Late onset muscle plasticity in the whisker pad of enucleated rats

Abstract

Blindness leads to a major reorganization of neural pathways associated with touch. Because incoming somatosensory information influences motor output, it is plausible that motor plasticity occurs in the blind. In this work, we evaluated this issue at the peripheral level in enucleated rats. Whisker muscles in enucleated rats 160 days of age or older showed increased cytochrome oxidase activity, capillary density, motor plate size, and amplitude of evoked field potentials as compared with their control counterparts. Such differences were not observed at ages 10 and 60 days, the capillary density was the exception being greater in the enucleated rat at the latter age. Interestingly, there was a trend to increased neurotrophin-3 concentrations in the whisker pads of enucleated rats throughout postnatal development. Our results show that neonatal enucleation leads to late onset plasticity of the whisker's motor system.

In mammals, the loss of vision leads to a large scale sensory reorganization of the brain. Somatosensory and auditory areas enlarge whereas the former visual regions become activated by somatosensory and auditory information (1). These functional modifications are associated with the reorganization of their anatomical substrate (2, 3). The reorganized brain allows the blind to develop the somatosensory and auditory abilities that distinguish them from sighted subjects (4, 5).

Somatosensory information calibrates motor output (5, 6–8). Hence, one might expect certain degree of motor plasticity in blind subjects. Accordingly, studies have shown incremental increases in the activity of glutamate dehydrogenase and in the size of neurons located in the motor cortex after visual deprivation (9). Magnetic resonance studies in blind humans have also documented a decreased functional connectivity to motor areas (9, 10), increments in the volume of cortical gray matter associated with sensory-motor cortices (11), and in the volume of the cortico-spinal tract (9). Whether motor plasticity in the blind extends to more peripheral levels is unknown.

A good model system to test whether peripheral motor plasticity occurs in the blind is the sensory-motor system controlling whisking in rodents. Sensory information arising from the whiskers calibrates their movements (12) while motor activity modulates sensory information influx (13, 14). Our work was aimed at evaluating the morpho-functional plasticity ongoing in the intrinsic motor apparatus of the facial whiskers of neonatally enucleated rats. The time course and the possible relationship between the plastic response and neurotrophin-3 concentrations were analyzed. Our results document late onset morpho-functional plasticity in the intrinsic musculature of whisker follicles in rats enucleated at birth. The evidence we provide suggests that this plastic response is associated with increments in the concentrations of neurotrophin-3 in the whisker pad. Hence, blind individuals display a plastic response that goes beyond the limits of the sensory sphere and reaches the motor area.

Results

Cytochrome Oxidase Activity.

Densitometric analyses of the histochemical reaction for cytocrome oxidase were carried out to evaluate oxidative metabolism in the intrinsic muscles of the whisker pad (15). Cytochrome oxidase activity increased from postnatal day (PD) 10 to PD60 in both groups (Fig. 1). After PD60, cytochrome oxidase activity increased significantly more in enucleated rats (21%; Fig. 1G). Such difference appeared to be due to increments in the oxidative metabolism of a subset of the intrinsic muscle fibers (Fig. 1F).

Fig. 1.

The relative activity of cytochrome oxidase increases more in intrinsic muscles of whisker pads of enucleated rats at PD160 than in those of the control animals. The figure illustrates the pattern of cytochrome oxidase histochemical staining in tangential sections of the whisker pad of control (A, C, and E) and enucleated (B, D, and F) rats at PD10 (A and B), PD60 (C and D) and PD160 (E and F). Scale = 60 μm. The bar graph in (G) depicts the percent difference in intensity of staining of the intrinsic muscles relative to the interfollicular dermis (Student's t test: *, P < 0.001).

Blood Vessel Area.

The area occupied by blood vessels correlates with ongoing oxidative metabolic demands in muscle fibers (16, 17). Although this value was similar in both groups at PD10, enucleated rats showed a higher percentage of area containing blood vessels than in their control counterparts at PD 60 (26%) and PD160 (36%). Overall, the area occupied by blood vessels in muscle fascicles increased significantly from PD10 to PD60 but tended to decrease from this age forward in both animal groups (Fig. 2G).

Fig. 2.

