In this study, the intensity of two-photon excited fluorescence (TPEF) of xanthene dye, Rose Bengal (RB), was significantly enhanced via bovine serum albumin (BSA) microstructures fabricated by the two-photon crosslinking (TPC) technique. The RB was utilized as the photoactivator in the TPC processing and the enhanced TPEF intensity correlates with the concentration of fabricated crosslinked BSA microstructures via the power control and pulse selection of the employed femtosecond laser. As a result, fabrication of three-dimensional BSA microstructures can be simultaneously monitored by the use of TPEF intensity. The crosslinked BSA microstructures synthesized may be used as an ordered biomaterial for fluorescence enhancement.
© 2011 OSA
Photopolymerization or photocrosslinking is a process which uses a combination of light with low molecular weight photoinitiators to trigger the polymerization or crosslinking reaction [1–3]. In particular, the use of multiphoton excited (MPE) photochemistry in creating microstructures offers a unique advantage. Specifically, since MPE photochemistry is confined to the focal volume, spatially-precise, sub-micron microstructures can be created in 3D [4–8]. This approach not only allows for the creation of structures that cannot be assembled by conventional single-photon lithography, but it also enables greater spatial resolution than other 3D microfabrication techniques to be achieved. Therefore, multiphoton polymerization and crosslinking have attracted widespread interest due to their potential use in the fabrication of 3D microstructures at sub-diffraction limited spatial resolution . Recently, 3D microfabrication has been demonstrated in polymerized resin- [3,4,9,10] and crosslinked protein-  substrates.
There are numerous applications that utilize biomaterials as a key functional component. One such material is deoxyribonucleic acid (DNA) biopolymer. DNA-doped biopolymers have excellent optical and electrical properties, such as low optical loss in the visible light and infrared regions. Furthermore, DNA films have a relatively high thermal stability of around 200-250 °C . It has also been observed that DNA-doped biopolymer thin films can enhance fluorescence emission. Based on this characteristic, studies targeting DNA light-emitting diodes and DNA distributed feedback Bragg lasers have been conducted [11–15]. However, the mechanisms of luminescence enhancement are not yet fully understood. A possible reason is the intercalation of dye molecules between base pairs of the DNA structure. This phenomenon can prevent the aggregation of dye molecules which can reduce the dye’s fluorescence emission. In addition, intercalated molecules in base-pair structures can also restrict conformation change of the fluorophores. Due to these features, DNA-doped biopolymers can effectively increase fluorophore photostability and results in luminescence enhancement.
In addition to DNA-doped biopolymers, other materials such as mesoporous silica  and dendrimers  can also enhance fluorophore luminescence. However, a limitation exists in that these materials cannot be easily fabricated into 3D structures. However, the 3D microstructures constructed by multiphoton photopolymerization or photocrosslinking can overcome this limitation. Specifically, we found that, compared to results in solution, the two-photon excited fluorescence (TPEF) of Rose Bengal (RB) decreased in two-photon polymerization (TPP) generated structures of ethoxylated trimethylolpropane triacrylate (ethoxylated TMPTA). In comparison, TPEF of RB is significantly enhanced in bovine serum albumin (BSA) structures produced from two-photon crosslinking (TPC). Specifically, RB was utilized as the photoinitiator in TPP and the photoactivator in TPC processes . The enhanced TPEF intensity correlates with the concentration of fabricated crosslinked BSA microstructures via modulation of the power and pulse number of the employed femtosecond laser. Therefore, the crosslinked BSA-structured biomaterial not only provides significant TPEF enhancement, but also provides an opportunity to develop 3D fluorescent microstructures. Furthermore, the fabrication of 3D gray-level BSA microstructures can be monitored in real time according to the localized intensity of the enhanced TPEF.
2. Sample preparation and microfabrication setup
2.1. Femtosecond laser microfabrication system and designing of 3D structures
Figure 1 shows the schematic of the femtosecond laser fabrication system. Key components of our instrument include a femtosecond, titanium-sapphire (ti-sa) laser (Tsunami, Spectra-Physics, USA), an inverted optical microscope (Axiovert 200, Zeiss, Germany), a galvanometer x-y scanner (6215H, Cambridge, USA), a triple-axis sample-positioning stage (ProScanTMII, Prior, UK), a z-axis piezoelectric nano-positioning stage (Nano-F100, Mad City Labs, USA), an acousto-optic modulator (AOM) (23080-x-1.06-LTD, Neos, USA), photomultiplier tubes (PMTs) (H5783P, Hamamatsu, Japan), and a data acquisition (DAQ) card with a field-programmable gate array (FPGA) module (PCI-7831R, National Instruments, USA). To overcome the group velocity dispersion of the femtosecond laser through the AOM and the objective, an SF-10 prism pair (PC-TS-KT, Newport, USA) was used for optimized laser operation in the wavelength region between 700 to 840 nm . A detailed description of the multiphoton fabrication instrument/microscope can be found in our previous study .
