A number of epidemiological studies have reported associations of beta-carotene plasma levels or intake with decreased lung cancer risk. However, intervention studies in smokers reported increased lung tumor rates after high long-term beta-carotene supplementation. For insight into these conflicting results, we studied the influence of beta-carotene on tobacco smoke carcinogen-induced lung cancer development in the A/J-mouse using 4-(N-Methyl-N-nitro samino)-1-(3-pyridyl)-1-butanone (NNK) as the initiator and lung adenoma multiplicity as the functional endpoint. Gene regulation of the putative tumor suppressor RARbeta in mouse lung was analyzed by quantitative real-time reverse transcriptase-polymerase chain reaction (RT-PCR) for its relevance in predicting the endpoint of lung cancer. A/J-mice achieved plasma beta-carotene levels of up to 3 micromol/L within 4 wk and up to 6 micromol/L after 6 mo of supplementation on a diet modified to enhance beta-carotene absorption. Despite high lung beta-carotene concentrations of up to 6 micromol/kg, tumor multiplicity was not significantly affected by the beta-carotene treatment, either in carcinogen-initiated or non-initiated mice, and was unrelated to beta-carotene dose and the time point of treatment during cancer formation. Tumor multiplicity did not correlate with beta-carotene plasma levels in NNK-treated animals. All RARbeta isoforms were significantly suppressed in the lungs of NNK- and NNK plus high dose beta-carotene-treated animals. However, the number of tumors per mouse did not correlate with the RARbeta-isoform expression levels. beta-carotene alone after 3 mo of supplementation mildly but significantly increased levels of RARbeta1, beta2, and beta4. This increase persisted for 6 mo for RARbeta2 and beta4. In summary, we found no effect of beta-carotene on tumor formation in the NNK-initiated A/J-mouse lung cancer model with respect to dose or time point of treatment. beta-Carotene-induced changes in RARbeta isoform gene expression levels were not predictive for the number of lung tumors but were indicative of intact beta-carotene metabolism and persistent sensitivity to retinoic acid in the mice. Down-regulation of RARbeta in NNK-induced adenoma-bearing lungs was similar to that observed in human lung cancer and further confirms the A/J-mouse as a valuable model for lung carcinogenesis.