Reduced gap junctional coupling leads to uncorrelated motor neuron firing and precocious neuromuscular synapse elimination

Personius et al. 10.1073/pnas.0703357104.

Supporting Information

Files in this Data Supplement:

SI Figure 7
SI Methods
SI Table 1




SI Figure 7

Fig. 7. SLD amplitude is insensitive to ventral root stimulation frequency. (A and B) Superimposed intracellular responses evoked by subthreshold stimulation of the ventral root at different frequencies (1, 2, and 5 Hz) in a motor neuron from WT (A) andCx40^-/- (B) mouse pups. In both cases, the membrane potential was held at -60mV, and the stimulus intensity was the same for each stimulus frequency. The red dotted line with arrows shows that the onset of the response time is locked to the stimulus artifact (black triangles). (C and D) Histograms show the mean peak amplitude of the response for motor neurons from WT (C) and Cx40^-/- (D) mice. The small amount of variance in the amplitude of successive responses is probably due to synaptic noise.

Table 1. Motor neuron and muscle fiber number is similar between Cx40^-/- and Cx40^+/+ littermates

Values are shown as mean ± SD; n = 3-4 mice in each experiment. No values were significantly different between genotypes or ages.

SI Methods

Animals and Genotyping. The original genetic background of Cx40^-/- animals was 129sv/B6 that was back-crossed to C57/B6 for at least seven generations. Genomic PCR, as described in ref. 1, was used to genotype the postnatal day (P)0 to P14 pups generated by breeding Cx40^+/- females with Cx40^+/- males. All animal procedures were performed according to NIH guidelines and reviewed by the Animal Care and Use Committee of the University of Pennsylvania, NINDS, NIH, and/or the State University of New York, Buffalo. With the exception of the intracellular recordings from motoneurons, all experiments and analyses were performed blind to genotype.

The primers used for genotyping were: Cx40 5'-TGTCACTATGGTAGCCCTGAG, Cx40 3'-TTTGGCAAGTCACGGCAGGG, Neo 5'-TTCGTCCAGATCATCCTGATC, and Neo 3'-AGAGGCTATTCGGCTATGACT. The 50-ml PCR mixture contained 5 ml of mouse tail genomic DNA, 4 ml of 2.5 mM deoxynucleotide triphosphates, 5 ml of each primer (2.5 mM stock), 2 mM magnesium, 5 ml of 10× PCR buffer (GIBCO-BRL) and 1.25 units of Taq polymerase (GIBCO-BRL). The PCR conditions were 94°C for 3 min; 94 ^oC for 30 sec, 60 ^oC for 30 sec, and 72 ^oC for 30 sec for 35 cycles, followed by 72 ^oC for 10 min.

In Situ Hybridization. Spinal cords were fixed in 4% paraformaldehyde (pH = 7.4) for 2 h, rinsed in PBS, cryoprotected in 20% sucrose in PBS overnight at 4°C, embedded in OCT (Tissue-Tek), frozen in an acetone/dry ice bath, and 20 mm of frozen sections obtained (CM3000 cryostat; Leica). Frozen tissue sections were rinsed with PBS, incubated with acetylation buffer (0.926 g of Triethanolamine, 112 ml of 10 N NaOH, 125 ml of acetic anhydride, and DEPC-H₂O to 50 ml), permeabilized with 1% TritonX-100 in PBS for 30 min, and rinsed with PBS. Slides were then incubated with prehybridization buffer (50% formamide, 4× SSC, 1× Denhardt's solution, 10% dextran sulfate, 250 mg/ml Baker's yeast RNA, and 500 mg/ml herring sperm DNA) at 68-72°C for 2 h. cRNA probe (200-500 ng/ml) was used for hybridization at 68-72°C overnight. Moist chambers and coverslips were used to prevent hybridization buffer from evaporating. The next day, the slides were washed with 0.2× SSC at 68-72°C for 1 h, rinsed with PBS, blocked with 1% BSA-0.1% Triton X-100 in PBS for 1 h at room temperature, and were incubated with alkaline phosphatase-conjugated anti-digoxigenin antibody (1:2,000 to 1:5,000; Boehringer-Mannheim) overnight at 4°C. Slides were then washed extensively with PBS, equilibrated with 0.1M Tris·HCl (pH = 9.5), 0.1M NaCl, and 50 mM MgCl₂ for 5 min. at room temperature. A colorimetric reaction for AP was then developed and stopped after the signal was apparent after examination under a dissecting microscope. Motor neurons were identified by their location in the ventral spinal cord, by their large soma size and, in some experiments, after immunostaining for choline acetyltransferase (Chemicon) or neurofilament (SMI-32; Sternberger Monoclonals) (3). Slides were photographed with a cooled color CCD camera (Hammamatsu) and images were acquired digitally by using interactive software (ImagePro). Composite images of overlapping fields were made with Adobe Photoshop.

