Regulation of BMP-induced transcription in cultured human bone marrow stromal cells

Abstract

Background: Adherent bone marrow stromal cells are inducible osteoprogenitors, giving rise to cells expressing osteoblast markers including alkaline phosphatase, osteopontin, osteocalcin, and bone sialoprotein. However, the potency of inducers varies in a species-specific manner. Glucocorticoids such as dexamethasone induce alkaline phosphatase activity in both human and rat mesenchymal stem cells, while mouse bone marrow stromal cells are refractory to dexamethasone-induced alkaline phosphatase activity. In contrast, BMP induces alkaline phosphatase activity in both mouse and rat bone marrow stromal cells, while BMP effects on human bone marrow stromal cells are poorly characterized.

Methods: Bone marrow samples were isolated from patients undergoing hip replacement. Mononuclear marrow cells were cultured and grown to confluence without or with 10 (-7) M dexamethasone. Cells from each isolate were passaged into medium containing 100 micro g/mL ascorbate phosphate and treated with dexamethasone, 100 ng/mL BMP, or no inducer. At day 6, alkaline phosphatase activity was assayed, and RNA was prepared for mRNA analyses by real-time polymerase chain reaction.

Results: Bone marrow stromal cells from twenty-four of twenty-six patients showed no significant osteogenic response to BMP-2, 4, or 7 as determined by alkaline phosphatase induction. However, BMPs induced elevated levels of other genes associated with osteogenesis such as bone sialoprotein and osteopontin as well as BMP-2 and noggin. If primary cultures of human bone marrow stromal cells were pretreated with dexamethasone, BMP-2 treatment of first-passage cells induced alkaline phosphatase in approximately half of the isolates, and significantly greater induction was seen in cells from males. Dexamethasone treatment, like BMP treatment, also increased expression of the BMP-binding protein noggin.

Conclusions: Most human femur bone marrow stromal cell samples appear incapable of expressing elevated alkaline phosphatase levels in response to BMPs. Since BMP treatment induced expression of several other BMP-regulated genes, the defect in alkaline phosphatase induction is presumably not due to impaired BMP signaling. We hypothesize that the mechanism by which BMPs modulate alkaline phosphatase expression is indirect, involving a BMP-regulated transcription factor for alkaline phosphatase expression that is controlled differently in humans and rodents.

Fig. 1

Alkaline phosphatase (AP) activity in day-6 first-passage human marrow-derived stromal cell cultures. First-passage human marrow-derived stromal cell cultures were treated and were harvested for analysis of alkaline phosphatase activity as described in Materials and Methods. A: First-passage cultures derived from untreated primaries did not show any response to BMP-2 treatment. Alkaline phosphatase activity was significantly (*) increased by dexamethasone (Dex). B: In first-passage cultures derived from dexamethasone-treated primaries, BMP-2 significantly increased alkaline phosphatase activity. Dexamethasone pretreatment also enhanced the dexamethasone stimulation of alkaline phosphatase seen in the first passage. C: The influence of dexamethasone pretreatment on BMP-2 induction of alkaline phosphatase in the first passage was greater in cultures of cells derived from male patients than it was in cultures of cells from female patients.

Fig. 2

Reverse transcriptase-polymerase chain reactions for osteoblastic and BMP-inducible genes in human marrow-derived stromal cells. Conventional reverse transcriptase-polymerase chain reaction was performed on total RNA isolated from day-6 first-passage human marrow-derived stromal cells not pretreated with dexamethasone. BMP-2 had no effect on alkaline phosphatase, but it consistently increased BMP-2 and noggin mRNA. BMP also induced osteopontin in most samples. Dexamethasone also increased noggin mRNA and potentiated the effect of BMP. However, dexamethasone attenuated the BMP-2 upregulation of BMP-2 and osteopontin mRNA.

Fig. 3

Expression of BMP-inducible genes in a representative sample of cultured first-passage human marrow-derived stromal cells. cDNA was prepared from mRNA of day-6 first-passage cultures and analyzed by real-time reverse transcriptase-polymerase chain reactions. SYBR Green fluorescence indicates formation of polymerase chain reaction product. The cycle number at which product fluorescence crosses the fluorescence threshold is the crossing point; larger amounts of initial mRNA result in detectable fluorescence at a lower crossing point. Both noggin and BMP-2 mRNA were increased in BMP-treated cultures. Dexamethasone (Dex) treatment increased noggin mRNA but decreased BMP-2 mRNA.

Fig. 4

Immunoblot analysis for noggin protein in human marrow-derived stromal cells. Whole cell protein lysates from day-6 cultures were evaluated as described in Materials and Methods with use of a rat anti-human noggin antibody. Two representative samples from non-pretreated first-passage cultures are shown. The results mirror the reverse transcriptase-polymerase chain reaction results, demonstrating increased noggin protein associated with either BMP-2 or dexamethasone (Dex) treatment and the additive effect of combined treatment.

Fig. 5

Expression of ostoblast genes in a representative sample of cultured first-passage human marrow-derived stromal cells. Results of real-time polymerase chain reactions were analyzed as described in the legend to Figure 3. While dexamethasone (Dex) treatment increased levels of alkaline phosphatase (AP) mRNA at both day 6 and day 10, BMP treatment did not. Bone sialoprotein (BSP) mRNA was increased by either dexamethasone or BMP, and osteocalcin mRNA was unaffected by either dexamethasone or BMP.