Fat Stem Cells vs. Bone Marrow Stem Cells for Knee Treatments

The Centeno clinic blog site has an article on the use of bone marrow or fat-derived mesenchymal stem cells to treat cartilage erosion and meniscus problems in the knee. Dr. Centeno notes that the literature is greatly in favor of the use of bone marrow stem cells for orthopedic treatments rather than fat-derived MSCs. The studies recorded in the literature largely examine animal studies, but there are some human clinical trials that have used MSCs from varying sources to treat knee cartilage lesions. Dr. Centeno is completely correct when he says that the papers that have been published show that the bone marrow-derived MSCs out perform fat-derived MSCs when it comes to cartilage production.

There are however is a caveat that Dr. Centeno does not mention and that is the conditions under which the MSCs were grown or induced to make cartilage. Research has pretty clearly shown that not all MSC populations are created equal. Each cell type has distinct requirements for cartilage making, and fat-derived MSCs performed poorly in cartilage assays because they were subjected to protocols that were standardized using bone marrow-derived MSCs. Therefore, it is no wonder that the fat-derived MSCs did poorly relative to the bone marrow-derived MSCs in such assays.

In order to get MSCs to make cartilage in culture, biologists use specific culture conditions that try to re-enact the conditions in embryos where cartilage is first made.  If MSCs are grown in a simple culture dish, they will clump together and start to make cartilage (Winter A, et al., Arthritis Rheum 2003;48(2):418-429).  To encourage cartilage formation in cultured MSCs, the cells are centrifuged and the cell pellets are deposited in a culture medium that is specifically designed to induce cartilage making in MSCs.  This culture system, which is one of the most widely used in cartilage research, is called a pellet, aggregate or spheroid culture.

What is the culture medium that is used?  A basal medium for MSC cartilage making contains a basic cell culture medium (known as Dulbecco’s Modified Eagles Medium or DMEM, or alternatively, fetal calf serum)  and then a laundry list of other chemicals are added.  These chemicals include: dexamethasone, ascorbate (vitamin C), the amino acid proline, insulin, the iron-binding protein transferrin and selenous acid (Johnstone B, Het al., Exp Cell Res1998; 238:265-272; Mackay AM, et al., Tissue Eng 1998;4:415-428; Puetzer, Petitte and Loboa, Tissue Eng Part B Rev 2010;16(4):435-444). That is more or less the “bare bones” medium used for cartilage making.

In addition to all these basic stuff, growth factors are added.  Growth factors are proteins that bind to the surfaces to specific cells and induce particular behaviors in those cells.  Growth factors bind to receptors, and growth factor receptor binding induces an entire host of reactions in cells that culminate in changes in gene expression.  Here’s the kicker:  Only specific growth factors induce MSCs to make cartilage, and they are transforming growth factor-β (TGF-β) proteins.  Specifically, TGF-β1, 2, and 3 are the only well-established full inducers of cartilage induction.  When any of these three growth factors are added to cultures of pelleted MSCs, they will induce the synthesis of cartilage-specific molecules (Mackay AM, et al., Tissue Eng 1998;4:415-428; Barry F, et al., Exp Cell Res 2001;268:189-200).  Other growth factors reportedly induce cartilage formation, but there is a controversy over whether or not these growth factors are bona fide cartilage inducers.

Despite what Centeno states in his blog post, there is a genuine controversy among stem cell scientists as to which type of MSC – bone marrow-derived or fat-derived – are better sources for orthopedic tissue repair (Frisbie DD, et al., J Orthop Res 2009;27(12):1675-1680.). Scientists have successfully used bone marrow-derived MSCs and fat-derived MSCs to make cartilage in culture (Johnstone B, et al., Exp Cell Res 1998; 238:265-272; Erickson GR, et al., Biochem Biophys Res Commun 2002;290:763-769).  Additionally, in experimental models, bone marrow-derived MSCs and MSCs from other tissue sources have been successfully used for orthopedic treatments (Wakitani S, et al., J Bone Joint Surg Am 1994;76;579-592; Im GI, et al., J Bone Joint Surg Br 2001;83:289-294; Koga H, et al., Cell Tissue Res. 2008 333(2):207-15; Centeno CJ, et al., Curr Stem Cell Res Ther 2011;6(4):368-378).

