Abstract

Diabetic complications and vascular disease are closely intertwined. Diabetes mellitus is a well-established risk factor for both large and small vessel vascular changes, and conversely other vascular risk factors confer increased risk for diabetic complications such as peripheral neuropathy, nephropathy and retinopathy. Furthermore, axons and blood vessels share molecular signals for purposes of navigation, regeneration and terminal arborizations. We examined blood vessel, Schwann cell and axonal regeneration using validated axotomy models to study and compare patterns and the relationship of regeneration among these different structures. Ten subjects with diabetes mellitus complicated by neuropathy and 10 healthy controls underwent 3 mm distal thigh punch skin biopsies to create an intracutaneous excision axotomy followed by a concentric 4-mm overlapping biopsy at different time points. Serial sections were immunostained against a pan-axonal marker (PGP9.5), an axonal regenerative marker (GAP43), Schwann cells (p75) and blood vessels (CD31) to visualize regenerating structures in the dermis and epidermis. The regenerative and collateral axonal sprouting rates, blood vessel growth rate and Schwann cell density were quantified using established stereology techniques. Subjects also underwent a chemical ‘axotomy’ through the topical application of capsaicin, and regenerative sprouting was assessed by the return of intraepidermal nerve fibre density through regenerative regrowth. In the healed 3 mm biopsy sites, collateral and dermal regenerative axonal sprouts grew into the central denervated area in a stereotypic pattern with collateral sprouts growing along the dermal–epidermal junction while regenerative dermal axons, blood vessels and Schwann cells grew from their transected proximal stumps into the deep dermis. Vessel growth preceded axon and Schwann cell migration into the denervated region, perhaps acting as scaffolding for axon and Schwann cell growth. In control subjects, Schwann cell growth was more robust and extended into the superficial dermis, while among subjects with diabetes mellitus, Schwann tubes appeared atrophic and were limited to the mid-dermis. Rates of collateral (P = 0.0001), dermal axonal regenerative sprouting (P = 0.02), Schwann cell migration (P < 0.05) and blood vessel growth (P = 0.002) were slower among subjects with diabetes mellitus compared with control subjects. Regenerative deficits are a common theme in diabetes mellitus and may underlie the development of neuropathy. We observed that blood vessel growth recapitulated the pattern seen in ontogeny and preceded regenerating nerve fibres, suggesting that enhancement of blood vessel growth might facilitate axonal regeneration. These models are useful tools for the efficient investigation of neurotrophic and regenerative drugs, and also to explore factors that may differentially affect axonal regeneration.

Introduction

Epidemiological studies indicate that 25–60% of diabetic patients develop a progressive peripheral neuropathy that leads to significant morbidity (Ziegler et al., 1992; Dyck et al., 1993). Axonal atrophy and degeneration are the pathological hall marks for the development of severe sensory loss and have been shown to occur both in nerve and skin biopsies of diabetic patients (Dyck et al., 1986; Lehtinen et al., 1989; Malik et al., 2001; Polydefkis et al., 2001; Singleton et al., 2005; Quattrini et al., 2008).

Many mechanisms including direct metabolic effects due to hyperglycaemia, oxidative stress, mitochondrial dysfunction and disruption of axonal transport by advanced glycation end products have been implicated in the pathogenesis of diabetic polyneuropathy (Jakobsen and Sidenius, 1979; Jakobsen et al., 1981; Katz et al., 1989; Jaap et al., 1996; Brownlee, 2001; Gibran et al., 2002; Yasuda et al., 2003; Kennedy and Zochodne, 2005). In addition to peripheral nerve abnormalities, vascular abnormalities are also present in diabetes mellitus, and human and experimental studies have suggested that diabetes impairs ischaemia-driven neovascularization (Yasuda and Dyck, 1987; Yarom et al., 1992; Rivard et al., 1999).

In addition to the loss of sensory fibres resulting in neurological deficits, regenerative failure has been observed in experimental nerve lesions of animals (Kennedy and Zochodne, 2000). We have also observed that regenerative axonal growth is impaired in human subjects with diabetes even in the setting of no evidence of peripheral neuropathy (Polydefkis et al., 2004). Previous studies have shown that molecular signals are shared by nerve and blood vessel growth for purposes of navigation, regeneration and terminal arborization (Carmeliet and Tessier-Lavigne, 2005) though the role of blood vessel growth in regenerative nerve failure in diabetic subjects is not clear.

Experimental studies have shown that topical capsaicin application and excision axotomy models can be used to study nerve regeneration (Rajan et al., 2003; Polydefkis et al., 2004; Hahn et al., 2006; Ebenezer et al., 2009). Capsaicin application produces a ‘pure’ chemical axotomy while excision of the 3 mm biopsy core produces a complete axonal and blood vessel injury. We used these models to define and compare the growth of blood vessels, Schwann cells and axonal regeneration in healthy control subjects and subjects with diabetes. This approach allows us to dissect and study the different components of axonal and blood vessel growth in the skin of human subjects.

Materials and methods

Study subjects

Healthy control subjects were determined to be neuropathy-free by neurological examination, symptom score, and were screened by laboratory testing for peripheral neuropathy risk factors (HIV serology, vitamin B12, fasting blood glucose, glycated haemoglobin, thyroid-stimulating hormone, complete blood count and metabolic panel). The median age of controls was 47 years, range 23–60 (Table 1).

Table 1

Demographics of study subjects

 n Age* Gender* (M/F) Height (cm) Race (% Caucasian) 
DM 10 51 (37–64) 5/5 170 (150–191) 70 
Control 10 47 (23–60) 6/5 166 (149–179) 70 
 n Age* Gender* (M/F) Height (cm) Race (% Caucasian) 
DM 10 51 (37–64) 5/5 170 (150–191) 70 
Control 10 47 (23–60) 6/5 166 (149–179) 70 

Median (range); DM = diabetes mellitus.

Subjects with diabetes had a median age of 51 years, range 37–64 (P = 0.24) (Table 1). Subjects with diabetes were examined by a neurologist trained in neuromuscular disease and determined to have peripheral neuropathy attributed to diabetes. All subjects had Michigan Diabetic Neuropathy Score 18 > (MDNS) >6. Diabetes was confirmed by review of medical records. Diabetes type and absence of retinopathy were determined by subject report and proteinuria was assessed through a urine dipstick. No subject had a history of retinopathy while four subjects had proteinuria (Table 2). Electrophysiology was performed on the subjects with diabetes using a uniform protocol in order to further objectively quantify the neuropathy (Table 3). All studies were approved by the Johns Hopkins University Institutional Review Board.

