Open Access
Add to BibSonomyAbstract
This paper discusses the application of objective-type total internal reflection fluorescence correlation spectroscopy (TIR-FCS) to the study of the kinetics of immobilized horseradish peroxidase on a single molecule level. Objective-type TIR-FCS combines the advantages of FCS with TIRF microscopy in a way that allows for simultaneous ultra-sensitive spectroscopic measurements using a single-point detector and convenient localization of single molecules on a surface by means of parallel imaging.
©2007 Optical Society of America
1. Introduction
The improvement and development of fluorescence based optical methods reached in the late 1980s a level where it was for the first time possible to directly detect single fluorophores under biologically relevant conditions [1]. At almost the same time the signal-to-noise ratio in fluorescence spectroscopy was dramatically increased by implementing the confocal detection scheme [2, 3]. These major breakthroughs boosted the development of an emerging class of optical methods named single molecule detection (SMD) [4]. It also opened brand new application fields for fluorescence spectroscopy in molecular biophysics (For a review on methods and applications see [4, 5, 6, 7]). One important cornerstone in the application of SMD was the investigation of the catalytic cycles of single enzymes. In the mid 1990s the group of Yanagida was able to observe individual enzymatic turnovers of ATP by single myosin molecules [8]. Soon afterwards, several groups investigated a number of different motor proteins and enzymes on a single molecule level e.g. kinesin [9], cholesterol oxidase [10], and horseradish peroxidase [11]. Making use of, and demanding for a permanent increase of the sensitivity of fluorescence spectroscopy techniques, ‘single enzyme kinetics’ developed to a very fertile research field, which is constantly increasing our knowledge about the physics of enzymatic catalysis.
With the single molecule approach features can be revealed that are otherwise hidden by the averaging process when a big ensemble of molecules is observed. For instance an ensemble of enzymes can be heterogeneous due to point mutations or folding into slightly different tertiary structures [12, 13]. The function and structure of enzymes are closely related and small differences in the structure of the molecules will give rise to a distribution of kinetic rates. The heterogeneity in the kinetic rates can be revealed by single molecule measurements. In addition single molecule measurements can show whether a heterogeneity observed in an ensemble measurement is due to a heterogeneous ensemble (static disorder) or rather a dynamic disorder, that is to say a heterogeneity over time for each single enzyme. Dynamic disorder is a characteristic frequently observed in measurements of single enzyme kinetics [10, 11, 14, 15, 16]. The widely accepted explanation states that an enzyme is not a rigid entity but that its conformation may fluctuate, which in turn translates into fluctuating kinetic rates [17].
One particularly useful spectroscopic technique with single molecule sensitivity is fluorescence correlation spectroscopy (FCS) [18, 2, 4, 19]. In 1999 the group of Rigler used confocal FCS to monitor the catalytic rate of single horseradish peroxidase (HRP) molecules [11]. From their measurements Rigler and co-workers concluded that the observed enzyme molecules displayed dynamic disorder in their enzymatic activity. HRP efficiently catalyzes the reduction of hydrogen peroxide in the presence of a reducing agent. Several reducing agents that can be processed by this protein are fluorogenic substances, which become highly fluorescent upon oxidation. HRP in the presence of hydrogen peroxide and an adequate fluorogenic substrate will turn over the substrate into a fluorescent product, thereby making the catalysis visible for fluorescence spectroscopy techniques.
The observation of the activity of single enzymes requires the immobilization of the molecules on a glass substrate. The requirement of measuring close to the surface leads to increased fluorescence background, stemming from the glass substrate, which complicates spectroscopic measurements. Recently, we applied objective-based total internal reflection (TIR) fluorescence excitation as an alternative to the commonly used confocal excitation in FCS [20, 21]. The surface immobilization giving an experimental complication thus also provides a possibility to use TIR-based excitation and oil-immersion objectives with high fluorescence collection efficiency. Thereby, significantly increased fluorescence signals and signal-to-noise ratios could be obtained. Here, we demonstrate the application of TIR-FCS to measure the kinetics of HRP. The used setup is an advancement of the setup presented in [20, 21], which has the additional feature that parallel imaging using a CCD camera and spectroscopic measurements using a single photon avalanche diode (SPAD) can be performed simultaneously. With this arrangement, fast and convenient localization of the immobilized enzymes can be performed by imaging of the produced fluorophores. The spectroscopic observation volume can then be positioned onto a specific enzyme by simultaneously imaging the point detector onto the CCD camera.
