Neural manifold analysis of brain circuit dynamics in health and disease.

Recent developments in experimental neuroscience make it possible to simultaneously record the activity of thousands of neurons. However, the development of analysis approaches for such large-scale neural recordings have been slower than those applicable to single-cell experiments. One approach that has gained recent popularity is neural manifold learning. This approach takes advantage of the fact that often, even though neural datasets may be very high dimensional, the dynamics of neural activity tends to traverse a much lower-dimensional space. The topological structures formed by these low-dimensional neural subspaces are referred to as "neural manifolds", and may potentially provide insight linking neural circuit dynamics with cognitive function and behavioral performance. In this paper we review a number of linear and non-linear approaches to neural manifold learning, including principal component analysis (PCA), multi-dimensional scaling (MDS), Isomap, locally linear embedding (LLE), Laplacian eigenmaps (LEM), t-SNE, and uniform manifold approximation and projection (UMAP). We outline these methods under a common mathematical nomenclature, and compare their advantages and disadvantages with respect to their use for neural data analysis. We apply them to a number of datasets from published literature, comparing the manifolds that result from their application to hippocampal place cells, motor cortical neurons during a reaching task, and prefrontal cortical neurons during a multi-behavior task. We find that in many circumstances linear algorithms produce similar results to non-linear methods, although in particular cases where the behavioral complexity is greater, non-linear methods tend to find lower-dimensional manifolds, at the possible expense of interpretability. We demonstrate that these methods are applicable to the study of neurological disorders through simulation of a mouse model of Alzheimer's Disease, and speculate that neural manifold analysis may help us to understand the circuit-level consequences of molecular and cellular neuropathology.

Development of T-cell engagers selective for cells co-expressing two antigens.

T cell-engaging bispecific antibodies (TCEs) are clinically effective treatments for hematological cancers. While the utility of TCEs in solid malignancies is being explored, toxicities arising from antigen expression on normal tissues have slowed or halted several clinical trials. Here, we describe the development of TCEs that preferentially drive T cell-mediated death against target cells co-expressing two tumor-associated antigens. We show that Ly6E and B7-H4 are simultaneously expressed on approximately 50% of breast cancers, whereas normal tissue expression is limited and mostly orthogonal. Traditional bispecific TCEs targeting a singular antigen, either Ly6E or B7-H4, are active when paired with high-affinity CD3-engagers, but normal tissue expression presents a toxicity risk. Treatment with a murine cross-reactive B7-H4-TCE results in rapid and severe weight loss in mice along with damage to B7-H4-expressing tissues. To overcome on-target toxicity, we designed trispecific antibodies co-targeting Ly6E, B7-H4, and CD3 and characterized the impact of dual-antigen binding and the relative placement of each binding domain on tumor killing in vitro and in vivo. In vitro killing of tumor cells co-expressing both antigens correlates to the placement of the higher affinity B7-H4 binding domain, with only modest enhancements seen upon addition of Ly6E binding. In xenograft models, avid binding of appropriately designed trispecific TCEs enables tumor growth inhibition while evading the poor tolerability seen with active bispecific TCEs. Collectively these data highlight the potential for dual-antigen targeting to improve safety and efficacy, and expand the scope of tumors that may effectively be treated by TCEs.Abbreviations: Chimeric antigen receptor T cells (CAR-Ts), dual-antigen targeted T cell engagers (DAT-TCE), Fragment antigen-binding (Fab), Hematoxylin and eosin (H&E), Institutional Animal Care and Use Committee (IACUC), Immunoglobulin G (IgG), immunohistochemistry (IHC), NOD SCID gamma (NSG), peripheral blood mononuclear cells (PBMCs), surface plasmon resonance (SPR), T cell-engagers (TCEs).

Valency of HER2 Targeting Antibodies Influences Tumor Cell Internalization and Penetration.

