|Home | About | Journals | Submit | Contact Us | Français|
Use of platelet-rich plasma (PRP) has shown promise in various orthopaedic applications, including treatment of anterior cruciate ligament (ACL) injuries. However, various components of blood, including peripheral blood mononuclear cells (PBMCs), are removed in the process of making PRP. It is yet unknown whether these PBMCs have a positive or negative effect on fibroblast behavior.
To begin to define the effect of PBMCs on ACL fibroblasts, ACL fibroblasts were cultured on three-dimensional collagen scaffolds for 14 days with and without PBMCs. ACL fibroblasts exposed to PBMCs showed increased type I and type III procollagen gene expression, collagen protein expression, and cell proliferation when the cells were cultured in the presence of platelets and plasma. However, addition of PBMCs to cells cultured without the presence of platelets had no effect. The increase in collagen gene and protein expression was accompanied by an increase in IL-6 expression by the PBMCs with exposure to the platelets. Our results suggest that the interaction between platelets and PBMCs leads to an IL-6 mediated increase in collagen expression by ACL fibroblasts.
Anterior cruciate ligament (ACL) rupture is a common injury with an annual incidence of 400,000 in the United States (1). The injury not only causes pain and joint instability, but also increases the risk of premature degenerative joint disease (2). Because the ACL fails to heal after suture repair, ACL reconstruction with a tendon autograft is the current gold standard of treatment (3). However, even ACL reconstruction has drawbacks, including loss of normal knee kinematics and high rates of premature osteoarthritis(4). Furthermore, failure of autografts is not uncommon; failure rates are 7.2% for patellar tendon grafts and 15.8% for hamstring grafts (5). These failures may be at least partly due to slow or incomplete healing. Therefore, there is an increasing interest in methods to enhance the healing of the ACL, such as the use of platelet-rich plasma (PRP).
PRP is plasma with a high concentration of platelets, which can serve as an autologous source of several growth factors, including platelet-derived growth factor (PDGF-AB), transforming growth factor beta (TGF-B), and vascular endothelial growth factor (VEGF) (6). These cytokines encourage wound healing by serving as mitogens, chemoattractants, and stimulators of cell proliferation (7). Furthermore, PRP activation by collagen leads to sustained release of these cytokines (8).
Prior clinical studies on the use of PRP for ACL reconstruction have shown little beneficial effect of PRP (9). This finding is consistent with our prior in vivo studies in animal models, which have also demonstrated PRP use alone to be ineffective. This is likely due to the presence of intra-articular plasmin in the post-traumatic joint. Plasmin readily degrades fibrin (the principal extracellular matrix molecule in PRP) and the PRP is lost before it can have an effect (10). Interestingly, further work in animal models has shown that when the PRP is combined with a collagen-based carrier, functional healing of the ACL is significantly enhanced (11; 12). This may be due to the fact that when the fibrin in the PRP is mixed with collagen, a co-polymer is formed which is more resistant to plasmin degradation(13). Thus, the use of a collagen-platelet composite in the joint may be more effective than supplementation with PRP alone.
PRP is known to stimulate collagen gene expression by ACL fibroblasts (14). Our previous studies have also shown that collagen-PRP composites stimulate ACL healing at early time points in large animal models (15-20). However, various blood components are removed from blood in the process of making PRP, and it is important to define the cellular components most beneficial to include in PRP. Erythrocytes are known to inhibit ACL fibroblast proliferation in collagen scaffolds and to encourage collagen production by fibroblasts (21). Leukocytes concentrations in PRP are largely influenced by the preparation method used; some systems leave large numbers of leukocytes in PRP, while others almost entirely remove leukocytes (22). There are some studies that suggest that inclusion of leukocytes in PRP may be harmful. For example, higher leukocyte concentration in PRP correlated with increased metalloproteinase gene expression (23). Neutrophils may particularly be hazardous as they can release reactive oxygen species and various proteases (24). While these molecules are important for fighting infectios agents, they can also damage tissue (25). On the other hand, other types of leukocytes may play beneficial roles in wound healing. Chamberlain et al. showed that depleting macrophages in rats limited early healing processes and compromised ligament strength (26).