Table 2

Glycaemic control and diabetes complications among subjects with diabetes mellitus

 n Age (years) A1C (%) Weight (kg) Neuropathy duration (years) Diabetes duration (years) Proteinuria (1+ or trace) 
Diabetes subjects (all) 10 51 (37–64) 7.4 (5.9–10.1) 103 (68–148) 7 (1–10) 13 (0.5–34) 
Type I 39 (37–47) 7.0 (6.4–9.9) 76 (68–82) 5 (1–10) 28 (5–34) 
Type II 60 (44–64) 7.7 (5.9–10.1) 116 (88–148) 8 (3–10) 10 (0.5–17) 
 n Age (years) A1C (%) Weight (kg) Neuropathy duration (years) Diabetes duration (years) Proteinuria (1+ or trace) 
Diabetes subjects (all) 10 51 (37–64) 7.4 (5.9–10.1) 103 (68–148) 7 (1–10) 13 (0.5–34) 
Type I 39 (37–47) 7.0 (6.4–9.9) 76 (68–82) 5 (1–10) 28 (5–34) 
Type II 60 (44–64) 7.7 (5.9–10.1) 116 (88–148) 8 (3–10) 10 (0.5–17) 

All values represent medians (range).

Table 3

Electrophysiological data of diabetic patients

Test Median motor
 
Median sensory
 
Peroneal motor
 
Sural sensory
 
 F-wave Amplitude c.v. Amplitude c.v. Amplitude c.v. Amplitude c.v. 
Median, range 30.5, 25.2–41.4 6.25, 3.8–12.3 52.8, 40.8–57.9 9.3, 1.8–38.2 45.3, 31.9–54.3 2.15, 0.1–4.7 43, 35.4–45.1 2.15, 0–9 38.4, 31.1–53.8 
Test Median motor
 
Median sensory
 
Peroneal motor
 
Sural sensory
 
 F-wave Amplitude c.v. Amplitude c.v. Amplitude c.v. Amplitude c.v. 
Median, range 30.5, 25.2–41.4 6.25, 3.8–12.3 52.8, 40.8–57.9 9.3, 1.8–38.2 45.3, 31.9–54.3 2.15, 0.1–4.7 43, 35.4–45.1 2.15, 0–9 38.4, 31.1–53.8 

F-wave latency in ms, motor amplitudes in mV. c.v. = conduction velocity in m/s. Sensory amplitudes are reported in microvolts. Normal values: ≥10 mV for age <60 years; ≥5 mV for age ≥60 years. CV ≥49 m/s in the upper extremity; ≥39 m/s in the lower extremity.

Capsaicin model

All subjects in the study underwent both capsaicin denervation as well as excision biopsies. For the capsaicin ‘chemical axotomy’, a standardized area was demarcated on the distal lateral thigh. An occlusive bandage measuring 35 × 50 mm, and containing 1.8 g of 0.1% capsaicin cream (Chattem Inc), was applied to the area for two consecutive 24 h periods as previously described (Polydefkis et al., 2004). Punch skin biopsies were obtained under local anaesthesia at baseline (Day 0), immediately after the 48 h capsaicin period and at 30 days, 60 days for all subjects and 90 days in subjects with diabetes.

Excision model

All subjects also underwent a 3 mm distal thigh punch skin biopsy (Acupunch™, Acuderm) to create an intracutaneous excision axotomy. This was followed by a 4 mm overlapping biopsy after 2–3 months (Fig. 1) (Rajan et al., 2003; Ebenezer et al., 2009). The first biopsy produces a denervated and avascular zone that subsequently fills in; the subsequent biopsy allows assessment of nerve and blood vessel regrowth and expansion (Rajan et al., 2003; Hahn et al., 2006; Ebenezer et al., 2009; Griffin et al., 2010).

Figure 1

Diagrammatic illustration of intracutaneous axotomy model. (A) Circular cutaneous axotomies were performed on the distal thigh using a 3-mm skin punch to transect epidermal axons. The central core containing epidermis and dermis was then removed (arrow) producing an excisional axotomy and vessel transection. (B) The site was allowed to heal by a process of granulation. Collateral nerve sprouts from uninjured epidermal nerve fibres grew centripetally (arrow). Regenerative sprouts and Schwann cells (arrow head) from the distal transected end of dermal axons and vessels (broken arrow) extended into the dermis. A second, concentric 4-mm circular biopsy punch was performed to capture these different tissues. (C) The insert identifies the central dermal region. Schwann cell bands are horizontally oriented (arrow head) and some ensheath dermal axons (arrow).

Figure 1

Diagrammatic illustration of intracutaneous axotomy model. (A) Circular cutaneous axotomies were performed on the distal thigh using a 3-mm skin punch to transect epidermal axons. The central core containing epidermis and dermis was then removed (arrow) producing an excisional axotomy and vessel transection. (B) The site was allowed to heal by a process of granulation. Collateral nerve sprouts from uninjured epidermal nerve fibres grew centripetally (arrow). Regenerative sprouts and Schwann cells (arrow head) from the distal transected end of dermal axons and vessels (broken arrow) extended into the dermis. A second, concentric 4-mm circular biopsy punch was performed to capture these different tissues. (C) The insert identifies the central dermal region. Schwann cell bands are horizontally oriented (arrow head) and some ensheath dermal axons (arrow).

Immunohistochemistry

Biopsies were fixed in paraformaldehyde/lysine/periodate, cryoprotected and sectioned at 50 µm intervals. Four sections from each biopsy were selected, bleached in 0.25% potassium permanganate to remove melanin pigment, rinsed in Tris-buffered saline and washed with 5% oxalic acid for 2 min. Blocking was performed in Tris-buffered saline 3% normal serum with 0.4% Triton X-100. The sections were incubated overnight at 4°C with the primary antibodies suspended in 0.1% Triton X-100 and 3% normal serum in Tris-buffered saline.

Primary antibodies used were PGP 9.5 (Chemicon, dilution 1:2000 in triton buffer), mouse anti-growth associated protein (GAP) -43/B-50 (Millipore, dilution 1:500 in triton buffer), mouse anti-p75 nerve growth factor receptor (Chemicon, dilution 1:500 in triton buffer) and mouse anti-CD31, PECAM (BD Pharmingen, dilution 1:100 in Triton buffer).

Non-specific binding of secondary antibodies was blocked with 4% normal goat serum, 1.0% Triton X-100, 0.5% non-fat powdered milk in Tris-buffered saline, pH 7.4. Sections were incubated in secondary antibody (goat anti-rabbit IgG and goat anti-mouse IgG) diluted 1:200 in Tris-buffered saline 0.1%Triton X-100, 3% normal serum for 1 h at room temperature. Endogenous peroxidase was blocked with 0.5% H2O2 in methanol and sections were then incubated with Avidin-Biotin Complex solution (Vector Labs) for 1 h followed by substrate incubation for 5–6 min (Vector SG substrate and diaminobenzidine). Eosin (1%) and Mayer's haematoxylin were used as counterstains.

