Nanofibrous spongy microspheres to deliver rabbit mesenchymal stem cells and anti-miR-199a to regenerate nucleus pulposus and prevent calcification

Abstract

Lower back pain is mainly caused by intervertebral disc degeneration, in which calcification is frequently involved. Here novel nanofibrous spongy microspheres (NF-SMS) are used to carry rabbit bone marrow mesenchymal stromal cells (MSCs) to regenerate nucleus pulposus tissues. NF-SMS are shown to significantly enhance the MSC seeding, proliferation and differentiation over control microcarriers. Furthermore, a hyperbranched polymer (HP) with negligible cytotoxicity and high microRNA (miRNAs) binding affinity is synthesized. The HP can complex with anti-miR-199a and self-assemble into "double shell" polyplexes which are able to achieve high transfection efficiency into MSCs. A double-emulsion technique is used to encapsulate these polyplexes in biodegradable nanospheres (NS) to enable sustained anti-miR-199 delivery. Our results demonstrate that MSC/HP-anti-miR-199a/NS/NF-SMS constructs can promote the nucleus pulposus (NP) phenotype and resist calcification in vitro and in a subcutaneous environment. Furthermore, injection of MSC/HP-anti-miR-199a/NS/NF-SMS can stay in place, produce functional extracellular matrix, maintain disc height and prevent intervertebral disc (IVD) calcification in a rabbit lumbar degeneration model.

Keywords: Anti-microRNA oligonucleotides; Calcification; Mesenchymal stem cells; Nucleus pulposus; Rabbit; Regeneration.

Fig 1.. SEM images of different microcarriers and rabbit mesenchymal stromal cell (MSC) attachment.

(A) solid microspheres (S-MS); (B) nanofibrous microspheres (NF-MS); (C) Nanofibrous spongy microspheres (NF-SMS). All the above microcarriers have a diameter ranging from 30 to 60 μm (Scale bars for A-C: 20 μm). MSCs were seeded on the three types of microspheres for 24h: (D) MSCs seeded on S-MS surface and spread over a large area; (E) MSCs seeded on the exterior of NF-MS; (F) MSCs seeded on both the exterior and the interior of the NF-SMS. (Scale bars for D-F: 10 μm).

Fig 2.. NF-SMS provided a supportive microenvironment for nucleus pulposus-like differentiation of rabbit MSCs.

(A-F) Safranin-O staining demonstrating GAG accumulation. Both NF-SMS and NF-MS were positively stained by safranin-O for GAG, while a weak staining was found in S-MS group. (G) Proliferation curves of MSCs cultured with NF-SMS, NF-MS, and S-MS, quantified using DNA content at each predetermined time. (H) GAG/DNA ratio in the NF-SMS, NF-MS, and S-MS groups. (I-K) After 2 and 4 weeks of induction, real time RT-PCR analysis showed different expression levels of nucleus pulposus-associated genes (Sox-9, type II collagen and aggrecan) in MSCs on different microcarriers (NF-SMS, NF-MS and S-MS). For all analyses, n=3, * indicates p<0.05.

Fig 3.. Two-stage delivery of anti-miR-199a.

(A-C) The HP-anti-miR-199a polyplexes were prepared, encapsulated in PLGA nanospheres (NS) via a double emulsion method, and then attached onto the PLLA NF-SMS. (D-i) TEM image showing anti-miRNA distribution (the darker rings): Anti-miRNA complexed with the positively charged outer shell of HP molecular core, the HP-anti-miRNA polyplexes self-assembled into a larger nanosized spherical shell sandwiched between an inner PEG core and an outer PEG layer. (D-ii) SEM image of PLGA NS that contain HP-anti-miRNA polyplexes. (D-iii) SEM image of a part of an NF-SMS. (D-iv) SEM image of a part of an NF-SMS with immobilized PLGA NS. (E) Confocal images of cells transfected for 2 days with Cy3-labeled anti-miR-199a (orange red) and DAPI stained nuclei (blue). A significantly higher transfection efficiency was shown in the HP-anti-miR polyplex group than in the anti-miR alone group. (F) Immunofluorescence staining for Hif-1α in MSCs cultured on NF-SMS with HP-carried anti-miR-199a, miR-199a, or control. Green: Hif-1α; red: NF-SMS. Enhanced HIF-1astaining was seen in the anti-miR-199a group, while decreased Hif-1α staining was seen in the miR-199a group. All scale bars in E&F are 50μm.

Fig 4.. Histological and immunohistochemical analyses of subcutaneous implanted constructs.