This work shows the versatility of objective-type TIR-FCS for single molecule spectroscopy. Our results support the hypothesis that HRP undergoes conformational changes while catalyzing the reduction of hydrogen peroxide, in agreement with the results obtained with confocal FCS by Edman et al. [11].
2. Materials and methods
2.1. Objective-type TIR-FCS
Objective-type TIR-FCS combines objective-type total internal reflection fluorescence microscopy (TIRFM) [22, 23, 24] with FCS. A detailed discussion of the used setup can be found in [20, 21]. We use a high NA oil-immersion objective (α-Plan-Fluar, 1.45NA, 100×, Carl Zeiss Jena GmbH, Jena, Germany) in an epi-illumination configuration to excite and detect fluorescence. In order to create an evanescent field at the interface between the coverslip and the sample, a laser beam is focused off-axis onto the back-focal plane of the objective (see Fig. 1). Due to this configuration the beam emerges collimated and at an oblique angle from the objective. By adjusting the focus position within the back focal plane, the angle between the beam and the optical axis can be adjusted to get total internal reflection at the coverslip-sample interface. Thereby, a circular evanescent field is created, which in our setup is approximately 15 μm in diameter (FWHM of the field intensity). For excitation, an ion argon laser with excitation filter is used, yielding a maximum power of 18 mW at 488 nm. The emitted fluorescence is detected by means of a fiber coupled SPAD (SPCM-AQR-13-FC, PerkinElmer, Wellesley, MA), with the fiber core (ø = 50 μm) acting as a pinhole. Prior to detection, scattered excitation light is blocked by a dichroic mirror and an interference filter. The applied laser power is typically ~ 7mW leaving the objective, corresponding to a maximum intensity in the center of the illuminated area of ~ 50 μW μm-2.
Fig. 1: Detailed scheme of the objective-type TIR-FCS setup. EXF: excitation filter; L1 - L4: lenses; GS: tiltable glass slab; BFP: back focal plane of objective; OL: objective lens; DM: dichroic mirror; EMF: emission filter; M: mirror; TL: tube lens; IP1, IP2: image planes; BS: beam splitter; WLS: white-light source.
The presented configuration allows for performing TIRFM and spectroscopic (e.g. FCS) measurements simultaneously by splitting the detection beam path using a beam splitter (BS, Fig. 1). Through the beam-splitter 80% of the incident fluorescence is directed onto the fiber end and 20% onto a CCD camera (DV434-BV, Andor Technology, Belfast, Northern Ireland). This enables us to locate active enzymes because fluorophores continuously produced by HRP molecules will show up on the CCD camera as bright spots at the enzymes positions.
The position of the fiber end can be aligned with the image of a HRP molecule in a convenient way by disconnecting it from the SPAD and plugging it to a white-light source. The bright (free) fiber end is then imaged onto the coverslip-sample interface, where it appears as a bright spot that can be observed by means of the CCD camera. By moving the fiber end (or alternatively, by moving the sample) using a motorized xy-translation table, the image of the fiber end is superimposed with the enzyme position, which is identified by strong, localized fluorescence emission. After connecting the fiber again to the SPAD, FCS measurements can be performed and intensity time traces can be recorded to investigate the kinetics of the enzyme.