Immunofluorescence

Sections were rinsed in Tris-buffered saline, pH 7.4, and incubated with secondary antibodies prepared in 2% normal goat serum, 0.5% Triton X-100, 0.5% non-fat powdered milk for 2 h at room temperature on a shaker table. The secondary antibodies Cy3 goat anti-rabbit IgG (Jackson Immuno Research Laboratories, Inc) were used at 1:150 dilution and Alexa-Fluor goat anti-mouse IgG (Molecular Probes) at 1:500 dilutions. The fluorescence-labelled sections were mounted in Mowiol 4-88 (Calbiochem) to prevent fluorescent quenching. DRAQ5 (Biostaus Ltd) was used for nuclear staining. Fluorescent samples were viewed and scanned using a Zeiss LSM510 confocal imaging system.

Intraepidermal nerve fibre density calculations were determined using established counting rules and the rate of regeneration among the capsaicin series biopsies was assessed by the return of intraepidermal nerve fibre density following confirmed denervation. The regeneration rate was determined from the slope of the regression line of the Day 2, 30, 60 and Day 90 biopsies, as previously described (Polydefkis, 2004).

Morphometric studies

During healing of the 3-mm biopsy, a collagen plug (scar) forms in the subepidermal dermis and the 3-mm excision site is easily identified histologically by the homogenous densely stained collagen in the dermis and the corresponding thinned overlying epidermis.

Measurement of blood vessels and axonal regeneration in dermis

The area of interest was defined as the dermal region extending from the epidermal/dermal junction to a dermal depth of 2000 μm. The length of blood vessels and dermal nerves growing into the excision site were measured using a design-based unbiased stereology methodology. With ×2.5/0.075 Plan-Neofluor objective of a Zeiss light microscope, the area of interest was marked and under ×63/1.40, oil Plan-Neofluor objective using hemisphere probe with a radius of 15 µm of the Stereo Investigator space ball probe (Fig. 2B) (MBF Bioscience) (Mouton et al., 2002; Tang et al., 2009), the growth rate was expressed as length of nerve or vessels divided by the number of days separating the 3 and 4 mm biopsies (mm/day). The Gundersen coefficient of error for the parameters described was 0.07 for blood vessels and 0.12 for dermal axons.

Figure 2

Immunohistochemical staining of skin sections for Schwann cells (p75) and vascular bundles (CD31) overlayed by design-based sampling grids for morphometric methods. (A) Skin section showing p75 positive Schwann cell bands (arrow, brown) with Schwann cell nuclei (the cells counted marked x) in the dermis overlaid by optical fractionator counting frame. Scale bar = 50 µm. (B) Skin sections showing CD31-positive blood vessels (arrow, brown) overlaid by space ball probe (star indicates counting sites). Scale bar = 50 µm.

Figure 2

Immunohistochemical staining of skin sections for Schwann cells (p75) and vascular bundles (CD31) overlayed by design-based sampling grids for morphometric methods. (A) Skin section showing p75 positive Schwann cell bands (arrow, brown) with Schwann cell nuclei (the cells counted marked x) in the dermis overlaid by optical fractionator counting frame. Scale bar = 50 µm. (B) Skin sections showing CD31-positive blood vessels (arrow, brown) overlaid by space ball probe (star indicates counting sites). Scale bar = 50 µm.

Measurement collateral axonal sprouts

The collateral sprouting distance was defined as the farthest centripetal distance from the 3-mm biopsy incision and was measured along the basement membrane of the epidermis with Bioquant software (R&M Biometrics). The collateral sprouting distance divided by the number of days between the 3 and 4 mm biopsies was calculated to yield the growth rate expressed in mm/day (Hahn et al., 2006).

Statistics

Comparisons between groups were made using the non-parametric Wilcoxon rank-sum test. To assess the effect of age or gender on growth rates, linear regression was performed using the growth rate as the dependent variable and the presence/absence of diabetes, age and gender as independent variables.

Histomorphometry of Schwann cells

Schwann cell bands within the dermis were identified in the skin sections by the presence of brown coloured cytoplasmic staining by p75 in the setting of elongated blue-staining nuclei (counter stained with Mayer's haematoxylin) (Fig. 2A).

Schwann cell nuclei were quantified using a similar approach as was done for the blood vessel and dermal axon measurements. A contour was made at the area of interest under a ×2.5/0.075 Plan-Neofluor objective of a Zeiss light microscope and the cells were counted using ×63/1.40, oil Plan-Neofluor objective. A sampling grid 200 µm × 150 µm, with a dissector height of 15 µm and guard zones of 2 µm (Fig. 2A) was used and the absolute numbers of Schwann cells were obtained using the ‘.DAT’ files of optical fractionator probe of Stereo Investigator (MBF Bioscience) (West et al., 1991; Peterson and Jones, 1993). The sampling design achieved a Gundersen coefficient of error 0.07. The density of Schwann cells in the dermis (cells/mm3) was determined and the rate of Schwann cell growth/day calculated.

All morphometric measurements were performed blinded of all clinical or identifying data.

Microscopic observation and photography

Preparations were viewed with Olympus BH2 and Axioscop microscopes and photographed with digital cameras. Using Adobe Photoshop CS v. 8.0 Software (Adobe Systems), the photomicrographs were adjusted evenly across the entire field for contrast and brightness levels, for optimal quality and positioned into the montages.

Results

Capsaicin model

There was complete epidermal denervation in both control and diabetes mellitus subject skin biopsies taken 48 h after capsaicin application (Fig. 3A and D). GAP43 expression was present in deep dermal nerve fibres in these biopsies and co-localized with PGP9.5 staining fibres (Fig. 3A and B). By Day 60, the papillary dermal nerve bundles showed increased GAP43 staining intensity, but only PGP9.5 staining fibres predominantly sprouted into the epidermis (Fig. 3C and D) confirming that PGP staining is the best measure for fibres for sprouting into the epidermis. From the baseline loss of fibres after chemical axotomy, the fibres progressively regenerated back into the epidermis (Fig. 3D) and by Day 60, attained the mean normative density (controls: 10 fibres/mm, diabetes mellitus: 7 fibres/mm) (McArthur et al., 1998). The controls showed a steady growth but the regenerative capacity lagged behind at all time points in subjects with diabetes mellitus (Fig. 3E) and findings has been consistent with our previous studies (Polydefkis et al., 2004) The median rate of post-capsaicin reinnervation among the subjects with diabetes was 0.092 (range 0.01–0.19) fibres/mm/day and was reduced compared with control subjects [median rate 0.19 fibres/mm/day, range (0.05–0.347)] consistent with our previous studies. This difference in regeneration rate persisted if baseline differences in epidermal nerve fibre density, age or gender were adjusted for.

