Field of Art

The present invention relates to the field of neuroscience, namely an in-vitro system and method for real-time assessment of axonal injury.

Background Art

Traumatic brain injury (TBI) occurs when an external force injures the brain. Common causes of TBI include traffic accidents, sport collisions, military combat injuries, violence and various falls. Most cases involve sudden acceleration or deceleration of the brain within the cranium or a combination of movement as well as impact. TBI represents the key process, which leads to chronic traumatic encephalopathy (CTE). CTE leads to behavioral changes, cognitive decline, suicidal ideations and often premature death. Mechanistically, TBI results in diffuse axonal injury (DAI), which consists in damage to the axons typically emanating from specific brain areas to target brain areas in bundles generally known as white matter. Axonal bundles appear to be particularly susceptible to TBI due to relatively long distances they travel without major anchoring with surrounding cells and processes on their way. These characteristics of axons contrast to dendrites, which are significantly shorter, do not bundle and create arborizations interacting richly with a plethora of surrounding processes and cells.

In order to understand mechanisms underlying axonal injury in TBI and DAI, which is needed to develop currently lacking diagnostic, preventive and therapeutic approaches, several experimental paradigms of TBI have been developed to date. These can be divided into in-vitro and in-vivo models.

Most currently available in-vitro axonal injury models are based on substrate deformation, such as stretching the surface on which cells are grown (Hemphill, M. A. et al. A possible role for integrin signaling in diffuse axonal injury. PLoS ONE 6, e22899- (2011)), or using negative pressure to produce a local substrate deformation, and consequently a stretching in the axons (Tang-Schomer, M. D. et al. Mechanical breaking of microtubules in axons during dynamic stretch injury underlies delayed elasticity, microtubule disassembly, and axon degeneration. FASEB J 24, 1401-1410 (2010)). These models lack the possibility of studying the changes in the axon during the injury, from the very moment it starts. An alternative approach is called puffing, which consists in infusing medium to the cell culture growth medium (Chung, R. S. et al. Mild axonal stretch injury in vitro induces a progressive series of neurofilament alterations ultimately leading to delayed axotomy. J Neurotrauma 22, 1081-1091 (2005) and Gu, Y. et al. Polarity of varicosity initiation in central neuron mechanosensation. J Cell Biol 216, 2179-2199 (2017)). Similarly to other existing approaches, this approach precludes critical assessment of the changes that take place at the very moment of the axonal injury in addition to difficulties in discriminating axons from dendrites in the culture. In fact, majority of these systems preclude distinguishing clearly axons from the dendrites.

Microfluidic chambers are a tool that has been developed in the last 15 years primarily for investigating neuronal growth, and for pharmacological purposes. Taylor AM et al. (Nat Methods. 2005 Aug;2(8):599-605, EP 1581612, EP 2719756, US 7419822) have developed a microfluidic culture platform for the polarization of axons in a fluidically isolated environment. They propose the potential of the system for investigating spinal cord injury and axotomy, however they don’t report its potential for studying axonal injury in real-time by means of shear stress produced by a syringe and a syringe pump in contrast to the use of vacuum induced axotomy.

To study axonal injury, microfluidic chambers were fitted with pneumatic channels (Yap YC et al. PLoS One. 2017 May 4;12(5):e0176997; Tang-Schomer, M. D., Patel, A. R., Baas, P. W. & Smith, D. H. Mechanical breaking of microtubules in axons during dynamic stretch injury underlies delayed elasticity, microtubule disassembly, and axon degeneration. FASEB J 24, 1401-1410 (2010)). Such system does not allow for real-time imaging of axonal changes during the injury.

In terms of in-vivo models, numerous animal models of TBI have been developed. Although larger animals are closer in size and physiology to humans, rodents are mostly used in TBI research due to their modest cost, small size and standardized outcome measurements, among other reasons. Whereas early models of TBI addressed the biomechanical aspects of brain injury, more recent models have been targeted at improving our understanding of the detrimental, complex molecular cascades that are initiated by head trauma. Among them, four specific models are widely used in research: fluid percussion injury (FPI), cortical impact injury (CCI), weight drop-impact acceleration injury, and blast injury. However, the biggest disadvantage of in-vivo models is the use of animals, which although of great value raises ethical issues and does not recapitulate response of human brain to injury. Another disadvantage is the inability to study axonal injury only (as opposed to inury of all cell types and all processes including axons and dendrites) as well as to address axonal injury from the very beginning of its development. To date, no animal model allows real time imaging or immediate harvesting of tissue for analysis. Therefore, despite a more physiological setting, current animal models of TBI do not allow one to learn what happens during and immediately after injury, but only several minutes afterwards.

