Patent Application
20240252726
Parkinson's disease (PD) is a disorder characterized by motor deficits, such as tremors, slowness of movement (bradykinesia), and rigidity, as well as non-motor symptoms including autonomic dysfunction, depression, and dementia. PD is the second most common neurodegenerative disease, affecting the quality of life of 10 million people worldwide, including 60,000 new cases per year in the United States and 1-2% of people 65 years or older. Projections indicate that the number of PD patients in the ten most populated countries will double by 2030 to 9.3 million. The Parkinson's Foundation has estimated that the total cost associated with treatment in the United States is $52 billion/year, including $2,500/year for medications and $100,000 for surgery per patient. Existing therapies can mitigate symptoms for a limited time, but there are currently no approved treatments to prevent or repair the underlying pathology. The search for an effective treatment is thus a crucial research endeavor. A key aspect of PD pathology is the loss of dopaminergic neurons in the substantia nigra pars compacta (SNpc) and the degeneration of their axonal projections along the nigrostriatal pathway. These dopaminergic axons synapse with medium spiny neurons (MSNs) in the dorsolateral striatum, a brain region associated with motor movements, decision making, and learning. In healthy patients, activated dopaminergic neurons release dopamine, which acts on MSNs expressing dopamine receptors. Through the direct and indirect pathways involving dopamine receptor 1 and 2, respectively, these synaptic connections modulate the gamma aminobutyric acid (GABA)-dependent inhibitory output of the basal ganglia (i.e., globus pallidus internus and the substantia nigra pars reticulata (SNpr)). With the loss of dopamine released by nigrostriatal terminals in PD, the net effect is less suppression of the basal ganglia by the striatum, increased inhibition of the thalamus, and reduced activation of the motor cortex. A loss of ˜30-50% of dopaminergic neurons and a reduction of 50-60% in striatal dopamine has been associated with the onset of symptoms. Moreover, axon degeneration on its own, years before the death of neuron bodies, has been interpreted as an early sign of PD.
Unfortunately, most patients are diagnosed when motor symptoms are evident and significant disease progression has already occurred. Current treatments focus on managing symptoms with drug therapy or deep brain stimulation (DBS). On the first front, the gold standard has been administration of L-3,4-dihydroxyphenylalanine (L-DOPA), the immediate precursor to dopamine, which has a short half-life and variable absorption and blood-brain barrier transport that prevent having sustained drug delivery. Possibly because of this discontinuous stimulation, long-term patients tend to suffer abnormal muscle movement (dyskinesia) and fluctuations between symptom control and recurrence. As alternatives, dopamine agonists (e.g., apomorphine, rotigotine) mimic the normal interaction of dopamine with receptors in the striatum and tend to have a longer half-life and less pulsatile effects than L-DOPA, but with less potency. On the other hand, DBS typically involves the implantation of electrodes to stimulate the subthalamic nucleus, which can improve motor scores and reduce dependence on drugs. Regardless, electrical stimulation is inherently non-specific, the mechanisms of DBS are unknown, and the effectiveness of electrodes can be affected by chronic foreign body responses. DBS patients may also be subject to bleeding, infection, and psychiatric symptoms. Current treatments also fail to intervene in the death of dopaminergic neurons and the loss of striatal innervation at the core of PD. There is a need in the art for new treatments for Parkinson's disease and other conditions of the nervous system that better satisfy patient needs. The present disclosure addresses this need.
Among other things, the present disclosure provides a construct comprising a pre-formed neural network, the construct comprising: a micro-column comprising an outer sheath comprising a hyaluronic acid (HA) hydrogel, and a core comprising an extracellular matrix (ECM); and a plurality of neurons within the micro-column.
In various embodiments, the construct is biocompatible.
In various embodiments, the construct is an implantable construct.
In various embodiments, the hydrogel sheath is cylindrical.
In various embodiments, the ECM core substantially fills a lumen of the hydrogel sheath.
In various embodiments, the micro-column is directed along a substantially straight line along its length.
In various embodiments, the micro-column is directed along a curved path along its length.
In various embodiments, the plurality of neurons have cell bodies substantially localized in proximity to a first end of the micro-column and extend axons longitudinally along at least a portion of a length of the micro-column.
In various embodiments, the plurality of neurons comprise one or more three dimensional aggregates.
In various embodiments, the axons are located within and extend longitudinally along a lumen of the hydrogel sheath from the neurons at the first end and towards the opposite end.
In various embodiments, the axons grow through the ECM of the core and/or along an interface between an inner surface of the hydrogel sheath and the ECM of the core.
In various embodiments, the neurons and axons extending therefrom have a cyto-architecture that replicates long-range axon tracts present in a subject.
In various embodiments, the subject is a human subject.
In various embodiments, the neuronal cells and axons extending therefrom have a cyto-architecture that replicates long-range axon tracts present in a brain of a human subject.
In various embodiments, the neuronal cells and axons extending therefrom have a cyto-architecture that mimics a native axon pathway between the substantia nigra and the striatum in a brain of a subject.
In various embodiments, the axon tracts are or have been pre-directed via the micro-column.
In various embodiments, the neurons together with the axons extending therefrom form a biofabricated micro-tissue.
In various embodiments, the axons of the plurality of neurons extend along at least 50% of the length of the micro-column.
In various embodiments, the axons extend along at least 75% of the length of the micro-column.
In various embodiments, the axons extend along 90% of the length of the micro-column.
In various embodiments, an outer diameter of the micro-column ranges from about 500 microns to about 2,500 microns.
In various embodiments, the outer diameter of the micro-column ranges from about 500 to about 1,500 microns.
In various embodiments, the outer diameter of the micro-column ranges from about 750 to about 1,000 microns.
In various embodiments, the outer diameter is a cross-sectional diameter of the hydrogel sheath.
In various embodiments, the outer diameter is a cross-sectional diameter of the hydrogel sheath and including any outer coatings thereon.
In various embodiments, an inner diameter of the micro-column ranges from about 250 microns to about 2,000 microns.
In various embodiments, the inner diameter of the micro-column ranges from about 250 to about 1,000 microns.
In various embodiments, the inner diameter of the micro-column is about 500 microns.
In various embodiments, the ECM comprises a polysaccharide.
In various embodiments, the ECM comprises one or more members selected from the group consisting of collagen, fibrin, fibronectin, gelatin, hyaluronic acid, laminin, and Matrigel.
In various embodiments, the ECM comprises collagen.
In various embodiments, the ECM comprises collagen at a concentration ranging from about 0.1 to 10 mg/ml.
In various embodiments, the ECM comprises collagen at a concentration of about 1 mg/ml.
In various embodiments, the ECM comprises laminin.
In various embodiments, the ECM comprises laminin at a concentration ranging from about 0.1 to 10 mg/ml.
In various embodiments, the ECM comprises laminin at a concentration of about 1 mg/ml.
In various embodiments, the hyaluronic acid (HA) hydrogel is or comprises a cross-linked modified hyaluronic acid.
In various embodiments, the modified hyaluronic acid is methacrylated HA (MeHA).
In various embodiments, the hyaluronic acid comprises about 0.5 to about 20% wt MeHA.
In various embodiments, the hyaluronic acid is or comprises 3-5% MeHA.
In various embodiments, the modified HA comprises one or more members selected from the group consisting of norbornene-modified HA, acrylated HA, maleimide HA, and hydroxyethyl methacrylate HA.
In various embodiments, the outer hydrogel sheath is a 3D printed and photopolymerized MeHA cylinder.
In various embodiments, the outer hydrogel sheath comprises one or more hydrolysis sensitive compounds within crosslinks.
In various embodiments, the one or more hydrolysis sensitive compounds comprise esters.
In various embodiments, the one or more hydrolysis sensitive compounds comprise lactic acid, caprolactone or an anhydride.
In various embodiments, the outer hydrogel sheath comprises methacrylated hyaluronic acid doped with the one or more hydrolysis sensitive compounds.
In various embodiments, one or more hydrolysis sensitive compounds are located between HA and the methacrylate groups.
In various embodiments, the outer hydrogel sheath comprises one or more di-thiol peptides.
In various embodiments, at least a portion of the one or more di-thiol peptides are sensitive to cleavage.
In various embodiments, at least a portion of the one or more di-thiol peptides are sensitive to cleavage by matrix metalloproteinases expressed by cells.
In various embodiments, the plurality of neurons comprise dopaminergic neurons.
In various embodiments, at least 50% of the plurality of neurons are dopaminergic neurons.
In various embodiments, the dopaminergic neurons are obtained via purification.
In various embodiments, the dopaminergic neurons comprise midbrain dopaminergic neurons.
In various embodiments, the midbrain dopaminergic neurons comprise A9 neurons.
In various embodiments, the plurality of neurons comprise GABAergic neurons.
In various embodiments, at least 50% of the plurality of neurons are GABAergic neurons.
In various embodiments, the GABAergic neurons are obtained via purification.
In various embodiments, the plurality of neurons comprise glutaminergic neurons.
In various embodiments, at least 50% of the plurality of neurons are glutaminergic neurons.
In various embodiments, the glutaminergic neurons are obtained via purification.
In various embodiments, the plurality of neurons comprise cholinergic neurons.
In various embodiments, at least 50% of the plurality of neurons are cholinergic neurons.
In various embodiments, the cholinergic neurons are obtained via purification.
In various embodiments, the plurality of neurons comprise midbrain neurons.
In various embodiments, the plurality of neurons comprise human neurons.
In various embodiments, the human neurons comprise induced pluripotent stem cell (iPSC)-derived neurons.
In various embodiments, the plurality of neurons comprise human A9 dopaminergic neurons.
In various embodiments, the human A9 dopaminergic neurons comprise stem cell-derived human A9 dopaminergic neurons.
In various embodiments, the stem cells from which the A9 dopaminergic neurons are derived are induced pluripotent stem cells (iPSCs).
In various embodiments, the plurality of neurons comprise at least 50,000 neurons.
In various embodiments, the plurality of neurons comprise at least 100,000 neurons.
In various embodiments, the plurality of neurons comprise at least 125,000 neurons.
In various embodiments, the plurality of neurons comprise at least 500,000 neurons.
In various embodiments, the plurality of neurons comprise at least 1,000,000 neurons.
In various embodiments, the micro-column has a length ranging from about 2 to about 5 centimeters.
In various embodiments, axons of the plurality of neurons extend a distance ranging from about 2 to about 5 centimeters along the length of the micro-column.
In various embodiments, the micro-column is adapted for implantation along a trajectory encompassing a substantia nigra (SN) region and a striatum region of a subject.
In various embodiments, the SN region is a ventrolateral SN region.
In various embodiments, the striatum region is a dorsal striatum region.
In various embodiments, the subject is a human subject.
In various embodiments, neurons comprise A9 dopaminergic neurons and exhibit dopamine release of at least 50 nM as measured by fast scan cyclic voltammetry.
In various embodiments, the plurality of neurons releases and/or has a quantity of dopaminergic neurons sufficient to release dopamine at a level sufficient to provide for a level of at least 4 ng/mg in tissue.
In various embodiments, the dopaminergic neurons release dopamine at a level sufficient to provide for a level of at least 4 ng/mg in tissue within six weeks after implantation in a subject.
In various embodiments, the plurality of neurons provides for increase in 18F-DOPA uptake in a putamen of a subject at a level of about 50-60% of a normal value.
In various embodiments, the neurons provide for an increase in 18F-DOPA uptake in the putamen of the subject at the level of about 50-60% of a normal value upon implantation in the subject.
In various embodiments, an increase at the level of about 50-60% of a normal value is achieved.
In another aspect, the present disclosure provides a method of manufacturing a construct comprising a pre-formed neural network, the method comprising:
In various embodiments, the construct is a biocompatible construct.
In various embodiments, the construct is an implantable construct.
In various embodiments, the construct is for use in an in-vitro test bed.
In various embodiments, step (b) comprises causing growth of axons from the neural cells, along a length of the micro-column, toward a second, opposite, end of the micro-column.
