Site-Specific Conjugation of Native Antibody: Transglutaminase-Mediated Modification of a Conserved Glutamine While Maintaining the Primary Sequence and Core Fc Glycan via Trimming with an Endoglycosidase.

A versatile chemo-enzymatic tool to site-specifically modify native (nonengineered) antibodies is using transglutaminase (TGase, E.C. 2.3.2.13). With various amines as cosubstrates, this enzyme converts the unsubstituted side chain amide of glutamine (Gln or Q) in peptides and proteins into substituted amides (i.e., conjugates). A pleasant surprise is that only a single conserved glutamine (Gln295) in the Fc region of IgG is modified by microbial TGase (mTGase, EC 2.3.2.13), thereby providing a highly specific and generally applicable conjugation method. However, prior to the transamidation (access to the glutamine residue by mTGase), the steric hindrance from the nearby conserved N-glycan (Asn297 in IgG1) must be reduced. In previous approaches, amidase (PNGase F, EC 3.5.1.52) was used to completely remove the N-glycan. However, PNGase F also converts a net neutral asparagine (Asn297) to a negatively charged aspartic acid (Asp297). This charge alteration may markedly change the structure, function, and immunogenicity of an IgG antibody. In contrast, in our new method presented herein, the N-glycan is trimmed by an endoglycosidase (EndoS2, EC 3.2.1.96), hence retaining both the core N-acetylglucosamine (GlcNAc) moiety and the neutral asparaginyl amide. The trimmed glycan also reduces or abolishes Fc receptor-mediated functions, which results in better imaging agents by decreasing nonspecific binding to other cells (e.g., immune cells). Moreover, the remaining core glycan allows further derivatization such as glycan remodeling and dual conjugation. Practical and robust, our method generates conjugates in near quantitative yields, and both enzymes are commercially available.

Methods for assessing and removing non-specific photoimmunotherapy damage in patient-derived tumor cell culture models.

Tumor-targeted, activatable photoimmunotherapy (taPIT) has been shown to selectively destroy tumor in a metastatic mouse model. However, the photoimmunoconjugate (PIC) used for taPIT includes a small fraction of non-covalently associated (free) benzoporphyrin derivative (BPD), which leads to non-specific killing in vitro. Here, we report a new treatment protocol for patient-derived primary tumor cell cultures ultrasensitive to BPD photodynamic therapy (BPD-PDT). Based on free BPD efflux dynamics, the updated in vitro taPIT protocol precludes non-specific BPD-PDT by silencing the effect of free BPD. Following incubation with PIC, incubating cells with PIC-free medium allows time for expulsion of free BPD whereas BPD covalently bound to PIC fragments is retained. Administration of the light dose after the intracellular free BPD drops below the threshold for inducing cell death helps to mitigate non-specific damage. In this study, we tested two primary ovarian tumor cell lines that are intrinsically chemoresistant, yet ultrasensitive to BPD-PDT such that small amounts of free BPD (a few percent of the total BPD dose) lead to potent induction of cell death upon irradiation. The modifications in the protocol suggested here improve in vitro taPIT experiments that lack in vivo mechanisms of free BPD clearance (i.e., lymph and blood flow).

Denoising multiplexed microscopy images in n-dimensional spectral space.

Hyperspectral fluorescence microscopy images of biological specimens frequently contain multiple observations of a sparse set of spectral features spread in space with varying intensity. Here, we introduce a spectral vector denoising algorithm that filters out noise without sacrificing spatial information by leveraging redundant observations of spectral signatures. The algorithm applies an n-dimensional Chebyshev or Fourier transform to cluster pixels based on spectral similarity independent of pixel intensity or location, and a denoising convolution filter is then applied in this spectral space. The denoised image may then undergo spectral decomposition analysis with enhanced accuracy. Tests utilizing both simulated and empirical microscopy data indicate that denoising in 3 to 5-dimensional (3D to 5D) spectral spaces decreases unmixing error by up to 70% without degrading spatial resolution.

Two-photon peak molecular brightness spectra reveal long-wavelength enhancements of multiplexed imaging depth and photostability.

The broad use of two-photon microscopy has been enabled in part by Ti:Sapphire femtosecond lasers, which offer a wavelength-tunable source of pulsed excitation. Action spectra have thus been primarily reported for the tunable range of Ti:Sapphire lasers (∼700-1000 nm). However, longer wavelengths offer deeper imaging in tissue via reduced scattering and spectral dips in water absorption, and new generations of pulsed lasers offer wider tunable ranges. We present the peak molecular brightness spectra for eight Alexa Fluor dyes between 700-1300 nm as a first-order surrogate for action spectra measured with an unmodified commercial microscope, which reveal overlapping long-wavelength excitation peaks with potential for multiplexed excitation. We demonstrate simultaneous single-wavelength excitation of six spectrally overlapping fluorophores using either short (∼790 nm) or long (∼1090 nm) wavelengths, and that the newly characterized excitation peaks measured past 1000 nm offer improved photostability and enhanced fidelity of linear spectral unmixing at depth compared to shorter wavelengths.

High-power light-emitting diode array design and assembly for practical photodynamic therapy research.

