Early changes at the site of repeated subcutaneous injection of food colourings

A study has been made of the local lesions produced in rats (and in one instance in mice) by a short series of subcutaneous injections of food colourings belonging to various chemical categories. The resulting pathological changes have been separated into four main categories. In the type I response (Patent Blue V and Amaranth in rats and Blue VRS in mice) no appreciable connective tissue reaction was produced, while in the type II response (Erythrosine and Green S in rats) the connective tissue reaction developed to a limited degree. The nature of these responses suggested an inability of the test material to produce injury in the normal subcutaneous tissue or to interfere with the process of connective tissue repair whenever the normal integument became necrotic. These responses were produced by colourings which failed to produce sarcoma when injected repeatedly at the same site for 10 months or longer. Type III response (Light Green SF Yellowish, Brilliant Blue FCF and Fast Green FCF in rats) consisted mainly of an intense macrophage response involving macrophage necrosis and considerable fibroblastic proliferation. This effect closely resembles the tissue reaction reported by other workers following repeated injections of high doses of iron-dextran and of carboxymethylcellulose. Type IV response (Blue VRS and Eosine G in rats) resulted in the development of thick collagenous tissue with foci of persistent fibroblastic reaction. This lesion was found to correspond with the previously reported early stages of the fibrous reaction around subcutaneous implants of films of solid material. Colourings that have been reported by previous authors to induce subcutaneous sarcoma at the site of repeated injection produced either a type III or a type IV response. Thus a close correlation has been found to exist between the category of initial tissue response and the long-term outcome of repeated injections, in terms of the development of subcutaneous sarcoma at the site of injection. These results indicate the possibility that sarcoma production by the colourings under consideration results from derangement of connective tissue repair rather than from induction by a process of chemical carcinogenesis.