Current experimental stroke research faces the same challenge as neuroscience: to transform correlative findings in causative ones. Research of recent years has shown the tremendous potential of the central nervous system to react to noxious stimuli such as a stroke: Increased plastic changes leading to reorganization in form of neuronal rewiring, neurogenesis, and synaptogenesis, accompanied by transcriptional and translational turnover in the affected cells, have been described both clinically and in experimental stroke research. However, only minor attempts have been made to connect distinct plastic remodeling processes as causative features for specific behavioral phenotypes. Here, we review current state-of the art techniques for the examination of cortical reorganization and for the manipulation of neuronal circuits as well as techniques which combine anatomical changes with molecular profiling. We provide the principles of the techniques together with studies in experimental stroke research which have already applied the described methodology. The tools discussed are useful to close the loop from our understanding of stroke pathology to the behavioral outcome and may allow discovering new targets for therapeutic approaches. The here presented methods open up new possibilities to assess the efficiency of rehabilitative strategies by understanding their external influence for intrinsic repair mechanisms on a neurobiological basis.
Although huge efforts have been made in recent years, both by clinicians and basic researchers, we have still gained limited insights into a neurological disease such as stroke preventing us from developing specific cures and resulting in poor statistical numbers: Of 15 million people suffering from an ischemic brain attack every year, a third dies, a third remains permanently disabled, and a third recovers as the stroke itself has not been too devastating. On the clinical side, stroke units have been created, which combine experts in intensive care medicine, neurology, physiotherapy, and speech therapy, to accelerate and coordinate the diagnostic and therapeutical processes aiming at improving recovery rates for patients. According to the neurologists’ saying “time is brain,” even mobile units have been established to bring the hospital to the patient [
On the side of basic research, we study the neurobiology of stroke, but seem to be stuck in a “black box” situation: We have accumulated data showing the tremendous capacities of the brain to reorganize by synaptogenesis and even neurogenesis and by neuronal circuit rewiring and new circuit formation. We find cortical map shifts and hyperactive brain regions after stroke; we detect genetic and proteomic turnover within a distinct spatiotemporal profile and sequence of events [
In recent years, new technology has been designed which is starting to fill the gap between correlative and causative research.
The aim of this review is to first discuss current state-of-the-art technology of experimental stroke research which enables a deeper understanding of neuronal reorganization and circuit formation. In the second part, techniques for distinct neuronal circuit manipulation are introduced which help to reveal causal relationships between anatomy and behavior. Finally, new approaches combining cytology with molecular profiles are provided which elucidate the underlying molecular mechanisms of neuronal rewiring and repair. The principles of the techniques are explained together with exemplary studies in experimental stroke research which have already applied the described methodology. The here discussed tools may not only enhance our understanding of stroke pathology but also help to identify crucial anchor points for new therapeutic interventions.
A classical approach to study reorganizational processes across the whole brain is functional resonance imaging (fMRI), allowing the monitoring of neuronal rewiring processes on a macroanatomical level within the same animal. However, although significant contributions to the understanding of the interplay between altered functional status and structural connectivity have been made in stroke models [
In contrast, intrinsic optical imaging sticks out by a high spatial resolution enabling the visualization of small domains within larger brain areas demonstrating the functional organization and the spatial relationships among those smaller domains, for example, in the barrel or visual cortex [
Another technique to study in particular sensory map shifts is millisecond-timescale voltage-sensitive dye (VSD) imaging which unlike functional fMRI and intrinsic optical signal imaging measures electrical activity with relatively high spatial and temporal resolution [
While VSD imaging was the first demonstration of wide-field optical imaging of neural activity [
That wide-field calcium imaging might be also a powerful tool to study brain-wide cortical reorganization on a high-resolution level over time in animals monitoring their rehabilitative courses or recording rehabilitative training after stroke was demonstrated by Murphy et al. [
Intracortical microstimulation (ICMS) and surface stimulation with electrode arrays have been used for many years to map cortical regions, to study cortical reorganization, and to find first hints if projections in the motor cortex are functionally relevant. This technique applies electrical stimulation of cortical sites to induce, for example, stimulus-evoked movement responses, which can be detected visually, by EMG responses or by the usage of accelerometers. Several studies have used this technique to either examine cortical map shifts in sensorimotor areas after spontaneous recovery or different therapeutical applications [