Figure 3

Double stained confocal images of PGP9.5 (red) with GAP43 (green) on the cutaneous regenerative fibres after chemical axotomy. (A) Forty-eight hours post capsaicin biopsy showing loss of epidermal fibres and few PGP9.5 expressed fibres (red arrow) remaining in the dermis. Note the absence of GAP43-positive fibres. (B) A large dermal nerve bundle reveals PGP 9.5-positive fibres (red arrow) co-localizing with GAP43 fibres (green arrow). (C) At Day 60, the PGP9.5 positive fibres regenerating (red arrow) into the epidermis and the Gap43-positive fibres remain at the level of the papillary dermis (yellow arrow). (D) The deep dermal larger nerve bundles revealing dense expression of GAP43 co-localizing with PGP9.5 fibres. Scale bar = 20 µm. (E) The regenerative rate after chemical axotomy was consistently lower in subjects with diabetes at all post capsaicin time points.

Figure 3

Double stained confocal images of PGP9.5 (red) with GAP43 (green) on the cutaneous regenerative fibres after chemical axotomy. (A) Forty-eight hours post capsaicin biopsy showing loss of epidermal fibres and few PGP9.5 expressed fibres (red arrow) remaining in the dermis. Note the absence of GAP43-positive fibres. (B) A large dermal nerve bundle reveals PGP 9.5-positive fibres (red arrow) co-localizing with GAP43 fibres (green arrow). (C) At Day 60, the PGP9.5 positive fibres regenerating (red arrow) into the epidermis and the Gap43-positive fibres remain at the level of the papillary dermis (yellow arrow). (D) The deep dermal larger nerve bundles revealing dense expression of GAP43 co-localizing with PGP9.5 fibres. Scale bar = 20 µm. (E) The regenerative rate after chemical axotomy was consistently lower in subjects with diabetes at all post capsaicin time points.

Excision axotomy model

The skin sections showed a thin layer of epithelialization and basement membrane formation in the central denervated zone and the excision edges showed epithelial hyperplasia. The central dermal core area was easily observed with identifiable fibrin, fibroblasts and interlacing bundles of collagen (Fig. 1). Blood vessel growth into the healed excision site was robust in both groups of subjects and preceded Schwann cell and axonal sprouting into denervated regions. The proliferating blood vessels were aligned perpendicularly, towards the epidermis and extended into the papillary dermis (Fig. 4A). Blood vessel growth was significantly delayed (P = 0.004) among subjects with diabetes compared with controls (Fig. 4B).

Figure 4

Regrowth of blood vessels at the excision site. (A) Photomicrograph from a control showing blood vessels (arrows) arranged in a longitudinal pattern and directed towards the epidermis. Scale bar = 100 µm. (B) The growth rate of blood vessels was significantly delayed (P = 0.002) among subjects with diabetes mellitus, compared with controls.

Figure 4

Regrowth of blood vessels at the excision site. (A) Photomicrograph from a control showing blood vessels (arrows) arranged in a longitudinal pattern and directed towards the epidermis. Scale bar = 100 µm. (B) The growth rate of blood vessels was significantly delayed (P = 0.002) among subjects with diabetes mellitus, compared with controls.

The earliest axonal growth into the excision site was the collateral sprouts from epidermal nerve fibres just peripheral to the excision margin; these sprouts grew in a stereotypic pattern along the dermal–epidermal junction close to the basement membrane on the side of the basal keratinocytes (Fig. 5A). Few axons extended vertically towards the skin surface to produce intraepidermal axons into the stratum spinosum and when this occurred these were seen only at the edge of the axotomy site. These axons were ensheathed by Schwann cells only for a short distance along the dermal–epidermal junction (Fig. 5B). The longest collateral sprouting distance was 2473.3 µm in length in control subjects and 1569.3 µm in subjects with diabetes and the rate of collateral sprouts lengthening rate was significantly slower (P = 0.003) among subjects with diabetes (median 0.01 mm/day, range 0.007–0.021) in comparison to controls (median 0.020 mm/day, range 0.016–0.044) (Fig. 5C).

Figure 5

Regrowth of axons at the axotomy site. (A) From the lateral edge of the 3 mm excision site (broken line), the collateral sprouts extended from the uninjured fibres on the periphery of the original 3 mm incision line along the basement membrane (arrows) of the epidermis. Scale bar = 100 µm. (B) Nerve bundles (arrows) lying in horizontal position and the Schwann cells (green) containing few axonal fragments co-localizing (yellow) with Schwann cells (red: PGP9.5). The broken line indicates the lateral edge of the 3 mm excision site. Scale bar = 50 µm. (C) Measurement of the length of sprouting fibres demonstrated that the collateral growth rate (mm/day) was significantly lower among subjects with diabetes mellitus. (D) Measurement of the length of dermal fibres demonstrated that the regenerative growth rate (mm/day) was significantly lower among subjects with diabetes mellitus.

Figure 5

Regrowth of axons at the axotomy site. (A) From the lateral edge of the 3 mm excision site (broken line), the collateral sprouts extended from the uninjured fibres on the periphery of the original 3 mm incision line along the basement membrane (arrows) of the epidermis. Scale bar = 100 µm. (B) Nerve bundles (arrows) lying in horizontal position and the Schwann cells (green) containing few axonal fragments co-localizing (yellow) with Schwann cells (red: PGP9.5). The broken line indicates the lateral edge of the 3 mm excision site. Scale bar = 50 µm. (C) Measurement of the length of sprouting fibres demonstrated that the collateral growth rate (mm/day) was significantly lower among subjects with diabetes mellitus. (D) Measurement of the length of dermal fibres demonstrated that the regenerative growth rate (mm/day) was significantly lower among subjects with diabetes mellitus.

In the central denervated zone, axons from the cut edge of the dermal nerves at the base of the excision grew through the dermal collagen in small bundles. These nerve bundles were oriented in a predominantly horizontal direction parallel to the epidermal surface (Figs 1B and 5B) and relatively few were directed perpendicularly towards the skin surface, suggesting attempted but incomplete regenerative spouting by the injured fibres. In both control and diabetes subjects, the regenerative sprouts never reached the papillary dermis and the growth rate was significantly lower among subjects with diabetes compared with controls, P = 0.01 (Fig. 5D).

