Recombinant TAT-Thymosin β-4 Promotes Angiogenesis by Activating VEGFR2 Signaling Pathway

Thymosin β-4 is a 43-amino acid intracellular polypeptide that was originally isolated from bovine thymus. Of the 16 known thymosin families, thymosin β-4 is the most common type found in all tissues. Thymosin β-4 regulates angiogenesis, cell differentiation, morphogenesis, migration, and organogenesis and is linked to a dynamic equilibrium between G-actin and F-actin. In particular, thymosin β-4 is well-known for its angiogenic and anti-apoptotic functions. In this study, we synthesized thymosin β-4 linked with the well-known cell-penetrating peptide TAT (YGRKKRRRQRRR). TAT-thymosin β-4 promotes angiogenesis and cell migration in vitro via the VEGFR2 signaling pathway and reduces apoptosis. To examine angiogenic potential in vivo, a Matrigel Plus assay was conducted that revealed the angiogenic effect of TAT-thymosin β-4. In conclusion, TAT-thymosin β-4 promotes blood vessels and is expected to be applicable in regenerative medicine for all organs requiring blood vessels.


Introduction
Angiogenesis is the process of forming blood vessels from the pre-existing vasculature 1 . It is necessary in the process of development, but is also crucial for the recovery of damaged or pathological tissue [1][2][3] . However, abnormal angiogenesis is associated with several diseases, including cancer, cardiovascular disease, diabetes, and age-related macular degeneration 1 . Thus, modulators of angiogenesis, such as vascular endothelial growth factor and its related molecules (inhibitors such as pegaptanib sodium, ranibizumab, bevacizumab, and antibodies) have been approved for treating diseases described above 1 .
Beta-thymosins (thymosin-β) comprise a structurally related family that is highly conserved in the animal kingdom 4,5 . There are 16 types of beta-thymosins, but only 3 types are found in humans (thymosin-β4, thymosin-β10, and thymosin-β15) 6,7 . Although highly homologous in genetic structure, these three proteins are derived from different genes and thus, are distinct products. NMR studies have revealed that thymosin-β4 is mostly unstructured in aqueous solutions with some preferential α-helix structures 8, 9 . Thymosin β-4 controls the assembly and disassembly of actin laments through its initial interaction with G-actin 8, 9 . Thus, the mechanism by which thymosin-β4 in uences angiogenesis, cell differentiation, morphogenesis, cell migration, and organogenesis is assumed to be linked to the maintenance of a dynamic equilibrium between G-actin and F-actin 8-10 .
As such, thymosin β-4 is a promising drug candidate for many diseases. However, unlike other angiogenic modulators, thymosin-β4 does not act through receptors on the cell surface but acts directly on cells 11 . This is a major obstacle to the use of thymosin-β4 as a drug target. For this reason, we combined the well-known cell-penetrating peptide TAT and thymosin-β4 for intracellular translocation from the extracellular area 12,13 .

Page 4/16
The recombinant protein produced in E. coli by molecular cloning techniques revealed the angiogenic effects of thymosin-β4 in vitro and in vivo. Here, we provide rst experimental evidence that recombinant TAT-thymosin-β4 is an angiogenic modulator of the VEGFR2 signaling pathway.

In vitro angiogenic effect of TAT-Thymosin ß-4
To investigate the angiogenic effect of TAT-thymosin β-4, a tube formation assay was performed. At 0-10 µg/mL, the angiogenic effect of TAT-thymosin β-4 increased proportionally ( Fig. 2a-f). However, at 15-50 µg/mL concentration, no angiogenic effect was observed; rather an unhealthy condition of the cells was observed ( Fig. 2g-j). We considered a concentration of 10 µg/mL to be the optimal concentration for angiogenesis. VEGF treated group was used as a positive control for angiogenesis ( Fig.  2k) In addition, cell migration, an important factor in angiogenesis, was examined using a wound healing assay. As in the tube formation assay, 10 µg/mL was ideal for cell migration assay (Fig. 3a, b), and cell migration increased proportionally at a concentration of 0-10 µg/mL.
To con rm the relationship between TAT-thymosin β-4 and VEGFR2, VEGF antagonists (bevacizumab and sorafenib) were treated in TAT-thymosin β-4-treated HUVECs. VEGFA secreted in the cell culture medium was increased in the TAT-thymosin β-4 treatment group but no change was observed in the group treated with both TAT-thymosin β-4 and VEGF antagonist (Fig. 5a, c). Western blotting of the cell lysates showed that the amount of VEGFA was similar to the change in the cell culture medium (Fig. 5b, c). The p-VEGFR2 level was increased in the TAT-thymosin β-4 treatment group, but reduced in the TAT-thymosin β-4 and antagonist co-treatment groups (Fig. 5b, c). Similar trends of increase and decrease were observed for p-Akt and p-ERK (Fig. 5b, c).