A new noninvasive strategy to study the reorganization of the motor cortex after stroke in the same animal over time is light-based motor mapping: This technique makes usage of the possibility to stimulate neurons by light, either by uncaging neurotransmitters [
However, although all described mapping approaches are powerful tools, they can only provide information about map shifts and representation of general movement dynamics. They stay far beyond cellular resolution and do not allow studying local neuronal circuitry or single neuron contribution to neuronal networks. In particular after stroke, it is not clear how activity in single neurons changes in relation to cortical map shifts. The analysis of single neurons in relation to the neuronal circuit in which they are embedded may elucidate whether stroke-induced plasticity is a result of the capacity of surviving neurons to process multiple functional streams. In vivo two-photon calcium imaging is a potent method which not only allows studying activity of a single neuron or ensembles of neurons in a network but also enables cell type and neuronal subtype-specific analysis. Only a few in vivo two-photon calcium imaging studies focusing on neuronal reorganization and circuit rewiring after stroke have been conducted so far. In an acute in vivo calcium imaging experiment during the induction of a transient global ischemia model in mice, Murphy et al. [
Two-photon calcium imaging enables recordings of individual neuronal activity within a neuronal network and allows subtype-specific functional analysis of brain tissue. However, neuronal activity with cellular resolution level can only be examined in small (<1 millimeter) fields of view. Collective dynamics across different brain regions are inaccessible. Recent advances in two-photon microscopy allow the simultaneous imaging of neuronal networks with cellular resolution level in the active animal in multiple areas, which are even not directly connected [
In a 1979 Scientific American article, Nobel laureate Francis Crick stated that the major challenge facing neuroscience was the need to control one type of cell in the brain while leaving others unaltered. In a lecture from 1999, he further confined: “One of the next requirements is to be able to turn the firing of one or more types of neuron on and off in the alert animal in a rapid manner. The ideal signal would be light, probably at an infrared wavelength to allow the light to penetrate far enough. This seems rather farfetched but it is conceivable that molecular biologists could engineer a particular cell type to be sensitive to light in this way” [
Manipulation of neuronal circuits or single neurons has two prerequisites: Manipulation has to be quick and very specific. Over the years, a very diverse set of tools has been developed to manipulate whole brain regions as well as the activity of individual cells and subtypes in the alert behaving animal.
The first constraint for specific manipulation implies a high spatial control allowing the selective modulation of a whole brain area, a distinct anatomical sub region (e.g., layer 5 pyramidal cells in the sensorimotor cortex) or a particular cell type (e.g., a parvalbumin-positive interneuron).
For inhibiting neuronal activity in whole brain regions, agents such as the GABA agonist muscimol have been used [
The Tet system uses at least two viral vectors plus an antibiotic drug which in a sequential way activate each other to induce the transcription and translation of the gene of interest: A tissue-specific promoter initiates the expression of a transcription factor, either the tetracycline transactivator (tTA) or the reverse tetracycline transactivator (rtTA). The tTA or rtTA then becomes the key player for the transcription of a tetracycline response element (TRE) promotor, which in dependence of the presence of tetracycline or doxycycline drives the expression of the gene of interest. Expression of the gene of interest is fully reversible as administration or removal of tetracycline or doxycycline will turn its expression on or off. In a study by Kinoshita et al. [
Similar to the Tet system, the Cre-flox system also requires two transgenes: A tissue-specific promotor regulates the expression of Cre recombinase, a bacteriophage enzyme which recombines DNA at specific recognition sequences called loxP sites. Cre recombinase then excises DNA within two loxP sites (“floxed”). As in most cases, floxed-stop constructs are knocked in by homologous recombination to a gene of interest [
As the second prerequisite for specific neuronal manipulation in addition to a high degree of spatial resolution, temporal resolution and directional modulation of signaling are required for remotely controlling neuronal firing. Temporal resolution implies the precise control when a receptor or pathway is active or inactive and for how long it should be in a specific active status. Temporal resolution can vary from milliseconds (see “opsins” described below) to hours (e.g., designer receptors exclusively activated by designer drugs, DREADDs). Important are also “onset” kinetics (the time between the experimental manipulation and the modulation of the receptor or signaling pathway) and “offset” kinetics (the time between the initiation of the signaling modulation and the termination of the modulation [
