Long-term functional engraftment of mesenchymal progenitor cells in a mouse model of accelerated aging

Abstract

Age-related osteoporosis is characterized by a decrease in bone-forming capacity mediated by defects in the number and function of osteoblasts. An important cellular mechanism which may in part explain osteoblast dysfunction that occurs with aging is senescence of mesenchymal progenitor cells (MPCs). In the telomere-based Wrn^−/−Terc^−/− model of accelerated aging, the osteoporotic phenotype of these mice is also associated with a major decline in MPC differentiation into osteoblasts. To investigate the role of MPC aging as a cell-autonomous mechanism in senile bone loss, transplantation of young wild-type whole bone marrow into Wrn^−/−Terc^−/− mutants was performed and the ability of engrafted cells to differentiate into cells of the osteoblast lineage was assessed. We found that whole bone marrow transplantation (BMT) in Wrn^−/−Terc^−/− mice resulted in functional engraftment of MPCs up to 42 weeks, which was accompanied by a survival advantage, as well as delays in microarchitectural features of skeletal aging.

Introduction

Skeletal aging of bone is characterized by loss of mineral content and microarchitectural changes that decrease bone strength and predispose bone to damage from repeated loading [1, 2].

Osteoporosis is common in the Werner and dyskeratosis congenita premature aging syndromes, both characterized by telomere dysfunction [3, 4]. One of the targets of WRN helicase is telomeric DNA, but need for WRN at telomeres is minimized in mice by long telomeres and abundant telomerase, making Wrn knockout mice relatively unaffected [5, 6]. However, combining Wrn mutation with shortened telomeres of telomerase (Terc) knockout mice results in an accelerated aging model [5, 7]. Deficiencies in Wrn^−/− Terc^−/− mutant mice cause a low bone mass phenotype due to impaired osteoblast differentiation in the context of intact osteoclast differentiation [8, 9]. This impaired differentiation is associated with telomere dysfunction, as measured by the association of DNA damage proteins with telomeres in mesenchymal progenitor cells (MPCs) isolated from double mutant mice [9]. MPCs from Wrn^−/− Terc^−/− mutants have a reduced in vitro lifespan but also display impaired osteogenic potential with dysfunctional telomeres independently of proliferative state [9].

Here we test the hypothesis that MPC aging contributes to bone loss in an accelerated aging mouse model that recapitulates many aspects of age-related bone loss.

Materials and Methods

Detailed Material and Methods are described in the supporting information.

Results

Wild-type whole bone marrow transplantation (BMT) into Wrn^−/−Terc^−/− mutants confers a survival advantage

To test the role of telomere-based MPC aging on age-related osteoporosis, Wrn^−/−Terc^−/− mutants were transplanted at 3 months of age with whole BM from young wild type donors. At 10.5 months after transplantation or when animals exhibited signs of significant distress and impending demise (whichever occurred first), mutants were sacrificed for analysis of functional MPC engraftment and concomitant measurements of skeletal microarchitectural features. As shown in Fig. 1, transplanted animals exhibited a survival advantage, having a mean life span about 30% longer than untransplanted controls (12.89 ± 0.21 months versus 10 ± 0.57 months). This difference is particularly remarkable given that telomerase-deficient mice are hypersensitive to ionizing radiation and otherwise would have been expected to incur substantial harm from irradiation associated with BMT [10].

Figure 1.

Overall survival advantage of Wrn^−/−Terc^−/− mice after bone marrow transplantation (BMT). Kaplan-Meier plot of Wrn^−/−Terc^−/− mutants with (n=13) and without (n=8) wild-type whole BMT. p= 0.00035 by log-rank test.