The real-time FPGA DAQ card based on our custom LabVIEW program can synchronously control the instrument through interfaces constructed in-house. The FPGA module was designed to perform a number of simultaneous tasks including control of the galvanometer scanner and the z-axis piezoelectric stage for 3D focal spot positioning; modulating the AOM for rapid on/off switching of the laser and pulse selection; and processing of the single photon counting (SPC) signals. A digitally controlled voltage converter was connected into the AOM for fast on/off laser control and pulse selection, for a speed of up to 9 MHz. Also, an analog output (from the DAQ card) was utilized to control the laser power level. The voltage converter reduces the 3.3 V command signal from the FPGA digital input/output (I/O) signal line to below 1.0 V to match the requirement of the AOM driver. The SPC-based pulse counting scheme was based on the FPGA digital I/O. Specifically, the number of high to low electronic transition signals from a discriminator-processed PMT was determined. The pulse counter records one count when the voltage level underwent a high to low transition. In addition to nonlinear optical imaging capabilities, CAD software such as AutoCAD, Pro/E, and Solidworks was used to design 3D structures for microfabrication. To transform 3D structures into 2D processing patterns, our design-transformation program was adopted to convert the 3D structures into sequential 2D DXF files. The 2D DXF files are then converted into bitmap files and downloaded into the FPGA module as laser processing commands. In this manner, we were able to create the desired 3D structures.
2.2. Sample preparation and selection
A protein, BSA, and a resin, ethoxylated TMPTA, were employed as reactive monomers. The first fabricated solution consisted of the photoactivator Rose Bengal (RB) (Avocado Research Chemicals, UK), mixed into 20 mg/ml BSA solution (Sigma-Aldrich, USA). The second solution consisted of RB as a photoinitiator and a 0.1 M triethanolamine (TEA) co-initiator (Sigma-Aldrich, USA) solution, and was mixed into ethoxylated TMPTA (Sartomer, USA) solution. All chemicals and reagents were of the analytical grade. In our previous two-photon absorption (TPA) spectrum measurement experiment, it was found that within the excitation wavelengths we examined, the excitation wavelength corresponding to the maximum value of the relative TPA of the RB was between 710 and 720 nm . Therefore, a fabrication laser wavelength of around 720 nm was adopted. According to our previous experimental results, the concentration of RB for 20 mg/ml BSA can be chosen to be around 2.0 mM, since a high RB concentration was not necessary for the TPC process. The fabrication solution was confined in a small chamber, which was created by using a 100 μm-thick spacer to separate a cover slip and a microscope slide. Before studying the enhanced TPEF in BSA fabricated structures, the TPEF intensities at different RB concentrations in deionized water (DIW), BSA, and ethoxylated TMPTA were tested (Fig. 2 ). The RB concentrations were 0.5, 1.0, 1.5, 2.0, and 2.5 mM, respectively. The TPEF intensity of RB in the BSA solution is slightly lower than that in the DIW solution. The TPEF intensity of RB in ethoxylated TMPTA solution is the highest among the three solutions.
3. Experimental results and discussions
3.1. Ethoxylated TMPTA polymerization versus BSA crosslinking
For BSA TPC processing, the xanthene dye (RB) was utilized as the photoactivator. The TPA process excites the electrons into the S2 level. Upon non-radiation decays to S1, subsequent intersystem crossing to the long-lived first triplet (T1) then occurs. The T1 electron efficiently converts triplet oxygen into singlet oxygen (~90%) . These reactive species can then react with a protein molecule, creating a radical that then binds to a second protein molecule, resulting in a covalently crosslinked structure. For ethoxylated TMPTA TPP process, the electrons of the activated photoinitiator (RB) have the same excitation to reach T1. These reactive species can then react with co-initiator TEA and form the TEA radical cation. Finally, the TEA radical cation initiates the polymerization reaction . Since TPP is a chain reaction, the efficiency of the 3D microfabrication via TPP is much higher than that through TPC.