For manipulation of neuronal networks for minutes to hours, designer G protein-coupled receptors have been developed. G protein-coupled receptor pathways are involved in a multitude of cellular functions. Unlike opsins, which are functionally silent without excitation in vivo as they are not directly activated by endogenous compounds, G protein-coupled receptors (GCPRs) are constantly modulated by endogenous ligands in vivo or reveal ligand-independent activity [
Since the first development of DREADD receptors, reports of their usage in vivo are now appearing: The pharmacokinetic properties of the DREADD ligands and the particular route of administration (oral administration, subcutaneous, intraperitoneal, or even local stereotaxic infusion) determine how quickly neurons response to experimental manipulation by ligand application. Responses typically emerge 5 to 15 min after systematic application, for example, of CNO and usually last for 2 h—but this time period can be further enlarged upon dose-dependent increase of CNO [
Although manipulation of GPCR signaling pathways by DREADD receptor induction and activation is highly efficient and shows a very specific spatial resolution (depending on the constructs or transgenic mouse lines used), the temporal resolution remains limited to an activation within minutes—due to the slower nature of GPCR signaling—and the necessary ligand delivery to the location of neuronal manipulation. In contrast, high-temporal (milliseconds) and cellular precision within intact mammalian neural tissue for fast, specific excitation or inhibition even within a freely moving animal can only be achieved with optogenetics [
Early approaches to use light to stimulate neuron activity included the selective photostimulation of neurons in
However, although the numerous advantages of optogenetics are evident—such as the highest specificity, the ultrafast millisecond time scale dissection, and basically no adverse effects due to the light (unless the light source is not too strong or applied too long)—optogenetics stays an invasive procedure for many in vivo experiments: As the light source has to be brought close to the neuronal tissue, targeting deep brain areas or diffuse neuronal populations remains challenging. New development of step function or bistable opsins and opsins such as Jaws—an inhibitory opsin, which is activated by light of infrared wavelength [
Only very few studies have applied optogenetics in experimental stroke research so far: Optogenetics was mainly used in the context of light-based motor mapping as described above [
That optogenetics can also be used to demonstrate a causal relationship between a rewiring neuronal circuit and recovery of a specific (sensorimotor) behavior was demonstrated by Wahl et al. [
New advanced technology in microscopy allows the precise optogenetic stimulation of individual neurons [
Although optogenetics have revolutionized the field of neuroscience, the examination of deeper, subcortical brain regions remains a challenge, as the light has to be somehow delivered to the tissue often requiring invasive implantation of fiber optics causing collateral damage of the surrounding brain tissue. A new emerging method which overcomes the spatial limitations is magnetogenetics: It relies on a principle known as thermal relaxation [
However, although individual neuron and specific neuronal circuit manipulation with high spatial and temporal control is possible as discussed above, there still is a need for good behavioral readouts: In particular, in experimental stroke research studying, for example, motor impairment and recovery, it is crucial to quantitatively understand true recovery versus compensation of impaired function [
So far, we have discussed how stroke reorganization can be examined on the macrolevel of map shifts or by studying single neurons in neuronal circuits using 2-photon calcium imaging approaches. We have reviewed how individual neurons and whole neuronal populations can be manipulated with high spatiotemporal resolution disclosing new possibilities of causally linking individual neuronal activity with a distinct behavioral phenotype. However, an understanding of the underlying molecular crosstalk which induces anatomical and behavioral changes is still lacking. Classically, tracing techniques (e.g., dextran tracers) have been applied to visualize cells involved in structural reorganization after stroke [
In addition, new constructs have been recently developed for molecular profiling of projecting neurons and thus bridging the gap between anatomical modifications and underlying molecular mechanisms. Using, for example, bacterial artificial chromosome (BAC) transgenic mice which express EGFP-tagged ribosomal protein L10a in defined cell populations allowed purification of polysomal mRNAs from genetically defined cell populations in the brain [
Here, we have reviewed current and new promising state-of-the-art techniques for studying reorganization after stroke, for the identification and manipulation of distinct neuronal populations and approaches which allow examining molecular profiles of neurons being part of the cortical reorganization process. While for decades studies in basic stroke research have only described and reported correlative findings, these techniques open up tremendous possibilities to analyze plastic processes and identify and target key players for the development of new therapies in stroke.
The author excludes any competing interest.
This work was supported by the Swiss National Science Foundation Grant no. 31003A-149315-1.