WO2024102460A1

WO2024102460A1 - Circuit analysis platforms for drug discovery

Info

Publication number: WO2024102460A1
Application number: PCT/US2023/037125
Authority: WO
Inventors: Ben GUNN; Gary Lynch
Original assignee: University of California Berkeley; University of California San Diego UCSD
Current assignee: University of California Berkeley; University of California San Diego UCSD
Priority date: 2022-11-11
Filing date: 2023-11-10
Publication date: 2024-05-16
Anticipated expiration: 2025-05-11

Abstract

Circuit analysis platforms for drug discovery in accordance with embodiments of the invention are disclosed. In one embodiment, a circuit analysis platform for drug discovery by predicting responses of complex systems to local changes is provided, the circuit analysis platform comprising a brain tissue prepared in vitro, wherein the brain tissue provides a brain circuit comprising a plurality of stages including at least one input stage and an output stage; wherein the brain circuit is configured to receive activations of inputs via the at least one input stage, wherein the activations of inputs comprises a collection of patterns corresponding to at least one behavioral activity; and wherein the brain circuit is configured to output responses via the output stage, wherein the output responses include at least one circuit level operation initiated by the collection of patterns.

Description

CIRCUIT ANALYSIS PLATFORMS FOR DRUG DISCOVERY

CROSS-REFERENCE TO RELATED APPLICATION

[0001] The current application claims priority to U.S. Provisional Patent Application No. 63/424,690 filed on November 11, 2022, the disclosure of which is incorporated herein by reference.

FEDERAL FUNDING SUPPORT

[0002] This invention was made with Government support under Grant No. N00014182114, awarded by the Office of Naval Research. The Government has certain rights in the invention.

FIELD OF THE INVENTION

[0003] The present invention generally relates to electrical circuits and more specifically to circuit analysis platforms for drug discovery.

BACKGROUND

[0004] A brain is an organ that may serve as the center of a nervous system. For example, the human brain is the central organ of the human nervous system, and with the spinal cord makes up the central nervous system. It is located in a head, usually close to the sensory organs such as the eyes, ears, nose, tongue, and skin.

[0005] In addition, a brain may control most of the activities of the body. The human brain includes 86 billion neurons and other cells (e.g., supportive glial cells). Brain activity is made possible by the interconnections of neurons and their release of neurotransmitters in response to nerve impulses. Neurons connect to form neural pathways, neural circuits (may also be referred to as “brain circuits” or “circuits”), and elaborate network systems. Circuits may be considered the complex processing units of the brain that generate, modulate, and execute essential operations. SUMMARY OF THE INVENTION

[0006] The various embodiments of the present circuit analysis platforms for drug discovery contain several features, no single one of which is solely responsible for their desirable attributes. Without limiting the scope of the present embodiments, their more prominent features will now be discussed below. In particular, the present circuit analysis platforms for drug discovery will be discussed in the context of hippocampal circuits. However, the use of particular parts of the brain and particular circuit elements are merely exemplary and various other implementations may be utilized as appropriate to the requirements of a specific application in accordance with various embodiments of the invention. After considering this discussion, and particularly after reading the section entitled “Detailed Description,” one will understand how the features of the present embodiments provide the advantages described here.

[0007] One aspect of the present embodiments includes the realization that brain disorders affect as much as one fourth of the population and may be associated with major disability leading to an enormous financial burden on individuals and society at large. Further, the development of therapeutics may be a slow and inexact process and the great majority of such efforts typically fail. Development of pharmaceutical agents as psychiatric therapies may begin with animals (e.g., rodents) and analysis of efficacy with regard to targeted cellular processes. Such processes are present in humans as well as the test animals. The experimental compound may then be tested on animal behavior, either normal or genetically altered, intended to reproduce symptoms of a particular neuropsychiatric disorder. Results from this second stage of drug development have a poor record of predicting human outcomes and the inadequacies of animal tests is a major contributor to the failure of drug development programs. There is therefore a pressing need for a new level of analysis that utilizes elements shared by rodents and humans and improves on the uncertainties associated with rodent behavioral models of psychiatric illnesses.

[0008] Another aspect of the present embodiments includes the realization that the brain uses elaborate circuits to generate complex behaviors of the type affected by psychiatric illnesses. However, little is known about throughput and signal processing by such circuits. A representative example of this involves the hippocampus, a structure that is critical to the acquisition and recall of everyday forms of memory and has been implicated in many psychiatric diseases. Neuroscientists have described the interconnections between the three major subdivisions of hippocampus and shown that these are conserved across the mammals including humans. The hippocampal circuit may be considered as being one level above detailed descriptions of the operations at its individual links and one step below the many difficulties associated with analyses of the behaviors it supports. There is accordingly a major unmet need for a method that enables quantitative analysis and manipulation of the operations carried out by the entire circuit.

[0009] Another aspect of the present embodiments includes the realization that the brain, including hippocampal and cortical circuits, operates across a range of frequencies (e.g., 5-10Hz, 0; 20-30Hz, 0; and 40-100Hz, y) and combinations thereof (e.g., 0y, 00). Transformations, such as filtering or amplification, of incoming signals by circuits (e.g., a hippocampal circuit) occur in a frequency and pattern-dependent manner. As with other complex systems, circuits may be prone to catastrophic changes in which small disturbances may have drastic functional consequences. Thus, while a drug, mutation, disorder, and/or disease may have a small effect at any single node, it can have major effects upon circuit throughput. Current technologies cannot test for such effects. In the specific instance of drug evaluation, a circuit level approach enables the identification and manipulation of the site(s) of maximal effect; an output not provided by prior methods. Furthermore, the operational state of the circuit, which is induced by specific frequencies/patterns of activation, may have profound effects upon the action of a drug. Conducting analyses and manipulations at the wrong site, or operational state could lead to incorrect conclusions regarding potency, utility, safety, and indeed, the basic nature of the effects a test compound, known drug, and the like, elicits. A related problem involves the possibility that the test compound, known drug, and the like, produces unexpected and potentially counterproductive effects at sites not sampled— again, these problems are addressed by the circuit analysis and manipulations enabled by the invention.

[0010] Another aspect of the present embodiments includes the realization that genetic mutations to mice are a primary means for studying inherited neurological and psychological disabilities. Such mutations are identical to, or very close to, those found in humans with the disorder of interest. The mutations typically trigger neurobiological effects that are found in human patients and thus constitute valuable models of specific conditions. Functional readouts disclosed in the prior art may be comparable to those for drug effects: changes to synaptic transmission and behavior. The extent to which behavioral disturbances in the animals in fact recapitulate those in people is questionable and there has not been good success in translating treatment results from animals to humans. This reflects the enormously greater complexity of cognitive, and related, processing in people relative to mice, and the strong effects of speciesspecific behavior in such rodents. Analysis and manipulation of brain circuits (far more alike in mice and humans than behavior) however, would provide readouts of the systems that generate behavior. As with drugs, mutations exert their effects to different degrees across the cell types found in brain. Even in cases where the mutation is restricted to a single class of neurons, conventional methods fail to define the functional consequences of the mutation on complex brain operations. Such functional analyses are also accomplished by the invention.

[0011] Another aspect of the present embodiments includes improvements that overcome major disadvantages found in prior art relating to the how drugs, life experiences (e.g., addiction, stress), injury, or genetic mutations alter brain operations. Such work has focused on single sites within the complex brain circuits that generate functionally significant outcomes. However, there is no reason to assume that the selected target for analysis is in fact the primary site of action for any of the conditions described herein. Moreover, prior art has used a limited set of input and output features to evaluate disturbances. This restriction narrows the range of detectable defects. An appropriate analogy involves testing for the locus of failure in a complex electronic circuit. An engineer would begin their analysis by activating the input to the circuit with various patterns and measuring the signals emitted by the circuit’s output. They would also be alert to diverse types of circuit operations. An informed search could then be initiated for defective elements. The invention enables this approach for brain circuits and accordingly constitutes a major advance over the prior art. In addition, the invention also describes the functional effect of manipulations on the operation of the entire circuit, something that is critical to the evaluation of therapeutic interventions and is not possible with the methods disclosed in the prior art.