Schwann cell growth and migration was more robust at the base of the healed excisional biopsy site with cells appearing as elongated bands. These became thinner and more clumped as they approached the superficial dermis (Fig. 6A). Schwann cell growth extended into the papillary dermis in control subjects. Among subjects with diabetes mellitus, growth was predominantly limited to the mid-dermis and relatively few Schwann cell bands in the superficial dermis contained axonal fragments and co-localized with PGP 9.5 (Fig. 6B). Among the subjects with diabetes mellitus, the Schwann cell migration rate was slowed (P < 0.05) compared with control subjects [median 143 cells/day (range 100–213) versus median 96 cells/day (range 42–185)] (Fig. 6C).

Figure 6

Regrowth of Schwann cells. (A) A photomicrograph from a control subject showing robust proliferation of Schwann cells (green, arrows) at the base of the scar, close to the injured fibre site. (B) Photomicrograph from a subject with diabetes mellitus showing the denervated site at Day 84, containing few Schwann cell bands (arrow) with axons at the scar site. Note the sprouts covered by Schwann cells entering the denervated site close to the basement membrane. The broken line indicates the edge of the excision site. (C) The figure shows the significant slow migration of Schwann cells in diabetes mellitus. Scale bar = all 50 µm.

Figure 6

Regrowth of Schwann cells. (A) A photomicrograph from a control subject showing robust proliferation of Schwann cells (green, arrows) at the base of the scar, close to the injured fibre site. (B) Photomicrograph from a subject with diabetes mellitus showing the denervated site at Day 84, containing few Schwann cell bands (arrow) with axons at the scar site. Note the sprouts covered by Schwann cells entering the denervated site close to the basement membrane. The broken line indicates the edge of the excision site. (C) The figure shows the significant slow migration of Schwann cells in diabetes mellitus. Scale bar = all 50 µm.

Because our groups differed slightly with respect to gender and age (Table 1), we investigated the effect of both these variables. In multivariable regression models, gender was not associated with the growth rates of any epidermal or dermal structure. Age did reach significance in several instances, though the magnitude of the effect was small compared with the association with diabetes, and diabetes remained a significant variable even after adjusting for age (Supplementary Tables 1 and 2).

Discussion

In this study, we efficiently measured different forms of axonal, Schwann cell and vascular growth using relatively non-invasive 3 mm and overlapping 4 mm skin punches in the same subjects. These tools will allow us to dissect different mechanisms and components of neurovascular outgrowth. There are several notable findings from this study. First, we observed that ‘pure’ axonal regenerative sprouting after capsaicin chemical axotomy occurred at a constant rate and was more complete than collateral sprouting, neovascularization or regenerative sprouting of the dermal Schwann cell/axon complex. Second, we observed that blood vessel growth into the excision site preceded regenerating dermal nerve fibres in both subject groups, perhaps providing scaffolding for subsequent axon and Schwann cell growth. Third, collateral sprouting from uninjured nerve fibres was more robust than regenerative sprouting of transected dermal nerve fibres. Finally, growth of blood vessels, dermal nerve fibres and Schwann cells into the axotomy site were significantly reduced among subjects with diabetes compared with healthy controls and appears to be a common theme in diabetes mellitus.

We assessed axonal sprouts through immunohistochemical staining against ubiquitin hydrolase, which is a cytoplasmic marker. It is possible that expression of this protein is altered during regeneration and that other markers would be preferable to assess nascent sprouts. In the past, we observed co-localization between PGP 9.5 and a membrane-bound axonal marker, Gα0, suggesting that this was not the case. In the present study, we also evaluated staining against GAP43, a known regenerative marker. Dermal axons were strongly GAP43/PGP9.5 positive while epidermal nerve fibres were visualized primarily with PGP9.5, which is consistent with PGP9.5 being an appropriate marker for regeneration studies.

Studies have shown that diabetes contributes to impaired axon sprouting (Bisby, 1980) and elongation (Ekstrom and Tomlinson, 1989) in experimental rat models. Schwann cells are essential for guiding regenerating axons to the denervated targets and proliferating Schwann cells within the basal lamina of axons and are regulated by nerve growth factors (Anton et al., 1994; Terenghi, 1999; Bentley and Lee, 2000; Hoke et al., 2002; Griffin, 2006; Song et al., 2006; Taniguchi et al., 2007).

In rodent models of diabetes, there are deficits in nerve growth factor expression (Tomlinson et al., 1996, 1997) and under experimental conditions, high glucose levels are implicated in neuritic outgrowth deficits (Tomlinson et al., 1996, 1997; Tosaki et al., 2008). The reduced regenerative and collateral sprouting that we observed in diabetes mellitus may be attributable to deficits in nerve growth factor production. Alternatively, or in addition, diabetes mellitus may lead to a cascade of events that result in an inhospitable environment for axonal outgrowth to occur in. Advanced glycation end product residue content of endoneurial extracellular matrix proteins increases markedly in STZ-induced diabetes, and glycation of extracellular matrix proteins causes loss of charge and structural distortion of extracellular matrix (Duran-Jimenez et al., 2009). Glycation of laminin and fibronectin causes a reduction in neurotrophin-stimulated neurite outgrowth (Charonis and Tsilbary, 1992; Thornalley, 2002). Studies also suggest that hypoxia causes injury of Schwann cells resulting in a significant decrease in Schwann cell survival (Zhu et al., 2010). The slow proliferation and migration of Schwann cells, and subsequently axons, through the dermal collagen in our study suggests that the surrounding extracellular matrix was relatively hostile and retarded axonal regrowth.

Vascular regrowth after injury is critical for re-establishment of metabolic support as well as production and delivery of growth factors. Vascular growth factors have been implicated in neuronal patterning, migration and neural repair (Hobson et al., 2000; Bates et al., 2003; Schwarz et al., 2004; Yu et al., 2008) and therefore may be well suited as a treatment target. We observed, as have others, that blood vessel and nerve growth recapitulated the pattern seen in ontogeny (Tesfaye et al., 1993; Carmeliet and Tessier-Lavigne, 2005; Eichmann et al., 2005). Subjects with diabetes mellitus exhibited slow proliferation and migration of endothelial cells in the dermal skin possibly attributable to extracellular matrix alterations, which supports previous observations that in diabetes, the endothelial progenitor cells are functionally deficient in their ability to respond to stresses (Tepper et al., 2002; Sigaudo-Roussel et al., 2004; Capla et al., 2007).

Our study indicates that regenerative deficits are a common theme in subjects with diabetes, extend beyond deficits in axonal regrowth and may underlie the development of neuropathy. The local environment in diabetic subjects appears to be deleterious to the growth, maturation, migration of both neuronal (axons and Schwann cells) and vascular structures. Vascular risk factors such as hypertension, elevated cholesterol and smoking have all been associated with the development of peripheral neuropathy in diabetes (Tesfaye and Selvarajah, 2005). Furthermore, axonal and vascular growth ligands, the semaphorins and vascular endothelial growth factors, are known to share a common receptor in neuropilin-1 (Gu et al., 2003). It is possible that diabetes mellitus leads to a cascade of events whereby advance glycation end product formation results in altered neuropilin-1 expression (Bondeva et al., 2009), which in turn mitigates semaphorin and vascular endothelial growth factor activity resulting in slowed axonal and vascular outgrowth.

