|dc.description.abstracteng||Ears achieve their exquisite sensitivity using positive mechanical feedback (Ashmore et al., 2010). Like pushing a swing augments its swing, sound-induced vibrations are enhanced by motile responses of auditory receptor cells on a cycle-by-cycle basis (Manley, 2001; Göpfert et al., 2005; Ashmore et al., 2010). This positive mechanical feedback boosts auditory sensitivity and is prone to feedback oscillations. The gain of this amplification needs to be controlled critically to adjust the sensitivity of hearing. Low amplification will hamper sensitive hearing, whereas excessive amplification can lead to large, self-sustained feedback oscillations that can be measured acoustically as spontaneous otoacoustic emissions, i.e. sounds emitted by the ear (Ashmore et al., 2010). The outer hair cells in the mammalian inner ear mediate the amplification through a voltage-dependent motor molecule, prestin (Fettiplace, 2006; Geurten et al., 2013). Four features of the active amplification are frequency-specific amplification, self-sustained oscillations (spontaneous otoacoustic emission), power gain, and compressive nonlinearity (Hudspeth, 2008). Amplification has been reported also in the hearing organ of non-mammalian vertebrates and insects (Manley, 2001; Nadrowski et al., 2011). The amplification in the Drosophila hearing organ is the focus of this thesis.
Stabilizing the gain of mechanical amplification is crucial for hearing, yet how this stabilization is achieved is little understood. In Drosophila, Nanchung (Nan)-Inactive (Iav) transient receptor potential vanilloid (TRPV) channels have been identified to negatively regulate amplification, whereby the loss of Nan-Iav leads to hyper-amplification, with the antenna displaying large, self-sustained oscillations in the absence of sound stimuli (Göpfert et al., 2006). Excess amplification also ensues from mutations in the flies’ single calmodulin gene (Senthilan et al., 2012), implicating TRPV channels and calcium in the endogenous regulation of the amplification gain. How TRPV channel activity and calcium levels are controlled, however, has remained mysterious. In the course of my work, I had a deeper look into this control, unraveling a metabolic feedback that regulates amplification (part 1) and identifying a novel molecular player in the regulation of the amplification gain (part 2).
Part I: Enzymatic control of mechanical amplification in fly hearing
The Drosophila auditory neurons are ciliated with Nan and Iav both localizing to the proximal cilium region in an interdependent manner, assembling into Nan-Iav heteromers (Gong et al., 2004). Hints that the open probability of the TRPV channel might be important comes from the action of pymetrozine, a synthetic insecticide whose molecular targets are Nan-Iav TRPVs (Nesterov et al., 2015). Pymetrozine activates Nan-Iav channels, leading to ciliary calcium influx and reducing the mechanical amplification gain (Nesterov et al., 2015). Apparently, activating Nan-Iav reduces the gain of mechanical amplification, but what controls internally the activity of Nan-Iav?
Nicotinamidase is a component of the nicotinamide adenine dinucleotide (NAD+) salvage pathway that generates NAD+ from niacin equivalents, such as nicotinamide. Nicotinamidase converts nicotinamide into nicotinic acid, whereby nicotinamide is an agonist of Nan-Iav TRPV channels that, at least in vitro, activates Nan-Iav (Upadhyay et al., 2016). The application of nicotinamide on the Drosophila larvae elicits calcium signals in the chordotonal neurons (Upadhyay et al., 2016), much like pymetrozine (Nesterov et al., 2015).
My work documented that, unlike pymetrozine, nicotinamide is an endogenous activator of Nan-Iav channels, linking Nan-Iav activity to metabolism. I identified the fly’s nicotinamidase (NAAM), as a central player in the regulation of mechanical amplification in fly hearing. By analyzing a null mutation that disrupts Naam expression, I found that loss of NAAM not only abolishes sound-evoked electrical Johnston’s organ responses and mechanical amplification but also affects TRPV channels expression and localization.
Judged from my results, loss of Naam leads to the accumulation of nicotinamide and, thus, excessive Nan-Iav opening, thereby reducing the gain of mechanical amplification in fly hearing. Consistent with such a scenario, supplementing the fly food of wild-type flies with nicotinamide also reduced the amplification gain, whereas feeding Naam mutant flies with NAD+ or nicotinic acid did not rescue mechanical amplification. Hence, rather than the lack of the NAAM product, it seems to be the accumulation of the NAAM substrate, nicotinamide, that reduces the amplification gain. Apparently, NAAM links mechanical amplification in hearing to metabolism, regulating the amplification gain by modulating nicotinamide levels and, thus, Nan-Iav activity.
Drosophila NAAM is particular in that it bears aminoterminal EF-hand domains, besides the isochorismatase-like domain. The evolutionary conserved EF-hand domains might regulate the enzyme activity by binding to Ca2+. In the course of this thesis, I found that the EF-hand domains have a direct effect on the NAAM enzymatic activity. NAAM with missense point mutation or deletion in the EF-hand domains did not rescue the hearing defect in the Naam mutant flies, and the same was observed in in vitro assays of the enzymatic reaction. Manipulating the EF-hand domains of Naam affected NAAM localization, leaving the question open, how exactly the EF-hand domains contribute to the NAAM function.
Part II: Drosophila GAP43 like (igl) in auditory neuron cilia
(IGL, Invertebrate GAP43 like), the fly ortholog of the vertebrate key “growth” and “plasticity” protein GAP43, is reportedly abundant in neurons (Neel and Young, 1994). I found that igl is expressed in auditory receptor neurons, with IGL protein localizing to their cilia. In effect, sequence analysis identified the binding motif of the ciliary transcription factor RFX in the igl promoter/enhancer region, putting IGL forward as a novel cilium compartment protein. Consistent with this notion, I could show that the expression of IGL protein is RFX-dependent, with the loss of RFX abolishing ciliary localization of the IGL. Moreover, I discovered an RFX DNA binding motif in human GAP43, adding a new twist to GAP43 regulation and suggesting that igl might be a conserved cilium gene. Precedence for a GAP43 cilium connection comes from newborn rat olfactory receptor neurons, whose cilia are strongly stained by antibodies against GAP43 (Verhaagen et al., 1989).
Notwithstanding these intriguing results, there was not a proper available mutant for this gene. The non-in-frame GFP cassette insertion in iglMI02290 flies leads to the reduction in the mechanical amplification as well as the power of the antenna’s mechanical free fluctuations. Hence, in contrast to TRPVs and calmodulin, Drosophila IGL seems to positively control the amplification gain in the fly’s auditory system. This seems intriguing given that the igl sequence comprises IQ motifs and reportedly binds calmodulin (Neel and Young, 1994).||de