Changes in the colloid and lysosomes of the thyroid of rats treated with clofibrate

in a subsequent and extensive study of rabbits concluded that the deposits, from 3 days to 18 months after administration of the gold-containing drug, have a constant composition with Au : P : S ratios of 1 : 1 : 3 (Ghadially et al., 1978). A recent extended X-ray absorption fine structure (EXAFS) study of isolated aurosomes from the rat kidney after both chronic and acute administration of gold has determined that gold is bound to sulphur in the ratio of 1 : 1.8 (Elder et al., 1983). We now report on the morphology and analysis of gold deposits in human synovial membranes removed from patients at the time of joint replacement. The patients had received therapy with Myocrisin (gold thiomalate) for rheumatoid arthritis for varying periods up to 18 months before the operation. Tissues for analysis were immediately fixed in 2% (v/v) glutaraldehyde buffered with 0.1 Mcacodylate at pH 7.3, dehydrated through graded alcohols and embedded in Epon. Thin sections of unstained material or of material stained with uranyl acetate and lead citrate were examined in a Philips 400 electron microscope at 80kV. Analysis of specimens tilted at 30" was with a Microtrace Si(Li) energy-dispersive detector and spectra were processed using Link systems 860 Computer and 'Quantem/FLS' program. Spectra of gold metal, Myocrisin (Au :S = 1) and Auranofin (Au : S : P = 1 : 1 : 1) were recorded and used as standards. These studies show that the gold deposits are not homogeneous in either morphology or composition. In the synovial membrane the gold is present in synovial-intimal cells and in macrophages. It is found in both electron-lucent and electron-dense lysosomes. Gold is present in both fine particles (c. 3 nm) on filamentous structures within the lysosomes or as larger particles (c. 30nm) which are often grouped in clusters. The region of the spectrum which has the PK, Au, and SK for the tissue deposits and for the standards is shown in Fig. 1. There was no indication that the gold in synovial-membrane cells was associated with phosphorus. The amount of sulphur associated with the gold was variable and in some deposits none was detected. Somewhat more sulphur appeared to be associated with the smaller particles than the larger. .All deposits were amorphous to electron diffraction. Gold deposits within the lysosomes of synovial-membrane cells do not, therefore, appear to possess a constant Au : S : P ratio like the deposits described in rabbits. This is of some interest as it has been suggested that gold-containing anti-rheumatic drugs may function by inactivating lysosoma1 hydrolases (Menninger & Burkhardt, 1983). Formation of gold-thiol complexes would seem a likely way in which gold could modify the structure of a protein and may explain the formation of the antibodies to gold-thiol compounds found in some human patients (Bretza et al., 1983). The results which we present in this communication show that the gold which accumulates within lysosomes is particulate and not always complexed to the most likely ligands phosphorus and sulphur, but is probably also present as metallic gold. It would seem likely that deposition in this form in lysosomes reflects the fate of inactivated drug.