We are at the beginning of an exciting new era for neuroscience, as our ability to probe neural circuits and their neuronal components is advancing rapidly due to genetic and optogenetic tools. Our research program applies these tools to address fundamental questions about how the same neural circuitry generates different motor patterns, and how such circuits develop and are maintained. We investigate these questions using the Drosophila larva, which has the following advantages:(i) The connectome of the larval motor circuit is near completion, enabling us to identify, at the single-synapse level, the pre and postsynaptic partners of each individual neuron embedded in it. This anatomical map has provided an excellent substrate to study the development, maintenance, and function of larval motor circuits as well as the cell biology of individual neurons embedded within it. (ii) The larval CNS generates multiple motor behaviors that can be studied at the single neuron/single muscle level. Moreover, using the modern optogenetic methods, it is possible to access individual neurons, monitor or alter their activity, and observe the behavioral consequences. (iii) It is also feasible to selectively inactivate or induce ectopic expression of any gene (e.g. those coding for transcription factors) in the neuron of interest, and examine its effect on intrinsic neural properties, morphology, connectivity pattern, and behavioral performance of the animal, thereby linking the gene to development and behavior.
Development and maintenance of sensory-motor circuits
Each neuron embedded within a circuit has its unique set of properties that are necessary for its proper function, such as neurotransmitter receptors, neurotransmitter synthesizing enzymes, ion channels, axonal/dendritic guidance molecules, and synaptic partners. In different animal models, terminal selector transcription factors (TTFs) have been shown to induce these signatures. For example, we previously discovered the mechanisms by which the evolutionarily conserved transcription factors (Even-skipped, Grain, and Zn finger homeodomain 1) regulate the expression of cell surface axon guidance molecules, thereby direct motor axons to their target muscles during embryogenesis (before larval hatching).
and to read the relevant papers.
To date, however, little is known about how neuronal and circuit properties are maintained throughout the Drosophila larval life span. Functionally and anatomically characterized larval motor circuits provide an excellent substrate to study the mechanisms of circuit formation and maintenance. Therefore, yet another main aim of the lab is to study the role of TTFs in development (embryogenesis) and maintenance (larval life) of sensory-motor circuits. We currently have two models for how these TTFs may function. To test these models, we will apply Drosophila classic genetics combined with modern tools such as single-cell genome engineering (CRISPR), RNA-seq, and high-resolution multiphoton imaging.
​
Welcome to the Zarin Laboratory
A cross-sectional view of the larval Central Nervous System (CNS) taken by transmission electron microscopy (TEM) (Ohyama et al 2015)
Larval Nervous System. The axons of motor neurons (MNs) exit the CNS and innervate body wall muscles required for locomotion.
Command circuits for switching between antagonistic locomotor programs.
Command-like descending neurons can induce many behaviors, such as backward locomotion, escape, feeding, courtship, egg-laying, or grooming (we define ‘command-like neuron’ as a neuron whose activation elicits or ‘commands’ a specific behavior). In most animals, it remains unknown how neural circuits switch between antagonistic behaviors: via top-down activation/ inhibition of antagonistic circuits or via reciprocal inhibition between antagonistic circuits.
​
We recently discovered a top-down excitation-inhibition command circuit, in which Drosophila larval ‘mooncrawler descending neurons’ (MDNs) coordinately induce backward locomotion and block forward locomotion: the former by stimulating a backward-active premotor neuron (A18b), and the latter by disynaptic inhibition of a forward-specific premotor neuron (A27h). (Panel A, B). Interestingly, larval MDNs persist into adulthood, where they can trigger backward walking (Panel C). Thus, MDNs induce backward locomotion in both limbless and limbed animals.
​
What we have discovered to date is just the tip of the iceberg, as there are tens of ascending and descending command neurons to be identified Drosophila larva.
Therefore, another main aim of the lab is to further study these command circuits in Drosophila larva, and determine how animals choose one behavior to the exclusion of others. To address this aim we use genetic screens, intersectional genetics, circuit reconstruction by electron microscopy, and functional optogenetics
​
Sensory-motor circuits underlying distinct muscle activation patterns during locomotor behaviors
B
A
Animals generate diverse motor behaviors, yet how the same motor neurons (MNs) generate distinct behaviors remains an open question. Our aim is to fully characterize neural circuits generating Drosophila forward and backward locomotion. We have recently shown that all body wall MNs (hence their downstream muscles) are activated during both behaviors, but a subset of MNs/Muscles change recruitment timing for each behavior (see movies A, B on the right).
To explore how these different MN phase relationships arise, we use a serial section TEM volume to reconstruct a comprehensive larval connectome. This connectome is composed of multiple layers of neurons, including MNs, premotor neurons (PMNs), interneurons (INs), and proprioceptive sensory neurons.
One of main research aims of the lab is to understand how this comprehensive connectome functions to generate the MN/Muscle activity patterns seen during each behavior. To address this aim, we apply different approaches:
​
1) Optogenetic loss of function (silencing) and gain of function (activation) of individual neurons combined with quantitative behavioral analysis of freely crawling larva.
​
2) Calcium imaging in individual neurons and muscles of intact behaving animals using high-resolution confocal microscopy.
​
3) Theoretical modeling of larval locomotion: In collaboration with a Theoretical Neuroscience Lab at Columbia University (Litwin-Kumar lab), we integrate the experimental and connectomic data described above to build a computational model capable of producing forward and backward locomotor patterns.
​
Motor neurons (A), pre-motor neurons (B), and MDN command descending neurons (C) reconstructed using electron microscopy
(Zarin* and Mark* et al, 2019; Carreira-Rosario*, Zarin* and Clark* et al, 2018).
(A) Schematic depiction of larval CNS and body wall muscles. The larva is composed of three thoracic segments (T1-T3)
and nine abdominal segments (A1-A9). (B) Larval body wall muscles stained with GFP. (C) Cartoon showing body wall muscles
in a half-segment (hemisegment). There are 58-60 muscles with different shapes and orientations in each of the A1-A7 segments
(29-30 per hemisegment) (Zarin* and Mark* et al, 2019).
A
B
C
(A) Mooncrawler Descending Neurons (MDNs) directly activate backward specific premotor neuron (A18b) and indirectly inhibits forward specific premotor neuron (A27h) via pair1. Optogenetic activation of MDNs in the larva (B) and adult fly (C) induces backward crawling and backward walking respectively (Carreira-Rosario*, Zarin* and Clark* et al, 2018).
Characterizing the activity pattern of individual body wall muscles during
forward (A) and backward crawling (B) of intact larva using muscle-calcium imaging (Zarin* and Mark* et al, 2019).
C
A
A
C
B
A
B
C
