(A) Equivalent concentrations of Wnt3a and liposomal Wnt3a were tested for their ability to stimulate luciferase activity in LSL cells. DMEM and PBS exhibited baseline activity. Wnt3a protein elicited volume-dependent activity. An equivalent concentration of liposome-packaged Wnt3a exhibited substantially greater activity. (B) Liposomal packaging sustained Wnt3a-dependent activity. The same volume of Wnt3a and Wnt3a liposomes elicited similar levels of activity after 24 h and 48 h in vitro. At 72 h, however, liposomal Wnt3a exhibited greater activity than Wnt3a protein.

(A) Structure of the 2-bromo 4,5-dimethoxysulfonamide analog (Ant 1.4Br) and 2-chloro 4,5-dimethoxysulfonamide analog (Ant 1.4Cl). (B) PA-1 cells transfected with a SuperTOPflash Wnt reporter were treated with Ant 1.4Br, Ant 1.4Cl, or hFzd8CRD. Frzd8CRD (yellow line) reduced Wnt-dependent luciferase activity. Ant 1.4Br (red) and Ant 1.4Cl (orange) were ineffective at blocking Wnt-induced activity. (C) 10T1/2 cells transfected with a TOPflash Wnt reporter construct were co-cultured with LSL cells secreting Wnt3a. No luciferase activity was detectable in 10T1/2 cells grown alone (control, brown line); luciferase activity was at a maximum when cells were co-cultured with LSL cells secreting Wnt3a (blue line). Fzd8CRD (yellow) inhibited paracrine-induced Wnt reporter activity in a dose-dependent manner, to a maximum of 90%. Ant 1.4Br (red) reduced luciferase activity by less than 20%. (D) LSL cells transfected with a TOPflash Wnt reporter construct were treated with either exogenous Wnt3a or liposomal Wnt3a in the presence of Ant 1.4Cl or hFzd8CRD. Ant 1.4Cl inhibited exogenous Wnt3a-induced reporter activity by 80% but inhibited liposomal Wnt-induced reporter activity by only 30%. hFzd8CRD (yellow) reduced Wnt3a-induced reporter activity by 80% in Wnt3 treated cells and by 70% in liposome treated cells. The symbols * and # reflect statistical significance (Student's T-test, p<0.05).

(A) Western analyses indicated that ∼70% of Wnt3a protein added during synthesis was incorporated into the liposomes; the remaining 30% remained in the supernatant. (B) Trypsin removed liposomal Wnt3a activity as measured in an in vitro dual-reporter assay (n = 3, mean+standard deviation). (C) Western analyses demonstrated that ∼20% of Wnt3a protein incorporated into the liposomal preparation still remained following trypsin digestion. This portion (∼20%) was localized to the endo-liposomal surface where it was inactive.

Liposomes were made with various lipid compositions as indicated (n = 3; mean+standard deviation). (A) Different lipid compositions exhibited radically different activities in vitro. (B) Western blot analysis showed that these differences in activity were not attributable to variations in total Wnt incorporation into the liposomal membrane.

(A) Equal volumes of PBS or Wnt3a liposomes were injected subcutaneously into Axin2LacZ/+ reporter mice. Xgal staining at 24 h revealed reporter activity in cells at the epidermal/dermal junction and surrounding hair follicles (arrows). (B) Treatment with Wnt3a liposomes resulted in stronger Xgal staining with no change in the distribution of reporter activity. Over the intervening time period, both groups received injections of PBS liposomes or Wnt3a liposomes every other day. On day 14, tissues were collected. (C) Repeated injection of PBS liposomes had no effect on dermal thickness or hair follicle number (n = 6). (D) In contrast, repeated injections of Wnt3a liposomes resulted in dermal thickening and a dramatic increase in the number of hair follicles at the site of injection (n = 6). e, epidermis; d, dermis; scf, subcutaneous fat; m, muscle.