One hundred and fifty-two fungal isolates from the air spores of 503 damp, mouldy dwellings in Glasgow and Edinburgh, Scotland, were screened for cytotoxicity against the human embryonic diploid fibroblast lung cell line, MRC-5 (ATTC No. CCL171). At least 37% of the isolates, principally of the genus Pénicillium, showed toxicity to the cells when subjected to the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] assay following a water extraction procedure. When spores of 23 of the isolates were subjected to extraction with the organic solvent dimethyl sulphoxide, an additional level of cytotoxicity was observed separate from the water extraction methods. Many mycotoxins are known to be sparingly soluble in water. When some of these isolate extracts were exposed to human Chang liver cells (ATCC No. CCL13) and to human normal sternal bone marrow cells, (Detroit 98, ATCC No. CCL18), some additional cytotoxicity was observed. A comparison was also made, in some cases, between results obtained in the MTT and neutral red assays. Organisms exhibiting the greatest cytotoxicity included Pénicillium viridicatum, P. expansum, P. chrysogenum, P. commune and Aspergillus fumigatus.

Based on an analysis of the best published data, critical limits for the growth of six commonly occurring indoor moulds (defined in terms of relative humidity and temperature) have been formulated into a mould prediction computer programme. The fungi were selected as representative of moulds which differ in their relative humidity and temperature requirements to sustain surface growth, and because several were known mycotoxin producing species and of potential health significance. Each growth limit curve was generated from a series of data points on a temperature-relative humidity (RH) plot and fitted using the third-order polynomial equation RH = a-)T3 + a2T2 + ajT + ao. The model was incorporated within the Environmental Systems Performance research programme for transient simulation of the energy and environmental performance of buildings, thereby enabling the system to predict the likely occurrence of mould development for fungi which exhibit similar temperature/ RH requirements to the reference moulds. The model predicts the interactive parameters which give rise to local environmental conditions that encourage mould growth. The system’s predictive capability was tested via laboratory experiments and by comparison with monitored data from a mouldy building.