Page 114 - The Vasculitides, Volume 1: General Considerations and Systemic Vasculitis

90 Alan D. Salama and Mark A. Little

Experimental animal models of vasculitis that may be contrived or manipulated to
produce the desired phenotype do not necessarily have to be mediated through identical
pathways. MPO-ANCA induced in MPO deficient animals by an alloimmune anti-MPO
response may be very different than the autoimmune anti-MPO response.

Indeed it is now appreciated that there is significant heterogeneity in anti-MPO antibodies
found in patients with different epitope reactivity and with interference of immune reactive
sites under certain circumstances [22]. It would therefore be of interest to investigate the
differences between induced anti-MPO antibodies from different rodent models. Epitope
differences may contribute to why the model cannot be transferred with anti-MPO T-cells.
This may relate in part to the short duration of follow up in comparison to the immunized rat
model which takes four or more weeks to develop similar glomerular lesions.

In the combined anti-MPO/nephrotoxic nephritis model, the anti-MPO response is a
recall response in a previously immunized animal.

When considered together, the rodent models demonstrate that vasculitis, as manifested
by pulmonary and renal disease, may be induced when a sufficient anti-MPO immunity
threshold is achieved. Patients rarely display an exclusively T- or B-cell response, or a
uniquely innate or adaptive immune response.

Rather, the observed disease represents the sum of these different pathways. Reductionist
models can provide critical insights into individual mechanisms or pathways of disease
induction, even though they may be incapable of defining them all. In particular, none of the
models currently explain why or how immune tolerance to self-antigens is broken in the first
instance, with later destructive anti MPO reactivity.

Rodent Models of PR3-AAV

There have been no spontaneous experimental models in rodents of glomerulonephritis or
vasculitis in association with anti-PR3 antibodies, nor PR3-ANCA in association with other
autoantibodies. This is likely related to differences between humans and rodents in PR3 and
its cellular expression. Despite immunizing rats and mice with combinations of human or
murine PR3, chimeric molecules, and demonstrating induction of PR3-ANCA, van der Geld
and colleagues [23] were unable to experimentally generate vasculitis. Pfister and colleagues
[24] immunized PR3- and neutrophil elastase-deficient mice with murine PR3 and transferred
the resultant antibodies into LPS-primed mice and did not observe significant renal or
pulmonary pathology. There was a mild increase in panniculitis following intradermal TNF
injection and PR3-ANCA transfer compared to controls. Primo and colleagues [25] described
the successful transfer of PR3-ANCA producing splenocytes isolated from recombinant
murine PR3-immunized autoimmune-prone non-obese diabetic (NOD) mice into naive NOD-
severe combined immunodeficient (SCID) recipients lacking endogenous T-cells and B-cells.

The immunized NOD mice developed PR3-ANCA without apparent disease despite
prolonged follow up while the NOD-SCID mice that received PR3-reactive splenocytes
rapidly developed signs of disease with vasculitis and necrotizing crescentic
glomerulonephritis. Moreover, there were no granulomata observed despite significant
anti-PR3 immune reactivity; and C57BL/6-RAG mouse recipients of the same splenocytes
did not develop disease.

Complimentary Contributor Copy