Observing single proteins in their native state is a major goal in molecular biology. While single-molecule fluorescence techniques have made significant strides, the use of fluorescent markers can significantly alter protein structure and interactions. Label-free alternatives are therefore actively pursued. Protein autofluorescence in the ultraviolet (UV) range, stemming from tryptophan and tyrosine residues present in most proteins, offers a potential solution. However, the low brightness of this autofluorescence has hampered single-molecule detection. The challenge lies in the inefficient collection of emitted light, especially at large angles (supercritical or forbidden light), and the limited availability of high numerical aperture objectives in the UV range. Optical antennas, analogous to radiofrequency antennas, can enhance the emission of single quantum emitters. While significant enhancement has been achieved with visible dyes, most designs are unsuitable for UV protein detection due to narrowband responses, fabrication challenges, or requirements for solid-state integration. This study introduces optical horn antennas as a novel nanophotonic platform for label-free UV detection of single proteins. The platform combines a conical horn reflector for high-angle light collection and a metal nanoaperture for fluorescence enhancement and background screening. The broadband nature of the design overcomes limitations of previous antenna designs. The researchers aim to demonstrate the platform's utility by detecting UV autofluorescence from immobilized and diffusing single proteins, and by monitoring protein unfolding and dissociation upon denaturation.

The authors extensively review existing single-molecule fluorescence techniques and their limitations due to the use of fluorescent labels. They discuss the challenges associated with label-free single-molecule detection and highlight the potential of using intrinsic protein autofluorescence in the UV range. They analyze previous attempts to employ optical antennas for fluorescence enhancement, emphasizing the limitations of these designs in the UV spectral range. They cite previous work demonstrating plasmonic enhancement of single protein autofluorescence in zero-mode waveguides. The literature review establishes the need for a novel nanophotonic platform that addresses the limitations of existing methods for label-free single protein detection in the UV.

The researchers designed and fabricated optical horn antennas. The fabrication process involves focused ion beam (FIB) milling to create the horn antenna structure, followed by the deposition of a 100 nm aluminum layer for UV reflectivity. A nanoaperture (65 nm or 200 nm diameter) is then milled in the center of the top plateau. The substrates used were NEG1 quartz microscope coverslips. Aluminum deposition was done by electron-beam evaporation, and FIB milling was performed using a gallium-based system. A 12 nm SiO2 layer was deposited to protect the aluminum from corrosion. To assess the optical performance, p-terphenyl, a UV fluorescent dye, was used. The fluorescence enhancement was determined using fluorescence correlation spectroscopy (FCS). The influence of the cone angle on collection efficiency was studied and numerically simulated using the finite-difference time-domain (FDTD) method. For single protein detection, the nanoaperture surface was functionalized with silane-polyethylene glycol-biotin to immobilize β-galactosidase-streptavidin proteins. Control experiments with Atto647N-biotin fluorescent labels were performed to quantify the number of proteins within each nanoaperture. UV autofluorescence time traces were acquired and analyzed using exponential decay fitting. Similar experiments were conducted with pure streptavidin. For diffusing protein detection, β-galactosidase-streptavidin proteins were used at low concentrations, and 55% sucrose was added to increase the viscosity and protein residency time. Autofluorescence time traces were analyzed, and FCS correlation functions were obtained to confirm single-molecule detection. Finally, urea-induced denaturation of β-galactosidase was studied using FCS to observe unfolding and dissociation. Hydrodynamic radius was determined from the FCS data. Autofluorescence spectra and fluorescence lifetimes were also analyzed.

The study's key findings include: 1. **Optical Horn Antenna Performance:** The optical horn antenna design demonstrated significant fluorescence enhancement and improved light collection compared to a bare nanoaperture. Optimal cone angles were identified, and experimental results correlated well with FDTD simulations. The system demonstrated a collection efficiency gain of around 15x compared to confocal reference using p-terphenyl. 2. **Immobilized Single Protein Detection:** The researchers successfully detected the UV autofluorescence from immobilized single β-galactosidase-streptavidin proteins. The signal amplitude scaled with protein concentration and the number of tryptophan residues. Control experiments with fluorescently labeled proteins validated the number of proteins in the detection volume. 3. **Diffusing Single Protein Detection:** The autofluorescence bursts from diffusing single β-galactosidase-streptavidin proteins were successfully resolved at low concentrations. FCS measurements confirmed the single-molecule nature of the detected signals, overcoming challenges from low signal to noise ratio by enhancing brightness. 4. **Protein Denaturation Study:** The horn antenna facilitated the label-free study of β-galactosidase denaturation using urea. The number of detected molecules and their hydrodynamic radius were monitored as functions of urea concentration. The results provided direct single-molecule evidence of protein unfolding and dissociation, confirming previous ensemble-averaged observations. The brightness improvements achieved with the optical horn antenna allowed for a 100-fold reduction in experiment integration time compared to a confocal reference.

The development of optical horn antennas provides a significant advancement in label-free single-molecule detection, circumventing the limitations of fluorescence labeling. The platform’s combination of plasmonic enhancement, efficient light collection, and background rejection allows for the detection of intrinsically weak UV autofluorescence from single proteins. The ability to monitor protein unfolding and dissociation at the single-molecule level opens up new possibilities for studying protein dynamics and interactions in their native state. The attained brightness improvement substantially reduces the time required for experiments compared to existing methods. The study demonstrates the feasibility of label-free single-protein detection in the UV range, establishing a new avenue for biophysical investigations. The successful application to the study of β-galactosidase denaturation showcases the technique's potential for investigating complex biomolecular processes.

This study successfully demonstrated label-free detection of single proteins using their intrinsic UV autofluorescence via a novel optical horn antenna platform. The design overcomes limitations of previous antenna approaches by combining plasmonic enhancement, high-angle light collection, and efficient background rejection. This work opens new avenues for studying protein dynamics and interactions in their native states, offering a powerful tool for single-molecule biophysics. Future research could explore the application of this technology to a wider range of proteins and biomolecular systems, as well as optimization for improved sensitivity and reduced photodamage.

One limitation of the current study is the potential for photodamage to the proteins due to UV illumination. While the short illumination time used for diffusing proteins mitigates this issue, immobilized proteins are susceptible to photodamage, limiting the achievable signal-to-noise ratio. Another limitation is the use of sucrose to increase viscosity, which introduces a higher background noise level and limits the maximum achievable viscosity. Further optimization of the experimental parameters and the antenna design could help to improve this. Finally, the focus on β-galactosidase limits the generalizability of findings to other proteins.