ERR2 and ERR3 promote the development of gamma motor neuron functional properties required for proprioceptive movement control

Abstract

The ability of terrestrial vertebrates to effectively move on land is integrally linked to the diversification of motor neurons into types that generate muscle force (alpha motor neurons) and types that modulate muscle proprioception, a task that in mammals is chiefly mediated by gamma motor neurons. The diversification of motor neurons into alpha and gamma types and their respective contributions to movement control have been firmly established in the past 7 decades, while recent studies identified gene expression signatures linked to both motor neuron types. However, the mechanisms that promote the specification of gamma motor neurons and/or their unique properties remained unaddressed. Here, we found that upon selective loss of the orphan nuclear receptors ERR2 and ERR3 (also known as ERRβ, ERRγ or NR3B2, NR3B3, respectively) in motor neurons in mice, morphologically distinguishable gamma motor neurons are generated but do not acquire characteristic functional properties necessary for regulating muscle proprioception, thus disrupting gait and precision movements. Complementary gain-of-function experiments in chick suggest that ERR2 and ERR3 could operate via transcriptional activation of neural activity modulators to promote a gamma motor neuron biophysical signature of low firing thresholds and high firing rates. Our work identifies a mechanism specifying gamma motor neuron functional properties essential for the regulation of proprioceptive movement control.

Fig 1

Direct electrophysiological interrogation of alpha and gamma motor neurons (A-D). (A) Transversal section of P21 (wild-type) mouse lumbar spinal cord ventral horn: motor neurons labeled by retrograde tracer Fluoro-Gold (FG). Putative alpha motor neurons retain low levels of FG in soma (FG^low) (open arrowheads), while putative gamma motor neurons retain high levels of FG in soma (FG^high) (arrowheads) (scale bar: 50 μm). (B) Imaris 3D reconstruction: VGLUT1⁺ synaptic varicosities (triangles) associated with FG^low and NeuN^high alpha motor neuron (open arrowheads), but not with adjacent FG^high and NeuN^{low or negligible} gamma motor neurons (arrowheads) (scale bar: 20 μm). (C) Whole-cell patch-clamp recordings: example traces of FG^high (black) and FG^low (gray) motor neurons upon 900 pA, 1 second square current pulse. (D) Scatter plot: FG^high (n = 24, N = 16) and FG^low (n = 22, N = 12) motor neurons exhibit divergent gamma and alpha subtype-defining electrophysiological signatures, respectively, including gamma subtype-specific combination with low rheobase and high gain by FG^high motor neurons (see S1 Table for details). (E) FG^high motor neurons have significantly lower rheobase (pA) (221.87 ± 31.34), higher firing frequency (Hz) (50.65 ± 3.23), higher gain (Hz/nA) (161.01 ± 9.77), higher input resistance (136.62 ± 14.63), and lower capacitance (76.07 ± 6.01) when compared to FG^low motor neuron rheobase (909.09 ± 82.11), firing frequency (20.41 ± 1.70), gain (32.64 ± 4.02), input resistance (36.55 ± 4.16), and capacitance (239.1 ± 17.17), respectively (see S1 Table for details). Data are presented as mean ± SEM. n = # of neurons, N = # of mice. Statistically significant differences between FG^high and FG^low neurons are indicated as ***p < 0.001, Student t test). Data for Fig 1D and 1E can be found in S1 Data.

Fig 2. High levels of correlated ERR2/3 expression by gamma motor neurons.

(A-D) Transversal section of P21 Chat^Cre; Rosa26^floxtdTomato mouse lumbar spinal cord ventral horn: motor neurons genetically labeled by tdTomato (scale bar: 50 μm). Arrowheads: high ERR2 and ERR3 levels in small NeuN^low, tdTomato⁺ motor neuron nuclei. Open arrowheads: relatively moderate-to-high levels ERR2/3 levels in NeuN^high tdTomato⁻ interneurons. Triangles: consistently lower but detectable levels in some large motor neurons with moderate-to-high NeuN levels. (E-H). Transversal section of P21 (wild-type) mouse lumbar spinal cord ventral horn: motor neurons labeled by retrograde tracer Fluoro-Gold (FG) (scale bar: 50 μm). Arrowheads: high ERR2 and ERR3 levels in small NeuN^low that retain high levels of FG^high. Open arrowheads: relatively moderate-to-high levels ERR2/3 levels in NeuN^high FG⁻ interneurons. Triangles: consistently lower but detectable levels in some large motor neurons with moderate-to-high NeuN levels. (I) Quantitative analysis: high ERR2 (red) and ERR3 (green) levels in motor neurons with small somas. (I’) L Linear regression: strong positive correlation of motor neuron Err2 and Err3 levels. (F) High Err2 (red) and NeuN (black) levels in nonoverlapping motor neuron populations (n = 93, N = 3, Pearson’s correlation coefficient r = 0.86). (J) High ERR2 (red) and low NeuN (blue) levels in motor neurons with small somas. (J’) Linear regression: lack of positive correlation between Err2 and NeuN levels in motor neurons (n = 93, N = 3, Pearson’s correlation coefficient r = −0.32). (K) Loss of high ERR2 (red) and ERR3 (green) levels by Egr3-deficient small motor neurons (n = 184, N = 3). (L-N) Imaris 3D reconstruction: VGLUT1⁺ synaptic varicosities associated with ERR2^low NeuN^high motor neuron, but not with adjacent ERR2^high NeuN^low motor neuron ERR2^low motor neuron (motor neurons retrogradely labeled by FG) (scale bar: 20 μm). (N) Quantification of VGLUT1⁺ synaptic varicosities associated with ERR2^high or ERR2^low motor neurons. N = # of mice and n = # of neurons. Statistically significant differences between motor neurons are indicated as: *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant, Student t test). Data for Fig 2I, 2J, 2K, and 2N can be found in S2 Data.

