Fig. 1. Evaluation of OGD-induced cell death. (A) Vital staining: The influence of OGD on the viability of Schwann cells was assessed by double-labeling fluorescence technique, whereby living cells, hydrolyzing fluorescein diacetate by intracellular esterases, generate a green-yellow fluorescence. Dead cells are labeled by propidium iodide, which interacts with DNA resulting in a red fluorescence of cell nuclei. It reflected that the cell death rate of Schwann cells was increased by OGD. In control cultures (1) and high dosed (50 μg/ml) minocycline-treated normoxid cultures (4), red nuclei were seen only sporadically. After 6 h of OGD (2), the number of red cell nuclei increased significantly, whereas minocycline treatment, even at a low dose (5 μg/ml), distinctly reduced the OGD-induced cell death rate (3). (B) Quantitative evaluation: The percentage of dead cells (propidium iodide positive, PI+) of the total number of cell profiles identified in the respective culture is given as mean ± SEM from n = 16 per treatment. p < 0.05, p < 0.001 in comparison with controls. (C) TUNEL staining: The number of apoptotic cells, demonstrated by the TUNEL technique of DNA fragments (dark brown nuclei), increased after OGD (1 versus 2). Again, this OGD-induced apoptosis was suppressed by the minocycline treatment (2 versus 3). Bars (A) = 200 μm, (B) = 50 μm. (D) Quantitative evaluation: The percentage of TUNEL positive cells of the total number of cell profiles identified in the respective culture is given as mean ± SEM from n = 16 per treatment. p < 0.05, p < 0.01 in comparison with controls.

Fig. 2. RT-PCR analysis of Schwann cell cultures. After 10 days in vitro, SCs were exposed to OGD (5% CO₂/95% N₂) for 6 h, followed by 1 h reoxygenation at 37 °C. RT-PCR demonstrates that OGD is able to induce the mRNA expression of BAX, caspase-3 and HIF-1α. Already at the lower concentration (5 μg/ml) minocycline (applied 30 min prior OGD) counteracted this effect. The expression of BCL-2, however, is reduced by OGD and normalized by minocycline. In PBS-treated control cultures, minocycline affected only the expression of BCL-2, which was slightly induced. Bands were quantified by densitometric analysis after normalizing the values with GAPDH. The differences to the respective control values (p < 0.05, p < 0.005) and between OGD and OGD+mino (^#p < 0.05) are indicated. For product sizes see Table 1.

Fig. 3. Western blot analysis of Schwann cell cultures. Western blot analysis of cultivated Schwann cells stressed by OGD (5% CO₂/95% N₂ for 6 h, followed by 1 h reoxigenation at 37 °C) indicates an induction of HIF-1α, BAX, and procaspase-3 (32 kDa-band) and a reduction of BCL-2. Moreover, it demonstrates the cleavage of procaspase-3 into the active 17 kDa-band and the subsequent cleavage of actin into fractin. All these effects are counteracted by minocycline (here demonstrated for 50 μg/ml, applied 30 min prior OGD), the BCL-2 expression even reaching levels above baseline. Cultures were used after 10 days in vitro. Immunoreactivity was quantified by densitometric analysis after normalizing the values with β-actin in the same blot. The differences to the respective control values (p < 0.05, p < 0.005) and between OGD and OGD+mino (^#p < 0.05) are indicated.

Fig. 4. Immunofluorescence staining of Schwann cell cultures. At cellular levels, immunocytochemistry showed the induction of BAX (A versus C) and the reduction of BCL-2 (E versus G), as well as the respective minocycline effects after OGD (5% CO₂/95% N₂ for 6 h, followed by 1 h reoxigenation at 37 °C). It demonstrates an activation and translocation into nuclei of HIF-1α (I versus K), AIF (M versus O) and caspase-3 (Q versus S); and the release of cytochrome c from mitochondria into the cytoplasm (U versus W), all also reversed by minocycline. The staining validates the generation of fractin (Y versus ZZ, co-staining with S100) again counteracted by minocycline. To support the efficiency of the treatment, minocycline effects against OGD are demonstrated for the lower dose of 5 μg/ml (D,H,L,P,T,X,ZZZ), whereas the non-toxicity of minocycline is demonstrated using the higher dose of 50 μg/ml (B,F,J,N,R,V,Z). Cultures were used after 10 days in vitro. Bar = 50 μm.

