Fig 2: (A) bovine IVD (bIVD) treatment by MSCs in spatial separation (0.22 µm porous membrane). (B-C) No changes in Tie2 expression (0.8- ± 0.5-fold) and fraction of dead cells (1.0- ± 0.3-fold) relative to donor corresponding ctrl disc was found at day 5 (n = 5 bovine tails, 2 MSC donors). IVD indicates intervertebral disc; MSCs, mesenchymal stem cells.

Fig 3: The intervertebral disc and degenerative disc disease. Schematic and a hematoxylin/eosin-stained histological sections of a healthy (A, B) and degenerated (C, D) canine intervertebral disc. (E) Human lumbar intervertebral disc obtained postmortem. Appropriate tissue selection for Tie2 fishing is indicated by dashed black lines, while degenerated or AF tissue areas are indicated by dashed red lines. (F) Human intervertebral disc (IVD)-derived tissues during discectomy are selected based on morphology, color and stiffness. Only gelatinous, white, and transparent tissue should be used for Tie2 fishing. Yellow and stiff tissue should be disregarded in order to enhance Tie2 detection

Fig 4: Colony forming units from assessed species. Nucleus pulposus progenitor cells from different species sorted on Tie2 expression were cultured in semi-solid methylcellulose medium. Fibroblastic colony forming units (CFU-F) and spherical colony forming units (CFU-S) emanate as 2 distinguishable colony types. Scale bar represents 50 µm

Fig 5: Affinity maturation of the Ang2-BDRGD-based library bi-specific for avß3 integrin and Tie2-Fc. a Ang2-BD was presented on the yeast cell surface as a fusion with agglutinin proteins. Display levels were detected using primary antibodies against the C-terminal cMyc tag (chicken anti-cMyc antibodies) and phycoerythrin (PE)-conjugated anti-chicken antibodies. Binding to Tie2-Fc was determined using fluorescein isothiocyanate (FITC)-conjugated anti-human Fc antibodies. Binding to avß3 integrin was determined using FITC-labeled mouse anti-av integrin antibodies. b–g FACS analysis of the binding of the bi-specific Ang2-BD-based library to avß3 integrin in different screening steps. Quadrant gate statistics are indicated in each panel b negative control. c Ang2-BDWT expression and avß3 integrin binding (10 nM). d Expression of the bi-specific Ang2-BDRGD-based library and avß3 integrin binding (10 nM) at pre-sorting and e–g expression of the bi-specific Ang2-BD-based library and avß3 integrin binding (10 nM) after sorts 1, 3, and 5, respectively. h Binding of isolated yeast-displayed bi-specific Ang2-BDRGD clones to Tie2 (20 nM). Data were normalized to the yeast surface expression levels of each clone and Tie2 binding of Ang2-BDWT. i Binding of isolated yeast-displayed bi-specific Ang2-BDRGD clones to avß3, a5ß1, avß5, a4ß7, aIIbß3, and a3ß1 integrins (50 nM). Data were normalized to the yeast surface expression level of each clone. Data shown represent the average of triplicates from independent experiments, and error bars represent the standard error of the mean