In various embodiments, the method comprises:
In various embodiments, the particular length is a predetermined desired length.
In various embodiments, the particular length ranges from about 2 to about 5 centimeters.
In various embodiments, step (c) comprises imaging the micro-columns and neural cells therein.
In various embodiments, step (c) comprises imaging via microscopy, fast-scan cyclic voltammetry (FSCV), staining, sectioning or measuring axon density.
In various embodiments, the plurality of neural cells with which the micro-column is seeded at step (a) comprise neural cell aggregates.
In various embodiments, the neural cell aggregates comprise a plurality of approximately spherical aggregates of neural cells.
In various embodiments, each neural cell aggregate comprises cells at a density ranging from about 100,000 to about 300,000 neurons per aggregate.
In various embodiments, a plurality of the neural cell aggregates exhibit a diameter of at least 500 μm.
In various embodiments, the micro-column comprises a hydrogel sheath and a core comprising an extracellular matrix (ECM), and wherein the neural cells are seeded to be in direct contact with the ECM of the core.
In various embodiments, the hydrogel sheath comprises MeHA.
In various embodiments, the hydrogel sheath of the micro-column is a 3D printed cylinder.
In various embodiments, the method comprises 3D printing the hydrogel sheath prior to step (a).
In various embodiments, the method comprises differentiating human induced pluripotent stem cells (iPSCs) for a particular differentiation period prior to step (a), thereby producing differentiated cells and, following differentiating the iPSCs for the particular differentiation period, performing step (a) using the differentiated cells as the neural cells
In various embodiments, the method comprises seeding the differentiated iPSCs in the micro-column after about 40 dd.
In various embodiments, the method comprises seeding the differentiated iPSCs in the micro-column after about 11 to about 20 dd.
In various embodiments, the method comprises seeding the differentiated iPSCs in the micro-column once dopaminergic precursor fate is established and when the cells are usually replanted and matured further.
In various embodiments, the micro-column and or cells have one or more features articulated in the preceding embodiments.
In another aspect, the present disclosure provides an in-vitro test bed comprising: the construct as described herein, and further comprising a first population of neurons and axons grown therefrom; and a second population of neurons, synapsed with the first population.
In various embodiments, the second population of neurons comprise striatal neurons.
In various embodiments, the second population of neurons are seeded at an end of the construct opposite to the end at which the first population of neurons were seeded.
In various embodiments, axons from the first population of neurons extend longitudinally from a first end of the construct along a length of the construct and synapse with the second population seeded at a second, opposite, end of the construct.
In various embodiments, cell bodies of the first population are localized in substantial proximity to the first end of the construct.
In another aspect, the present disclosure provides a method of at least partially replacing a population of neurons forming a pathway between the substantia nigra and striatum in a subject, the method comprising implanting at least one construct as described herein into a brain of the subject.
In various embodiments, the method comprises ameliorating one or more conditions of the subject.
In various embodiments, ameliorating the one or more conditions comprises restoring motor function of the subject.
In various embodiments, the ameliorating the one or more conditions comprises reducing pain of the subject.
In various embodiments, the ameliorating the one or more conditions comprises reducing tremors of the subject.
In various embodiments, the method comprises implanting the at least a portion of one construct within a substantia nigra of the subject.
In various embodiments, following implantation, the neurons of the construct synapse with host neurons in a brain of the subject.
In various embodiments, the host neurons with which the neurons of the construct synapse comprise, medium spiny neurons (MSNs) in a dorsolateral striatum of the subject.
In various embodiments, the subject is a human subject.
In various embodiments, implanting the at least one construct comprises using MRI-guided neurosurgery.
In various embodiments, implanting the at least one construct comprises implanting a plurality of constructs.
In various embodiments, the implanting the plurality of constructs comprises implanting a plurality of constructs in a single hemisphere of the brain of the subject.
In various embodiments, implanting the plurality of constructs comprises implanting one or more constructs in each hemisphere of the brain of the subject.
In various embodiments, the method comprises implanting 1 to 3 constructs in each hemisphere of the brain of the subject.
In various embodiments, implantation in a first hemisphere is performed via a first surgery and implantation in a second hemisphere is performed via a second surgery, performed at a different time than the first surgery.
In various embodiments, the second surgery is performed about 6 months after the first surgery.
For a fuller understanding of the nature and desired objects of the present invention, reference is made to the following detailed description taken in conjunction with the accompanying drawing figures wherein like reference characters denote corresponding parts throughout the several views.
The instant invention is most clearly understood with reference to the following definitions.
As used herein, the singular form “a,” “an,” and “the” include plural references unless the context clearly dictates otherwise.
Unless specifically stated or obvious from context, as used herein, the term “about” is understood as within a range of normal tolerance in the art, for example within 2 standard deviations of the mean. “About” can be understood as within 10%, 9%, 8%, 7%, 6%, 5%, 4%, 3%, 2%, 1%, 0.5%, 0.1%, 0.05%, or 0.01% of the stated value. Unless otherwise clear from context, all numerical values provided herein are modified by the term about.
As used in the specification and claims, the terms “comprises,” “comprising,” “containing,” “having,” and the like can have the meaning ascribed to them in U.S. patent law and can mean “includes,” “including,” and the like.
As used herein, the term “biofabricated” refers to the process of creating, building, and/or assembling as related to biological, biocompatible, and/or biologically inspired components
As used herein, the term “cylinder” or “cylindrical” includes a surface consisting of each of the straight lines that are parallel to a given straight line and pass through a given curve. In some embodiments, cylinders have an annular profile. In other embodiments, the cylinder has a cross-section selected from the group consisting of: a square, a rectangle, a triangle, an oval, a polygon, a parallelogram, a rhombus, an annulus, a crescent, a semicircle, an ellipse, a super ellipse, a deltoid, and the like. In other embodiments, the cylinder is the starting point of a more complex three-dimensional structure that can include, for example, complex involutions, spirals, branching patterns, multiple tubular conduits, and any number of geometries that can be implemented in computer-aided design, 3-D printing, and/or in directed evolutionary approaches of secretory organisms (e.g., coral), including of various fractal orders.
As used herein, the term “micro-tissue” means a three-dimensional construct containing some combination of cells and biomaterials that has at least one dimension measuring hundreds to thousands of microns.
As used herein, the term “modified hyaluronic acid” means hyaluronic acid that has been chemically modified to at least a fraction of the disaccharide repeat units, such as, without limitation, to allow for crosslinking of the hyaluronic acid into a hydrogel.
As used herein, “synapse” refers to a junction between a neuron and another cell, across which chemical communication flows.
As used herein, “synapsed” refers to a neuron that has formed one or more synapses with one or more cells, such as another neuron or a muscle cell.
Unless specifically stated or obvious from context, the term “or,” as used herein, is understood to be inclusive.
Ranges provided herein are understood to be shorthand for all of the values within the range. For example, a range of 1 to 50 is understood to include any number, combination of numbers, or sub-range from the group consisting 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, or 50 (as well as fractions thereof unless the context clearly dictates otherwise).
In one aspect, the present disclosure provides a construct comprising a pre-formed neural network, the construct comprising a micro-column comprising an outer sheath comprising a hyaluronic acid (HA) hydrogel, and a core comprising an extracellular matrix (ECM); a plurality of neurons within the micro-column. In various embodiments, the construct is biocompatible. In various embodiments, the construct is an implantable construct. In various embodiments, the hydrogel sheath is cylindrical or substantially cylindrical. In various embodiments, the ECM core substantially fills a lumen of the hydrogel sheath. In various embodiments, the micro-column is directed along a substantially straight line along its length. In various embodiments, the micro-column is directed along a curved path along its length.
In various embodiments, the plurality of neurons have cell bodies substantially localized in proximity to a first end of the micro-column and extend axons longitudinally along at least a portion of a length of the micro-column. In various embodiments, the plurality of neurons comprise one or more three dimensional aggregates. In various embodiments, the axons are located within and extend longitudinally along a lumen of the hydrogel sheath from the neurons at the first end and towards the opposite end, extending in some embodiments along the entire length of the construct and terminating at the end opposite the first end and in other embodiments extending a portion of the distance between the first end and the opposite end. In various embodiments, the axons of the plurality of neurons extend along at least 50% of the length of the micro-column, extend along at least 75% of the length of the micro-column, or extend along 90% of the length of the micro-column. In various embodiments, the axons grow through the ECM of the core and/or along an interface between an inner surface of the hydrogel sheath and the ECM of the core. As described further in Examples 1 and 2, in some embodiments the inner surface of the hydrogel sheath provides a substrate that facilitates axon extension through the ECM lumen. In various embodiments, the neuronal cells and axons extending therefrom have a cyto-architecture that replicates long-range axon tracts present in a subject. In various embodiments, the subject is a mammal. In various embodiments, the subject is a human subject. In various embodiments, the neuronal cells and axons extending therefrom have a cyto-architecture that replicates long-range axon tracts present in a brain of a human subject. In various embodiments, the neuronal cells and axons extending therefrom have a cyto-architecture that mimics a native axon pathway between the substantia nigra and the striatum in a brain of a subject. In various embodiments, the axon tracts are or have been pre-directed via the micro-column. In various embodiments, the neurons together with the axons extending therefrom form a biofabricated micro-tissue.
In various embodiments, the outer diameter of the micro-column ranges from about 500 microns to about 2,500 microns. In various embodiments, the outer diameter of the micro-column ranges from about 500 to about 1,500 microns. In various embodiments, the outer diameter of the micro-column ranges from about 750 to about 1,000 microns. In various embodiments, the outer diameter is a cross-sectional diameter of the hydrogel sheath. In various embodiments, the outer diameter is a cross-sectional diameter of the hydrogel sheath, including any outer coatings thereon. In various embodiments, the inner diameter of the micro-column ranges from about 250 microns to about 2,000 microns. In various embodiments, the inner diameter of the micro-column ranges from about 250 to about 1,000 microns. In various embodiments, the inner diameter of the micro-column is about 500 microns. In various embodiments, the ECM comprises a polysaccharide. In various embodiments, the ECM comprises one or more members selected from the group consisting of collagen, fibrin, fibronectin, gelatin, hyaluronic acid, laminin, and Matrigel. In various embodiments, the ECM comprises collagen. In various embodiments, the ECM comprises collagen at a concentration ranging from about 0.1 to 10 mg/ml. In various embodiments, the ECM comprises collagen at a concentration about 1 mg/ml. In various embodiments, the ECM comprises laminin. In various embodiments, the ECM comprises laminin at a concentration ranging from about 0.1 to 10 mg/ml. In various embodiments, the ECM comprises laminin at a concentration of about 1 mg/ml.
In various embodiments, the hyaluronic acid (HA) hydrogel present in various embodiments of the construct is or comprises a modified hyaluronic acid (e.g., a cross-linked modified hyaluronic acid). In various embodiments, the modified hyaluronic acid is methacrylated HA (MeHA). In various embodiments, the hyaluronic acid is or comprises about 0.5 to about 20% wt MeHA. In various embodiments, the hyaluronic acid is or comprises 3-5% MeHA. In various embodiments, the modified HA comprises one or more members selected from the group consisting of norbornene-modified HA, acrylated HA, maleimide HA, and hydroxyethyl methacrylate HA. In various embodiments, the outer hydrogel sheath is a 3D printed and photopolymerized MeHA cylinder.
MeHA hydrogels may be further optimized to have the combination of molecular weight, macromer concentration, degree of methacrylation, and light irradiation time to yield the desired mechanical properties. The molecular weight of HA can be selected before the methacrylation process by using the HA with the desired weight as one of the reactants. In various embodiments, the HA acid has a molecular weight in the range between about 50 kDa and about 1.5 MDa. In various embodiments, the HA has a molecular weight of 50 kDa, 60 kDa, 70 kDa, 80 kDa, 90 kDa, 1 MDa, 1.1 MDa, 1.2 MDa, 1.3 MDa, 1.4 MDa, or 1.5 MDa.
