Development of 4T1 breast cancer mouse model system for preclinical carbonic anhydrase IX studies

Abstract

Triple-negative breast cancer (TNBC) is the most aggressive type of breast cancer, for which targeted treatment is currently lacking. Carbonic anhydrase IX (CAIX) is a known cancer target due to its selective overexpression in hypoxia, a hallmark of many solid cancers including TNBC. This study aimed to develop a robust murine TNBC cell line 4T1-based model system that could be used in the comprehensive preclinical evaluation of targeting CAIX. The model is based on the original 4T1 breast cancer cell line and two genetically edited versions of it-one with biallelic CRISPR/Cas9-mediated Car9 inactivation and another with constitutively expressed Car9, thus ensuring negative and positive controls for CAIX production in the model system, respectively. The generated cell lines were validated for CAIX production and characterised functionally in vitro and in vivo after orthotopic implantation in syngeneic BALB/c mice. Results demonstrated significantly reduced primary tumour growth and metastatic progression rates in animals with CAIX-deficient tumours, while the CAIX-expressing tumours had vascularised phenotypes with prominent central areas of coagulative necrosis. The differential CAIX expression levels in the model were preserved during tumour growth in syngeneic mice, as verified by in vivo imaging using a novel high-affinity CAIX-specific near-infrared (NIR) fluorescent imaging probe, GZ22-4. Constitutive overexpression of autologous CAIX did not elicit specific autoantibody responses in vivo, demonstrating the suitability of this model for evaluating the efficacy of anti-CAIX vaccination as a therapeutic strategy. The in vivo study was repeated as an independent experiment and demonstrated good robustness of the developed model.

Keywords: 4T1 model; breast cancer; cancer imaging; carbonic anhydrase IX; hypoxia; preclinical models.

Fig. 1

In vitro characterisation of the 4T1 cell lines edited for mCAIX expression. (A) Western blot results showing the level of mCAIX protein production in the generated 4T1 cell lines after 30 h incubation in normoxia and hypoxia (exposition time for mCAIX – 6 min, β‐Actin – 1 min). (B) Changes in extracellular Ph (pHe) after 30 h incubation in serum starvation conditions in normoxia and hypoxia, relative to baseline at 0 h, normalized to cell number (10⁶ cells) (n = 3, mean, SD); ANOVA multiple comparison‐adjusted P‐value: ****< 0.0001 Relative to 4T1 in normoxia, ^####< 0.0001 Relative to 4T1 in hypoxia. (C) The left block shows the 4T1 cell line variant morphology in inverted phase contrast light microscopy (scale bar—100 μm) and the right block – Immunocytochemical detection of mCAIX production (green) side by side with DAPI staining of nuclei (blue) in the generated 4T1 cell line variants after 30 h incubation in normoxia and hypoxia (scale bar—50 μm; small upper left sections – Secondary antibody control). No nonlinear adjustments have been made in the original microphotographs; all have been acquired by using the same settings. 4T1‐CAIX, 4T1 cell line with no mCAIX production; 4T1‐CAIX⁺, 4T1 cell line with constitutive mCAIX production; Car9, murine carbonic anhydrase IX gene; mCAIX, murine carbonic anhydrase IX; M, size marker.