In our study, the experimental results were utilized to investigate the differences between the RB TPEF intensities at the same RB concentration in the TMPTA TPP and BSA TPC processes. To implement multiphoton fabrication of 3D microstructures, the power of the 100 fs laser at the repetition rate of 80 MHz must be sufficient to support MPE photochemistry processing. According to our experience, the use of an NA 1.3 objective and an x-galvanometer scan rate of 1 kHz, the laser power at the TPA wavelength of RB needed to be at least 1.0 mW in order to initiate multiphoton fabrication. Furthermore, we noted that the power needed in the TPP process is lower than that in the TPC case. For this study, the laser powers were settled as 0.67 mW for TPP and 1.73 mW for TPC. Moreover, the laser power under such fast scan fabrication processes should not induce photobleaching in RB. Figures 3(a) and 3(b) show the TPEF images of fabricated TMPTA polymerization and BSA crosslinked microstructures with a size of 10 × 10 × 3 μm3 within the respective fabrication solutions. For TPEF imaging, the laser power was lowered to 0.13 mW at a 5 kHz x-galvanometer scan rate to avoid further fabrication. The average photon counts of 10 by 10 pixels at four selected locations A to D (A & B in Fig. 3(a) and C & D in Fig. 3(b)) are 263, 562, 599, and 13, respectively. The number of the photon count was based on the accumulation of 5 scans. The TPEF of the ethoxylated TMPTA solution with 2.0 mM RB is much more intense than that of the BSA solution with the same RB concentration (Figs. 3(a) and 3(b)). These results are consistent to that found in Fig. 2. TPEF intensity in the ethoxylated TMPTA fabricated square structure is weaker than that in the TMPTA fabrication solution (Fig. 3(a)). Conversely, the TPEF intensity is significantly enhanced in the BSA structures compared to that in the BSA fabrication solution (Fig. 3(b)). More than 40-fold enhancement was achieved. Therefore, two advantages are revealed of our technique. First, TPEF intensity can be significantly enhanced via crosslinked BSA microstructures. Furthermore, the fabrication process can be monitored according to the local TPEF intensity.
3.2. Enhanced TPEF from modulation of laser power and pulse number
During the BSA TPC process, we found that the intensity of the RB TPEF is depended on different fabrication parameters including laser power, laser dose, and the concentration of the BSA solution. Therefore, we attempted to find the relation between the TPEF intensity and the concentration of fabricated BSA structures, and then regionally control the enhancement factors to develop TPEF BSA microstructures. Herein, the TPEF intensity of fabricating BSA structures is modulated by using different laser powers and different pulse numbers in the same concentration of BSA solution (20 mg/ml). The AOM in Fig. 1 acted as an intensity modulator to adjust the laser power and as a pulse selector to tune the pulse numbers on the fabrication pixel or area. When the AOM is operated at the A/O mode of the FPGA controller, the laser power can be controlled through different A/O voltages. However, the updating rate of the A/O channel was not fast enough for pulse selection. Therefore, in order to achieve accurate pulse selection, a digital I/O pin was implemented to command the AOM driver. The amount of pulses was controlled to hit the fabrication area by modulating the on/off duty cycle of the AOM.
Figure 4(a) is the enhanced TPEF image of 5 fabricated BSA crosslinked squares, each with an area of 12 × 12 μm2 and a 3 μm interval separated adjacent squares. The fabrication laser powers of 1.07, 1.2, 1.33, 1.47, and 1.6 mW (top) and a pulse duty cycle of 20%, 40%, 60%, 80%, and 100% at 1.6 mW (bottom) with 1 kHz scan rate was employed. A pulse duty cycle of 100% corresponds to a selected pulse number of 80,000. Similarly, the pulse duty cycle of 20% has a pulse number of 16,000. Furthermore, the TPEF laser power was lowered to 0.13 mW at the 5 kHz x-galvanometer scan rate and the photon count was based on the accumulation of 5 scans Fig. 4(b) is the corresponding bright-field images of Fig. 4(a). Figures 4(c) and 4(d) are the corresponding TPEF photon count variations of two line-cuts from the top and bottom in Fig. 4(a), respectively. Figures 4(a)–4(d) shows that the RB TPEF intensity becomes brighter when the BSA structures were fabricated with stronger laser doses. However, an excessive laser power will saturate the value of TPEF intensity and even leads to its decrease the value due to photobleaching. If the laser power was too weak, an effective reaction spot cannot be formed, and the formation of the microstructures would not initiate. Although enhanced TPEF microstructures can be made by both modulating laser power and pulse number, the mixing scheme is not suitable to dynamically fabricate a complex structure. From Fig. 4(d), a near linear relationship between the TPEF intensity and the pulse number can be found when the laser power was high enough to fully develop the BSA crosslinked structures. Hence, the pulse modulation is a more suitable scheme to fabricate complex BSA structures with different local TPEF enhancements.