In summary, current in-vitro and in-vivo models offer significant opportunities to study TBI and DAI, however, not at the exact time it happens, but only several minutes later. Most importantly, all systems known in art preclude critically testing what happens exclusively in the axons versus dendrites, thus preventing learning about the mechanisms that trigger several cascades of intracellular events that govern axonal response to injury. The present invention provides the first in-vitro system and method that allows studying exclusively axonal injury from the very moment it happens and allows for real time imaging as well as studying its genetic, biochemical and other events.

Disclosure of the invention

Subject matter of the invention is an in-vitro system for real-time assessment of axonal injury comprising:

- a syringe pump fitted with a syringe;

- a microfluidic chamber, said microfluidic chamber containing a somatic compartment and an axonal compartment, wherein the somatic compartment and the axonal compartment are mutually connected by axonal channels, wherein the axonal channels are transected by a flow channel arranged substantially perpendicularly to the axonal channels and wherein the flow channel has a first end and a second end, and the first end of the flow channel has an opening receiving an output from the syringe of the syringe pump and the second end of the flow channel is connected to at least one reservoir.

The combination of a microfluidic chamber coupled to a syringe pump is the unique feature of the present invention. The combination of these components allows to cause an axonal injury and to observe it directly and immediately. Unlike in known systems, in the present invention there is no need for stretching or deformation of the substrate on which the neural cells are grown to cause axonal injury. There is thus no change of focal distance between an imaging device and the injured axon during the injury stage and there is no need to limit the substrate to materials having the necessary deformability as in the prior art. The combination with the syringe pump and the direct connection of the syringe pump into the flow channel allows to use microfluidic chambers which are very suitable for microscopic imaging. No prior art system so far allowed direct imaging and real-time assessment of the axons during the injury stage and immediately afterwards.

Axonal channels have dimensions suitable for accommodating neuronal axons. In some embodiments, axonal channels have the length of 450 to 900 micrometers and the width within the range of 2 to 20 micrometers, more preferably 5 to 7 micrometers.

The flow channel has dimensions suitable for injection or withdrawal of liquid by means of the opening and of the syringe pump. In some embodiments, the flow channel has the width within the range of 30 to 300 micrometers, more preferably 55 to 80 micrometers, most preferably about 60 micrometers.

It is strongly preferred that the flow channel has the same depth as axonal channels. Output means from the syringe pump may be adapted by adapters and tubing to the size of the flow channel opening.

In some embodiments, the flow channel may be provided with one or a plurality of reservoirs on the second end. Additional reservoirs may be connected with the somatic compartment and/or with the axonal compartment. The reservoirs are configured for holding medium which flows from the reservoirs into the channels and from the channels to the reservoirs when a flow of medium is caused by the syringe pump.

The system according to the present invention is suitable for use with a microscope for observing the processes occuring in the microfluiding chamber before, during and after the flow of medium as manipulated by the syringe pump.

In some embodiments, the system also contains a microscope where the microfluidic chamber is placed for real-time imaging of the axonal injury.

In some embodiments, the syringe pump (more particularly, the syringe fitted in the syringe pump) and the corresponding opening of the microfluidic chamber are connected by tubing. The tubing may, when needed, contain an adapter portion or a separate adapter element to adjust the diameters of the openings of the tubing, so that the tubing fits the syringe opening on one side and the opening of the microfluidic chamber on the other side. The fitting of all the tubing should be tightly sealed, in a way that no air enters the tubing and no liquid leaks from it (i.e., in an air-proof and liquid-proof way).

It is preferred that the tubing is made of a wettable material which prevents disruptions in the flow and decreases the formation of air bubbles. It may also be advantageous that the tubing is transparent so that the formation of air bubbles can be observed when it occurs and suitable known measures can be taken to remove the air bubbles.

The microfluidic chamber is typically made of a transparent material, or at least the portion comprising the axonal channels is made of a transparent material to allow for real-time imaging of the axons. The microfluidic chamber shape, material and construction can be, for example, as described in EP 1581612, EP 2719756 or US 7419822. Any other suitable microfluidic chamber fulfilling the aforementioned requirements can be used.