Fig. 2

The in vivo study design and tumour growth comparison in mice bearing 4T1 tumours with differential mCAIX expression. (A) An overview of the in vivo study design. In the exploratory study, adult BALB/c female mice (n = 7) were orthotopically implanted with 4T1 murine breast cancer cells in 50% Matrigel and monitored for tumour growth dynamics. Depending on the study group, animals received either parental 4T1 cells (group 1) or 4T1‐CAIX⁻ (group 2), or 4T1‐CAIX⁺ (group 3) cells. This exploratory study was repeated as an independent procedure 7 months later by using another set of BALB/c female mice (n = 7); the terminal procedure timing is depicted with the dotted line in this case. (B) Comparison of the tumour growth dynamics between study groups (n = 7, mean values within the groups ± SD are shown). One‐way repeated‐measures Mann–Whitney test P‐value shown (group 1 and 2 data compared). (C) The time needed to reach tumour volume of 500 mm³ (median, 95% CI), dots represent individual animals (n = 7); Kruskal–Wallis multiple comparison‐adjusted P‐values shown. The data from a repeated study were compared to that from the exploratory study – (D) shows the comparison of tumour growth dynamics (n = 7; mean, SD) and (E) shows comparison of time needed to reach tumour volume of 500 mm³ (n = 7; median, 95% CI; Mann–Whitney test P‐values shown). 4T1‐CAIX⁻, 4T1 cell line with no mCAIX production; 4T1‐CAIX⁺, 4T1 cell line with constitutive mCAIX production; Car9, gene encoding for murine carbonic anhydrase IX; D, day; G, group; R, replicate; *P < 0.05, **P < 0.01, ***P < 0.001, ns – non‐significant.

Fig. 3

In vivo epifluorescence imaging of murine CAIX in tumour‐bearing animals. (A) The chemical structure of the novel high‐affinity CAIX‐specific near‐infrared fluorescent imaging probe GZ22‐4 used for in vivo imaging. (B) Epifluorescence imaging results of tumour‐bearing animals at day 27 post tumour cell implantation. Results demonstrate fluorescent signals detected in living animals by using the IVIS Spectrum in vivo imaging system 24 (upper panel) and 48 h (middle panel) after intravenous administration of the GZ22‐4 probe. Results are displayed in a yellow hot logarithmic scale. Grayscale photographs demonstrating the tumour size differences are shown below for comparison. 4T1‐CAIX⁻, 4T1 cell line with no mCAIX production; 4T1‐CAIX⁺, 4T1 cell line with constitutive mCAIX production; ROI, region of interest.

Fig. 4

Macroscopic evidence of mCAIX expression‐associated changes in primary tumours, lung metastases, and spleen size. (A) Representative examples of ulcerative lesions of primary tumours at the repeated study endpoint (i.e., day 37 for groups 1 and 2, and day 31 for group 3). (B) Macroscopic appearance of extirpated primary tumours at study endpoint showing gross differences in vascularisation. (C) Representative examples of lung metastatic lesions at the repeated study endpoint; black arrows indicate macrometastases. (D) Comparison of spleen size between study groups at the repeated study endpoint (representative examples). (E) Comparison of spleen size between study groups in the exploratory and repeated study (mean, SD); dots represent individual animals (n = 6–7). Data for the exploratory study groups were compared (one‐way ANOVA multiple comparison‐adjusted P‐values shown). 4T1‐CAIX⁻, 4T1 cell line with no mCAIX production; 4T1‐CAIX⁺, 4T1 cell line with constitutive mCAIX production; G, group; R, replicate; ***P = 0.0001.

Fig. 5

Histological representation of the developed 4T1 breast cancer primary tumours and lung metastases. Upper panel (A–I): representative microphotographs of primary tumour tissue sections from the three study groups, stained with H&E. (A–C) Cross‐sectional view of the primary tumours demonstrating areas of coagulative necrosis (scale bar—200 μm). (D–I) Differential features of the primary tumour tissues (main panel scale bar—100 μm, inner panel scale bar—50 μm). Lower panel (J–L): H&E staining of lung tissue sections harvested from tumour‐bearing mice at study endpoint [i.e., day 26 after tumour cell injection for animals bearing 4T1 (J) and 4T1‐CAIX⁺ tumours (L), and day 39 after 4T1‐CAIX⁻ tumour cell injection (K); scale bar—200 μm, the black arrows indicate micrometastases]. 4T1‐CAIX⁻, 4T1 cell line with no mCAIX production; 4T1‐CAIX⁺, 4T1 cell line with constitutive mCAIX production; CN, coagulative necrosis; LS, lymphoid structures; V, vasculature.