Prox1 induces new lymphatic vessel formation and promotes nerve reconstruction in a mouse model of sciatic nerve crush injury

Abstract The peripheral nervous system lacks lymphatic vessels and is protected by the blood–nerve barrier, which prevents lymphocytes and antibodies from entering the neural parenchyma. Peripheral nerve injury results in degeneration of the distal nerve and myelin degeneration causes macrophage aggregation, T lymphocyte infiltration, major histocompatibility complex class II antigen expression, and immunoglobulin G deposition in the nerve membrane, which together result in nerve edema and therefore affect nerve regeneration. In the present paper, we show myelin expression was absent from the sciatic nerve at 7 days after injury, and the expression levels of lymphatic vessel endothelial hyaluronan receptor 1 (LYVE‐1) and Prospero Homeobox 1 (Prox1) were significantly increased in the sciatic nerve at 7 days after injury. The lymphatic vessels were distributed around the myelin sheath and co‐localized with lymphatic endothelial cells. Prox1 induces the formation of new lymphatic vessels, which play important roles in the elimination of tissue edema as well as in morphological and functional restoration of the damaged nerve. This study provides evidence of the involvement of new lymphatic vessels in nerve repair after sciatic nerve injury.

and macrophages, which provide a supportive microenvironment for successful regrowth of the proximal nerve fiber ending (Caillaud et al. 2019).
The lymphatic system plays an important role in the immune system and fluid balance. Disruption of the lymphatic system leads to poor fluid drainage (especially in inflammatory tissues), which results in acute edema (Jamieson et al. 2005;Shin and Rockson, 2008;Swartz et al. 2008). The injury causes hypertonicity of the blood vessels and aggregation of neutrophils, macrophages, and other reactive cells that help in the clearing of dead cells and debris and promote angiogenesis (Nahrendorf et al. 2007; Barbay et al. 2015).
However, several inflammatory factors and free radicals were also found to inhibit lymphatic function, resulting in lymphatic obstruction, edema, and chronic inflammation (Zawieja et al. 1991;Aldrich and Sevick-Muraca, 2013). As lymphatic vessels are absent from the peripheral nerves and there have been no reports of lymphatic vessel involvement in nerve repair after damage, it was worth noting that edema was clinically apparent after peripheral nerve injury (Olmarker et al. 1989).
The formation of new lymphatic vessels is an important link between inflammation and the repair of damaged tissues (Watari et al. 2008). Prospero Homeobox 1 (Prox1) and vascular endothelial growth factor receptor 3 (VEGFR-3) are known to play important roles in inflammation-induced lymphangiogenesis [formation of new lymphatic vessels (Hong et al. 2002;Wigle et al. 2002)], and the nuclear factor-κB (NF-κB) family of transcription factors are key factors regulating inflammation-induced transcription (Karin, 2006). Recent studies reported that inflammation induces increased expression of VEGFR-3 on the lymphatic endothelium, and this effect is mediated by NF-κB and Prox1. This results in a significant increase in the binding affinity of VEGFR-3 to vascular endothelial growth factor (VEGF)-D and VEGF-D on lymphatic endothelial cells, leading to lymphangiogenesis (Olmarker et al. 1989).
Evidence has also shown that adequate intraneural vascularization is found concurrently with the regeneration of nerves (Mizisin and Weerasuriya, 2011;Goedee et al. 2013). Therefore, the present study investigated the expression of Prox1 after sciatic nerve crush injury as well as the presence of new lymphatic vessels in the sciatic nerve and their effects on nerve regeneration and functional recovery.

| Mouse model of sciatic nerve crush injury
Forty mice were randomly divided into control groups (days 7 and 14) and two sciatic nerve injury groups analyzed after 7 and 14 days (n = 10/group). The mice were deeply anesthetized for surgery using a combination of ketamine (50 mg/kg) and xylazine (40 mg/kg). The mice were placed in prone position on mouse plates, and all limbs were taped down. Along the midline of the back of the thigh, an incision was made in the skin. The biceps femoris and semimembranosus muscle were separated, revealing the sciatic nerve at the edge of the gluteus maximus. A length of 1 cm of the sciatic nerve was clamped and unclamped using a vascular clamp for a duration of 10 s each time, for a total of three times. Non-invasive clamps (using #0 surgical sutures) were used to mark the upper and lower end of the sciatic nerve. After the surgery, the skin incision was sutured using #7 surgical sutures. For the control group, the sciatic nerve was exposed without clamping.