In the current study, we decided to focus on a specific subtype of leukocytes: peripheral blood mononuclear cells (PBMCs). PBMCs include lymphocytes, monocytes, and macrophages, which are often removed from blood when making PRP. In the past, platelet gels have been shown to activate PBMCs to release pro-inflammatory cytokines, including IL-6 (27; 28). Because IL-6 is known to stimulate collagen production by tendon fibroblasts(29) and dermal fibroblasts(30), we hypothesized that co-culture of the PBMCs with platelets might increase the IL-6 production of PBMCs and subsequent collagen production by the ACL fibroblasts.
Bovine knee capsular tissue was harvested using sterile technique. The tissue was solubilized in acidic pepsin solution to make acid soluble, type I collagen slurry as previously described (15). The pepsin solubilization process cleaves the species-specific ends of the collagen fragment, greatly reducing any potential for immunogenic reaction(31). The collagen concentration was adjusted to 8mg/ml and neutralized with 0.1 M HEPES, 5× PBS, and 7.5% sodium bicarbonate.
Porcine whole blood was centrifuged for 6 minutes at 150 × g. The supernatant was collected and centrifuged at 1000 × g for 10 minutes. Approximately one third of the supernatant was collected as PPP. The platelet pellet was resuspended in the remaining supernatant to make PRP. The platelet concentration of the blood was 181 × 106 /mL and of PRP was 911 × 106 /mL (enrichment of 5.03x).
PBMCs were isolated from porcine whole blood. The blood was collected into syringes containing 10% acid-citrate dextrose; then, PBMCs were isolated on a Ficoll column using Histopaque®-1077 (Sigma, St. Louis, MO). The isolated PBMCs were resuspended in PBS. The final cell concentration was 5 × 106 cells/mL. All PBMC samples contained less than 0.3 × 106 platelets/mL.
ACL explants were obtained from porcine knees using sterile technique. The explants were cultured in completed Dulbecco’s modified Eagle’s medium (DMEM), which was prepared by supplementing DMEM (Invitrogen, Carlsbad, CA) with 10% fetal bovine serum (Invitrogen, Carlsbad, CA) and 1% antibiotic/antimycotic (Invitrogen, Carlsbad, CA). Once the primary outgrowth cells were 80% confluent, they were trypsinized and frozen. Then, the first passage cells were thawed, expanded, and passaged. Second passage cells were used for the experiment.
ACL fibroblasts were resuspended in PBS, PRP, or PPP at a concentration of 2.0 × 106 cells/mL. 5 mL of cell suspension, 13 mL of neutralized collagen slurry, and 2 mL of PBS or PBMCs were combined to form the cell-hydrogel composites. The final collagen density in all groups was 3 mg/mL, the final ACL fibroblast concentration in all groups was 5.0 × 105 cells/mL, and the final PBMC concentration in the groups with PBMCs was 5.0 × 105 cells/mL. The collagen-cell mixture was placed in designated 3-cm-long semicylindrical molds with a polyester mesh at each end to anchor the gels (Table 1). Control groups of each group of constructs without ACL cells were also cultured (n=5 for each group). All constructs were incubated in a humidified 5% CO2 incubator at 37°C for 1 hour to achieve gelation. Thereafter, the constructs were cultured in completed DMEM. Medium was changed every 3 days.
An additional series of 5 × 105 PBMCs were cultured in collagen gel constructs as described above. Constructs were harvested on days 0, 1, 4, 7, 10, and 14 (n=3). Then, the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazoliumbromide (MTT) assay was performed. The constructs rinsed with PBS and incubated in a solution of 0.5mg/mL MTT (Invitrogen, Carlsbad, CA) in DMEM for 4 hours in a 5% CO2 37°C incubator. Constructs were incubated in a solution of 0.1N HCl (Sigma, St. Louis, MO) in isopropyl alcohol (Sigma, St. Louis, MO) for an additional 4 hours, and the optical density of the resulting supernatant was measured at 570nm using a microplate reader (Molecular Devices, Sunnyvale, CA).
PBMC conditioned media were prepared as follows. First, PBMCs were collected as described above. Then, the PBMCs were cultured in DMEM for 48 hours (1 × 106 PBMCs and 20ml of DMEM in each flask) and removed from the media by centrifugation. The conditioned media were also used to culture an additional set of ACL fibroblast constructs for the same six groups (n=5) used above, with the PBMC media used in place of the PBMCs for the final three groups in Table 1, which were prepared as described above.
At day 14, the constructs were harvested, and MTT assay was performed as described above (n=3 per group).
