The experimental design is shown in figure 13, with arthritis initiated on day 1, 
the autologous cells, with or without the IRAP gene, transplanted on day 2 and the 
animals euthanized on day 4. Lavages are tested for human IRAP and GAGs, and the 
number of leukocytes counted. Cartilage matrix synthesis is measured by the covalent 
incorporation of SO 4 ' into proteoglycan, and the synovium is evaluated histologically. 
Initial results show that IRAP expression is higher in arthritic knees than non- 
arthritic knees (fig. 14), and that measures of cartilage breakdown (fig. 15) and 
suppression of cartilage synthesis (fig. 16) are normalized by IRAP. Leukocytosis is also 
suppressed by the IRAP gene but to a lesser degree (fig. 17). These data are consistent 
with the literature suggesting that IL-1 is the predominant mediator of cartilage loss, but 
only one mediator among several other, of inflammation. The observation that IRAP 
expression is higher in arthritic knees is an encouraging, unexpected result of obvious 
utility in the present context. 
(iv) Safety Studies 
Possible adverse effects of these procedures have been monitored in New Zealand 
white rabbits. Two safety aspects have received particular attention. First: Do the 
transplanted cells migrate from the joint into which they were injected? Second: Is 
there any change inhematologic parameters or blood chemistry? 
Synovia were obtained from 27 rabbits by surgical partial synovectomy carried out 
under anesthesia, using a medial approach. A retroviral vector, BAG, carrying two 
marker genes, lac Z and neo, was used to transduce the synoviocytes grown from these 
samples. Autologous neo-selected cells (10^ cells suspended in 1ml Gey's balanced salt 
solution) were injected into the left knee of each rabbit. At 1, 2 or 6 weeks post- 
transplant, 7mls of blood was withdrawn from the ear vein prior to euthanasia. 
Blood samples were analyzed for the hematological and chemical parameters 
listed in table 2. Two-tailed t-tests have been performed for all values, and fail to show 
a statistically significant difference between control and cell-injected rabbits at any time. 
After euthanasia, tissue was taken from the injected, left knee, the uninjected, 
right knee, liver, lung, heart and spleen. Samples of each tissue were treated with 
collagenase and then recovered cells grown in vitro under neo-selective conditions. Any 
neo"^ cells were stained with X-gal for expression of the lac Z gene. As shown in Table 
3, neo'^, lac Z^ cells were only recovered from the site of injection. No such cells could 
be recovered from the contralateral knee, liver, lung, heart or spleen. 
Further samples of each organ are presently being examined by PCR for the 
presence of lac Z gene sequences. 
(v) Transduction of Human Synoviocytes 
[ 438 ] 
Recombinant DNA Research, Volume 19 