Whole-Cell Recordings from Lumbar Spinal Motor Neurons. ACSF composition was 128.35 Mm NaCl, 4 Mm KCl, 0.58 Mm NaH₂PO₄-H₂0, 21 Mm NaHCO₃, 30 Mm D-Glucose, 1.5 Mm CaCl₂-H₂0, and 1 Mm MgSO₄-7H₂0). Ca²⁺-free aCSF composition was 128.35 Mm NaCl, 4 Mm Mm KCl, 0.58 Mm NaH₂PO₄-H₂0, 21 Mm NaHCO₃, 30 Mm D-Glucose, and 2.5 Mm MgSO₄-7H₂0. The mean age for wild-type and mutant mice used in these experiments was similar (wild type 3.4 ±0.4 and Cx40^-/-: 3.5 ± 0.3 days; not significantly different, Student's t test). Patch electrodes were filled with 10 Mm NaCl, 130 Mm potassium gluconate, 10 Mm Hepes, 11 Mm EGTA, 1 Mm MgCl₂, 0.1 Mm CaCl₂, and 1 Mm Na₂ATP, pH adjusted to 7.2-7.3 with KOH; ~305-310 mOsm. The recordings were not corrected for liquid junction potentials.

Motor neurons were identified by the presence of an all-or-none antidromic action potential after a short (0.2 ms) current pulse delivered to the ventral root by a stimulus isolator (A365; WPI, Sarasota, FL). The intracellular responses to either ventral or dorsal root stimulation were recorded (dc to 4 KHz) by using a Multiclamp 700A intracellular amplifier (Molecular Devices). Extracellular recordings from the ventral roots (dc to 3 KHz) were recorded with a Cyberamp extracellular amplifier (Molecular Devices). Both signals were digitized at 10 KHz (Digidata 1320A; Molecular Devices) and were acquired with Clampex software (version 9; Molecular Devices). Intracellular recordings were analyzed offline by using Clampfit (Molecular Devices).

Graded antidromic stimulation of the ventral root evoked several discrete depolarizing components that were sometimes contaminated by synaptic noise. To minimize the effects of such noise, three traces were averaged at each stimulus intensity (see Fig. 2D). The latency of short latency depolarizations (SLDs) and the latency of the antidromically-evoked action potential (AP) were measured as the time from the onset of the stimulus to the onset of the response (see Fig. 2 A and B). The peak amplitude of the SLDs was measured from three averaged responses in a 1- to 1.2-ms window »3.5 ms after the stimulus (see vertical dashed lines in Fig. 2D). Motor neuron input resistance (R_in) was measured by injection of depolarizing and hyperpolarizing current pulses (100-ms duration) at a holding potential of -60 mV and was calculated from the slope of the I/V relationship within its linear range. Motor neuron input resistance (R_in) was not significantly different between Cx40^-/- (49.2 ± 13.1 MW, range: 9.1-131.2 MW) compared with wild-type littermates (38.4 ± 6.7 MW, range: 8.8-63.7 MW; P > 0.05, Student's t test).

To confirm that the SLDs were electrical and not chemical in nature, several tests were performed. First, SLDs were evoked in Ca²⁺-free aCSF to block chemical synaptic transmission in one motor neuron from a wild-type and one from a Cx40^-/- mouse. The number and amplitude of the SLDs were unchanged in Ca²⁺-free aCSF (Fig. 2 A and B Inset; WT seven SLDs; Cx40^-/- three SLDs examined). Second, SLD amplitude was measured at several stimulation frequencies (1, 2, 5, and 10 Hz; WT five motor neurons; Cx40^-/- five motor neurons) because chemical synaptic transmission is rapidly depressed at 5 and 10 Hz (3). SLD peak amplitude was not depressed at 5 or 10 Hz stimulation in any of the WT or Cx40^-/- motor neurons examined (SI Fig. 7; P = 0.11, one-way ANOVA). Finally, the sensitivity of SLD amplitude to changes in membrane potential (-95 mV to 40 mV) was examined (WT four motor neurons; Cx40^-/- five motor neurons) at a stimulus frequency of 1 Hz. In each motor neuron, SLD amplitude was statistically unchanged as membrane potential was altered (P = 0.22, one-way ANOVA; data not shown). Taken together, these data demonstrate that the SLDs reflect electrical coupling among motor neurons, consistent with other work (3, 4, 5).

