The difficulties to detect intact noncovalent complexes involving proteins and peptides by MALDI-TOF mass spectrometry have hindered a widespread use of this approach. Recently, "intensity fading MS" has been presented as an alternative strategy to detect noncovalent interactions in solution, in which a reduction in the relative signal intensity of low molecular mass binding partners (i.e., protease inhibitors) can be observed when their target protein (i.e., protease) is added to the sample. Here we have performed a systematic study to explore how various experimental conditions affect the intensity fading phenomenon, as well as a comparison with the strategy based on the direct detection of intact complexes by MALDI MS. For this purpose, the study is focused on two different protease-inhibitor complexes naturally occurring in solution, together with a heterogeneous mixture of nonbinding molecules derived from a biological extract, to examine the specificity of the approach, i.e., those of carboxypeptidase A (CPA) bound to potato carboxypeptidase inhibitor (PCI) and of trypsin bound to bovine pancreatic trypsin inhibitor (BPTI). Our results show that the intensity fading phenomenon occurs when the binding assay is carried out in the sub-M range and the interacting partners are present in complex mixtures of nonbinding compounds. Thus, at these experimental conditions, the specific inhibitor-protease interaction causes a selective reduction in the relative abundance of the inhibitor. Interestingly, we could not detect any gaseous noncovalent inhibitor-protease ions at these conditions, presumably due to the lower highmass sensitivity of MCP detectors. Several major effects have been described to result in nonnative conditions for the noncovalent complexes: MALDI matrix [3][4][5], sample preparation [4,6,7], crystal morphology [4,8,9], pH of the solution [6, 10 -12], organic solvent [7,8], ionic strength [13,14], matrix/analyte ratio [15,16], speed of solvent evaporation [4], and sample concentration [9,15]. The effect of some instrumental parameters in the detection of complexes has also been studied; extraction delay time, positive/negative ion mode, linear/reflector and acceleration mode were found to be of minor importance [7,17], whereas laser pulse energy [8,17,18] and the number of laser shots (i.e., "first shot phenomenon") [4,8,13,19,20] was reported to be a decisive factor in a number of cases.Also, analyses by electrospray ionization (ESI) [21] are not carried out at physiological conditions as only solutions of very low ionic strength can be analyzed. However, ESI generates "colder" ions than MALDI and it keeps the sample in aqueous "biological" environment before ionization. For these reasons, ESI has been used in numerous studies to detect noncovalent complexes [22].Relatively few cases have been reported where specific intact noncovalent complexes were successfully observed with MALDI. Besides the frequent dissociation of noncovalent complexes due to the experimental conditions employed, a further...