Directing osteogenic and myogenic differentiation of MSCs: interplay of stiffness and adhesive ligand presentation
Article Link: http://ajpcell.physiology.org/cgi/content/abstract/295/4/C1037
Citation: Rowlands AS, George PA, Cooper-White JJ. Directing osteogenic and myogenic differentiation of MSCs: interplay of stiffness and adhesive ligand presentation. Am J Physiol Cell Physiol 2008;295:C1037–44.
Summary:
Mesenchymal stem cells (MSCs) have been shown to differentiate into multiple cell types including: chondrocytes, adipocytes, myocytes, and osteocytes, depending on growth media constituents and reagents particular to cell type differentiation. MSCs have been known to alleviate inflammation by migrating to affected sites and play an important role in tissue regeneration through secretion of chemokines as well as engrafting onto damaged tissue. More importantly, the authors of the paper cite the interaction between the cell and the extracellular matrix (ECM) for influencing the differentiation capability of MSCs. Integrins are the transmembrane receptors responsible for binding to the ECM and generating cytoskeletal tension via action-myosin contraction, thereby forming a focal adhesion complex, which plays a major role in cell mobility, adhesion and cell signaling.
The main objective of the study was to determine whether both substrate stiffness and ECM protein content affected the differentiation capability of MSCs into osteocytes and myocytes. The researchers initially coated the surface of polyacrylamide gels with tissue specific ECM proteins: collagen I, collagen IV, laminin, and fibroncetin. The polyacrylamide gels were constructed with varying stiffness: 0.7 kPa, 9 kPa, 25 kPa, and 80 kPa, by altering the density of cross-links. This was done by increasing the amount of bis-acrylamide that resulted in a higher young’s modulus which is the prime indicator of the stiffness of a material. ECM proteins were bound to the polyacrylamide gel using a photoactivatable cross-linker, N-Succinimidyl-6-(4'-azido-2'-nitrophenylamino) hexanoate (SANPAH). Differentiation of MSCs into myocytes were measured using antibody stains for MyoD1, a myogenic marker whereas, differentiation of MSCs into osteocytes were measured using an antibody stain for Runx2, a bone marker.
Significance:
The results show that generally, higher substrate stiffness resulted in greater cell attachment as well as greater area of cell occupation or spreading. The stiffness of the substrate also affected the morphology of the cells, as cells tended to be rounded and occupied much less space on softer or less stiffer substrates (0.7 kPa). On more stiffer substrates, the morphology of cells was more elongated and spiny-like(> 9 kPa). The authors also noted that ECM protein content also affected cell attachment in conjunction with substrate stiffness. For example, comparing the stiffness vs. cell attachment for the various proteins, it can be seen that fibronectin produced the greatest cell attachment in substrates with stiffness of 25 kPa (Refer Fig. 4). Similarly, comparing stiffness vs. cell spread area, it can be seen that collagen IV produces the greatest cell area for stiffness of 25 kPa. (Refer Fig. 4). These results show that cell attachment and cell spread area are dependent on both ligand protein adhesion in the ECM along with substrate stiffness.
Fig. 4. MSC attachment (A) and spread area (B) on the gel-protein substrates after 24 h
Potential differentiation of MSCs into osteogenic or myogenic cells were characterized based on the detection of corresponding transcription factors Runx2 and MyoD1 through fluorescent antibody stains. The researchers found that Runx2 was strongly expressed in the 80kPa collagen I-coated polyacrylamide gel. The researchers justified this finding based on the natural microenvironment of bone which is comprised of 80% collagen I and is consistent of a stiffness close to 80kPa. MyoD1 was found to be strongly expressed in the 80kPa collagen I-coated gel as well as the 25kPa fibronectin-coated gel (refer Fig. 8). This again is consistent with the microenvironment of muscle cells in which 90% of the perimysial ECMis composed of collagen I and fibronectin also plays a major role in the ECM. These results show that a variance in both the ECM protein along with stiffness affect the differentiation potential of MSCs into both myogenic and osteogenic cells. Fig. 8. Relative intensity data for the expression of MyoD1 and Runx2 by MSCs cultured on substrates of varying stiffness and ligand.
Fig. 8. Relative intensity data for the expression of MyoD1 and Runx2 by MSCs cultured on substrates of varying stiffness and ligand
The implications of this research are immense as they provide a safe new outlet for differentiating MSCs, compared to the current protocol that uses toxic DNA demethylation agents as well as expensive growth factors. Also, potentially, based on the specificity of the substrate stiffness and ECM protein coating, MSCs can be more viably differentiated and proliferated into desired cell tissue types. More importantly, the results of the research stress the importance of providing a proper microenvironment, similar to one found in vivo, to study the effects of stem cells. Producing a microenvironment that simulates one found in the body may help produce better pathology models and accurately study how tissue regeneration works.
