Cyclic nucleotide phosphodiesterases (PDEs) are a family of isoenzymes that degrade the second messengers adenosine 3′,5′-cyclic monophosphate (cAMP) and guanosine 3′,5′-cyclic monophosphate (cGMP). Phosphodiesterase type 4 (PDE4), a cAMP-specific and principal PDE in proinflammatory cells, is a target for antiasthmatic and antiinflammatory agents. The archetypal PDE4 inhibitor rolipram has been the starting point for numerous structure-activity relationship (SAR) studies aimed at optimizing selectivity and potency against PDE4 inhibition and tumor necrosis factor receptor (TNFR)-release (a cytokine associated with arthritis, endotoxic shock, and AIDS). Despite evidence of a zinc binding domain in the PDE4 active site and rolipram being a competitive inhibitor, previous SAR studies did not consider inhibitor-metal interactions. We now report on two novel series of rolipram analogues, 3-aryl-2-isoxazoline-5-hydroxamic acids 1 and their acyclic counterparts N-aroyl amino hydroxamic acids 2. Their SARs demonstrate that the hydroxamic acid group, a well-known metal chelator, makes a unique, tight, and highly stereospecific interaction with PDE4. This effect not only results in a dramatic increase in PDE4 inhibitory activity over that of rolipram but also translates into potent inhibition of TNFR-release in purified human monocytes (HM) and human whole blood (HWB). The close structural resemblance of the non-catechol regions of these series to the ribose 3′,5′-phosphate group of cAMP (putatively bound to a metal in the active site) provides circumstantial evidence that they bind to PDE4, in part, as substrate analogues, which has implications for developing a new active site model of PDE4.