The design of this project is the characterization of optical readouts by mapping them to specific metabolic pathways affected by injury.

This can be achieved with a 3-compartments system: cell culture, imaging, and computational model. The objectives are based on each compartment. A specific injury will be induced in monoculture and 3D co-culture and its impact will be examined via microscopy, metabolic assays, and mass spectrometry. Computational models will be created to identify affected metabolic pathways from biochemical data. These objectives can be tested and evaluated. Once a comprehensive cell culture protocol has been developed, we will consider our cultures as viable if they remain stable at passage 3 and are healthy as observed in baseline imaging readouts, which will be compared to imaging readouts from validated cultures in the lab. Mass spectrometry will follow the validated protocols from the Lee lab. The acquisition of optical images will be adapted from the imaging protocols of the Georgakoudi lab. We will evaluate the successful induction of our injury conditions by asserting that trends of optical readouts and mass spectral data converge, indicating that the experimental treatment successfully induced a consistent metabolic shift. As an additional safety net, results from metabolic assays should agree with data from mass spec given both methods measure metabolomics concentrations. 

Multiple engineering principles are applied in this project.

First being two-photon microscopy (TPEF) – an imaging modality for injury assessment. Compared to standard fluorescence microscopy, TPEF utilizes a pulsed, non-linear excitation process where 2 photons are used to excite the fluorophore. By lowering the amount of energy needed per photon, TPEF uses a longer wavelength, which generates less tissue damage and penetrates deeper. Sufficient laser intensity for this excitation is only achievable in the focal plane. This restricts the volume of the signal generation as out-of-focus signals from the planes above and below the focal plane of the sample are removed.

These characteristics make TPEF depth-resolved, facilitating the imaging of thick and highly scattering specimens like engineered brain tissue (EBT) without the need for slicing or biopsy. For this project, endogenous fluorophores such as FAD and NADH will be used so the imaging process is label-free where samples can be live imaged.

There are 2 realistic constraints: ethical concerns and translatability of the computational model.

There are ethical concerns about incurring TBI in human brains or postmortem samples. As a solution, we will use 3D-engineered brain tissues which show pathophysiology observed in an in-vivo model [15]. While there are ethical concerns due to the use of human cells, this is necessary to accurately determine if our results are clinically translatable. Additionally, we plan to use a model of brain metabolism at baseline derived from literature because there is no complete TBI metabolic model. Metabolic model source code is difficult to obtain, and models may be designed based on assumptions and conditions specific to the institution which published that model. It may be a non-trivial task to adapt existing models to assist our project. A solution would be to write our own model based on the key differential equations governing the metabolic processes of interest to us (central metabolism, glutamate-glutamine conversion, and oxidative stress). This would be outside the scope of our capstone but would be doable by masters students.

DESIGN ELEMENT TABLE

Design Elements	Success Measures
LPS – microglia monoculture	Study 1: Optimize injury conditions: – Optical readouts indicate decrease in free NADH and glycolysis shift – Phasor shifts to bottom right – Redox ratio increases – Spectral constituents have an increased NADH concentration
LPS – 3D neuron-microglia co-culture Glutamate injury – neuron-astrocyte 3D co-culture	Study 1: Optimize culture conditions (NM scaffolds only) – No significant difference in optical readouts between Kaplan Lab’s and our scaffolds  Study 2: Optimize injury conditions (NM and NA scaffolds) – Verify that, using statistical analysis, injury occurs – Cell viability decreases after injury – Glutathione assay indicates that glutathione is down-regulated in glutamate excitotoxicity – Glutathione assay indicates an increased oxidized-reduced glutathione ratio for LPS condition (oxidative stress) Study 3: Induce and asses injury – The results of mass spectra and metabolic assays should be consistent with each other – Student’s T-Test shows significantly different peak heights in glutamate and glutamine at 0h and 24h for glutamate excitotoxicity  – The trend of optical readouts is similar to that of monoculture
Computational Model Differential Equations MATLAB Functions	– Review literature for ODE simulation MATLAB format – Establish a set of ODEs and reactions with Km and Vmax values  – Create a MATLAB function using ODE MATLAB solvers to simulate microglia pathways involving LPS, glutamate, glutathione or glutamine. – Analyze the simulated upstream pattern for TBI indicators Verify the output of our function with experimental results

Table 1. Table describing design elements and success measures to validate and verify them.

Design Flowchart