Increasing the level of cytoskeletal protein Flightless I reduces adhesion formation in a murine digital flexor tendon model

Abstract

Background: Surgical repair of tendons is common, but function is often limited due to the formation of flexor tendon adhesions which reduce the mobility and use of the affected digit and hand. The severity of adhesion formation is dependent on numerous cellular processes many of which involve the actin cytoskeleton. Flightless I (Flii) is a highly conserved cytoskeletal protein, which has previously been identified as a potential target for improved healing of tendon injuries. Using human in vitro cell studies in conjunction with a murine model of partial laceration of the digital flexor tendon, we investigated the effect of modulating Flii levels on tenocyte function and formation of adhesions.

Methods: Human tenocyte proliferation and migration was determined using WST-1 and scratch wound assays following Flii knockdown by siRNA in vitro. Additionally, mice with normal and increased levels of Flii were subjected to a partial laceration of the digital flexor tendon in conjunction with a full tenotomy to immobilise the paw. Resulting adhesions were assessed using histology and immunohistochemistry for collagen I, III, TGF-β1and -β3 RESULTS: Flii knockdown significantly reduced human tenocyte proliferation and migration in vitro. Increasing the expression of Flii significantly reduced digital tendon adhesion formation in vivo which was confirmed through significantly smaller adhesion scores based on collagen fibre orientation, thickness, proximity to other fibres and crimping. Reduced adhesion formation was accompanied with significantly decreased deposition of type I collagen and increased expression of TGF-β1 in vivo.

Conclusions: These findings suggest that increasing the level of Flii in an injured tendon may be beneficial for decreasing tendon adhesion formation.

Keywords: Adhesions; Flightless I; Flii; Tendon; Tenocyte.

Fig. 1

Reduced Flii levels impair human tenocyte cell proliferation and migration. (a) Expression of Flii protein in human tenocytes and fibroblasts treated with siRNA or control illustrating reduced Flii levels in response to treatment. Effect of attenuated Flii expression in human tenocytes and fibroblasts using siRNA and subsequent effects on cell proliferation (b-c) and migration (d-e) assessed using a WST-1 proliferation assay or scratch wound assay respectively. *p ≤ 0.05. Data represented as mean ± SEM. n = 6

Fig. 2

Murine digital tendon adhesion model. (a) Schematic of surgical techniques used for murine digital tendon adhesion model. (b) Image of 50% laceration surgery in fourth (shown) digit between the A1 and A3 pulleys through a transverse skin incision. Dotted line indicates the exposed tendon. Scale bar = 2 mm. (c) A 50% partial laceration (visible between the arrows) is carried out in the exposed tendon. Dotted line indicates the exposed tendon. Scale bar = 500 μM. (d) The skin over the laceration is closed with a single suture. Scale bar = 2 mm. (e) The deep and superficial flexor tendons are exposed distal to the ankle in the left hind limb by a transverse skin incision. The superficial flexor tendon can be visualised as the white area within the dotted lines. Scale bar = 2 mm. (f) A tenotomy is performed, fully dividing the deep and superficial tendons. The ends are buried, the arrow indicates the buried superficial flexor tendon still visible under the skin. Dotted line indicates transverse skin incision. Scale bar = 1 mm. Scale bar = 2 mm. (g) H&E image showing 50% of divided tendon fibres (arrows). T indicates the tendon and S the surrounding structures. Magnification × 4, scale bar = 200 μM. (h) Graphical representation of laceration validation. n = 6. Immobilisation vs mobilisation adhesion formation validation was performed. H&E stained section of tendon in WT mouse 21 days after partial laceration and (i) freely mobilised (j) immobilised. Dotted lines indicate adhesions. T, tendon; E, epidermis; D, dermis. Magnification × 4. (k) Graphical representation of the percentage of WT digits that formed adhesions at D21 in both mobilised and immobilised paws ± SEM. n = 6. (l) Graphical representation of adhesion size in WT digits at D21 in both mobilised and immobilised paws ± SEM. n = 6

