A Simplified Method for the Aspiration of Bone Marrow from Patients Undergoing Hip and Knee Joint Replacement for Isolating Mesenchymal Stem Cells and In Vitro Chondrogenesis

The procedure for aspiration of bone marrow from the femur of patients undergoing total knee arthroplasty (TKA) or total hip arthroplasty (THA) may vary from an OR (operating room) to OR based on the surgeon's skill and may lead to varied extent of clotting of the marrow and this, in turn, presents difficulty in the isolation of mesenchymal stem cells (MSCs) from such clotted bone marrow. We present a simple detailed protocol for aspirating bone marrow from such patients, isolation, and characterization of MSCs from the aspirated bone marrow specimens and show that the bone marrow presented no clotting or exhibited minimal clotting. This represents an economical source and convenient source of MSCs from bone marrow for use in regenerative medicine. Also, we presented the detailed protocol and showed that the MSCs derived from such bone marrow specimens exhibited MSCs characteristics and generated micromass cartilages, the recipe for regenerative medicine for osteoarthritis. The protocols we presented can be used as standard operating procedures (SOPs) by researchers and clinicians.


Introduction
The posterior iliac crest is a readily accessible site for the bone marrow aspiration that is safe and psychologically less traumatic and affords representative samples of bone marrow similar to that obtained from the sternum, the vertebral spine, and the anterior iliac crest [1]. In general, for biopsy and clinical diagnosis, clinicians collect bone marrow from posterior and anterior iliac crest and lumbar pedicle. For osteoarthritis research and for developing bioengineering tools for repairing OA or use in other diseases, bone marrow can be aspirated from the femur of patients undergoing TKA or THA and can be an alternative and economical source of mesenchymal stem cells (MSCs). There are reports of researchers using bone marrow aspirated from patients for research though protocol details have never been published [2,3]. Bone marrow aspiration procedure from patients may vary from surgeon to surgeon. In the beginning, surgeons in our OR tried to aspirate bone marrow from patient's femur with a pipette and that resulted in clotting of bone marrow. By introducing this simple procedure, the bone marrow is aspirated directly into the specimen trap and gets mixed with anticoagulant instantly under the influence of vacuum resulting in minimum or no clotting of bone marrow. Second, we present a simple lab procedure for isolating MSCs and detail protocol for generating micromass cartilages that can be used for osteoarthritis research and repair. As far as we know, this is the first detailed report for a simplified procedure for aspirating bone marrow.
Thomas Scientific). MSCs' characteristic positive and negative markers and their ability to differentiate into three different lineages were evaluated.
(   (ii) Negative control was without primary antibody and with mouse IgG (for types II, I, and X) and with rabbit IgG, polyclonal (ab27478, abcam), for aggrecan and COL VI. in Histo-Clear II (3x, 3 min) and mounted in Omnimount mounting medium (17997, EM sciences). Proteoglycans stain red, cytoplasm stains gray green, and nuclei stain black. Knee paraffin section (4 m thick) from a rat was stained for safranin O staining as a positive control.
(4) Immunohistochemistry (IHC). Micromass cartilages were characterized by localizing specific antigens by IHC. Slides were heated at 58-60 ∘ C for 60 min. The slides at RT were deparaffinized and hydrated as described earlier in the text.
The sections went through antigen retrieval process. Sections were incubated with blocking buffer for 30 min and then with primary antibody ( Table 2) overnight at 4 ∘ C, followed by step A or B as shown below depending upon the primary antibody used: (A) for COL II, COL X, COL I, and aggrecan, the sections were washed with PBS-T (4x, 5 min), treated with blocking buffer for 30 min, and incubated with secondary antibody (

Bone Marrow Aspiration and MSCs Characterization.