Long-term functional engraftment of MPCs in Wrn^−/−Terc^−/− mice

Enhanced green fluorescent protein-positive (GFP+) wild-type mice were used as donors in all BMT experiments. GFP+ MPCs in BM aspirates were identified from transplanted animals by fluorescent immunohistochemistry and represented 54.0 ± 7.1 percent of plastic adherent stromal cells after 30 hours in culture. Expanded MPC cultures from young wild-type donor animals are essentially CD45⁻ Sca-1⁺ cells and can differentiate in vitro into osteoblasts and adipocytes (Fig. S1). CD45⁻ Sca-1⁺ MPC are present after long-term BMT in Wrn^−/−Terc^−/− mice (Fig. 2).

Figure 2.

Long-term engraftment of Sca-1⁺CD45⁻ MPCs after BMT. Serial bone sections were stained with the indicated antibodies at left. Representative examples of engrafted MPCs in a Wrn^−/−Terc^−/− recipient are shown by arrows. Scale bar = 20 μm.

MPCs demonstrate functional engraftment as differentiated GFP⁺ osteoblasts and osteocytes in bone sections from recipient Wrn^−/−Terc^−/− mice (Fig. 3, Fig. S2). MPC functional engraftment is present up to 10.5 months after BMT. In femur sections from transplanted animals 20 ± 8 percent of cortical osteocytes and 6 ± 2 percent of trabecular osteocytes were derived from engrafted precursors. Among endocortical and trabecular bone-lining osteoblasts, 15 ± 6 percent and 5 ± 1 percent were from precursors of donor origin, respectively. Differentiation of engrafted MPCs was most evident as endocortical osteocytes (Fig. S2).

Figure 3.

Donor MPCs differentiate into bone-lining osteoblasts (and subsequently osteocytes) that are incorporated into bone. Arrowheads indicate bone-lining cells. Arrows indicate osteocytes located in their lacunae. Scale bar = 10 μm.

Delays in microarchitectural features of skeletal aging in Wrn^−/−Terc^−/− mutants after BMT

Despite being ~30% older than non-transplanted double mutants, Wrn^−/−Terc^−/− BMT recipients had preserved or improved measures of bone microarchitecture. Transplanted animals showed no statistically significant changes in trabecular bone volume/total volume, trabecular number, or in cortical thickness (Fig. 4). Interestingly, there was a statistically significant increase in the ratio of cortical area to total area, suggesting that transplanted mice were able to improve endocortical bone mass. Similarly, Wrn^−/−Terc^−/− BMT recipients had preserved osteoblasts/bone surface and no increase in the number of osteoclasts/bone surface [Fig. S3].

Figure 4.

Microarchitectural features of skeletal aging are delayed in Wrn^−/−Terc^−/−mutants after BMT. (Top panels) Representative micro-CT 3-dimentional reconstructions of trabecular and cortical bone in Wrn^−/−Terc^−/− mutants without (n=6) and with BMT (n=6) are shown. (Bottom panels) Quantification of trabecular and cortical bone parameters in Wrn^−/−Terc^−/− mutants demonstrates preservation of microarchitectural features with BMT. Note that BMT recipients were about 30% older than non-BMT controls. BV, bone volume; TV, total volume; Tb.N, trabecular number; Ct Th, cortical thickness; Ct.Ar; cortical area; Tt.Ar, total area. ****, p< 0.002.

Discussion

Although the causal role(s) of telomere dysfunction in age-related osteoporosis are not completely established, there is evidence of its importance. Telomere lengthening mechanisms are not present in human BM MPCs [11]. Exogenous telomerase expression extends in vitro proliferative capacity, accelerates osteogenic differentiation, and enhances bone formation upon subcutaneous transplantation into mice [12, 13]. In addition, progeriod syndromes on which the Wrn^−/− Terc^−/− mutants are based (Werner syndrome and dyskerostis congenita, respectively) display premature osteoporosis [3, 4].

