mast cells, endothelial cells, type A synoviocytes or lymphocytes present in the original 
tissue are lost and the cultures are uniformly fibroblastic. 
When 3rd passage cells are approximately 75% confluent, they are infected with 
MFG-IRAP. To do this, media are removed, and replaced by 1ml of viral suspension 
(10^ infectious particles) in the presence of SjLtg/ml Polybrene. After a 2h incubation 
with intermittent shaking, an additional 3ml of medium is added and cultures returned to 
the incubator for 72h. Cultures were then trypsinized and reseeded at a 1:5 split ration. 
Media conditioned by the transduced cells are then tested for human IRAP both 
by a commercial ELISA assay and by Western blotting. Up to 500ng/10^ cells/3 days of 
IRAP are produced by the cells. The protein generally runs as a single, glycosylated 
band corresponding to a molecular weight of approximately 25 kDa (Fig. 8). Sometimes 
a doublet is seen, with bands of approximately 22 and 25 kDa (not shown). 
Having confirmed high expression of the transgene, 10^ cells transduced are 
injected back to one or other of the transduced knee joints of the donor rabbit. 
(Repeated experiments have shown that it doesn't matter which knee is used). 
Combined histological and immunohistochemical staining of the transplanted cells 
suggest that they are captured by the synovium. Some of the cells insinuate themselves 
into the synovial interstitium, while other remain attached to the synovial surface where 
they appear to surround themselves by a collagenous capsule. Either way, the object of 
the transplantation has been achieved, with the transplanted cells remaining associated 
with the synovium and secreting IRAP (30). 
Serial lavage of the joint space with 1ml of saline followed by ELISA assay 
permits assessment of in vivo expression of IRAP. As shown in fig. 12, one week after 
transplant there are several ng/knee of IRAP, an amount which declines progressively 
with time such that by 5 weeks, the values are close to the limit of detection of the assay. 
The IRAP produced intraarticularly by this ex vivo method is biologically active 
and able to suppress the pathological changes that follow a single intraarticular injection 
of human IL-1/3. These changes include the influx of leukocytes into the joint space, 
synovial thickening and inflammation, and release of glycosaminoglycans from the 
cartilaginous matrix (28,29 - see reprints in appendix D). 
(iii) Efficacy in Antigen-Induced Arthritis in Rabbits knee 
Experiments are underway to evaluate the efficacy of ex vivo IRAP gene transfer 
in antigen-induced arthritis in the rabbit knee. In this model, animals are pre-sensitized 
to ovalbumin by intradermal injection, and then given an intraarticular injection of 
ovalbumin (100/ig) in one knee joint. A severe, erosive arthritis ensues which has an 
acute phase of approximately 1 week, superseded by a chronic phase which can last 
several weeks. We have begun by testing the ability of the IRAP gene to suppress the 
acute phase of the disease. 
Recombinant DNA Research, Volume 19 
[437] 