BRIEF DESCRIPTION OF THE DRAWINGS

[0012] The various embodiments of the present circuit analysis platforms for drug discovery now will be discussed in detail with an emphasis on highlighting the advantageous features. These embodiments depict the novel and non-obvious features of circuit analysis platforms shown in the accompanying drawings, which are for illustrative purposes only. These drawings include the following figures:

[0001] Fig 1 is a schematic diagram illustrating a hippocampal circuit with direct and indirect cortical inputs highlighted in accordance with an embodiment of the invention.

[0002] Fig 2 is a diagram illustrating the 4 key operations performed by the hippocampus and having subfields that have unique anatomical and functional properties in accordance with an embodiment of the invention.

[0003] Fig 3 is a diagram illustrating responses to afferent stimulation patterns occurring in a synapse-specific manner, while thresholds for plasticity similarly vary across the hippocampal nodes, in accordance with an embodiment of the invention.

[0004] Fig 4 is a diagram illustrating representative examples of three primary output measures recorded from CAI in accordance with an embodiment of the invention.

[0005] Figs. 5A-C are diagrams illustrating activation of the LPP input to the hippocampal circuit in accordance with an embodiment of the invention.

[0006] Figs. 6A-C are diagrams illustrating transient activation of cholinergic projections to Cornu Ammonis (CA) 3 enhances mossy fiber (MF)-evoked reverberating activity in accordance with an embodiment of the invention.

[0007] Figs. 7A-B are diagrams illustrating epileptiform activity detected during (Fig. 7A) and following (Fig. 7B) a stimulation train in accordance with an embodiment of the invention.

[0008] Fig 8 is a diagram illustrating activation of the mossy fiber (MF) input producing a complex response in accordance with an embodiment of the invention.

[0009] Figs. 9A-C are diagrams illustrating a somato-dendritic amplifier in CA3 may be dependent upon mobilization of the local recurrent connections in accordance with an embodiment of the invention.

[0010] Fig 10 is a diagram illustrating a CA3 C/A system operating as spatial amplifier following activation of the MF inputs in accordance with an embodiment of the invention.

[0013] Figs. 11A-B are diagrams illustrating transient synapse-specific plasticity within the hippocampal circuit in accordance with an embodiment of the invention. [0014] Fig. 12 is a diagram illustrating aging may be associated with impaired throughput in accordance with an embodiment of the invention.

[0015] Fig. 13 is a diagram illustrating impaired amplification of responses evoked by the LPP-DG-CA3 (indirect) pathway may be evident during aging in accordance with an embodiment of the invention.

[0016] Figs. 14A-B are diagrams illustrating aging may be associated with impaired filtering of direct LPP inputs in accordance with an embodiment of the invention.

[0017] Figs. 15A-B are diagrams illustrating a CA3 filter may be selectively impaired during aging in accordance with an embodiment of the invention.

[0018] Figs. 16A-G are diagrams illustrating application of Ketamine and circuit analysis in accordance with an embodiment of the invention.

[0019] Figs. 17A-F are diagrams illustrating application of CX614 (Ampakines) and circuit analysis in accordance with an embodiment of the invention.

[0020] Figs. 18A-G are diagrams illustrating application of neurosteroid allopregnanolone and circuit analysis in accordance with an embodiment of the invention.

[0021] Figs. 19A-E are diagrams illustrating circuit analysis of aging in accordance with an embodiment of the invention.

[0022] Figs. 20A-F are diagrams illustrating circuit analysis of Fragile-X Syndrome (FXS) in accordance with an embodiment of the invention.

[0023] Figs. 21A-D are diagrams illustrating circuit analysis of major depressive disorder in accordance with an embodiment of the invention.

[0024] Fig. 22 is diagram illustrating a process for implementing a circuit analysis platform in accordance with an embodiment of the invention.

DETAILED DESCRIPTION OF THE DRAWINGS

[0025] The following detailed description describes the present embodiments with reference to the drawings. In the drawings, reference numbers label elements of the present embodiments. These reference numbers are reproduced below in connection with the discussion of the corresponding drawing features. [0026] Turning now to the drawings, circuit analysis platforms for drug discovery may include preparation of a portion of a brain for activation of inputs. For example, many embodiments may include a rodent hippocampal slice preparation in which activations of inputs are applied to the first stage of a circuit in order to measure responses at the output stage. In various embodiments, the invention may also include inputting a collection of patterns to the circuit, corresponding to those patterns occurring during various behavioral activities and which shift the operational state of the circuit (i.e. amplification, filter, plasticity). In several embodiments, the invention may include measuring (e.g., up to four, or more) essential circuit level operations initiated by patterned input such as, but not limited to, filtering, amplification, reverberating activity, plasticity, and the like. In a variety of embodiments, the invention may also include determining if any of circuit level operations are affected by a treatment of interest and identifying circuit links and nodes responsible for one or more changes in circuit operations. Thus, the invention may allow for the measurement of circuit functioning while including a capacity for pinpoint localization of the site and nature of any inputs of disturbances.

[0027] As described herein, the invention may enable an entirely new level of description, analysis, and modulation regarding the manner in which drugs, mutations, and other condition of neuropsychiatric interest exert their effects on brain operations. Importantly, such descriptions concern the systems that are critically involved in the production of behavior and psychological states. The invention will therefore enable dramatically improved predictions related to clinical outcomes than those obtained with extant technologies. Thus, the invention may enable greatly improved, accelerated, and accurate drug discovery, development, and safety testing for therapeutics across multiple brain disorders. Circuit analysis platforms in accordance with embodiments of the invention are discussed further below.

OVERVIEW OF CIRCUIT ANALYSIS PLATFORMS

[0028] In many embodiments, the invention may utilize a customized in vitro brain slice preparation to analyze operations across the entirety of the hippocampal circuit. A schematic diagram illustrating a hippocampal circuit 100 with direct and indirect cortical inputs highlighted in accordance with an embodiment of the invention is shown in Fig. 1. The hippocampus circuit 100 may include a series of three interconnected stages: a) the dentate gyrus (DG) 102, b) field CA3 104, and c) field CAI 106 (output stage). The cortex provides the major inputs via the lateral and medial perforant paths (LPP and MPP, respectively), both of which project to the DG 102 and CA3 104. In various embodiments, the DG 102 may have but one output that terminates in CA3 104; this projection is highly specialized in several regards. Thus, the cortex may have direct connections 110 with CA3 and via the DG an indirect link 112.

[0029] Field CA3 104 may generate a massive feedback system that interconnects the neurons within the field. Branches of this system may form a very dense input to all of the cells within field CAI 106. Output 114 from CAI 106 may go to the subicular complex and via this intermediary back to cortex. These connections constitute the major links in the hippocampal circuit 100 with the DG 102, CA3 104, and CAI 106 acting as local processing nodes. The nodes may have local circuitry most of which involves a diverse array of interneurons that use the inhibitory (hyperpolarizing) amino acid GABA as a transmitter - in contrast, links and feedback connections may release the excitatory (depolarizing) transmitter glutamate.

[0030] A diagram illustrating the hippocampus performing 4 key operations and having subfields that have unique anatomical and functional properties in accordance with an embodiment of the invention is shown in Fig. 2. In many embodiments, various parts of the brain may be represented by a brain circuit 200 (e.g., a hippocampal circuit) that may be utilized as a circuit analysis platform for drug discovery by predicting responses of such complex systems to local changes. In some embodiments, the brain circuit 200 may include various circuit representations including, but not limited to, the DG circuit node 202, CA3 circuit node 204, and the CAI circuit node 206. In various embodiments, the brain circuit 200 may be configured for various operations (e.g., circuit level operations), including, but not limited to, amplification, filtering, reverberating activity, plasticity, etc. For example, the various nodes (e.g., DG circuit node 202, CA3 circuit node 204, CAI circuit node 206) may include various electrical elements such as, but not limited to, amplifiers, feedback loops, etc. Local connections may differ markedly between the three nodes 202, 204, 206 (see Fig. 2) indicating that each performs a specialized type of signal processing. The present embodiments may be configured to analyze, assemble, manipulate, and predict outcomes across all of these connections and local processing operations. [0031] Because of its relative anatomical simplicity, the hippocampus is the most commonly used region for attempts to derive general rules for the much larger neocortex. Despite this, the invention constitutes the first use of the structure for circuit analysis. While hundreds of studies have analyzed the physiology and biochemistry of discrete synaptic connections within the structure, descriptions of the flow of activity across the circuitry — to say nothing of how this occurs with naturalistic input — are conspicuously missing in prior work. The absence of a circuit platform has had serious consequences for investigations into the effects of drugs and other treatments that affect the operation of brain networks. Take, for example, the case of benzodiazepines (BZ), drugs that enhance GABA transmission and which are among the most commonly prescribed of psychiatric medicines. It is well established that the compounds increase inhibition within each of the hippocampal nodes but how this affects the operation of the entire circuit is entirely unknown. Moreover, it is possible, and indeed likely, that a given BZ has different effects across the nodes simply because the DG, CA3, and CAI contain different types of interneurons with different functional properties. An analysis at any given site could therefore generate misleading conclusions about thresholds, side effects, general outcomes, etc. The present embodiments of the invention enable the analysis of net effects of compound(s) on circuit operations and thereby guides drug discovery, development, and safety analysis with regard to specific links and/or nodes. Such methods greatly enhance drug classification, discovery, comparison, and the like. Similar arguments apply to effects of genetic mutations and aberrant life history (drug abuse, stress, etc.).