T-cell-dependent bispecific antibodies (TDB) have been a major advancement in the treatment of cancer, allowing for improved targeting and efficacy for large molecule therapeutics. TDBs are comprised of one arm targeting a surface antigen on a cancer cell and another targeting an engaging surface antigen on a cytotoxic T cell. To impart this function, the antibody must be in a bispecific format as opposed to the more conventional bivalent format. Through in vitro and in vivo studies, we sought to determine the impact of changing antibody valency on solid tumor distribution and catabolism. A bivalent anti-HER2 antibody exhibited higher catabolism than its full-length monovalent binding counterpart in vivo by both invasive tissue harvesting and noninvasive single photon emission computed tomography/X-ray computed tomography imaging despite similar systemic exposures for the two molecules. To determine what molecular factors drove in vivo distribution and uptake, we developed a mechanistic model for binding and catabolism of monovalent and bivalent HER2 antibodies in KPL4 cells. This model suggests that observed differences in cellular uptake of monovalent and bivalent antibodies are caused by the change in apparent affinity conferred by avidity as well as differences in internalization and degradation rates of receptor bound antibodies. To our knowledge, this is the first study to directly compare the targeting abilities of monovalent and bivalent full-length antibodies. These findings may inform diverse antibody therapeutic modalities, including T-cell-redirecting therapies and drug delivery strategies relying upon receptor internalization.

Place Cells in Head-Fixed Mice Navigating a Floating Real-World Environment.

The hippocampal place cell system in rodents has provided a major paradigm for the scientific investigation of memory function and dysfunction. Place cells have been observed in area CA1 of the hippocampus of both freely moving animals, and of head-fixed animals navigating in virtual reality environments. However, spatial coding in virtual reality preparations has been observed to be impaired. Here we show that the use of a real-world environment system for head-fixed mice, consisting of an air-floating track with proximal cues, provides some advantages over virtual reality systems for the study of spatial memory. We imaged the hippocampus of head-fixed mice injected with the genetically encoded calcium indicator GCaMP6s while they navigated circularly constrained or open environments on the floating platform. We observed consistent place tuning in a substantial fraction of cells despite the absence of distal visual cues. Place fields remapped when animals entered a different environment. When animals re-entered the same environment, place fields typically remapped over a time period of multiple days, faster than in freely moving preparations, but comparable with virtual reality. Spatial information rates were within the range observed in freely moving mice. Manifold analysis indicated that spatial information could be extracted from a low-dimensional subspace of the neural population dynamics. This is the first demonstration of place cells in head-fixed mice navigating on an air-lifted real-world platform, validating its use for the study of brain circuits involved in memory and affected by neurodegenerative disorders.

Calicheamicin Antibody-Drug Conjugates with Improved Properties.

Calicheamicin antibody-drug conjugates (ADCs) are effective therapeutics for leukemias with two recently approved in the United States: Mylotarg (gemtuzumab ozogamicin) targeting CD33 for acute myeloid leukemia and Besponsa (inotuzumab ozogamicin) targeting CD22 for acute lymphocytic leukemia. Both of these calicheamicin ADCs are heterogeneous, aggregation-prone, and have a shortened half-life due to the instability of the acid-sensitive hydrazone linker in circulation. We hypothesized that we could improve upon the heterogeneity, aggregation, and circulation stability of calicheamicin ADCs by directly attaching the thiol of a reduced calicheamicin to an engineered cysteine on the antibody via a disulfide bond to generate a linkerless and traceless conjugate. We report herein that the resulting homogeneous conjugates possess minimal aggregation and display high in vivo stability with 50% of the drug remaining conjugated to the antibody after 21 days. Furthermore, these calicheamicin ADCs are highly efficacious in mouse models of both solid tumor (HER2+ breast cancer) and hematologic malignancies (CD22+ non-Hodgkin lymphoma). Safety studies in rats with this novel calicheamicin ADC revealed an increased tolerability compared with that reported for Mylotarg. Overall, we demonstrate that applying novel linker chemistry with site-specific conjugation affords an improved, next-generation calicheamicin ADC.

Systematic Variation of Pyrrolobenzodiazepine (PBD)-Dimer Payload Physicochemical Properties Impacts Efficacy and Tolerability of the Corresponding Antibody-Drug Conjugates.