Fig. 3

Flii^Tg/Tg mice have slower skin re-epithelialisation and increased epidermal thickness. (a-d) Representative images of H&E stained digits, showing the area of skin injury carried out to allow access to the tendon in WT and Flii^Tg/Tg mice at days 3 and 7. Images illustrate the significantly delayed re-epithelialisation and increased epidermal thickness in Flii^Tg/Tg mice compared with WT mice. W, wound; D, dermis; E, epidermis. Arrows indicate edge of wound. Magnification × 10. Scale bar = 200 μM. n = 6. (e) Graph showing average wound gape at day 3 in WT and Flii^Tg/Tg mice. (f) Graph showing average epidermal thickness of WT and Flii^Tg/Tg mice at days 3 and 7. n = 6, mean ± SEM. *p ≤ 0.05, **p ≤ 0.01

Fig. 4

Flii^Tg/Tg mice form smaller adhesions following a 50% partial laceration tendon injury compared with WT counterparts. Representative images of H&E stained digits showing the area of tendon injury and the surrounding adhesions and structures in WT (a-e) and Flii^Tg/Tg (f-j) mice. (k) Graphical representation of adhesion size following a 50% partial laceration, on days 3, 7, 14, 21 and 28 following injury. n = 6. Data represented as mean ± SEM. **p ≤ 0.01. t, tendon; dotted line, adhesion area. Magnification × 4. Scale bar = 200 μM

Fig. 5

Adhesions in Flii^Tg/Tg mice have upregulated Flii levels. Representative images immunostained for Flii expression in WT (a-e) and Flii^Tg/Tg (f-j) mice from days 3, 7, 14, 21 and 28 post-injury. Flii expression is detected as red fluorescence and DAPI staining detected as blue fluorescence. t, tendon; d, dermis; dotted line, adhesion area. Magnification × 10. Scale bar = 200 μM and refers to all images. (k) Graphical representation of Flii expression in the tendon adhesions of WT and Flii^Tg/Tg mice at 7, 14, 21 and 28 days post 50% partial laceration injury. No adhesions were detected at day 3 post injury. n = 6. Data represented as mean ± SEM. *p ≤ 0.05

Fig. 6

Flii^Tg/Tg mice have slower, more organised collagen deposition within the adhesion and significantly decreased adhesion score. (a-d) Representative images of tendon adhesions in WT (a-b) and Flii^Tg/Tg (c-d) mice at days 21 and 28 post-injury, stained with Masson’s trichrome. Dotted line represents adhesion area; t, tendon. Magnification × 20, scale bar = 200 μM. (e) Graph showing significantly decreased collagen deposition in the healing tendons of Flii^Tg/Tg mice compared with WT at day 21 and 28. (f) Graph showing decreased adhesion score indicative of collagen fibres appearing more similar to those of unwounded tendons. n = 6. Data represented as mean ± SEM. *p ≤ 0.05

Fig. 7

WT mice have significantly increased collagen type I expression in the tendon adhesions. Representative images of collagen type I expression (gold) in the tendon adhesions of WT (a-b) and Flii^Tg/Tg (e-f) mice and collagen type III expression (gold) in the tendon adhesions of WT (c-d) Flii^Tg/Tg (g-h) at 21 and 28 days post 50% partial laceration injury. (i) WT mice have significantly upregulated collagen type I levels in the adhesions compared with Flii^Tg/Tg mice. (j) No significant difference was noted in collagen type III levels in WT and Flii^Tg/Tg mice, although detectable levels of collagen type III were found in the Flii^Tg/Tg mice. DAPI is represented as blue fluorescence; t, tendon; d, dermis. Dotted line represents tendon adhesion area. Magnification × 10. Scale bar = 200 μM and refers to all images. Data represented as mean ± SEM. *p ≤ 0.05. n = 6

Fig. 8

Flii^Tg/Tg mice adhesions express higher levels of TGFβ1 and lower levels of TGFβ3 than WT mice. (a) TGFβ1 expression levels in the tendon adhesions and (b) surrounding dermis show significantly elevated expression in Flii^Tg/Tg mice compared with WT mice. (c) TGFβ3 expression levels in the tendon adhesions and (d) surrounding dermis show significantly elevated expression in WT mice compared with Flii^Tg/Tg mice. Data represented as mean ± SEM. *p ≤ 0.05, **p ≤ 0.01. n = 6