Shawn Tan

Osmoregulated periplasmic glucans are needed for competitive growth and biofilm formation by Salmonella enterica serovar Typhimurium in leafy-green vegetable wash waters and colonization in mice

Osmoregulated periplasmic glucans (OPGs) are major periplasmic constituents of Gram-negative bacteria. The role of OPGs has been postulated in symbiotic as well as pathogenic host-microorganism interactions. Here, we report the role of OPGs from Salmonella enterica serovar Typhimurium during growth and biofilm formation in leafy-green vegetable wash water. The opgGH mutant strain, which was defective in OPG biosynthesis, initiated the growth at a slower rate in wash waters obtained from spinach, lettuce and green collard and severely impaired biofilm formation. The lack of OPG synthesis did not influence biofilm formation by the opgGH mutant in low-nutrient low-osmolarity laboratory media. In coculture experiments initiated with equal proportions of cells, the opgGH mutant was outnumbered by the wild-type strain under the planktonic as well as the biofilm growth conditions. The opgGH mutant strain poorly colonized mouse organs when introduced orally along with the wild-type strain. This is the first report demonstrating the role of OPGs of Salmonella in competitive colonization of biofilms, planktonic cultures and mouse organs.

Improving the interpretability of deep neural networks with stimulated learning

Deep Neural Networks (DNNs) have demonstrated improvements in acoustic modelling for automatic speech recognition. However, they are often used as a black box, and not much is understood about what each of the hidden layers does. We seek to understand how the activations in the hidden layers change with different input, and how we can leverage such knowledge to modify the behaviour of the model. To this end, we propose stimulated deep learning where stimuli are introduced during the DNN training process to influence the behaviour of the hidden units. Specifically, constraints are applied so that the hidden units of each layer will exhibit phone-dependent regional activities when arranged in a 2-dimensional grid. We demonstrate that such constraints are able to yield visible activation regions without compromising the classification of the network and suppressing the activations for a region affects the classification accuracy of the corresponding phone more than the others.

The effect of the mGlu5 negative allosteric modulator MTEP and NMDA receptor partial agonist D-cycloserine on Pavlovian conditioned fear

The metabotropic glutamate receptor 5 (mGlu5) and N-methyl-D-aspartate (NMDA) receptor are critical for processes underlying synaptic plasticity, such as long-term potentiation. mGlu5 signaling increases neuronal excitability and potentiates NMDA receptor currents in the amygdala and the hippocampus. The present study examined the involvement of mGlu5 in the acquisition and consolidation of conditioned fear to a tone and context in mice, and explored the functional relationship between mGlu5 and NMDA receptors in this regard. Experiment 1 showed that systemic administration of the mGlu5 negative allosteric modulator 3-[(2-methyl-1,3-thiazol-4-yl)ethynyl]pyridine (MTEP) prior to conditioning significantly attenuated cue-elicited freezing during fear conditioning, which suggests that mGlu5 is necessary for the formation of a tone-shock association. This effect was dose-related (Experiment 2) and not due to any effects of MTEP on shock sensitivity or state-dependency (Experiment 3). Post-conditioning injection of MTEP had no effects (Experiment 4). Although post-conditioning injection of the NMDA receptor partial agonist D-cycloserine (DCS) alone facilitated consolidation of conditioned fear (Experiment 6), it was not able to rescue the acquisition deficit caused by MTEP (Experiment 5). Taken together, these findings indicate a crucial role for mGlu5 signaling in acquisition and NMDA receptor signaling in consolidation of conditioned fear.

Regulation of Brain-Derived Neurotrophic Factor Exocytosis and Gamma-Aminobutyric Acidergic Interneuron Synapse by the Schizophrenia Susceptibility Gene Dysbindin-1

BACKGROUND Genetic variations in dystrobrevin binding protein 1 (DTNBP1 or dysbindin-1) have been implicated as risk factors in the pathogenesis of schizophrenia. The encoded protein dysbindin-1 functions in the regulation of synaptic activity and synapse development. Intriguingly, a loss of function mutation in Dtnbp1 in mice disrupted both glutamatergic and gamma-aminobutyric acidergic transmission in the cerebral cortex; pyramidal neurons displayed enhanced excitability due to reductions in inhibitory synaptic inputs. However, the mechanism by which reduced dysbindin-1 activity causes inhibitory synaptic deficits remains unknown. METHODS We investigated the role of dysbindin-1 in the exocytosis of brain-derived neurotrophic factor (BDNF) from cortical excitatory neurons, organotypic brain slices, and acute slices from dysbindin-1 mutant mice and determined how this change in BDNF exocytosis transsynaptically affected the number of inhibitory synapses formed on excitatory neurons via whole-cell recordings, immunohistochemistry, and live-cell imaging using total internal reflection fluorescence microscopy. RESULTS A decrease in dysbindin-1 reduces the exocytosis of BDNF from cortical excitatory neurons, and this reduction in BDNF exocytosis transsynaptically resulted in reduced inhibitory synapse numbers formed on excitatory neurons. Furthermore, application of exogenous BDNF rescued the inhibitory synaptic deficits caused by the reduced dysbindin-1 level in both cultured cortical neurons and slice cultures. CONCLUSIONS Taken together, our results demonstrate that these two genes linked to risk for schizophrenia (BDNF and dysbindin-1) function together to regulate interneuron development and cortical network activity. This evidence supports the investigation of the association between dysbindin-1 and BDNF in humans with schizophrenia.