The degree of methacrylation can be changed by decreasing or increasing the duration of the reaction between HA and methacrylic anhydride. In various embodiments, the degree of HA methacrylation is about 40%. In various embodiments, the degree of HA methacrylation is between about 25% and 100%.
The macromer concentration refers to the percentage of MeHA in the hydrogel solution that is created before photopolymerization. In various embodiments, the concentration of MeHA is between about 1% and about 10%. In various embodiments, the concentration of MeHA is between about 3% and about 5%.
In some embodiments, properties of the hydrogel change as a function of time while exposed to the ultraviolet light. Therefore, in various embodiments, the exposure time is varied to modulate the properties of the hydrogel. In various embodiments, the hydrogel is exposed to ultraviolet light until all available crosslinks are formed. In various embodiments, the exposure time is about 5 min. In various embodiments, the exposure time is about 1 minute, about 2 minutes, about 3 minutes or about 4 minutes. Overall, higher molecular weights, degrees of methacrylation, macromer concentrations, and light irradiation times result in MeHA hydrogels that are stiffer and slower to degrade in the presence of hyaluronidases. In addition to MeHA, other types of modified HA can be used to confer other crosslinking behaviors and degradation properties. For instance, HA hydrogels can be made with increased degradation through hydrolysis or through matrix metalloproteinase degradation with peptide crosslinkers. Varying different types of crosslinkers within a hydrogel can also change properties. MeHA hydrogels crosslinked by radical photopolymerization are mainly degradable by enzymes known as hyaluronidases. When implanted, the environment in which the hydrogel is found may not have the hyaluronidases to the extent required for meaningful degradation to occur. Other types of modified HA may be used to connect the HA chains with modified methacrylates or other chemical groups to make the hydrogel susceptible to ester hydrolysis or with enzymatically degradable crosslinker groups. These two would allow for significant degradation to occur, which can be tuned similarly as explained above by tuning the crosslinking density of the hydrogel to get slower or faster degradation. In various embodiments, the outer hydrogel sheath comprises one or more hydrolysis sensitive compounds within the crosslinks. In various embodiments, the one or more hydrolysis sensitive compounds comprise esters such as lactic acid, caprolactone or anhydrides. In various embodiments, hydrolysis sensitive moieties are incorporated into the crosslinks. In various embodiments, at least about 50%, at least about 60%, at least about 70%, at least about 90%, at least about 95% of the crosslinks comprise hydrolysis sensitive moieties. In various embodiments, the outer hydrogel sheath comprises methacrylated hyaluronic acid doped with the one or more hydrolysis sensitive compounds. In various embodiments, the hydrolysis sensitive compounds are located between HA and the methacrylate groups. In various embodiments, the outer hydrogel sheath comprises one or more di-thiol peptides.
In various embodiments, the constructs present various biochemical cues to the host. In various embodiments, an addition reaction (non-light based) is performed to consume some but not all methacrylates of MeHA. In various embodiments, subsequent exposure to light consumes the remaining methacrylates and crosslinks the gel.
In various embodiments, the HA is norbornene-HA. In various embodiments, a protein or other thiol containing molecule is cross-linked to norbornene-HA by exposure to light. In various embodiments, the protein or other thiol containing molecule is linked to some but not all norbornenes. These molecules would be covalently bound to the HA backbone through the norbornene. In various embodiments, a crosslinker having di-thiols to crosslink norbornenes and form the hydrogel. NorHA is added. In various embodiments, at least a portion of the one or more di-thiol peptides are sensitive to cleavage. In various embodiments, at least a portion of the one or di-thiol peptides are sensitive to cleavage by matrix metalloproteinases expressed by cells. In various embodiments, the modified HA is maleimide or acrylate functionalized hyaluronic acid.
In various embodiments, growth factors and drugs can be loaded into the hydrogel sheath by mixing with the hydrogel solution during fabrication, passively by swelling the hydrogel in media containing those molecules, or by covalently binding those molecules to the HA chains. In this case, the molecules may be released during the normal degradation process of the hydrogel, with faster degradation rates leading to faster release. As explained above, there are several ways to tune the degradation rate of the HA in the construct and to make it susceptible to hyaluronidases, hydrolysis, and/or the action of enzymes released by cells. In other instances, modified HA has been covalently bound to structures containing drugs like dexamethasone that can be released later on in response to a trigger such as mechanical stress. HA has been investigated for light-triggered drug release, where nanoparticles release drugs after irradiation with near-infrared light leads to hydrolysis of susceptible bonds in modified HA. Modified HA has also been covalently bound to heparin to allow for the retention and slow release of growth factors that have heparin-binding domains. In this case, thiolated heparin can be bound to the chemical groups added to HA, and the release profile can be tuned according to the concentration of heparin and its molecular weight.
In various embodiments, the plurality of neurons comprises dopaminergic neurons. In various embodiments, at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80% or at least 90% of the plurality of neurons are dopaminergic neurons. In various embodiments, the dopaminergic neurons are obtained via microdissection. In various embodiments, the dopaminergic neurons are obtained via differentiation from stem cells. In various embodiments, the dopaminergic neurons are obtained via cell sorting or purification. In various embodiments, the dopaminergic neurons comprise midbrain dopaminergic neurons. In various embodiments, the midbrain dopaminergic neurons comprise A9 neurons.
In various embodiments, the plurality of neurons comprise GABAergic neurons. In various embodiments, at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80% or at least 90% are GABAergic neurons. In various embodiments, the GABAergic neurons are obtained via microdissection. In various embodiments, the GABAergic neurons are obtained via differentiation from stem cells. In various embodiments, the GABAergic neurons are obtained via cell sorting or purification.
In various embodiments, the plurality of neurons comprise glutaminergic neurons. In various embodiments, at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80% or at least 90% of the plurality of neurons are glutaminergic neurons. In various embodiments, the glutaminergic neurons are obtained via microdissection. In various embodiments, the glutaminergic neurons are obtained via differentiation from stem cells. In various embodiments, the glutaminergic neurons are obtained via cell sorting or purification.
In various embodiments, the plurality of neurons comprise cholinergic neurons. In various embodiments, at least 10%, at least 20%, at least 30%, at least 40%, at least 50%, at least 60%, at least 70%, at least 80% or at least 90% of the plurality of neurons are cholinergic neurons. In various embodiments, the cholinergic neurons are obtained via microdissection. In various embodiments, the cholinergic neurons are obtained via differentiation from stem cells. In various embodiments, the cholinergic neurons are obtained via cell sorting or purification. In various embodiments, the plurality of neurons, whether dopaminergic, GABAergic, glutaminergic or cholinergic, comprise midbrain neurons.
In various embodiments, the plurality of neurons comprise human neurons. In various embodiments, the human neurons comprise induced pluripotent stem cell (iPSC)-derived neurons. In various embodiments, the plurality of neurons comprise human A9 dopaminergic neurons. In various embodiments, the human A9 dopaminergic neurons comprise stem cell-derived human A9 dopaminergic neurons. The construct of any one of the preceding claims, wherein the neurons comprise A9 dopaminergic neurons and exhibit dopamine release of at least 50 nM as measured via fast scan cyclic voltammetry. In various embodiments, the plurality of neurons releases and/or has a quantity of dopaminergic neurons sufficient to release dopamine at a level sufficient to provide for a level of at least 4 ng/mg in tissue. In various embodiments, the dopaminergic neurons release dopamine at a level sufficient to provide for a level of at least 4 ng/mg in tissue within two weeks, within four weeks or within six weeks after implantation in a subject. In various embodiments, the plurality of neurons provides for an increase in 18F-DOPA uptake in a putamen of a subject at a level of about 50-60% or of about 80-100% of a normal value, in various embodiments within about 4, about 6 or about 8 weeks following implantation.
In various embodiments, neurons of constructs described herein provide for increase in 18F-DOPA uptake in the putamen of the subject at the level of about 50-60% of a normal value upon implantation in the subject.
In various embodiments, the stem cells from which the A9 dopaminergic neurons are derived are induced pluripotent stem cells (iPSCs). In various embodiments, the plurality of neurons comprises at least 50,000 neurons. In various embodiments, the plurality of neurons comprises at least 100,000 neurons. In various embodiments, the plurality of neurons comprises at least 125,000 neurons. In various embodiments, the plurality of neurons comprises at least 500,000 neurons. In various embodiments, the plurality of neurons comprises at least 1,000,000 neurons.
In various embodiments, the micro-column has a length ranging from about 2 to about 5 centimeters. In various embodiments, the axons of the plurality of neurons extend a distance ranging from about 2 to about 5 centimeters along the length of the micro-column. In various embodiments, the micro-column is suitable for implantation along a trajectory encompassing a substantia nigra (SN) region and a striatum region of a subject. In various embodiments, the SN region is a ventrolateral SN region. In various embodiments, the striatum region is a dorsal striatum region. In various embodiments, the subject is a mammal. In various embodiments, the subject is a human subject.
In another aspect, the present disclosure provides a method of manufacturing a construct comprising a pre-formed neural network, the method comprising: seeding a first end of a micro-column with a plurality of neural precursor cells and/or neurons selected from the group consisting of dopaminergic, GABAergic, glutaminergic or cholinergic neuronal precursors or neurons; and culturing the micro-column and plurality of neural cells seeded therein in-vitro. In various embodiments, the construct is a biocompatible construct. In various embodiments, the construct is an implantable construct. In various embodiments, the construct is for use in an in-vitro test bed. In various embodiments, the step of culturing the micro-column and plurality of neural cells seeded therein in-vitro comprises causing growth of axons from the neural cells, along a length of the micro-column, toward a second, opposite, end of the micro-column.
In various embodiments, the method further comprises determining that axon growth from the plurality of neural cells has reached a particular length; and responsive to the particular length of axon growth being determined to have been reached, packaging (e.g., in a sterile package) and/or providing the micro-column for implantation. In various embodiments, the particular length is a predetermined desired length. In various embodiments, the particular length ranges from about 2 to about 5 centimeters. In various embodiments, the step of determining that axon growth from the plurality of neural cells has reached a particular length comprises imaging the micro-columns and neural cells therein. In various embodiments, determining that axon growth from the plurality of neural cells has reached a particular length comprises imaging via fluorescence or other microscopy, fast-scan cyclic voltammetry (FSCV), staining, sectioning, or determining axon density. In various embodiments, the plurality of neural cells with which the micro-column is seeded comprises neural cell aggregates. In various embodiments, the neural cell aggregates comprise a plurality of approximately spherical aggregates of neural cells. In various embodiments, each neural cell aggregate comprises cells at a density ranging from about 100,000 to about 300,000 neurons per aggregate. In various embodiments, a plurality of the neural cell aggregates exhibit a diameter of at least 500 μm. In various embodiments, the micro-column comprises a hydrogel sheath and a core comprising an extracellular matrix (ECM), and wherein the neural cells are seeded to be in direct contact with the ECM of the core. In various embodiments, the hydrogel sheath comprises MeHA. In various embodiments, the hydrogel sheath of the micro-column is a 3D printed cylinder. In various embodiments, the method comprises, 3D printing the hydrogel sheath prior to seeding the first end of the micro-column. In various embodiments, the method comprises differentiating human induced pluripotent stem cells (iPSCs) for a particular differentiation period to produce differentiated cells and, following differentiating the iPSCs for the particular differentiation period, and seeding the micro-column using the differentiated cells as the neural cells. In various embodiments, the method comprises seeding the differentiated iPSCs in the micro column after about 40 dd. In various embodiments, the method comprises seeding the differentiated iPSCs in the micro-column after about 11 to about 20 dd. In various embodiments, the method comprises seeding the differentiated iPSCs in the micro-column once dopaminergic precursor fate is established and when the cells are usually replanted and matured further.
In various embodiments, the present disclosure provides an in vitro test bed comprising the construct as described herein, comprising a first population of neurons and axons grown therefrom; and a second population of neurons, synapsed with the first population. In various embodiments, the second population of neurons comprises or are striatal neurons. In various embodiments, the second population of neurons are seeded at an end of the construct opposite to the end at which the first population of neurons were seeded. In various embodiments, axons from the first population of neurons extend longitudinally from a first end of the construct along a length of the construct and synapse with the second population seeded at a second, opposite, end of the construct. In various embodiments, cell bodies of the first population are localized in substantial proximity to the first end of the construct.