Vascular endothelial growth factor-A is a well-known angiogenic factor, now considered to have neurotrophic, neuroprotective and neurogenesis functions (Sondell et al., 1999; Sondell et al., 2000; Jin et al., 2002; Gu et al., 2003; Schwarz et al., 2004; Gora-Kupilas and Josko, 2005; Brockington et al., 2006). Preliminary studies have shown that plasmid-mediated delivery of vascular endothelial growth factor-A can reverse neuropathy in diabetic animals and has been shown to improve diabetic neuropathy symptoms in a clinical trial (Schratzberger et al., 2001; Ropper et al., 2009). Our results suggest that such a strategy may be attractive as it may address the initial abnormality in neurovascular repair that appears to be central to diabetic poly-neuropathy. Our observations might also have implications for the design of cancer chemotherapy regimens since bevacizumab (Avastin™) is increasingly used for treatment of subjects with metastatic cancer of the colon or rectum, vestibular schwannomas and glioblastoma multiforme. To our knowledge, no studies have been conducted to evaluate peripheral nerve function in bevacizumab recipients.

This study demonstrates slowed neuronal and vascular growth in subjects with diabetes compared with age-matched controls following standardized injury models. This finding is consistent with alterations in the extracellular matrix and/or alterations in the shared molecular signals responsible for axonal and vascular growth playing a central role in the development of diabetic polyneuropathy. We suggest that the methods developed here are well suited to study vascular and axonal growth in regenerative treatment trials and to dissect mechanisms influencing vascular and nerve growth.

Funding

Juvenile Diabetes Research Foundation (to M.P., G.J.E., R.O.); 1P30MH075673 (to J.C.M.); UL1RR025005 (CRC to J.H.U.).

Supplementary material

Supplementary material is available at Brain online.

Acknowledgements

The authors thank Erin Burke Voelkel and Scott Sultzaberger for expert technical assistance and Dr Jack Griffin for helpful discussions.

Abbreviations

    Abbreviations
  • GAP

    growth-associated protein

References

Anton
ES
Weskamp
G
Reichardt
LF
Matthew
WD
Nerve growth factor and its low-affinity receptor promote Schwann cell migration
Proc Natl Acad Sci USA
 , 
1994
, vol. 
91
 (pg. 
2795
-
9
)
Bates
D
Taylor
GI
Minichiello
J
Farlie
P
Cichowitz
A
Watson
N
, et al.  . 
Neurovascular congruence results from a shared patterning mechanism that utilizes Semaphorin3A and Neuropilin-1
Dev Biol
 , 
2003
, vol. 
255
 (pg. 
77
-
98
)
Bentley
CA
Lee
KF
p75 is important for axon growth and schwann cell migration during development
J Neurosci
 , 
2000
, vol. 
20
 (pg. 
7706
-
15
)
Bisby
MA
Axonal transport of labeled protein and regeneration rate in nerves of streptozocin-diabetic rats
Exp Neurol
 , 
1980
, vol. 
69
 (pg. 
74
-
84
)
Bondeva
T
Ruster
C
Franke
S
Hammerschmid
E
Klagsbrun
M
Cohen
CD
, et al.  . 
Advanced glycation end-products suppress neuropilin-1 expression in podocytes
Kidney Int
 , 
2009
, vol. 
75
 (pg. 
605
-
16
)
Brockington
A
Wharton
SB
Fernando
M
Gelsthorpe
CH
Baxter
L
Ince
PG
, et al.  . 
Expression of vascular endothelial growth factor and its receptors in the central nervous system in amyotrophic lateral sclerosis
J Neuropathol Exp Neurol
 , 
2006
, vol. 
65
 (pg. 
26
-
36
)
Brownlee
M
Biochemistry and molecular cell biology of diabetic complications
Nature
 , 
2001
, vol. 
414
 (pg. 
813
-
20
)
Capla
JM
Grogan
RH
Callaghan
MJ
Galiano
RD
Tepper
OM
Ceradini
DJ
, et al.  . 
Diabetes impairs endothelial progenitor cell-mediated blood vessel formation in response to hypoxia
Plast Reconstr Surg
 , 
2007
, vol. 
119
 (pg. 
59
-
70
)
Carmeliet
P
Tessier-Lavigne
M
Common mechanisms of nerve and blood vessel wiring
Nature
 , 
2005
, vol. 
436
 (pg. 
193
-
200
)
Charonis
AS
Tsilbary
EC
Structural and functional changes of laminin and type IV collagen after nonenzymatic glycation
Diabetes
 , 
1992
, vol. 
41
 