The flow channel is provided with an opening adapted to receive an output from the syringe pump on one side (first end), and with at least one reservoir on the other side (second end). The present invention further provides an axonal injury assay method using the system comprising a microfluidic chamber and a syringe pump fitted with a syringe, said microfluidic chamber containing a somatic compartment and an axonal compartment, wherein the somatic compartment and the axonal compartment are mutually connected by axonal channels, wherein the axonal channels are transected by a flow channel arranged substantially perpendicularly to the axonal channels and wherein the flow channel has on one end an opening adapted to receive an output from the syringe pump and at least one reservoir on the other end, wherein CNS cell culture is provided in the somatic compartment, with axons reaching through the axonal channels to the axonal compartment, said method comprising the steps of a) causing by means of the syringe pump a flow of medium in withdrawal direction (i.e., the medium flows from the microfluidic chamber to the syringe in the syringe pump), said flow of medium having a speed in the flow channel of 10-66 mm/sec, preferably 17-33 mm/sec. This is a normal flow speed and does not cause damage to axons; b) causing by means of the syringe pump a flow of medium at a speed in the flow channel of at least 166 mm/sec, and preferably between 166 and 1666 mm/sec or between 333 and 1000 mm/sec. This flow speed bends and injures the axon, and depending on the parameters (such as duration of the flow rate, cell type, age of the axons, etc.) can even cause axotomy; c) causing by means of the syringe pump a flow of medium in inverse direction to the direction used in step b). This can be at the same flow rate as in step b) or at a flow rate differing at most by 20 %, preferably at most by 10 % from the flow rate used in step b). This flow in inverse direction normalizes the flow rate in the channel; d) optionally causing by means of the syringe pump a flow of medium in withdrawal direction (i.e., the medium flows from the microfluidic chamber to the syringe in the syringe pump), said flow of medium having a speed in the flow channel of 10-66 mm/sec, preferably 17-33 mm/sec.

The term “CNS cell culture” encompasses cultures of cells which are part of the central nervous system (CNS) and which have processes, and their precursor cells. In particular, such cells include neurons, astrocytes, microglia, oligodendrocytes.

The method may optionally include a step of infusing substances by means of the syringe pump directly to a single axon or a subset of axons; this allows testing drugs effects on axonal damage or axonal recovery after injury. The step of infusing substances may be carried out before step a) or in between any of the steps a) to d), or after step d). The substance may be selected from medicaments, candidate medicaments (i.e., substances to be tested for their therapeutic effects), inhibitors or potentiators of axonal function, substances naturally occurring in human body (such as proteins, neurotransmitters, nucleic acids), more specifically in brain or peripheral nervous system. The method may also optionally include a step of taking a sample of the medium and/or cells from the somatic and/or axonal compartment, wherein the step of taking the sample may be carried out before step a) or in between any of the steps a) to d), or after step d). The sample may then be analyzed using, for example, biochemical, analytical, proteomic or nucleic techniques. The whole content of the compartment(s) may constitute such a sample. The techniques may include Mass Spectrometry, RNA or DNA analysis or other.

In step a), the flow of medium is at a speed of 10-66 mm/sec, preferably 17-33 mm/sec in the flow channel. The flow rate is preferably maintained for at least 0.5 minute or for 0.5-5 minutes, more preferably for 1-3 minutes. 0.5 to 1 minute is usually a sufficient time to stabilize the flow through the channel. Longer times than 3 minutes may in some embodiments result in progressive decrease of the levels of medium from the microfluidic chamber.

In step b), the flow of medium at a speed in the flow channel of at least 166 mm/sec or preferably between 166 and 1666 mm/sec or preferably between 333 and 1000 mm/sec is preferably maintained for 30-180 seconds, more preferably for 60-120, or for about 90 seconds. Preferably, the flow direction is withdrawal direction. However, in some embodiments, it may also be inverse direction to the withdrawal direction.

In step c), the flow of medium in inverse direction than in step b) and at about the same flow speed in the flow channel as step b) is preferably performed for a time necessary to bring the flow to about the same resulting flow speed in the flow channel as the speed used in step a) (the term “about” corresponds to the deviation of at most 20 %, preferably at most 10 %). Typically, the time is the same or shorter than the time used in step b), more preferably, it is at least 15 seconds.

In step d), the flow of medium at a speed in the flow channel of 10-66 mm/sec, preferably 17-33 mm/sec, is preferably performed for the necessary time to record post injury effects. At least 0.5 min is a typical time. For shortterm effects a range between 0.5-5 minutes is sufficient.

The medium used during the injury is usually the same medium that is used to maintain the cells (most typically, Neuron Differentiation Media). However, alternative physiological media without phenol red can be used, especially for imaging purposes. For example, Artificial CerebroSpinal Fluid (aCSF), which is commonly used in electrophysiology, is particularly suitable, as it allows better imaging.