We presented a simplified procedure for the aspiration of bone marrow from the femur of patients undergoing TKA and THA (Figures 1(a)-1(d)). The bone marrow was collected without clotting or with minimal clotting in all the aspirations we conducted so far using this procedure. Further, we showed that fatty material was easy to remove in the first step of centrifugation before loading the marrow over Ficoll-Paque PLUS for gradient formation (Figure 1(e)). Passing the bone marrow through wide-holed sieve proved to be a useful step since it removed any bone spur or fibrous tissue or any clotted pieces of bone marrow that are likely to disrupt gradient separation of mononuclear cells (MNCs). As far as we know, this is the first of its kind in a simple protocol form that we are reporting here. The volume of bone marrow retrieved from femur varied from patient to patient (≈2-20 mL). Here, we report MSCs isolation from bone marrow aspirated from proximal femur from 28 THA patients (M : F; 1 : 1) and distal femur from 10 TKA patients (M : F; 1 : 1). The MSCs that were obtained using this procedure showed positive markers (CD44, CD73, CD105, CD90, CD146, and STRO-1) and negative markers (CD14, CD19, CD34, and CD45) as shown by immunofluorescence and/or flow cytometry (Figures 2 and 3). MSCs were able to differentiate into three lineages in vitro: adipocytes (Figure 4(a)), osteoblasts (Figures 4(b)-4(d)), and chondrocytes ( Figure 5). High density MSCs that were plated in 30 L drops (Figure 5(a)) underwent chondrogenic differentiation and the resulting micromass cartilage in 24-well plate ( Figure 5(b)) increased in size from week 1 through week 4 (≈2 mm in diameter at week 4; Figure 5(c)).

Micromass Chondrogenesis and Assessment.
Masson's trichrome staining showed that collagen (blue colored fibrils), intercellular distance, and chondrocyte size increased from week 1 to week 4 in micromass cartilages (Figure 6(a)). Safranin O staining showed the presence of proteoglycans at weeks 1 and 2 that became abundant at weeks 3 and 4 ( Figure 6(b)). Rat knee served as a positive control for safranin O staining that showed proteoglycans in articular cartilage ( Figure 6(b)). COL II began to be expressed at week 2 and became prominent at subsequent time points (weeks 3 and 4) in micromass cartilages (Figure 7(a)). Rat knee articular cartilage along with meniscal cartilage and human tracheal cartilage served as positive staining for COL II (Figure 7(a)). Col II was present around chondrocytes and in intercellular matrix (Figure 7(a)). COL I was expressed at all of the time points in micromass cartilage (week 3 not shown, Figure 7(b)). Rat knee served as positive control for COL I, the expression of which was shown in trabeculae only, and the native articular cartilage did not express COL I (Figure 7(b)). Negative controls are shown in Figure 7(c). COL X showed weak expression at weeks 3 and 4 and that is a sign of low level of hypertrophy of chondrocytes (Figure 8(a)). Hypertrophic chondrocyte zone in undecalcified limb section of P1 mouse was used as positive control (Figure 8(a)). Micromass cartilages did not show apoptosis (Figure 8(b)). Mouse involuting mammary gland (day 4 postpartum) served as positive control for apoptosis (Figure 8(b)).
Aggrecan and COL VI showed expression in micromass cartilages (Figures 9(a) and 9(b)). Rat knee articular cartilage served as positive control for aggrecan (Figure 9(a)). Human tracheal cartilage served as positive control for COL VI; higher magnification shows that the COL VI is centered at pericellular regions of chondrocytes or lacunae containing chondrocytes (Figure 9(b)). Transmission electron microscopy images of micromass cartilage from the upper middle area at week 4 are shown (Figures 10(a)-10(e)) and that indicates that chondrocytes are normal in week 4's micromass cartilage. Chondrocytes appears to be metabolically active with extensive Golgi network, well-developed rough endoplasmic reticulum (rER), euchromatin nucleus with nucleolus, a large number of secretory vesicles and coated vesicles, a number of lipid droplets, and glycogen, mitochondria, and lysosomes. Centriole is present. Extracellular matrix (ECM) shows collagen fibers (Figures 10(a)-10(e)).

Statistical Analysis.
