The postnatal development of ultrasonic vocalization-associated breathing is altered in glycine transporter 2-deficient mice.

KEY POINTS: Newborn mice produce ultrasonic vocalization to communicate with their mother. The neuronal glycine transporter (GlyT2) is required for efficient loading of synaptic vesicles in glycinergic neurons. Mice lacking GlyT2 develop a phenotype that resembles human hyperekplexia and the mice die in the second postnatal week. In the present study, we show that GlyT2-knockout mice do not acquire adult ultrasonic vocalization-associated breathing patterns. Despite the strong impairment of glycinergic inhibition, they can produce sufficient expiratory airflow to produce ultrasonic vocalization. Because mouse ultrasonic vocalization is a valuable read-out in translational research, these data are highly relevant for a broad range of research fields. ABSTRACT: Mouse models are instrumental with respect to determining the genetic basis and neural foundations of breathing regulation. To test the hypothesis that glycinergic synaptic inhibition is required for normal breathing and proper post-inspiratory activity, we analysed breathing and ultrasonic vocalization (USV) patterns in neonatal mice lacking the neuronal glycine transporter (GlyT2). GlyT2-knockout (KO) mice have a profound reduction of glycinergic synaptic currents already at birth, develop a severe motor phenotype and survive only until the second postnatal week. At this stage, GlyT2-KO mice are smaller, have a reduced respiratory rate and still display a neonatal breathing pattern with active expiration for the production of USV. By contrast, wild-type mice acquire different USV-associated breathing patterns that depend on post-inspiratory control of air flow. Nonetheless, USVs per se remain largely indistinguishable between both genotypes. We conclude that GlyT2-KO mice, despite the strong impairment of glycinergic inhibition, can produce sufficient expiratory airflow to produce ultrasonic vocalization.

Mixed miniature postsynaptic currents resulting from co-release of glycine and GABA recorded from glycinergic neurons in the neonatal respiratory network.

Inhibitory neurons are involved in the generation and patterning of the respiratory rhythm in the adult animal. However, the role of glycinergic neurons in the respiratory rhythm in the developing network is still not understood. Although the complete loss of glycinergic transmission in vivo is lethal, the blockade of glycinergic transmission in slices of the medulla has little effect on pre-Bötzinger complex network activity. As 50% of the respiratory rhythmic neurons in this slice preparation are glycinergic, they have to be considered as integrated parts of the network. We aimed to investigate whether glycinergic neurons receive mixed miniature inhibitory postsynaptic currents (mIPSCs) that result from co-release of GABA and glycine. Quantification of mixed mIPSCs by the use of different objective detection methods resulted in a wide range of results. Therefore, we generated traces of mIPSCs with a known distribution of mixed mIPSCs and mono-transmitter-induced mIPSCs, and tested the detection methods on the simulated data. We found that analysis paradigms, which are based on fitting the sum of two mIPSC templates, to be most acceptable. On the basis of these protocols, 20-40% of all mIPSCs recorded from respiratory glycinergic neurons are mixed mIPSCs that result from co-release of GABA and glycine. Furthermore, single-cell reverse transcriptase polymerase chain reaction revealed that 46% of glycinergic neurons co-express mRNA of glycine transporter 2 together with at least one marker protein of GABAergic neurons. Our data suggest that significant co-transmission occurs in the pre-Bötzinger complex that might be involved in the shaping of synaptic inhibition of respiratory glycinergic neurons.

Development of synaptic inhibition in glycine transporter 2 deficient mice.

Mice deficient for the neuronal glycine transporter subtype 2 (GlyT2) die during the second postnatal week after developing neuromotor deficiencies, which resembles severe forms of human hyperekplexia. This phenotype has been attributed to a dramatic reduction in glycinergic neurotransmission. In the present study we analyzed the development of GABAergic and glycinergic synaptic transmission in GlyT2-knockout mice during early postnatal life. Anti-glycine immunohistochemistry in spinal cord and brainstem slices and whole-cell voltage-clamp recordings of glycinergic inhibitory postsynaptic currents (IPSCs) from hypoglossal motoneurons revealed strikingly reduced levels of synaptic glycine already at birth. Since GABA and glycine use the same vesicular inhibitory amino acid transporter (VIAAT or VGAT) we also analysed GABAergic neurotransmission. No increase of GABA immunoreactivity was observed in the spinal cord and brainstem of GlyT2(-/-) mice at any stage of postnatal development. Correspondingly no up-regulation of GABAergic IPSCs was detected in GlyT2(-/-) hypoglossal motoneurons. These data suggest that in the first postnatal week, loss of the glycine transporter 2 is neither compensated by glycine de-novo synthesis nor by up-regulation of the GABAergic transmission in GlyT2(-/-) mice.

Hyperekplexia phenotype of glycine receptor alpha1 subunit mutant mice identifies Zn(2+) as an essential endogenous modulator of glycinergic neurotransmission.

Zn(2+) is thought to modulate neurotransmission by affecting currents mediated by ligand-gated ion channels and transmitter reuptake by Na(+)-dependent transporter systems. Here, we examined the in vivo relevance of Zn(2+) neuromodulation by producing knockin mice carrying the mutation D80A in the glycine receptor (GlyR) alpha1 subunit gene (Glra1). This substitution selectively eliminates the potentiating effect of Zn(2+) on GlyR currents. Mice homozygous for Glra1(D80A) develop a severe neuromotor phenotype postnatally that resembles forms of human hyperekplexia (startle disease) caused by mutations in GlyR genes. In spinal neurons and brainstem slices from Glra1(D80A) mice, GlyR expression, synaptic localization, and basal glycinergic transmission were normal; however, potentiation of spontaneous glycinergic currents by Zn(2+) was significantly impaired. Thus, the hyperekplexia phenotype of Glra1(D80A) mice is due to the loss of Zn(2+) potentiation of alpha1 subunit containing GlyRs, indicating that synaptic Zn(2+) is essential for proper in vivo functioning of glycinergic neurotransmission.