Structural and functional features of medium spiny neurons in the BACHDΔN17 mouse model of Huntington’s disease

In the BACHD mouse model of Huntington’s disease (HD), deletion of the N17 domain of the Huntingtin gene (BACHDΔN17, Q97) has been reported to lead to nuclear accumulation of mHTT and exacerbation of motor deficits, neuroinflammation and striatal atrophy (Gu et al., 2015). Here we characterized the effect of N17 deletion on dorsolateral striatal medium spiny neurons (MSNs) in BACHDΔN17 (Q97) and BACWTΔN17 (Q31) mice by comparing them to MSNs in wildtype (WT) mice. Mice were characterized on a series of motor tasks and subsequently whole cell patch clamp recordings with simultaneous biocytin filling of MSNs in in vitro striatal slices from these mice were used to comprehensively assess their physiological and morphological features. Key findings include that: Q97 mice exhibit impaired gait and righting reflexes but normal tail suspension reflexes and normal coats while Q31 mice do not differ from WT; intrinsic membrane and action potential properties are altered -but differentially so- in MSNs from Q97 and from Q31 mice; excitatory and inhibitory synaptic currents exhibit higher amplitudes in Q31 but not Q97 MSNs, while excitatory synaptic currents occur at lower frequency in Q97 than in WT and Q31 MSNs; there is a reduced total dendritic length in Q31 -but not Q97- MSNs compared to WT, while spine density and number did not differ in MSNs in the three groups. The findings that Q31 MSNs differed from Q97 and WT neurons with regard to some physiological features and structurally suggest a novel role of the N17 domain in the function of WT Htt. The motor phenotype seen in Q97 mice was less robust than that reported in an earlier study (Gu et al., 2015), and the alterations to MSN physiological properties were largely consistent with changes reported previously in a number of other mouse models of HD. Together this study indicates that N17 plays a role in the modulation of the properties of MSNs in both mHtt and WT-Htt mice, but does not markedly exacerbate HD-like pathogenesis in the BACHD model.

Methods
Behavior Data Videos: 

Mice were weighed and assessed for quality of their coat prior to being assessed on three different tests of motor function including: tail suspension, self-righting, and gait assessment [49]. Coat quality was rated on a scale of 0 (normal, shiny)- 4 (matted, unkempt, yellowing). During testing, mice were videotaped and subsequently videos were analyzed by three researchers blinded to gender and genotype. For tail suspension, mice were suspended by their tails, one third of the length away from the base of the tail. Mice were suspended and videotaped for 15 seconds. The tail suspension task was scored on a scale of 0 (normal splaying of toes and legs)- 4 (curled toes, both legs close to body) For assessment of self-righting ability, mice were placed onto their backs and then released. This was repeated twice and videotaped for 20 seconds each time. Self-righting was scored as 0 (mouse rights self under 5 seconds)- 4 (mouse takes more than 5 seconds to right itself). For assessment of gait, mice were placed in an 18 inch x 18 inch x 5-inch arena and videotaped for 60 seconds. Mice were assigned a gait score from 0 (normal limb movement, feet under body)- 4 (limbs splay when walking, unsteady) based on the scoring system previously reported [49]. Total score was obtained by summing the scores on tail suspension, self-righting, gait, and coat quality.

Electrophysiology .DAT files:

