FKBPL and peptide derivatives: novel biological agents that inhibit angiogenesis by a CD44-dependent mechanism

Abstract

Purpose: Antiangiogenic therapies can be an important adjunct to the management of many malignancies. Here we investigated a novel protein, FKBPL, and peptide derivative for their antiangiogenic activity and mechanism of action.

Experimental design: Recombinant FKBPL (rFKBPL) and its peptide derivative were assessed in a range of human microvascular endothelial cell (HMEC-1) assays in vitro. Their ability to inhibit proliferation, migration, and Matrigel-dependent tubule formation was determined. They were further evaluated in an ex vivo rat model of neovascularization and in two in vivo mouse models of angiogenesis, that is, the sponge implantation and the intravital microscopy models. Antitumor efficacy was determined in two human tumor xenograft models grown in severe compromised immunodeficient (SCID) mice. Finally, the dependence of peptide on CD44 was determined using a CD44-targeted siRNA approach or in cell lines of differing CD44 status.

Results: rFKBPL inhibited endothelial cell migration, tubule formation, and microvessel formation in vitro and in vivo. The region responsible for FKBPL's antiangiogenic activity was identified, and a 24-amino acid peptide (AD-01) spanning this sequence was synthesized. It was potently antiangiogenic and inhibited growth in two human tumor xenograft models (DU145 and MDA-231) when administered systemically, either on its own or in combination with docetaxel. The antiangiogenic activity of FKBPL and AD-01 was dependent on the cell-surface receptor CD44, and signaling downstream of this receptor promoted an antimigratory phenotype.

Conclusion: FKBPL and its peptide derivative AD-01 have potent antiangiogenic activity. Thus, these agents offer the potential of an attractive new approach to antiangiogenic therapy.

Figure 1

rFKBPL protein inhibits angiogenesis in vitro and ex vivo (A) Transient transfection of an FKBPL cDNA construct inhibits migration of wounded HMEC-1 monolayers compared to empty vector controls. Representative images of wounded monolayers and overexpression of FKBPL following transfection. The histogram shows the wound size relative to wound size at time=0 h ± SEM; n=3. Significance was determined by ANOVA. (B) Inhibition of HMEC-1 wound closure (compared to time matched control) after exposure to a range of concentrations of rFKBPL. Data points show means ± SEM; n=3. (C) Inhibition of HMEC-1 tubule formation in matrigel following exposure to increasing concentrations of rFKBPL; data are corrected to a sham treated control. Data points are means ± SEM; n=3. (D) Microvessel sprouting from rat aortic rings incubated with increasing concentrations of rFKBPL (left panel). Quantitative determination of vessel length and number of vessels after 7 days compared to time matched controls (right panel). Data points are means ± SEM; n=3.

Figure 2

FKBPL inhibits angiogenesis in vivo and prevents the growth of DU145 human tumor xenografts (A) rFKBPL (5 ng injected directly into the sponge on alternate days) inhibited β-FGF (10 ng)-induced angiogenesis in C57 black mice. 14 days after implant there was a marked decrease in vessel density and cellular ingrowth in rFKBPL-treated sponges (left panel; representative image-arrows indicate vessels containing bright eosin stained erythrocytes). Graph shows quantification of microvessel density in β-FGF alone or β-FGF+rFKBPL-injected sponges. Each symbol represents the average number of vessels per x40 field, with 10 fields counted blindly in 5 sponges. n=5 mice/sponges per treatment group. (B) Immunohistochemistry showing FKBPL expression in DU145 tumors grown in SCID mice after injection with a cDNA construct expressing FKBPL. (C) DU145 tumors grown in SCID mice were intra-tumorally injected twice weekly for the duration of the experiment with 25 μg of a cDNA construct expressing either FKBPL or endostatin (as a positive control), or pcDNA3.1 empty vector (as a negative control). Graph shows tumor volume over time ± SEM; n=4-7 mice per condition. (D) Kaplan Meier survival curves; *significance was determined by the Logrank test.