Inputs

[0032] In several embodiments, the invention may involve the use of exogenously applied multiple input patterns to engage different circuit operations. The patterns may be delivered in brief trains of pulses and include, among others: theta (5Hz), beta (25Hz), gamma (50 Hz), and two interleaved patterns (theta-beta and theta-gamma). As noted above, these may be used to mimic patterns known to be active during various behaviors and thus allow the circuit analysis to closer approximate real world conditions. But there is another reason for incorporating patterns into the invention: the excitatory synapses at different links in the circuit respond in surprisingly different ways to the patterns (see Fig. 3 for examples). A diagram 300 illustrating responses to afferent stimulation patterns occurring in a synapse-specific manner, while thresholds for plasticity similarly vary across the hippocampal nodes, in accordance with an embodiment of the invention is shown in Fig. 3. The diagram 300 illustrates an LPP-DG response 302 and an LPP- CA3 response 304. In addition, diagram 300 also illustrates a chart 306 that provides relative long-term potentiation (LTP) thresholds. In some embodiments, the LTP thresholds may be classified as high, medium, and low. For example, LPP-DG is high, MPP-DG is high, LPP-CA3 is medium, MF-CA3 is high, CA3-CA1 is low. The use of patterned input thus reveals two dominating factors - filtering and amplification - for circuit throughput, while the invented circuit analytical approach reveals that synaptic operations of this type can occur in different ways across the various circuit links.

Outputs

[0033] In a variety of embodiments, the analysis may use multiple output measures. Each of these may be considered conventional but the combination may be considered inventive. The first measure samples baseline (no stimulation) communication across the circuit and involves large field potentials (sharp waves: SPWs) generated by self-organized activity of neurons within CA3. SPWs are then transferred to CAI. As described herein, stochastic release from the DG- CA3 connections may prompt CA3 to produce SPWs. A diagram 400 illustrating representative examples of three primary output measures recorded from CAI in accordance with an embodiment of the invention is shown in Fig. 4. One to three measures are typically used to evaluate responses evoked in the CAI output stage by patterned stimulation of the perforant path. These may include: a) field excitatory postsynaptic potentials (fEPSPs) 402 which reflect synaptic currents in dendrites; b) synchronized action potentials generated multiple target neurons (‘population spikes’) 404; and c) patterns of individual spikes from several neurons (e g., single units 406 showing a raw signal 408 and a filtered signal 410). In some embodiments, custom software developed as part of the invention may be utilized for this last measure. This collection, analysis, and manipulation of baseline and multiple evoked measures has not been used previously.

[0034] Examples of input/output relationships: Activation of the perforant path input (LPP) may produce a complex, composite response in CAI. As described, the perforant path may have two branches, one going to the DG and the other terminating in CA3. Diagrams illustrating activation of the LPP input to the hippocampal circuit in accordance with an embodiment of the invention are shown in Figs. 5A-C.

[0035] The present invention teaches that the isolation of these revealed that the initial CAI response is largely driven by the direct LPP-CA3 input, while the indirect (LPP-DG-CA3) path produces the delayed secondary component (see Fig. 5A). Fig. 5A illustrates LPP activation promotes a complex response (i.e., combined LPP input 502) where the initial component is attributable to the direct (LPP-CA3) input 504 and the secondary response is mediated by the indirect (LPP-DG-CA3) pathway 506. Activating the two LPP inputs together using brief stimulation trains across a range of frequencies (5Hz, 20Hz and 50Hz) and patterns (e.g. 0y) produced some surprising results. Specifically, activation at 5Hz promoted a robust amplification of the response in CAI, at the level of the fEPSP and population spike, while a consistent output was evident at the single unit level (see Fig. 5B graph 512 for 5Hz and graph 514 for 20Hz). In contrast, activation at higher frequencies produced a biphasic response where an initial facilitation was followed by a marked suppression of the fEPSP and a complete loss of cell spiking (see Fig. 5C graph 522 for 5Hz and graph 524 for 20Hz). Figs. 5B-C illustrate that frequency dependent amplification and filtering of population (Fig. 5B) and single unit (Fig. 5C) events occur in CAL Of note, the amplification and suppression of responses was primarily evident on the component of the response driven by the LPP-DG-CA3 route 506, while the more direct LPP-CA3 path 504 is responsible for the initial component of CAI output. In summary, the multiple measures, analysis, and manipulation of circuit output to multiple inputs, which are easily and rapidly collected (timeline of minutes), may provide deep drug discovery, development, and safety insights regarding the operation of diverse circuit constituents.

Operations

[0036] 1) Multiple measures of normal circuit function: The present embodiments of the invention may include the definition and sampling of essential circuit level operations. These operations can include: a) Throughput in which activation of the perforant path input generates the output responses summarized immediately above (‘output’). Filtering and amplification occurring in the links and nodes of the circuit shape throughput, b) Reverberation or ‘buffer memory’ in which activity persists after termination of a brief input, c) Plasticity: Modifications to the strength of synaptic transmission such that a second presentation of an input elicits a different output than that produced by the first presentation of the same input. This operation may be related to the short and long-term encoding of information. While these operations are evident in circuit output, they are largely determined by the local circuits and interactions within each subfield.

[0037] In many embodiments, the invention may also involve the inclusion into the circuit platform of marked differences between links and nodes in the execution of the above operations. (Fig. 3 describes dramatic variations in amplification and filtering for three synaptic populations.) In some embodiments, the circuit platform may utilize different thresholds for plasticity at the different network links with the lowest thresholds being in field CAI and the highest in the DG, as one example. While signal transformations and plasticity may be evident within each subfield, the ability to generate prolonged reverberating activity may be restricted to field CA3 and may require activation of the dense recurrent commissural-associational system (C/A). Field CA3 may also be unique with regard to its two primary inputs, which in addition to innervating distinct subcellular domains of pyramidal cells, activate the cells along the proximo- dorsal axis of the subfield in a temporally distinct manner. In various embodiments, these features may also be included in the circuit platform.

[0038] The complexity of local circuit operations within subfields may make it extremely difficult to predict output from any of the subfields and how they might interact to produce the overall hippocampal output. The current invention allows all of this complexity to be captured using the end point measures. In addition, by incorporating the information summarized above, it enables the interrogation of the operations that occur in each circuit link and node. As a result, the invention may provide for identifying the site (or sites) and nature of action (or actions) produced by any treatment, compound, or therapeutic manipulation of interest as well as the net effect on overall circuit operations.

[0039] 2) External gain control for circuit operations: Electronic circuits commonly have inputs that serve to increase/decrease throughput. The extrinsic input in these cases does not act as a link or node but instead modulates overall operation of the circuit (e.g., volume up or down for an acoustic circuit). In a variety of embodiments, the invention may further involve adding such a gain control feature to the hippocampal circuit. The present invention teaches that such implementation may be achieved by activating ascending projections from the lower brain. Diagrams illustrating transient activation of cholinergic projections to CA3 enhances mossy fiber (MF)-evoked reverberating activity in accordance with an embodiment of the invention are shown in Figs. 6A-C. In Fig. 6A, image 602 shows the projections and image 604 shows a close up. In several embodiments, projections may generate sparse populations of axons whose synapses release one or more of the following transmitters: acetylcholine (ACh), norepinephrine, dopamine, or serotonin. Each of these forms a scattered population of synapses located in one or more nodes of the primary hippocampal circuit. Despite being few in number, activation of one the ascending inputs (ACh) has a potent effect on output measures (reverberation: see Figs. 6A- C), thereby confirming successful implementation of the invented gain control feature. In Fig. 6B, the baseline response 612 and the 5 minutes post stimulation response 614 are provided. In Fig. 6C, diagram 622 provides the normalized frequency as a function of bins (1 second) for MF & optical stimulation 624 and MF stimulation (5 Hz) 626. Numerous psychoactive drugs target the noted ascending systems and such compounds have important clinical usages. Despite this, very little is known about the manner in which the drugs affect the complex networks that generate psychological outcomes. In many embodiments, the invention may include combining circuit analysis with optogenetic activation of a specific ascending projection.