In various embodiments, the present disclosure provides methods of at least partially replacing a population of neurons forming a pathway between the substantia nigra and striatum in a subject, the method comprising implanting at least one construct as described herein into a brain of the subject. Without meaning to be limited by theory, as illustrated in
MeHA was synthesized by the esterification of hyaluronic acid (Lifecore, molecular weight) and methacrylic anhydride (Sigma, 276685) in deionized water for ˜3.5 hr at pH 8.5 for a ˜40% degree of methacrylation. The reaction was purified by dialysis for 5-7 days, and the product was recovered by lyophilization. The degree of modification was assessed from the integration of the area below the two vinyl singlets at ˜5.8 and ˜6.3 ppm relative to the area representative of the sugar ring in 1H-NMR spectra of lyophilized MeHA. To study gelation kinetics, MeHA solutions were prepared in Dulbecco's phosphate buffered saline (DPBS; ThermoFisher, 14190136) with 0.05% weight/volume (w/v) Irgacure 2959 (12959; Sigma, 410896) photoinitiator. The solution was transferred between the fixed plate and the geometry of a controlled-stress rheometer (TA Instruments, AR2000) fitted to an ultraviolet (UV) lamp (Excelitas Technologies, OmniCure S1500). Using a time sweep with 0.5% strain and 1 Hz oscillations, the storage and loss moduli were recorded before and after exposure to 10 mW/cm2 UV light at 2 min to observe crosslinking in situ. The data presented are representative examples after taking at least duplicate profiles for each hydrogel. Hydrogels were also cast in ˜4.78 mm cylinders (n=5), and the compressive moduli were calculated from the slope of the stress vs. strain curve (10-20% region) using a dynamic mechanical analyzer (TA Instruments, Q800) at a 0.2 N/min loading rate.
MeHA micro-columns were created by drawing up MeHA solution (with 0.05% 12959 in DPBS) by capillary action into glass capillary tubes (Drummond Scientific; inner diameters: 398, 701, 973 μm) containing an inserted acupuncture needle (Lhasa OMS; outer diameters: 160, 300, 500 μm, respectively). The needle created the inner diameter (ID) or lumen of the micro-column while the remaining space formed the outer diameter (OD) or shell. The solution within each tube was then gelled with 10 mW/cm2 UV for 5 min, a time sufficient to obtain plateaued mechanical properties. The acupuncture needles were manually pulled from the tubes, leaving the gelled micro-columns inside. The hydrogels were removed by pushing them out of the tubes and into DPBS using 20, 23, or 30 gauge needles (BD, 305178, 305120, 305128), depending on the diameter of the tubes (
Rat dopaminergic neurons were isolated from embryonic day 14 (E14) rat pups as previously published. Pregnant Sprague Dawley rats (Charles River) were euthanized with carbon dioxide and decapitation, the pups were extracted, and the ventral midbrain was dissected from the brains in Hank's balanced salt solution (HBSS; ThermoFisher, 14175079). The tissue was rinsed, dissociated with accutase (ThermoFisher, A1110501) for 10 min at 37° C., and triturated with a pipette. After centrifugation at 200 g for 5 min, the cells were resuspended at 1-2×106 cells/mL using media with Neurobasal, 2% B27 (Invitrogen, 12587010), 1% fetal bovine serum (FBS; Sigma, F0926), 2 mM Glutamax (ThermoFisher, 35050061), 0.3% penicillin-streptomycin (ThermoFisher, 15140122), 4 ng/mL mouse recombinant basic fibroblast growth factor (bFGF; Fisher PMG0034), and 100 μM ascorbic acid (Sigma, A5960). Cell solutions with rat embryonic striatal neurons were obtained similarly by isolating and dissociating the striatum of E18 rat pups and cultured in the same media. In some cases, the rat ventral midbrain neurons were transduced virally to express green fluorescent protein (GFP) to facilitate visualization in vivo. In these cases, rat TE-NSPs were incubated overnight at 5 DIV with media including 1/2000 of pAAV1.hSyn.eGFP.WPRE.bHG (Addgene, 105539-AAV1), with a full media change occurring on the next day.
The C1.2 human iPSC line (passages 22-35) was derived from the commercially available BJ line (ATCC® CRL-2522™). Stem cells were differentiated based on a floor-plate midbrain dopaminergic induction protocol. The iPSCs were maintained on CF-1 mouse embryonic fibroblast feeder cells (Molecular Transfer, A34181) and cultured in embryonic stem cell (ESC) media containing DMEM/F12 (ThermoFisher, 11330032), 20% knockout serum replacement (KSR; ThermoFisher, 10828010), 1 mM non-essential amino acids (ThermoFisher, 11140076), 2 mM Glutamax, 0.1 mM β-mercaptoethanol (ThermoFisher, 21985-023), 1% penicillin-streptomycin, and 6 ng/mL bFGF (R&D Systems, 233-FB-CF). Human iPSCs were then detached with accutase and cultured with ESC media on dishes coated with 1/20 Matrigel (Corning, 354234) in DMEM/F12. When confluent, the cells were exposed to 100 nM LDN193180 (ReprocellUSA, 04-0074) and 10 μM SB431542 (ReprocellUSA, 04-0010-10) in DMEM/F12, 15% KSR, 2 mM Glutamax, and 10 UM β-mercaptoethanol (KSR media). These two compounds were used for dual SMAD inhibition to induce differentiation into neuroectoderm. Then at 1 dd the media was supplemented with 100 ng/mL sonic hedgehog (SHH; R&D Systems, 464SH025CF) and 2 μM purmorphamine (ReprocellUSA, 04-0009). At 3 dd, the media also included 3 μM CHIR99021 (ReprocellUSA, 04-0004), an activator of Wnt signaling required for coexpression of the floor- and roof-plate markers FOXA2 and LMX1A, respectively. Starting on 5 dd, the media was changed to 75% KSR media and 25% N2 media, with the proportion of N2 media increasing by 25% every two days. N2 media consisted of DMEM/F12 with L-glutamine and HEPES (ThermoFisher, 11330-032), 55 μM β-mercaptoethanol, 23.8 mM sodium bicarbonate (Sigma, S5761), 0.156% w/v D-[+]-glucose (Sigma, G7021), 22 nM progesterone (SIgma, P8783), and 1% N2 Supplement (Stem Cell Technologies, 07156). The dopaminergic precursor fate was established at 11 dd after which differentiation media was added, consisting of Neurobasal with 2% B27, 2 mM Glutamax, 200 μM ascorbic acid, 20 ng/mL GDNF (Fisher Scientific, 212GD050CF), 20 ng/ml BDNF (Peprotech, 450-02), 1 ng/mL TGFβ3 (R&D Systems, 243B3002CF), 500 μM dibutyryl cAMP (Sigma, D0260), and 10 μM DAPT (Fisher, 263410). After confirming a bipolar morphology and tight packing at 20 dd, the cells were lifted using accutase and cultured on dishes coated with 15 μg/mL poly-L-ornithine (Sigma, P4957) and 5 μg/cm2 laminin. At ˜40 dd the cells were detached with accutase and resuspended in differentiation media at 1×106 cells/mL for subsequent aggregation.
For micro-column seeding, dissociated dopaminergic or striatal neurons were first aggregated by adding 15 μL to each micro-well in a custom-made array of 3×3 inverted pyramidal micro-wells fitting in 12-well plates. The cells formed aggregates at the bottom after centrifugation at 1,500 rpm for 5 min and were subsequently incubated overnight (
The effect of biomaterial encasement on neurite growth length and rate in rat TE-NSPs (OD: 398 μm; ID: 160 μm; length: ˜5 mm) was evaluated at 1, 3, 7, and 10 DIV for 1% agarose (n=7), 3% MeHA (n=13), and 5% MeHA (n=14). Human TE-NSPs with those dimensions were similarly characterized at 8, 14, and 21 DIV for 3% MeHA (n=10, 4, 2 for three independent runs, respectively). Scaled-up human constructs in 3% MeHA were imaged for growth measurements at 8, 14, 21, 28, and 42 DIV (length: ˜1 cm; n=4, 3 for OD/ID combinations of 700/300 and 973/500, respectively). The neurite growth lengths were measured from phase contrast images as the distance between the inner edge of the aggregate and the tip of the longest observable neurite using Fiji. The growth rates were estimated with the backward difference method using the neurite lengths at the current and previous timepoints. The images were obtained using a Nikon Eclipse Ti-S microscope and QiClick camera integrated with NIS Elements software (Nikon). Where applicable, the results were analyzed using repeated measures two-way ANOVA and post-hoc Tukey's multiple comparison test for differences between groups, with P<0.05 for statistical significance.
The cytoarchitecture of TE-NSPs and the presence of neuronal and dopaminergic phenotypes were verified by immunolabeling and confocal imaging. TE-NSPs and other cultures were stained at several time points: 14 DIV for rat TE-NSPs; around 21, 30, and 60 DIV for human TE-NSPs. Cultures were fixed in 4% paraformaldehyde (Electron Microscopy Sciences, 15710) in DPBS for 35 min. Whole constructs were permeabilized for 1 hr with 0.3% Triton X-100 (Sigma, T8787) in 4% horse serum (ThermoFisher, 16050122) and then incubated overnight with primary antibodies in 4% horse serum at 4° C. The primary antibodies used in this study included tyrosine hydroxylase (TH; 1/500, sheep, Abcam, ab113; 1/500, rabbit, Pel-Freez Biologicals, P40101), the enzyme in the rate-limiting reaction of dopamine synthesis, and β-tubulin III ( 1/500, mouse, Sigma, T8578), a neuronal microtubule protein. Then, the cultures were exposed to appropriate fluorescent secondary antibodies ( 1/500, ThermoFisher, Alexa-488, Alexa-568, Alexa-647) in 4% horse serum for 2 hr at room temperature. Finally, the constructs were incubated for 10 min with Hoechst ( 1/10000, ThermoFisher, 33342), to stain the nuclei, and rinsed thoroughly. Immunolabeled cultures were imaged with a Nikon A1RSI laser scanning confocal microscope. Except where noted, all images used for analysis or included in this manuscript represent the maximum intensity projection of the full-thickness z-stacks.
In Vitro Functional Characterization with Fast Scan Cyclic Voltammetry
TE-NSPs were assessed for functionality by measuring electrically-evoked dopamine release with fast scan cyclic voltammetry, with the focus being the effect of biomaterial encasement and/or time. Rat TE-NSPs encased in 1% agarose (n=5), 3% MeHA (n=4), and 5% MeHA (n=5) were evaluated at ˜36 DIV. These constructs were incubated with 100 μM L-DOPA (Sigma, D9628) for 30 min before transferring to the recording chamber and perfused continuously throughout the analysis. In this case, dopamine release was analyzed using the Kruskal-Wallis test with Dunn's multiple comparisons test. Constructs fabricated with human iPSC-derived dopaminergic neurons (OD: 398 μm; ID: 160 μm) were assessed based on the following groups: 23-24 DIV (3% MeHA: n=3; 5% MeHA: n=3), 29-32 DIV (1% agarose: n=1; 3% MeHA: n=2; 5% MeHA: n=1), and 56-58 DIV (3% MeHA: n=3). In certain cases, L-DOPA was added to determine its effect on dopamine release.
