Engineering Tumor-Targeting Bacteria and Characterizing Their Interactions with Tumor Cells in Therapy-Resistant Breast Cancers

Breast cancer (BC) accounts for one-third of malignancies among women in 157 countries and ~15% mortality among diagnosed cases, a burden projected to reach 1.1 million deaths per year by 2050. Five-year survival drops from >90% in early-stage BC to ~32% for therapy-resistant subtypes such as triple-negative (TNBC), hormone-receptor–variable or resistant ER+, and Luminal B tumors. For these high-risk subtypes, molecular heterogeneity undermines targeted therapies, and clinical management still relies on maximum tolerable doses of systemic chemotherapy, causing severe dose-limiting toxicities. Moreover, the dense collagen-rich extracellular matrix (ECM) of solid tumors restricts intratumoral drug transport, motivating strategies that function across BC subtypes, overcome ECM barriers, and inform lowered clinical dosing. Bacteria-based cancer therapy (BBCT) with cancer-selective bacteria combines motility, self-replication, and on-board biosynthesis with the programmability of synthetic biology, enabling local release of therapeutic factors within the tumor microenvironment. Attenuated Salmonella Typhimurium VNP20009 (ST) exhibits ~10³-10⁴-fold tumor selectivity with respect to liver and spleen. It has a favorable clinical safety profile but remains inefficacious due to poor colonization. Our lab previously showed that ECM-targeting ST with collagenase secretion improved tumor penetration without gross collagen disruption, but at the cost of reduced bacterial fitness and motility. In this dissertation, we hypothesized that a fitness-restored, ECM-targeting ST enhances bacterial intratumoral transport and colonization, as well as chemotherapy penetration. We evaluated the engineered strains in perfused 3D tumor models that represent in vivo intratumor transport properties and investigated cancer cells-neutrophil interactions in presence of bacterial factors. First, we developed a high-motility, fitness-improved collagenase-expressing strain (HM-CEST ΔydcP) that preserves 100% motility under sub-cytotoxic chemotherapy. This strain improves intratumoral transport, and reduces spheroid viability and tumor migration relative to chemotherapy alone while maintaining tumor specificity and safety in preclinical murine models in vivo. Second, we validated a perfusion-enabled microfluidic spheroid platform that supports at least 14-day culture of murine TNBC and ER+ spheroids, enabling long-term BBCT screening. We demonstrated that perfused spheroids require lower chemotherapy doses than static cultures. Third, we biophysically characterized the crosstalk between BBCT, neutrophils, and Luminal B BC cells, demonstrating that neutrophils in the presence of bacteria supernatant suppress cancer cell growth, viability, and migration. Collectively, this work delivers a fitness-restored ECM-targeting Salmonella chassis, validates a perfusion-enabled 3D tumor-spheroid microphysiological platform, and develops a quantitative framework for neutrophil–bacteria–cancer cell interactions, contributing to the design pipeline for next-generation BBCT and supporting the long-term goal of safer, more affordable, lower-dose, and more broadly applicable therapies for therapy-resistant breast cancers.

Name thesis:degree_name

Doctor of Philosophy

Level thesis:degree_level

doctoral

Discipline thesis:degree_discipline

Macromolecular Science and Engineering

Department dc:contributor.department

Graduate School

Grantor dc:publisher

Virginia Tech

Year dc:date.issued

2026

Author dc:creator

• Saha, Binita

Chair dc:contributor.committeechair

• Behkam, Bahareh

Committee members dc:contributor.committeemember

• Li, Liwu
• Bortner, Michael J.
• Johnson, Blake
• Senger, Ryan S.

• Breast cancer
• microphysiological 3D cancer models
• combinatorial therapy
• bacteria-based cancer therapy
• extra cellular matrix targeting
• immunotherapy

Statement dc:rights

• Creative Commons Attribution 4.0 International

Licence dc:rights.uri

• http://creativecommons.org/licenses/by/4.0/

Language dc:language.iso

en

Dc Identifier Other

vt_gsexam:45381

OAI identifier oai:identifier

oai:vtechworks.lib.vt.edu:10919/140686