the interactions between T-cell receptors and MHC-class II molecules present on the 
surfaces of antigen-presenting cells, or inhibiting the activities of IL-1 or TNF-a, The 
present application clearly falls into the latter category. We have preferred IL-1 over 
TNF-a because it is a stronger inducer of cartilage breakdown (19) and because there 
exists a safe , natural antagonist of IL-1, the interleukin-1 receptor antagonist protein (IL- 
Ira or IRAP), which has already completed phase II clinical trials without showing any 
toxicity. The properties of IRAP are discussed in the following section. 
C. The Interleukin-1 Receptor Antagonist 
The interleukin-1 receptor antagonist protein is a naturally occuring inhibitor of 
the biological actions of both IL-la and IL-1)3. Originally identified as a product of 
macrophages, several additional cell types have been shown to produce this molecule. 
An extensive review has been published recently by Arend (21). 
Secreted human IRAP is a glycoprotein, whose core protein has a M.W. of 
approximately 17,000 Da. Addition of N-linked polysaccharides produces mature 
molecules of 22-25 kDa. IRAP binds to the type I IL-1 receptor with approximately the 
same affinity as IL-la and IL-1/3 but, because it lacks agonist activity, acts as an inhibitor 
of cellular responses to IL-1 (22). Because of the involvement of IL-1 in many 
inflammatory diseases, IRAP has considerable therapeutic potential and has shown 
efficacy in a variety of animal models of disease, including arthritis (23-25). A human 
phase II human trial of IRAP in human arthritis has just been completed. Preliminary 
data from the trial have been presented in abstract form (26) but a full length account 
has not yet appeared. In this trial, patients self-administered IRAP by daily, 
subcutaneous injection at 0.5, 1, 2, 4 or 6mg/kg. By this method, mean plasma 
concentrations of IRAP were 39 36ng/ml at 0.5mg/kg and 600 4-/- 240ng/ml at 
6mg/kg 24 hours after injection. IRAP did not accumulate with daily injection. At 
6mg/kg the mean number of tender joints decreased, erythrocyte sedimentation rates 
improved, and c-reactive protein levels dropped. These preliminary data suggest that 
IRAP has potential as an anti-arthritic protein, but is difficult to deliver as a drug. Daily 
self-injection at high doses is clearly not an acceptable option for a chronic disease such 
as RA. It is precisely for these reasons that we suggest gene delivery (24). 
D. Gene Therapy for Rheumatoid Arthritis 
Our group first suggested the use of gene therapy for RA in 1989 (27). This was 
based upon the realization that, although several novel proteins held great potential for 
use in treating arthritic joints, there was no satisfactory way to deliver them to the 
intraarticular site of disease. Because proteins cannot be taken orally, or delivered 
transdermally, and because proteins are, in general, metabolically labile, they would need 
to be injected into patients at regular intervals. Furthermore, because proteins are not 
freely diffusible into joints, they would need to be injected at very high doses. The 
diffusibility problem can be obviated by direct, intraarticular (i.a.) injection but, 
ironically, proteins are cleared from joints very rapidly via the lymph. Besides, repeated 
i.a. injection is not feasible. 
Recombinant DNA Research, Volume 19 
[429] 