Figure 3

The active domain within FKBPL resides between aa-34-58 and a 24mer-based peptide, AD-01, spanning this domain inhibits angiogenesis and is more potent than rFKBPL. (A) Wound size of HMEC-1 monolayers 7 h after transfection with truncated DNA constructs; n=3. (B) Inhibition of migration (left panel; compared to time matched controls) and tubule formation (right pane; compared to time matched controlsl) of HMEC-1 cells after treatment with AD-01/AL-57 peptides across a range of concentrations. Data points are means ± SEM; n=3. (C) Inhibition of microvessel sprouting from rat aortic rings (compared to time matched controls ± SEM) incubated for 7 d with a range of concentrations of AD-01/AL-57; n=3. (D) AD-01 inhibited β-FGF-induced angiogenesis in the sponge assay in vivo. Microvessel densities in implanted sponges treated with in β-FGF alone (10 ng) or or β-FGF + AD-01 (0.35 μg or 0.11 ng). Each symbol represents the average number of vessels per x40 field, with 10 fields counted blindly in each sponge; n= 5 mice/sponges for β-FGF alone; n=3 mice/sponges for AD-01 treatment.

Figure 4

Systemic delivery of AD-01 inhibits blood vessel development and tumor growth (A) Growth curves for DU145 xenografts or MDA-231 (B) with or without daily i.p injection of a range of doses of AD-01. Data points are means ±SEM; n= 5-8 mice per treatment group and Kaplan Meier survival curves with time to tumor tripling volume as the endpoint for survival; *significance was determined by the Logrank test. (C) Intravital images (compressed Z-stacks) of DU145 xenografts showing blood vessels at 7 and 14 days after the start of treatment with AD-01 (0.3mg/kg/day) or PBS control, i.p. (left panel). Number of vessel branch points or average vessel diameter (μm) at 7 and 14 days after initiating treatment with AD-01 (0.3mg/kg/day, i.p.) or PBS ± SEM; n=5 mice per treatment group, 4 fields of view per tumor and 30 vessels per field. Significance was determined by the two-tailed T -test.

Figure 5

Systemic delivery of AD-01 inhibits DU145 tumor growth in combination with docetaxel. (A) Fold change in tumor volume following treatment with AD-01 (0.3 and 0.003 mg/kg/d, i.p.) in combination with docetaxel (20 mg/kg once every 15 days in 3 cycles) compared to PBS alone, AD-01 alone and docetaxel alone controls ± SEM; n= 5-7 mice per treatment group. (B) Kaplan Meier survival curves with time to tumor doubling as the endpoint for survival. *significance was determined by the Logrank test. (C) Haematoxylin and eosin staining of FFPE tissue sections taken from control animals and animals exposed to high dose AD-01 (0.3 mg/kg/day) at the end of the experiment; no obvious toxicity was observed.

Figure 6

FKBPL is secreted and is dependent on CD44 for its anti-angiogenic activity (A) IP western blots using conditioned medium from HMEC-1, L132 and MDA-231 cells 24 h after plating. FKBPL was IPed using an anti-FKBPL antibody and then run on a western blot and probed for FKBPL. Control medium (no cell exposure) or immunoprecipitation with an IgG control were used as negative controls. Whole cell lystates were used as positive controls. (B) Inhibition of migration of HMEC-1 compared to time matched control 72 h after transfection with non-targeted siRNA compared to CD44 targeted siRNA in untreated cells (upper graph) and treated (10⁻⁹M AD-01; lower graph) ± SEM; n=5. (C) Migration of tumor cells after treatment with AD-01 (left panel) or rFKBPL (right panel) across a range of concentrations (± SEM; n=3). (D) Representative gel showing inhibition of rac activity in HMEC-1 cells after treatment with AD-01 (10⁻⁹M) for 10 min/60 min prior to serum/HA activation.