Culture of Neurons in Microfluidic chambers: Microfluidic chambers are adaptable for use with a wide range of cell cultures, from primary neurons to stem -cell derived post-mitotic neurons and other cells developing processes. Neuronal cells may be enriched in the axonal channel by addition of other cell types such as oligodendrocytes or astrocytes. For the purpose of illustration of the process the use of neural progenitor cells (NPCs) derived from neural stem cells (NSCs) is described herein. Alternatively, NPCs can be derived starting from induced pluripotent stem cells or other suitable types of cells as previously described (Lacovich V, Espindola SL, Alloatti M, Pozo Devoto V, Cromberg LE, Cama ME, Forte G, Gallo JM, Bruno L, Stokin GB, A vale ME, Falzone TL. Tau Isoforms Imbalance Impairs the Axonal Transport of the Amyloid Precursor Protein in Human Neurons. J Neurosci. 2017 Jan 4;37(l):58-69; Pozo Devoto VM, Dimopoulos N, Alloatti M, Pardi MB, Saez TM, Otero MG, Cromberg LE, Marin-Burgin A, Scassa ME, Stokin GB, Schinder AF, Sevlever G, Falzone TL. aSynuclein control of mitochondrial homeostasis in human- derived neurons is disrupted by mutations associated with Parkinson's disease. Sci Rep. 2017 Jul 11;7(1):5042). The cells are seeded in the somatic compartment and cultured so that axons reach through the axonal channels to the axonal compartment. The cells are maintained in neuronal differentiation media (e.g. DMEM/F12, 1% v/v N2 supplement (Thermo Fisher), 2% v/v B27 supplement (Thermo Fisher), and 1% v/v penicillin-streptomycin, laminin (1 ug/ml), cAMP (100 nM), ascorbic acid (200 ng/ml), BDNF (10 ng/ml), GDNF (10 ng/ml), IGF (10 ng/ml)) until they are ready for the experimental procedure. All before mentioned components for the media are from Gibco, but can also be from others.

The cells can be genetically manipulated by transfection or transduction or other means prior to the axonal injury assay and subsequent imaging or other experiments, i.e. biochemistry, electrophysiology, etc.

Manipulations of cultures prior to axonal injury assay:

Fluorescent probes (such as MitoTracker, LysoTracker, Fluo4) can be used to visualize the organelles, cytoskeleton or cytoplasm of the axons and allow for imaging during injury. In some embodiments, the CNS cells (e.g., neurons) or the axons may be pre-treated before step a) by transfection, transduction or incubation with fluorescent probes. As an alternative pre-treatment, protein expression can be enhanced by various constructs or knocked down by shRNA or by other means. Most typically, incubation with fluorescent probes is performed in the somatic and axonal compartment, while transduction(s) or transfection(s) are carried out only in the somatic compartment. Alternative approaches are possible.

The system of the present invention is unique in coupling a microfluidic chamber with a syringe pump to produce axonal injury or to test substance(s) applied directly and selectively to axons or subset(s) of axons. Such system has not been developed yet. The proposed assay allows for experimental conditions that no current system provides. These conditions consist in generating a flow stress in a localized region of the axonal tract, stress that can be controlled in duration and intensity. Applying a negative pressure in the flow channel affects directly only the axonal segment that passes through that channel. In these conditions there is no deformation or stretching of the substrate as in the other models, only a fluid flow that exerts a perpendicular force to axonal trajectory. The stretching of the substrate, which occurs in the systems known in the prior art, hinders direct and immediate imaging of the axon during the injury, by changing the focal distance to the axon. Also the composition of the membrane that requires to present elasticity for stretching is not optimal for microscopy visualization. Thus, this model overcomes these two disadvantages by presenting a solid and optically efficient glass as substrate for the axons and by applying a defined force that keeps the axon relatively stable in the focal point during the whole experiment. Furthermore, the device of the present invention may further comprise a means for observing or recording axons and/or axonal changes in real-time arrangement. The observation can be performed during the preparatory period (step a)) and/or during the injury period (step b)) and/or during the post-injury period (steps c), d)).

The means for observing or recording axons and/or axonal changes is preferably a microscope, alternatively it can be electrophysiological or biochemical systems. The microscope can be an optical microscope or a fluorescence microscope. In the case of a fluorescent microscope it should contain a source of light and filter sets that are necessary to image the fluorescent probe or tag of interest.

Microscope settings for real-time recording should be in accordance to the sample preparation. It is recommended to test these settings beforehand in a common culture and/or in the somatic compartment.

Preferably, the microscope has a stage with an incubator provided with means configured to maintain the temperature of the stage at 34-39 degrees C. The temperature is important for the maintenance of the neurons or other CNS cells, as they are susceptible to temperature changes, and too low or too high temperatures can already affect the state of the cell culture. An optimum temperature is 37 degrees C.