Individual TE-NSPs were transferred to a recording chamber and perfused at 37° C. with media bubbled with 95% O2 and 5% CO2 and containing Neurobasal, 2.0 mM L-glutamine, and ascorbic acid (100 and 200 μM for rat and human cells, respectively). A bipolar stimulating electrode (Plastics One) was placed to span the region of interest, and a carbon fiber electrode (150-200 μm outer fiber length, 7 μm diameter) was located in this same area. In all cases the region of interest was the dopaminergic aggregate, except where noted. For stimulation, the bipolar electrode was set to apply a monophasic+ electrical train of 10 pulses of 5 ms width at 20 Hz and an amplitude of 8 V. For recording, the potential of the carbon fiber electrode was scanned linearly from −0.4 V to 1.2 V to −0.4 V vs. Ag/AgCl at a rate of 400 V/s with a voltammeter/amperometer (Chem-Clamp; Dagan Corporation). The stimulation train was applied at 8 s, and the cyclic voltammograms (CVs) were recorded during the entire session every 100 ms using the Demon Voltammetry and Analysis Software. The current at the peak oxidation potential of dopamine in consecutive CVs, as well as all the CVs obtained during the session, were averaged from 4-6 recording runs within the same location 8-12 min apart. This averaging was done to minimize the noise and emphasize the signal. The currents were converted to dopamine concentration with a calibration curve. We recorded the elicited current after injection of 1.5, 3, and 6 μM dopamine hydrochloride (Sigma, H8502) and obtained the slope of the regression of concentration vs. current. The dopamine in this case was dissolved in deionized water with 126 mM NaCl, 2.5 mM KCl, 1.2 mM NaH2PO4-H2O, 2.4 mM CaCl2-2H2O, 1.2 mM MgCl2-6H2O, 25 mM NaHCO3, and 0.4 mM L-ascorbic acid (all from Sigma).
Male Sprague Dawley rats (8-11 weeks, 275-290 g; Charles River) were used for transplants of MeHA-encased rat TE-NSPs at 14 DIV, with a terminal time point of 2 weeks (3% MeHA: n=3; 5% MeHA: n=3). Rats were anesthetized with 5% isoflurane (Midwest Veterinary Supply, 193.33165.3) and mounted in a stereotactic frame, with anesthesia being maintained at 2.5%. The scalp was disinfected with betadine, and 2.0 mg/kg bupivacaine (Hospira, 0409-1161-19) was injected in the incision area. The incision was made along the midline of the scalp, and a 5 mm craniectomy was made at −5 mm AP and +2.2 mm ML relative to Bregma. TE-NSPs were pulled into a needle (Vita Needle; OD: 534 μm, ID: 420 μm) containing a rod that lay directly in contact with the construct. A protective polyurethane tubing (OD: 686 μm; ID: 559 μm; Microspec Corporation) was cut to fit the needle, sealed on one end, and used to cover the outer area of the needle. The needle was inserted into a Hamilton syringe attached to a stereotactic arm, and its plunger was put directly in contact with the exposed rod coming out of the needle. The stereotactic arm was adjusted to be at 38° relative to the horizontal, the dura was opened, and the needle was lowered ˜12 mm into the brain. Then, the sheath was pulled up with forceps to break the seal, and the TE-NSPs were laid out by pulling up the needle ˜5 mm while a stationary arm maintained the plunger in place. After 2 min, the needle was pulled out entirely, the scalp was sutured, and the animals received a subcutaneous injection of 2.0 mg/kg meloxicam (Midwest Veterinary Supply, 577.30200.3). The animals were sacrificed with transcardial perfusions with heparinized saline followed by 10% formalin. The brains were extracted, post-fixed overnight in 10% formalin, and stored in 1×PBS at 4° C.
Thick brain sections were subjected to a clearing and immunohistochemistry protocol to visualize the entire constructs and assess their cytoarchitecture and survival. A vibratome (Leica, VT1000S) was utilized to obtain 2 mm thick sagittal sections containing the implanted constructs. The sections were dehydrated with increasing concentrations of methanol (Fisher, A412) and then incubated overnight at room temperature with 66% dichloromethane (Sigma, 270997) and 33% methanol for delipidation and refractive index matching. Afterwards, the sections were bleached overnight in chilled 5% H2O2 (Sigma, H1009), rehydrated with decreasing concentrations of methanol, permeabilized for 2 days at 37° C., and blocked for 2 days at 37° C. Permeabilization was performed using 0.16% Triton X-100, 20% dimethyl sulfoxide (DMSO; Sigma, 276855), and 0.3 M glycine (Bio-Rad, 161-0718) in 1×PBS, while the blocking solution consisted of 0.16% Triton X-100, 6% horse serum, and 10% DMSO in 1×PBS. Afterwards, the sections were exposed to primary and secondary antibodies for 7 days at 37° C. each, with washing steps in between. The primary antibody buffer was 3% horse serum and 5% DMSO dissolved in 1×PBS with 0.2% Tween-20 (Sigma, P2287) and 10 μg/mL heparin (Sagent Pharmaceuticals, 25021-400-10), while the secondary buffer contained 3% horse serum and the same solvent. We used primary antibodies for TH ( 1/100, sheep, Abcam, ab113) and GFP ( 1/100, mouse, Aves, GFP-1020) and the appropriate Alexa-Fluor secondaries ( 1/250). The sections were then dehydrated with increasing concentrations of methanol, exposed to Visikol® HISTO-1™ overnight at room temperature, and incubated with Visikol® HISTO-2™ at least for 2 hr before imaging.
We have traditionally fabricated TE-NSPs encased in 1% agarose hydrogels (1), but for the next generation of these constructs we chose HA hydrogels because of their bioactivity and the modifications available to expand the capabilities and functionality of TE-NSPs. We modified ˜40% of the hydroxyl groups in the N-acetyl-D-glucosamine units of HA to obtain MeHA following existing protocols. MeHA hydrogels can be created by radical photopolymerization reactions where light and an initiator molecule generate free radicals that react with the methacrylate reactive group. Indeed, hydrogels were formed by irradiating solutions with 10 mW/cm2 UV light, as confirmed with rheological measurements showing an increase in the storage (G′) modulus as a function of time after exposure to light to values much greater than the loss (G″) modulus (
We evaluated the feasibility of using MeHA to encase TE-NSPs by fabricating these constructs with rat embryonic dopaminergic neurons, our conventional cell source, and testing outcomes of growth, cytoarchitecture, and functionality in vitro. Initially, we confirmed that MeHA hydrogels could be produced in the shape of a hollow micro-column that could be filled with polymerized collagen and laminin. Qualitatively, we have consistently observed a high yield of MeHA micro-columns with ECM filling the lumen axially and longitudinally. We were then able to produce MeHA-encased TE-NSPs with a rat dopaminergic aggregate in ˜0.5 cm micro-columns. We characterized the role of biomaterial type and time on TE-NSP neurite growth, with the former having a significant effect on length (P=0.0002) and rate (P=0.0011) and the latter on both outcomes (P<0.0001). All mean neurite lengths were higher for MeHA groups relative to agarose, with significant differences being observed between 1% agarose and one of the MeHA groups at all time points (
Following these structural characterization studies, we examined the functionality of TE-NSPs encased in MeHA in terms of electrically-evoked dopamine release. We used fast-scan cyclic voltammetry (FSCV), a technique with high temporal/spatial resolution used to measure changes in neurotransmitter concentration based on the currents elicited by the oxidation or reduction of the released molecules. For this purpose, we introduced a carbon fiber electrode into the aggregates in TE-NSPs to record extracellular dopamine release after electrical stimulation (
A crucial consideration for the feasibility of using MeHA to encase TE-NSPs was whether these constructs could be implanted in rats, maintain their structure during and after delivery, and survive. Thus, after 14 DIV MeHA-encased rat TE-NSPs were stereotactically implanted to span the region between the SNpc and the striatum in rats (
Human iPSC-Derived Dopaminergic Neurons can be Employed to Fabricate TE-NSPs with Various Dimensions
Ultimately our goal was to transition to more clinically relevant TE-NSPs fabricated from human dopaminergic neurons. We obtained these cells after differentiating human iPSCs in 2D and aggregating them by centrifugation in custom-made micro-wells at ˜40 dd, after which they were plated in 2D or within micro-columns. This time point was chosen balancing the need for dopaminergic fate specification, sufficient TH expression (>50%) when plating, proper maturation at the time of implantation, and timeliness in TE-NSP fabrication. We first needed to confirm that human dopaminergic neurons could be cultured as aggregates in laminin-coated surfaces. These aggregates exhibited extensive neurite growth in all directions from the entire circumference (
Having confirmed that the neurons survive and grow after aggregation, we proceeded to create TE-NSPs with 3% MeHA and these human aggregated neurons. We only made human constructs with 3% MeHA because 3 and 5% MeHA yielded generally similar growth and functional results with rat neurons, as previously discussed, and we wanted to maintain the encasement type consistent when testing human neurons. In addition, 3% MeHA micro-columns could be easily handled manually even though they were noticeably softer than 5% MeHA. We started making TE-NSPs with dimensions relevant for implantation to span the endogenous nigrostriatal pathway in rats, having an OD and ID of 398 and 160 μm, respectively, and a length of ˜0.5 cm. Representative phase contrast images of these human TE-NSPs showed neurite growth throughout the entire length of the micro-column (
We also fabricated human TE-NSPs with greater diameters and lengths to investigate if these constructs could be scaled up to accommodate the cell number and structural requirements for implantation in larger animal models and humans. MeHA-encased human TE-NSPs could be made to have axon tracts spanning the lumen of micro-columns with an ID of 300 μm (
We employed FSCV to detect the release of dopamine after electrical stimulation in the aggregate of human TE-NSPs and to corroborate the functionality of these constructs. After placing both the stimulating and carbon fiber electrodes in the aggregate, we observed examples of evoked release of 56 nM (
Standard treatments for PD (e.g., L-DOPA) therapy generally involve discontinuous drug delivery, which can cause severe motor fluctuations with chronic exposure. DBS electrodes provide inherently non-specific stimulation and are implanted downstream of the striatum, thus circumventing the cause of PD. Moreover, these stiff inorganic electrodes are invasive and typically produce a chronic foreign body response diminishing their effectiveness. Cell replacement approaches have traditionally involved the transplantation of fetal- or stem cell-derived dopaminergic neurons only in the striatum. In contrast, TE-NSPs represent a biologically-inspired strategy to reconstruct the dopaminergic fiber tracts in the nigrostriatal pathway, thereby targeting both the loss of dopaminergic neurons/axons and striatal innervation and dopamine input at the core of the pathophysiology of PD. TE-NSPs would ideally improve upon the discontinuous effects of L-DOPA and the lack of specificity of DBS by delivering dopamine in a natural, synaptic-based, and spatially-specific manner. Apart from this, our human TE-NSPs can possibly advance conventional cell replacement approaches by employing a clinically relevant cell source that is more available than fetal grafts and that may allow the development of personalized implantable dopaminergic axons. These constructs also featured an HA hydrogel encapsulation, which serves to protect cells and axons during and after implant in comparison with current stem cell-derived transplants delivered as cell suspensions. This may allow us to tune degradation profiles and to provide other functionalities like the local presentation or controlled release of pro-survival molecules to improve in vivo outcomes.
Our previous iterations of TE-NSPs have been fabricated with agarose hydrogels and rat embryonic ventral midbrain neurons. Here, we transitioned to using HA (in its modified form MeHA) as the biomaterial for the encasement of the dopaminergic axon tracts. HA is an unsulfated glycosaminoglycan that is the major backbone of the brain ECM and can bind to neural cells through cell surface receptors such as CD44 and CD168. It forms complexes with lecticans, a type of chondroitin sulfate proteoglycan that includes aggrecan, versican, neurocan, and brevican. These complexes offer the structural support and modulation of development, cell signaling, intercellular interactions, and synaptic plasticity that are key to brain function and participate in the response to injury. Degradation of HA in injury has been associated with angiogenesis and neuroimmune and inflammatory responses according to the molecular weight of the fragments. HA hydrogels have been applied for 3D stem cell encapsulation and control of their expansion, migration, and differentiation. In terms of neural applications, HA hydrogels have been utilized to differentiate human iPSC-sourced neural progenitor cells (NPCs) and augment survival and behavioral outcomes in models of stroke. In a study related to ours, human ESC-derived dopaminergic neurons were encapsulated in HA to mature the cells in vitro and protect them during transplantation. Despite this, there are no published data characterizing human iPSC-sourced dopaminergic neurons in HA. This biomaterial can be used to form hydrogels ideal for our application when the backbone is modified with certain functional groups. We used MeHA given its status as the most common form of modified HA for photocrosslinking. Photopolymerization paradigms permit having spatial control of gelation and properties and in situ cell encapsulation. MeHA and other modified HA hydrogels have also been used extensively for 3D-bioprinting, which should allow for standardized manufacturing of micro-column as described herein. Previous manufacturing methods often produced micro-columns inherently have off-center lumens with variable wall thicknesses along the constructs and between batches. This could increase the likelihood of ruptured micro-columns and exposure of cells and axons to differential stiffnesses. Thus, using HA and 3D-bioprinting to expected to allow the manufacture of TE-NSPs with less batch-to-batch variability and a greater potential in the clinic.