[0040] 3) Epileptiform activity: Epilepsy is a major clinical issue and the potential for triggering seizures is an important concern for drug development and safety testing. Relatedly, seizures are also associated with various brain disorders and figure into the design of therapeutic strategies for these conditions. Diagrams illustrating epileptiform activity detected during and following a stimulation train in accordance with an embodiment of the invention is shown in Figs. 7A-B. Specifically, diagram 702 of Fig. 7A illustrates epileptiform activity detected during a stimulation train and diagram 712 of Fig. 7B illustrates epileptiform activity detected following a stimulation train. The present circuit platforms may include the generation of epileptiform discharges as one of its operations and endpoint measures, a feature that relates naturally to the gain control feature. The use of multiple input patterns may also facilitates testing for minimum circumstances that initiate epileptiform activity. Simultaneous testing for several levels of seizures along with multiple normal circuit operation may be a component of circuit analysis enabled by the disclosed methods.

[0041] Although circuit analysis platforms are discussed above with respect to Figs. 1-7B, any of a variety of circuit analysis platforms including various circuits and circuit configurations as appropriate to the requirements of a specific application may be utilized in accordance with embodiments of the invention. Analysis of individual circuit nodes in accordance with embodiments of the invention are discussed further below.

ANALYSIS OF INDIVIDUAL CIRCUIT NODES

[0042] The present embodiments of the invention may include sampling circuit outputs in response to diverse inputs and then pinpointing the contributions of individual links and nodes to the outputs. This feature may greatly increase the utility of circuit analysis. For example, it is not known if the Fragile-X Syndrome mutation, a common cause of intellectual disability and autism, produces accumulating problems across the circuit or instead exerts disproportionate effects at a particular site. The same can be said for various psychoactive drugs. By the use of the present invention, effects on circuit output may enable the location of within-circuit sites of action - the following sections further describe aspects of the invention that provide targets for nodal analyses and manipulations to be performed after assessing multiple levels of circuit output.

[0043] LPP-DG LINK. In many embodiments, the present invention may include adding to the circuit analysis for this link: 1) the synapses function as low pass fdters (low frequency input is transmitted to dendrites while high frequency input is not), and 2) a nonlinear amplifier between dendritic responses and the spiking output. This filtering/amplification concept for this link illustrates the concept of recasting biological operations as electronic circuit functions. A third electronic feature for the DG node may include a novel filter that acts over seconds during processing of naturalistic, interleaved input patterns (e.g., short high frequency packets separated by various intervals).

[0044] FIELD CA3. This node may receive direct and indirect (via the DG) perforant path inputs and may be differentiated from other hippocampal and cortical circuits by a dense recurrent (feedback) commissural associational (C/A) system. CA3 pyramidal cells thus have three primary inputs that innervate specific subcellular domains. Current descriptions of CA3 operations are not sufficiently detailed for use in the present invention. It is therefore necessary to introduce a series of features to characterize circuit operations of the CA3 node. These features can include:

[0045] 1. Filters and amplifiers: The output of the hippocampal circuit followed (and amplified) low frequency stimulation applied to LPP but was blocked at higher frequency. While this is consistent with filtering in the LPP-DG link, the invention discloses a powerful low pass filter within CA3. This may allow for a method for routinely assessing this unusual filtering operation and its inclusion in the nodal analysis for CA3. The technique uses activation of the dense feedback excitatory system of connections within the node at 25-50Hz and recording of synchronized discharges from the cell body layer, a test that can be performed in minutes. The CA3 node may also include a serial step amplifier such that the DG input elicits a short latency response that triggers multiple inputs from the feedback system (see Fig. 8). A diagram illustrating activation of the mossy fiber (MF) input producing a complex response in accordance with an embodiment of the invention is shown in Fig. 8. The LPP-evoked fEPSP 802 is provided for baseline 801 and clozapine N-oxide (CNO) lOpM 803. Further, the MF-evoked fEPSP 804 is provided for baseline 805 and CNO lOpM 807. In baseline 805 of the MF-evoked fEPSP 804, the initial monosynaptic MF response (circles 806, 809) is followed by a larger secondary response mediated by the C/A circuit. The total response to the DG input is thus greatly amplified and thereby rendered far more useful for drug development and testing. The graph 808 illustrates the normalized fEPSP amplitude as a function of time (mins) for monosynaptic MF response 810, LPP-evoked fEPSP 812, and MF-evoked fEPSP 814.

[0046] Diagrams illustrating a somato-dendritic amplifier in CA3 may be dependent upon mobilization of the local recurrent connections in accordance with an embodiment of the invention is shown in Figs. 9A-C. Fig. 9A provides a graph 902 that illustrates the normalized response as a function of pulse number for dendritic fEPSP 904 and pop spike 906. In a variety of embodiments, direct stimulation of the feedback input at the same frequency may not result in facilitation of the dendritic responses but may be associated with a clear enhancement of the spiking output by those cells (see Fig. 9A). Thus, the disclosed concept of a dendrite to cell body amplifier (see DG node above) may also apply to the CA3 node (see Fig. 9C). For example, Fig. 9C illustrates the circuit representation of pulse #1 933 and the circuit representation of pulse #10 924. Notably, the methods demonstrate that the spiking response of CA3 persists long after the initiating C/A input ends (see Fig. 9B pulse #1 912 with fEPSP 914 and pulse #10 919 with fEPSP 918). Utilizing this observation for circuit analysis may require a ‘temporal amplifier’, by which persistence of the nodal response serves to amplify the effects of an input. Surprisingly, responses elicited by low frequency activation of the DG inputs (mossy fibers) leading to activation of C/A feedback are much larger at the far end of CA3 (i.e., CA3a the part closest to CAI) than at the segment proximal to the DG (i.e., CA3c; see Fig. 10). Translating this into circuit analysis terms led to a ‘spatial amplifier’ located in the CA3 node. A diagram 1000 illustrating a CA3 C/A system operating as spatial amplifier following activation of the MF inputs in accordance with an embodiment of the invention is shown in Fig. 10. The diagram 1000 provides the CA3c 1002 with MF response 1004, CA3b 1012 with MF response 1014, and CA3a 1022 with MF response 1024. Analyzing and modulating the spatial amplifier may lead to novel and enhanced methods for drug discovery, development, and safety analysis.

[0047] 2. Reverberating activity. The CA3 C/A system is capable of generating self-sustained activity for remarkably long periods (e.g., minutes) following a brief activation at 5Hz and behavioral studies show that silencing of this system eliminates acquisition of temporal information (the order in which events occurred) but leaves intact the encoding of cue identity and spatial locations. The sustained activity seen in the CA3 node is not described for any other brain region. The effects that such activity has on the operation of the full hippocampal circuit may readily be analyzed using the present invention, enabling enhanced methods for drug discovery, development, and safety analysis.

