University of Connecticut

My research career began in the laboratory of Joe LoTurco at the University of Connecticut where I studied somatotopic map development in the barrel system of the citron-kinase (citron-K) null mutant rat, flathead. The flathead rat has severe micrencephaly resulting from substantial deficits in embryonic neurogenesis that are due to failed cytokinesis and cell death. I found that the flathead exhibited scaled reductions in map size proportional to its reduced cortical volume, but displayed enhanced somatotopic plasticity in response to vibrissae removal [1].

My thesis work was titled “Neurogenic potential and microglial-neuronal fusion in postnatal cerebral cortex” First, I worked on a collaborative project with Ottorino Belluzzi's group at the University of Ferrara, Italy where we looked at the fate and function of newly generated neuroblasts in the postnatal olfactory bulb (OB). Combining retroviral-mediated labeling of newly generated neurons together with patch-clamp electrophysiology and immunocytochemical characterization, we demonstrated that soon after new cells enter the layers of the olfactory bulb, they express functional GABA and glutamate receptor channels, respond synaptically to stimulation of the olfactory nerve, and establish both axodendritic and dendrodendritic synaptic contacts within the olfactory bulb. These experiments provided a basic description of the physiology of newly generated cells in the olfactory bulb and showed that such new cells are functional neurons that synaptically integrate into olfactory bulb circuitry soon after their arrival [2, 3]. This work was published in the Journal of Neuroscience and has become a highly cited part of the adult neurogenesis literature.

Figure 1. New neurons in the glomerular layer of the postnatal olfactory bulb (J. Ackman, from work with Belluzzi et al., 2003). Newborn periglomerular interneurons labeled with GFP retrovirus (green). Calbindin positive interneurons and neuron specific beta-tubulin labeled in blue and red respectively.

Next, I embarked on a collaborative effort with other members of Joe's lab to create a rat model for a human cortical malformation disorder, double-cortex syndrome, by knocking down expression of doublecortin (DCX RNAi) via in utero electroporation. My contribution was to perform much of the confocal microscopy and image analysis for the work, including the cover image for Nature Neuroscience, vol 6, no. 12. We found that DCX is required for radial migration of neurons during cortical development and that DCX RNAi causes both cell-autonomous and non-cell autonomous disruptions in neuronal migration. Neurons that prematurely stopped migrating formed a subcortical band heterotopia within the white matter, analogous to the malformation seen in humans with DCX mutations. Our study resulted in a highly cited work [4] that highlighted the advantages of spatiotemporal control of genetic expression when assessing gene function. Furthermore, this work established in utero electroporation and shRNA plasmid delivery as premier tools that are now widely employed in neuroscience laboratories around the world.

Figure 2. Neurons migrating radially into the cortical plate, 4 days after in utero electroporation with EGFP and doublecortin shRNA plasmids (J. Ackman, from Bai et al., 2003).

I next continued with the second part of my dissertation work where I investigated postnatal neurogenesis in the flathead mutant rat. I found that the loss of citron-K function in flathead had a severe effect on postnatal neurogenesis in the dentate gyrus. Analysis of postnatal neurogenic regions of the flathead mutant revealed a complete lack of both mitotic cells and neuroblasts in the dentate gyrus and a large reduction in the number of dividing cells in the subventricular zone during the second postnatal week. In addition, the postnatal flathead dentate gyrus also lacked the glial scaffold that defines the neurogenic niche in postnatal subgranular zone. This work indicated that postnatal neurogenesis in the dentate gyrus is eliminated by loss of citron-K function, and suggested that a citron-K dependent progenitor lineage forms the postnatal neuronal progenitor population in the dentate gyrus [5].

Figure 3. Development of the dentate gyrus. Granule neuron progenitors for the internal blade of the dentate (DGi) migrate from the secondary dentate germinal zone (dgs). The lineage giving rise to these cells are abolished in citron-K-/- animals. From Ackman et al., 2007.

For the final part of my dissertation I analyzed the neurogenic potential of the postnatal rat neocortex. Since the evidence for adult generated neocortical neurons has remained controversial, I performed a systematic test for postnatal neocortical neurogenesis using a replication-incompetent retrovirus that labels mitotic cells. I found that after injections of eGFP retrovirus into postnatal rat cerebral cortex, a small fraction of GFP labeled cells were neocortical pyramidal neurons. However, these GFP labeled cortical pyramidal neurons, unlike GFP labeled astrocytes, oligodendrocytes, and microglia, did not become labeled with a secondary method to label newly generated cells using BrdU incorporation. Closer inspection of retrovirally labeled cells revealed microglia fused to the apical dendrites of labeled pyramidal neurons. Retroviral infection of mixed cultures of cortical neurons and microglia confirmed the presence of neuronal-microglial fusions. Furthermore, activation of the innate immune response in microglia greatly increased the fusion of microglia to neurons in cultures. This work indicated a novel form of cell fusion between neuronal dendrites and microglia, and illustrated the need for additional caution when interpreting evidence for neurogenesis in postnatal neocortex [6, 7] in stem cell studies of regeneration and repair. Furthermore this work suggested that a unique interaction exists between microglia and neuronal arborizations during the period of development when the rate of synaptogenesis is at its peak. This study resulted in a highly regarded Journal of Neuroscience article that is 'Recommended' on Faculty of 1000 and which is now a fundamental part of the adult neurogenesis, stem cell, and repair literature.

