The Effect of Polyethylene Glycol Linked Multi-Walled Carbon Nanotubes on Myogenic Differentiation of Human Mesenchymal Stem Cells for Skeletal Muscle Engineering

The effect of polyethylene glycol linked multi-walled carbon nanotubes on myogenic differentiation of human mesenchymal stem cells for skeletal muscle engineering

Chunyan Zhao, Giorgia Pastorin, Han Kiat Ho

Department of Pharmacy, National University of Singapore, Singapore

INTRODUCTION

The reconstruction of skeletal muscle defects has been a continuing challenge and the need for improved treatments has thus motivated the research on skeletal muscle tissue engineering. Human mesenchymal stem cells (hMSCs) have attracted much attention in tissue engineering due to their wide range of sources, unique capabilities of self-renewal in an undifferentiated state for prolonged time, and multi-lineage differentiation upon proper stimuli [1]. They are attractive, being readily isolated from bone marrow, umbilical cords, umbilical cords blood, lacrimal glands, glomeruli and other easily accessible sources such as adipose tissue and peripheral blood. Advances in stem cell biology have shown that hMSCs can differentiate into a variety of connective tissue cells including osteocytes, chondrocytes, adipocytes and myocytes, which envisage the versatile applications of hMSCs. Hence, to repair and regenerate skeletal muscle tissue, the employment of hMSCs is promising, triggering investigative works to improve its differentiation toward myogenic lineage.

Carbon nanotubes (CNTs), cylindrical tubes of rolled graphene sheets comprising of honeycomb structure of carbon atoms, have been at the forefront of nanotechnology due to their unique mechanic, thermal, electrical and chemical features. However, the pristine CNTs are extremely hydrophobic and rapidly precipitate in aqueous solutions, mitigating their physiological relevance. To improve the property of CNTs, polyethylene glycol (PEG) is often used to modify CNTs¡¯ surfaces and increase their hydrophilicity and biocompatibility. For the first time ever reported, our group found that the thin films of PEG linked multi-walled CNTs (PEG-CNTs) were not cytotoxic and accelerated the osteogenic differentiation of hMSCs, to a similar extent of hMSCs cultured with a commonly used growth factor, the bone morphogenetic protein-2 (BMP-2) [2]. Similarly, we found that graphene sheets also accelerated osteogenic differentiation of hMSCs, which was comparable to the one achieved with BMP-2 [3]. It was postulated that the intrinsic nanostructure of PEG-CNTs and the honeycomb structure of carbon atoms in the outer layer of PEG-CNTs helped in accelerating the osteogenic differentiation of hMSCs. Therefore, this study is driven by exploiting the breadth of this effect with investigation of PEG-CNT films¡¯ role in the myogenic differentiation of hMSCs, with the hope of improvements in skeletal muscle engineering.

METHODS

The PEG-CNT films were prepared by dropping PEG-CNTs suspension onto pre-heated cover slips. When dried, the films were characterized for surface roughness (helium ion microscopy, HIM; and atomic force microscopy, AFM), thickness (scanning electron microscopy, SEM) and stiffness (AFM). Subsequently, the hMSCs were cultured on PEG-CNT films and were induced for myogenic differentiation by incubating with normal medium for one week and myogenic medium (normal medium with dexamethasone and hydrocortisone) for another two weeks. The myogenically-induced and non-induced hMSCs were examined by cell morphology (fluorescent rhodamine-phalloidin staining) and cell viability (CellTiter-Glo assay). The quantitative real-time polymerase chain reaction (RT-PCR) was then employed to determine the extent of myogenesis by monitoring changes of hMSCs specific genes (CD73, CD90 and CD105), myogenic genes (Myf-5, MyoD, desmin and myosin heavy chain, MHC) and skeletal muscle specific genes (fast skeletal troponin C, TnC; and ryanodine receptor 1, Ryr) in the differentiated hMSCs. To verify the skeletal myogenic differentiation of hMSCs, the non-induced hMSCs on PEG-CNT films were also characterized with osteogenic genes (collagen-1, Col-1; osteocalcin, OCN; alkaline phosphatase, ALP; osteopontin, OPN).  

RESULTS

The HIM image of PEG-CNT films showed that the PEG-CNT film exhibited smooth surface and orderly fashion of single PEG-CNT (Figure 1 A). The AFM analysis substantiated the HIM result and the PEG-CNT films had small values in the roughness parameters (Table 1). The estimated thickness of PEG-CNT films was around 5¦Ìm (Figure 1 B) while the average elastic modulus for the PEG-CNT films was 944¡À474MPa, indicative of high stiffness.

Over the three weeks of incubation, rhodamine-phalloidin staining of hMSCs showed that PEG-CNT films did not alter the hMSCs morphology (Figure 2). The non-induced hMSCs retained a spindle shape, which is similar with that observed in human skeletal muscle cells (SKMCs). However, the induced hMSCs adopted rounder shape, which is different from the shape of SKMCs.