[0048] 3. Plasticity. There is a long-standing consensus among experts that everyday forms of memory are encoded by modifications to the strength of synaptic communication. There is an enormous amount of literature showing that discrete populations of synapses undergo predicted changes when activated with certain input patterns. The development of circuit analysis as disclosed here enables the rapid, readily implemented techniques for pinpointing where in a circuit such changes occur. Diagrams illustrating transient synapse-specific plasticity within the hippocampal circuit in accordance with an embodiment of the invention is shown in Figs. 11A- B. Fig. 11A illustrates an example in which the DG-CA3 link was stimulated with a train of 10 electrical pulses at either 5 or 25Hz. Single pulses were applied at 3/min to measure synaptic strength (size of postsynaptic responses) prior to vs. after the trains. In Fig. 11 A, a monosynaptic MF response diagram 1102 is provided showing individual responses prior to and following (6 for each) a train of 10 electrical pulses at 5 Hz 1104 and 10 electrical pulses at 25Hz 1106. Further, a di-synaptic pop. spike (C/A) diagram 1112 is provided showing individual responses prior to and following (6 for each) a train of 10 electrical pulses at 5 Hz 1114 and 10 electrical pulses at 25Hz 1116. The 5Hz train caused a transient (minutes time scale) increase in the feedback C/A-mediated population response, with no change to the monosynaptic DG-CA3 response. Thus, repetitive activation of the basic within-node circuit results in plasticity at one link but not in the other. Surprisingly, 25Hz produced a larger effect despite the blocking effect of the above-described CA3 filter. Fig. 11B describes an example involving the operation of the circuit from LPP input to the DG to the dendritic response of the CAI output stage. In this case, brief patterns of LPP input (theta-gamma) proved effective. The initial phase of the CA3 response 1122 (evoked by single pulses) was unchanged, indicating that the brief stimulation did not affect the LPP-DG and DG-CA3 links in the circuit. But the delayed component that is generated by the CA3 feedback projection was transiently enhanced. The right side of Fig. 11B shows that the within-node modification in CA3 was relayed to CAI 1132. Accordingly, the invention discloses that the use of multiple input patterns and output measures provides actionable insight into unexpected circuit operations and enables a platform useful in the evaluation of treatment consequences. Note also that an CAI output measure points to a node (CA3) in which a key operation had occurred.

[0049] FIELD CAI . The output node of the circuit (CAI) may receive the vast majority of its input from CA3. There is also a small projection from the cortex via the perforant path and still weaker connections from lower brain areas. Analysis of the node indicates that it amplifies signals from CA3 while local interneurons strongly suppress spiking output. These features, among others, and methods for analyzing them are part of the invention. Nodal analysis also shows that memory related synaptic modifications (plasticity) has a much lower threshold in CA3-CA1 than is found at other sites in the circuit. This feature is included in the disclosed methods as the placement of different thresholds at various links in the circuit. [0050] Although analysis of specific circuit nodes is discussed above with respect to Figs. 8- 1 IB, any of a variety of circuit nodes and analysis of specific circuit nodes as appropriate to the requirements of a specific application can be utilized in accordance with embodiments of the invention. Detecting defects in circuit operations in accordance with embodiments of the invention are discussed further below.

DETECTING DEFECTS IN CIRCUIT OPERATIONS

[0051] As described herein, the present embodiments of the invention include circuit platforms that enable an unprecedented technology for defining how experimental compounds, drugs, mutations, manipulations, and other inputs affect complex brain operations. The following section illustrates some of these capabilities using an exemplary example of aging.

[0052] Memory loss, particularly of those forms dependent on the hippocampus, is one of the earliest psychological concomitants of aging. Despite this, little is known about the manner in which normal and pathological aging affects the transfer of information from cortex to the output stage of the hippocampus. And the critical question of whether aging acts uniformly throughout the structure or instead produces different kinds of functional alterations at different sites is unknown. Resolution of these issues may be a necessary step in designing therapeutics. The present embodiments provide the first answers. The disclosed results show an analysis that compared 3 month vs. 12-14 month old mice. The latter are considered by experts to correspond to middle age in humans.

[0053] Baseline function: The first measure used in the platform is the measurement of selforganized spontaneous activity generated in the CA3 node, prompted by stochastic release by the DG-CA3 link. The measure of this operation is Sharp Waves (SPWs). In young adults, a SPW in CA3 was shortly followed by a similar wave in CAI; this correlation was clearly weakened in the older animals. Conclusion: transfer of signals from CA3 to CAI is negatively affected by normal aging.

[0054] Throughput: Low frequency activation of its cortical input may engage the entire circuit producing ever-larger responses in the CAI output node. The steady enhancement of output during the stimulation train may be due to the multiple amplifiers discussed above. A diagram illustrating aging may be associated with impaired throughput in accordance with an embodiment of the invention is shown in Fig. 12. Diagram 1200 illustrates a control group’s response 1202 and an aging group’s response 1204 for a first and tenth electric pulses at 5 Hz. The facilitation of output during low frequency input may be greatly reduced in middle-aged mice as illustrated in Fig. 12. Conclusion: One or more amplifiers in the circuit are impaired by middle age.

[0055] Localizing defects in the circuit: A diagram 1300 illustrating impaired amplification of responses evoked by the LPP-DG-CA3 (indirect) pathway may be evident during aging in accordance with an embodiment of the invention is shown in Fig. 13. Diagram 1300 illustrates a control group’s response 1302 and an aging group’s response 1304 for a first and tenth electric pulses at 5 Hz. Responses of the CA3 node to indirect cortical input failed to facilitate during low frequency stimulation in the middle age cases as illustrated in Fig. 13. Conclusion: age related loss of throughput from cortical input to CAI output is due to defects located in early stages of the circuit.

[0056] Nodal analysis of CA3: Diagrams illustrating aging may be associated with impaired filtering of direct LPP inputs, most likely due to a reduction in dendritic inhibition, in accordance with an embodiment of the invention are shown in Figs. 14A-B. In Fig. 14A, diagram 1400 illustrates a control group’s CA3 response 1402 and an aging group’s CA3 response 1404. Diagram 1400 further illustrates the CAI “filter” response 1406 and the CAI no “filter” response 1408. In Fig. 14B, circuit representation 1420 is provided illustrating the possible effects. The direct cortical (perforant path) input to CA3 (LPP-CA3) is filtered in young adults and this effect is absent in older animals, likely due to a failure of local interneuron operations (see Figs. 14A- B). Diagrams illustrating a CA3 filter may be selectively impaired during aging in accordance with an embodiment of the invention is shown in Figs. 15A-B. In Fig. 15A, diagram 1500 illustrates the norm pop. spike amplitude for a train of 10 electrical pulses at 5 Hz for 3-4 month 1502 and 12-14 month 1504. In Fig. 15B, diagram 1510 illustrates the norm pop. spike amplitude for a train of 10 electrical pulses at 25 Hz for 3-4 month 1512 and 12-14 month 1514. The potent low pass nodal filter is also reduced with aging (see Figs. 15A-B). These filter defects are of interest in their own right but cannot explain the failure of CA3 to relay information to CAI. Conclusion: the throughput failure in middle age mice is likely due an impaired DG amplifier. [0057] The present invention therefore illustrates how circuit modulation and analysis enables the identification of clinically important conditions that affect complex brain operations. Much more can be learned by using the full array of capabilities included in the present circuit analysis platforms but the above data suffice to describe therapeutic targets. They also describe different defects and thus difficulties likely to be encountered by potential treatments.

[0058] In many embodiments, the present embodiments of the invention enable the broad analysis and modulation of conditions, disorders, diseases, and other external and internal factors that affect the operation of brain circuits - thereby providing novel, actionable, and clinically- relevant tools to advanced drug discovery, development, and safety analysis.

[0059] Although specific detection of defects in circuit operations are discussed above with respect to Figs. 12-15B, any of a variety detection of defects in circuit operations as appropriate to the requirements of a specific application can be utilized in accordance with embodiments of the invention. Example applications of circuit analysis platforms in accordance with embodiments of the invention are discussed further below.

EXAMPLE APPLICATIONS OF CIRCUIT ANALYSIS PLATFORMS

[0060] Example 1 : Ketamine. Ketamine is the first rapidly acting antidepressant drug. However, despite significant efforts, the mechanism(s) of action remain unclear. As a result, the development of novel, more selective rapid antidepressant compounds has been hindered.