Suppl 2
(pg. 
49
-
51
)
Duran-Jimenez
B
Dobler
D
Moffatt
S
Rabbani
N
Streuli
CH
Thornalley
PJ
, et al.  . 
Advanced glycation end products in extracellular matrix proteins contribute to the failure of sensory nerve regeneration in diabetes
Diabetes
 , 
2009
, vol. 
58
 (pg. 
2893
-
903
)
Dyck
PJ
Karnes
JL
O'Brien
P
Okazaki
H
Lais
A
Engelstad
J
The spatial distribution of fiber loss in diabetic polyneuropathy suggests ischemia
Ann Neurol
 , 
1986
, vol. 
19
 (pg. 
440
-
9
)
Dyck
PJ
Kratz
KM
Karnes
JL
Litchy
WJ
Klein
R
Pach
JM
, et al.  . 
The prevalence by staged severity of various types of diabetic neuropathy, retinopathy, and nephropathy in a population-based cohort: the Rochester Diabetic Neuropathy Study
Neurology
 , 
1993
, vol. 
43
 (pg. 
817
-
24
)
Ebenezer
GJ
Laast
VA
Dearman
B
Hauer
P
Tarwater
PM
Adams
RJ
, et al.  . 
Altered cutaneous nerve regeneration in a simian immunodeficiency virus / macaque intracutaneous axotomy model
J Comp Neurol
 , 
2009
, vol. 
514
 (pg. 
272
-
83
)
Eichmann
A
Le Noble
F
Autiero
M
Carmeliet
P
Guidance of vascular and neural network formation
Curr Opin Neurobiol
 , 
2005
, vol. 
15
 (pg. 
108
-
15
)
Ekstrom
AR
Tomlinson
DR
Impaired nerve regeneration in streptozotocin-diabetic rats. Effects of treatment with an aldose reductase inhibitor
J Neurol Sci
 , 
1989
, vol. 
93
 (pg. 
231
-
7
)
Gibran
NS
Jang
YC
Isik
FF
Greenhalgh
DG
Muffley
LA
Underwood
RA
, et al.  . 
Diminished neuropeptide levels contribute to the impaired cutaneous healing response associated with diabetes mellitus
J Surg Res
 , 
2002
, vol. 
108
 (pg. 
122
-
8
)
Gora-Kupilas
K
Josko
J
The neuroprotective function of vascular endothelial growth factor (VEGF)
Folia Neuropathol
 , 
2005
, vol. 
43
 (pg. 
31
-
9
)
Griffin
JW
Nguyen
HA
TT
Michael
ES
Stephanie
C
Leonardo
C
Pamela
D
Fred
G
Neural repair and plasticity
Textbook of neural repair and rehabilitation
 , 
2006
Cambridge University Press
(pg. 
293
-
303
)
Griffin
JW
Pan
B
Polley
MA
Hoffman
PN
Farah
MH
Measuring nerve regeneration in the mouse
Exp Neurol
 , 
2010
, vol. 
223
 (pg. 
60
-
71
)
Gu
C
Rodriguez
ER
Reimert
DV
Shu
T
Fritzsch
B
Richards
LJ
, et al.  . 
Neuropilin-1 conveys semaphorin and VEGF signaling during neural and cardiovascular development
Dev Cell
 , 
2003
, vol. 
5
 (pg. 
45
-
57
)
Hahn
K
Sirdofsky
M
Brown
A
Ebenezer
G
Hauer
P
Miller
C
, et al.  . 
Collateral sprouting of human epidermal nerve fibers following intracutaneous axotomy
J Peripher Nerv Syst
 , 
2006
, vol. 
11
 (pg. 
142
-
7
)
Hobson
MI
Green
CJ
Terenghi
G
VEGF enhances intraneural angiogenesis and improves nerve regeneration after axotomy
J Anat
 , 
2000
, vol. 
197
 
Pt 4
(pg. 
591
-
605
)
Hoke
A
Gordon
T
Zochodne
DW
Sulaiman
OA
A decline in glial cell-line-derived neurotrophic factor expression is associated with impaired regeneration after long-term Schwann cell denervation
Exp Neurol
 , 
2002
, vol. 
173
 (pg. 
77
-
85
)
Jaap
AJ
Shore
AC
Stockman
AJ
Tooke
JE
Skin capillary density in subjects with impaired glucose tolerance and patients with type 2 diabetes
Diabet Med
 , 
1996
, vol. 
13
 (pg. 
160
-
4
)
Jakobsen
J
Brimijoin
S
Skau
K
Sidenius
P
Wells
D
Retrograde axonal transport of transmitter enzymes, fucose-labeled protein, and nerve growth factor in streptozotocin-diabetic rats
Diabetes
 , 
1981
, vol. 
30
 (pg. 
797
-
803
)
Jakobsen
J
Sidenius
P
Decreased axonal flux of retrogradely transported glycoproteins in early experimental diabetes
J Neurochem
 , 
1979
, vol. 
33
 (pg. 
1055
-
60
)
Jin
K
Zhu
Y
Sun
Y
Mao
XO
Xie
L
Greenberg
DA
Vascular endothelial growth factor (VEGF) stimulates neurogenesis in vitro and in vivo
Proc Natl Acad Sci USA
 , 
2002
, vol. 
99
 (pg. 
11946
-
50
)
Katz
MA
McCuskey
P
Beggs
JL
Johnson
PC
Gaines
JA
Relationships between microvascular function and capillary structure in diabetic and nondiabetic human skin
Diabetes
 , 
1989
, vol. 
38
 (pg. 
1245
-
50
)
Kennedy
JM
Zochodne
DW
Impaired peripheral nerve regeneration in diabetes mellitus
J Peripher Nerv Syst
 , 
2005
, vol. 
10
 (pg. 
144
-
57
)
Kennedy
JM
Zochodne
DW
The regenerative deficit of peripheral nerves in experimental diabetes: its extent, timing and possible mechanisms
Brain
 , 
2000
, vol. 
123
 
