Creating a stem cell niche in the inner ear using self-assembling peptide amphiphiles

Abstract

The use of human embryonic stem cells (hESCs) for regeneration of the spiral ganglion will require techniques for promoting otic neuronal progenitor (ONP) differentiation, anchoring of cells to anatomically appropriate and specific niches, and long-term cell survival after transplantation. In this study, we used self-assembling peptide amphiphile (PA) molecules that display an IKVAV epitope (IKVAV-PA) to create a niche for hESC-derived ONPs that supported neuronal differentiation and survival both in vitro and in vivo after transplantation into rodent inner ears. A feature of the IKVAV-PA gel is its ability to form organized nanofibers that promote directed neurite growth. Culture of hESC-derived ONPs in IKVAV-PA gels did not alter cell proliferation or viability. However, the presence of IKVAV-PA gels increased the number of cells expressing the neuronal marker beta-III tubulin and improved neurite extension. The self-assembly properties of the IKVAV-PA gel allowed it to be injected as a liquid into the inner ear to create a biophysical niche for transplanted cells after gelation in vivo. Injection of ONPs combined with IKVAV-PA into the modiolus of X-SCID rats increased survival and localization of the cells around the injection site compared to controls. Human cadaveric temporal bone studies demonstrated the technical feasibility of a transmastoid surgical approach for clinical intracochlear injection of the IKVAV-PA/ONP combination. Combining stem cell transplantation with injection of self-assembling PA gels to create a supportive niche may improve clinical approaches to spiral ganglion regeneration.

Fig 1. Schematic summary of the protocol and timeline for deriving the SGN lineage from undifferentiated hESCs.

DIV: day in vitro; NNE: nonneuronal ectoderm; PPE: preplacodal ectoderm; ONP: otic neural progenitor; BMP4: bone morphogenetic protein 4; SHH: Sonic hedgehog; ATRA: all-trans retinoic acid; EGF: epidermal growth factor; BDNF: brain-derived neurotrophic factor; NT-3: neurotrophin 3; IGF-1: insulin-like growth factor 1; FGF2: fibroblast growth factor 2; E8: Essential 8^™ medium; N2B27-CDM: chemically defined medium containing N2 and B27 supplements; MACS: magnetic-activated cell sorting; FACS: fluorescence-activated cell sorting; p75: low-affinity neurotrophin receptor (p75^NTR); IKVAV-PA: IKVAV-containing peptide amphiphile; VVIAK-PA: VVIAK-containing peptide amphiphile. Protocol adapted and modified from Matsuoka et al. [8].

Fig 2. Immunocytochemical assessment of differentiation of mid-stage hESC-derived ONPs into late-stage ONPs.

(A): An experimental paradigm. Immunocytochemistry of late-stage ONPs derived from H1, H7, and H9 undifferentiated ESCs for SOX2 (B), PAX2 (C), PAX8/NEUROD1 (D), and GATA3/PAX8 (E). Human hESC cell lines used were H7 (C), H9 (B), and H1 (D, E). Scale bars: 50 μm (B and D) and 20 μm (C and E). m-ONP: mid-stage ONPs; l-ONP: late-stage ONPs; SGN: spiral ganglion neurons; S/R/E/F/I: SHH/ATRA/EGF/FGF2/IGF-1.

Fig 3. Neuronal differentiation of hESC-derived late-stage ONPs in IKVAV and VVIAK 3D matrices.

(Aa): An experimental paradigm on neuronal differentiation of hESC-derived mid-stage ONPs treated with S/R/E/F/I in IKVAV & VVIAK PA gels (3-D) and Matrigel™ (2-D) for seven days. DIV: days in vitro. (Ab): A photomicrograph of phase-contrasted image of a mid ONP treated with S/R/E/F/I in IKVAV-PA gels. A White arrow indicates a neurite. Scale = 20 μm. (B): Immunocytochemistry of nestin and β-III tubulin for hESC-derived mid-ONPs treated with S/R/E/F/I in IKVAV-PA gels. DAPI stain is shown in blue. A white arrow indicates neurites. Scale bar: 20 μm. (C): Quantification of nestin and β-III tubulin-immunopositive cells on derived mid-stage ONPs treated with S/R/E/F/I in IKVAV-PA gels, VVIAK-PA gels, and Matrigel™ for 7 days. β-III: β-III tubulin. (D): Quantification of neurite-bearing cells on derived mid-stage ONPs treated with S/R/E/F/I. in IKVAV-PA gels, VVIAK-PA gels, and Matrigel™. (E): Quantification of average neurite length on S/R/E/F/I-treated hESC-derived mid-stage ONPs. ** p < 0.01, * p < 0.05 by one-way ANOVA with Tukey-Kramer’s post-hoc test.

Fig 4. CRISPR-Cas9-modified EGFP⁺ hESCs that differentiated towards late-stage ONP lineage possess late-stage otic neuronal progenitor characteristics.