[0061] Diagrams illustrating application of Ketamine and circuit analysis in accordance with an embodiment of the invention are shown in Figs. 16A-G. Experiments may be aimed to characterize the acute effects of ketamine (e.g., 50pM) on 1) the spontaneous, self-organized activity known as Sharp Waves (SPWs) known to be generated within the hippocampal circuit, 2) circuit output from the CAI end station of the circuit elicited by single pulse activation of a cortical input to the circuit, and 3) output triggered by cortical input delivered at 5 or 25Hz. These two frequencies respectively may correspond to the theta and beta rhythms routinely used by forebrain networks during behavior, and which engage different circuit operations (e.g., amplification, filtering). As illustrated in Fig. 16A, diagrams 1602, 1604 show that ketamine (50pM) had no effect upon the incidence or size of the spontaneous SPWs (diagrams 1602, 1604 show minimal drug effect on sharp waves (SPWs). In Figs. 16B-C, the drugs lack of effect on circuit throughput are illustrated, as measured by output from the CAI end station, to single stimulation pulses applied to the cortical input. In reference to Fig. 16B, the baseline response 1612 in comparison to the ketamine response 1614 show that the spike output to single pulses is unchanged. In reference to Fig. 16C, the diagram 1622 plots the number of spikes for baseline 1624 and for ketamine 1626, and shows that the interval between spikes is unchanged.

[0062] However, ketamine may cause a frequency-dependent increase in output to repetitive input as illustrated in Figs. 16D-G. In reference to Fig. 16D, diagram 1632 plots the number of spikes as a function of pulse number. Diagram 1632 shows that the spike output to 5Hz (theta) input is unaffected. In reference to Fig. 16E, diagram 1642 plots the mean latency to first spike (ms) as a function of pulse number. Diagram 1642 shows that latency to output is reduced during a 5Hz input. Interpretation of these results using the circuit design and circuit operation specified in the embodiments of the invention indicates that the likely site of action for ketamine lies within field CA3 and, in particular, the filtering of the responses generated by its massive recurrent collateral system (see Figs. 16F-G). In reference to Fig. 16F, diagram 1652 plots the number of spikes as a function of pulse number at 25Hz. Diagram 1652 indicates normal suppression of beta frequency (25 Hz) is converted into enhanced output by ketamine. In reference to Fig. 16G, diagram 1662 provides examples of normal filtering during 25 Hz input (‘baseline’) and enhancement by ketamine. Analysis within the CA3 node of the circuit confirmed that ketamine suppresses a CA3 filter. The circuit description specifies the synapses and cells responsible for the CA3 filtering operation, which greatly narrows the search for molecular mechanisms.

[0063] In many embodiments, circuit analysis may be utilized to identify a likely site and mode of action for a valuable but poorly understood drug. The results are not predicted from current ideas about the drug. In some embodiments, analysis may be conducted to confirm these points and specify additional molecular details about mode of action. The present invention provides a novel, information rich, and functionally meaningful target for testing the efficacy of various drugs including, but not limited to, new ketamine variants.

[0064] Example 2: CX614 (Ampakines). Ampakines are a novel family of memory enhancing drugs that facilitate AMP A receptor function. Studies have tested the effects of these compounds on monosynaptic responses in vivo and in vitro and concluded that 5pM concentration of the gold standard variant CX614 is close to the threshold for synaptic facilitation. Using the circuit analysis platform, reliable effects were obtained at 1 pM concentrations and thus at much lower values than those needed for monosynaptic effects as further described below.

[0065] Diagrams illustrating application of CX614 (Ampakines) and circuit analysis in accordance with an embodiment of the invention are shown in Figs. 17A-F. Figs. 17A-F illustrate that the effects of CX614 may depend on the input frequency. Application of CX614 at 1 pM to the circuit preparation caused a dramatic reduction in the amplitude and an increase in the frequency of SPWs, while the CAI spike output following single-pulse cortical stimulation was reduced (Fig 17A-D). In Fig. 17A, diagrams 1702, 1704 show a reduction of SPW amplitude. In Fig. 17B, the baseline response 1712 in comparison to the CX614 response 1714 show suppression of output to single pulse input. In reference to Figs. 17C-D, the diagrams 1722, 1732 show a decreased spike output and increase in latency between spikes, respectively.

[0066] In contrast to the single pulse stimulation, the drug greatly enhanced throughput following repetitive cortical activation at 5Hz as illustrated in Figs. 17E-F. In Figs. 17E-F, diagrams 1742, 1752 show an increased output to theta (5 Hz) input. There are no data or a priori arguments that would predict the SPW results or frequency dependency of throughput effects observed following single-pulse or repetitive stimulation. While effects are clear at a IpM dose, the actual threshold for drug actions remains to be determined. In some embodiments, circuit analysis may determine if the drug produces its effects by small actions at multiple links and nodes or if instead there are low threshold sites within the circuit for the compound. In accordance with embodiments of the invention, it may be anticipated that recurrence in CA3 may be particularly sensitive to low concentrations of the drug.

[0067] Example 3 : Neurosteroid. Steroid hormones play diverse and critical roles in maintaining bodily homeostasis but also manufactured in brain where they serve as small lipid messengers between neurons. Allopregnanolone (ALLO) is a compound of this type that is currently of intense interest because of work showing that it acts as a positive modulator of the GABAA receptors that mediate inhibitory transmission throughout the central nervous system. Its actions thus resemble those of the benzodiazepine class of drugs widely used to treat anxiety disorders. The possibility thus exists of using peripheral treatments with a naturally occurring compound to achieve many of the benefits associated with benzodiazepines. However, there is considerable uncertainty about which of the many GABAA receptor subtypes are targeted by ALLO and the nature of effects that might emerge from complex systems.

[0068] Diagrams illustrating application of ALLO and circuit analysis in accordance with an embodiment of the invention are shown in Figs. 18A-G. Figs. 18A-G show that ALLO causes pronounced but discrete changes to circuit operations. The compound at 100 nM had no evident effects on SPWs but substantially reduced the output (CAI cell discharges) produced by single pulse stimulation of the cortical input to the circuit (see Figs. 18A-C). In Fig. 18A, diagrams 1802, 1804 show an absence of reliable effects on SPWs. In Fig. 18B, the baseline response 1812 in comparison to the ALLO response 1814 is provided. In Fig. 18C, diagram 1822 provides the single pulse response for baseline and ALLO. Figs. 18C-D show a depression of spike output to single pulse activation of cortical input to circuit.

[0069] Despite this, ALLO enhanced throughput elicited by 5Hz activation of the cortical input (see Figs. 18D, E). In Fig. 18D, diagram 1832 shows an increased output to input arriving at 5 Hz. In Fig. 18E, diagram 1842 shows output to the 1^st and 10^th pulses in a theta (5Hz) train. A pattern in which throughput is depressed to single pulses but enhanced to theta (5Hz) input is not readily explained by current ideas about brain operations. More surprising still, this pattern was elicited by a drug that enhances excitatory receptors (ampakine; CX614) and by a compound (ALLO) that enhances inhibitory receptors. A feature of great utility embodied in the invention concerns the specification of treatment actions in terms of circuit operations performed by particular links and nodes of the circuit.

[0070] In Figs. 18F-G, diagram 1852 and 1862 show recordings from field CA3, the region which drives the CAI output station. Responses to single activations of circuit input are greatly reduced relative to the no drug, baseline condition. ALLO depressed the recurrent amplification of signals arriving from cortex that occurs in CA3 (see Figs. 18F, G), a result that explains why the steroid reduced throughput elicited by single pulse activation of cortical inputs. In some embodiments, the circuit locations and operations targeted by ALLO to enhance circuit throughput at 5HZ may be considered.

[0071] A brief summary of the primary outcomes from circuit tests for three agents with therapeutic potential is listed below: [0072] 1. The results describe previously unknown effects for ketamine, ampakine, and allopregnanolone. As these effects occur in a large circuit located in a structure critical for mood and memory, it is reasonable to assume that they will have major consequences on behavior.

[0073] 2. The combination of a complex circuit and multiple input patterns, as embodied in the invention, served to differentiate the three drugs but also identified unexpected overlaps in their effects. Results for the drugs that enhance excitatory (ampakine) or inhibitory (allopregnanolone) transmission provided a striking example: though having opposite functional effects at the level of individual neurons, each compound increased throughput to theta (5 Hz) pattern input.

[0074] 3 The description of the test system in terms of circuit components and operations (e g., multiple types of amplifiers and filters, recurrent feedback) as embodied in the invention proved to be very useful in specifying site and mode of action for a drug.

[0075] 4. The results support the idea that psychoactive drugs act differentially rather than uniformly across the many links and nodes of the circuit. For example, the effect of ketamine on throughput to 25 Hz input is strongly suggestive of a loss of specific filtering operation executed within field CA3. The circuit operations included in the invention provide means for testing the conclusion and further narrowing of the description of drug actions.