In some embodiments, the device comprises a microfluidic chamber, a syringe pump, tubings and optionally adapters and a microscope to assess real-time axonal changes. The chamber consists of a somatic compartment that is connected to the axonal compartment by axonal channels. The axonal channels are transected by a flow channel perpendicular to axonal channels that ends in one side in a reservoir, while the other side is provided with an opening adapted to receive an output from the syringe pump through the tubing. The syringe pump is fitted with a syringe that is connected through the tubing and optionally through the adapters to the opening of the flow channel of the microfluidic chamber. In some embodiments, after assembling of the microfluidic chamber by pressure bonding with a coverslip and subsequent matrigel coating, neural progentior cells (NPCs) are seeded in the somatic compartment and differentiated to mature neurons with specific media. During the neuronal maturation axons start to grow in axonal channels. Typically, about the 14 ^th day of cultivation (DIV14) usually first axons start to appear in the axonal compartment (ac). At later stages more axons cross to the axonal compartment, until approximately DIV30. To perform the injury, the syringe and the tubing are loaded with PBS in such a way that liquid continuity is not interrupted by any bubble. All the reservoirs of the microfluidic chamber are topped up with a medium, such as a growth medium or aCSF to avoid any compartment to get dry during the experiment. The chamber is placed in the microscope stage with the coverslip side down and the tubing from the syringe pump inserted. By DIC (differential interference contrast microscopy) illumination, axons can be located in the perpendicular channels. A slow flow (step a)) (for example 10 ul/min, which corresponds to a flow speed of 33 mm/sec in the flow channel) is selected in the syringe pump in withdrawal direction, so that media flows from the chamber to the syringe. For the injury, the flow rate (preferably in withdrawal direction) and the duration are selected (step b)). After this period the same flow rate but inverse flow is applied (step c)) (for example during 20 seconds), to bring the flow to normal speeds. A steady flow rate (for example 10 ul/min, which corresponds to a flow speed of 33 mm/sec in the flow channel) is then set. Before, during and after the injury, the axons can be imaged in real-time by brightfield or fluorescence detection, depending on the previous preparation of the cultures. In some embodiments, after the injury, cultures can be fixed and microfluidic chambers detached gently from the coverslip for further immunofluorescence assays.

Brief Description of Drawings

Figure 1 is a schematic depiction of the microfluidic chamber with the syringe pump. On the rights side, an exploded view of a part of the microfluidic chamber is shown.

Figure 2 shows the mean measured speed in flow channel for each tested pump flow rate during time. Figure 3 presents a plot showing the decrease of intact axons and increase of enlargement formation with increasing pump flows (n=31 chambers).

Figure 4 shows an example of axotomy occuring at pump flow rates higher than 100 mΐ/min, which corresponds to flow speeds in the flow channel higher than 333 mm/sec for interval equal or longer than 90 secs.

Figure 5 is a time series showing the morphological changes of a GFP labelled axon before, during and after injury (t = time from injury (in seconds)).

Figure 6 shows representative images of axons labelled with GFP (upper panels) or mCherry (lower panels) before injury (Control) and during injury (Injury). The arrows mark the axonal enlargements which appear as a consequence of applying a flow rate 100 mΐ/min corresponding to flow speed in the flow channel of 333mm/sec. Scale bars are 10 um.

Figure 7 is a graph showing the maximum number of enlargements during injury versus the control setting.

Figure 8 is a graph and an illustrative image, showing the number of enlargements in the treatment with Ryanodine and with Ca++.

Examples of carrying out the Invention

In this example we present the full procedure, step by step to obtain mature neurons with axons extended in the channels of the microfluidic device. We also subject the axons to physical stress and to assess the damage we use two different constructs to image the membrane and the cytosol of the axon.

Equipment, materials, consumables

Materials:

Microfluidic chamber (Xona microfluidics, uLP)

Coverslips (Coming, thickness 1, 22 x 40 mm, #2975-224) p60

Syringe Pump (New Era Pump Systems Inc. NE-1002X)

Syringe (12 ml syringe, Chirana)

Tubings (Green) GE Healthcare PEEK tubing. 0,75 mm i.d.; 1/16 o.d. #18-1112-53, transparent Adaptors, syringe to tube.

Microscope (Zeiss Confocal LSM780, Zeiss Live LSM7, Heated stage, C02 stage)

Media, factors, consumables:

Paraffin Poly-0 (Sigma)

Matrigel (Coming)

Neuronal Differentiation Media: DMEM/F12 (cat. 31330), 1% v/v N2 supplement, 2% v/v B27 supplement, 1% v/v penicillin-streptomycin, laminin (Invitrogen, 1 pg/ml final), cAMP (Sigma, 100 nM), L-ascorbic acid (Sigma, 200 ng/ml), BDNF (10 ng/ml final), GDNF (10 ng/ml final), IGF (10 ng/ml final). Media, factors and supplements are from Gibco unless stated otherwise.

