Advanced Tri-culture Wound System for Studying Androgen Influences on Angiogenesis and Inflammation

Introduction: Transgender males (i.e., born female and identifying as male) have been clinically observed to exhibit greater hypertrophic scarring—typically associated with an inflammatory wound healing environment—while on gender-affirming testosterone (T) therapy. Our lab, closely connected to gender-affirming clinical surgery, has recapitulated this phenomenon in vivo in rodent and pig wound healing (WH) models. However, T, an androgen, has been reported to exert anti-inflammatory effects in vitro. This discrepancy has stunted the scientific community’s ability to test the mechanisms responsible for T and WH in an accessible, high-throughput manner.

T is known to enhance angiogenesis in multiple contexts, for example after myocardial infarction. We hypothesized that treatment with T promotes angiogenesis via macrophage polarization and kinetics. We tested this in an in vitro wound model when fibroblasts and macrophages are co-cultured with endothelial cells, a cell type that was lacking in our previous investigations .

Materials and

Methods: Pooled female and male human dermal fibroblasts (HDF), human umbilical vein endothelial cells (HUVEC), and THP-1 cells were grown and maintained in FGM-2, EGM-2, and RPMI culture media, respectively, under standard conditions. THP-1 cells were differentiated into macrophages using Phorbol-12-myristate-13-acetate (PMA) and, for the polarized model, induced into an M2-like phenotype using IL-4 at 20ng/mL.

On day one, each well of a black 96-well plate was coated with a 0.2% gelatin solution for 2 hours at 37°C after which excess solution was aspirated away and left to dry in a sterile cell culture hood to promote monolayer adhesion. 10,000 HDF were seeded into each well. The next day, 2,000 HUVEC and either 10,000, 25,000, or 100,000 THP-derived macrophages were seeded per well of the monolayer according to its experimental group (Fig. 1).

Serum-containing treatment media was composed of Testosterone R1881 at 0nM and 10nM concentrations as well as vascular endothelial growth factor 165 (VEGF165) at 10ng/mL. R1881, a testosterone analogue with greater solubility and no capacity for aromatization to estradiol, was used for androgen receptor stimulation.

Brightfield images were taken daily. Supernatant was collected and the treatment media was replaced every two days. Plates were harvested on days 5 and 10. Immunofluorescent imaging was performed on Zeiss AxioObserver after staining with DAPI and CD31 antibodies (Fig. 2A, B). Percent area, branching, and junctions were quantified using ImageJ. Total protein and gene expression analyses were performed through BCA and qPCR assays.

Results, Conclusions, and Discussions: Percent area of CD31+ staining, branching, and junctions decreases with increasing T for the THP 10k and 25k conditions on Days 5 (Fig. 2C, E, G) and 10 (Fig. 2D, F, H). A similar effect is observed in a polarized tri-culture but only on Day 5 (Fig. 3A). By Day 10, this effect is reversed, with branching increasing with increasing amounts of T (Fig. 3B).

A T dose-dependent increase in protein concentration was observed when THP was 25k and 100k on Day 5 (Fig. 4A) only. By Day 10, however, no difference in protein concentration among conditions was observed (Fig. 4B).
In the IL-4 polarized model, Day 5 levels of TNFa gene expression decrease in T-treated groups (Fig. 5A). The opposite trend is observed at the later timepoint Day 10 (Fig. 5B).

The decrease in the endothelial cell marker CD31 when T increased was a surprising finding in the unpolarized model, given that macrophages derived from THP via PMA tend to default towards an M1-like phenotype. In the IL-4 polarized model, contrary to expectations of decreased angiogenesis, we in fact saw more, suggesting there may be other cell types missing from this model that would account for these observations.

The increase in protein concentration due to T while there being a simultaneous decrease in CD31+ protein expression could be due to a proliferation of cell types that do not express CD31, in this case the HDF or the THP-derived macrophages.

The increase in TNFa on Day 10 correlates with dose escalation of T as well as with cell numbers of THP-derived macrophages. Furthermore, when T is held constant, an increase in THP-derived macrophages led to an increase in TNFa expression.

We created a unique in vitro tri-culture system mimicking a wound environment to investigate the effects of T on inflammation and angiogenesis. With easy-to-visualize angiogenic metrics, this platform affords investigators worldwide high-throughput screening of various treatment conditions. Further probing of other pro-inflammatory genes such as IL-6 and IFNg are underway. Future work includes using other macrophage sources such as PBMCs.

Acknowledgements (Optional):