Cytotoxic pyrrolobenzodiazepine (PBD)-dimer molecules are frequently utilized as payloads for antibody-drug conjugates (ADCs), and many examples are currently in clinical development. In order to further explore this ADC payload class, the physicochemical properties of various PBD-dimer molecules were modified by the systematic introduction of acidic and basic moieties into their chemical structures. The impact of these changes on DNA binding, cell membrane permeability, and in vitro antiproliferation potency was, respectively, determined using a DNA alkylation assay, PAMPA assessments, and cell-based cytotoxicity measurements conducted with a variety of cancer lines. The modified PBD-dimer compounds were subsequently incorporated into CD22-targeting ADCs, and these entities were profiled in a variety of in vitro and in vivo experiments. The introduction of a strongly basic moiety into the PBD-dimer scaffold afforded a conjugate with dramatically worsened mouse tolerability properties relative to ADCs derived from related payloads, which lacked the basic group.

Evaluation and use of an anti-cynomolgus monkey CD79b surrogate antibody-drug conjugate to enable clinical development of polatuzumab vedotin.

BACKGROUND AND PURPOSE: Polatuzumab vedotin is an antibody-drug conjugate (ADC) being developed for non-Hodgkin's lymphoma. It contains a humanized anti-CD79b IgG1 monoclonal antibody linked to monomethyl auristatin E (MMAE), an anti-mitotic agent. Polatuzumab vedotin binds to human CD79b only. Therefore, a surrogate ADC that binds to cynomolgus monkey CD79b was used to determine CD79b-mediated pharmacological effects in the monkey and to enable first-in-human clinical trials. EXPERIMENTAL APPROACH: Polatuzumab vedotin, the surrogate ADC, and the corresponding antibodies were evaluated in different assays in vitro and in animals. In vitro assessments included binding to peripheral blood mononuclear cells from different species, binding to a human and monkey CD79b-expressing cell line, binding to human Fcγ receptors, and stability in plasma across species. In vivo, ADCs were assessed for anti-tumour activity in mice, pharmacokinetics/pharmacodynamics in monkeys, and toxicity in rats and monkeys. KEY RESULTS: Polatuzumab vedotin and surrogate ADC bind with similar affinity to human and cynomolgus monkey B cells, respectively. Comparable in vitro plasma stability, in vivo anti-tumour activity, and mouse pharmacokinetics were also observed between the surrogate ADC and polatuzumab vedotin. In monkeys, only the surrogate ADC showed B-cell depletion and B-cell-mediated drug disposition, but both ADCs showed similar MMAE-driven myelotoxicity, as expected. CONCLUSIONS AND IMPLICATIONS: The suitability of the surrogate ADC for evaluation of CD79b-dependent pharmacology was demonstrated, and anti-tumour activity, pharmacokinetics/pharmacodynamics, and toxicity data with both ADCs supported the entry of polatuzumab vedotin into clinical trials.

A compact holographic projector module for high-resolution 3D multi-site two-photon photostimulation.

Patterned two-photon (2P) photolysis via holographic illumination is a powerful method to investigate neuronal function because of its capability to emulate multiple synaptic inputs in three dimensions (3D) simultaneously. However, like any optical system, holographic projectors have a finite space-bandwidth product that restricts the spatial range of patterned illumination or field-of-view (FOV) for a desired resolution. Such trade-off between holographic FOV and resolution restricts the coverage within a limited domain of the neuron's dendritic tree to perform highly resolved patterned 2P photolysis on individual spines. Here, we integrate a holographic projector into a commercial 2P galvanometer-based 2D scanning microscope with an uncaging unit and extend the accessible holographic FOV by using the galvanometer scanning mirrors to reposition the holographic FOV arbitrarily across the imaging FOV. The projector system utilizes the microscope's built-in imaging functions. Stimulation positions can be selected from within an acquired 3D image stack (the volume-of-interest, VOI) and the holographic projector then generates 3D illumination patterns with multiple uncaging foci. The imaging FOV of our system is 800×800 µm2 within which a holographic VOI of 70×70×70 µm3 can be chosen at arbitrary positions and also moved during experiments without moving the sample. We describe the design and alignment protocol as well as the custom software plugin that controls the 3D positioning of stimulation sites. We demonstrate the neurobiological application of the system by simultaneously uncaging glutamate at multiple spines within dendritic domains and consequently observing summation of postsynaptic potentials at the soma, eventually resulting in action potentials. At the same time, it is possible to perform two-photon Ca2+ imaging in 2D in the dendrite and thus to monitor synaptic Ca2+ entry in selected spines and also local regenerative events such as dendritic action potentials.