2.2. Horseradish peroxidase (HRP)
Horseradish peroxidase is a ~ 40kDa protein which efficiently reduces hydrogen peroxide to water and oxygen. This protein is routinely used in enzyme-linked immunosorbent assays (ELISA). The catalytic reaction includes a multiple step reaction (for details see e.g. [25]). In a first step, the active site reacts with H2O2, which is reduced to water and oxygen. Thereby, the active site is oxidized leading to a state called Compound I containing an oxyferryl (Fe(IV)=O) and an organic cation radical. In this state, the enzyme can react with a substrate molecule. This oxidizes the substrate molecule, reduces the cation radical and yields Compound II. A second reaction reduces the iron to iron(III) and oxidizes a second substrate molecule. The reaction is schematically depicted in Fig. 2(left).
Fig. 2: Above: simple scheme for the catalytic cycle of horseradish peroxidase. Right: a more complete description proposed by L. Edman et al. [11] taking into account different conformational states of the enzyme and their supposed influence on its catalytic activity. S: substrate; P: product; Ei+ j: enzyme with i = 3, 4, 5, indicating the trivalent, tetravalent and pentavalent redox states, respectively. The subscript j indicates the conformational state.
A more complete picture includes the hypothesis that the protein undergoes conformational fluctuations on time scales comparable or longer than that of the enzymatic reaction. According to Edman et al. [11], the enzyme exhibits different catalytic rates in different conformations making the conformational fluctuations measurable by spectroscopic techniques (see Fig. 2, right).
HRP conjugated to streptavidin was purchased from Invitrogen (S-911, Invitrogen, Basel, Switzerland). As a substrate we used in our experiments the fluorogenic substance dihydrorho-damine 123 (dhRh123) (D-632, Invitrogen, Basel, Switzerland). This is turned over by the enzyme into the very bright and stable fluorophore rhodamine 123 (Rh123).
2.3. Enzyme immobilization
The HRP molecules we investigated are conjugated to streptavidin. The immobilization on microscope coverslides is achieved via the very stable binding to biotin molecules, attached to the surface. We functionalized the surface of standard microscope coverslips using poly(ethylene glycol)biotin grafted to poly(L-lysine) (PLL-PEG-biotin). The lysine, as a cationic amino acid, binds by electrostatic interaction to the negatively charged glass surface. Polyethylene chains, partly containing biotin at one end are grafted to the lysine molecules. The PLL-PEG-biotin builds a dense layer on the surface, as could be verified by dual polarization interferometry (DPI) [26] measurements. The function of this copolymer is twofold. Firstly, the biotin binds very tightly to the HRP-streptavidin compound (K aff = 1013-15M-1 in solution, [27]) implicating that single HRP molecules can be observed over many hours. Secondly, the copolymer layer passivates the surface, reducing interactions of the hydrophobic reaction product with the surface [27].
In order to get a small surface density of HRP we applied a mixture of two different species of PLL-PEG to the surface (both purchased from Surface Solutions GmbH, Zürich, Switzerland). PLL(20)-g[3.6]-PEG(2) (containing no biotin) was mixed withPLL(20)-g[3.5]-PEG(2)/PEG(3.4)biotin(50%) in a 1.7 × 10~-5 ratio. The molecular weight of the constituents is specified within parentheses. The value in brackets is the grafting ratio and indicates that 2 out of 7 lysine molecules are attached to poly(ethylene glycol) chains. At the given surface density of ~ 10-8molm-2 as inferred from the DPI data this should result in a sparse PLL-PEG-biotin distribution of ~ 1 per 10 μm-2. The HRP-streptavidin compound binds tightly to the biotin, resulting in a similar sparse occupancy. These predictions are in agreement with a HRP surface density in the order of 1 per 10 μm-2 as inferred from TIRFM images (Fig. 3). According to the manufacturers specification, the PLL-PEG-biotin should contain on average 14 biotin molecules per copolymer (50% of the PEG being biotynilated). We can therefore not exclude that more than one HRP molecule binds per PLL-PEG-biotin compound. However, the average number of bound HRP molecules per copolymer will be smaller than the number of biotins due to sterical hindrance and possible binding of more than one biotin to a streptavidin. Indeed, data published by Ruiz-Taylor et al. [28] suggests that on average not more than one streptavidin molecule binds to each PLL-PEG-biotin compound.