EMG Recordings and Analyses of Single Motor Unit Activity. A small incision was made in the distal hindlimb to expose the soleus muscle, and a single 75-mm stainless steel microelectrode (500 MW; FHC) together with a 125-mm Teflon-coated tungsten ground wire (A-M Systems) were placed within or upon the soleus muscle near the endplate band. The entire field was then bathed in mineral oil to reduce electrical cross-talk, followed by lidocaine application to the incision site. Recordings began when the animal recovered from anesthesia and began spontaneous movement. Recording sessions generally lasted between 0.5 and 1.0 h. EMG signals were filtered (500 Hz high pass), and differentially amplified (TDT digital bioamplifier; DT Tucker-Davis Technologies). Signals were acquired on a PC using an A/D converter (CED), and single motor units were discriminated offline byusing Spike2 (CED).

On average, three motor units with a sufficient number of spikes for further analysis were identified per EMG record. Mean and median motor unit firing frequency were determined from interspike interval histograms (1,000 60-ms bins spanning a period of 60 s). Cross-correlation analyses were used to determine whether any discriminated motor unit pair showed correlated activity (2). Cross-correlograms were constructed with 100-ms bins over a period of ±25 sec. The index k'-1, defined as the number of events in excess of chance within the peak of the cross-correlogram divided by the events expected within the same region due to chance, was used to determine the strength of correlation between motor unit pairs (4). To examine the correlations at specific time scales, we determined a correlation index, as described by (5), within time-windows of 5, 10, 20, 50, and 100 ms. A correlation index of 1 indicates that the activity of the motor units in a pair is independent, whereas a value >1 indicates that the motor units in a pair are more likely to fire together within a time window than expected by chance.

Structural Analyses of Neuromuscular Innervation. Postsynaptic acetylcholine receptors were labeled with TRITC-conjugated a-bungarotoxin (red in Fig. 5; Molecular Probes, Eugene, OR). Presynaptic motor axons and terminals were labeled by a combination of mouse monoclonal anti-neurofilament SMI 31 (Sternberger Monoclonals) and mouse monoclonal anti-SV2 (Developmental Studies Hybridoma Bank, Iowa City, IA) antibodies, followed by incubation with FITC-conjugated secondary antibodies (green in Fig. 5). Muscles were mounted on slides in Vectashield (Vector Laboratories, Burlingame, CA), and coverslips were sealed with nail polish.

The number of axons innervating neuromuscular junctions was evaluated by using confocal microscopy (TCS 4D system; Leica), and the observer was blind to genotype. At least 30 junctions from each of three muscles (one muscle per mouse) at each age for each genotype were examined. The number of z planes acquired was adjusted so that axons could be followed from a parent axon in the muscle nerve to each junction. Only junctions in which the number of motor axon inputs could be easily counted were included for analysis.

Because both Cx36 and Cx37 are expressed in motor neurons from embryonic through adult life, the number of multiply and singly innervated junctions was counted in the soleus, EDL, and sternomastoid muscles from Cx36^-/- and Cx37^-/- mice and wild-type littermates from P0-P14. The time course of synapse elimination was similar between mutant and wild-type littermates (data not shown), suggesting that the alterations in the time course of synapse elimination are specific to the loss of Cx40.

Electrophysiological Analyses of Neuromuscular Innervation. The soleus muscle and its innervation to the ventral roots were dissected under oxygenated (95% O₂, 5% CO₂) Rees' Ringer's solution (6) in anesthetized mice (7). The muscle was pinned in a Sylgard-lined Petri dish and superfused with oxygenated Ringer's, and ventral root fibers from L3-L5 were carefully split and divided between two suction electrodes and stimulated with square pulses (0.2 mV-2.0 V, 0.2 ms duration). By adjusting the stimulus voltage to each suction electrode and visually monitoring muscle contractions, all ventral root bundles that contained motor axons innervating the soleus muscle were identified.

Muscle contractions were prevented by using Ringer's solution with 2.5-4.5 mM curare, a dose that reduces but does not completely block neuromuscular synaptic transmission. Muscle membrane potentials were amplified by using an Axoprobe 1A amplifier (Axon Instruments), low pass filtered at 1 kHz, digitized at 10 kHz using an A/D converter (DigiData; Axon Instruments) and interactive software (Axoscope; Axon Instruments). By recording from a muscle fiber while independently altering the stimulation voltage to each ventral root bundle, the number of distinct steps in the amplitude of the endplate potential (epp), reflecting the number of motor axon inputs to each muscle fiber, was counted and recorded. A minimum of 20 fibers per muscle, from each of three animals for each genotype, was analyzed. Preparations that contained denervated muscle fibers, indicative of possible damage to axons during the dissection, were not studied further.