Figure 4. Labeled pyramidal neurons several days after fusion with microglia infected with GFP retrovirus in postnatal rat neocortex. Right panel, labeled pyramidal neurons in mixed primary cultures via fusion with microglia. GFP, green; neuronal MAP-2, red; nuclear DAPI, blue. From Ackman et al., 2006.

INMED, Marseille, France

Next I pursued a short (1.5 year) postdoc fellowship at the Mediterranean Institute of Neurobiology (INMED/INSERM U901) in Marseille, France formerly directed by Yesekiel Ben-Ari. I had previously interacted with several investigators from INMED when they visited Joe's lab to learn in utero electroporation techniques. I became aware of short-term fellowships available from INSERM for foreign postdocs to work in France where I wanted to train in multineuronal two-photon calcium imaging methods that Rosa Cossart was developing at INMED and apply these techniques to epilepsy models with Alfonso Represa's group. First, I collaborated with members of Rosa's group where we characterized the dynamics of spontaneous network activity in developing neocortical circuits, resulting in a publication in Journal of Neuroscience [8].

My primary project in Marseille consisted of studying the role of neuronal migration disorders in epilepsy with Alfonso Represa's group. I applied multineuron calcium imaging and patch-clamp recordings to the DCX RNAi in utero electroporation model for double cortex syndrome from previous work in the LoTurco lab, where we analyzed the network activity in rat neocortical slices containing GFP+ dysplasias. We found increased excitability in the 'normotopic' neocortical networks overlying white matter heterotopias, a retention of immature activity patterns in heterotopic neurons, coherent population activity within normotopic and heterotopic networks, and synaptic integration of the heterotopia and overlying neocortex.

Figure 5. Network physiology of normal and dysplastic cortical circuits. a, Rodent model for human double cortex syndrome, an intractable childhood epilepsy, using RNAi knockdown of doublecortin expression. b, Heterotopic cortical dysplasia in white matter at P7. c, Whole cell current clamp recording and biocytin reconstructions from heterotopic neurons. d, Raster plot of calcium event spike times during a 3 min two-photon calcium imaging recording from 100s of neurons in a cortical dysplasia. From Ackman et al., 2009.

This work, which was published in the Journal of Neuroscience, provided the first detailed physiological analysis of a targeted genetic model for a human cortical malformation disorder and suggested that aberrant activity in neocortex coupled with delayed maturation of heterotopic networks may underlie neurological deficits seen in patients with double cortex syndrome [9].

Yale University

After working at INSERM in Marseille, I pursued a second postdoc in the U.S. where I wanted to train more extensively with in vivo neurophysiological and calcium imaging approaches and examine activity-dependent circuit development in the visual system. At the time, Michael Crair had recently moved to Yale and had just begun setting up a laser and microscope for multiphoton imaging in studies of activity-dependent sensory map development. I joined Mike's lab and first worked to setup the experimental system for successful electrophysiological and optical recordings in live mice. I established a flexible system for two-photon and widefield CCD calcium imaging with extracellular recordings in neonatal and adult mice as well as programmed a considerable amount of image processing and data analysis software in Matlab, R, Python, and ImageJ to analyse spontaneous and stimulus evoked activity patterns in the datasets.

In Mike's lab, I established methods for examining the spontaneous and evoked activity of large groups of neurons (100’s) at single-cell resolution in normal and mutant mouse lines so we can study the development of topographic maps in the visual system. I developed techniques for functional calcium imaging from pre-synaptic axon terminals in the visual system in vivo, multineuron calcium imaging in the superior colliculus, simultaneous optogenetic stimulation and multineuron calcium imaging, as well as simultaneous population level imaging of the midbrain and neocortex using traditional calcium sensitive dyes and mice expressing genetic calcium indicators. By performing calcium imaging and multiunit recordings in the superior colliculus together with channelrhodopsin-2 stimulation of the retina in neonatal mice in vivo, I made key contributions to work from Mike's lab published in Nature Neuroscience which demonstated a role for precise temporal patterns of retinal activity in visual map development before the onset of vision [10].

Figure 6. Stimulation of the developing mouse visual system in vivo using chR2. a, Bright field image of bilateral craniotomy over the superior colliculus. b, Raw calcium signals for individual ROIs in each collicular hemisphere bulk loaded with the calcium indicator OGB1AM in response to synchronous optogenetic stimulation of both eyes. Light stimuli (1 s long) occurred at the times indicated by blue shading. c, Raster plots for responses from all ROIs and the fraction of active ROIs to four consecutive optogenetic stimulations of both eyes.

My primary project in Mike's lab involved examining the properties and role of spontaneous network activity in the development of the visual system. It's been known since the early 1990s that the isolated retina in vitro, can display patterns of spontaneous activity ('retinal waves') during development that are thought to be instrumental in the development and refinement of visual system connectivity. However it had remained unknown whether retinal waves actually occur in vivo and whether circuits throughout the visual system exhibit similar patterns of activity.

My research involved the first direct optical recordings of the developmental activity pattern known as 'retinal waves' in live, unanesthetized mice via presynaptic calcium imaging from retinal ganglion cell terminals using anterograde or genetically encoded calcium indicators. This work was published in Nature as an Article (Ackman et al., Nature 2012) [11]. The work not only demonstrated that retinal waves do exist in vivo, but also through postsynaptic calcium imaging showed that waves propagate to and activate neurons throughout developing midbrain and visual cortex. Surprisingly, retinal waves have unique properties (directional, nucleation site preference in binocular retina, and bilateral symmetry) that could not have been anticipated from previous in vitro recordings. Furthermore, retinal waves play a differential role in activating primary versus extrastriate visual cortical areas.