The relative area occupied by blood vessels nourishing intrinsic muscle fascicles of the whisker pad was greater in enucleated rats than in control ones at PD60 and PD160. The figure illustrates the pattern of alkaline phosphatase stained blood vessels in tangential sections of control (A, C, and E) and enucleated (B, D, and F) rats at postnatal PD10 (A and B), PD60 (C, D) and PD160 (E and F). Scale = 60 μm. The bar graph in (G) shows the percent difference in the area occupied by blood vessels at the different ages (Student's t test: *, P < 0.001).

Myosin Isoform Immunocytochemistry.

The metabolic phenotype and physiological properties of the sarcomeres in the muscle fibers are tightly linked with the expression of different myosin isoforms (18–20). Broadly, the sarcomeres of oxidative slow twitch type I muscle fibers contain myosin heavy chain isoform Iβ, whereas those of the glycolytic fast twitch type II muscle fibers are composed of myosin heavy chain isoforms IIa, d, and/or b (20). The latter are ordered from the slower to the faster fiber contraction velocity (18). Intrinsic whisker muscles in rats are formed predominantly by type IIb/d fibers. There is also a reduced percentage of type I fibers (21). To investigate whether the metabolic shifts observed in intrinsic whisker muscles of enucleated rats are associated with muscle phenotypic transitions, we conducted immunocytochemical analyses aimed at evaluating the percentage of the area occupied by muscle fibers displaying the myosin heavy chain isoform II or Iβ in control and enucleated rats at different ages (Fig. 3). The analyses revealed that this parameter was similar in control and enucleated rats irrespective of age (Fig. 3). At PD10, 92% percent of the sampled area was occupied by fast fibers, whereas only 8% of the sampled area was taken up by slow fibers. At PD60, the values were 94% for fast fibers and 6% for slow fibers. At PD160, the values reached 99% for fast fibers and 1% for slow fibers (Fig. 3D). These last observations agree with previous findings (21).

Fig. 3.

The relative area occupied by intrinsic slow twitch muscle fibers in the whisker pad does not differ between control and enucleated rats. The figure illustrates tangential sections in which slow (dark) and fast (light brown) twitch fibers were immunocytochemically stained at PD10 (A), PD60 (B) and PD160 (C) (Scale bar, 500 μm; inset scale bar, 40 μm). The line graph in the lower panel (D) depicts the area occupied by the slow relative to the fast twitch muscle fibers at different ages.

Motor Plate Size.

Metabolic properties, myosin isoform subtypes and plate size in striate muscles are all correlated positively (22–24). In general, oxidative slow twitch type I muscle fibers have larger motor plates than those observed in association with glycolytic fast twitch type II muscle fibers (25–27); although see (28–30). We evaluated the area of the presynaptic and postsynaptic elements of the motor plate in control and enucleated rats at different ages (Fig. 4). At PD160 these areas increased by 38% and 28% respectively (Fig. 4 M and N). No differences between groups were found at PD10 and PD60. Our data also showed that under normal conditions, presynaptic terminals grew continuously from PD 10 to PD160, while postsynaptic sites have stopped growing by PD60. After enucleation, however, postsynaptic elements continue to grow, so both synaptic elements participate in the reorganization of the motor plate.

Fig. 4.

The area of pre- and postsynaptic elements of the motor plates increased in enucleated rats at PD160. The figure illustrate tangential sections in which acetyl cholinesterase stained terminals (A–F) (scale bars: A and B, 10 μm; C–F, 40 μm) and rhodamine-conjugated α- bungarotoxin labeled postsynaptic component (G–L) (scale bar, 10 μm) are seen in control and enucleated rats at PD10 (A, B, G, and H), PD60 (C, D, I, J) and PD160 (E, F, K, L). The line graphs in the lower panels depict the average area of the terminals (M) and of postsynaptic elements (N) (Student's t test: *, P < 0.001; †, P < 0.01).

Electrophysiological Recordings.

There is a positive correlation between motor plate size and the amount of neurotransmitter released (31). We measured the extracellular evoked field potentials (EFPs) in the intrinsic musculature of the facial whiskers. The graph in Fig. 5 shows a pooled graph of EFPs amplitude of the whisker pad muscles versus stimulus intensity (from 1 to 3 times threshold, xT) following buccolabialis nerve stimulation. The amplitude of the EFPs elicited by increasing stimulus intensity (from 1.5 to 2.5 xT) was generally higher in enucleated rats than in control animals. At the highest stimulus intensity (>2.5 xT), the amplitude was similar in control and enucleated rats.