Here, we were able to show that MeHA can be applied to build hollow micro-columns containing a lumen with polymerized collagen and laminin. Neuronal aggregates sourced from rat embryonic ventral midbrains were able to extend long axon tracts that stained for TH, a key dopaminergic marker. Growth and functional outcomes were compared to 1% agarose micro-columns given that this biomaterial has been our historical standard when fabricating TE-NSPs and other micro-tissue constructs. Our results suggested that the type of hydrogel had a significant effect on growth length and rate, with MeHA-encased constructs exhibiting significantly greater outcomes relative to 1% agarose. At 1 DIV, which may imply that the aggregates may require time to properly adhere and adapt to their environment prior to extending neurites at higher growth rates. Growth rates were lowest We confirmed that MeHA-encased TE-NSPs retained their functionality by releasing dopamine upon electrical stimulation in levels comparable to time-matched constructs encased in 1% agarose. Overall, we validated that MeHA can be used to make TE-NSPs that structurally and functionally resemble the nigrostriatal pathway.
The effects of the hydrogel on growth could be attributed to MeHA itself providing a microenvironment more conducive for neurite growth than previous materials, for example, agarose. For example, HA could be more amenable to cells given its crucial role in vivo in the brain ECM. It may be better at retention and local presentation of growth factors present in the media to cells in comparison to agarose. MeHA could also promote a more consistent superior polymerization of the collagen and laminin core, thus providing a 3D space more agreeable for neurite growth. On the other hand, the improved growth in MeHA could be a result of the distinct mechanical properties of MeHA and agarose, as exemplified by the greater compressive moduli for the former. We have previously shown that axons have the highest density of growth along the hydrogel-ECM interface in the circumference of the lumen. As such, while the neurites use the ECM core for growth throughout the width of the lumen, they also interact with and use the hydrogel as a substrate for growth, and the mechanical properties of the biomaterial may thus directly impact cell behavior. Indeed, it has been extensively studied that neural cells and other types sense and respond to substrate stiffness. Specifically, it has been widely reported that there is an inverse relationship between neurite growth and substrate stiffness in studies using mouse hippocampal NPCs seeded on a 3D hydrogel of HA-pentenoate and a poly(ethylene glycol)-bis(thiol) crosslinker, mouse ventral midbrain NPCs in MeHA hydrogels, and chick dorsal root ganglia in 3D agarose hydrogels. In addition, human iPSC-derived NPCs, as single cells or spheroids, exhibited more spreading and attachment in acrylated HA hydrogels and greater neurite outgrowth/density and neuronal differentiation in 3D MeHA hydrogels that were soft relative to stiffer substrates. While our results coincide with stiffness influencing neurite growth, we observed an opposite relationship: the stiffer MeHA micro-columns benefitted growth length and rate when compared to the less stiff 1% agarose. This may imply that other mechanical/physical properties are the driving force in provided constructs, that indeed MeHA itself provides a better environment than agarose separate from stiffness, and/or that the stiffer MeHA provides a more structurally sound encasement that maintains the integrity of the ECM core. On the latter point, it is indeed common to see shrinking or detachment of ECM inside agarose hydrogels, which may disrupt the neurite growth process. Our results may also reflect that we are using a different type of scaffold, one in which the hydrogel substrate is found in the periphery of a 3D micro-column, while previous studies have relied on 2D surfaces or 3D growth in whole blocks/pieces of hydrogel. Indeed, we previously demonstrated improved neurite health in engineered micro-tissues similar to TE-NSPs when grown in 3-4% agarose micro-columns relative to 1-2% agarose. Thus, this direct relationship between stiffness and growth outcomes may be a feature of our micro-column scaffolds.
The present disclosure also confirmed that MeHA micro-columns could be successfully delivered into rat brains along the nigrostriatal pathway and that the cells could survive and retain their axonal cytoarchitecture for at least 2 weeks in vivo. As such, we deemed that MeHA was a feasible encasement for TE-NSPs in terms of structure, function, and transplantation. MeHA is mainly susceptible to enzymatic degradation by hyaluronidases. However, our results indicate that hyaluronidase presence around the implant was not sufficient for noticeable degradation to occur at observed time points. In some embodiments, it may be advisable to have a controllable degradation profile in which there is little to no degradation during the in vitro culture and early implant periods, with degradation becoming more prominent in vivo as time goes on. This would allow for TE-NSPs to grow in 3D with the proper cytoarchitecture and to have sufficient integrity to be properly implanted and to integrate with the host circuitry acutely. Subsequent degradation may enable better and more seamless integration of the dopaminergic axon tracts with the surrounding tissue. Moreover, the rate of degradation is inversely related to stiffness and has other implications. For example, it has been established that implants should have mechanical properties matching the surrounding tissue for better outcomes such as the inflammatory response. Indeed, softer acrylated HA hydrogels with a stiffness matching the mouse brain caused significantly less reactive astrocyte and microglia presence in naive and stroke mouse models, respectively, relative to stiffer hydrogels. Still, the stiffness must be sufficient to ensure the hydrogels hold the shape of a hollow micro-column without disrupting the interior before, during, and after implantation. Key properties, such as degradation and stiffness, can be tuned by optimizing the MeHA macromer concentration, molecular weight of HA, and the degree of methacrylation. Greater control of degradation may be granted by using hydrogels that are more susceptible to hydrolysis or cell-based degradation. For example, HA hydrogels sensitive to hydrolysis have been created by modifying HA to have methacrylates bound to lactic acid or caprolactone, which are responsive to ester hydrolysis. Di-thiol peptides sensitive to cleavage by matrix metalloproteinases expressed by cells have also been used as part of the sequential crosslinking of acrylated HA hydrogels.
Other types of HA could also be used to make TE-NSPs depending on the desired properties (e.g., degradation, stiffness, toughness, binding of cell-responsive peptides, retention and release of growth factors). Previous studies have characterized double-network hydrogels consisting of a primary network resulting from the non-covalent guest-host assembly of HA modified with β-cyclodextrin and HA with adamantane and a secondary one with thiol-based crosslinking of MeHA. The two networks could also be covalently tethered when the polymers in the primary network were also methacrylated. The guest-host assembly conferred self-healing properties by dissipating and transferring stress to protect the covalent bonds. On the other hand, changing the thiol-to-methacrylate ratio tuned the compressive elastic moduli of the hydrogels. These hydrogels could allow TE-NSPs to recover from mechanical deformation without failure and provide greater resilience to handling during culture and implantation. Notably, this could be a possible avenue for shaping TE-NSPs during implantation to fit the more complex curved geometry of the human nigrostriatal pathway. For example, with norbornene-modified HA, the norbornenes can have crosslinks only with the addition of di-thiol molecules. By modulating the ratio of thiols to norbornenes, it can be ensured that there will be unreacted norbornenes remaining for subsequent reactions. This paradigm could be applied for TE-NSPs encased in micro-columns that are first crosslinked and then bound with thiol-containing peptides or proteins in a secondary reaction. These molecules could influence survival, adhesion, neurite growth, among other relevant in vitro and in vivo outcomes. In addition, TE-NSPs could be further modified to retain or deliver growth factor and drugs, as HA hydrogels have been widely used for this purpose based on diffusion, protease-sensitive, or hydrolytic degradation, on-demand force-mediated and light-triggered processes, and affinity to chemical groups. We foresee that HA will significantly improve TE-NSPs and other engineered micro-tissues by expanding their range of applications, design control, and prospects for successful clinical outcomes.
Having validated the use of an HA-based hydrogel to encase TE-NSPs, we then created these constructs with human iPSC-derived dopaminergic neurons. These neurons allow for fabricating patient-derived, implantable dopaminergic axon tracts, circumventing the ethical and practical limitations of human fetal- or ESC-sourced neurons. These neurons were differentiated in 2D for ˜40 days following established protocols and then aggregated for seeding within the micro-columns. The present disclosure describes the feasibility of combining an HA-based hydrogel encasement and human iPSC-derived dopaminergic neurons to fabricate TE-NSPs. We found that the human dopaminergic aggregates could grow axon tracts to ˜1 cm by 28 DIV within micro-columns. To our knowledge, this is the first report of human iPSC-derived dopaminergic engineered tissue exhibiting unidirectional axon growth reaching lengths up to ˜1 cm in vitro. Human TE-NSPs exhibited the desired cytoarchitecture of neuronal cell bodies restricted to the end of the micro-columns with aligned axon tracts longitudinally spanning the lumen and expressing a dopaminergic phenotype at least up to 2 months in vitro. The human TE-NSPs released dopamine to the extracellular space within the aggregates upon electrical stimulation, with concentrations increasing with time. Notably, we observed dopamine release in the 50-110 nM range between 30-60 DIV using FSCV, which is the range that is estimated to cause a functional effect in parkinsonian rats. Specifically, constructs cultured for longer were able to release more dopamine after stimulation, suggesting ongoing maturation and concomitant increase in dopamine release machinery. This assessment of somatodendritic dopamine release served as verification for the health, phenotype, and functionality of TE-NSPs. Somatodendritic release in the SNpc is also an important aspect of the native nigrostriatal pathway, where it is involved in the modulation of neuronal activity in the SNpc and SNpr and distal dopamine release in the striatum. It was not surprising that we did not detect evoked dopamine in the axonal tract region given that TE-NSP axons were in their developing, growth phase and did not have an end-target for integration. Studies have highlighted the absence of or low capacity for evoked neurotransmitter release in developing mammalian neurons of the central nervous system. Another noteworthy observation during FSCV analysis was that the CV colorplots for human TE-NSPs appeared to be more noisy or show signal peaks other than those characteristic for dopamine when compared to the much cleaner plots for rat TE-NSPs. This may highlight differences in fate commitment, functional maturation, or cell content between the rat embryonic and the human iPSC-derived dopaminergic neurons.
We utilized an existing protocol to differentiate human ESCs and iPSC into dopaminergic neurons based on the coexpression of the floor-plate and roof-plate markers FOXA2 and LMX1A, respectively. These neurons had more than 50% expression of TH and exhibited dopamine release, membrane potential oscillations at frequencies characteristic of SNpc neurons, and markers such as G-protein-regulated inward-rectifier potassium channel 2 (GIRK2) and calbindin, which are characteristic for A9 and A10 dopaminergic neurons, respectively.
Another aspect of construct manufacturing is consideration around the ratio of A9 to A10 dopaminergic neurons after differentiation and present within the TE-NSPs, given the role of these cell types in functional dopamine restoration. The A9 dopaminergic neurons extend axons mainly to the dorsolateral striatum and are selectively lost in PD, while the A10 cells of the ventral tegmental area project to the ventral striatum and limbic regions and are not heavily impacted. Certainly, the intrinsic capacity of cells to recognize proper outgrowth targets has been highlighted by the inability of non-A9 dopaminergic neurons (e.g., dopaminergic neurons of the olfactory bulb, ventral forebrain neurons) to innervate the dorsolateral striatum. Consequently, studies in 6-OHDA rats have shown that mouse-sourced intrastriatal grafts enriched in A9 neurons (˜65% GIRK2+/calbindin− cells) innervated a significantly greater area of the dorsolateral striatum and led to the improvement of rotational and forelimb asymmetry scores to intact lesion levels when compared to grafts deficient in this cell type (˜15% GIRK2+/calbindin-cells). Nevertheless, most if not all of transplantation studies with stem cell-derived neurons have both populations and show innervation of both A9 and A10 target regions. It would be advisable to control the relative content of GIRK2+ and calbindin+ dopaminergic neurons in TE-NSPs to ensure a therapeutic effect and that the axons innervate the proper striatal targets, a capacity that has been previously identified for dopaminergic subtypes. For example, a different protocol used lentiviral vectors to drive the expression of LMX1A in human ESCs and iPSCs, resulting in over 60% A9 neurons compared to only 10% in controls. Future iterations of TE-NSPs could incorporate protocols that account for differentiation into specific dopaminergic subtypes. Apart from the differentiation protocol, fabrication steps specific to TE-NSPs may also influence cell behavior.