The aspirable volume of bone marrow from femur, number of viable MNCs collected from bone marrow, patient age, number of viable MSCs at passage 3 after expansion, and the day at which successful expansion of viable MSCs was achieved were compared between proximal ( = 28) and distal ( = 10) femur bone marrow aspirations by using the Wilcoxon test for paired samples. A value of < 0.05 was considered as statistically significant, with a 95% confidence interval.

Bone Marrow Aspiration and MSCs Characterization.
Bone marrow is a flexible tissue in the interior of bones. On an average, bone marrow constitutes 4% of the total body mass of humans. Bone marrow constitutes hematopoietic and lymphatic system. Bone marrow transplants can be conducted to treat severe diseases of the bone marrow, including certain forms of cancer such as leukemia [10]. Additionally, bone marrow stem cells can be differentiated into other lineages and used to treat diseases [11]. The posterior iliac crest is a readily accessible site for bone marrow aspiration that is safe [1]. In general, for biopsy and clinical diagnosis, clinicians collect bone marrow from posterior and anterior iliac crest and lumbar pedicle. Alternatively, bone marrow can be aspirated from the femur of patients undergoing TKA or THA. There are reports of researchers using bone marrow aspirated from patients though standard operating room protocol has never been published [2,3]. Bone marrow aspiration procedure from patients may vary from surgeon to surgeon and may lead to partial or full clotting of bone marrow within seconds. By introducing this procedure, the bone marrow is aspirated directly into the specimen trap and that gets mixed with anticoagulant instantly resulting in minimum or no clotting of bone marrow (Figures 1(b)-1(d)). The bone marrow provides acceptable number of MNCs. Second, we introduced a step of filtering diluted bone marrow through wide-holed sieve to remove any fibrous tissue or small piece of bone in the marrow (Figure 1(e)). The bone marrow aspiration from patients undergoing TKA and THA is a very economical procedure. Purchasing bone marrow-derived MSCs from a standard biotechnology company can be very expensive and can cost up to 2,000 USD per million MSCs at passage 2. MSCs isolated from patients' bone marrow were able to undergo trilineage differentiation (adipogenesis, osteogenesis, and chondrogenesis) and exhibited specific positive and negative cell surface markers (Figures 2-5) similar to that reported earlier [12,13]. The chondrogenesis was further assessed in detail (Figures 6-10).
An interesting study was published recently by Narbona-Carceles and colleagues [14], in which the authors aspirated bone marrow from three different sites from the patients prior to TKA in the same operating room for the purpose  of isolation and characterization of MSCs. They obtained bone marrow aspirates through bone puncture of the iliac crest, distal femur, and proximal tibia, and the samples were aspirated by a single investigator. To avoid hemodilution of the aspirate, the depth and angle of the trocar were changed after every 2 mL of material had been aspirated. Samples were dispensed in heparinized tubes. We aspirated bone marrow from a single site either proximal or distal femur from patients undergoing THA and TKA, respectively. Hence, we will exclude iliac crest for comparison purpose. We did not use a separate session for bone marrow aspiration from patients in the operating room before THA or TKA surgery; rather, we aspirated bone marrow during surgical procedure. Since our procedure of aspiration was simple, any surgeon on operating duty was suitable for these aspirations. The bone marrow was aspirated from the femur cavity by  gliding a suction needle (13 ) attached at the rubber tubing of the specimen trap containing heparin. The nozzle of the trap was attached to vacuum system as shown in Figure 1. The procedure we devised, for the bone marrow aspiration from femur during THA or TKA, is simple, ethical, and novel due to the following reasons and has advantages over bone puncturing procedure in the knee [14]. (1) In our procedure, there was no need of special-skilled surgeon because of its simplicity. Any available surgeon was able to do aspiration. (2) In the event, bone marrow aspiration is proposed to be collected by bone puncturing method of knee [14] or hip in a separate