Retrograde Labeling of Motor Neurons and Muscle Fiber Counts. The leftward shift in the time course of synapse elimination observed in Cx40^-/- mice might have arisen because of fewer motor neurons innervating each muscle. This would result in a lesser degree of overlap among competing motor units and, thus, a higher proportion of singly innervated muscle fibers. To address this possibility, the motor neurons innervating the soleus, EDL, and sternomastoid muscles were retrogradely labeled with Fluoro-Gold at P7 and P14. P7 and P14 mice were anesthetized as described above, and the sternomastoid, soleus, and/or EDL muscles were exposed. Care was taken during surgery to avoid damaging surrounding muscles. A single injection of 400 nl of 4% Fluoro-Gold in saline was made into each muscle on the left side of the animals. In some animals, the soleus muscle was injected on the right side, whereas the EDL muscle was injected on the left side. The wounds were sutured closed, and the pups were returned to their mother. A week later, the pups were killed with an overdose of a solution of 17.4 mg/ml ketamine and 2.6 mg/ml xylazine (Phoenix Pharmacia, St. Joseph, MO). Their brainstems and spinal cords were dissected and processed as described above. Serial horizontal sections were made at 35 mm, mounted on glass slides in Vectashield (Vector Laboratories), and coverslips were sealed with nail polish. Fluoro-Gold-labeled motor neurons were counted by using a fluorescence microscope with an UV filter set (A cube; Leica), imaged with a cooled color CCD camera (Hammamatsu), and images were acquired digitally (ImagePro). To prevent double counting, only cells in which a nucleus was seen were counted. In young neonatal mice, leakage of Fluoro-Gold from the injected to surrounding muscle is common, thus the number of motor neurons innervating a single muscle can be overestimated. Such leakage occurred in the experiments reported here, resulting in a large number of labeled motor neurons, and a small change in the number of motor neurons between genotypes may thus have been missed. However, the pattern and extent of motor neuron labeling was indistinguishable between genotypes. The counts of labeled cell bodies in the spinal cord showed that Cx40^-/- mice had the same number of motor neurons as wild-type animals (P = 0.27; Student's t test; Table 1).

Given that the number of retrogradely labeled motor neurons was unaffected by the absence of Cx40 expression, the time course of synapse elimination might also be altered by an increase in the number of muscle fibers. To address this possibility, the number of muscle fibers was counted from cross-sections of the soleus, EDL, and sternomastoid muscle in each genotype. These muscles were dissected, pinned to styrofoam, covered with OCT (Tissue Tek), and frozen in isopentane cooled in liquid nitrogen to at least -80°C. Twenty-micron-thick cross-sections were obtained through the middle of each muscle, mounted on glass slides, stained with hematoxylin and eosin as described in Gonzalez et al. (8), and imaged by using a cooled color CCD camera (Hammamatsu). Images were acquired by using interactive software (ImagePro). This analysis showed that a similar number of muscle fibers was present in muscles from Cx40^-/- mice compared with wild-type littermates at P7 and P14 (P = 0.21; Student's t test; Table 1).

Statistical Analyses. All data are presented as mean ± SEM., unless otherwise noted. Data expressed as percentages were processed through an arc-sine transformation before Student's t test for significance was performed. The index k'-1 and correlation index values were analyzed by a Mann-Whitney rank sum test or Kruskal-Wallis one-way analysis of variance on ranks, because these values had a nonnormal distribution. Statistical analyses were performed in Sigma Stat (SPSS) and Matlab (MathWorks).

1. Bevilacqua LM, Simon AM, Maguire CT, Gehrmann J, Wakimoto H, Paul DL, Berul CI (2000) J Interv Card Electrophysiol 4:459-467.
2. Lev-Tov A, Pinco M (1992) J Physiol 447:149-69.

3. Chang Q, Gonzalez M, Pinter MJ, Balice-Gordon RJ (1999) J Neurosci 19:10813-28.

4. Mentis GZ, Diaz E, Moran LB, Navarrete R (2002) J Physiol 544, 757-64.

5. Walton KD, Navarrete R (1991) J Physiol 433:283-305.
6. Nordstrom MA, Fuglevand AJ, Enoka RM (1992) J Physiol 453:547-574.
7. Sears TA, Stagg D (1976) J Physiol 263:357-381.
8. Gonzalez M, Ruggiero FP, Chang Q, Shi YJ, Rich MM, Kraner S, Balice-Gordon RJ (1999) Neuron 24:567-583.

• Early Edition
• Archives
• Online Submission

• Feature Articles

  (Expanded research articles of exceptional breadth)

• Commentaries

  (Expert commentary on leading research)

• Letters

  (Online-only letters to the editor)

• Inaugural Articles

  (Articles by recently elected NAS members)

• PNAS Profiles

  (Biographical profiles of Academy members)

• Sustainability Science

  (Interactions of human and environmental systems)

• Collected Papers

  (Editorials, Perspectives, Reviews, Colloquia, etc.)

• Special Features

  (Solicited articles on exceptional research topics)

• Sackler Colloquia

  (Scientific reports held under NAS auspices)