Fig. 5.

The amplitude of extracellular field potentials recorded in the intrinsic musculature of the whisker pad elicited by increasing stimulation of the buccolabialis motor branch of the facial nerve was in general larger in enucleated rats at PD200 (times threshold, xT) (Student's t test: *, P < 0.01).

Concentrations of Neurotrophin-3.

Neurotrophin-3 is synthesized, among other peripheral sources, by skeletal muscles (32) and influences the maintenance of presynaptic motor nerve terminals and acetylcholine packaging and release (33). Thus, NT-3 content was estimated in the whisker pad of control and enucleated rats. The amount of NT-3 of enucleated rats was reduced by 12% at PD2. However, by PD7 the concentrations of NT-3 began to increase in enucleated rats. Significant differences were found at PD30 (119%) and PD200 (144%; Fig. 6).

Fig. 6.

The content of neurotrophin-3 in the whisker pads tended to be higher in enucleated rats throughout postnatal life. (A) Western blot that illustrates the pattern of immunoreactive bands for neurotrophin-3 in control (c) and enucleated (e) rats at different ages. (B) Quantification of NT-3 content expressed as the fraction of the density relative to the optical density of PD2 controls (Student's t test: *, P < 0.02; †, P < 0.007; ‡, P < 0.03, respectively).

Discussion

Most research carried out in blind subjects focus at understanding sensory compensation. Only recently have researchers begun to look for signs of motor plasticity in the blind (9–11, 34). We now can report that intrinsic muscle fibers of the rat's whisker pad undergo late onset plasticity following neonatal enucleation. These plastic changes include (a) an increment of oxidative metabolism, (b) the size of pre- and post synaptic elements and (c) the amplitude of EFPs in intrinsic muscles of the whisker pad between PD160 and PD200. Hence, our observations support the conjecture that blindness may induce plasticity at peripheral levels of the motor system. Whether central and peripheral motor changes concur in time is still an open question. It might be that peripheral motor plasticity is the final stage of a process of top-down plasticity that begins at cortical levels at the early stages of development following enucleation. This circumstance could explain why we observed muscle plasticity after PD60.

Under normal functional loading, intrinsic muscles of the rat whisker pad are formed predominantly by fast twitch glycolytic fibers (21). Muscle fascicles formed by these fibers have poorly developed capillary beds. The intrinsic muscle fiber fascicles in enucleated rats (PD160) showed both greater the activity of cytochrome oxidase and increments in the area occupied by capillaries as compared with age-matched littermate control rats. Although immunocytochemical studies did not document a difference in the area occupied by slow twitch/fast twitch muscle fibers in the whisker pads of control and enucleated rats at different ages, we think that our observations indicate that intrinsic muscle fibers in enucleated rats tend to develop a fast oxidative phenotype. This idea is further supported by the facts that motor plate size and the amplitude of EFPs increased in enucleated rats at PD160 and PD200, respectively. It is known that motor terminals impinging upon slow twitch muscle fibers have larger motor plates than those contacting fast twitch muscles (25–27) and that larger motor endings display larger mean quantum content and increased neurotransmitter output (35–37). In addition, the fact that the mean amplitude of EFPs increased following greater stimulus intensity suggests that acetylcholine release and/or axon recruitment is facilitated after enucleation. Finally, our observation that the amplitude of the extracellular field potential at the highest stimulus intensity was similar in control and enucleated rats suggests that the number of motor fibers does not differ between both groups of animals.

