We investigated a fluconazole-sensitive (MICflu = 5 micrograms ml-1) clinical isolate and a fluconazole-resistant (MICflu > 80 micrograms ml-1) laboratory mutant Candida albicans strain developed from the sensitive one. We studied putative virulence factors including germination, adherence ability to either buccal epithelial cells or acrylate surface, the secreted aspartic proteinase, and the extracellular phospholipase activity of the two strains as well as their growth. The fluconazole-resistant strain proved to be superior to the original strain in all the virulence traits tested. The higher virulence of the fluconazole-resistant strain was also supported by a mouse model. These results suggest that the development of fluconazole resistance can be accompanied by serious morphological and physiological changes: several putative virulence traits, moreover the in vivo virulence can increase simultaneously.

SUMMARY The absorption of macromolecules from the small intestine of rats was studied in terms of the amount of peroxidase activity that appeared in thoracic duct lymph after a 10 mg dose of horseradish peroxidase had been injected directly into the lumen of the duodenum. When the horseradish peroxidase was injected as a solution in saline no peroxidase activity was detected in the lymph. When ethyl alcohol was included in the dose at final concentrations of 12.5-16% the flow rate of the lymph increased markedly for an hour or so and during this time peroxidase activity was detected in the lymph. An electronmicroscope study of the duodenal epithelium that had been exposed to alcoholic solutions of horseradish peroxidase showed that the enzyme had penetrated between the enterocytes. It was concluded that the presence in the intestine of substantial amounts of alcohol temporarily destabilises the intercellular junctions of the epithelium and thus promotes the absorption of materials which are normally excluded.Absorption from the healthy adult gastrointestinal tract of immunologically significant amounts of macromolecular material is well documented and has been reviewed.' Recent investigations of this phenomenon have used the administration of test antigens by gavage to experimental mice2 or by normal ingestion in human volunteers.3 In such situations it is not clear at which point in the gastrointestinal tract that the absorption occurs, and the role of minor traumatic lesions in the gums, oesophagus, or stomach cannot be excluded totally. In order to lessen these problems we decided to measure the absorption of antigen in rats in terms of its appearance in thoracic duct lymph, which comes principally from the small intestine and which is an obligatory pathway for any macromolecular material which has traversed intestinal wall.5 In spite of recent claims for substantial uptakes of protein antigens by tile gut of rats6 we were unable to detect in our preliminary experiments any absorption of horseradish peroxidase after a 10 mg dose of the purified enzyme had been instilled into the lumen of the duodenum. We observed, ' On sabbatical leave from the

Summary.-Since the tumour-selective cytotoxic activity of activated macrophages in vitro can be attributed to depletion of the culture medium of L-arginine by macrophage arginase, a series of experiments was designed to determine whether such a mechanism could operate in vivo.Extracellular fluid obtained from Gullino chambers within established tumours contained high levels of arginase, no detectable arginine and high levels of ornithine. When tumours were disaggregated into single-cell suspensions, arginase was readily detected within tumour macrophages but not within malignant cells. Inflammatory ascites induced in mice by Corynebacterium parvum was rich in arginase, depleted of L-arginine and cytotoxic in vitro to L5178Y and V79 cells. High levels of arginase in the ascites fluid were associated with resistance to challenge with syngeneic L5178Y cells.Lymph collected from the cisterna chyli in rats bearing a macrophage-rich sarcoma on the small bowel contained elevated levels of arginase, was depleted of arginine and contained increased concentrations of ornithine.We conclude that in sites of macrophage infiltration there is microenvironmental arginine depletion due to the action of arginase, and that arginase release could represent an important macrophage effector mechanism against a variety of targets, including malignant cells, virus-infected cells, fungi and parasites.

Summary Rats were prepared surgically so that peripheral intestinal lymph could be collected from them while a syngenieic tumour (the HSN sarcoma) was growing in each major Peyer's patch of the small intestinie. Dendritic lymph cells were isolated from the lymph and injected i.p. into naive, syngeneic rats. Each of the 16 recipients received just under 106 such cells and was challenged 10 days later with a subcutanieous dose of 104 viable HSN cells. Six weeks after this challenge only 7 of the recipients had a tumour anid these \vere small (mean weight 1.8g), while 17 controls (which had each been treated with 106( thoracic duct lymphocytes from the same donors, and given the same challenge) all had large tumours (mean weight 8.8g). The remaining 9 test rats were still free of tumours when they were killed and autopsied 4 months after challelnge.Dendritic lymph cells from normal rats were senlsitised' by incubating them overnight on a monolayer-of HSN cells. They were then tranisferred to 5 naive recipients which received the usual challenge. Six weeks later they all had tumours (mean weight 1.3g) but these were much smaller than those in the 5 controls (mean weight 9.3 g).The occurrence in peripheral lymph of free-floating, nonlymphoid, mononuclear cells with a dendritic morphology was noted first by Morris (1968) (Hall, 1971). However, decisive experiments were hard to do in the outbred animals in which these phenomena had been demonstrated. Steinman and Cohn (1973) obtained appairently similar cells from the peripheral lymphoid organs of mice, and performed a series of experiments which showed that dendritic cells had important accessory and antigen-presenting functions in viva) and in vitro (Steinman & Nussenzweig, 1980). The role of such cells in the context of tumour immunology is uncertain.By excising the mesenteric lymph nodes from experimental animals, and then collecting intestinal lymph sometime later when the lymphatics have repaired themselves, it is possible to obtain peripheral intestinal lymph which, like peripheral lymph from any source (Smith et al., 1970), contains dendritic cells (Hall cet al., 1977). This can be done in rats bearing intestinal tumours (Moore et al., 1982;Gyure, et al.. 1985) and this type of preparation has allowed us to collect dendritic cells coming directly from the area of tumour growtll. The ability of such cells to induce anti-tumour immunity after adoptive transfer to naive recipient rats is the subject of the experiments reported below. Materials and methods GenCerail exvperimnental clesignYoung (5-6 weeks old) male hooded rats were subjected to mesenetric lymphadenectomy so that 6-8 weeks later, after they had attained adult weight (200-250 g) ol' the tumour-bearing rats were cannulated and the riats were placed in Bollman cages so that the lymph could be collected quantitatively for the next five days. Each morning the lymph from several donor rats was pooled and the cells centrifuged over a layer of Nycodenz (S.G. 1.065); the dendritic cells and macrophages rema...

No abstract