(A): Schematic of AAVS1 safe harbor locus within PPP1R12C gene and targeting vector containing puromycin-resistance cassette (puro^®), CAG enhancer, and EGFP within 5’ and 3’ homology arms (HA). Arrows indicate PCR primers. (B): EGFP targeting was validated by PCR. H7: H7 human ESCs were used as a control (untargeted clone). (C): CRISPR-Cas9 modified EGFP⁺ undifferentiated hESCs in culture. (Ca): phase contrast, (Cb): fluorescence (EGFP⁺) imaging. CRISPR-CAs9 modified EGFP⁺ late-stage ONPs: (Cc): phase contrast, (Cd): fluorescence imaging. Scale bar = 100 μm. (D): RT-PCR quantification of OCT3/4 and NANOG expression measurements for H7 hESC, H7 late-stage ONPs (lONPs), and CRISPR-Cas9 modified EGFP⁺ late-stage ONPs. n.s.: not statistically significant, *** p < 0.001, and ** p < 0.01 by one-way ANOVA with Tukey-Kramer’s post-hoc test. (E): Immunocytochemistry of CRISPR-Cas9 modified EGFP⁺ late-stage ONPs for NEUROD1 (Ea), nestin (Eb), SOX2, and PAX8 (Ec). Scale bar = 25 μm.

Fig 5. IKVAV-PA gels enhance survival of transplanted EGFP⁺ in the X-SCID rat cochlea.

(A): An experimental paradigm. DPI: days post injection. (B): Anti-EGFP antibody labeling of EGFP⁺ cells in the basal turn of the XSCID cochlea. BT: basal turn. Scale bar = 200 μm. (C): Control experiment showing DMEM only injection into the cochlea. Scale bar = 200 μm. (D): Example high-power image of DAPI and EGFP profiles at cellular level used for subsequent quantification. Scale bar = 20 μm. (E): The percentage of EGFP⁺ profiles per section injected with IKVAV-PA gels in and with DMEM (n = 6 for each condition). (F): Comparison of EGFP⁺ profiles in each turn and the modiolus injected with IKVAV-PA gels in and with DMEM (n = 6 for each condition). Abbreviation: MO: modiolus; BT: basal turn; MT: mid turn; and AT: apical turn. * p < 0.05, ** p < 0.01, and *** p < 0.001.

Fig 6. Ex vivo cadaveric human temporal bone study.

(A): Schematized midmodiolar cross-section of a mammalian cochlea, showing injection needle approaching the modiolus. (B-C): Photomicrograph of the basal turn of the left cochlea in a cadaveric temporal bone (B) and in humans (C). Black arrows indicate direction of injection with IKVAV-PA gels. Note the direction of the fine needle for injection of IKVAV gel containing hESCs. A blue-dotted line shows the cochleostomy site. A white dotted line: round window. (D and E): Endoscopic IKVAV-PA gel injection into human modiolus. View of cochleostomy site using a 16-mm 0° rigid endoscope. A black dashed line marks cochleostomy boundary. A white dashed line indicates presumed plane of the scala vestibuli. (F): Artist’s rendition of the superior view of the human skull base, with black arrow showing the direction of injection of IKVAV gels with hESCs and anatomical landmarks. Black square corresponds to sectioned area in subsequent figures. (G-L): IKVAV-PA gels with hESCs in the IAC in two sets of human cadaveric temporal bones. (G and J): Middle cranial fossa view of the human cadaveric temporal bone before the hESC injection. TAMRA-tagged IKVAV gels (H) and corresponding autofluorescence measurement observed in normal temporal bone tissue abutting the IAC (I). TRA-1-81 tagged hESCs with magnified inset (K) and corresponding autofluorescence measurement (L). Abbreviations: RW: round window; C: cochleostomy site; ST: scala tympani; SV: scala vestibule; M: modiolus; N: spinal needle; ACF: anterior cranial fossa; MCF: middle cranial fossa; PCF: posterior cranial fossa. Cranial nerves are denoted by Roman numerals.

Fig 7. Electron microscopy study of IKVAV-PA gels in vitro and ex vivo human cadaveric tissue.

(A): SEM image of aligned IKVAV nanofibers in vitro (note predominant orientation in the direction of the shear force, indicated by red arrow). Scale bar = 1 μm. (B): Non-aligned IKVAV nanofibers in vitro showing a randomly oriented configuration due to fluid turbulence. Scale bar = 1 μm. (C): TEM image of self-assembled IKVAV-PA gel overlying the facial-vestibulocochlear nerve complex (indicated by a black circle and black arrows) following endoscopic transmastoid injection into the modiolus. Note the parallel alignment of self-assembled peptides indicated by the red arrow (11,000× magnification). Scale bar = 1 μm. (D): Control nerve specimen with no evidence of peptide scaffold for comparison (1,900× magnification). Scale bar = 1 μm. Note putative epineurium of the facial-vestibulocochlear nerve complex indicated by the white dotted line (C & D). Abbreviations: VIIVIII: facial-vestibulocochlear nerve complex.