Previous work in mouse bulbocavernosus and sternomastoid muscles suggest that the growth of motor pre- and postsynaptic elements is tightly coupled (38, 39; although see 40). An observation that might be considered somewhat contradictory is that in our study, motor terminals continued to grow well after the postsynaptic site had stopped growing in control rats. There is evidence, however, that postsynaptic sites in the mouse tensor fasciae latae muscle enlarge faster and far exceed motor terminal growth during postnatal development, thus suggesting that the differentiated postsynaptic site permits or promotes nerve terminal growth (41). We think that this might be happening in the motor plates of intrinsic muscles of the whisker follicles. Indeed, on one hand, the growth of motor terminals proceeds at slower pace than that of the postsynaptic sites. On the other hand, the postsynaptic sites roughly triple those of the motor terminals along postnatal development. This heterochronic growth may cause the postsynaptic sites to reach maturity and stop growing before motor terminals, obligating the latter to grow for a longer period. In addition, other sources of discrepancies among the studies may come from species differences (42) and differences among muscle types (43). Finally, it is worth emphasizing that in enucleated rats both synaptic elements participate in motor plate plasticity. Nevertheless, motor terminals grew faster than postsynaptic sites after PD60. This finding supports that motor terminals participate more actively in the plastic response in the enucleated rats, as suggested by Hill et al. (44)

Lastly, there are three issues that deserve consideration. The first two relate to the mechanisms that lead to muscle fiber plasticity. First, our results showed that overall concentrations of NT-3 increased in the whisker pad of enucleated rats from PD10 to PD200. One might think that increased nerve-muscle trophism could induce motor plate plasticity in the whisker pad of enucleated rats (30, 45) because NT-3 is the predominant neurotrophin expressed in skeletal muscles (32), and it enhances spontaneous and impulse evoked synaptic activity at the neuromuscular junction (33). However, the time frame in which the changes in NT-3 concentrations occurred did not match precisely that of the ongoing muscular plasticity. This indicates that NT-3, and hence the increment in nerve-muscle trophism, is not the primary factor inducing late onset motor plasticity in the enucleated rats. Another possibility is that late onset plasticity in the intrinsic musculature of the whisker follicles in enucleated rats could result from a long lasting increment in the use of these sensors (46). So, the daily mobility training might help refining motor plate morphology and function, and thus behavior, at later stages of postnatal development (5, 47). In addition, because plasticity along the trigeminal sensory pathway appears to occur at earlier stages of postnatal development after enucleation in rodents, and since sensory information calibrates motor output, it is possible that the plasticity ongoing along the somatosensory pathways is “transferred” to the motor pathways and that this process will take a bit longer than 100 days to fully occur in enucleated rats. The second issue we must consider relates to the functional consequences of the changes described in the motor system of the enucleated rat. We think that fast oxidative fibers that sustain contractions for longer times could slow down the speed of whisker scanning and adjust whisker stiffness thereby improving tactile abilities in enucleated rats. Clearly more research is needed to evaluate the merits of all of the possibilities discussed here. In the mean time, it would appear fair to suggest that motor plasticity of the sort reported here might underlie differences in the display of motor behavior observed between sighted and blind subjects while exploring objects and/or new environments (46–50). Also, we think that our results imply that successful rehabilitation in the blind must seek for measures not only oriented to re-establishing the visual channel, but to adjusting the body's motor output through mobility training (see also ref. 5 for a similar implication).

Materials and Methods

Animals.

The experiments were carried out in male Wistar rats raised in the animal facility of the Instituto de Investigaciones Biomédicas, Universidad Nacional Autónoma de México. The animals were kept in light (lights on at 6:00 a.m./lights off 6:00 p.m.) and temperature controlled rooms with free access to food and water until being killed. Litters were reduced to eight pups each; only four out of the eight pups were enucleated 6–8 h after birth as described elsewhere (51). Neonate and young rats were maintained with their mothers from birth until sacrifice at PD10. Young and mature adult rats were weaned at PD 21. After PD30, the animals were maintained until sacrifice in groups composed of two control and two enucleated rats. All of the experiments were carried out in littermates and analyzed using blind procedures. The experimental procedures were approved by the local animal rights committee and carefully followed the guidelines of the Código Bioético para la Investigación con Animales published by the Instituto de Investigaciones Biomédicas. All efforts were made to minimize the number of animals used and their potential suffering.

Cytochrome Oxidase Histochemistry.