Growth and differentiation of stem cells for seeding in chambers

Human neurons were differentiated from human neural stem cells (NSCs, Gibco) to Neural Progenitor Cells (NPCs) through change in media composition. NSCs are seeded in a plOO dish coated with Matrigel (Coming) and maintained until confluency in NSC media (KO DMEM/F12, StemPro Neural Supplement 2% v/v, Glutamax 1% v/v, bFGF 20 ng/ml and EGF 20 ng/ml). After reaching confluency media is changed to NPC media (DMEM F12, B27 supplement 1% v/v, N2 Supplement 0.5% v/v, Glutamax 1% v/v) for 7 days, changing media every 2 days. After 7 days NPCs develop small projections and are ready for the terminal differentiation to neurons. Alternatively, the same protocol of differentiation can be applied to induced Pluripotent Stem Cells (iPSCs) derived NSCs. hiPSC colonies were maintained in pi 00 adherent culture plates (Coming) with a feeder layer of irradiated E13 mouse embryonic fibroblasts (iMEF) at 37°C, 5% C02. Cells were split weekly 1:3. Expansion medium was comprised of Dulbecco’s Modified Eagle’s KO Medium (KO-DMEM) supplemented with 10% v/v Knockout Semm Replacement (KSR), 2 mM non-essential amino acids, 2 mM L-glutamine, 100 U/ml penicillin, 50 pg/ml streptomycin, 0.1 mM b-mercaptoethanol and 4 ng/ml of bFGF. All reagents were obtained from Life Technologies (Carlsbad, CA, USA). Efficient neural derivation of the pluripotent cells was achieved by adaptation of a previously published protocols. Pluripotent colonies were enzymatically detached from its feeder layer using Collagenase IV (lmg/ml) (Gibco) for 25 min at 37°C . Cell aggregates obtained from two p 100 culture plates, were then suspended in 25 ml of expansion medium (without bFGF) in a T75 culture flask and kept in flotation for a total of 4 days at 37°C 5% C02, changing half of medium volume every other day. During this period, the cell aggregates acquire a spherical shape. The fifth day, medium was replaced for 25 ml Neural Induction Medium (NIM), composed of DMEM/F12 media supplemented with non-essential amino acids (1: 100, Invitrogen), N2 supplement (1: 100, Gibco), and 1 mg/ml Heparin at 37°C 5% C02. After 2 days the cell aggregates were attached in laminin pre-coated (20pg/ml) 6-well plate. About 30-40 aggregates per well were then kept in 2 ml of NIM at 37°C 5% C02 changing most of the medium every other day for 9 days. On the attachment stage, formation and maturation of neural tube-like structures and neural progenitors enriched rosettes was observed. On the 9th day, those multi-cell layered structures were mechanically dissected. These aggregates of neural rosettes cells were then transferred to T25 flasks with 15 ml of NIM supplemented with B27 supplement (1:50, Gibco) changing medium every other day. After 12 days in suspension, the NPCs are ready to dissociate and terminally differentiate to neurons.

Description of the system

The system used in the examples is composed of a microfluidic chamber, a syringe pump, tubings and adapters. A microscope 8 is used to assess real-time axonal changes. The chamber J_, shown in Fig. 1, consists of a somatic compartment 10 connected to the axonal compartment JT by 500 pm length channels (axonal channels 6) having the width of 6 pm. Each of the somatic compartment and the axonal compartment is provided with two reservoirs 3, 4. These axonal channels 6 are transected by one perpendicular flow channel 5 of 60 pm width; the flow channel 5 ends in one side in three reservoirs 9 while the other side has an opening where the tubing is connected. The syringe pump 2 holds a syringe that through a series or adapters and tubing is connected to the opening of the flow channel 5 of the microfluidic chamber. The detail on Figure 1 shows the axonal channels 6 transected by the flow channel 5, wherein axons are disposed in axonal channels and neuron cell bodies 7 are disposed in the somatic compartment J_0. while the endings of the axon reach the axonal compartment J .