SIGNIFICANCE: Commercial lasers, lamps, and light-emitting diode (LED) light sources have stimulated the clinical translation of photodynamic therapy (PDT). Yet, the continued exploration of new photosensitizers (PSs) for PDT often requires separate activation wavelengths for each agent being investigated. Customized light sources for such research frequently come at significant financial or technical cost, especially when compounded over many agents and wavelengths. AIM: LEDs offer potential as a cost-effective tool for new PS and multi-PS photodynamic research. A low-cost-per-wavelength tool leveraging high-power LEDs to facilitate efficient and versatile research is needed to further accelerate research in the field. APPROACH: We developed and validated a high-power LED array system for benchtop PDT with a modular design for efficient switching between wavelengths that overcome many challenges in light source design. We describe the assembly of a low-cost LED module plus the supporting infrastructure, software, and protocols to streamline typical in vitro PDT experimentation. RESULTS: The LED array system is stable at intensities in excess of 100 mW / cm2 with 2.3% variation across the illumination field, competitive with other custom and commercial devices. To demonstrate efficacy and versatility, a primary ovarian cancer cell line was treated with two widely used PSs, aminolevulinic acid and verteporfin, using the LED modules at a clinically relevant 50 J / cm2 light dose that induced over 90% cell death for each treatment. CONCLUSIONS: Our work provides the community with a tool for new PS and multi-PS benchtop photodynamic research that, unlike most commercial light sources, affords the user a low barrier to entry and low-cost-per-wavelength with the goal of illuminating new insights at the forefront of PDT.

Site-specific Bioconjugation and Convergent Click Chemistry Enhances Antibody-Chromophore Conjugate Binding Efficiency.

Photosensitizer (PS)-antibody conjugates (photoimmunoconjugates, PICs) enable cancer cell-targeted photodynamic therapy (PDT). Nonspecific chemical bioconjugation is widely used to synthesize PICs but gives rise to several shortcomings. The conjugates are heterogeneous, and the process is not easily reproducible. Moreover, modifications at or near the binding sites alter both binding affinity and specificity. To overcome these limitations, we introduce convergent assembly of PICs via a chemo-enzymatic site-specific approach. First, an antibody is conjugated to a clickable handle via site-specific modification of glutamine (Gln) residues catalyzed by transglutaminase (TGase, EC 2.3.2.13). Second, the modified antibody intermediate is conjugated to a compatible chromophore via click chemistry. Utilizing cetuximab, we compared this site-specific conjugation protocol to the nonspecific chemical acylation of amines using N-hydroxysuccinimide (NHS) chemistry. Both the heavy and light chains were modified via the chemical route, whereas, only a glutamine 295 in the heavy chain was modified via chemo-enzymatic conjugation. Furthermore, a 2.3-fold increase in the number of bound antibodies per cell was observed for the site-specific compared with nonspecific method, suggesting that multiple stochastic sites of modification perturb the antibody-antigen binding. Altogether, site-specific bioconjugation leads to homogenous, reproducible and well-defined PICs, conferring higher binding efficiency and probability of clinical success.

Multichannel correlation improves the noise tolerance of real-time hyperspectral microimage mosaicking.

Live-subject microscopies, including microendoscopy and other related technologies, offer promise for basic biology research as well as the optical biopsy of disease in the clinic. However, cellular resolution generally comes with the trade-off of a microscopic field-of-view. Microimage mosaicking enables stitching many small scenes together to aid visualization, quantitative interpretation, and mapping of microscale features, for example, to guide surgical intervention. The development of hyperspectral and multispectral systems for biomedical applications provides motivation for adapting mosaicking algorithms to process a number of simultaneous spectral channels. We present an algorithm that mosaics multichannel video by correlating channels of consecutive frames as a basis for efficiently calculating image alignments. We characterize the noise tolerance of the algorithm by using simulated video with known ground-truth alignments to quantify mosaicking accuracy and speed, showing that multiplexed molecular imaging enhances mosaic accuracy by leveraging observations of distinct molecular constituents to inform frame alignment. A simple mathematical model is introduced to characterize the noise suppression provided by a given group of spectral channels, thus predicting the performance of selected subsets of data channels in order to balance mosaic computation accuracy and speed. The characteristic noise tolerance of a given number of channels is shown to improve through selection of an optimal subset of channels that maximizes this model. We also demonstrate that the multichannel algorithm produces higher quality mosaics than the analogous single-channel methods in an empirical test case. To compensate for the increased data rate of hyperspectral video compared to single-channel systems, we employ parallel processing via GPUs to alleviate computational bottlenecks and to achieve real-time mosaicking even for video-rate multichannel systems anticipated in the future. This implementation paves the way for real-time multichannel mosaicking to accompany next-generation hyperspectral and multispectral video microscopy.

Illuminating the Numbers: Integrating Mathematical Models to Optimize Photomedicine Dosimetry and Combination Therapies.

Cancer photomedicine offers unique mechanisms for inducing local tumor damage with the potential to stimulate local and systemic anti-tumor immunity. Optically-active nanomedicine offers these features as well as spatiotemporal control of tumor-focused drug release to realize synergistic combination therapies. Achieving quantitative dosimetry is a major challenge, and dosimetry is fundamental to photomedicine for personalizing and tailoring therapeutic regimens to specific patients and anatomical locations. The challenge of dosimetry is perhaps greater for photomedicine than many standard therapies given the complexity of light delivery and light-tissue interactions as well as the resulting photochemistry responsible for tumor damage and drug-release, in addition to the usual intricacies of therapeutic agent delivery. An emerging multidisciplinary approach in oncology utilizes mathematical and computational models to iteratively and quantitively analyze complex dosimetry, and biological response parameters. These models are parameterized by preclinical and clinical observations and then tested against previously unseen data. Such calibrated and validated models can be deployed to simulate treatment doses, protocols, and combinations that have not yet been experimentally or clinically evaluated and can provide testable optimal treatment outcomes in a practical workflow. Here, we foresee the utility of these computational approaches to guide adaptive therapy, and how mathematical models might be further developed and integrated as a novel methodology to guide precision photomedicine.