MeHA was synthesized by the esterification of hyaluronic acid (Lifecore Biomedical) with methacrylic anhydride (Sigma, 276685) in cold deionized water for ˜3.5 hr, with the pH maintained at 8.5 throughout with the addition of NaOH. The MeHA solid was obtained after dialysis for 5-7 days and lyophilization. This reaction produced a degree of methacrylation of the hyaluronic acid backbone of ˜40%, as confirmed by 1H-nuclear magnetic resonance spectroscopy. The degree of methacrylation corresponded to the ratio between the area below the two vinyl singlets at ˜5.8 and ˜6.3 ppm and the area below the region corresponding to the sugar ring in the spectra. MeHA solutions were created by adding MeHA for a 3% weight/volume (w/v) to Dulbecco's phosphate buffered saline (DPBS; ThermoFisher, 14190136) having 0.05% w/v Irgacure 2959 (12959; Sigma, 410896) photoinitiator. We characterized the reaction product for its capacity to form hydrogels by light-induced radical polymerization. This was achieved by measuring the storage and loss moduli of the MeHA hydrogel solution over time before and after exposure to 10 mW/cm2 ultraviolet light (UV) using a controlled-stress rheometer (TA Instruments, AR2000) connected to an UV lamp (Excelitas Technologies, OmniCure S1500).
2. Fabrication of MeHA Hydrogel Micro-Columns with an ECM Core
The hydrogel micro-columns used as the encasement for TE-NSPs were fabricated by first inserting an acupuncture needle with an outer diameter (OD) of 500 μm (Lhasa OMS, TC1.50×100) inside a glass capillary tube with an inner diameter (ID) of 973 μm (Drummond Scientific, 1-000-0750). The former and the latter create the lumen and the outer shell of the micro-column, respectively, which has an OD of 973 μm and an ID of 500 μm. The MeHA solution was drawn into the tubes by capillary action, and they were then placed horizontally below the UV lamp and irradiated with 10 mW/cm2 light for 5 min. The needles were removed, and the gelled micro-columns within the capillary tubes were pushed out into DPBS using a 20 gauge needle (BD, 305178). The micro-columns were then inspected with light microscopy to remove any that did not form properly and sterilized under UV light for 30-60 min. The micro-columns were cut to the desired length (1-2 cm) under a dissection microscope.
The extracellular matrix (ECM) core of the micro-columns was created by first making a solution of 1 mg/mL rat-tail type I collagen (Corning, 354236) and 1 mg/mL mouse laminin (Corning, 354232) in Neurobasal media (ThermoFisher, 21103049) adjusted to a pH of 7.2-7.4 with 1 N NaOH. After suctioning the DPBS remaining within the lumen of the hydrogel micro-columns, excess ECM solution was added to the lumen to fill it (around 10 μL for a 1 cm long column). The ECM was allowed to polymerize in an incubator at 37° C. for 15 min and, after this period, fresh media was added and floating, excess gelled ECM not inside the columns was removed with forceps. The hydrogel-ECM micro-columns were kept in the media at 37° C. for 1 or 2 days before seeding the neuronal aggregates.
Human iPSC-derived dopaminergic neurons were obtained already differentiated and cryopreserved (iCell DopaNeurons from Fujifilm Cellular Dynamics, R1032), having a purity of 98% MAP2+/NES−, 96% FOXA2+, and 90% TH+ cells as reported by the company. The source somatic cells were obtained from a healthy male of 50-59 years, and the generated iPSCs were differentiated based on the midbrain floor plate protocol. The culture media consisted of iCell Neural Base Medium 1 (Fujifilm Cellular Dynamics, M1010) with 2% Neural Supplement B (M1029), 1% iCell Nervous System Supplement (M1031), and 1% Penicillin-Streptomycin (ThermoFisher, 15140122). After receipt, the cells were thawed by following the protocol suggested by the company. The vial was placed for 3 min in a bead bath at 37° C., after which its contents were slowly transferred to a 50 mL centrifuge tube. Then, after rinsing the vial, 9 mL of culture media were slowly added to the tube and gently mixed by swirling. This cell solution was then moved to a 15 mL tube and centrifuged at 400×g for 5 min. The supernatant was aspirated, the cell pellet was resuspended in 2 mL of media, and the viable cell concentration was quantified with a hemocytometer and Trypan Blue exclusion. New cell solutions were created as needed to form aggregates with the desired number of cells.
The neuronal aggregates were created with custom-made polydimethylsiloxane (PDMS) molds containing nine inverted pyramidal micro-wells. Around 15 μL of cell solution were transferred to each micro-well, and the plate containing the molds was centrifuged at 1500 rpm for 5 min in order to concentrate the cells at the tip of the micro-well. Culture media was added to cover each PDMS mold, and the plate was incubated overnight at 37° C. On the next day, the aggregates were pipetted out of the micro-wells and seeded within one end of the hydrogel micro-columns with fine forceps. In some cases, a microknife was used to make enough space in the ECM to contain the aggregates. For the TE-NSPs presented in this manuscript, we started with aggregates containing 13,269 neurons. Thus, for TE-NSPs containing aggregates with 2.7×104, 5.3×104, and 1.1×105 cells we seeded 2, 4, and 8 aggregates, respectively, which over time merged to become one within the lumen. TE-NSPs were maintained in culture at 37° C. and 5% CO2, and 75-100% of the media was changed every 3 days.
In some cases, an aggregate of striatal neurons was seeded at the end of the hydrogel micro-column opposite to the one containing the dopaminergic aggregate 5 days after seeding the latter. For this study, two striatal aggregates were seeded to have a merged total of 2×105 cells. These striatal neurons were isolated from embryonic day 18 rat pups. Pregnant Sprague-Dawley rats (Charles River) were euthanized by CO2 exposure and decapitation, and the pups were extracted. Under a dissection microscope, the brains were removed from the heads of the pups and the striatum was cut from each hemisphere, taking care to remove any pieces of cortex. The striata were rinsed with cold Hank's balanced salt solution (HBSS; ThermoFisher, 14175079) and then exposed to accutase (ThermoFisher, A1110501) for 10 min at 37° C., with shaking every 3 min. A Pasteur pipette was used to triturate the tissue until dissolved and then the solution was centrifuged at 200×g for 5 min. The supernatant was removed and the pellet was resuspended in 2 mL of media. This cell solution was utilized to form the aggregates as explained in the previous section.
TE-NSPs were imaged at 4, 7, 10, 14, 20, and 31 days in vitro using phase contrast microscopy with a Nikon Eclipse Ti-S microscope and QiClick camera integrated with NIS Elements software (Nikon). These images were analyzed to quantify the neurite growth length and rate using ImageJ/Fiji. The neurite growth length corresponded to the distance between the edge of the aggregate and the tip of the longest neurite that could be observed in the lumen. The growth rate was estimated based on the difference between the neurite length at a time point and the previous one relative to the difference in the number of days. The effects of variables such as cell number, time, and/or culture type were assessed using repeated measures ANOVA or a mixed-effects model, and differences between groups were determined with Tukey's or Sidak's multiple comparisons test with the Prism 8 software (GraphPad). In all cases p<0.05 was considered statistically significant.
6. Assessment of Structure and Phenotype with Immunocytochemistry
The structure and phenotype of TE-NSPs were confirmed by immunolabeling these constructs to identify relevant neuronal, dopaminergic, and striatal markers. At terminal time points, TE-NSPs were incubated for 35 min in 4% paraformaldehyde (Electron Microscopy Sciences, 15710) in DPBS to fix the tissue. The constructs were then permeabilized for 1 hr with 0.3% Triton X-100 (Sigma, T8787) in 4% horse serum (ThermoFisher; 16050122). Afterwards, the TE-NSPs were incubated overnight at 4° C. with the primary antibody solution in 4% horse serum. The primary antibodies used in this study were: (1) tyrosine hydroxylase (TH; 1/500, rabbit, Pel-Freez Biologicals, P40101), the enzyme in the rate-limiting reaction of dopamine synthesis; (2) β-tubulin III/Tuj1 ( 1/500, mouse, Sigma, T8578), a neuronal microtubule protein seen in somata and axons; (3) dopamine-and-cAMP-regulated neuronal phosphoprotein (DARPP-32; 1/500, mouse, Santa Cruz Biotechnology, sc-271111), a marker of striatal medium spiny neurons. After rinsing, the constructs were incubated for 2 hr at room temperature with the secondary antibody solution in 4% horse serum ( 1/500, ThermoFisher, donkey anti-mouse 488, donkey anti-rabbit Alexa-568). The nuclei were then stained for 10 min with Hoechst ( 1/10000, ThermoFisher, 33342). After rinsing with DPBS, the full thickness of these constructs was imaged using a Nikon A1RSI laser scanning confocal microscope and the maximum intensity projections were obtained from the z-stacks.
TE-NSPs were analyzed at 33-42 days in vitro for electrically-evoked dopamine release using fast-scan cyclic voltammetry. These constructs were placed in a perfusion chamber at 37° C. with an inlet flow of Neurobasal, 2.0 mM L-glutamine (ThermoFisher, 35050061), and 200 μM ascorbic acid (Sigma, A5960) bubbled with a mixture of 95% 02 and 5% CO2. Stimulation and recording were both done in the same area within the dopaminergic aggregate, axon tracts, or the striatal end. Stimulation was carried out with a bipolar electrode (Plastics One) applying a monophasic+ electrical train of 10 pulses of 5 ms width at 20 Hz and an amplitude of 8 V. Recordings were performed with a carbon fiber electrode (150-200 μm outer fiber length, 7 μm diameter) that had its potential scanned linearly from −0.4 V to 1.2 V to −0.4 V vs. Ag/AgCl at a rate of 400 V/s with a voltammeter/amperometer (Chem-Clamp, Dagan Corporation). Cyclic voltammograms (CVs) were obtained every 100 ms during the recording session using the Demon Voltammetry and Analysis Software (Wake Forest Baptist Medical Center), and the representative current for the session corresponded to the current at the peak oxidation potential of dopamine. For each TE-NSP and recording/stimulation region we obtained the current, the CV at peak release, and a colorplot representing all CVs over time; all these represented the average signal from 3-6 runs within the same area taken 8-12 min apart. Then we processed the data with principal component regression, a chemometric tool combining principal component analysis with inverse least-squares regression, to improve the isolation of the dopamine analyte from CVs that contained other interfering signals. This process was carried out with the Demon software using a training set from awake rats having the CVs corresponding to different known dopamine concentrations and known pHs. The carbon fiber electrodes in use during our recording sessions were calibrated over time, before and after recordings, by determining the evoked current corresponding to injections of 1.5, 3, and 6 μM dopamine hydrochloride (Sigma, H8502). During these calibrations the solvent was deionized water with 126 mM NaCl, 2.5 mM KCl, 1.2 mM NaH2PO4—H2O, 2.4 mM CaCl2-2H2O, 1.2 mM MgCl2-6H2O, 25 mM NaHCO3, and 0.4 mM L-ascorbic acid (all from Sigma). The slope of concentration vs. current was used to convert the currents from processed TE-NSP recordings to concentrations. The maximum dopamine release in each case was measured as the difference between the peak concentration during the 2 s after stimulation and the average baseline signal before stimulation. Two-way ANOVA was used to assess the effect of the type of construct (different cell numbers, unidirectional or bidirectional) and recording/stimulation region (dopaminergic aggregate, axon tracts, or striatal end) on the evoked dopamine concentration. Sidak's multiple comparisons test was used to determine differences between groups. In all cases, p<0.05 was considered to be statistically significant.