The cell viability assay showed a slight decrease in the viability after the myogenic induction of hMSCs on cover slips (Figure 3). Significant inhibition of cell viability in hMSCs on PEG-CNT films was observed compared to the control (non-induced hMSCs on cover slips), regardless of presence or absence of myogenic induction. As cells differentiate, their rate of proliferation usually decreases, thus cell proliferation potential decreased with increasing differentiation [4, 5]. Coupling to the lack of overt cell death based on fluorescent staining described earlier, the decrease of cell viability on PEG-CNT films might be due to myogenic differentiation of hMSCs and hence suppression of proliferation.

The non-induced hMSCs on cover slips and PEG-CNT films presented a set of highly expressed CD genes (CD73, CD90, CD105) (Figure 4 A). With myogenic induction, it was shown that the hMSCs seeded on both cover slips and PEG-CNT films had a reduction of these CD genes. The suppressed hMSCs feature gene expression indicated that the induced hMSCs has diminished hMSC characteristics, thus may be more prone to myogenic differentiation. Comparing to the control (non-induced hMSCs on cover slips), the expression of myogenic genes and skeletal muscle specific genes was weakly increased in the myogenically-induced hMSCs on cover slips and highly increased on PEG-CNT films (Figure 4 B). The up-regulation of myogenic markers verified the myogenic phenotype of investigated cell culture and indicated that the differentiation protocol was capable of guiding myogenic differentiation of hMSCs. The TnC and Ryr are primarily expressed in skeletal muscle among the three types of muscle (skeletal muscle, cardiac muscle and smooth muscle) [6]. Therefore, the higher expression of TnC and Ryr in myogenically-induced hMSCs indicated that the hMSCs derived myobalsts may be committed towards skeletal muscle cells. Interestingly, these myogenic genes and skeletal muscle specific genes were also significantly increased in the non-induced hMSCs on PEG-CNT films compared to the control. This occurrence suggests that PEG-CNT films alone, in the absence of myogenic inducers, such as dexamethasone and hydrocortisone, could trigger the myogenesis of hMSCs. The advantage of this effect is the possible removal of potentially noxious dexamethasone and hydrocortisone from the induction protocol. Today, there is no well recognized way of controlling the optimal concentrations of the inducers for efficient differentiation with reduced or no side effects in vivo. Therefore, the observed myogenic commitment of hMSCs by PEG-CNT films in the absence of myogenic inducers is a major advancement of hMSCs application in skeletal muscle engineering. Furthermore, the non-induced hMSCs on PEG-CNT films did not demonstrate higher expression of Col-1, OCN and ALP compared to the control, indicating a lack of concurrent osteogenesis during the skeletal myogenesis (Figure 4 C).

CONCLUSION

In this study, we created PEG-CNT films with slight surface roughness with the PEG-CNTs aligned in orderly fashion on the surface. The PEG-CNT film thickness was approximately 5¦Ìm with exhibition of significant stiffness (944¡À474MPa). The films supported the culturing of hMSCs and preserved the hMSCs morphology as compared to cover slip controls. Upon myogenic induction, hMSCs adopted new morphology with rounder shape while the non-induced hMSCs retained the spindle shape. Based on RT-PCR results, myogenic induction of hMSCs was successful on cover slips and PEG-CNT films with myogenic medium. Moreover, the PEG-CNT films alone could specifically trigger the myogenic differentiation of hMSCs (without concurrent osteogenesis), although somewhat at the expense of the viability of hMSCs, an expected secondary effect during differentiation process. Therefore, the PEG-CNT films show promise as a potential scaffold in guiding myogenic differentiation of hMSCs for skeletal muscle tissue engineering. This was the first study to investigate the CNTs¡¯ influence in the myogenesis of hMSCs. Further studies will be conducted to substantiate this finding with myogenic protein expression and physiological function tests.

REFERENCE

[1] C. Zhao, et. al, Biotechnology advances, 31(2013) 654-668.

[2] T.R. Nayak, et. al, ACS nano, 4 (2010) 7717-7725.

[3] T.R. Nayak, et. al, ACS nano, 5 (2011) 4670-4678.

[4] M. Olbrich, et. al, PloS one, 7 (2012) e47176.

[5] C. GM, The Cell: A Molecular Approach. 2nd edition. , 2000.

[6] R. Gahlmann, et. al, Journal of molecular biology, 201 (1988) 379-391.

Figures:

Figure 1. (A) HIM image of the surface topography of PEG-CNT films; (B) SEM image at the cross section of PEG-CNT films.

Figure 2. hMSCs morphology on cover slips and PEG-CNT films during three weeks incubation.

Figure 3. Viability of hMSCs induced to myogenic differentiation as measured by CellTiter-Glo assay. The asterisks (*) indicate statistically significant difference (p<0.05) between the sample and the control (non-induced hMSCs on cover slips).

Figure 4. (A) Fold change of (A) hMSCs specific genes, (B) myogenic genes and (C) osteogenic genes with 2^{- ∆∆CT}. The asterisks (*) indicate statistically significant difference (p<0.05) between the samples and the control (non-induced hMSCs on cover slips).

Table:

Table 1. Surface roughness of PEG-CNT films with AFM