Fig 3. ERR2/3 are required for the acquisition of a gamma motor neuron biophysical signature.

(A) Example traces of whole-cell patch-clamp recordings upon 240, 380, or 680 pA, 1 second square current pulses from small soma motor neurons exhibiting high levels of FG incorporation (FG^high) from control mice or ERR2/3^cko mice. Control FG^high motor neurons (A’) exhibit higher firing rates compared to FG^high ERR2/3^cko motor neurons (A”) and gear up their firing rates more rapidly in response to current pulses. (B) Scatter plot: lack of segregation of ERR2/3^cko FG^high gamma motor neuron and ERR2/3^cko FG^low alpha motor neuron electrophysiological signatures. (C) ERR2/3^cko FG^high motor neurons show higher rheobase (419.72 ± 84.11), lower firing frequency (27.61 ± 3.15), and lower gain (84.01 ± 12.34) compared control FG^high motor neuron rheobase (221.87 ± 31.34), firing frequency (50.65 ± 3.23), gain (161.01 ± 9.77), respectively. (D) Example traces of whole-cell patch-clamp recordings upon 900, 1,040, or 1,340 pA, 1 second square current pulses from large soma motor neurons with lower levels of FG incorporation (FG^low) from control mice or ERR2/3^cko mice. ERR2/3^cko FG^high motor neurons exhibit comparable firing rates and properties to control FG^low motor neurons. (E) Scatter plot: segregation of electrophysiological signatures between control FG^high gamma motor neurons versus control FG^low alpha motor neurons. (Note: Data are from Fig 1D). (F) No significant differences seen in ERR2/3^cko FG^low alpha motor neuron subtype rheobase (855.55 ± 124.85), firing frequency (25.03 ± 3.76), and gain (43.04 ± 9.81) when compared to control FG^low alpha motor neuron subtype rheobase (909.09 ± 82.11), firing frequency (20.41 ± 1.70), gain (32.64 ± 4.02), respectively (see S1 Table for details). Data are presented as mean ± SEM. n = # of neurons and N = # of mice. Statistically significant differences between FG^high and FG^low neurons are indicated as *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant, Student t test). Data for Fig 3B, 3C, 3E, and 3F can be found in S1 Data.

Fig 4. Generation of distinguishable gamma motor neurons in the absence of ERR2/3.

(A-I) Imaris 3D reconstruction: VGLUT1⁺ synaptic varicosities (green) associated with large FG^low, NeuN^high motor neurons in control (A, B) and ERR2/3^cko mice (C, D) (scale bar: 20 μm). (E-H) Absence of VGLUT1⁺ puncta on FG^high, NeuN^low motor neurons in control (E, F) and ERR2/3^cko mice (G, H) (scale bar: 20 μm). (I) Significant difference in VGLUT1⁺ varicosities associated with FG^high, NeuN^low and FG^low, NeuN^high motor neurons from control mice. Lack of significant differences in VGLUT1⁺ synaptic varicosities associated FG^low, NeuN^high motor neurons in control (20 ± 3.48, n = 9) versus ERR2/3^cko mice (23.1 ± 4.06, n = 10). Lack of significant differences in VGLUT1⁺ varicosities between control FG^high, NeuN^low (1.1 ± 0.36, n = 10) and ERR2/3^cko FG^high, NeuN^low motor neurons (2.4 ± 0.78, n = 10). (J-O) P300 mouse extensor digitorum longus (EDL) muscle spindles of control (J-L) and ERR2/3^cko (M-O) mice. (J-L) Distribution of Ia sensory annulospiral endings in the central spindle segment (visualized by VGLUT1, green), motor innervation (VACHT, magenta) and their postsynaptic sites (arrowheads) on intrafusal fibers (BTX, alpha bungarotoxin, grey) (scale bar: 100 μm). (K-O) Normal appearance of ERR2/3^cko muscle spindles, including Ia sensory annulospiral endings and motor innervation of BTX⁺ postsynaptic sites (arrowheads) on intrafusal fibers (scale bar: 100 μm). Data are presented as mean ± SEM. n = # of neurons. Statistically significant differences control and ERR2/3^cko motor neurons are indicated as: *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant, Student t test). Data for Fig 4I can be found in S2 Data.