The relative levels of the activity of cytochrome oxidase were estimated in the intrinsic muscles of facial whiskers in control and enucleated rats by using computer-assisted imaging densitometry as described in (52). Control and enucleated rats from different ages were euthanized (sodium pentobarbital; 45 mg/kg). Whisker pads were bilaterally dissected, flattened and frozen in 2-methyl butane prechilled with dry ice. Serial tangential sections (20 μm) were obtained in a cryostat, mounted, and processed through the histochemical staining for cytochrome oxidase (53). Images of stained sections were captured with a CoolPix 4.0 megapixel digital camera (Nikon) attached to a bright field Optiphot Nikon Microscope. Tangential sections of the whisker pad rendered images of the intrinsic muscles oriented longitudinally (PD10) or transversally (PD60 and PD160). The images were segmented digitally and the densitometric analyses carried out in individual muscles fibers using Image J (Scion Image). To correct for uneven illumination and for variations of the staining unrelated with the experimental conditions, light transmittance values in muscle fibers were normalized by reporting them as the percent difference relative to transmittance values obtained for the interfollicular dermis. A total of 4 fields per section and 10 sections per animal (n = 6 for PD10 and PD60; n = 4 for PD160) taken randomly were analyzed. Magnification (40× for PD60 and PD160 and 100× for PD10), illumination and contrast were all held fixed during image acquisition. Because variations in enzyme activity might result from developmental changes of its availability, we determined the saturation point of the histochemical reaction; this was reached after 2 h of incubation (37 °C) irrespective of animal age and manipulation. We then stopped the histochemical reaction after 30 min.

Area Occupy by Blood Vessels.

The area occupied by blood vessels in muscle fascicles of whiskers follicles was estimated in control and enucleated rats. Samples were obtained and processed as described previously. Once mounted on the slides, the sections were fixed with buffered paraformaldehyde (PF; 4%) for 2 h at room temperature, washed thoroughly with phosphate buffer (PB; 0.1M, pH 7.4), and incubated with a substrate kit (Vector) to reveal the activity of alkaline phosphatase for 25 min at room temperature. Images of the stained sections were captured, digitized, and segmented as described above. The percentage of the area occupied by blood vessels for each muscle fascicle was estimated. The fields photographed were always taken from regions chosen randomly but within intrinsic muscles. A total of 20 fields per animal were analyzed (n = 6 for PD10 and PD60; n = 4 for PD160). The magnification (40× for PD60 and PD160; and 100× for PD10), illumination, and contrast were all kept fixed during image acquisition.

Double Immunocytochemistry.

To label fast and slow myosin isoforms, control and enucleated rats were anesthetized and perfused with saline followed by PF. Their whisker pads were dissected, postfixed (8 h at 4 °C), cryoprotected, frozen flattened, and cut tangentially (20 μm) in a cryostat. Sections were mounted and incubated with hydrogen peroxide (1% in PB) during 45 min. The primary monoclonal antibody raised against rabbit fast myosin isoforms (1:400; Biomeda) diluted in blocking phosphate solution (3% normal horse serum, 0.1% triton X-100) was incubated over night at room temperature. Corresponding biotinylated secondary antibodies (1:800 in blocking solution; Chemicon) were incubated for 2 h at room temperature. The avidin-peroxidase complex (Vector) was incubated for an hour at room temperature and the activity of peroxidase was revealed with the DAB substrate kit omitting the use of nickel (Vector) for 10 min. Double immunostaining was achieved by incubating the sections with the avidin/biotin blocking kit (Vector) and then over night at room temperature with the primary monoclonal antibody raised against the slow myosin isoform (1:400, Chemicon). The activity of peroxidase was revealed by using nickel intensification. Double-stained slides were air dried and coverslipped with Cytoseal 60 (Richard-Allan Scientific). Images of the stained sections were captured, digitized, and segmented as described above. The percentage per frame of the area occupied by slow twitch fibers relative to fast twitch fibers was estimated. The fields photographed were always taken randomly from regions containing intrinsic muscle fibers. A total of 6 fields per section and 10 sections per animal (n = 6 for PD10 and PD60 and n = 4 for PD160) were analyzed.

Size of Motor Plates.