After 14 days, the constructs were harvested and washed with PBS. RNA was extracted from constructs using RNeasy mini kit (Qiagen, Valencia, CA). The RNA samples were reverse-transcribed into cDNA with RETROscript kit (Ambion, Austin, TX). Real-time PCR was performed with StepOnePlus™ Real-Time PCR System (Applied Biosystems, Carlsbad, CA) using SYBRGreen PCR Master Mix Kit (Applied Biosystems, Carlsbad, CA). Types I and III procollagen genes (COL1A1 and COL3A1) were the targets, and GADPH was used as the reference gene. The primer sequences were:
COL1A1 forward 5′-CAGAACGGCCTCAGGTACCA-3′
COL1A1 reverse 5′-CAGATCACGTCATCGCACAAC-3′
COL3A1 forward 5′-CCTGGACTTCCTGGTATAGC-3′
COL3A1 reverse 5′-TCCTCCTTCACCTTTCTCAC-3′,
GAPDH forward 5′-GGGCATGAACCATGAGAAGT-3′
GAPDH reverse 5′-GTCTTCTGGGTGGCAGTGAT-3′
The transcript levels of the procollagen genes normalized to GAPDH were calculated using the 2−ΔΔCt formula.
At day 14, the SIRCOL collagen assay kit (Biocolor, Carrickfergus, UK) was used to assess the collagen expressed by the cells into the media. To account for collagen released into the media as a result of scaffold degradation during culture, additional constructs were cultured without cells and the collagen protein release was measured using the SIRCOL assay.
At day 14, the concentrations of IL-6 in media were measured using Quantikine Porcine IL-6 Immunoassay (R&D Systems, Minneapolis, MN). The ELISA assay was conducted according to the manufacturer’s manual. The optical densities were measured at 450nm using a microplate reader (Molecular Devices, Sunnyvale, CA).
All results are given as mean ± SD. Statistical analysis was performed with t-test. A p value less than 0.05 was considered significant. All calculations were done using intercooled STATA 10 (StataCorp, College Station, TX).
The metabolic function of the PBMCs in the collagen gel increased from 0.017 OD units on the first day of culture to over 0.026 OD units at the seven day time point. The metabolic function then remained stable until fourteen days (Figure 5). There were no significant differences between day 7 and day 10 (p=0.279) or between day 10 and day 14 (p=0.859).
There was a significant increase in metabolic activity by the ACL fibroblasts when they were cultured with both PRP and PBMCs as opposed to PRP alone (Figure 1, p<0.01). However, adding PBMCs alone to the PBS group or the PPP group did not significantly affect cell metabolic activity (Figure 1, p=0.615, p=0.293). The constructs containing ACL fibroblasts had significantly higher MTT values than their corresponding controls without ACL fibroblasts (Figure 1, p < 0.001 for all comparisons).
Both type I and type III procollagen gene expressions by ACL cells were significantly higher in cultures with both PBMCs and PRP compared to cultures with only PRP (Figure 2, p<0.05 for both). The addition of PBMC-conditioned medium to the fibroblasts did not have a significant effect on type I or type III procollagen gene expression by the fibroblasts (Figure 3, p>0.05 for all comparisons).
Co-culture of the ACL cells with both PBMCs and PRP resulted in an increase in the amount of collagen released into the media compared to cultures without PBMCs (Figure 4, p<0.01). However, adding PBMCs to cells cultured in PBS or PPP did not increase the amount of collagen in the medium (p=0.440, p=0.213). The amount of collagen released into the media from the constructs without ACL fibroblasts was minimal (all values less than 2 μg/mL). The collagen protein expression was also normalized by MTT. After normalization, the production of collagen per cell was still highest in the PRP+PBMC+ACL cell group with a significant difference between PRP+PBMC group and PRP only group (Figure 5, p<0.05).
The amount of IL-6 released into the culture media was highest for the constructs containing PRP and PBMCs. This was true for constructs cultured with and without ACL fibroblasts (Figure 6; p<0.001 for both).
In a previous study, we had shown that the addition of PRP to ACL fibroblasts cultured in 3D collagen scaffolds increased collagen expression and cell proliferation (14). The results of the current study show that adding PBMCs in addition to PRP further amplifies collagen gene and protein expression by ACL fibroblasts as well as increasing cell metabolic activity.