Fig. 5. Revascularization. In all PBS-treated grafts, the vascular architecture (demonstrated by i.v. injected Evans blue bound to albumin, red fluorescence) is well reestablished by the fifth post-operative day, whereas in the respective minocycline-treated grafts the reduced number of endoneurial vessels indicates a hampering effect of minocycline. Bar = 400 μm.

Fig. 6. Light microscopy of nerve/graft material. Light microscopical evaluation of toluidine blue stained semi-thin cross-sections reveals that the uninjured PBS-treated peripheral nerve (A) is not affected by minocycline administered once daily for five consecutive days (A versus B). Before grafting, the predegenerated nerve (C) is free of myelin structures. Five days after implantation, the nerve graft (D) of PBS-treated rats is characterized by irregular myelin sheaths and cell debris whereas the predegenerated nerve graft (E) and the muscle/Schwann cell graft (F) first showed newly-formed thin myelin sheaths (arrows). In the minocycline-treated nerve graft (G), an abnormal persistence of residual, virtually intact, myelinated fiber profiles is evident, whereas in the artificial grafts minocycline treatment is without conspicuous effects (H,I). After 6 weeks, all PBS-treated groups show a clear regeneration, indicated by a high number of substantially re-myelinated fibers (J,K,L), whereas in minocycline-treated animals (M,N,O), an increased number of small axons with thinner myelin sheaths can be seen. Regenerated fibers are seen also in all distal nerve segments of PBS-treated animals (P,Q,R) (Figure continued on next page). Minocycline treatment reduces the number of regenerated fibers in the nerve graft (P versus S) and in the predegenerated nerve graft (Q versus T) group. In the muscle/Schwann cell group, however, the outcome of re-myelinated fibers is enhanced (R versus U). Predeg, predegenerated; ST, survival time; bar = 50 μm.

Fig. 7. Morphometric analysis. (A) After 6 weeks, nerve counts in all PBS-treated grafts did not differ significantly from those of the respective proximal stumps. In minocycline-treated animals, axon counts of the grafts are significantly enhanced compared with the proximal nerve stump, indicating an unusual persistence of collateral sprouts. Low axon counts of the distal nerve illustrate the poor regenerative outcome in the minocycline-treated nerve graft group. In the predegenerated nerve graft group, this phenomenon was less pronounced, and in the muscle/Schwann cell graft group it was significantly reversed. (B) The g-ratio (axon diameter/total fiber diameter; the normal value is between 0.6 and 0.7) shows only non-significant differences between PBS- and minocycline-treated animals. (C) Muscle to weight ratio of the gastrocnemius muscle indicates reduced reinnervation in the minocycline-treated nerve and predegenerated nerve graft groups compared with the respective PBS-treated animals. The muscle/Schwann cell group, however, shows a tendentially enhanced ratio after minocycline treatment. Data are mean values ± S.E.M. Asterisks indicate significant differences (p < 0.05, p < 0.005, p < 0.0005).

Fig. 8. RT-PCR analysis of nerve/graft material. After a regeneration time of 6 weeks, RT-PCR demonstrates an injury-induced up-regulation of MMP9 and VEGF mRNA expression in all grafts which is inhibited by minocycline treatment. In the contralateral (control) nerve minocycline is ineffective. HIF-1α mRNA expression, also massively induced by nerve grafting, however, is unaffected by minocycline treatment. Bands were quantified by densitometric analysis after normalizing the values with GAPDH. For product sizes see Table 1. (−) PBS-treated, (+) minocycline-treated. The differences to the respective control values (p < 0.05, p < 0.005, p < 0.0005, horizontal comparison) and between the respective PBS (upper line-/minocycline (lower line)-treated values (^#p < 0.05, vertical comparison) are indicated.

Fig. 9. Western blot analysis of nerve/graft material. After a regeneration time of 6 weeks, Western blot analysis indicates that nerve grafting induces expression of MMP9, HIF-1α and VEGF protein in the grafts. The induction of MMP9 and VEGF is suppressed by minocycline, whereas the enhanced expression of HIF-1α seems to be uninfluenced. Immunoreactivity was quantified by densitometric analysis after normalizing the values with β-actin in the same blot. (−) PBS-treated, (+) minocycline-treated. The differences to the respective control values (p < 0.05, p < 0.005, p < 0.0005, horizontal comparison) and between the respective PBS (upper line)-/minocycline (lower line)-treated values (^#p < 0.05, vertical comparison) are indicated.

Sequences of primers used for RT-PCR

Timetable of the return of thermal sensitivity and active spreading of digits

(−) non reaction, (+) positive reaction; SC, Schwann cell, mino, minocycline.