Student Research Award in the Doctoral Degree Candidate Category, 28th Annual Meeting of the Society for Biomaterials, Tampa, FL, April 24–27, 2002 Quantitative analysis of binary adsorbed protein films by time of flight secondary ion mass spectrometry M. S. Wagner, 1 M. Shen, 2 T. A. Horbett, 1,2 David G. Castner 1,2 1 National ESCA and Surface Analysis Center for Biomedical Problems, Department of Chemical Engineering, Box 351750, Seattle, Washington 98195-1750 2 National ESCA and Surface Analysis Center for Biomedical Problems, Department of Bioengineering, University of Washington, Box 351750, Seattle, Washington 98195-1750 Received 15 October 2001; revised 6 February 2002; accepted 6 February 2002 Abstract: Time of flight secondary ion mass spectrometry (ToF-SIMS) is an ideal technique for the analysis of adsorbed protein films because of its surface sensitivity and chemical specificity. In this study, we examined ToF-SIMS with the multivariate calibration method partial least squares regres- sion (PLSR) for the determination of the relative abundance of the components in binary protein films adsorbed onto mica, PTFE, and heptyl amine plasma polymer substrates. These results have been compared with independently mea- sured 125 I-radiolabeled protein adsorption experiments. By applying PLSR to the ToF-SIMS data, the relative abundance of the components in the binary adsorbed protein films was quantified, and the agreement between the ToF-SIMS and 125 I-radiolabeling data was measured by the root mean square prediction error (RMSPE). Differences in protein quantification by PLSR and 125 I-radiolabeling ranged from 5 to 25 mass % RMSPE and were highly dependent on the structure of the adsorbed protein film, the substrate surface chemistry and morphology, and the number of latent vari- ables retained in the PLSR model. The limit of detection for the minor component in the adsorbed protein film was found to be approximately 10 mass %. This study demon- strates that the combination of ToF-SIMS and multivariate calibration provide complementary information to 125 I- radiolabeling about the composition and structure of binary adsorbed protein films. © 2002 Wiley Periodicals, Inc. J Biomed Mater Res 64A: 1–11, 2003 Key words: protein adsorption; ToF-SIMS; partial least squares regression INTRODUCTION Many techniques have been applied to study pro- tein adsorption. 1, 2 Typically, analysis of multicompo- nent adsorbed protein films involves a labeling step for each of the proteins detected, such as radiolabel- ing 3,4 or fluorescence labeling combined with total in- ternal reflection fluorescence 5,6 or fluorescence mi- croscopy. 7 Detection methods utilizing a reaction step such as ELISA, 8 which can be coupled with surface plasmon resonance 9 or ellipsometry, 10,11 are also com- mon. Elution of the adsorbed proteins and subsequent electrophoretic and immunoblotting analysis has also been used to analyze complex adsorbed protein films. 12 ELISA experiments pose difficulties for quan- tification of the amounts of adsorbed protein because the results are related to a reference material. 8 Recent developments in surface matrix-assisted laser desorp- tion ionization (MALDI) mass spectrometry has en- hanced the ability to detect adsorbed proteins in com- plex mixtures, 13 but this technique requires the appli- cation of a photoactive matrix and is not able to detect very large proteins. 14 Correspondence to: D. G. Castner; e-mail: castner@nb.engr. washington.edu Contract grant sponsor: National ESCA and Surface Analysis Center for Biomedical Problems (NESAC/BIO), National Institutes of Health; contract grant number: RR- 01926. Contract grant sponsor: National Center for Research Re- sources. Contract grant sponsor: NHLB; contract grant number: HL19419. © 2002 Wiley Periodicals, Inc.