For negative control measurements, surfaces were prepared according to the same protocol as discussed above with the difference that the HRP-streptavidin compound was replaced with streptavidin.
3. Theoretical background
For data evaluation we adapted the model presented in [11] to our specific experimental situation. This model is based on the fact that HRP cycles between two spectroscopically distinguishable states, which are the bright enzyme-product (EP) complex and all other states, which are not visible, denoted E (scheme 1).
If the grand scheme shown in Fig. 2 describes the kinetics of HRP correctly and if the rates of conformational changes are not much higher than the kinetic rates, then a distribution of rates leading from E to EP would be observed rather than a single rate kb. The resulting normalized autocorrelation function of the fluorescence fluctuations G(τ) would consist of a weighted sum of exponential functions. This sum is in the present model well approximated by a stretched exponential term of the form exp[-(kbτ)β], with β ∈ [0,1]. The parameter β is a measure for the width of the rate distribution and decreases with increasing width. For β = 1 the stretched exponential term transforms into a single exponential. It turns out that the enzyme kinetics in G(τ) is fully represented by two exponential terms (with coefficients a 2 and a 3):
The exponential term with coefficient a 2 represents the dissociation of the enzyme-product complex, where ka is a first order approximation to the dissociation rate. Equation 2 was adapted from the original model presented in [11] by including triplet kinetics and by adjusting the term accounting for diffusion GD(τ) (Eq. 3). Here, T is the fraction of fluorophores in the triplet state and τT is the relaxation time of the singlet-triplet state transitions [29]. Parameters a 1, a 2 and a 3 are proportionality factors. The function GD(τ) can be derived from the diffusion equation and the molecule detection efficiency function of the microscope [30, 21]. GD(τ) can be expressed as:
where N is the mean number of molecules in the observation volume, τz is the axial diffusion time and ω is the ratio between the height and the radius of the observation volume. The proportionality constant γ relates the amplitude of the correlation function, GD(0), to N and can be calculated from the spatial distribution of the molecule detection efficiency. In our case γ = 0.29. The function w is defined by w(x) = exp(-x 2)erfc(-ix). For further details see [21].
Note that ka appears in an exponential term that is not stretched. This is reasonable, since one would assume that conformational changes affect rather the catalytic rate than the dissociation rate. The second exponential term is in this model associated with the forward reaction in scheme 1.
4. Results
4.1. TIRFM data
Fig. 3: A false color TIRF image of immobilized streptavidin, HRP conjugate in the presence of substrate and cosubstrate (H2O2) is shown on the left. The right image shows a surface where streptavidin was immobilized instead of streptavidin, HRP conjugate. Substrate and cosubstrate concentrations were the same in both cases. Intensities are given in arbitrary units.
Figure 3 shows a typical example of a TIRFM image of enzymes turning over product to substrate and a corresponding negative control. The surfaces for the negative control and for the enzyme measurement were functionalized in the same way. To prepare the coverslip for the negative control, streptavidin was immobilized in the last step instead of streptavidin, HRP conjugate. The integration time of the CCD camera for the measurement and the negative control was set to two seconds. For the data shown in Fig. 3, 30 images were accumulated in each case resulting in a total measurement time of ~ 1min. The applied sample consisted of 100nM dhRh123 and 1mM H2O2 in PBS buffer.
One problem that we faced during our experiments was the interaction of Rh123 with the coverslip surface. Rh123 is always present in the sample solution due to enzyme mediated product formation and further due to auto-catalysis of the substrate. This fluorophore binds stochastically to random locations on the glass surface even after passivating the latter with PLL-PEG. This can possibly be attributed to hydrophobic and/or electrostatic interactions. The effect can be reduced by thorough cleaning of the surface before building up the PLL-PEG layer and by increasing the ionic strength of the buffer. However, the binding of fluorophores to the surface could not be suppressed completely. The binding events show up as short flashes observable on the CCD image, or through the ocular of the microscope. A single flash due to a binding event followed by de-binding or bleaching is a priori not distinguishable from a flash due to the production and subsequent release of a Rh123 molecule. In order to single out locations where substrate is turned over to fluorescent product we accumulated CCD images over long time spans (~ 1min). In this way, flashes occurring stochastically in space were drowned in the background due to the long measurement time. Positions, where fluorescence was produced continuously appeared as bright spots on the images (Fig. 3).