Fig 5. ERR2/3 expressed by gamma motor neurons is required for normal locomotion and precision movements.

(A, B) (A) Polygon graphs based on partial least squares (PLS) analysis of 58 gait variables measured during treadmill locomotion at 25 m•s⁻¹ reveals significant gait alterations in ERR2/3^cko mice (n = 8, 3 trials/animal) when compared to control mice (n = 9, 3 trials/animal). Each dot represents a single animal, polygons group animals of the same genotype, the segregation of which along the x-axis indicate that ERR2/3^cko mice exhibit significant gait alterations at all speeds tested (ERR2/3^cko mice, 3 trials each). (B) Negative (black) or positive (light blue) changes (arbitrary units) in gait variables in ERR2/3^cko compared to control mice ranked by predictive value independent of sign, including reduced stance-swing phase ratio (gait phase related), reduced propulsion velocities (force related), decreased stance width (posture related), and increased paw angle variability (stride related). (C) Still images of ERR2/3^cko mouse navigating a horizontal beam. Red arrow: (example of a “miss”): foot missing the beam during swing-stance transition, causing the hind limb and animal to slip during swing phase (red arrow in C). (D) ERR2/3^cko (18.0 ± 1.75, 55.0 ± 2.20, 79.0 ± 3.49) but not control (1.0 ± 0.175, 1.0 ± 0.125, 1.0 ± 0.125) mice exhibit dramatically increasing erratic locomotion (18-, 55-, and 79-fold increase in the # of misses) upon navigating horizontal beams with decreasing width 30 mm, 25 mm, and 20 mm, respectively (control N = 4, ERR2/3^cko N = 4, 4–5 trials/animal). (E) Still images of ERR2/3^cko mouse navigating a horizontal ladder. Examples of a “miss” (red arrow in upper panel): foot missing a rung during swing-stance transition, causing the hind limb to slip during swing phase. Note: Animals frequently attempted to compensate such misses by using the “slipped” hind limb to push against the rung and propel itself forward (red arrow in lower panel). (F) ERR2/3^cko (7.12 ± 2.06) exhibit significantly more Erratic locomotion (7-fold increase in the # of misses) when compared to control (1.0 ± 0.55) upon navigating horizontal ladder (control N = 5, ERR2/3^cko N = 5, 4–5 trials/animal). (G) Example traces showing Ia spindle afferent responses during resting length (REL) or stretch (STR) applied by force transducer: normal STR responses, but reduced REL firing of Ia afferents in ERR2/3^cko mice (N = 10) when compared to control mice (N = 10). (G’) Pie charts: ratio of Ia afferents firing above or below 2 Hz at REL (total trials: control n = 84, ERR2/3^cko n = 118; control >2 Hz n = 79, ERR2/3^cko >2 Hz n = 68). Data are presented as mean ± SEM. N = # of mice, n = # of 1a afferents. Statistically significant differences between control and ERR2/3^cko mice are indicated as: *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant, Student t test). Data for Fig 5A, 5B, 5D, and 5F can be found in S3 Data.

Fig 6. ERR2/3 promote a gamma motor neuron-like biophysical signature in chick through transcriptional activation.