Our combinatorial approach centers on engineering the entire nigrostriatal pathway in vitro for implantation into the host brain as a unit to replace/reconstruct the damaged fibers. This platform, which we call the tissue-engineered nigrostriatal pathway (TE-NSP), consists of a hydrogel micro-column with controllable inner/outer diameter and length that has a lumen filled with extracellular matrix (ECM) and seeded on one end with an aggregate of dopaminergic neurons that extends long axon tracts throughout the column. Thus, TE-NSPs mimic the cytoarchitecture of the native nigrostriatal pathway: dopaminergic neurons restricted to one end (the SNpc) projecting axonal tracts (the nigrostriatal fibers) that in turn release dopamine and form synapses along medium spiny neurons (MSN) neurons in the striatum in vivo. The hydrogel encases the engineered dopaminergic tissue, provides necessary physical cues to drive longitudinal axon growth in vitro, and protect it during and after implantation. The ECM ensures neurons have the proper environment to survive and grow axons throughout the lumen.
Herein, we have characterized TE-NSPs fabricated with hyaluronic acid (HA) hydrogels, particularly methacrylated hyaluronic acid (MeHA), as a way to augment the functionality of these constructs and the degree of control we have over their properties. HA hydrogels have been widely used both clinically and in research for various applications. When compared to the bioinert agarose hydrogels, HA is cell-responsive and is one of the main components of the ECM in the brain. There is also a vast trove of research into chemical modifications that can be applied to these hydrogels to enable fine-tuning of physical properties, the presentation of biochemical cues, and the release of growth factors and drugs. Thus, many types of modified HA hydrogels could be explored for TE-NSPs depending on the need and application. The use of these hydrogels may make TE-NSPs more biocompatible, versatile, adaptable to specific design criteria, and optimizable for their purpose of treating PD. Particularly, this may contribute to significantly improved outcomes as a result of better graft survival, preservation of the structure of the engineered axons in vivo, and integration with brain tissue. Moreover, expanding the functionality of the hydrogel micro-column beyond encasement and protection to deliver drugs and other molecules may turn TE-NSPs into a more inclusive approach that addresses other aspects of PD pathophysiology. Using embryonic rat dopaminergic aggregates, we confirmed that MeHA can be utilized to create TE-NSPs that mimic the nigrostriatal pathway in terms of structure, phenotype, and dopamine release and that can be implanted and survive in the rat brain. Notably, MeHA-encased TE-NSPs featured faster and longer axon growth and elicited a lesser astrocyte and microglia response in the brain after 6 weeks in vivo when compared to traditional 1% agarose encasement.
TE-NSPs have been designed to improve upon and circumvent the limitations of both conventional PD interventions and approaches centered on cell replacement and pathway reconstruction. On the first front, TE-NSPs directly address the underlying causes of PD by replacing the lost dopaminergic neurons in the SNpc and by reconstructing the cytoarchitecture of the dopaminergic fibers that encompass the nigrostriatal pathway. In doing so, TE-NSPs target both ends of the nigrostriatal circuit and may thus release dopamine in the striatum in a biologically- and temporally-controlled manner in response to their interaction with the host brain and with the spatial specificity provided by targeted innervation and synaptic communication. This contrasts with L-DOPA, other drug therapies, and electrical stimulation through DBS, which indirectly treat PD by managing symptoms in a nonspecific manner and tend to cause debilitating off-target effects.
In the case of biomaterials- and/or cell-based strategies, research efforts and clinical trials have concentrated on the injection of suspensions of dopaminergic neurons into the striatum to provide innervation in situ, overlooking the loss of cells in the SNpc and their nigrostriatal fibers. On the other hand, TE-NSPs involve implanting an entire nigrostriatal pathway substitute that matches the actual neuroanatomy and may synaptically integrate with targets around the SN and in the striatum. Having the dopaminergic neurons and axons in their proper location may restore the integration with other regions associated with the motor circuit, which should ultimately augment the therapeutic benefit of these constructs over cell replacement alone. Furthermore, TE-NSPs have a protective hydrogel encasement that may promote better cell survival during and after implantation and greater cell retention in the intended anatomical areas, in contrast with the low cell survival and the possibility for off-target cell presence observed with traditional implants.
Studies looking at ways to repair nigrostriatal fibers have focused on injecting neurons in the SN and either passively allowing them to extend axons in vivo or actively guiding their growth by making the local microenvironment amenable to growth and by implanting other cells along the pathway or in the striatum. Regardless, as previously mentioned, the extent of this in vivo growth would not be enough for the distances required in humans and oftentimes the resulting innervation lacked specificity. TE-NSPs ultimately avoid needing axons to grow through the perilous environment in vivo, as these constructs are implanted as a unit after growing axons with the required length, diameter, and density in vitro. This complete biofabrication in vitro bypasses the need for precise growth conditions, signaling cues, or axon pathfinding along existing pathways in the brain microenvironment that are necessary for extensive growth and innervation. The use of a tubular hydrogel for growing and encasing the axons may also help ensure that the growth is restricted to occur in a direction similar to the native pathway and that the innervation is spatially targeted to the striatum after implantation. The hydrogel may also be further modified to accommodate the more curved trajectory of the human nigrostriatal pathway. With TE-NSPs we can ideally control the phenotypic distribution in the aggregate and the kinds of projected axons in order to guarantee the type of striatal reinnervation that maximizes motor improvements. We can also exploit fabrication happening entirely in vitro to fully validate the final structure and function of the constructs before implantation, specifically in terms of axon growth as a function of time, viability, expression of mature/functional dopaminergic markers, and dopamine release. Thus, TE-NSPs can be a more reproducible, controllable, and consistent alternative than other less reliable methods that rely on the correct circumstances presenting themselves in vivo.
As previously suggested, the presence of the hydrogel encasement can also expand the capabilities of TE-NSPs to tackle the precise drivers of neurodegeneration. For example, the hydrogel micro-column could also be employed to deliver and release neurotrophic factors, such as GDNF, neurturin, and BDNF, that have been shown to possess neuroprotective effects on midbrain dopaminergic neurons and functional benefits in animals models and clinical trials after injection/infusion, viral-induced expression, or delivery in combination with cells. Similarly, the hydrogel could be leveraged to target other components of PD pathology, such as α-synuclein aggregation and mitochondrial dysfunction, as has been investigated through the delivery of heat-shock proteins, vectors to reduce α-synuclein gene transcription or expression, drugs to increase clearance of α-synuclein, among other approaches. Thus, our TE-NSP technology features several key benefits over existing and nascent approaches to treat PD. By leveraging different approaches taken thus far, inspiring its design on neuroanatomy and PD pathophysiology, incorporating tissue engineering, featuring an expansive degree of control and versatility, and being fully fabricated, optimized and assessed for quality in vitro before use, we believe TE-NSPs will be the ideal method to repair nigrostriatal fibers.
First, we characterized the growth length and rate of neurites projected from human dopaminergic aggregates within 1 cm-long micro-columns as a function of cell dose/number. We used aggregates having a total of 2.7×104, 5.3×104, or 1.1×105 neurons. According to the manufacturer, these neurons were 90% TH+; therefore, these aggregates did contain mostly dopaminergic neurons. Based on phase contrast imaging, it could be qualitatively determined that with higher cell numbers the neurites grow longer distances and have greater density (
A key necessity for any pathway reconstruction strategy is that grafts must be capable of releasing dopamine. Thus, we tested human-scale TE-NSPs for electrically-evoked dopamine release using fast scan cyclic voltammetry (FSCV). We transferred individual constructs to a recording chamber where stimulation and carbon fiber electrodes were both placed either within the dopaminergic aggregate or the axon tract region (
We sought to simulate the connectivity between dopaminergic axons and their striatal targets seen in the native nigrostriatal pathway within TE-NSPs. Thus, we fabricated human-scale TE-NSPs with an aggregate of around 2×105 striatal neurons, isolated from E18 rat pups and seeded 5 days after the dopaminergic aggregate within the end of the micro-column opposite to the dopaminergic one. Moreover, we sought to examine the effect of the presence of these cells on growth and dopamine release. On the first front, phase contrast images showed a qualitative improvement on axon growth length and density at the same distance from the dopaminergic aggregate in bidirectional constructs having the striatal aggregate (
Bidirectional, human-scale TE-NSPs with a striatal aggregate were also stained for dopaminergic (TH) and striatal MSN (DARPP-32) markers and imaged to evaluate the cytoarchitecture of the engineered tissue and physical connectivity between the aggregates. Both phenotypes were observed, with TH expressed strongly in the dopaminergic aggregate and axon tracts and DARPP-32 mainly seen in the striatal aggregate, adjacent neurites (assumed to be dendrites), and migrating neurons (
This Example describes the in vivo biomaterial response and the in vitro biofabrication of human stem cell derived TE-NSPs encased in methacrylated hyaluronic acid (MeHA) microcolumns, and characteristics of the human TE-NSPs.
To evaluate the rat brain response to implantation of MeHA columns along the nigrostriatal pathway, acellular hydrogel micro-columns fabricated with % agarose (n=3), 3% MeHA (n=5), and 5% MeHA (n=5), respectively, were implanted along the nigrostriatal pathway of athymic rats for 6 weeks. The host brain response was then assessed. For NeuN+ counts, an increase in counts was observed with increasing distance from the implant (
Next, the in vitro cytoarchitecture and growth profiles of rat-scale human TE-NSPs encased in MeHA hydrogel was evaluated. The human iPSC-derived dopaminergic neurons displayed aggregate and axonal tracts as well as expression of dopaminergic neuron markers (
The human TE-NSPs tested evoked dopamine release in vitro. In experiments using rat-scale human TE-NSP with a dopaminergic aggregate at one end (see
In summary, the data in this Example shows that in vitro biofabricated human stem cell derived TE-NSPs encased in MeHA microcolumns release dopamine, display dopaminergic phenotypes, and promote neurite growth.
This Example describes the in vivo implantation and efficacy of human stem cell derived TE-NSPs encased in MeHA microcolumns in athymic rat models of Parkinson's Disease (PD). The implanted TE-NSPs had neurons that persisted long-term in vivo (for at least 6 months), were able to innervate out of the construct and into the striatum, and demonstrated physiological functions.
Human TE-NSPs were implanted in a rat model of PD. At 12 weeks post implantation, the animals were sacrificed and brain sections were obtained for histological assessments. As shown in
Axon preservation and striatal innervation was also observed in this rat model upon implantation of TE-NSPs. At 12 weeks post-implantation, preserved inner axon tracts and axons ending near the striatum were observed (
Next, scaled-up human TE-NSPs were generated.
In addition to persisting in vivo for 6 months, the scaled-up TE-NSPs also produced outgrowths of human and dopaminergic projections into the dorsal striatum from the implanted scaled-up TE-NSPs was observed (
In summary, the data in this Example shows the in vivo efficacy of human-scale, human iPSC-derived TE-NSPs. The demonstration of long-term persistence, outgrowth, and functionality of TE-NSPs (e.g., as assessed by reinnervation (see
The entire contents of all patents, published patent applications, and other references cited herein are hereby expressly incorporated herein in their entireties by reference.
The present application is entitled to priority under 35 U.S.C. § 119(e) to U.S. Provisional Patent Application No. 63/190,581 filed May 19, 2021, the content of which is incorporated by reference herein in its entirety.
This invention was made with government support under I01-BX003748 awarded by Department of Veterans Affairs, 1845298 awarded by the National Science Foundation, IK2-RX002013 awarded by the Department of Veterans Affairs, U01 NS094340 awarded by the National Institutes of Health. The government has certain rights in the invention.
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/US22/29908 | 5/18/2022 | WO |
| Number | Date | Country | |
|---|---|---|---|
| 63190581 | May 2021 | US |
