Several studies have shown that fibroblast growth factor-2 (FGF2) can directly affect axon regeneration after peripheral nerve damage. In this study, we performed sensory tests and histological analyses to study the effect of recombinant human FGF-2 (rhFGF2) treatment on damaged mental nerves. The mental nerves of 6-week-old male Sprague-Dawley rats were crush-injured for 1 minute and then treated with 10 or 50 μg/mL rhFGF2 or PBS in crush injury area with a mini Osmotic pump. Sensory test using von Frey filaments at 1 week revealed the presence of sensory degeneration based on decreased gap score and increased difference score. However, at 2 weeks, the gap score and difference score were significantly rebounded in the mental nerve crush group treated with 10 μg/mL rhFGF2. Interestingly, treatment with 10 μg/mL rhFGF had a more obviously positive effect on the gap score than treatment with 50 μg/mL rhFGF2. In addition, retrograde neuronal tracing with Dil revealed a significant increase in nerve regeneration in the trigeminal ganglion at 2 and 4 weeks in the rhFGF2 groups (10 μg/mL and 50 μg/mL) than in the PBS group. The 10 μg/mL rhFGF2 group also showed an obviously robust regeneration in axon density in the mental nerve at 4 weeks. Our results demonstrate that 10 μg/mL rhFGF induces mental nerve regeneration and sensory recovery after mental nerve crush injury.

BackgroundComparative proteomics in bacteria are often hampered by the differential nature of dataset quality and/or inherent biological deviations. Although common practice compensates by reproducing and normalizing datasets from a single sample, the degree of certainty is limited in comparison of multiple dataset. To surmount these limitations, we introduce a two-step assessment criterion using: (1) the relative number of total spectra (R TS) to determine if two LC-MS/MS datasets are comparable and (2) nine glycolytic enzymes as internal standards for a more accurate calculation of relative amount of proteins. Lactococcus lactis HR279 and JHK24 strains expressing high or low levels (respectively) of green fluorescent protein (GFP) were used for the model system. GFP abundance was determined by spectral counting and direct fluorescence measurements. Statistical analysis determined relative GFP quantity obtained from our approach matched values obtained from fluorescence measurements.Results L. lactis HR279 and JHK24 demonstrates two datasets with an R TS value less than 1.4 accurately reflects relative differences in GFP levels between high and low expression strains. Without prior consideration of R TS and the use of internal standards, the relative increase in GFP calculated by spectral counting method was 3.92 ± 1.14 fold, which is not correlated with the value determined by the direct fluorescence measurement (2.86 ± 0.42 fold) with the p = 0.024. In contrast, 2.88 ± 0.92 fold was obtained by our approach showing a statistically insignificant difference (p = 0.95).ConclusionsOur two-step assessment demonstrates a useful approach to: (1) validate the comparability of two mass spectrometric datasets and (2) accurately calculate the relative amount of proteins between proteomic datasets.Electronic supplementary materialThe online version of this article (doi:10.1186/s12859-015-0561-9) contains supplementary material, which is available to authorized users.