T-lymphocytes enable osteoblast maturation via IL-17F during the early phase of fracture repair

Abstract

While it is well known that the presence of lymphocytes and cytokines are important for fracture healing, the exact role of the various cytokines expressed by cells of the immune system on osteoblast biology remains unclear. To study the role of inflammatory cytokines in fracture repair, we studied tibial bone healing in wild-type and Rag1(-/-) mice. Histological analysis, µCT stereology, biomechanical testing, calcein staining and quantitative RNA gene expression studies were performed on healing tibial fractures. These data provide support for Rag1(-/-) mice as a model of impaired fracture healing compared to wild-type. Moreover, the pro-inflammatory cytokine, IL-17F, was found to be a key mediator in the cellular response of the immune system in osteogenesis. In vitro studies showed that IL-17F alone stimulated osteoblast maturation. We propose a model in which the Th17 subset of T-lymphocytes produces IL-17F to stimulate bone healing. This is a pivotal link in advancing our current understanding of the molecular and cellular basis of fracture healing, which in turn may aid in optimizing fracture management and in the treatment of impaired bone healing.

Figure 1. T-cell presence and increased bone marker expression are demonstrated in early fracture repair.

(A) Although no cartilage or new bone formation is expected at 3 days post tibial fracture, low magnification (20X) Safranin O staining of WT mice at this time point helps to illustrate the region of interest (inset). (B) Higher magnification images corroborate the histology in this region and (C) positive CD3 staining of T-cells localizing to the endosteal fracture hematoma in contiguous sections (200X). Immunohistochemistry analysis confirms presence of T-cell infiltration (CD3 – black arrows) in WT mice. (D) In a similar region, no positive staining was seen in Rag1^−/− mice (200X). (E) T-cell presence is shown to correlate with gene expression of mature bone markers detected on quantitative RT-PCR normalized to GAPDH. Expression of Col1 (p<0.01), Col2 (p<0.05), BSP (p<0.01), BMP2 (p<0.01) and Runx2 (p<0.05) were increased in WT compared to Rag1^−/− mice at 3 days post fracture.

Figure 2. Cytokine expression during the early phase of fracture repair.

(A) WT and Rag1^−/− mice serums drawn 2 days post-fracture show elevated IL-6 and G-CSF levels compared to unfractured baseline mice. However, WT mice were found to have twice the increase in cytokine levels of IL-6 and G-CSF compared to Rag1^−/− mice during this early phase of fracture healing (p<0.05). (B) Fracture callus RNA expression analysis of cytokine levels at a similar early time point post fracture, shows a greater than 2 fold up-regulation of pro-inflammatory cytokines IL-6, IL-17F and IL-23 in WT mice compared to Rag1^−/− and baseline mice (p<0.05). Conversely, expression of anti-inflammatory cytokine IL-10 is up-regulated in Rag1^−/− and TGFβ is down-regulated in WT mice. (C) Immunohistochemistry analysis during the early phase of fracture healing at 3 days post fracture confirmed IL-17F positive staining (black arrows) in WT mice similar to the distribution of CD3 staining shown previously with no positive IL-17F staining in Rag1^−/− mice (200X).

Figure 3. IL-17F promotes osteoblast maturation.

(A) Treatment of MC3T3-E1 pre-osteoblast cell line cultures directly with pro-inflammatory cytokine IL-17F showed increased bone marker gene expression of Col1, Col2, BSP and osteocalcin, whereas, anti-inflammatory cytokine TGFβ treatment inhibited osteoblast maturation and showed significant decreases in expression of the entire panel of bone markers analyzed. (B) Left: Treatment of Rag1^−/− mice primary mesenchymal stromal cell cultures directly with pro-inflammatory cytokine IL-17F showed increased bone marker gene expression of Col1, Col2, and Runx2. Right: In contrast, treatment with anti-inflammatory cytokine TGFβ inhibited WT mice primary mesenchymal stromal cell maturation and showed significant decreases in expression of all the bone markers analyzed. (*p<0.05 and **p<0.01).

Figure 4. Osteogenesis is reduced in Rag1

^{− /−} mice during fracture repair. (A) In vitro cultures of primary mesenchymal stromal cells differentiated to osteoblast colony-forming units (CFU) showed lower gene expression levels of bone markers Col1, Col2 and BSP and osteocalcin in Rag1^−/− mice compared to WT. *p<0.05, **p<0.01 (n = 10). (B) Alizarin red staining of osteoblast CFU showed significant decrease in mineralization at 20 days of culture in Rag1^−/− compared to WT. CFU-osteoblast was quantified by direct counting of all stained nodules positive to Alizarin Red using light microscopy. (C) Digital fluorescent microscopy images of calcein green administered mice 2 and 9 days prior to harvest at 21 and 28 days post fracture exhibited a smaller distance measured between mineralization fronts and hence, less bone formation in Rag1^−/− compared to WT mice (p<0.05). Unfractured limbs showed no significant differences in bone formation.

Figure 5. Healing of the fracture callus in Rag1

^{− /−} mice is delayed. (A) Safranin O staining of WT and Rag1^−/− mice 28 days post fracture showed persistence of a cartilage template with more proteoglycan staining (black arrows) and less bridging bone across the fracture gap in Rag1^−/− compared to WT mice. (B) µCT analysis confirmed the presence of a wider, lower density callus in Rag1^−/− mice tibias at 28 days. (C) This was reflected by increased total callus volume (TV) measurements (p = 0.002) and decreased tissue mineral density (TMD) at 28 days in the Rag1^−/− compared to WT mice (p = 0.004). Torsional rigidity (TR) was significantly higher in Rag1^−/− mice compared to WT (p = 0.002). (D) Mechanical assessment of the samples using torsional mechanical testing showed a significantly higher ultimate torque and torsional stiffness in the Rag1^−/− mice at 28 days compared to WT (p = 0.002).

Figure 6. Proposed mechanistic scheme of T-cell mediated osteoblast differentiation and maturation.

Upstream IL-6 increase early post fracture promotes naïve CD4+ T-cells to Th17 cells stimulating pre-osteoblast cell differentiation via IL-17F. Concomitantly, the Treg pathway is suppressed, decreasing TGFβ and IL-10 and inhibiting pre-osteoblast differentiation. The effect of IL-6 also has a direct role in the later stages of osteoblast differentiation in fracture healing.