Induction of anti-Ro60/anti-La by immunisation with spectrin and induction of anti-spectrin by immunisation with Ro60 and 4-hydroxy-2-nonenal-modified Ro60 immunisation

Abstract

Objectives: The Ro ribonucleoprotein particle, targeted in systemic lupus erythematosus (SLE) and Sjögren's syndrome (SS), includes Ro60 (SSA) and La (SSA) autoantigens. Anti-Ro60 occurs in SLE and SS. The importance of α-fodrin and spectrin as well as anti-Ro and anti-fodrin/spectrin antibodies in SS and SLE, led us to hypothesise that rabbit immunisation with Ro60 or 4-hydroxy-2-nonenal-modified Ro60 would induce anti-spectrin. In addition, we hypothesised that antibodies to Ro60 and La will develop in animals immunised with spectrin.

Methods: Two NZW rabbits each were immunised with 4-hydroxy-2-nonenal-modified Ro60 or unmodified Ro60. Methods used included ELISA, including an inside-out RBC membrane ELISA, and Crithidia lucilae assays.

Results: Commercial anti-spectrin sera bound significantly to Ro60 (OD 2.6 ± 0.1), Ro60 multiple antigenic peptides (MAPs) (3 out of 21 Ro60 MAPs), La (OD 4.4±0.5), and La fragments as well as to double stranded DNA but not to BSA (OD 0.6±0.1). Anti-spectrin binding to purified spectrin could be inhibited by spectrin (>95%), and Ro60 or La (70%). When the binding of anti-spectrin was tested against a nested set of La fragments we found that a N4 fragment representing the C-terminal 250 aa (aa 159 to 408) bound the strongest (OD=4.12) followed by a N9 fragment (the C-terminal 36aa; aa373 to 408 (OD=1.36). Also, significant anti-spectrin antibody levels were induced by Ro60 and HNE-modified Ro60 immunisation.

Conclusions: We found intermolecular epitope spreading from Ro60/La to spectrin and vice versa, and this may have pathological significance in these animal models of autoimmunity.

Figure 1

Serial bleeds of rabbits immunized with Ro60 or 4-hydroxy nonenal (HNE)-modified Ro 60 analyzed for binding to purified spectrin. Spectrin was coated on ELISA plates and binding of the rabbit anti-sera to spectrin was determined by ELISA. Anti-spectrin was used as the positive control. Arrow indicates time of boost given to the animals.

Figure 2

Serial bleeds of rabbits immunized with Ro 60 or HNE-modified Ro 60 analyzed for binding to inside-out cell membrane immobilized to polylysine coated ELISA plates. Inside-out cell membrane was prepared as mentioned in Materials and methods. The plates were blocked and anti-Ro60 or anti-HNE Ro60 rabbit sera were added (1: 100 dilution). After addition of appropriate alkaline phosphatase conjugate, paranitrophenol phosphate substrate was added and the color developed was read.

Figure 3

Anti-spectrin binding to purified recombinant human La and to bovine serum albumin by ELISA. Ro 60, La or BSA were coated on ELISA plates and binding by anti-spectrin to these antigens was determined.

Figure 4

Inhibition of anti-spectrin binding to spectrin in the absence and presence of spectrin, Ro or La. The antigens were incubated at room temperature for one hour with anti-spectrin (1:500 dilution) and binding of anti-spectrin or anti-spectrin + antigen to spectrin was investigated.

Figure 5

Intermolecular epitope spreading of anti-spectrin to Ro MAPs. MAPs constructed from the 60 kD Ro sequence were coated on ELISA plates, blocked and incubated with anti-spectrin rabbit antisera. A MAP from the Sm autoantigen, PPPGMRPP was used as control.

Figure 6

Anti-spectrin binding to fragments of La showing intermolecular epitope spreading. BSA, La and fragments of La were coated on ELISA plates and the binding of anti-spectrin to these antigens was determined by ELISA. N4 corresponds to the C-terminal 256 amino acids (aa 159 to 408). N9 corresponds to the C-terminal 46 amino acids (aa 373 to 408).

Figure 7

Antibodies to dsDNA detected by CLIF assay in rabbits immunized with spectrin, hemoglobin, or Freund’s control. A: Hemoglobin immunized; B: Freund’s immunized. C: Spectrin immunized; D: Spectrin immunized.