Rabbit MSC/HP-anti-miR-199a/NS/NF-SMS were induced by TGF-β1 for 3 weeks and then implanted into subcutaneous pockets of nude mice (n=4. Note: HP and NS/NF-SMS are omitted in the images to simplify the labels). Rabbit MSC/HP-anti-miR-199a/NS/NF-SMS without TGF-β1 induction and rabbit MSC/HP-miR-NC/NS/NF-SMS with TGF-β1 induction were used as controls. A) Histological Analysis: After 8 weeks of implantation, the rabbit MSC/HP-anti-miR-199a/NS/NF-SMS constructs with TGF-β1 induction were stained positive by safranin-O and negative by von Kossa. In contrast, rabbit MSC/HP-anti-miR-199a/NF-SMS constructs without TGF-β1 induction were stained negative by either safranin-O or von Kossa, while rabbit MSC/HP-miR-NC/NS/NF-SMS constructs with TGF-β1 induction were stained positive by both safranin-O and von Kossa methods. B) Immunohistochemical Analysis: Eight weeks after implantation, the rabbit MSC/HP-anti-miR-199a/NS/NF-SMS constructs with TGF-β1 induction were stained positive by type II collagen and Hif-1α antibodies, but stained weakly by type X collagen antibody. In MSC/HP-miR-NC/NS/NF-SMS with TGF-β1 induction group, the constructs were stained positive by type X collagen antibody but negative by either type II collagen or Hif-1α antibodies. Rabbit MSC/HP-anti-miR-199a/NS/NF-SMS constructs without TGF-β1 induction were stained negative by type II collagen, type X collagen and Hif-1α antibodies.

Fig 5.. Anti-miR-199a targets Hif-1α to functionally inhibit calcification.

(A) The Hif-1α gene expression (upper panel) and protein level (lower panel) in rabbit MSCs in response to the treatment of either miR-199a or its negative control (miR-NC). (B) Schematic illustration of designing the luciferase reporters with the WT Hif-1α 3’UTR (Hif-1α 3’UTR WT) or the site-directed mutant Hif-1α 3’UTR (Hif-1α 3’UTR mutated (MUT)). (C) The effect of miR-199a or miR-NC on luciferase activity in either the Hif-1α 3’UTR WT or the Hif-1α 3’UTR MUT transfected MSCs. (D) The Hif-1α gene expression (upper panel) and the Hif-1α protein level (lower panel) in MSCs after silencing Hif-1α with a Hif-1α-specific siRNA (siR-Hif-1α). (E) Western blot analysis of Sox9, Col-II and Col-X proteins in MSCs after siR-Hif-1α or its NC treatment for 1 week. (F) The Sox9 gene expression (upper panel) and Sox9 protein level (lower panel) in MSCs after treatment with a Sox9-specific siRNA (siR-Sox9). (G) Real-time RT-PCR analysis of the mRNA levels of Col II and Col X in MSCs after silencing Sox-9 with siR-Sox9 and being treated with miR-199a, anti-miR-199a, or their corresponding NCs. (H) Western blot analyses of Sox-9, Col II and Col X protein contents in MSCs treated with miR-199a, anti-miR-199a, and their corresponding NCs. (I) Schematic diagram of calcification activity mediated by anti-miR-199a via modulating Sox-9 signaling pathway in MSCs. Intracellular miR-199a level was suppressed by anti-miR-199a. Subsequently, Hif-1α activity was enhanced by reduced miR-199 expression, and therefore to promote Sox-9 activity, which resulted in inhibition of calcification. * p<0.05. Triplicates were analyzed for all experiments, n=5 for every group. Data are expressed as mean ± SD.

Fig 6.. Representative gross images (A-D) and fluoroscopy images (E-H) of rabbit lumbar spines at 6 weeks after in vivo gene therapy.

In the sham group (punctured without treatment), the disc space nearly completely disappeared in the central area of the NP. Ectopic bone tissue formed as indicated by high intensity shadows in X-ray images in the sham group. Regeneration characteristics were found in the discs treated with MSC/HP-anti-miR199a/NS/NF-SMS, showing higher disc heights as compared to the sham group. Notably, this treatment group also showed more regular disc space with matrix translucency in gross appearance and less ectopic bone formation (observed under X-ray) compared to the sham group and MSC/NF-SMS groups.

Fig 7.. NP regeneration and disc height maintenance after different treatments.

A) Three-dimensional reconstructions of μCT scans of rabbit lumbar spines and representative cross-sections 6 weeks after in vivo MSC/HP-anti-miR-199a/NS/NF-SMS treatment along with three control groups. B) Quantification of osteophyte bone volumes after different treatments. C) The disc height index (DHI) changing over time. * p<0.05, ** p<0.01 for comparison against MSC/HP-anti-miR-199a/NS/NF-SMS group.

Fig 8.. HE and safranin-O (SO) staining of disc samples of different treatment groups 6 weeks after injection (A-P):

In the normal control group (A, E, I, M), the volume of the oval-shaped NP was large as shown in the midsagittal cross-section of the HE staining (A), the safranin-O staining in the NP area was strong (E&I). In the sham group (B, F, J, N), there was minimal disc space (B), substantial fibrous tissue invasion and weak GAG staining (F&J). In the discs treated with MSC/HP-anti-miR-199a/NS/NF-SMS (D, H, L, P), the NP area was similar to that in the normal control and safranin-O staining was similarly strong. In the MSC/NF-SMS group (C, G, K, O), the NP area was stained positively by safranin-O, but the staining intensity was weaker than MSC/HP-anti-miR-199A/NS/NF-SMS group. The bony endplate (BEP) thickness and calcification were substantially greater in the sham group (N) and MSC/NF-SMS (O) group compared to the MSC/HP-anti-miR-199a/NS/NF-SMS group (P). Assessment of histological appearance and calcification (Q&R): The histological appearance of the treated discs was semi-quantitatively scored 12 weeks after treatment (Q). The calcification area/endplate area ratio was quantified at 12 weeks after treatment (R). Scale bar: 200μm for A-H & M-P, 500 μm for I-L. **p <0.01.