Although we cannot exclude that hematopoietic precursors including hematopoietic stem cells (HSCs), transplanted along with MPCs, may support mesenchymal engraftment and/or differentiation into cells of the osteoblast lineage, it is unlikely that CD45⁻ Sca-1⁺ MPCs are derived directly from HSCs. However, there may be an early common BM progenitor for hematopoietic cells and osteoblasts delineated as Lin⁻Sca-1⁺cKit⁺CD45⁺. [14, 15].

Soluble hematopoietic factors alone may be sufficient to exert effects on bone remodeling, or development of osteoblasts may depend on the proximity of hematopoietic cells. Sca-1, a cell-surface molecule also expressed on HSCs, appears to be necessary to maintain self-renewal of MPCs and suggests that the latter is plausible. In support of this, Sca-1 knockout mice develop age-dependent osteoporosis [16].

There could be systemic effects of BMT which may influence skeletal engraftment and differentiation of MPCs. Also, microenvironmental factors (including oxidative stress) may be at play. For example, we previously showed that osteoblast differentiation of MPCs from Wrn^−/−Terc^−/− mutants is rescued by reducing oxidative stress [9]. Increased oxidative stress favoring senescence in BM MPCs has also been postulated based on proteome screening of these cells from young and old rodents [17]. Muscle-derived stem/progenitor cells from young wild-type mice transplanted into a murine progeria model extended life span and improved degenerative changes in tissues where donor cells are not detected [18]. Although we cannot be sure that extra-skeletal effects of BMT are responsible for life span extension in Wrn^−/−Terc^−/− mutants, the fact that GFP⁺Sca-1⁺CD45⁻ MPCs are present in bone tissue after long-term transplantation suggests that they play a role in maintenance of bone microarchitectural features over the period of extended survival.

To the extent that bone loss with physiologic aging involves telomere-based aging, our data indicates MPC senescence is a contributory mechanism. However, BMT as a therapeutic strategy may be limited by inadequate MPC engraftment [19]. It is debatable whether donor MPCs from human or mouse sources have sustained engraftment in host BM; however, our and other reports indicate that this is so [20–26]. Therefore, it is reasonable to suggest decreased bone regeneration with age may be partially reversed by transplantation of young donor MPCs.

Wrn^−/− Terc^−/− mutants may have altered BM stroma which permits effective engraftment of donor MPCs. Thus BMT, performed for other reasons in telomere-based accelerated aging syndromes (e.g. aplastic anemia), may also confer beneficial effects in stabilizing or delaying premature osteoporosis. In fact, aplastic anemia in individuals with telomere-based progeroid syndromes may actually occur due to defective (telomerase-deficient) stroma which cannot support HSCs. Whole BMT in these patients may be successful [27] because it theoretically will correct both the stromal as well as the hematopoietic defects.

Conclusions

Replacement of aging MPCs with young cells results in delays or amelioration of aspects of skeletal aging. Wrn^−/− Terc^−/− recipients of whole BMT have functional reconstitution of MPCs and stable or improved bone microarchitectural features compared to much younger, non-transplanted mice.

Supplementary Material

Supp Fig S1&S3

Figure S1. (A) Adherent MPCs from young wild-type donors are CD45⁻ Sca-1⁺ cells. Gated cells are shown in the left panel. Cells are > 99.5% CD45⁻ and > 96.5% Sca-1⁺ (right panel). SSC, side scatter; FSC, forward scatter. (B) Differentiation of GFP⁺ CD45⁻Sca-1⁺ mesenchymal progenitor cells (MPCs) into osteoblasts and adipocytes. Original magnification, 200X.

Figure S2. Donor MPCs differentiate into bone-lining osteoblasts (and subsequently osteocytes) that are incorporated into bone. Arrowheads indicate representative bone-lining cells. Arrows indicate representative osteocytes in their lacunae. Scale bar = 10 μm.

Figure S3. BMT in Wrn^−/−Terc^−/− animals preserves bone osteoblasts/bone surface (N. Ob/BS) and number of osteoclasts/bone surface (N. Oc/BS).