(A-D’) Overview of E12 chick spinal cord: stable transfection with expression vector driving eGFP expression (A) in ventral horn motor neurons. Higher magnification of eGFP expressing neurons further identified as motor neurons by retrograde cholera toxin B (CTxB) tracing upon in ovo injection into the hind limb (A’) (scale bar: 100 μm). Examples of expression and nuclear localization of mouse ERR2 (B) and ERR3 (C) in chick motor neurons (detected via N-terminal V5 epitope tags) (scale bar: 30 μm). eGFP⁺ motor neuron recorded with patch pipette containing Alexa Fluor 568 dye (red) (D and D’) (scale bar: 50 μm). (E, F) Example traces of current clamp recordings of chick motor neurons (in acute spinal cord slice preparations) forcedly expressing eGFP only (control) (E) or ERR2:VP16 and eGFP (F), upon 100, 200, and 400 pA, 1 s square current pulses: motor neurons expressing elevated ERR2 levels exhibit higher firing rates and gear up their firing rates more rapidly in response to current pulses (F). (G) Forced ERR2 and ERR2:VP16 expression (n = 21, n = 20, respectively) shifts chick motor neuron properties towards a gamma motor neuron-like biophysical signature (high firing rates, low rheobases) compared to control (white, n = 23). (H) Decrease in rheobase upon ERR2, ERR3 (n = 21) transfection, yet not significant, but significant decrease in rheobase upon ERR2:VP16 (n = 20) or ERR3:VP16 (n = 10). No decrease in rheobase upon ERR2:EnR (n = 17) transfection. (I) Significant increases in gain upon transfection of ERR2, ERR3, ERR2:VP16 or ERR3:VP16, but not upon ERR2:EnR transfection. (H, I) Similar decreases in rheobase (H), increase in gain (I) upon ERR2+ERR3 (n = 13) cotransfection, compared to ERR2 or ERR3 single transfection. No differences detected between control 1 and control 2 with parameters recorded during two different experiments at different time points upon expressing “eGFP only” control vectors, thus demonstrating robustness of the assay. Data are presented as mean ±SEM. n = # of neurons. Statistically significant differences are indicated as: *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant, Student t test). Data for Fig 6G-J can be found in S4 Data.

Fig 7. ERR2/3 may in part operate through Kcna10 to promote a gamma motor neuron-like biophysical signature.

(A) Heatmap based on transcript reads (transcript per million (TPM)) detected by RNA sequencing: ERR2 (n = 4) and Dlk1 (n = 4) promote different gene expression signatures in chick motor neurons when compared to control (n = 4), including Kcna10 (red) by ERR2 (only up-regulated genes are shown). (B) Kcna10 genomic locus: 3 clustered ERR2/3 binding sites within promoter region. (C) Reporter tdTomato fluorescence driven by wild-type (WT) Kcna10 promoter is boosted by ERR2 cotransfection (red, n = 102) in chick motor neurons when compared to control (gray, n = 100) and decrease in tdTomato reporter fluorescence upon mutating the ERR2/3 binding sites (black, n = 101). (D) Whole-cell patch-clamp recordings: forced ERR2:VP16 expression (red, n = 20) shifts chick motor neuron properties towards a gamma motor neuron-like biophysical signature (high firing rates, low rheobases) compared to control (white, n = 23). Forced Kcna10 (black, n = 24) expression partially recapitulates the promotion of a gamma motor neuron biophysical signature by ERR2 in chick motor neurons, when compared to control (white, n = 23) motor neurons. (E) Significant decreases in rheobase upon Kcna10 (n = 24) transfection. (F) Significant increase in firing frequency upon Kcna10 (n = 24) transfection. (G) No significant increase in gain upon Kcna10 (n = 24) transfection. Data are presented as mean ± SEM, n = number of experiments or neurons. Statistically significant differences are indicated as: *p < 0.05, **p < 0.01, ***p < 0.001, n.s. = not significant, Student t test). Data for Fig 7A and 7C can be found in S5 Data and data for Fig 7D–7G can be found in S2 Data. The complete set of values for the RNAseq experiments can be found in S6 Data.

Fig 8. ERR2/3-dependent expression of Kv1.8 (protein product of Kcna10) in gamma motor neurons in mouse.

(A-D). Transversal section of adult (mouse lumbar spinal cord ventral horn (scale bar: 50 μm)). Bold arrowheads: high Kv1.8 levels in small ERR3^high NeuN^low tdTomato⁺ motor neurons. Triangles: low or not-detectable Kv1.8 levels in ERR3^low NeuN^high tdTomato⁺ motor neurons. Open arrowheads: some ERR3^high NeuN^low tdTomato⁺ motor neurons possess low or not-detectable Kv1.8 levels. (E-H) Loss of detectable Kv1.8 levels in NeuN^low tdTomato⁺ motor neurons (arrowheads) in ERR2/3^cko mice. (I) Two-dimensional scatter plot of relative fluorescence intensities over soma sizes: high Kv1.8 levels in NeuN^low motor neurons with small somas and low Kv1.8 levels in NeuN^high motor neurons with larger somas in control mice (n = 84, N = 3). (J) Loss of high Kv1.8 levels in NeuN^low motor neurons with smaller somas in ERR2/3^cko mice (n = 84, N = 4). (K) Summary: ERR2/3 promote a gamma motor neuron biophysical signature required for spindle-dependent proprioceptive movement control, possibly by operating through the activation of neural activity modulator genes including Kcna10 (and likely others), while the generation of morphologically distinguishable gamma motor neurons proper may involve other factors operating earlier in development. n: number of neurons, N: number of mice. Data for Fig 8I and 8J can be found in S2 Data.