session in the patients undergoing TKA or THA; an ethical concern would be evident as compared to aspiration from open wound during THA or TKA. Also, patients may hesitate to provide consent for puncturing of knee or hip just before their painful THA or TKA surgery due to the assumptive pain perception [15]. (3) An aspiration of bone marrow, from multiple points of a bone marrow site, is considered as better aspiration as compared to that from a single point [16]. The bone puncturing method and maneuvering of needle at different depths [14] in femur may require multiple punctures of bone to fulfill that requirement. That may add additional time on patient under anesthesia as compared to the current procedure. The aspiration of bone marrow from an open wound procedure during THA or TKA takes ≈10-15 secs. (4) In the bone puncturing method [14], the bone marrow is collected in a syringe and then dispensed into a container with an anticoagulant. This procedure may be inefficient for stopping part of bone marrow clotting. In addition, if the multiple puncturing is required, that will be of additional concern, in this regard. This problem does not exist in our current procedure since bone marrow is aspirated and mixed instantly with heparin, already present in the specimen trap, under the influence of suction pressure. (5) In bone puncturing method of bone marrow collection from femur, there are more chances of getting bone marrow contaminated with blood or other tissue fluids or tissues pieces depending upon the accuracy of needle insertion, that is, hemodilution concerns [14]. In open wound femur aspiration during THA or TKA, bone marrow sample can be aspirated without this concern since the suction needle is less likely to puncture any other tissue except bone marrow itself and also there is only a small force applied on suction needle for gliding along the bone marrow cavity as compared to the force applied in bone puncturing method. Above all, the suction needle end is not beveled.
It was very interesting that Narbona-Carceles and colleagues [14] were able to filter bone marrow through 100 m filters. On the other hand, we diluted the bone marrow with equal volume of 2 mM EDTA and then filtered it through a sieve of pore size 1.25×1.25 mm 2 . Given the much larger pore size we used, filtration of bone marrow was still slow. That indicated that the bone marrow in our aspirations was more viscous than the bone marrow aspirated by bone puncturing method [14]. Age could be a factor for this observation. Narbona-Carceles and colleagues [14] used aspirations from the patients in the age range of 64-75 years. On contrary, we used much younger patients in the age group of 51-64 years.  Figure 9: IHC showed that micromass cartilage expressed aggrecan and COL VI at all the time points (weeks 1-4, a, b). Articular cartilage of rat knee expressed aggrecan and served as positive control (a). Human tracheal cartilage served as positive control for COL VI and showed that COL VI is pericellular in location in native cartilage (b). Higher magnification of week 4 bioengineered micromass cartilage also shows the pericellular localization of COL VI (arrowheads, b). Rat knee (a) and micromass cartilage (b) were used as negative control with rabbit IgG polyclonal and without primary antibody. M, meniscus; AC, articular cartilage; B, bone; mm, micromass.
Narbona-Carceles and colleagues [14] aspirated 5 mL bone marrow from iliac crest, distal femur, and proximal tibia. In our procedure, the average volume aspirated from distal femur (during TKA procedure) was 4.87 ± 0.63 mL; additional gliding of suction needle in marrow cavity did not increase the aspirated volume (Figure 1(g)). On the other hand, in proximal femur (during THA procedure), the aspirated volume was 11.17±1.15 mL which was 2.29 times higher than the distal femur. Narbona-Carceles and colleagues [14] reported recovery of MNCs per mL bone marrow as 0.67 ± 1.1 × 10 6 in distal femur aspirate and 1.70 ± 4.8 × 10 6 in proximal tibia aspirates. On the other hand, we showed the MNCs per mL bone marrow as 11.74 ± 1.21 × 10 6 and 12.37 ± 2.28 × 10 6 from proximal femur and distal femur, respectively (Figure 1(f)). MNCs number was 18.46-fold higher in our procedure in distal femur group than that in bone punctured procedure [14]. This could be due to the reason that we conducted investigation in much younger