Latency of phototransduction limits transfer of higher-frequency signals in cockroach photoreceptors

Abstract

Visual transduction in rhabdomeric photoreceptors is compartmentalized and discretized. Signals of individual microvilli, the quantum bumps, are electrotonically summed, producing a graded response. Intrinsic dispersion of quantum bump latencies is thought to introduce noise and degrade signal transfer. Here, we found profound differences in the information rate and signaling bandwidth between in vitro patch-clamp and in vivo intracellular recordings of Periplaneta americana photoreceptors and traced them to the properties of quantum bumps and membrane resistance. Comparison of macroscopic and elementary light responses revealed differences in quantum bump summation and membrane resistance in vivo versus in vitro. Modeling of voltage bumps suggested that current bumps in vivo should be much bigger and faster than those actually recorded in vitro. Importantly, the group-average latency of dark-adapted photoreceptors was 30 ± 8 ms in intracellular ( n = 34) versus 70 ± 19 ms in patch-clamp ( n = 57) recordings. Duration of composite responses increased with mean latency because bump dispersion depended on mean latency. In vivo, latency dispersion broadened the composite response by 25% on average and slowed its onset. However, in the majority of photoreceptors, the characteristic durations of multiphoton impulse responses to 1-ms stimuli did not exceed the durations of mean voltage bumps. Consistently, we found strong associations between the latency and onset kinetics of the macroscopic response, on the one hand and the higher-frequency signal gain and information rate of the photoreceptor, on the other hand, indicating a direct connection between quantum bump latency and its dispersion and the signaling bandwidth. NEW & NOTEWORTHY When stimulated by light, microvilli of rhabdomeric photoreceptors produce discrete signals characterized by variable latencies. We show that this intrinsic latency dispersion restricts signaling bandwidth and information rate of photoreceptors in Periplaneta americana. Profound differences are found between the properties of photoreceptor responses in vivo and in vitro.

Suppression of Gq and PLC gene expression has a small effect on quantum bumps in vivo in Periplaneta americana

“…The modifications to the original model are described in "Methods" and Table 2. This model allowed us to generate a mean current bump with kinetics similar to those of the previously inferred (from simulations) P. americana current bump (Ignatova et al 2019), and with the mean latency matching our in vivo group-average control results (Fig. 5a).…”

“…While the latency itself may not be important for signal transfer, unless it is so prolonged that it limits the reaction speed, the intrinsic latency dispersion, with its standard deviation being a linear function of mean photoreceptor latency ( Figs. 2e, 6j), is important (Ignatova et al 2019). The noise arising from the variability in the timing of quantum bump latency is one of the three components of phototransduction noise in microvillar photoreceptors (the other two are due to the variabilities in quantum bump amplitude and duration).…”

“…6), 3 and 4 µm 2 s −1 for most of the analysis (including data in Fig. 5; "intermediate (Ignatova et al 2019) and D. melanogaster (grey) (Nikolic et al 2010) are superimposed with a current bump derived by fitting from the group-average voltage bump in vivo (black, see description in (Ignatova et al 2019)), and with a bump obtained using our model (blue); bumps were positioned so that their peaks coincide; the cockroach current bump amplitudes (experimental bumps were recorded at a holding potential of -82 mV) were adjusted for proper comparison with the average D. melanogaster bump (recorded at a holding potential of -70 mV) using a reversal potential value of + 10 mV. b Dependencies of mean bump latencies on concentration of Gqα and PLC proteins in experiments and simulations.…”

“…Our previous analysis suggested that the kinetics of the cockroach current bump in vivo should be much faster than the current bumps recorded in vitro and faster than the current bump in D. melanogaster (Fig. 5a) (Ignatova et al 2019;Nikolic et al 2010). Therefore, we altered the model to accelerate the bump kinetics.…”

Vision does not impact walking performance in Argentine ants

“…Given the relatively slow transduction speeds of walking insect 56 photoreceptors (Frolov et al, 2017), fast movements of the limbs during walking may preclude 57 visual feedback (Full and Koditschek, 1999). Nocturnal and subterranean species particularly 58 confront feedback constraints since transduction speeds depend on both light levels and 59 temperature (Frolov and Ignatova, 2020;Heimonen et al, 2012;Warrant, 2017;Warzecha and 60 Egelhaaf, 2000). These data suggest that vision likely does not influence intra-stride walking 61 coordination but may inform obstacle avoidance at intermediate distances.…”