To document changes in the size of the motor plate of intrinsic muscle fibers of facial whiskers, the area of the presynaptic and postsynaptic elements was measured in control and enucleated rats. Animals were euthanized and their whisker pads dissected and processed as described earlier. Cryostat tangential section were obtained, mounted, and stained with the acetylcholinesterase histochemistry. Images of motor terminals were captured (PD10 = 60×; and PD60 and PD160 = 20×) and digitized as described previously. Individual motor plates were segmented automatically with only minor adjustments and the area was estimated using Image J (n = 6 for PD10 and PD60 and n = 4 for PD160). To evaluate the size of postsynaptic motor elements rats were perfused and their whisker pads were dissected, postfixed, and frozen flattened. Tangential sections (20 μm) were cut, mounted, and incubated with a solution containing rhodamine conjugated α-bungarotoxin (1 μg/ml, In Vitrogen) at 37 °C during one hour. After this step, the sections were washed three times with PB at room temperature and the slides cover-slipped with anti-fade mounting medium (Dako). Image stacks of the labeled postsynaptic motor elements were acquired (40×) and digitized using a scanning confocal microscope (Carl Zeiss LSM Pascal). The stacks were then used to generate bi-dimensional reconstructions that were used to estimate the area of each postsynaptic motor element with the aid of Image J (PD10, PD60 and PD160; n = 4).

Electrophysiology.

The amplitude of EFPs in the intrinsic musculature of facial whiskers was determined in control and enucleated rats (PD 200; n = 6) following the stimulation of the buccolabialis branch of the facial nerve. The right whisker pad of anesthetized rats (ketamine/xylazine; 50 mg/10 mg/kg) was shaved off and each animal's head held in a stereotaxic frame. Additional doses of anesthetics were administered when needed. To avoid sensory feedback to the vibrissa musculature, the infraorbital branch of the trigeminal nerve was transected (54). The facial nerve was exposed at the level of the junction between the buccal and upper marginal mandibular branches before they enter the mystacial pad. An incision was made on the upper border of the whisker mystacial pad and the intrinsic muscles of the A row were carefully exposed. Glass microelectrodes filled with 1.2 M NaCl (1.2 MOhms resistance) were used to record EFPs following the application of electrical stimuli of increasing intensity which were delivered (50 μs pulses/2 s through a 1 mm diameter concentric bipolar electrode) to the buccolabialis nerve. The reference electrode was placed in the midline of the snout. EFPs were amplified and band-pass filtered between 0.3 Hz and 10 kHz with a P511 Grass Amplifier. Signals were digitized with a sampling rate of 33 kHz with an analog-to-digital converter Digidata from Axon Instruments. Average EFPs were obtained from 64 consecutive responses.

Western Blot Analysis for Neurotrophin-3.

The content of neurotrophin-3 in the whisker pads in control and enucleated rats (PD2, PD5, PD7, PD10, PD15, PD30, PD60, and PD200) was estimated. Animals were euthanized and their whisker pads were dissected, frozen, and stored at −70 °C until use. Two whisker pads from different rats of the same age were pooled and then powdered in a mortar with liquid nitrogen. The sample was sonicated in lysis buffer (100 mM Sodium Chloride, 10 mM Tris, 1 mM PMSF, 1% Triton X-100, and a tablet of protease inhibitors; Roche). Samples were centrifuged three times at 18300 g during 15 min at 4 °C. The supernatants were collected and protein content measured by using the Bradford′s assay (Bio-Rad). Equal amounts (40 μg) of protein were electrophorased on 16% SDS/PAGE/Tris-Tricine minigels and blotted semidry onto nitrocellulose membranes (BioRad). Membranes were blocked with Tris buffer-saline (TBS) containing 5% skim milk and 0.3% Tween-20 and incubated with sheep anti-human neurotrophin-3 antibody (1:5000; Chemicon) overnight at 4 °C. After three washes of 5 min with TBS, membranes were incubated with anti-sheep IgG biotinylated secondary antibody (1:10000; 2 h; Chemicon) and then with avidin-peroxidase (1.5 h; Vector) at room temperature. Immunopositive bands were visualized using the ECL chemiluminsescence detection system (Amersham) and recorded on films (Kodak). The films were digitized using a FS BioRad gel documenting system. The quantification of the density of NT-3 was done with the Quantity One Software (version 4.4.1; BioRad). Data were averaged from three experiments and are presented here as the fraction value relative to the optical density obtained for PD2 control rats.

Statistics.

Quantitative data are presented as averages ± standard errors. Group comparisons were carried out by using Student's t tests. Analyses were performed blindly.