An Anti-GDNF Family Receptor Alpha 1 (GFRA1) Antibody-Drug Conjugate for the Treatment of Hormone Receptor-Positive Breast Cancer.

Luminal A (hormone receptor-positive) breast cancer constitutes 70% of total breast cancer patients. In an attempt to develop a targeted therapeutic for this cancer indication, we have identified and characterized Glial cell line-Derived Neurotrophic Factor (GDNF) Family Receptor Alpha 1 (GFRA1) antibody-drug conjugates (ADC) using a cleavable valine-citrulline-MMAE (vcMMAE) linker-payload. RNAseq and IHC analysis confirmed the abundant expression of GFRA1 in luminal A breast cancer tissues, whereas minimal or no expression was observed in most normal tissues. Anti-GFRA-vcMMAE ADC internalized to the lysosomes and exhibited target-dependent killing of GFRA1-expressing cells both in vitro and in vivo The ADCs using humanized anti-GFRA1 antibodies displayed robust therapeutic activity in clinically relevant cell line-derived (MCF7 and KPL-1) tumor xenograft models. The lead anti-GFRA1 ADC cross-reacts with rodent and cynomolgus monkey GFRA1 antigen and showed optimal pharmacokinetic properties in both species. These properties subsequently enabled a target-dependent toxicity study in rats. Anti-GFRA1 ADC is well tolerated in rats, as seen with other vcMMAE linker-payload based ADCs. Overall, these data suggest that anti-GFRA1-vcMMAE ADC may provide a targeted therapeutic opportunity for luminal A breast cancer patients. Mol Cancer Ther; 17(3); 638-49. ©2017 AACR.

Multiphoton minimal inertia scanning for fast acquisition of neural activity signals.

OBJECTIVE: Multi-photon laser scanning microscopy provides a powerful tool for monitoring the spatiotemporal dynamics of neural circuit activity. It is, however, intrinsically a point scanning technique. Standard raster scanning enables imaging at subcellular resolution; however, acquisition rates are limited by the size of the field of view to be scanned. Recently developed scanning strategies such as travelling salesman scanning (TSS) have been developed to maximize cellular sampling rate by scanning only select regions in the field of view corresponding to locations of interest such as somata. However, such strategies are not optimized for the mechanical properties of galvanometric scanners. We thus aimed to develop a new scanning algorithm which produces minimal inertia trajectories, and compare its performance with existing scanning algorithms. APPROACH: We describe here the adaptive spiral scanning (SSA) algorithm, which fits a set of near-circular trajectories to the cellular distribution to avoid inertial drifts of galvanometer position. We compare its performance to raster scanning and TSS in terms of cellular sampling frequency and signal-to-noise ratio (SNR). MAIN RESULTS: Using surrogate neuron spatial position data, we show that SSA acquisition rates are an order of magnitude higher than those for raster scanning and generally exceed those achieved by TSS for neural densities comparable with those found in the cortex. We show that this result also holds true for in vitro hippocampal mouse brain slices bath loaded with the synthetic calcium dye Cal-520 AM. The ability of TSS to 'park' the laser on each neuron along the scanning trajectory, however, enables higher SNR than SSA when all targets are precisely scanned. Raster scanning has the highest SNR but at a substantial cost in number of cells scanned. To understand the impact of sampling rate and SNR on functional calcium imaging, we used the Cramér-Rao Bound on evoked calcium traces recorded simultaneously with electrophysiology traces to calculate the lower bound estimate of the spike timing occurrence. SIGNIFICANCE: The results show that TSS and SSA achieve comparable accuracy in spike time estimates compared to raster scanning, despite lower SNR. SSA is an easily implementable way for standard multi-photon laser scanning systems to gain temporal precision in the detection of action potentials while scanning hundreds of active cells.

Modulating Therapeutic Activity and Toxicity of Pyrrolobenzodiazepine Antibody-Drug Conjugates with Self-Immolative Disulfide Linkers.