Microfluidic Chamber assembly

The microfluidic chambers (Xona microfluidics, uLP) were cleaned with Ethanol and coverslips coated overnight with a poly-omithine solution (0,1 mg/ml in PBS). Before bonding, poly -ornithine from coverslips and ethanol from chambers were removed by thoroughly washing with distilled water. Bonding was performed by placing the microfluidic device over the dry coverslip and applying a uniform pressure on the upper surface of the microfluidic chamber, the process being detailed in the manufacturer’s instructions. After assembling the reservoirs of the microfluidic chamber are filled with PBS, avoiding to generate any bubble in the channel system. PBS is removed (not completely to not generate bubbles) and Matrigel is added in all of the reservoirs to completely coat all of the chambers surfaces. Chambers are maintained in incubator at 37°C for at least 1 hour.

Seeding and growth of neurons on chambers

NPCs obtained from previous steps were seeded in the somatic compartment. NPCs derived from iPSCs at the neurosphere stage were enzymatically digested with 500 ul of trypsin-EDTA [0,25%] and Accutase (1:1) for 3 min at 37°C. The reaction was ended by the addition of 500ul of Soybean Tripsin inhibitor (0.5 mg/ml) for 3 min at 37°C and DMEM (500 ul). Cells were washed by centrifugation twice (250 g for 2 min), discarding the media and adding DMEM again. After the second centrifugation, pellet were suspended in Neuron Differentiation Media (NDM composed of DMEM/F12, 1% v/v N2 supplement, 2% B27 v/v supplement, and 1% v/v penicillin-streptomycin, laminin (1 ug/ml), cAMP (100 nM), ascorbic acid (200 ng/ml), BDNF (10 ng/ml), GDNF (10 ng/ml), IGF (10 ng/ml)) and mechanically dissociated by pipetting up and down with pi 000 for approximately 20 times. Cells were counted and 50 ul of a cell suspension of 500.000 cells/100 ul was pipetted in the upper reservoir of the somatic compartment of the chamber.

NPCs derived from NSCs were digested with 5ml of Accutase solution for 5 min in incubator. 5 ml DMEM is added and cells are lifted and collected in a falcon tube, counted and centrifuged at 250 g for 5 min. Media was discarded and cells resuspended in NDM in a volume to obtain 500.000 cells/100 ul. As in previous case, 50 ul of cells suspension was seeded in the upper reservoir of somatic compartment of the microfluidic chamber. The seeding procedure is done according to the microfluidic chamber manufacturer protocol (Xona). After seeding cells are kept in incubator for 30 min to allow flow equilibration and attachment on the somatic compartment. This is followed by adding NDM in all of the reservoirs and importantly it should be ensured that the volume of media in the somatic compartment is approximately 100 ul higher than in the opposing axonal compartment.

During neuronal growth media is balanced (withdrawing media from axonal compartment and adding in the somatic compartment) every three days and completely changed every six days.

During the neuronal maturation axons start to grow in axonal channels and at DIV14 usually first axons start to appear in the axonal compartment. At later stages more axons cross to the axonal compartment.

Transduction of neurons with cytosolic and membrane fluorescent reporters

At least 5 days prior to the injury assay on axons, transduction is carried out to allow an abundant expression of the fluorescent protein in the axons. DIV 30 chambers are transduced with mCherry or GFP (Vectalys) lentiviral particles at a MOI of 10. Culture media from the somatic compartment is withdrawn and 200 ul of fresh NDM with the proper volume of virus is added to the upper reservoir. Chambers are kept in incubator over night or the equivalent to 12-18 hours. After incubation, media containing virus is withdrawn from somatic compartment and discarded. This compartment is refilled with fresh NDM media, and media change and balance is performed as previously described until the day of the injury. The resulting efficiency of transduction using this protocol is approximately 80%.

Axonal Injury

Syringe pump and tubing setting was assembled. The end of the tubing that connects to microfluidic chamber was a short segment (15 mm) of GE Healthcare PEEK tubing (green). 0,75 mm i.d.; 1/16 o.d. #18-1112-53, that was fitted inside a transparent small and flexible tube in a way that 5 mm of the green tube remains out of the transparent tube. By a series of adaptors (Transparent small, Transparent Big, Adaptors, Syringe to tube) the tubbing connects to the syringe (12 ml syringe, Chirana), placed in an electronic syringe pump (New Era Pump Systems Inc. NE-1002X), which controls the pressure of the syringe and ultimately controls the flow through the channel. This setting is mounted on the confocal microscope (Zeiss Confocal LSM780, Zeiss Live LSM7) and allows real time visualization of axonal transport by taking advantage of fluorescently labeled surrogate markers of axonal transport.

Before starting, the microscope stage was pre-heated to 37 °C and 5% CO2. The syringe and the tubing were loaded with PBS in a way that liquid continuity was not interrupted by any bubble. If needed, the tubing was purged so no bubble remains. All the reservoirs of the microfluidic chamber are topped up with growth media, to avoid any compartment to get dry during the experiment. The system was kept in the incubator until temperature was equilibrated.