| Effects of sciatic nerve injury on the myelin sheath
The four groups of mice were killed at their respective time points (7 or 14 days after surgery), and the control mice were killed on days 7 and 14 after surgery. The clamped sciatic nerve of the right thigh was removed, and the tissues were fixed in 4% formaldehyde and embedded in paraffin. The paraffin blocks were then sectioned at 4 µm, stained with Luxol fast blue (myelin stain), and viewed under a DP-72 optical microscope (Olympus). Some tissues were also fixed in 4% paraformaldehyde solution for 3 h before immersion in 30% sucrose (in phosphate-buffered saline) for dehydration. The tissues were rapidly frozen in liquid nitrogen and sectioned at 20 µm. The sections were further fixed in 1% osmium tetroxide for 30 min, dehydrated using ethanol, and embedded in EPON 812. The sections were then cut into ultra-thin slices and viewed under an HC-1 electron microscope (Hitachi).

| Protein expression of lymphatic vessel endothelial hyaluronan receptor 1, Prox1 and myelin basic protein in the sciatic nerve after injury
The paraffin-embedded sections of mouse tissues were deparaffinized and subjected to antigen retrieval in 0.01 M citrate buffer at 121°C for 15 min. Endogenous peroxidase was inactivated using 3% of hydrogen peroxide. The sections were incubated in rabbit antimouse Prox1 antibody (diluted 1:100; Boster Biological Technology) or goat anti-mouse lymphatic vessel endothelial hyaluronan receptor 1 (LYVE-1) antibody (diluted 1:100; R&D Systems) at 4°C for 24 h, followed by incubation in horseradish peroxidase-labeled goat anti-rabbit IgG or rabbit anti-goat IgG. 3'3'-Diaminobenzidine (DAB) staining was performed using a DAB immunostaining kit (Boster Biological Technology), and the sections were viewed under a DP-72 optical microscope (Olympus). The sections were viewed at 400 × magnification, and the images for each tissue section were captured at 50 different points. Semi-quantitative analysis was performed using Image-Pro Plus 6.0, and the average optical density value was analyzed.

| Statistical analysis
All statistical analysis was performed using SPSS 19.0 statistical software. Comparison between groups was performed using analysis of variance with Tukey post hoc analysis. A value of P < 0.05 was considered statistically significant.

| Myelin reduction and morphological changes on day 7 after sciatic nerve injury
In the control group, the nerve fiber (red) and myelin sheath (blue) were clearly seen in the sciatic nerve ( Figure 1a). However, on the 7th day after nerve injury, the nerve fiber (red) was observed, with degeneration of the myelin sheath in the sciatic nerve ( Figure 1b). On the 14th after nerve injury, the myelin sheath (blue) had reappeared (Figure 1c), and the distance between the nerve membranes had returned to normal (data not shown). Note that as the sciatic nerve tissues were very small, paraffin embedding was difficult, which resulted in variable orientation and positioning of the tissues in the paraffin wax. Thus, obtaining a consistent set of sections for each group was difficult, and this accounts for the different orientations in Figure 1a-c.
Electron microscopy images showed a tightly packed and clearly defined myelin sheath in the sciatic nerve of the control groups ( Figure 2a). However, on the 7th day after nerve injury, the myelin sheath showed morphological changes, appearing disorganized and folded (Figure 2b). On the 14th after the nerve injury, most of the myelin structure had returned to normal (Figure 2c).

| Increased protein expression of Prox1 on day 7 after sciatic nerve injury
A few Prox1-positive cells were observed in the sciatic nerve of the control group ( Figure 4a). However, on the 7th day after F I G U R E 1 Luxol blue staining of the sciatic nerve: (A) control, (B) 7 days after sciatic nerve injury, and (C) 14 days after sciatic nerve injury. The nerve fiber and myelin sheath were clearly seen in the sciatic nerve of the controls. However, only the nerve fiber was observed on the 7th day after nerve injury, with absence of the myelin sheath in the sciatic nerve. On the 14th after nerve injury, the myelin sheath had reappeared. Blue, myelin; red, nerve fiber.

| Co-localization of LYVE-1-positive lymphatic vessels with Prox1-positive lymphatic endothelial cells on day 7 after sciatic nerve injury
The relationship between LYVE-1-positive lymphatic vessels and lymphatic endothelial cells was examined using double immunolabeling with a lymphatic endothelial cell marker, Prox1. The LYVE-1positive cells were labeled in green (Figure 6a), and the Prox1-positive cells were labeled in red (Figure 6b). The results showed that the lymphatic vessels were co-localized with the lymphatic endothelial cells (Figure 6c).