4.2. TIR-FCS data
Fig. 4: Correlogram and part of the intensity time trace for enzyme turnover. A fit to the correlogram is shown in red. The used model is given by Eq. 2.
Following enzyme localization as outlined above, intensity time traces and FCS data were recorded from immobilized enzymes. Figure 4 shows a typical example for a correlogram and a detail from the corresponding intensity time trace obtained for enzyme mediated catalysis. The intensity fluctuations were measured at the position of the bright spot highlighted by the red rectangle shown in Fig. 3. The used sample was a mixture of 100nM dhRh123 and 1mM H2O2 in PBS buffer. In all FCS measurements the data was collected over a period of 30s and the model given by Eq. 2 was fit to the correlograms using a global search algorithm. The rate kb was found to be distributed over several orders of magnitude, with β typically ranging between 0.15 and 0.3 at substrate concentrations ≤ 500 nM. The fact that β takes values well below one, indicates dynamic disorder and supports the hypothesis of slow conformational motions affecting the turnover rate of HRP. The parameter estimates are given in table 1.
Figure 5 shows a typical correlogram and the corresponding intensity time-trace measured ~ 1μm aside from the position highlighted in Fig. 3. Equation 2 could be fit to the correlogram with the factors a 2 and a 3 fixed to zero, i.e. with a model accounting only for diffusion. The parameters estimated from these control measurements were used to fit the data recorded for enzymatic turnover (see table 1).
Fig. 5: Correlogram and part of the intensity time trace measured at a position where no enzyme is located. A fit to the correlogram is shown in red.
Table 1:. Parameter estimates for data from single enzyme measurements and a control measurement. i: Estimates for the data displayed in Fig. 4. ii, iii: Results from successive measurements for the same enzyme. iv: Results from a control measurement. The data corresponding to the control measurements is displayed in Fig. 5. *fixed parameters.
The estimates for the rates ka and kb are 1 and 2 orders of magnitude higher as compared to values that have been obtained earlier with different substrates [11, 31]. In the present experiments we used as a substrate Rh123 instead of Rh6G that was used in the earlier experiments. We attribute the differences in the obtained rates to differences in the properties i.e. hydrophobicity, charge density and size of these molecules. In ongoing work the differences in the observed rates are further investigated.
5. Summary
Objective-type TIR-FCS is a technique for investigating fluorescence fluctuations close to a surface, which is simple to implement and characterized by an excellent signal-to-noise ratio. Furthermore, this method can be easily combined with TIRFM. This adds a flexibility to the method that can be used as an advantage in different single molecule spectroscopy applications. We applied objective-type TIR-FCS to investigate the turnover of dihydrorhodamine 123 to rhodamine 123 by single, immobilized HRP molecules. Thereby, we demonstrated the applicability of this technique to the investigation of the kinetics of single enzymes immobilized on a glass surface. The combination with TIRFM in this application greatly facilitates the localization and identification of single immobilized enzymes. Our results are in principle in accordance with previously reported results and support in particular the hypothesis that slow conformational fluctuations of the enzymes can influence their catalytic activity. However, the rates we observed were higher compared to what was observed in earlier work. We attribute this to the fact that we used a different substrate for the experiments reported in this paper.
Acknowledgments
This work was supported by the Swiss National Science Foundation (SNSF), the Wenner-Gren Foundation, the Swedish National Research Council and the Swedish Foundation for Strategic Research. We are grateful to Andreas Sonesson for helping us to perform the DPI measurements.