[0076] Example 4: Aging. Normal aging is associated with deficits in memory function, and increased probability for the development of age-related dementias including Alzheimer’s disease. However, the defective circuit level operations underlying the decline in memory processing during normal aging are unknown. Most work in this area has focused on the synaptic plasticity that encodes new memories but the lack of a circuit preparation has precluded testing of the very real possibility that aging impairs the information processing steps that precede storage. In some embodiments, such concerns may be addressed by investigating circuit level operations in slices prepared from mice of various ages, including, but not limited to, early middle-aged mice.

[0077] Diagrams illustrating circuit analysis of aging in accordance with an embodiment of the invention are shown in Figs. 19A-E. Figs. 19A-E show that throughput may severely impaired in 12-month old mice. In Figs. 19A-B, diagrams 1902, 1912 show latency to output spiking to single activations of input is reduced and number of spikes is decreased in middle aged slices, respectively. In Fig. 19C, diagram 1922 shows output during a 5Hz train is also reduced in older mice. There was a sizeable reduction in the frequency and size of SPWs in 12-14 month old mice (life span is about 30 months) and CAI output produced by single pulse or 5 Hz cortical input was clearly less than that found in young adults (see Figs. 19A-C). The circuit design embodied in the invention states that the throughput is driven by a sub-circuit composed of DG to CA3 and then amplification via cycling with CA3 with output to CAI (the ‘indirect path’). The design further specifies that the sub-circuit is modulated by direct input to CA3 from cortex. It can be deduced from this that the observed effects of aging are likely due to a defect in amplification or excessive filtering within CA3.

[0078] Indeed, deficits in the CA3 spike response were evident following single-pulse and repetitive cortical stimulation (see Figs. 19D-E). In Fig. 19D, diagram 1932 shows number of CA3 spikes triggered by single pulse activation of circuit input is markedly reduced by age. In Fig. 19E, diagram 1942 shows the response of CA3 to 5Hz circuit input is decreased in middle- aged mice. These results indicate that throughput in a critical memory circuit deteriorates to a surprising degree by early middle age. There is a high probability that these disturbances are directly responsible for memory problems that begin to develop during middle age. In this instance, circuit analysis has pinpointed an entirely novel and functionally critical target for a therapeutic program.

[0079] Example 5: Fragile-X Syndrome (FXS). FXS is the leading cause of inherited intellectual disability and affects about 1 in 7000 males. It is commonly associated with autism. The gene mutation blocks a protein needed for intra-neuronal transport and as expected from this causes a variety of behavioral disturbances. Mice lacking the pertinent gene exhibit several characteristics of the human disease. Work on neurobiological defects found that memory related synaptic plasticity is defective in these animals but surprisingly, has failed to identify abnormalities in excitatory transmission. However, circuit analysis reveals a dramatic disruption that could not have been predicted with conventional technology.

[0080] Diagrams illustrating circuit analysis of FXS in accordance with an embodiment of the invention are shown in Figs. 20A-F. Figs. 20A-F, indicate that throughput is increased in a mouse model of FXS. In Fig. 20A, diagrams 2002, 2004 show increased frequency of SPWs. While no changes were evident in SPW properties (see Fig. 20A), the CAI spike response to single pulse cortical stimulation were enhanced in FXS mice (see Fig. 20B). For example, in Fig. 20B, diagram 2012 shows latency to onset of circuit output (CAI spiking) is reduced and number of spikes is increased in FXS mice. In Fig. 20C, diagram 2022 shows pronounced increase in output to theta (5 Hz) pattern input in FXS mice. Throughput following 5Hz stimulation was relatively unaffected by the FXS mutation.

[0081] Normally inputs arriving at the beta (~25Hz) or gamma (50Hz) frequency ranges are not transmitted across the serial stages of the circuit. This low pass filtering operation is absent in slices prepared from FXS mice (see Figs. 20D-E). In reference to Figs. 20D-E, diagrams 2032, 2042 show dramatic increase in output produced by beta (25Hz) and gamma (50Hz) frequency input, respectively. Furthermore, throughput associated with theta-gamma patterned stimulation was enhanced in slices prepared from FXS mice (see Fig. 20F). In Fig. 20F, diagram 2052 shows a complex pattern of input (3 pulses at 50 Hz occurring every 200 msec) associated with learning produced greater output in FXS mice than in controls. The present embodiments may include a powerful low pass CA3 filter that blunts spiking responses to recurrent activity arriving in the beta frequency range or higher. In some embodiments, a second phase of analysis may include testing the prediction that the FXS mutation causes defective operation of the filter. The circuit analysis led to a new explanation for certain of the symptoms associated with FXS and autism.

[0082] Example 6: Major depressive disorder (MDD). Diverse lines of evidence implicate the hippocampus in MDD, which implies that neuropsychiatric disorder disturbs circuit level operations in the structure. Tests of this using the invented analytical system with its claimed operational features produced interesting results. Hippocampal slices were prepared from mice that had been singly housed in a colony room for 7-10 days and then administered a brief test for depression such as social interaction.

[0083] Figs. 21A-D are diagrams illustrating circuit analysis of MDD in accordance with an embodiment of the invention. Figs. 21A-D indicate that filtering within the circuit is impaired in a mouse model of psychological depression. In Fig. 21 A, diagrams 2102, 2104 show a reduction in the frequency but not amplitude of SPWs. The frequency of spontaneous SPWs was reduced relative to group-housed animals, which is suggestive of a failure of self-organized activity within field CA3. A more striking difference between the control vs. ‘depressed’ mice was observed following activation of the circuit using different stimulation frequencies. In Fig. 2 IB, diagram 2112 shows output during a 5Hz train was not significantly affected by the experimental treatment (single housing for 7-10 days). Cortical stimulation at 5Hz produced a similar CAI spike output in the two groups (see Fig. 21B). In contrast, the rejection (i.e., fdtered) of beta frequency (25 Hz) input that characterizes normal circuit operation was absent in the latter, “depressed” group of animals. In Fig. 21C, diagram 2122 shows normal filtering of 25 Hz input is absent in the depression model. In Fig. 2 ID, diagram 2132 shows examples of output elicited by the 1^st and 10^th input activations in a control and depressed animal. The loss of filtering in the single housed mice resembles that seen in the FXS animals and as in that case could be due to failure in the conceptualized filtering operations executed within CA3. In some embodiments, additional tests may be conducted using the embodiment of the invention for confirmation.

[0084] A brief summary of the results for three models of neuropsychiatric disorders is presented below:

[0085] 1. Aging causes a marked reduction of throughput across the hippocampal circuit. It follows from this that the essential contribution of hippocampus to memory processing in cortex are defective shortly after the beginning of middle age. Tests of circuit operations described in the invention can now be used to pinpoint the locus and nature of the age-related problem(s) that depress throughput.

[0086] 2. The Fragile-X mutation produces a surprisingly discrete disruption of overall circuit operations: impairment to the CA3 filtering operation that blocks transfer of high frequency signaling from the dentate gyrus to CAI. Thus, the circuit processes signals that would normally be rejected.

[0087] 3.The results from the depression model resemble those obtained in Fragile-X mice. Thus, analysis in terms of circuit operations can be used to reveal otherwise unrecognized similarities between the neurobiological concomitants of neuropsychiatric disorders.

[0088] Although specific applications of circuit analysis platforms are discussed above with respect to Figs. 16A-21, any of a variety of applications of circuit analysis platforms as appropriate to the requirements of a specific application can be utilized in accordance with embodiments of the invention. Processes for implementing circuit analysis platforms in accordance with embodiments of the invention discussed further below.

