Date of Award

January 2023

Document Type


Degree Name

Medical Doctor (MD)



First Advisor

Julius Chapiro


In this study, we investigated the effects of systemic glycolysis inhibition with 2-deoxy-D-glucose (2-DG), a glucose analogue that inhibits hexokinase, in the Hepa1-6 orthotopic mouse model of hepatocellular carcinoma (HCC) and characterized changes in metabolism using biosensor imaging of redundant deviation in shifts (BIRDS) for extracellular pH (pHe) mapping as well as immunohistochemistry (IHC).

In vitro efficacy of 2-DG was tested in the Hepa1-6 murine HCC cell line via the Promega CellTiter-Glo luminescent cell viability assay and the Abcam colorimetric assay of hexokinase activity from cell lysates. For an in vivo assessment of 2-DG, ten C57BL/6 mice underwent orthotopic subcapsular implantation of Hepa1-6 HCC tumor cells. After liver tumors were growth for 1 week, mice were treated with daily intraperitoneal injections of either control (N = 5) or 2000 mg/kg 2-DG daily (N = 5) for 1 week. Mouse livers and tumors were then harvested and underwent IHC-staining for hexokinase 2 (HK2), lysosomal-associated membrane protein 2 (LAMP2), and lactate dehydrogenase A (LDHA). For BIRDS pHe experiments, six C57BL/6J mice were implanted with Hepa1-6 tumors. After 1 week of tumor growth, mice were treated with daily intraperitoneal injections of either control (N = 3) or 2000 mg/kg 2-DG (N = 3) for 1 week. On Day 7 of treatment, mice were subject to BIRDS pHe imaging on a horizontal-bore 9.4 Tesla Bruker scanner. A MATLAB script was used to analyze chemical shifts and generate pHe maps of tumors and livers. After scanning IHC slides using the Aperio Scope XT Digital Slide Scanner, quantitative IHC analysis was achieved with the Aperio Positive Pixel Count Algorithm to assess strong and medium intensities of IHC staining. Comparisons of two independent groups were analyzed using an unpaired t-test with Welch’s correction. Comparisons of more than two independent groups were done using a Brown-Forsythe and Welch ANOVA test with Dunnett’s T3 or Games-Howell’s multiple comparisons test. A 2-tailed P-value of <0.05 was considered statistically significant.

Hepa1-6 exhibited greater than 80% viability across a 1000-fold range of 2-DG concentrations and a IC50 value of 111.5 mM. At a 2-DG concentration corresponding to 75% viability, no significant differences in hexokinase activity were noted. When pooling all pHe voxels for each experimental group that underwent BIRDS imaging, 2-DG-treated mice had a significantly lower liver pHe (6.806 ± 0.311) than control-treated mice (6.905 ± 0.290, P = <0.0001). However, tumor pHe of 2-DG-treated and control-mice were similar [6.880 ± 0.352 (2-DG) vs. 6.828 ± 0.257 (control), P = 0.1720]. These changes in pHe gradients were accompanied by a significant increase in medium intensity of LDHA expression, and non-significant increases in strong intensities of HK2 and LDHA expression.

In conclusion, we demonstrated Hepa1-6 was resistant to 2-DG in vitro and established a safe 2-DG dose that could be administered in vivo to examine effects on IHC metabolic markers and BIRDS pHe imaging. We discovered that pharmacological manipulation of glycolysis could be detected with BIRDs pHe imaging. Notably, 2-DG causes a significant pHe decrease within the hepatic parenchyma, a finding that is essential for future models that seek to evaluate metabolic changes in the tumor microenvironment. Ultimately, these findings have implications for the potential of non-invasive MRI pHe mapping to evaluate the dynamic response to therapy at high spatiotemporal resolution.


This thesis is restricted to Yale network users only. This thesis is permanently embargoed from public release.