patients. Narbona-Carceles and colleagues [14] reported recovery of MSCs (per mL original bone marrow) as 2.90 ± 6.6 × 10 5 in distal femur aspirates and 3.25 ± 2.6 × 10 5 in proximal tibia aspirates on days 40.1 and 40.7, respectively. On the other hand, we showed the viable MSCs (per mL original bone marrow) in proximal and distal femur as 4.10 ± 0.72 × 10 6 and 2.56 ± 0.69 × 10 6 (Figure 1(f)) on days 18.39 ± 0.41 and 18.7 ± 1.13, respectively. Hence, comparing with bone punctured procedure for distal femur [14], our procedure provided 8.82-fold higher number of viable MSCs and that was on day 18.7 ± 1.13 as compared to bone punctured method on day 40.1 [14]. In brief, we presented a simplified procedure for bone marrow aspiration that is novel, less cumbersome, and more ethical towards patients undergoing joint replacement. By comparing distal femur aspirations, we obtained 18.46-fold higher number of viable MNCs per mL bone marrow than that obtained by Narbona-Carceles and colleagues [14]. By comparing distal femur aspirations, we obtained 8.82-fold higher number of viable MSCs per mL bone marrow than that obtained by Narbona-Carceles and colleagues [14] and at much earlier day. Based on our data, we strongly recommend proximal femur of THA patients as the preferred site for bone marrow collection for MSCs isolation simply because more volume can be aspirated from this site as compared to distal femur of TKA patients.

Micromass Chondrogenesis and Assessment.
We generated micromass cartilage (also referred to as bioengineered cartilage like material) from the MSCs we isolated from bone marrow and characterized it (Figures 5-10). In the bioengineered micromass cartilages we generated ( Figure 5(c)), we showed that collagen was present at week 1 to 4 as determined by Masson's trichrome staining (Figure 6(a)). Higher magnification (lower panels, Figure 6(a)) showed that collagen (blue colored fibers) increased from week 1 through week 4 indicating that cells secreted extracellular collagen in an increasing manner as micromass cartilage grew in size along with cell size (Figures 5(c) and 6(a)). This showed that chondrocytes are highly metabolic in function. This was supported by TEM studies of micromass cartilage chondrocytes at week 4 in which we showed that chondrocytes were metabolically active ( Figure 10). Other groups have shown that human bone marrow MSCs-derived chondrogenic pellets, at day 14, showed the presence of collagen determined by Masson's trichrome staining [2]. Micromass cartilage showed the presence of proteoglycans as determined by safranin O staining, at low level at weeks 1 and 2 but prominently at weeks 3 and 4 ( Figure 6(b)) similar to staining in native articular cartilage (Figure 6(b)). Pellet cartilage showed the presence of proteoglycans at week 2 determined by safranin O staining [2]. Zhang lab showed the presence of proteoglycans in micromass cartilage at weeks 1-3 as determined by toluidine blue staining [17].
Zhang lab showed the expression of COL II in micromass cartilage at week 2 [17]. We reported that COL II began to be expressed at week 2 and became prominent afterwards in micromass cartilage (Figure 7(a)). Tuan laboratory showed that COL2A1 gene was expressed at day 14 and increased at days 21 and 28, the similar pattern of COL II protein expression we showed in micromass cartilages by IHC (Figure 7(a)). In human bone marrow MSCs-derived chondrogenic pellets, COL II was expressed at day 14 shown by IHC [2]. We showed that micromass cartilages express COL I at week 1 through week 4 though it has less intensity than COL II (week 3 not shown, Figure 7(b)). By using RTPCR technique, Tuan laboratory showed the expression of COL1A1 gene at days 7-28 in chondrogenic pellets [18]. Chondrogenic cultures at weeks 1, 2, and 3 presented COL I expression [19]. COL I is not present in native articular cartilage and is present in bone as shown in our positive control (Figure 7(b)). Why is COL I expressed in in vitro chondrogenesis? There are two possibilities: (i) it is difficult to mimic chondrogenesis conditions (pH, oxygen tension, and autocrine factors) of native cartilage with those of in in vitro chondrogenesis and some cells can be reprogrammed to produce COL I [20][21][22]; (ii) during development, there is a cellular condensation event prior to joint development that has been shown to be rich in type I collagen [23].
