Parabens (alkyl esters of p-hydroxybenzoic acid) are used extensively as preservatives in consumer products, and intact esters have been measured in several human tissues. Concerns of a potential link between parabens and breast cancer have been raised, but mechanistic studies have centred on their oestrogenic activity and little attention has been paid to any carcinogenic properties. In the present study, we report that parabens can induce anchorage-independent growth of MCF-10A immortalized but non-transformed human breast epithelial cells, a property closely related to transformation and a predictor of tumour growth in vivo. In semi-solid methocel suspension culture, MCF-10A cells produced very few colonies and only of a small size but the addition of 5 × 10-4 M methylparaben, 10–5 M n-propylparaben or 10–5 M n-butylparaben resulted in a greater number of colonies per dish (P < 0.05 in each case) and an increased average colony size (P < 0.001 in each case). Dose-responses showed that concentrations as low as 10–6 M methylparaben, 10–7 M n-propylparaben and 10–7 M n-butylparaben could increase colony numbers (P = 0.016, P = 0.010, P = 0.008, respectively): comparison with a recent measurement of paraben concentrations in human breast tissue samples from 40 mastectomies (Barr et al., 2012) showed that 22/40 of the patients had at least one of the parabens at the site of the primary tumour at or above these concentrations. To our knowledge, this is the first study to report that parabens can induce a transformed phenotype in human breast epithelial cells in vitro, and further investigation is now justified into a potential link between parabens and breast carcinogenesis. Copyright © 2012 John Wiley & Sons, Ltd.