References and links
1. E. B. Shera, N. K. Seitzinger, L. M. Davis, R. A. Keller, and S. A. Soper “Detection of single fluorescent molecules,” Chem. Phys. Lett. 174, 553–557 (1990). [CrossRef]
2. R. Rigler and J. Widengren, “Ultrasensitive detection of single molecules by fluorescence correlation spec-troscopy,” in BioScience, B. Klinge and C. Owman, eds., (Lund University Press, Lund, Sweden, 1990), pp. 180–183.
3. R. Rigler, U. Mets, J. Widengren, and P. Kask, “Fluorescence Correlation Spectroscopy with High Count Rate and Low Background - Analysis of Translational Diffusion,” Eur. Biophys. J. Biophys. Lett. 22, 169–175 (1993). [CrossRef]
4. M. Eigen and R. Rigler, “Sorting single molecules - application to diagnostics and evolutionary biotechnology,” Proc. Natl. Acad. Sci. USA 91, 5740–5747 (1994). [CrossRef] [PubMed]
5. S. Weiss, “Fluorescence spectroscopy of single biomolecules,” Science 283, 1676–1683 (1999). [CrossRef] [PubMed]
6. W. P. Ambrose, P. M. Goodwin, J. H. Jett, A. Van Orden, J. H. Werner, and R. A. Keller, “Single molecule fluorescence spectroscopy at ambient temperature,” Chem. Rev. 99, 2929–2956 (1999). [CrossRef]
7. J. Enderlein, R. A. Keller, and C. Zander, “Single Molecule Detection in Liquids and on Surfaces under Ambient Conditions: Introduction and Historical Overview,” in Single Molecule Detection in Solution, C. Zander, J. Enderlein, and R. A. Keller, eds., (Wiley-VCH, Berlin, 2002), pp 1–19.
8. T. Funatsu, Y. Harada, M. Tokunaga, K. Saito, and T. Yanagida, “Imaging of Single Fluorescent Molecules and Individual Atp Turnovers by Single Myosin Molecules in Aqueous-Solution,” Nature 374, 555–559 (1995). [CrossRef]
9. R. D. Vale, T. Funatsu, D. W. Pierce, L. Romberg, Y. Harada, and T. Yanagida, “Direct observation of single kinesin molecules moving along microtubules,” Nature 380, 451–453 (1996). [CrossRef] [PubMed]
10. H. P. Lu, L. Y. Xun, and X. S. Xie, “Single-molecule enzymatic dynamics,” Science 282, 1877–1882 (1998). [CrossRef]
11. L. Edman, Z. Fö;ldes-Papp, S. Wennmalm, and R. Rigler, “The fluctuating enzyme: a single molecule approach,” Chem. Phys. 247, 11–22 (1999). [CrossRef]
12. R. Polakowski, M. Eggertsson, D. B. Craig, and N. J. Dovichi, “The Chemistry of a Single Enzyme Molecule,” in Single Molecule Detection in Solution, C. Zander, J. Enderlein, and R. A. Keller, eds. (Wiley-VCH Verlag Berlin GmbH, Berlin, 2002), pp. 293–301.