Several head-to-head comparisons of bone marrow-derived MSCs and fat-derived MSCs have produced contradictory results,.  There are some studies that show equivalent cartilage-making capacities for the two types of MSCs (Zuk PA, et al., Tissue Eng 2001;7(2):211-228; De Ugarte DA, et al., Cells Tissues Organs 2003;174(3):101-109; Rebelatto CK, et al., Exp Biol Med 2008;233(7):901-913).  However other studies have concluded that human and equine bone marrow MSCs have superior cartilage-making abilities when compared to fat-derived MSCs (Winter A, et al., Arthritis Rheum 2003;48(2):418-429.; Im GI, Shin YW, Lee KB. Osteoarthritis Cartilage 2005;13(10):845-853; Sakaguchi Y, et al., Arthritis Rheum 2005;52(8):2521-2529;Vidal MA, et al., Vet Surg 2008; 37(8):713-724). Now these experiments have a potential problem and it is this:  these experiments utilize MSCs from human or animal donors and several experiments have shown that the same MSC populations from different donors show different differentiation potentials (Bieback K, et al., Stem Cells 2004;22:625–634; Chang YJ, Set al., Stem Cells 2006a;24:679-685; Kern S, et al., Stem Cells 2006;24:1294–1301).  Therefore, to address this problems, some papers have made head-to-head comparisons of donor-matched MSC populations, and these experiments have shown that bone marrow-derived MSCs show superior cartilage-making potential over fat-derived MSCs (Huang JL, et al., J Orthop Res 2005;23:1383-138; Afizah H, et al., Tissue Eng 2007;13(4):659-666).  On top of all that, genetic studies that examine the expression of the whole genome at one time (microarray studies) have shown that bone marrow-derived MSCs have a gene expression pattern that more closely resembles that of native cartilage-making cells than that of fat-derived MSCs  (Winter et al 2003).

Another way to grow MSCs for cartilage production is in three-dimensional cultures.  In this case, the MSCs are seeded into a mass of molecules that resemble the types of matrices in which cartilage is normally deposited.  Such culture systems are thought to be very important for defining the conditions under which cartilage is made (Johnstone et al 1998; Yoo JU, et al., J Bone Joint Surg Am 1998;80(12):1745-1757; Erickson GR, et al., Biochem Biophys Res Commun 2002;290:763-769). When MSCs are grown in three-dimensional cultures, once again bone marrow-derived MSCs outperform fat-derived MSCs (Jakobsen RB, et al., Knee Surg Sports Traumatol Arthosc 2010 18(10):1407-16; Mehlhorn AT, et al., Tissue Eng 2006;12(10):2853-62).

With all this work, you might think that Centeno’s warning about fat-derived MSCs is completely warranted.  Well,, not so fast.  Let’s remember that the head-to-head comparisons treated both MSC populations with the same cartilage induction protocol, and this implicitly assumes that the culture conditions optimized for bone marrow MSCs are also optimal for fat-derived MSCs.  This assumption, however, ignores the intrinsic differences between these two MSC populations.  Two scientists, Kim and Im, have shown that fat-derived MSCs have a cartilage-making capacity that is equal to that of bone marrow-derived MSCs if they are treated with higher concentrations of growth factors (Kim HJ, Im GI. J Orthop Res 2009;27: 612-61).  Additionally, Diekman and colleagues have shown that cartilage formation in MSCs from bone marrow and fat is highly dependent on the presence and concentration of particular growth factors, the presence or absence of serum, and the composition of the scaffold in which the cells are embedded.  Fat-derived MSCs made significantly more cartilage-specific molecules in response to a growth factor called Bone Morphogen Protein-6 (BMP-6) than to TGF-β, but the exact opposite was true for bone marrow-derived MSCs. Also, fat-derived MSCs produced more cartilage in the presence of serum whereas bone marrow MSCs produced more without serum. Thus the culture system is crucially important for cartilage induction, and once the culture conditions for fat-derived MSCs are well worked out, they can make cartilage as well as bone marrow-derived MSCs.  Some labs have some a very long way in designing cartilage-inducing culture systems for fat-derived MSCs (Estes BT, Wu AW, Guilak F.  Arthritis Rheum 2006;54(4):1222-1232).

Therefore, even though bone marrow MSCs do a better at making cartilage, it is due to the fact that the cartilage induction uses a culture system that was designed for bone marrow MSCs.  MSCs from other tissue sources are more than capable of forming cartilage and might even be good candidates for orthopedic applications if they were developed for such a purpose.  Therefore, we must agree with Centeno that presently, Bone marrow MSCs are the best candidates for orthopedic work, but that does not mean that other MSC populations are incapable.

Published by


Professor of Biochemistry at Spring Arbor University (SAU) in Spring Arbor, MI. Have been at SAU since 1999. Author of The Stem Cell Epistles. Before that I was a postdoctoral research fellow at the University of Pennsylvania in Philadelphia, PA (1997-1999), and Sussex University, Falmer, UK (1994-1997). I studied Cell and Developmental Biology at UC Irvine (PhD 1994), and Microbiology at UC Davis (MA 1986, BS 1984).