PROCESSES FOR IMPLEMENTING CIRCUIT ANALYSIS PLATFORMS [0089] A diagram illustrating a process for implementing a circuit analysis platform in accordance with an embodiment of the invention is shown in Fig. 22. In many embodiments, the process 2200 may include a Phase I 2202 that tests how a signal recorded from the output stage (e.g., CAI node) of a brain circuit (e.g., the hippocampal circuit) is influenced by specific pharmacological compounds or in mouse models of disease. In various embodiments, at this initial phase 2202 effects may be assessed upon an i) “idling” state 2204 and ii) one or more operational states 2206. For example, the outputs 2208 under the “idling” state 2204 may include spontaneous events such as, but not limited to, sharp waves (SPWs) and single units as well as the response to single pulse stimulation of the cortical input (e.g., LPP). In some embodiments, the operational states 2206 of the hippocampal circuit may be assessed following activation of the cortical input (e.g., LPP) across a range of frequencies and patterns (e.g., Theta 2210, Beta 2212, Gamma 2214, and Theta-gamma 2216). In some embodiments, the one or more operational states 2206 may provide outputs 2218, 2220, 2224, 2226 which may be characterized as throughputs, filters, etc. In some embodiments, the conceptualized hippocampal circuit may include a number of frequency-dependent components (e.g., amplifiers, filters, etc.) that may be selectively engaged by the different stimulation frequencies and patterns. Given the differing nature of the components underlying these frequency-dependent operations the effects of drugs and/or alterations associated with animal models of disease may only be evident in certain operational states. Both the “idling” and operational states 2204, 2206 may be assessed in the absence or presence of specific extrinsic (e.g., extra-hippocampal) inputs.

[0090] In reference to Fig. 22, the process 2200 may also include a Phase II 2226 that may utilize the findings from the initial circuit analysis conducted in Phase I 2202, in conjunction with knowledge of the conceptualized hippocampal circuit to pin point the specific location(s) of the experimental effects. In several embodiments, during Phase II 2226 of the circuit analysis, monosynaptic responses may be recorded (e.g., single units, fEPSPs, etc.) from the desired links and nodes. At this stage, in some embodiments, it may be possible to drill down to specific cell types using single-cell recordings where appropriate.

[0091] Although specific processes for implementing circuit analysis platforms are discussed above with respect to Fig. 22, any of a variety of processes for implementing circuit analysis platforms as appropriate to the requirements of a specific application can be utilized in accordance with embodiments of the invention. While the above description contains many specific embodiments of the invention, these should not be construed as limitations on the scope of the invention, but rather as an example of one embodiment thereof. It is therefore to be understood that the present invention may be practiced otherwise than specifically described, without departing from the scope and spirit of the present invention. Thus, embodiments of the present invention should be considered in all respects as illustrative and not restrictive.

Claims

WHAT IS CLAIMED IS:

1. A circuit analysis platform for drug discovery by predicting responses of complex systems to local changes, the platform comprising: a brain tissue prepared in vitro, wherein the brain tissue provides a brain circuit comprising a plurality of stages including at least one input stage and an output stage; wherein the brain circuit is configured to receive activations of inputs via the at least one input stage, wherein the activations of inputs comprises a collection of patterns corresponding to at least one behavioral activity; and wherein the brain circuit is configured to output responses via the output stage, wherein the output responses include at least one circuit level operation initiated by the collection of patterns.

2. The circuit analysis platform of Claim 1, wherein the in vitro preparation is a rodent hippocampal slice preparation and the brain circuit is a hippocampal circuit.

3. The circuit analysis platform of Claim 2, wherein the plurality of stages comprises a series of three interconnected stages.

4. The circuit analysis platform of Claim 3, wherein the three interconnected stages includes a dentate gyrus (DG) input stage, a field Cornu Ammonis 3 (CA3) stage, and a field CAI output stage.

5. The circuit analysis platform of Claim 4, wherein the brain circuit is configured to receive the activations of inputs via a lateral perforant path (LPP) and a medial performant path (MPP), wherein the LPP and the MPP connect to the DG input stage and the CA3 stage.

6. The circuit analysis platform of Claim 5, wherein the LPP and MPP has direct connections with the CA3 stage.

7. The circuit analysis platform of Claim 5, wherein the LPP and MPP has indirect connections with the DG input stage.

8. The circuit analysis platform of Claim 5, wherein the DG input stage has an output that terminates in the CA3 stage.

9. The circuit analysis platform of Claim 5, wherein the CA3 stage generates a feedback system that interconnects neurons within the CA3 stage.

10. The circuit analysis platform of Claim 9, wherein branches of the CA3 stage form a dense input to the CAI output stage.

11. The circuit analysis platform of Claim 10, wherein an output of the Cl output stage links to a subicular complex back to a cortex.

12. The circuit analysis platform of Claim 11, wherein the LPP, the MPP, the output of the DG input stage, the branches of the CA3 stage, and the output of the CAI output stage act as links in the hippocampal circuit.

13. The circuit analysis platform of Claim 11, wherein the DG input stage, the CA3 stage, and the CAI output stage act as local processing nodes of the hippocampal circuit.

14. The circuit analysis platform of Claim 13, wherein the local processing nodes have local circuity comprising a diverse array of interneurons that use an inhibitory (hyperpolarizing) amino acid GABA as a transmitter.

15. The circuit analysis platform of Claim 14, wherein the links and the feedback system release an excitatory (depolarizing) transmitter glutamate.

16. The circuit analysis platform of Claim 15, wherein the DG input stage, the CA3 stage, and the CAI output stage each comprise local connections performing a specialized type of signal processing.

17. The circuit analysis platform of Claim 16, wherein the brain circuit is configured to measure the at least one circuit level operation initiated by the inputting of the collection of patterns.

18. The circuit analysis platform of Claim 17, wherein the at least one circuit level operation comprises filtering, amplification, reverberating activity, or plasticity.

19. The circuit analysis platform of Claim 17, wherein the brain circuit is configured to measure if the at least one circuit level operation is affected by a treatment of interest.

20. The circuit analysis of Claim 19, wherein the brain circuit is configured to identify at least one node of the local processing nodes for one or more changes in the at least one circuit level operation.

21. A method for predicting responses of complex systems to local changes, the method comprising: preparing an in vitro preparation of a brain tissue, wherein the in vitro preparation of the brain tissue describes conditions for activating a brain circuit, and wherein the brain circuit comprises a plurality of stages including at least one input stage and an output stage; applying activations of inputs to the at least one input stage including inputting a collection of patterns to the brain circuit, wherein the collection of patterns corresponds to at least one behavioral activity; and measuring responses at the output stage including measuring at least one circuit level operation initiated by the collection of patterns.

22. The method of Claim 21, wherein the in vitro preparation is a rodent hippocampal slice preparation and the brain circuit is a hippocampal circuit.

23. The method of Claim 22, wherein the plurality of stages comprises a series of three interconnected stages.

24. The method of Claim 23, wherein the three interconnected stages includes a dentate gyrus (DG) input stage, a field Cornu Ammonis 3 (CA3) stage, and a field CAI output stage.

25. The method of Claim 24, wherein applying the activations of inputs includes providing inputs via a lateral perforant path (LPP) and a medial performant path (MPP), wherein both of the LPP and the MPP connect to the DG input stage and the CA3 stage.

26. The method of Claim 25, wherein the LPP and MPP has direct connections with the CA3 stage.

27. The method of Claim 25, wherein the LPP and MPP has indirect connections with the DG input stage.

28. The method of Claim 25, wherein the DG input stage has an output that terminates in the CA3 stage.

29. The method of Claim 25, wherein the CA3 stage generates a feedback system that interconnects neurons within the CA3 stage.

30. The method of Claim 29, wherein branches of the CA3 stage form a dense input to the CAI output stage.

31. The method of Claim 30, wherein an output of the Cl output stage links to a subicular complex back to a cortex.

32. The method of Claim 31, wherein the LPP, the MPP, the output of the DG input stage, the branches of the CA3 stage, and the output of the CAI output stage act as links in the hippocampal circuit.

33. The method of Claim 31, wherein the DG input stage, the CA3 stage, and the CAI output stage act as local processing nodes of the hippocampal circuit.

34. The method of Claim 33, wherein the local processing nodes have local circuity comprising a diverse array of interneurons that use an inhibitory (hyperpolarizing) amino acid GABA as a transmitter.

35. The method of Claim 34, wherein the links and the feedback system release an excitatory (depolarizing) transmitter glutamate.

36. The method of Claim 35, wherein the DG input stage, the CA3 stage, and the CAI output stage each comprise local connections performing a specialized type of signal processing.

37. The method of Claim 36 further comprising measuring at least one circuit level operation initiated by the inputting of the collection of patterns.

38. The method of Claim 37, wherein the at least one circuit level operation comprises filtering, amplification, reverberating activity, or plasticity.

39. The method of Claim 37 further comprising determining if the at least one circuit level operation is affected by a treatment of interest.

40. The method of Claim 39 further comprising identifying at least one node of the local processing nodes for one or more changes in the at least one circuit level operation.