhered to very closely; we used the same peptide doses, adjuvants, and injection routes. We allowed more than adequate time for the development of autoimmunity, since we followed the NZW rabbits and A/J mice for 402 and 429 days, respectively, after immunization with PPPGMRPP-MAP.
There are many examples of animal models of autoimmunity that have proven difficult to reproduce; some of these are discussed by Isenberg et al. (26) when describing their attempt to reproduce the 16/6 Id lupus model (27). Leiter et al. (28) report variability in the frequency of development of diabetes in NOD mice, probably due to variable diet or microbial exposure. Goverman et al. (29) report how spontaneous EAE can develop in transgenic mice housed in a nonsterile facility, but not in those maintained in a specific pathogen-free facility. Mouse hepatitis virus, in particular, has a long history of interference with the results of animal experiments, especially those involving the immune system (25). It is interesting to note that although our rabbits were all female and housed in the same environment, only half of them exhibited significant epitope spreading, and this occurred to varying degrees and to different epitopes in the individual rabbits. Also, of a total of nine rabbits (five males and four females) immunized by James et al. (13) with PPPGMRPP-MAP using the same immunization schedule that we employed, only one developed Abs to dsDNA and clinical symptoms reminiscent of SLE. However, sera from all nine of these rabbits did bind to octapeptides of spliceosomal proteins beyond Sm B/B'. Perhaps if we had immunized larger numbers of rabbits, some of these may have developed further autoimmunity. It is notable in MRL/Mp-lpr/lpr mice, a spontaneous model of SLE, only approximately 25% of these mice develop anti-Sm Abs despite being a genetically inbred strain (30).
Although James et al. (13, 14) have successfully used the MAP system to generate autoimmunity in normal animals, there are several authors who report the limited value of MAPs for raising anti-peptide Abs that are cross-reactive with the cognate protein. These limitations are reviewed by Briand et al. (31). These authors reported that this lack of cross-reactivity occurred especially when the MAP was constructed from C-terminal peptides. PPPGMRPP is found at three positions within Sm B/B', all of these are toward the C-terminal end, with the 231 position being at the very terminus. McLean et al. (32) report that none of their animals immunized with different MAPs produced sera that recognized the parent protein in Western blots. It is thought likely that the peptides in the MAP construction assume an unusual conformation that does not mimic the structure of either the monomeric peptide or the parent protein. This may explain why we were able to demonstrate ANAs by Western blotting but not by ELISA in the sera from our PPPGMRPP-immunized rabbits.
In a situation where comparable numbers and as identical as possible conditions have been employed to reproduce a model of autoimmune disease but with differing results, it is important to isolate and define those factors that might allow the expression of autoimmunity. The immunological mechanism should have been the same in these two experimental sets; however, it is possible that a slight alteration to the level of signaling or in the presentation of the Ag resulted in autoimmune disease in one set of animals but not in the other. Dissection of any subtle environmental or genetic variables might provide clues to the pathogenesis of autoimmunity in lupus.
In conclusion, we observed limited epitope spreading in some NZW rabbits but not in any A/J mice, as a result of immunization with PPPGMRPP-MAP. Immunization with this Sm B/B' peptide did not lead to the development of autoantibodies to dsDNA or to the development of autoimmune disease in our normal animals. We were thus unable to confirm the generation of autoimmune disease in normal animals following immunization with PPPGMRPP-MAP, as reported by James et al. (13, 14). This once again underlines the difficulties often experienced in reproducing animal models in different laboratories.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. J. K. Kalsi, Center for Rheumatology/Bloomsbury Rheumatology Unit, Arthur Stanley House, 40-50 Tottenham St., London, United Kingdom W1P 9PG.
3 Abbreviations used in this paper: SLE, systemic lupus erythematosus; ANA, anti-nuclear autoantibody; snRNP, small nuclear ribonucleoprotein; NZW, New Zealand White; MAP, multiple Ag peptide; ENA, extractable nuclear Ag.
4 J. K. Kalsi, W. Ng, S. Muller, and D. A. Isenberg. Expansion of the autoantibody response in patients with systemic lupus erythematosus. Submitted for publication.
Received for publication October 6, 1998. Accepted for publication February 5, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  M Ishii, Y Muramoto, H Kosaka, S Ohshima, T Mima, Y Katada, S Hirohata, and Y Saeki A serological switching from anti-dsDNA to anti-Sm antibodies coincided with severe clinical manifestations of systemic lupus erythematosus (hemophagocytosis, profundus and psychosis) Lupus, January 1, 2007; 16(1): 67 - 69. [PDF]
|
|
|
|
 |
|
 G. Rai, S. Ray, R. E. Shaw, P. F. DeGrange, R. G. Mage, and B. A. Newman Models of Systemic Lupus Erythematosus: Development of Autoimmunity Following Peptide Immunizations of Noninbred Pedigreed Rabbits J. Immunol., January 1, 2006; 176(1): 660 - 667. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Pal, U. S. Deshmukh, Y. Ohyama, Q. Fang, C. C. Kannapell, F. Gaskin, and S. M. Fu Evidence for Multiple Shared Antigenic Determinants within Ro60 and Other Lupus-Related Ribonucleoprotein Autoantigens in Human Autoimmune Responses J. Immunol., December 1, 2005; 175(11): 7669 - 7677. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J K Kalsi, J Grossman, J Kim, P Sieling, D W Gjertson, E F Reed, F M Ebling, M Linker-Israeli, and B H Hahn Peptides from antibodies to DNA elicit cytokine release from peripheral blood mononuclear cells of patients with systemic lupus erythematosus: relation of cytokine pattern to disease duration Lupus, July 1, 2004; 13(7): 490 - 500. [Abstract] [PDF]
|
|
|
|
|
|
B H Hahn Lessons in lupus: the mighty mouse Lupus, September 1, 2001; 10(9): 589 - 593. [PDF]
|
|
|
|
|
|
M Lidar, A Braf, N Givol, P Langevitz, R Pauzner, A Many, and A Livneh Anti-insulin antibodies and the natural autoimmune response in systemic lupus erythematosus Lupus, February 1, 2001; 10(2): 81 - 86. [Abstract] [PDF]
|
|
|
|
|
|
M Chang, S E Walker, and R W Hoffman Immunization with a bacterial ATP-binding cassette transporter fragment suppresses autoimmunity and prolongs survival in MRL/lpr lupus-prone mice Lupus, November 1, 2000; 9(9): 655 - 663. [Abstract] [PDF]
|
|
|
|
 |
|
 M. L. Stoll and J. Gavalchin Systemic lupus erythematosus--messages from experimental models Rheumatology, January 1, 2000; 39(1): 18 - 27. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
). OD values were measured at 410 nm, with
a reference filter of 490 nm.
). A, A 1/500 dilution of sera from a pool of three
naive unimmunized (aged 18 wk) MRL/Mp-lpr/lpr mice;
B–D, a 1/500 dilution of sera from a pool of 4 A/J mice
immunized with PPPGMRPP-MAP on day 27 (B), day 84
(C), and day 300 (D). OD values were
measured at 410 nm, with a reference filter of 490 nm.