The microfluidic chamber was placed in the microscope stage with the coverslip side down and the tubing from the syringe pump inserted into the opening of the flow channel. With 63X objective and brightfield the desired position of axons crossing the flow channel is located. For live imaging purposes location of axons that express the membrane marker (mCherry) or cytosolic marker (GFP) is perform through a Mercury lamp and the respective filtersets for these fluorophores. In the software that controls the microscope a time lapse movie is recorded with suitable pre-defmed settings (i.e. total time: 420 sec, frame rate: 1 FPS, exposure time: 100 msec, laser: 488 nm (GFP) or 560 nm (mCherry)). In our example the movie consisted of 60 secs of control flow speed (33 mm/sec), 90 secs of high flow speed (333 mm/sec), 20 secs of reversion (inverse flow) (-333 mm/sec) and remaining time at control flow speed (33 mm/sec). Several images from the time lapse movie are presented in Figure 5.

In the syringe pump, the flow rate was set to 10 pi /min (which corresponds to a flow speed of 33 mm/sec in the flow channel) in withdrawal direction. One minute is enough to stabilize the flow through the channel. Longer times than 3 minutes are not recommended due to a progressive decrease of the media levels from the deposits of the chamber.

For the injury, the flow rate and the duration were set to 100 mΐ/min (which corresponds to a flow speed of 333 mm/sec in the flow channel) for 90 seconds in withdrawal direction. After this period, the same flow rate of 100 mΐ/min but inverse flow is applied for 20 seconds to bring the flow to normal speeds. A flow rate of 10 mΐ/min (which corresponds to a flow speed of 33 mm/sec in the flow channel) is then set for the time needed for post injury observation (280 secs in our example). Once the recording was finished, the tubing was separated from the chamber. After the injury, cultures can be fixed and microfluidic chambers detached gently from the coverslip for further immunofluorescence assays, or the chamber can be restored to the incubator.

Enlargements were not detected during the first minute of control flow, while several appear during the high flow rate (arrows in figure 6).

Results

To characterize the hydrodynamics of the flow in the injury channel, fluorescent beads were loaded in the chamber and different pump flow speeds were selected in the syringe pump (30, 50, 100, 150 and 200 ul/min). Starting from basal flow of 10 ul/min, movies were recorded in the channel for 90 seconds. On the movies, the speed of 10 beads in intervals of 10 seconds was tracked for each pump flow (n=3). The plot in Figure 2 shows the mean measured speed for each flow rate during time. The maximum speed measured for each flow rate is presented in Table 1.

Table 1 To measure the damage on neurons, mature neurons were transduced with a construct expressing a plasma membrane directed Cherry. Stained axons were subjected to different pump flows for 90 seconds and live-imaging movies recorded. Figure 3 presents a plot showing the decrease of intact axons and increase of enlargement formation with increasing pump flows (n=31 chambers). It was found that in the herein used setting, pump flow rates higher than 100 ul/min (which corresponds to a flow speed of 333 mm/sec in the flow channel) for interval equal or longer than 90 secs cause axotomy in the majority of the axons that cross the flow channel (Figure 4).

In the system of the present invention, and using the assay proposed in the present invention, changes on the axon morphology due to 333 mm/sec flow speed during 90 seconds can be observed by the formation of enlargements (marked with arrows in figure 5) in the axonal membrane or accumulation of cytosolic fluorophore in specific positions during and after injury. This is shown in Figures 5 and 6. The graph in Figure 7 summarizes the maximum number of enlargements during injury versus the control setting. Three movies were analyzed per setting. The mean number of enlargements visualized with the cytosolic GFP was 13 ± 4.3 and 9.3 ± 1.76 with the membrane mCherry.

As an example of how drugs can modulate the injury outcome, neurons transduced with membrane mCherry were incubated with ryanodine (100 uM for 40 min) (a drug that blocks Ryanodine Calcium channels in the Endoplasmic Reticulum) previously to perform the injury. In addition, medium used during injury (333 mm/sec for 90 secs) was depleted of Calcium (aCSF 0.2 mM Ca++). As shown in Figure 8 the number of enlargements in the treatment with Ryanodine and low Ca++ is significantly lower than the non-treated. Bars show the mean ±SEM of at least 5 independent experiments.

Neurons were incubated with fluorescent probes (e.g. MitoTracker, MitoSOX, ERTracker, LisoTracker, Fluo 4Am, etc, Invitrogen) that label different components of neurons prior to injury, to assess their changes through injury. All these labels are compatible with the present invention.