String database analysis
The protein group interacting with VEGF and the protein group interacting with thymosin β-4 were largely identical to each other (Fig. 7a). As a result of analyzing only highly correlated proteins, thymosin β-4 was detected upstream of VEGF (Fig. 7b). This result is consistent with the results of western blotting using VEGF inhibitors (Fig. 5). These data strongly suggest that TAT-thymosin β-4 activates the VEGFR2 signaling pathway through VEGF, thereby exhibiting an angiogenic effect.

In vivo angiogenic effect of TAT-Thymosin ß4
The Matrigel plug, which was dissected from the animal 2 weeks after injection, showed newly formed blood vessels in the TAT-thymosin β-4-treated group and the VEGF-treated group, while the non-treated group showed no blood vessels (Fig. 8). The hemoglobin content of the Matrigel plug was measured using the Drabkin's reagent. A total of 5.25 ± 0.64 ug/mg of hemoglobin was detected in VEGF-containing Matrigel, and 5.84 ± 0.82 ug/mg of hemoglobin was detected in TAT-thymosin β-4-containing Matrigel. In the Matrigel plug of control group, only 1.07 ± 0.24 ug/mg of hemoglobin was detected (Fig. 8b, c). These data strongly suggest that TAT-thymosin β-4 has as much angiogenic capacity as VEGF in vivo.

Discussion
Thymosin β-4 is capable of angiogenesis similar to VEGF 12,13 . In the case of VEGF, receptors on the cell surface interact with VEGF, so it is possible to target it using a drug without cell penetration 15,16 .
Thymosin β-4 has many clinical bene ts such as angiogenesis, cell migration, cell adhesion, differentiation, prevention of apoptosis, cell survival, and tissue regeneration. It is a promising new drug candidate, if it is easy to transfect into cells.
To achieve this, we used molecular cloning techniques. We combined TAT, a well-known transmembrane molecule, and thymosin β-4 at the DNA level to produce a fusion protein in E. coli. When HUVECs were treated with the fusion protein, an increase in the cytoplasmic fusion protein was con rmed compared to that of the control. HUVECs expressing the cytosolic fusion protein exhibited increased angiogenesis and migration ability and prevented apoptosis in vitro. In addition, angiogenic effects have been observed in vivo. We found that cytosolic thymosin β-4 promoted VEGFA secretion. Secreted VEGFA binds to VEGFR2 in autocrine or paracrine regulation, which activates the VEGFR2 signaling pathway.

Recently, dermal wound healing and regeneration effects of thymosin β-4 have been demonstrated in a
full-thickness excisional rat model and impaired-healing model [17][18][19] . Furthermore, based on these regenerative properties, RGN-137 and RGN-259 have been clinically studied for several diseases such as pressure ulcers, venous stasis ulcers, and corneal wounds 4 .
However, compared to VEGF, the e cacy of angiogenesis is poor and studies on the topic are scarce. Although the molecular weight of thymosin β-4 is small, it is partially introduced into the cytosol. This explains why the number of studies on thymosin β-4 is small. Translocation to the cytosol of thymosin β-4 has numerous potential applications.

Recombinant protein and cell penetration
We used an advanced TAT sequence (YGRKKRRRQRRR). Its tricodon DNA sequences (33 bp) were ligated with PCR-ampli ed TMSB4X sequences (136 bp) in the pBluescript Sk2 + plasmid multiple cloning site using T4 ligase. Thereafter, the plasmids were transformed into E. coli (RBC Bioscience, New Taipei City, Taipei) for protein production. Recombinant proteins were isolated using a Sepharose FF column (Merk-Millipore, Burlington, VT, USA).

Tube formation and wound healing assay
Pre-chilled growth factor-reduced Matrigel was added to a 24-well culture plate. The cells were incubated at 37°C for 30 min to solidify the Matrigel. HUVECs (7.5 × 10 4 cells per well) were trypsinized and added to Matrigel-coated 24-well culture plates. TAT-thymosin β-4 was added according to the group composition (0, 2, 4, 6, 8, 10, 15, 20 µg/mL concentration). Four hours later, tube formation images were obtained, processed, and calculated using ImageJ software.
Poly-L-lysine (50 µg/mL) was coated onto a 60-mm cell culture dish. Fibroblasts (8 × 10 5 cells per well) were seeded to create a con uent monolayer. The scratch was created with a p200 tip and incubated for 16 h. Wound healing images were obtained, processed, and calculated using ImageJ software.

Protein analysis
Total cellular protein samples were collected from each group. Western blotting was performed according to the guidelines supplied by Bio-Rad (Hercules, CA, USA). Primary antibodies (Table 1) were diluted 1:1000 in 3% bovine serum albumin (BD Biosciences Franklin Lakes, NJ, USA) and visualized by chemiluminescence imaging. Band density was measured using ImageJ software. We performed analysis of protein-protein interactions for proteins related to TAT-thymosin β-4 and VEGF using a stringdatabase platform (http://string-db.org).