Expression of cartilage hypertrophy markers by MSCs undergoing chondrogenesis raises concern for tissue engineering application for MSCs, because hypertrophy would result in apoptosis and ossification [18]. COL X is expressed at low level at day 7 and day 14 but increased at days 21 and 28 in pellet culture [18]. Besides low level of expression of COL X at weeks 3 and 4, we did not see any apoptosis in our micromass cartilages (Figures 8(a) and 8(b)). Apoptosis was present in pellet culture but not in micromass culture at week 2 as shown by IHC and TEM [17]. We showed that there was no apoptosis in micromass cartilages at any time point (Figure 8(b)) and also we did not see any apoptotic bodies in chondrocytes by TEM studies at week 4 in micromass cartilage ( Figure 10). We used involuting mammary gland as control since it showed apoptotic epithelial cells for comparison (Figure 8(b)).
Aggrecan is the major proteoglycan in the articular cartilage. It is important in the proper functioning of articular cartilage because it provides a hydrated gel structure that endows the cartilage with load-bearing properties. Cartilage contains up to 10% proteoglycan consisting of mainly the large aggregating chondroitin sulfate proteoglycan aggrecan. Aggrecan and collagen II are the major proteins of cartilage [24]. Micromass cartilages showed aggrecan expression at all the time points (Figure 9(a)). Aggrecan expression has been shown in pellet and micromass culture by other investigators [17]. Eyre reported that COL VI accounts for less than 2% of total collagen in bovine cartilage [25,26]. In articular cartilage, COL VI is found and is maintained at low levels, forming a microfibrillar network once secreted by the cell [27]. COL VI is expressed in normal cartilage and increases in OA cartilage [28,29]. Micromass cartilage showed COL VI expression (Figure 9(b)). Knockdown of COL VI affected gene expression of aggrecan, biglycan, and SOX9 during in vitro chondrogenesis. COL VI was shown to be important in resisting applied strains [30].
Transmission electron microscopy images of micromass cartilage from the upper middle area at week 4 are shown (Figures 10(a)-10(e)) which indicated that chondrocytes are normal and are metabolic in nature. Zhang lab showed welldeveloped Golgi in chondrocytes and collagen fibers in extracellular matrix in week 3's chondrogenic cultures [17]. Similarly, we showed well-developed Golgi and collagen fibers in micromass cartilage (Figures 10(b) and 10(c)). Ichinose lab showed well-developed rER and abundant collagen fibers in day-14 pellet and well-developed rER and moderate collagen fibers in micromass cartilages [31]. Extensive rER is present in week-4 generated micromass chondrocytes (Figure 10(b)). Ultrastructure detail of micromass cartilage shows many similarities to the ultrastructure of human articular cartilage [32,33]. Chondrocyte from human femur articular cartilage shows vacuoles, glycogen deposits, and rER similar to the structures we observed in the chondrocyte of micromass cartilage [33]. Roy and Meachim noted glycogen deposits, many mitochondria, and extensive rER in human nonfibrillated femur articular cartilage [32]. Lipid droplets are present in in vitro chondrogenesis from MSCs or chondrocytes [34,35]. We observed lipid droplets in our micromass cartilages at week 4 (Figures 10(a), 10(c), and 10(d)). Presence of lipid droplets is not considered a sign of degeneration since lipid droplets are present in normal cartilage though they increase with aging [36,37] and the exact role of lipid droplets is not yet clear. TEM of chondrocytes derived from infrapatellar fat derived stem cells showed a number of lipid droplets [35].

Conclusion
A simplified method for the aspiration of bone marrow was established from the femur of patients undergoing TKA or THA. The method provides an economical source of MSCs. A simplified procedure was described to isolate MSCs from the aspirated bone marrow. MSCs were characterized including extensive characterization through chondrogenesis. The detailed SOPs we provided will help researchers and clinicians.