3.3. 3D BSA microstructures with localized TPEF enhancements
Based on the above discussion, 3D BSA crosslinked microstructures with different local TPEF enhancements can be fabricated by using the pulse modulation at the fully developed laser power such as 1.6 mW. To demonstrate the capability of developing 3D gray-level BSA microstructures with enhanced TPEF, a 3D 12 × 12 × 12 μm3 solid container enclosing a 2 × 2 × 3 μm3 cuboid on the top and four 3 × 3 × 3 μm3 cubes on the bottom was first designed based on CAD. A CAD draft is shown in Fig. 5(a) . Sequential 2D bitmap files slicing from the designed 3D pattern was loaded into the FPGA controller. The BSA structure is defined by various BSA concentrations inside the structure. Localized BSA concentration can be achieved by modulating the laser pulse number incident upon the designed local region. Different BSA concentrations can provide proportional TPEF enhancement effects. A 3D BSA microstructure, based on the Fig. 5(a) pattern design, was fabricated (pulse duty cycle of 66.7% for the 5 inside tiny structures and 100% for the solid container). Figures 5(b) and 5(c) are the 2D TPEF images for the four 3 × 3 × 3 μm3 cubes on the bottom and for the 2 × 2 × 3 μm3 cuboid on the top, respectively. The average photon counts of 10 by 10 pixels at the indicated locations of A to F are 453 (A), 283 (B), 15 (C), 260 (D), 512 (E), and 18 (F). Figure 5(d) is the top view of the 3D rendered TPEF image. Additionally, another 3D gray-level BSA microstructure based on the Fig. 5(a) design with a pulse duty cycle of 100% for the 5 inside tiny structures and 66.7% for the container was fabricated. The 2D TPEF images for the four 3 × 3 × 3 μm3 cubes on the bottom and for the 2 × 2 × 2 μm3 cuboid on the top are shown in Figs. 5(e) and 5(f), respectively. The average photon counts of 10 by 10 pixels at the indicated locations of G to L are 495 (G), 512 (H), 18 (I), 414 (J), 405 (K), and 20 (L). The imaging laser power was 0.89 mW at the 20 kHz scan rate for the two BSA microstructures. In these microstructures, the number of the photon count was directly obtained without multiple accumulations.
The main solid container should be strong enough to support the whole device; however, too strong a laser power will saturate the fabrication process, and loses the image contrast. Therefore, the fabrication laser power at the latter BSA microstructure was chosen to be 2.4 mW, which is higher than that of the previously used value of 1.6 mW (fabrication scan rates were set at the 1 kHz scan rate). The localized TPEF intensity was significantly enhanced and correlated with the concentration of fabricated 3D crosslinked BSA microstructures by modulating the pulse duty cycle of the femtosecond laser at the fully developed fabrication laser power. The fabricated 3D BSA microstructures can be instantaneously monitored by using the contrast of the enhanced TPEF intensity. The crosslinked BSA microstructures we created may be used as structured biomaterial with fluorescence enhancement capabilities.
TPEF intensity was significantly enhanced via 3D crosslinked BSA microstructures fabricated by the TPC technique (RB as the photoactivator). In contrast, the TPEF intensity in ethoxylated TMPTA polymer by TPP with RB photoinitiator was found to have decreased. The enhanced TPEF intensity was proportional to the concentration of the fabricated BSA microstructure, laser power, and the pulse number of the ti-sa femtosecond laser. In situ and real time monitoring of the 3D fabricating microstructure can be achieved by utilizing the TPEF as a contrast mechanism.
This work was supported by the National Research Program for Genomic Medicine (NRPGM) of the National Science Council (NSC) of Taiwan (NSC 97-3112-B-006-013), (NSC 97-3111-B-006-004), and by the Advanced Optoelectronic Technology Center in National Cheng Kung University.