The inclusion of white blood cells in PRP has been somewhat controversial. Some studies support the inclusion of leukocytes in PRP. A recent level 1 clinical trial by Gosens et al.(32) found that leukocyte-enriched PRP outperformed corticosteroids in the treatment of chronic lateral epicondylitis. However, other studies have reported a deleterious effect of white blood cells on tissue healing. McCarrell et al.(23) showed that increased leukocytes in PRP are associated with more metalloproteinase (MMP3 and MMP-13) gene expression and less cartilage oligomeric matrix protein (COMP) and decorin gene expression. However, they also reported an increase in type III procollagen gene expression with more leukocytes. In addition, Anitua et al. (24) discussed how neutrophils express metalloproteases (MMP-8 and MMP-9) and release reactive oxygen species (ROS). ROS can kill cells and may worsen healing. It should be noted that neutrophils were not present in our cultures; PBMCs consist of lymphocytes, monocytes, and macrophages. Therefore, while PBMCs appear to have an anabolic effect on the ACL fibroblasts, other types of white blood cells may be detrimental and the types of leukocytes included in a PRP preparation is important to note in studies of these heterogeneous cells.
It is interesting to note that the anabolic effect of PBMCs was present only when the ACL fibroblasts were also exposed to platelets. There was no such effect noted when then ACL fibroblasts were cultured with PBMCs and platelet-poor plasma (which contains all the plasma proteins in PRP, just with the platelets removed), or with PBMCs and saline (which contains neither the plasma proteins nor platelets found in PRP). This suggests that for ACL fibroblasts, the presence of platelets is required for the PBMCs to significantly affect fibroblast behavior. This is contrary to work previously published which has reported direct interactions between PBMCs and fibroblasts. Co-culturing PBMCs and gingival fibroblasts has been shown to increase collagen synthesis without the presence of platelets(33). PBMCs alone also have been shown to stimulate dermal fibroblasts to proliferate and to produce more collagen and non-collagen proteins (34). The unique response of ACL fibroblasts to PBMCs may help explain why PRP alone was not sufficient to enhance suture repair of ACL in animal model (35).
The increase in cellular metabolism and collagen production seen in the fibroblasts cultured with both PRP and PBMCs may be a result of the platelets stimulating the PBMCs to secrete IL-6, which in turn stimulates the collagen production by the ACL fibroblasts. Without the PRP, the PBMCs secreted very little IL-6, while with the PRP, the level of secretion was significantly higher (Figure 6). Collagen gene expression by the ACL fibroblasts followed the same pattern. This finding is consistent with prior reports of IL-6 playing a key role in the wound healing process(36) and inducing collagen expression(29).
We also conducted an experiment to see if the soluble factors released from PBMCs were sufficient for these interactions. PBMCs were cultured by themselves, and then removed from media. Then, the media were used to culture the constructs. While constructs cultured in PBMC-conditioned media showed small increases in collagen gene expressions compared to unconditioned media, the differences were not statistically significant. There are two possible explanations for these results. It may be that direct cell-to-cell contact plays an important role. An alternate explanation is that the interaction between PBMCs and PRP is reciprocative. In this experimental design, PBMCs can stimulate PRP via soluble factors, but the PRP is not able to stimulate the PBMCs. Additional studies to delineate these hypotheses are needed.
The study of soluble factors influencing cell behaviors could also be studied in a modified Boyden chamber or similar apparatus where the cells are cultured separately(33). In this paper, we elected to try to model a three-dimensional provisional wound scaffold composed of many of the extracellular matrix molecules and cytokines which might be present in a healing wound. The cellular interactions which would be present in this biologically complex hydrogel system may be less representative of the two-dimensional case of separate chambers, but may have some translational relevance.
One of the additional limitations to this study is we were limited in our ability to measure complete collagen protein production. As the cells were cultured in a hydrogel, some of the newly produced collagen may have been trapped in the gel, and would be indistinguishable from the collagen of the hydrogel using our analysis methods. In addition, while very little collagen was released from constructs cultured without cells, there may have been additional collagen released from the cell-seeded constructs due to cell-based degradation of the constructs.
PBMCs are a byproduct of PRP synthesis that is typically discarded, but our study suggests that they may be a useful additive for ACL repair or other orthopaedic treatments where collagen production by fibroblasts is desirable. PBMCs can be isolated easily and quickly by density-gradient centrifugation, so the use of autologous PBMCs for ACL surgeries is logistically feasible. Thus, we believe that the combination of PRP, PBMCs, and a collagen scaffold may be a reasonable candidate for a biomaterial that can improve ACL repair.
This study was supported by NIH-NIAMS Grant RO1 AR052772 and NIH R01 AR054099 (MMM).