13. H. Frauenfelder and B. H. McMahon, “Energy landscape and fluctuations in proteins,” Ann. Phys. 9, 655–667 (2000). [CrossRef]
14. K. Velonia, O. Flomenbom, D. Loos, S. Masuo, M. Cotlet, Y. Engelborghs, J. Hofkens, A. E. Rowan, J. Klafter, R. J. M. Nolte, and F. C. de Schryver, “Single-enzyme kinetics of CALB-catalyzed hydrolysis,” Angew. Chem. Int. Ed. 44, 560–564 (2005). [CrossRef]
15. O. Flomenbom, K. Velonia, D. Loos, S. Masuo, M. Cotlet, Y. Engelborghs, J. Hofkens, A. E. Rowan, R. J. M. Nolte, M. Van der Auweraer, F. C. de Schryver, and J. Klafter, “Stretched exponential decay and correlations in the catalytic activity of fluctuating single lipase molecules,” Proc. Natl. Acad. Sci. USA 102, 2368–2372 (2005). [CrossRef] [PubMed]
16. B. P. English, W. Min, A. M. van Oijen, K. T. Lee, G. B. Luo, H. Y. Sun, B. J. Cherayil, S. C. Kou, and X. S. Xie, “Ever-fluctuating single enzyme molecules: Michaelis-Menten equation revisited,” Nat. Chem. Biol. 2, 87–94 (2006). [CrossRef] [PubMed]
17. H. Frauenfelder, S. G. Sligar, and P. G. Wolynes, “The Energy Landscapes and Motions of Proteins,” Science 254, 1598–1603 (1991). [CrossRef] [PubMed]
18. D. Magde, W. W. Webb, and E. Elson, “Thermodynamic fluctuations in a reacting system - Measurement by Fluorescence Correlation Spectroscopy,” Phys. Rev. Lett. 29, 705–708 (1972). [CrossRef]
19. R. Rigler and E. L. Elson, Fluorescence Correlation Spectroscopy: Theory and Applications (Springer, Berlin, 2001). [CrossRef]
20. K. Hassler, T. Anhut, R. Rigler, M. Gösch, and T. Lasser, “High count rates with total internal reflection fluorescence correlation spectroscopy,” Biophys. J. 88, L1–L3 (2005). [CrossRef]
21. K. Hassler, M. Leutenegger, P. Rigler, R. Rao, R. Rigler, M. Gösch, and T. Lasser, “Total internal reflection fluorescence correlation spectroscopy (TIR-FCS) with low background and high count-rate per molecule,” Opt. Express 13, 7415–7423 (2005). [CrossRef] [PubMed]
22. D. Axelrod, T. P. Burghardt, and N. L. Thompson, “Total Internal-Reflection Fluorescence,” Ann. Rev. Biophys. Bioeng. 13, 247–268 (1984). [CrossRef]
23. D. Axelrod, E. H. Hellen, and R. M. Fulbright, “Total Internal Reflection Fluorescence,” in Topics in Fluorescence Spectroscopy: Biochemical Applications, J. R. Lakowicz, ed., (Plenum Press, New York, 1992), vol. 3, pp. 289–343.
24. D. Axelrod, “Total internal reflection fluorescence microscopy in cell biology,” Traffic 2, 764–774 (2001). [CrossRef] [PubMed]
25. L. Banci, “Structural properties of peroxidases,” J. Biotechnol. 53, 253–263 (1997). [CrossRef] [PubMed]
26. M. J. Swann, L. L. Peel, S. Carrington, and N. J. Freeman, “Dual-polarization interferometry: an analytical technique to measure changes in protein structure in real time, to determine the stoichiometry of binding events, and to differentiate between specific and nonspecific interactions,” Anal. Biochem. 329, 190–198 (2004). [CrossRef] [PubMed]
27. N. P. Huang, J. Voros, S. M. De Paul, M. Textor, and N. D. Spencer, “Biotin-derivatized poly(L-lysine)-g-poly(ethylene glycol): A novel polymeric interface for bioaffinity sensing,” Langmuir 18, 220–230 (2002). [CrossRef]
28. L. A. Ruiz-Taylor, T. L. Martin, and P. Wagner, “X-ray photoelectron spectroscopy and radiometry studies of biotin-derivatized poly(L-lysine)-grafted-poly(ethylene glycol) monolayers on metal oxides,” Langmuir 17, 7313–7322 (2001). [CrossRef]
29. J. Widengren, Ü. Mets, and R. Rigler, “Fluorescence Correlation Spectroscopy of Triplet States in Solution: A Theoretical and Experimental Study,” J. Phys. Chem. 99, 13368–13379 (1995). [CrossRef]
30. T. E. Starr and N. L. Thompson, “Total internal reflection with fluorescence correlation spectroscopy: Combined surface reaction and solution diffusion,” Biophys. J. 80, 1575–1584 (2001). [CrossRef] [PubMed]
31. K. Hassler, “Single molecule detection and fluorescence correlation spectroscopy on surfaces,” Ph.D. Dissertation, École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland (2005).