Critique:
Based on the experimental results, the authors conclude that both substrate thickness and ECM protein content affect the differentiation of MSCs into osteocytes or myocytes. Although both of the factors do seem to affect differentiation potential, the degree of the effect is not known. The conclusion cannot be completely verified because the presence of the transcription factors/markers Runx2 and MyoD1 does not guarantee that the stem cell will in fact differentiate into corresponding osteocytes or myocytes. The problem is, as the authors admit, the concentrations of Runx2 as compared to MyoD1 needed to cause differentiation of MSCs is not well known. More research needs to be done regarding the concentrations of transcription factors and their effect on differentiation capability. Additionally, it is possible that other significant cofactors that are necessary for differentiation of MSCs are not taken into account and simply comparing RunX2 and MyoD1 might not be accurate. Taking into account of these cofactors can help validate the differentiation potential of MSCs.
11 comments:
They test a range of mechanical stiffness - how do these compare with actual tissues in the body?
Also: what actual mechanics would be changed by substrate thickness?
It sounds like the authors of this paper are primarily interested in identifying myogenic vs osteogenic culture factors. Did they try to visually identify whether the MSCs differentiated into bone or muscle progenitors? I would expect that the morphology between an osteocyte and a myocyte to be significantly different. This may cut down on some of the ambiguity of the fluorescent probe results.
A few other side notes: I looked online a bit but couldn't find data on the crosslinker (SANPAH). What wavelength activates it? Is there any chance the acrylamide gel is further crosslinked via UV polymerization during that process (thus affecting modulus)? Also, were the cells seeded in serum media? The presence of FBS may have confounding effects on adhesion studies.
-Steven Zhao
"For example, comparing the stiffness vs. cell attachment for the various proteins, it can be seen that fibronectin produced the greatest cell attachment in substrates with stiffness of 25 kPa (Refer Fig. 4)."
Looking at Figure 4A, the error bar for fibronectin is huge and overlaps with Collagen I and Laminin. Do you think this changes the conclusions drawn from the authors?
Also, if Figure 8 shows the relative levels of MyoD1 and Runx2, what does it mean if cells in a particular condition express both differentiation markers at a high level (i.e. Collagen I, 80 kPa)?
There are definitely large standard error bars for cell attachment, but the trend seems to be that cells attach better around 9kPa than at other stiffnesses. I wonder if this will limit our ability to try to induce differentiation at other stiffnesses. If cell attachment is a problem then substrate stiffness becomes a variable we can't mess with too much.
Since mesenchymal stem cells (MSCs) can differentiate into many cell types, do you think merely changing the substrate stiffness and protein will be enough to fully differentiate the MSCs into osteoblasts and myocytes? I think it may be inadequate, which is why the researchers looked at markers to see whether the cells start to lean towards a type of cell, rather than trying to see fully differentiated cells.
Also, the stiffness ranges of muscle and cartilage overlap. How do the researchers make sure that the stiffnesses they tested won't promote cartilage differentiation?
@ Terry:
According to the article, stiffness of brain ranges from 0.1 - 1 kPa, muscle: ranges from 8-17 kPa, and bone is greater than 34 kPa. The authors tested stiffnesses ranging from 0.7 to 80 kPa which covers the range of in vivo tissue stiffness. Also by changing substrate thickness, the interaction between the cells and ECM would change depending on what type and how much ECM proteins are being secreted.
@ Steven:
The authors did not just visually try to determine whether the MSCs had differentiated into osteocytes or myocytes based on morphology.The authors relied on expression of differentiation markers Runx2 and MyoD1 as the primary evidence. I looked online and SANPAH is activated at about 350 nm and there is a possibility that the gel is further crosslinked by UV. Also, the cells were grown in FBS and since the control group was also grown in FBS, it is a possibility that it could have affected adhesion studies.
@ michelle ho:
It is true that the error is high for fibronectin however the authors are not focused on which ECM protein produces the greatest differentiation rate but rather are trying to justify the use of substrates of varying stiffness for MSC differentiation instead of using expensive growth factors. Also for figure 8, the scaling for both the graphs is different so in the case of Collagen 1, MyoD1 expression is about twice as much.
@ michelle marcus:
Looking at Figure 4, 25 kPa stiffness is the most optimal for cell attachment even if the greatest error is applied but yes, substrate stiffness cannot be varied too much.
@ mimi:
I agree with you that only using substrate stiffness and different protein ligands will not guarantee the path of differentiation MSC takes. The researchers were interested in finding markers that may indicate the eventual destination of differentiation. The authors cannot ensure that the MSC develops into muscle by using a stiffness similar to that of cartilage but can test for the myod1 marker and see whether the protein ligand affects differentiation because muscle and cartilage will have different ECM protein compositions.
Post a Comment