References and links
1. C. R. Lambert, I. E. Kochevar, and R. W. Redmond, “Differential reactivity of upper triplet states produces wavelength-dependent two-photon photosensitization using Rose Bengal,” J. Phys. Chem. B 103(18), 3737–3741 (1999). [CrossRef]
2. J. D. Pitts, P. J. Campagnola, G. A. Epling, and S. L. Goodman, “Submicron multiphoton free-form fabrication of proteins and polymers: studies of reaction efficiencies and applications in sustained release,” Macromolecules 33(5), 1514–1523 (2000). [CrossRef]
3. P. J. Campagnola, D. M. Delguidice, G. A. Epling, K. D. Hoffacker, A. R. Howell, J. D. Pitts, and S. L. Goodman, “3-dimensional submicron polymerization of acrylamide by multiphoton excitation of xanthene dyes,” Macromolecules 33(5), 1511–1513 (2000). [CrossRef]
5. P. Galajda and P. Ormos, “Complex micromachines produced and driven by light,” Appl. Phys. Lett. 78(2), 249–251 (2001). [CrossRef]
6. T. Tanaka, H. B. Sun, and S. Kawata, “Rapid sub-diffraction-limit laser micro/nanoprocessing in a threshold material system,” Appl. Phys. Lett. 80(2), 312–314 (2002). [CrossRef]
7. M. Miwa, S. Juodkazis, T. Kawakami, S. Matsuo, and H. Misawa, “Femtosecond two-photon stereo-lithography,” Appl. Phys., A Mater. Sci. Process. 73(5), 561–566 (2001). [CrossRef]
9. T. Watanabe, M. Akiyama, K. Totani, S. M. Kuebler, F. Stellacci, W. Wenseleers, K. Braun, S. R. Marder, and J. W. Perry, “Photoresponsive hydrogel microstructure fabricated by two-photon initiated Polymerization,” Adv. Funct. Mater. 12(9), 611–614 (2002). [CrossRef]
10. Z. B. Sun, X. Z. Dong, W. Q. Chen, S. Nakanishi, M. Duan, and S. Kawata, “Multicolor polymer nanocomposites: in situ synthesis and fabrication of 3D microstructures,” Adv. Mater. (Deerfield Beach Fla.) 20(5), 914–919 (2008). [CrossRef]
11. A. J. Steckl, “DNA – a new material for photonics?” Nat. Photonics 1(1), 3–5 (2007). [CrossRef]
12. J. A. Hagen, W. Li, A. J. Steckl, and J. G. Grote, “Enhanced emission efficiency in organic light-emitting diodes using deoxyribonucleic acid complex as an electron blocking layer,” Appl. Phys. Lett. 88(17), 171109 (2006). [CrossRef]
13. Y. Ner, D. Navarathne, D. M. Niedzwiedzki, J. G. Grote, A. V. Dobrynin, H. A. Frank, and G. A. Sotzing, “Stabilization of fluorophore in DNA thin films,” Appl. Phys. Lett. 95(26), 263701 (2009). [CrossRef]
15. J. Mysliwiec, L. Sznitko, A. Sobolewska, S. Bartkiewicz, and A. Miniewicz, “Lasing effect in a hybrid dye-doped biopolymer and photochromic polymer system,” Appl. Phys. Lett. 96(14), 141106 (2010). [CrossRef]
16. P. Yang, G. Wirnsberger, H. C. Huang, S. R. Cordero, M. D. McGehee, B. Scott, T. Deng, G. M. Whitesides, B. F. Chmelka, S. K. Buratto, and G. D. Stucky, “Mirrorless lasing from mesostructured waveguides patterned by soft lithography,” Science 287(5452), 465–467 (2000). [CrossRef]
17. A. Otomo, S. Yokoyama, T. Nakahama, and S. Mashiko, “Supernarrowing mirrorless laser emission in dendrimer-doped polymer waveguides,” Appl. Phys. Lett. 77(24), 3881–3883 (2000). [CrossRef]
18. D. C. Neckers, “Rose Bengal,” J. Photochem. Photobiol., A 47(1), 1–29 (1989). [CrossRef]
19. Z. Zhang and T. Yagi, “Observation of group delay dispersion as a function of the pulse width in as mode locked Ti:sapphire laser,” Appl. Phys. Lett. 63(22), 2993–2995 (1993). [CrossRef]
20. W.-S. Kuo, C.-H. Lien, K.-C. Cho, C.-Y. Chang, C.-Y. Lin, L. L. H. Huang, P. J. Campagnola, C. Y. Dong, and S.-J. Chen, “Multiphoton fabrication of freeform polymer microstructures with gold nanorods,” Opt. Express 18(26), 27550–27559 (2010). [CrossRef]