Pt 10
(pg. 
2118
-
29
)
Lehtinen
JM
Uusitupa
M
Siitonen
O
Pyorala
K
Prevalence of neuropathy in newly diagnosed NIDDM and nondiabetic control subjects
Diabetes
 , 
1989
, vol. 
38
 (pg. 
1307
-
13
)
Malik
RA
Veves
A
Walker
D
Siddique
I
Lye
RH
Schady
W
, et al.  . 
Sural nerve fibre pathology in diabetic patients with mild neuropathy: relationship to pain, quantitative sensory testing and peripheral nerve electrophysiology
Acta Neuropathol
 , 
2001
, vol. 
101
 (pg. 
367
-
74
)
McArthur
JC
Stocks
EA
Hauer
P
Cornblath
DR
Griffin
JW
Epidermal nerve fiber density: normative reference range and diagnostic efficiency
Arch Neurol
 , 
1998
, vol. 
55
 (pg. 
1513
-
20
)
Peterson
DA
Jones
DG
Determination of neuronal number and process surface area in organotypic cultures: a stereological approach
J Neurosci Methods
 , 
1993
, vol. 
46
 (pg. 
107
-
20
)
Polydefkis
M
Hauer
P
Griffin
JW
McArthur
JC
Skin biopsy as a tool to assess distal small fiber innervation in diabetic neuropathy
Diabetes Technol Ther
 , 
2001
, vol. 
3
 (pg. 
23
-
8
)
Polydefkis
M
Hauer
P
Sheth
S
Sirdofsky
M
Griffin
JW
McArthur
JC
The time course of epidermal nerve fibre regeneration: studies in normal controls and in people with diabetes, with and without neuropathy
Brain
 , 
2004
, vol. 
127
 (pg. 
1606
-
15
)
Quattrini
C
Jeziorska
M
Boulton
AJ
Malik
RA
Reduced vascular endothelial growth factor expression and intra-epidermal nerve fiber loss in human diabetic neuropathy
Diabetes Care
 , 
2008
, vol. 
31
 (pg. 
140
-
5
)
Rajan
B
Polydefkis
M
Hauer
P
Griffin
JW
McArthur
JC
Epidermal reinnervation after intracutaneous axotomy in man
J Comp Neurol
 , 
2003
, vol. 
457
 (pg. 
24
-
36
)
Rivard
A
Silver
M
Chen
D
Kearney
M
Magner
M
Annex
B
, et al.  . 
Rescue of diabetes-related impairment of angiogenesis by intramuscular gene therapy with adeno-VEGF
Am J Pathol
 , 
1999
, vol. 
154
 (pg. 
355
-
63
)
Ropper
AH
Gorson
KC
Gooch
CL
Weinberg
DH
Pieczek
A
Ware
JH
, et al.  . 
Vascular endothelial growth factor gene transfer for diabetic polyneuropathy: a randomized, double-blinded trial
Ann Neurol
 , 
2009
, vol. 
65
 (pg. 
386
-
93
)
Schratzberger
P
Walter
DH
Rittig
K
Bahlmann
FH
Pola
R
Curry
C
, et al.  . 
Reversal of experimental diabetic neuropathy by VEGF gene transfer
J Clin Invest
 , 
2001
, vol. 
107
 (pg. 
1083
-
92
)
Schwarz
Q
Gu
C
Fujisawa
H
Sabelko
K
Gertsenstein
M
Nagy
A
, et al.  . 
Vascular endothelial growth factor controls neuronal migration and cooperates with Sema3A to pattern distinct compartments of the facial nerve
Genes Dev
 , 
2004
, vol. 
18
 (pg. 
2822
-
34
)
Sigaudo-Roussel
D
Demiot
C
Fromy
B
Koitka
A
Leftheriotis
G
Abraham
P
, et al.  . 
Early endothelial dysfunction severely impairs skin blood flow response to local pressure application in streptozotocin-induced diabetic mice
Diabetes
 , 
2004
, vol. 
53
 (pg. 
1564
-
9
)
Singleton
JR
Smith
AG
Russell
J
Feldman
EL
Polyneuropathy with impaired glucose tolerance: implications for diagnosis and therapy
Curr Treat Options Neurol
 , 
2005
, vol. 
7
 (pg. 
33
-
42
)
Sondell
M
Lundborg
G
Kanje
M
Vascular endothelial growth factor stimulates Schwann cell invasion and neovascularization of acellular nerve grafts
Brain Res
 , 
1999
, vol. 
846
 (pg. 
219
-
28
)
Sondell
M
Sundler
F
Kanje
M
Vascular endothelial growth factor is a neurotrophic factor which stimulates axonal outgrowth through the flk-1 receptor
Eur J Neurosci
 , 
2000
, vol. 
12
 (pg. 
4243
-
54
)
Song
XY
Zhou
FH
Zhong
JH
Wu
LL
Zhou
XF
Knockout of p75(NTR) impairs re-myelination of injured sciatic nerve in mice
J Neurochem
 , 
2006
, vol. 
96
 (pg. 
833
-
42
)
Taniguchi
M
Matsuzaki
S
Tohyama
M
P75 plays a key role in the induction of the sprouting of sensory nerve fibers in inflamed skin
J Invest Dermatol
 , 
2007
, vol. 
127
 (pg. 
2062
-
5
)
Tepper
OM
Galiano
RD
Capla
JM
Kalka
C
Gagne
PJ
Jacobowitz
GR
, et al.  . 
Human endothelial progenitor cells from type II diabetics exhibit impaired proliferation, adhesion, and incorporation into vascular structures
Circulation
 , 
2002
, vol. 
106
 (pg. 
2781
-
6
)
Terenghi
G
Peripheral nerve regeneration and neurotrophic factors
J Anat
 , 
1999
, vol. 
194
 
Pt 1
(pg. 
1
-
14
)
Tesfaye
S
Harris
N
Jakubowski
JJ
Mody
C
Wilson
RM
Rennie
IG
, et al.  . 
Impaired blood flow and arterio-venous shunting in human diabetic neuropathy: a novel technique of nerve photography and fluorescein angiography
Diabetologia
 , 
1993
, vol. 
36
 (pg. 
1266
-
74
)
Tesfaye
S
Selvarajah
D
Morphine, gabapentin, or their combination for neuropathic pain
N Engl J Med
 , 
2005
, vol. 
352
 (pg. 
2650
-
1
)
Thornalley
PJ
Glycation in diabetic neuropathy: characteristics, consequences, causes, and therapeutic options
Int Rev Neurobiol
 , 
2002
, vol. 
50
 (pg. 
37
-
57
)
Tomlinson
DR
Fernyhough
P
Diemel
LT
Neurotrophins and peripheral neuropathy
Philos Trans R Soc Lond B Biol Sci
 , 
1996
, vol. 
351
 (pg. 
455
-
62
)
Tomlinson
DR
Fernyhough
P
Diemel
LT
Role of neurotrophins in diabetic neuropathy and treatment with nerve growth factors
Diabetes
 , 
1997
, vol. 
46
 
Suppl 2
(pg. 
S43
-
9
)
Tosaki
T
Kamiya
H
Yasuda
Y
Naruse
K
Kato
K
Kozakae
M
, et al.  . 
Reduced NGF secretion by Schwann cells under the high glucose condition decreases neurite outgrowth of DRG neurons
Exp Neurol
 , 
2008
, vol. 
213
 (pg. 
381
-
7
)
West
MJ
Slomianka
L
Gundersen
HJ
Unbiased stereological estimation of the total number of neurons in the subdivisions of the rat hippocampus using the optical fractionator
Anat Rec
 , 
1991
, vol. 
231
 (pg. 
482
-
97
)
Yarom
R
Zirkin
H
Stammler
G
Rose
AG
Human coronary microvessels in diabetes and ischaemia
Morphometric study of autopsy material. J Pathol
 , 
1992
, vol. 
166
 (pg. 
265
-
70
)
Yasuda
H
Dyck
PJ
Abnormalities of endoneurial microvessels and sural nerve pathology in diabetic neuropathy
Neurology
 , 
1987
, vol. 
37
 (pg. 
20
-
8
)
Yasuda
H
Terada
M
Maeda
K
Kogawa
S
Sanada
M
Haneda
M
, et al.  . 
Diabetic neuropathy and nerve regeneration
Prog Neurobiol
 , 
2003
, vol. 
69
 (pg. 
229
-
85
)
Yu
CQ
Zhang
M
Matis
KI
Kim
C
Rosenblatt
MI
Vascular endothelial growth factor mediates corneal nerve repair
Invest Ophthalmol Vis Sci
 , 
2008
, vol. 
49
 (pg. 
3870
-
8
)
Zhu
H
Li
F
Yu
WJ
Wang
WJ
Li
L
Wan
LD
, et al.  . 
Effect of hypoxia/reoxygenation on cell viability and expression and secretion of neurotrophic factors (NTFs) in primary cultured schwann cells
Anat Rec (Hoboken)
 , 
2010
, vol. 
293
 (pg. 
865
-
70
)
Ziegler
D
Gries
FA
Spuler
M
Lessmann
F
The epidemiology of diabetic neuropathy. Diabetic Cardiovascular Autonomic Neuropathy Multicenter Study Group
J Diabetes Complications
 , 
1992
, vol. 
6
 (pg. 
49
-
57
)