Statistical analysis
SPSS software, v20.0 (SPSS, Inc., Chicago, IL, USA) was used. Values are reported as mean ± standard deviation (SD). Statistical differences were assessed using a non-parametric one-way Kruskal-Wallis test. Statistical signi cance was set at P < 0.05.

Matrigel plug assay
To investigate the angiogenic properties of TAT-thymosin β-4, an in vivo Matrigel plug angiogenic assay was performed. Six-week-old BALB/c nude mice were purchased from Orient Bio (Seongnam, Gyeonggido, Korea). All experiments in mice were performed according to the guidelines of the Korea Polytechnic University Institutional Animal Care and Use Committee (KPU-2020-0001, 1 May 2020). All animals were maintained in a 12 h light/dark cycle at 23 ± 1°C and 50 ± 10% relative humidity with ad libitum feeding.
One hundred micrograms of Matrigel (BD Biosciences, San Jose, CA, USA) containing 60 U/mL heparin (Sigma-Aldrich, CA, USA) mixed with TAT-thymosin β-4 (n = 3) or VEGFA (n = 3) was injected subcutaneously into BALB/c nude mice. Two weeks later, the Matrigel plugs were removed from the animals after euthanasia. Thereafter, vessel formation in Matrigel plug was quanti ed by measuring the hemoglobin using Drabkin's reagent (Sigma-Aldrich, CA. USA) 20 .

Conclusions
Several previous studies reported the angiogenic effect of thymosin β-4. However, it is functional only when it is located in the cytoplasm. For this reason, we prepared a fusion protein by combining thymosin β-4 with the cell-penetrating peptide TAT for cytosolic translocation. Translocated TAT-thymosin β-4 is located in the cytoplasm, promotes angiogenesis, and reduces the apoptosis of HUVECs. Furthermore, TAT-thymosin β-4 also promotes angiogenesis in vivo.
In this study, we con rmed angiogenesis in vivo. However, we did not con rm whether angiogenic effects were maintained under pathological conditions. In future studies, we plan on employing a foot ulcer   Tube formation assay. TAT-thymosin β-4 of concentrations a 0 µg/mL, b 2 µg/mL, c 4 µg/mL, d 6 µg/mL, e 8 µg/mL, f 10 µg/mL, g 15 µg/mL, h 20 µg/mL, i 30 µg/mL, and j 50 µg/mL was treated. k VEGF-treated positive control. a-f Tube formation ability was increased in proportion to the TAT-thymosin β-4 concentration. f,k The 10 µg/mL treatment group showed similar tube formation ability to the VEGF treated group. g-j the tube formation ability was lost. a Images were processed sequentially using ImageJ software from original images (a-k). Scale bar indicates 50 µm. Figure 3 Cell migration assay. a TAT-thymosin β-4 treated to HUVEC cells in concentration of 0, 5, 10 μg/mL. b Images were processed using ImageJ software from original images to (a) highlight the areas where cells have not migrated. Scale bar indicated 50 μm. Activation of VEGFR2 signaling pathway according to TAT-thymosin β-4 concentration. a VEGFR2 and p-VEGFR2 indicate VEGFR2 signaling pathway. Angiopoietin 1 and Angiopoietin 2 showed vascular endothelial differentiation through VEGFR2 signaling pathway. b Relative band densities compared to βactin and were measured at least three times for statistical analysis and all data were expressed as mean ± standard deviation. Asterisk indicates signi cant difference (p<0.05) compared to control group (0 µg/mL ).

Figure 5
Loss-of-function study using a VEGF antagonist. a Western blotting of cell culture medium for VEGF. PonceauS was used as standard because there was no β-actin in cell culture medium. b Western blotting of cell lysates for VEGF, p-VEGFR2, t-VEGFR2, p-AKT, t-AKT, p-ERK, t-ERK, CD31. c Relative band density compared to Ponceau S or β-actin and were measured at least three times for statistical analysis and all data were expressed as mean ± standard deviation. Asterisk indicates signi cant difference (p<0.05) compared to control group (0 µg/mL ).

Figure 6
Anti apoptotic effect of TAT-thymosin β-4. a Western blotting of apoptotic markers using apoptosis western blot cocktail consisting of cleaved PARP, β-actin, procaspase 3, and cleaved caspase 3. b Relative band densities compared to β-actin and were measured at least three times for statistical analysis and all data were expressed as mean ± standard deviation. Asterisk indicates signi cant difference (p<0.05) compared to control group (0 µg/mL ).

Figure 7
String database analysis using thymosin β-4 (TMSB4X) and VEGF (VEGFA). a Medium con dence 0.400 no more than 20 interactions. b Sorting only those with high relevance. Figure 8