Learning utterance-level normalisation using Variational Autoencoders for robust automatic speech recognition

This paper presents a Variational Autoencoder (VAE) based framework for modelling utterances. In this model, a mapping from an utterance to a distribution over the latent space, the VAE-utterance feature, is defined. This is in addition to a frame-level mapping, the VAE-frame feature. Using the Aurora-4 dataset, we train and perform some analysis on these models based on their detection of speaker and utterance variability, and also use combinations of LDA, i-vector, and VAE-frame and utterance features for speech recognition training. We find that it works equally well using VAE-frame + VAE-utterance features alone, and by using an LDA + VAE-frame +VAE-utterance feature combination, we obtain a word-errorrate (WER) of 9.59%, a gain over the 9.72% baseline which uses an LDA + i-vector combination.

The Pathogenesis of Saffold Virus in AG129 Mice and the Effects of Its Truncated L Protein in the Central Nervous System.

Saffold Virus (SAFV) is a human cardiovirus that has been suggested to cause severe infection of the central nervous system (CNS). Compared to a similar virus, Theiler’s murine encephalomyelitis virus (TMEV), SAFV has a truncated Leader (L) protein, a protein essential in the establishment of persistent CNS infections. In this study, we generated a chimeric SAFV by replacing the L protein of SAFV with that of TMEV. We then compared the replication in cell cultures and pathogenesis in a mouse model. We showed that both SAFV and chimeric SAFV are able to infect Vero and Neuro2a cells well, but only chimeric SAFV was able to infect RAW264.7. We then showed that mice lacking IFN-α/β and IFN-γ receptors provide a good animal model for SAFV infection, and further identified the locality of the infection to the ventral horn of the spine and several locations in the brain. Lastly, we showed that neither SAFV nor chimeric SAFV causes persistence in this model. Overall, our results provide a strong basis on which the mechanisms underlying Saffold virus induced neuropathogenesis can be further studied and, hence, facilitating new information about its pathogenesis.

Spatial Learning Requires mGlu5 Signalling in the Dorsal Hippocampus.

We examined the role of hippocampal metabotropic glutamate receptor 5 (mGlu5) in spatial learning and memory. Although it has been shown that mGlu5 signalling is required for certain forms of learning and memory, its role in spatial learning is unclear since studies using pharmacological or knockout mice models provide inconsistent findings. Additionally, the location in the brain where mGlu5 signalling may modulate such learning is yet to be precisely delineated. We stereotaxically injected rAAV-Cre into the dorsal hippocampus of mGlu5loxP/loxP mice to knockdown mGlu5 in that region. We show for the first time that knockdown of mGlu5 in the dorsal hippocampus is sufficient to impair spatial learning in Morris Water Maze. Locomotor activity and memory retrieval were unaffected by the mGlu5 knockdown. Taken together, these findings support a key role for dorsal hippocampal mGlu5 signalling in spatial learning.

Saffold virus, an emerging human cardiovirus.

Saffold virus (SAFV) is an emerging human cardiovirus that has been shown to be ubiquitous. Initial studies of SAFV focused on respiratory and gastrointestinal infection; however, it has also recently been associated with diverse clinical symptoms including the endocrine, cardiovascular, and neurological systems. Given the systemic nature of SAFV, and its high prevalence, understanding its pathogenicity and clinical impact is of utmost importance. This comprehensive review highlights and discusses recent developments in epidemiology, human pathogenicity, animal, and molecular studies related to SAFV. It also provides detailed insights into the neuropathogenicity of SAFV. We argue that human studies have been confounded by coinfections and therefore require support from robust molecular and animal research. Thereby, we aim to provide foresight into further research to better understand this emerging virus.

Tetratricopeptide repeat domain 9A modulates anxiety-like behavior in female mice

Tetratricopeptide repeat domain 9A (TTC9A) expression is abundantly expressed in the brain. Previous studies in TTC9A knockout (TTC9A−/−) mice have indicated that TTC9A negatively regulates the action of estrogen. In this study we investigated the role of TTC9A on anxiety-like behavior through its functional interaction with estrogen using the TTC9A−/− mice model. A battery of tests on anxiety-related behaviors was conducted. Our results demonstrated that TTC9A−/− mice exhibited an increase in anxiety-like behaviors compared to the wild type TTC9A+/+ mice. This difference was abolished after ovariectomy, and administration of 17-β-estradiol benzoate (EB) restored this escalated anxiety-like behavior in TTC9A−/− mice. Since serotonin is well-known to be the key neuromodulator involved in anxiety behaviors, the mRNA levels of tryptophan hydroxylase (TPH) 1, TPH2 (both are involved in serotonin synthesis), and serotonin transporter (5-HTT) were measured in the ventromedial prefrontal cortex (vmPFC) and dorsal raphe nucleus (DRN). Interestingly, the heightened anxiety in TTC9A−/− mice under EB influence is consistent with a greater induction of TPH 2, and 5-HTT by EB in DRN that play key roles in emotion regulation. In conclusion, our data indicate that TTC9A modulates the anxiety-related behaviors through modulation of estrogen action on the serotonergic system in the DRN.

Maternal Separation Does Not Produce a Significant Behavioral Change in Mice

Early life adversities together with genetic predispositions have been associated with elevated risks of neuropsychiatric disorders during later life. In order to investigate the underlying mechanisms, many chronic, early-life stress paradigms in multiple animal models have been developed. Previously, studies reported that maternal separation (MS) in the early postnatal stages triggers depression-and/or anxiety-like behaviors in rats. However, similar studies using mice have reported inconsistent behavioral outcomes. In this study, we sought to assess behavioral outcomes from two different early-life stress paradigms; a conventional 3-hour MS and a maternal separation with early weaning (MSEW) paradigm using C57BL/6J male mice with independent cohorts. Our data demonstrated that both MS and MSEW paradigms did not produce reported behavioral anomalies. Therefore, MS paradigms in mice require further validation and modification.