Direct identification of bacteria in urine samples by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry and relevance of defensins as interfering factors.

Standard methods for the identification of uropathogens that are based on the determination of metabolic activity require cultivation on agar plates, which often takes more than 1 day. If microbial growth on agar plates is slow, or if metabolic activity is impaired by adverse interactions resulting from the patient's condition or from medical treatment, the application of standard methods may lead to delayed or erroneous identification of bacteria. In recent studies, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) has proven to be able to rapidly identify bacteria obtained from cultures. We tested the applicability of this analytical technique for the rapid identification of bacteria collected directly from urine samples and compared the results with those of conventional identification methods, such as the Vitek system, the MicroScan WalkAway system and the API system, and in some cases with the gas chromatographic determination of the bacterial long-chain fatty acid pattern. We analysed a total of 107 urine samples with bacterial counts ranging from 10(2) to ≥10(5) c.f.u. ml(-1). Mass spectrometric identification of bacteria was accomplished for 62 of these samples. In the mass spectra obtained from 40 of the 45 urine samples for which no identification result was achieved, a triplet of very intense peaks corresponding to the human α-defensins 1, 2 and 3 occurred at m/z values of around 3440 Da. This signal suppressed the intensity of the bacterial protein peaks and thus impaired database matching. Our results show that MALDI-TOF MS allows the reliable direct identification of bacteria in urine samples at concentrations as low as 10(3) c.f.u. ml(-1). In a subset of samples, human defensins may occur and impair the mass spectrometric identification of bacteria.