The small molecule S9a was derived from an established tumor necrosis factor-alpha (TNF-alpha) inhibitor (Canventol) by replacement of the isopropylidine group with a phenyl ring. S9a at 10 to 100 nM inhibited HIV production as potently as 3'-azido-3'-deoxythymidine (AZT), an inhibitor of viral reverse transcriptase. Furthermore, S9a and AZT in combination, at noncytoxic concentrations strongly inhibited HIV-1 replication that was more than additive and substantially prolonged the appearance of virus both in acutely infected CD4+ lymphocytes (SupT) in culture and in peripheral blood mononuclear cells (PBMCs) infected with a primary HIV-1 isolate. S9a inhibited TNF-alpha promoter-driven reporter gene activity. It was proposed that the mechanism of antiviral action of S9a was on the host cell, by blocking TNF-alpha transcription via a Tat-induced tar-independent loop, which decreases downstream NF-kappaB activation of HIV-1 long terminal repeat (LTR). S9a was superior to the first generation compound Canventol, which was superior to the natural compound sarcophytol A, demonstrating that further structure-based enhancement of potency of these compounds is feasible. This study suggests a therapeutic approach against AIDS by application of two drugs, one against a cellular and the other a viral target, which may provide an approach to the problem of frequent emergence of resistant variants to combinations of drugs that target only HIV genes.