A novel disulfide linker was designed to enable a direct connection between cytotoxic pyrrolobenzodiazepine (PBD) drugs and the cysteine on a targeting antibody for use in antibody-drug conjugates (ADCs). ADCs composed of a cysteine-engineered antibody were armed with a PBD using a self-immolative disulfide linker. Both the chemical linker and the antibody site were optimized for this new bioconjugation strategy to provide a highly stable and efficacious ADC. This novel disulfide ADC was compared with a conjugate containing the same PBD drug, but attached to the antibody via a peptide linker. Both ADCs had similar efficacy in mice bearing human tumor xenografts. Safety studies in rats revealed that the disulfide-linked ADC had a higher MTD than the peptide-linked ADC. Overall, these data suggest that the novel self-immolative disulfide linker represents a valuable way to construct ADCs with equivalent efficacy and improved safety. Mol Cancer Ther; 16(5); 871-8. ©2017 AACR.

Targeted pruning of a neuron's dendritic tree via femtosecond laser dendrotomy.

Neurons are classified according to action potential firing in response to current injection. While such firing patterns are shaped by the composition and distribution of ion channels, modelling studies suggest that the geometry of dendritic branches also influences temporal firing patterns. Verifying this link is crucial to understanding how neurons transform their inputs to output but has so far been technically challenging. Here, we investigate branching-dependent firing by pruning the dendritic tree of pyramidal neurons. We use a focused ultrafast laser to achieve highly localized and minimally invasive cutting of dendrites, thus keeping the rest of the dendritic tree intact and the neuron functional. We verify successful dendrotomy via two-photon uncaging of neurotransmitters before and after dendrotomy at sites around the cut region and via biocytin staining. Our results show that significantly altering the dendritic arborisation, such as by severing the apical trunk, enhances excitability in layer V cortical pyramidal neurons as predicted by simulations. This method may be applied to the analysis of specific relationships between dendritic structure and neuronal function. The capacity to dynamically manipulate dendritic topology or isolate inputs from various dendritic domains can provide a fresh perspective on the roles they play in shaping neuronal output.

Four-dimensional multi-site photolysis of caged neurotransmitters.

Neurons receive thousands of synaptic inputs that are distributed in space and time. The systematic study of how neurons process these inputs requires a technique to stimulate multiple yet highly targeted points of interest along the neuron's dendritic tree. Three-dimensional multi-focal patterns produced via holographic projection combined with two-photon photolysis of caged compounds can provide for highly localized release of neurotransmitters within each diffraction-limited focus, and in this way emulate simultaneous synaptic inputs to the neuron. However, this technique so far cannot achieve time-dependent stimulation patterns due to fundamental limitations of the hologram-encoding device and other factors that affect the consistency of controlled synaptic stimulation. Here, we report an advanced technique that enables the design and application of arbitrary spatio-temporal photostimulation patterns that resemble physiological synaptic inputs. By combining holographic projection with a programmable high-speed light-switching array, we have overcome temporal limitations with holographic projection, allowing us to mimic distributed activation of synaptic inputs leading to action potential generation. Our experiments uniquely demonstrate multi-site two-photon glutamate uncaging in three dimensions with submillisecond temporal resolution. Implementing this approach opens up new prospects for studying neuronal synaptic integration in four dimensions.

Simultaneous multi-site two-photon photostimulation in three dimensions.

We demonstrate simultaneous multi-site two-photon photolysis of caged neurotransmitters with close to diffraction-limited resolution in all three dimensions (3D). We use holographic projection of multiple focal spots, which allows full control over the 3D positions of uncaging sites with a high degree of localized excitation. Our system incorporates a two-photon imaging setup to visualize the 3D morphology of the neurons in order to accurately determine the photostimulation sites. We show its application to studies of synaptic integration by performing simultaneous and controlled glutamate delivery at multiple locations on dendritic trees.

Optimal complex field holographic projection.

We describe a technique that uses complex field holograms to project three-dimensional light patterns. Holographic projection commonly uses phase-only encoding since accurately representing complex holograms using both amplitude and phase spatial light modulators reduces the optical throughput significantly. Here, we use a lossless projection via the generalized phase contrast method to produce the necessary amplitude pattern required for complex field holographic projection. We numerically evaluate the technique and demonstrate high optical throughput with reduced undesired high diffraction orders.

