Possibility for Transcriptional Targeting of Cancer-Associated Fibroblasts-Limitations and Opportunities

Abstract

Cancer-associated fibroblasts (CAF) are attractive therapeutic targets in the tumor microenvironment. The possibility of using CAFs as a source of therapeutic molecules is a challenging approach in gene therapy. This requires transcriptional targeting of transgene expression by cis-regulatory elements (CRE). Little is known about which CREs can provide selective transgene expression in CAFs. We hypothesized that the promoters of FAP, CXCL12, IGFBP2, CTGF, JAG1, SNAI1, and SPARC genes, the expression of whose is increased in CAFs, could be used for transcriptional targeting. Analysis of the transcription of the corresponding genes revealed that unique transcription in model CAFs was characteristic for the CXCL12 and FAP genes. However, none of the promoters in luciferase reporter constructs show selective activity in these fibroblasts. The CTGF, IGFBP2, JAG1, and SPARC promoters can provide higher transgene expression in fibroblasts than in cancer cells, but the nonspecific viral promoters CMV, SV40, and the recently studied universal PCNA promoter have the same features. The patterns of changes in activity of various promoters relative to each other observed for human cell lines were similar to the patterns of activity for the same promoters both in vivo and in vitro in mouse models. Our results reveal restrictions and features for CAF transcriptional targeting.

Keywords: fibroblasts; gene therapy; promoter; transcriptional targeting; tumor microenvironment.

Figure 1

Determination of Transcription Level of selected genes in human cells. The Y-axis indicates the relative level of mRNA content in cells. The values obtained were normalized to the transcription levels of the 18S RNA. The measurements were performed in three independent experiments for each sample and are represented as mean ± s.e.m. The Y-axis serif designate scale break. Below the graph is a heat map of the expression values (in TPM, Transcripts Per Kilobase Million) of the genes in the studied cell lines according to the RNA-seq E-MTAB-2706 database. The names of genes are indicated below the heat map. N/D—no data available.

Figure 2

Schemes of promoter regions used in the work. Regions with several binding sites for transcription factors (TFs) are shown in green; deletion of such regions is critical for the activity of the promoter; sites of single binding TFs, important for the activity of the promoter, are marked in red; TATA-like regions are marked in black; the curved arrow indicates the translation initiation site.

Figure 3

Scheme of the reporter constructs (based on pGL3). Promoter—FAP, CXCL12, IGFBP2, CTGF, JAG1, SNAI1, or SPARC, which are the promoters of the respective genes. Luciferase: firefly luciferase gene.

Figure 4

Relative promoter activities of the studied promoters in different human cell lines. The upper graph represents relative promoter activities (Y-axis) as ratios of the luciferase activity expressed by plasmids containing promoters under study to the R. reniformis luciferase activity. Luciferase activity expressed by promoterless pGL3-basic plasmid was subtracted. Mean values (±s.e.m.) of relative luciferase activity were calculated from three independent experiments. The Y-axis serifs designate scale breaks. The names of promoters are indicated below the top X-axis and correspond to both graphs. The numbers under the names of promoters indicate the median activity of promoters in cell lines relative to the activity of the CMV promoter. The lower graph represent the activity of the promoters under the study (Y-axis, expressed in %) relative to the activity of the CMV promoter in each cell line. Graph for the CMV activity (100%) is omitted.

Figure 5

Relative promoter activities of the studied promoters in mouse cell lines. The graph represents relative promoter activities (Y-axis) as ratios of the luciferase activity expressed by plasmids containing promoters under study to the R. reniformis luciferase activity. Luciferase activity expressed by promoterless pGL3-basic plasmid was subtracted. Mean values (± s.e.m.) of relative luciferase activity were calculated from at least three independent experiments. The Y-axis serifs designate scale breaks. The names of promoters are indicated below the X-axis. The numbers under the names of promoters indicate the median activity of promoters in mouse cell lines relative to the activity of the CMV promoter. Median promoter activities higher than SV40 (purple) are shown in red.

Figure 6

Relative promoter activities of the studied promoters in mouse CMT 167 tumors. The graph represents relative promoter activities (Y-axis) as ratios of the luciferase activity expressed by plasmids containing promoters under the study normalized to milligram of protein. The Y-axis serifs designate scale breaks. The numbers under the names of promoters indicate the activity of promoters in tumors relative to the activity of the CMV promoter. The measurements were performed in 3–4 technical replicates for each sample and are represented as mean ± s.e.m. RLU/mg—relative light units per milligram of protein.