Controlled delivery of foreign cargo into cells is a critical step in many biological studies and in cell engineering and analysis workflows. Recent advances in micro and nanotechnology, specifically in microfluidics and microfabrication have added significantly to the precision, accuracy, resolution and throughput of cell manipulation and analysis pipelines. These improvements have enabled efficient molecular delivery into cells with minimal cell damage compared to traditional methods of intracellular delivery. Moreover, the capability of both high throughput and single cell manipulation and analysis has facilitated the design and execution of experimental workflows catered towards applications like gene editing, study of disease pathology and drug development without losing the heterogeneity of single cell information.Over the last decade localized electroporation has emerged as one of the most promising modern techniques of intracellular delivery. It has been shown to be capable of delivering exogenous molecular cargo into cells with higher delivery efficiency and precision than commercial techniques while preserving cell health and function. Although the proof of concept has been demonstrated in several studies, there are still a few knowledge gaps and engineering challenges that is preventing its widespread scaling up as a commercially available technology. First, the efficacy of molecular delivery via localized electroporation depends on a multitude of experiments factors like pulse voltage, pulse duration, electroporation buffer, cargo concentration and cell type to name a few and arriving at the optimized conditions requires several rounds of experiments. There is a lack of experimental demonstration of a systematic way to enable rapid optimization of intracellular delivery for a particular application. Secondly, localized electroporation involves creating an electric field induced temporary permeabilized state of the cell membrane which reseals after the electric pulse application. Therefore, the amount of cargo delivered/extracted also depends on the time it takes for the cell membrane to reseal post electroporation. The knowledge of how the temporal dynamics of the cell membrane resealing can be controlled by experimental variables like pulse voltage and number of pulses can help in the design of experiments ensuring optimal delivery/sampling while keeping cell toxicity to a minimum. Thirdly, until now localized electroporation has mostly been used for delivery of highly charged moieties like plasmid DNA or smaller molecules like mRNA. There is a lack of studies showing the optimization of localized electroporation for delivery of large proteins, which are not highly charged. Additionally, depending on the nature of the cargo there can be differences in their intracellular localization post-delivery which can be important for applications like intracellular sensing. The literature is devoid of such investigations in the context of localized electroporation. In this thesis, we will seek to fill some of the knowledge gaps and address some of the challenges about designing a localized electroporation-based cell engineering and analysis workflow that can be potentially used in a clinical setting. We first developed a Multiphysics model to aid in guiding the experimental design of delivering proteins and plasmids using a localized electroporation based single cell manipulation probe known as the Nano fountain Probe electroporation (NFP-E) system. We then used the same system to investigate the dynamics of cell membrane resealing after localized electroporation and developed an analytical model to explain the variation of molecular transport across the cell membrane with respect to pulse parameters. Next to address the issue of lack of a systematic way to optimize localized electroporation-based delivery, we first designed and developed a localized electroporation based high throughput, multiplexed intracellular delivery platform known as the 24-well LEPD system. We combined the platform with a custom-built deep learning-based image analysis pipeline to quickly and accurately quantify image based experimental data. The 24-well LEPD workflow enabled both multiplexing of delivery experiments, quantification of multiple intensity and cell morphology-based measurements, leading to quick identification of optimal delivery conditions. We demonstrated multi-cargo delivery in several cell types with higher efficiency and viability compared to some commercial systems. Moreover, we also demonstrated the clinical potential of the platform by using the 24-well LEPD to genetically manipulate human induced pluripotent stem cells (hiPSCs) via siRNA mediated protein knockdown. Overall, the workflow developed and presented here can be an inspiration to further optimize intracellular molecular delivery in highly sensitive and difficult to transfect clinically relevant/patient derived cell types for applications in areas like in-vitro disease modeling and personalized drug development. In the intracellular delivery focused section of the thesis, we used a machine learning based pipeline to help quickly arrive at optimal experimental conditions for intracellular delivery. In the latter part of this thesis, we built a machine learning based workflow to tackle the challenge of inverse design of Kirigami-based metamaterials. Kirigami-engineering has become an avenue for realizing multifunctional metamaterials that tap into the instability landscape of planar surfaces embedded with cuts. Recently, it has been shown that two-dimensional Kirigami motifs can unfurl a rich space of out-of-plane deformations, which are programmable and controllable across spatial scales. Notwithstanding Kirigami’s versatility, arriving at a cut layout that yields the desired functionality remains a challenge (Inverse design). Here, we introduce a comprehensive machine learning framework to shed light on the Kirigami design space and to rationally guide the design and control of Kirigami-based materials from the meta-atom to the metamaterial level. We employ a combination of clustering, tandem neural networks, and symbolic regression analyses to obtain Kirigami that fulfills specific design constraints and inform on their control and deployment. Our systematic approach is experimentally demonstrated by examining a variety of applications at different hierarchical levels, effectively providing a tool for the discovery of shape shifting Kirigami metamaterials.

Creator

• Pathak, Nibir

DOI

• https://doi.org/10.21985/n2-ws9n-bn49

Subject

• Mechanical engineering

Language

• en

Alternate Identifier

• http://dissertations.umi.com/northwestern:16641
• etdadmin_upload_997980

Date created

• 2023-01-01

Resource type

• Dissertation

Rights statement

• In Copyright