| D ISCUSS I ON
Compression of the peripheral nerve causes damage to the nerve structure and function, which leads to sensory and motor dysfunction, such as carpal tunnel syndrome and cervical and lumbar disc herniation . Clinically, treatment for peripheral nerve compression has achieved favorable outcomes; however, the pathological changes as well as the repair and regeneration mechanisms after peripheral nerve compression remain incompletely understood. In addition, protecting the pinched nerve, reducing secondary damage to the nerve, and restoring its conductive function are difficult problems that have yet to be completely solved in the clinical field. In the early stage of nerve compression, nerve ischemia, hypoxia, and microcirculatory disorders lead to the destruction of the blood-nerve barrier, causing nerve edema (Kaiser and Haninec, 2012). Research has indicated that the degree of neuroinflammatory responses after nerve injury is inversely correlated with functional recovery. Prolonged nerve compression results in nutrient deficiency of the nerve. This leads to thickening of the outer membrane, hyperplasia of the connective tissue, Wallerian degeneration, fibroblast invasion, adhesion formation and permanent scarring, and severe neurological impairment, which affects nerve conduction function (Rydevik et al. 1980;Bouldin et al. 1991;Aziz et al. 1999).
During the early stage of injury, macrophages proliferate and accumulate at the injury site, clearing the ruptured axons and myelin sheath via phagocytosis. At the same time, Schwann cells secrete various neurotrophic factors such as brain-derived neurotrophic factor, nerve growth factor, fibroblast growth factor and leukemia inhibitory factor, forming bands of Bünger in the peripheral nerve.

The neuronal cell body undergoes chromatolysis (axonal reaction)
and develops growth cones on the surface of the retraction ball, which grow toward the target organs at a rate of 2-4 mm/day. The Schwann cells gradually wrap around the axons to form the regenerated myelin sheath (Funakoshi et al. 1993). Elimination of the inflammatory response and edema of the nerve in the early stage of injury could prevent myelin degeneration and fibrous scar formation in the later stage of the injury, which would be beneficial to the functional recovery of the nerve. However, how nerve edema is relieved and whether lymphatic vessels are involved in relieving nerve edema after injury have not been well studied. This is because it is generally believed that the peripheral nerves lack lymphatic vessels and are restricted to outside the neuroepithelium. Research has shown that Prox1 promotes lymphangiogenesis in adult inflammatory conditions (Wigle et al. 2002), and Prox1 siRNA injection in rats was found to inhibit lymphangiogenesis (Rho et al. 2015). During the early stages of embryonic development, VEGFR-3 is expressed on vascular endothelial cells, and its ligand, VEGF-C, plays an important role in the eventual establishment of lymphatic endothelial cells Wigle et al. 2002). VEGFR-3 is a tyrosine kinase receptor mainly expressed on lymphatic endothelial cells. It is one of the key proteins in regulating lymphangiogenesis, and its ligands include VEGF-C and VEGF-D . VEGF-C and VEGF-D to VEGFR-3 (Ristimaki et al. 1998;Cursiefen et al. 2004;Baluk et al. 2005). Although a large body of evidence supports the notion of inflammation-induced lymphangiogenesis, the molecular mechanisms of this association have not been elucidated in detail. In addition, an in vivo study showed that NF-κB and Prox1 activate transcription of VEGFR-3, suggesting important roles for NF-κB and Prox1 in the regulation of VEGFR-3-dependent lymphangiogenesis (Flister et al. 2010).
In the present study, morphological changes and reduction of myelin were observed 7 days after sciatic nerve injury, together with the distribution of lymphatic vessels around the myelin.
Increased expression of Prox1 was also observed in the sciatic nerve at 7 days after injury. Prox1 is known to play a key regulatory role in lymphatic vessel-mediated elimination of tissue edema and morphological and functional restoration of damaged nerve. In addition, Prox1 induced lymphatic endothelial cell proliferation around the sciatic nerve, resulting in formation of new lymphatic vessels, which was observed through an increase in LYVE-1 expression and co-localization of Prox1 and LYVE-1 at 7 days after sciatic nerve injury. This is consistent with a study that showed an increase in the number of vessels after sciatic nerve crush injury (Caillaud et al. 2019). The new lymphatic vessels could effectively reduce tissue edema by increasing the removal of excess interstitial fluid, inflammatory cells, and myelin debris, and therefore, promote the formation of new myelin by the glial cells to enhance the functional recovery of the sciatic nerve. This is supported by the results observed on day 14 after nerve injury, when the myelin sheath had reappeared and the structure had returned to normal.
In conclusion, our results showed increased expression of Prox1 and the presence of lymphatic vessels (LYVE-1) at 7 days after sciatic nerve injury. This increase in lymphatic vessels helped to eliminate edema and restore the morphology and function of the damaged nerve. This study provides evidence of the presence of new lymphatic vessels in nerve repair after sciatic nerve injury in mice.
Further research using Prox1 siRNA is necessary to elucidate the physiological role of Prox1 in the sciatic nerve after injury.

ACK N OWLED G EM ENTS
This work was supported by the Key Subject Grant for the "Integrated Chinese and Western Medicine Foundation" from Shandong College of Traditional Chinese Medicine (No. XZDXK03).

CO N FLI C T S O F I NTE R E S T
None declared.

DATA AVA I L A B I L I T Y S TAT E M E N T
The datasets generated and analyzed during the present study are available from the corresponding author upon reasonable request.