Quantifying antivascular effects of monoclonal antibodies to vascular endothelial growth factor: insights from imaging.

PURPOSE: Little is known concerning the onset, duration, and magnitude of direct therapeutic effects of anti-vascular endothelial growth factor (VEGF) therapies. Such knowledge would help guide the rational development of targeted therapeutics from bench to bedside and optimize use of imaging technologies that quantify tumor function in early-phase clinical trials. EXPERIMENTAL DESIGN: Preclinical studies were done using ex vivo microcomputed tomography and in vivo ultrasound imaging to characterize tumor vasculature in a human HM-7 colorectal xenograft model treated with the anti-VEGF antibody G6-31. Clinical evaluation was by quantitative magnetic resonance imaging in 10 patients with metastatic colorectal cancer treated with bevacizumab. RESULTS: Microcomputed tomography experiments showed reduction in perfused vessels within 24 to 48 h of G6-31 drug administration (P

Therapeutic potential of an anti-CD79b antibody-drug conjugate, anti-CD79b-vc-MMAE, for the treatment of non-Hodgkin lymphoma.

Here we describe the generation of an antibody-drug conjugate (ADC) consisting of a humanized anti-CD79b antibody that is conjugated to monomethylauristatin E (MMAE) through engineered cysteines (THIOMABs) by a protease cleavable linker. By using flow cytometry, we detected the surface expression of CD79b in almost all non-Hodgkin lymphoma (NHL) and chronic lymphocytic leukemia patients, suggesting that anti-CD79b-vcMMAE could be widely used in these malignancies. By using NHL cell lines to simulate a patient population we discovered that a minimal cell-surface expression level of CD79b was required for in vitro activity. Within the subpopulation of cell lines above this minimal threshold, we found that sensitivity to free MMAE, mutation of cancer genes, and cell doubling time were poorly correlated with in vitro activity; however, the expression level of BCL-XL was correlated with reduced sensitivity to anti-CD79b-vcMMAE. This observation was supported by in vivo data showing that a Bcl-2 family inhibitor, ABT-263, strikingly enhanced the activity of anti-CD79b-vcMMAE. Furthermore, anti-CD79b-vcMMAE was significantly more effective than a standard-of-care regimen, R-CHOP (ie, rituximab with a single intravenous injection of 30 mg/kg cyclophosphamide, 2.475 mg/kg doxorubicin, 0.375 mg/kg vincristine, and oral dosing of 0.15 mg/kg prednisone once a day for 5 days), in 3 xenograft models of NHL. Together, these data suggest that anti-CD79b-vcMMAE could be broadly efficacious for the treatment of NHL.

Quantification of viable tumor microvascular characteristics by multispectral analysis.

Tumor heterogeneity complicates the quantification of tumor microvascular characteristics assessed by dynamic contrast-enhanced MRI (DCE-MRI). To address this issue a novel approach was developed that combines DCE-MRI with diffusion-based multispectral (MS) analysis to quantify the microvascular characteristics of specific tumor tissue populations. Diffusion-based MS segmentation (feature space: apparent diffusion coefficient, T(2) and proton density) was performed to identify tumor tissue populations and the DCE-MRI characteristics were determined for each tissue class. The ability of this MS DCE-MRI technique to detect microvascular changes due to treatment with an antibody (G6-31) to vascular endothelial growth factor-A (VEGF) was evaluated in a tumor xenograft mouse model. Anti-VEGF treatment resulted in a significant reduction in K(trans) for the MS viable tumor tissue class (-0.0034 +/- 0.0022 min(-1), P < 0.01) at 24 hr posttreatment that differ significantly from the change observed in the control group (0.0002 +/- 0.0025 min(-1)). Viable tumor K(trans) for the anti-VEGF group was also reduced 62% relative to the pretreatment values (P < 0.01). Necrotic tissue classes were found to add only noise to DCE-MRI estimates. This approach provides a means to measure physiological parameters within the viable tumor and address the issue of tumor heterogeneity that complicates DCE-MRI analysis.