Whole-cell patch clamp recordings were obtained from visually identified MSNs in the dorsolateral quadrant of the striatum [53, 55]. Electrodes were pulled on a Flaming and Brown horizontal pipette puller (model P87, Sutter Instrument) and filled with potassium methanesulfonate (KMS) internal solution, concentrations inmMas follows: (KCH3SO3 122, MgCl2 2, EGTA 5, NaHEPES 10, Na2ATP 5 and 1% biocytin). Electrodes in Ringer’s solution had a resistance of 4–6MO. Electrophysiology data was obtained using PatchMaster software (HEKA Elektronik) and EPC-9/EPC-10 amplifiers (HEKA Elektronik). Assessment of intrinsic membrane and action potential properties. Passive membrane properties (resting membrane potential–Vr-, input resistance–Rn- and membrane time constant –τ-) and action potential firing properties were assessed under current clamp [50–53]. Vr was measured as the voltage in the absence of current injection. A series of 200 ms or 2s hyperpolarizing and depolarizing current steps was applied for the rest of the measures. The voltage responses to each step were measured at steady state and plotted on a voltage-current graph: Rn was calculated as the slope of the best-fit line through the linear portion of the plot. Membrane time constant was measured by fitting a single exponential function to the membrane voltage response to the -10 pA hyperpolarizing step. Rheobase was determined with a 10 s depolarizing current dual ramp (0–100 pA, 0–200 pA; 3.03 kHz sampling frequency). Single AP properties, including threshold and amplitude, were measured on the second evoked AP in a 200 ms current-clamp series in which the current step elicited 3 or more action potentials. An expanded timescale and the linear measure tool were used in FitMaster analysis software (HEKA Elektronik). Finally, a series of 2 s hyperpolarizing and depolarizing steps (-200 to +450 pA, using 25 or 50 pA increments, 12.5kHz sampling frequency) was used to assess repetitive AP firing. Firing rate in response to current steps was analyzed with a generalized linear model, using the genotype, MSN type, rheobase, input resistance, injected current level and their respective interactions as independent variables. Assessment of spontaneous excitatory postsynaptic currents. AMPA receptor-mediated spontaneous excitatory currents (sEPSCs) were recorded for 2 min at a holding potential of -80 mV (6.67 kHz sampling frequency). MiniAnalysis software (Synaptosoft) was used to assess synaptic current properties including: frequency, amplitude, area, time to rise and time to decay. For assessment of kinetics, the rise and decay of averaged traces were each fit to a single- exponential function. For all synaptic current measurements, the event detection threshold was set at the maximum root mean squared noise level (5 pA). Morphology XLS files (reconstructions are posted at NeuroMorpho.org: Streptavidin-Alexa labeling of biocytin filled neurons Following recordings, brain slices were sandwiched between filter paper in fixative (4% paraformaldehyde) overnight at 4˚C. Next, slices were washed in 0.1Mphosphate buffered saline (PBS) 3x 5 minutes. Slices were then incubated in 0.1% Tx-100/PBS for 2 h at room temperature (RT), then incubated in Streptavidin-Alexa 568 (1:500, 0.1% Tx-100/PBS) for 2d at 4˚C, followed by subsequent washes in 0.1MPBS and stored in anti-freeze solution (30% glycerol, 30% ethylene glycol in 0.05Mphosphate buffer) [50–53].

Confocal imaging

For verification of eGFP labeling of biocytin-filled cells, slices were placed in an inverted well slide and temporarily coverslipped for an initial imaging of the soma using a 40x oil immersion objective on a Leica SPE confocal microscope. Cell somata were scanned in their entirety in two channels, 568 and 488, to detect filled cells and the presence or absence of somal eGFP, respectively. Cells were classified as D2/eGFP+ if the 488 and 568 signals overlapped along the x-, y-, and z- planes, D1/eGFP- cells lacked eGFP signal in their soma. Brain slices that contained cells met rigid criteria for morphometric analysis (zero to minimal dendritic varicosities, high signal-to-noise ratio indicative of a well filled cell, few cut branches) were either mounted in Prolong Antifade (Life Technologies) after streptavidin-Alexa staining or used for immunohistochemistry. MSNs were scanned for dendritic morphometric analyses in their entirety using a Leica SPE confocal microscope with a 40x oil immersion objective obtaining a voxel size of 0.27 x 0.27 x 0.5 μm (as described previously [50,51]). Images were deconvolved with AutoQuant Software and 8-bit images imported into NeuronStudio for reconstruction and quantitative analysis. For assessment of spines, 3 dendrites from each cell were chosen for imaging of spines. Each dendrite was selected by dividing the entire dendritic arbor into equal thirds and selecting the dendrite that were located perpendicular to the z-plane. In order to obtain the necessary resolution for spine sub-typing, dendrites were imaged with a 63x oil emersion objective (1.4 NA) with a 2.5 zoom using a Leica SPE laser scanning confocal microscope. The resulting voxel size was 0.034 x 0.034 x 0.17 μm. Each dendrite was imaged in its entirety from the soma to the distal dendrite ending. Z-stacks were deconvolved using Auto-Quant software and 8-bit images imported into Neurolucida 360 for reconstruction, spine sub-typing, and analysis. Spine sub-types were classified based on spine head diameter and spine head distance from the dendritic shaft. Spines were classified as: thin (diameter .0.6 μm), mushroom (diameter > 0.6 μm), stubby (spines lacking a neck), or filopodia (length > 3 μm) [50,52].