Association of Runx2, the bone forming protein, with progression of prostate cancer


Cancer is one of the most versatile diseases because abnormal cell growth results can be achieved through abnormalities in numerous pathways and may be spread through metastasis to the other regions of the body. One of the most commonly diagnosed cancers in men is Prostate cancer (PCa) (Akech et al. 2010, American Cancer Society 2014, Horoszewicz et. al. 1983). Statistics show that 80% of PCa metastatic occurrences happen in bones, which strongly affects the normal bone remodeling – resorption and formation. It has been noted that certain transcription factors involved in bone formation are expressed abnormally in cancer cell.  (Akech et. al., 2010, Cher et. al. 2003, Nameth at. al. 2001). Signal proteins, like  RANKL, a receptor activator of NF-kB ligand, parathyroid hormone related protein (PTH-rP), bone resorption promoter interleukin (IL8), promoters of osteoblastic lesions endothelial and Wnt pathways are frequently upregulated in tumor cells. Bone formation transcription factors are also highly expressed in tumors that metastasize in bone are pathological calcification suppressor Msx2 and a master regulator Runx2. Runx2 is especially highly expressed, which makes it a primary candidate to be the key regulator for facilitating adhesion and migration of the cancer cells (Akech et. al., 2010).


RUNX2 transcription factor protein model



Core-binding factor subunit alpha-1 (Cbfa1), or the human homologue RUNX2, gene belongs to the runt-domain genetic family and its organization is very similar in both humans and mice (Fig 0). It is composed of eight exons and contains at least two promoter regions. Several variants of the gene have been identified generated by the alternative gene splicing.  (Komori and Kishimito, 1998).  Runx2 been identified as the key transcription factor in osteoblast, or bone forming cell, differentiation and a master regulator of osteoblast specific expression of Osteocalcin and Ospetpontin during the embryonic development and after birth. (Komori, 2014). It interact with bone morphogenic proteins, hormone regulators and growth factors, and integrates them into an intricate signaling pathway to induce the specific cell fate (Figure 0 and Table 1) (Lian et. al., 2004) The unique property of Runx2 is in the mediation of Wnt, Src, BMP, and TGFbeta signaling pathways, which are activated in tumor cells (Fig 0). Therefore, it is hypothesized that Runx2 contributes to the progression and invasiveness of prostate cancer cells, as well as to the aggressiveness of osteolytic bone disease that frequently accompanies prostate cancer (Akech et. al., 2010).

Figure 0: Runx2 pathway for bone remodeling regulation during embryonic and after birth development.


Table 1

Table 0: Runx2 target genes for regulation (Lian et. al., 2004).


The Study: 

Step 1: Experiments in vitro to assure the functional activity.

  • Because cancer cells exhibit wide genomic instability, cell lines of prostate cancer cells with different metastatic and tumor growth potentials were compared, which include highly metastatic cancer cell line PC3 from bone metastases, non-metastatic  androgen-sensitive human prostate adenocarcinoma cells LNCaP that do not proliferate in bones, and osteotropic prostate cancer subline C4-2B derived from LNCaP cells that form osteoblastic lesions.
  • Western blot analysis and quantitative reverse transcription PCR (qRT-PCR) showed that PC3 cells expressed the highest levels of Runx2 transcription factor (Fig 1). PC3 cells were analyzed closer and were split into high, medium, and low levels of Runx2 mRNA expression. The whole cell lysate showed low levels of expression, and therefore nuclear extracts were examined. qRT-PCR revealed three cell groups that were given names according to their relative Runx2 expression: PC3-h (high), PC3-m (medium), and PC3-l (low) (Fig1b). PC3-h showed 2 fold greater expression of the gene than PC3-m and 15-fold greater expression than PC3-l.
  • Furthermore, there was a comparison of expression of Runx2 genes to the family members, in order to determine if they have similar expression patterns, since they recognize the same regulatory sequence. The two family members were the differentiation regulator of hematopoietic stem cells into mature blood cells Runx1 and tumor suppressor Runx3. This experiment showed that very low expression of Runx1 in all the PC3 cells and higher expression of Runx3 in the PC3-low cells.
  • In the next step RNA interference was used to determine if Runx2 is indeed linked to the metastasis of the PC3 cell lineage. Prevention of mRNA expression with mRNA interference showed an 80-90% decrease in the bone-related gene expression (Fig 1c). Furthermore, a 30% increase in fibronectin adhesion was observed, which correlates to decrease of invasive properties of PC3 cell, and a 50% decrease in Survivin, a protein strongly associated with tumor cell proliferation (Fig 1d).

Fig 1:  Runx2 is responsible for prostate cancer cell proliferation and matastasis a. Comparison of Runx2 expression between the different cancer cell strains using qRT-PCR (upper panel) and Western blot analysis (lower panel)    b. PC3 sublines and their relative Runx 2 mRNA expression patterns, as well as protein analysis in cell lysate and nuclear extrct. c. Western blot analysis of PC3 cell that shows that there is an 80% knockdown of endogenous Runx2 (CDK1 cells as control)   d.  siRNA-treated PC3-H cells show reduction of invasion and increase of adhesion in matrigel medium.   e. qRT-PCR aalysis of expression of Runx2 and its target genes 48 hours after knockdown.    f. Runx2 siRNA reduces expression of Survivin protein in triplicate assays.

Step 2: Experiments in vivo to identify involvement in osteolytic disease promotion.

  •  In the second part of the study, Runx2 functional activity in PC3 sublines was explored in vivo. In order to do so, the prostate cancer cells were inoculated into the intramedullary cavity of immunocompromised SCID mice(Figure 2). PC3-high and medium cells showed much faster osteolysis and tumor cell invasion rate compared to the PC3-low cells (Fig 2a).
  • Immunochemistry  assay, which identifies a protein with a specific antigen, of PC3-medium tumor revealed that in human cytokeratin cells Runx2 was largely expressed (Fig 2b). Such method helped to distinguish between the implanted tumor cells and mouse bone marrow cells.
  • This study furthermore showed a gradient of osteolysis, or bone resorption, in the PC3 cells, where PC3-high resulted in the most aggressive disease occurrence. To support further the claim that Runx2 indeed promotes osteolytic disease, expression of osteoclast activating molecules IL8 and parathyroid hormone-related protein (PTHrP) were analyzed (Figure 2d). PC3-H and –M showed the most prevalent expression of both factors. Furthermore, analysis of expression after implantation revealed that the sublines retained those properties, thus supporting the previous assumptions.
Figure 2:

Figure 2: Experiments on PC3 lines reviealed that Runx2 pmotes osteolytic bone disease in vivo and osteoclast activation in vitro. a. Comparison of expressions of high, medium and low levels of Runx2 in mouse tibiae at 4 and 6 weeks that show erosion areas and osteoblastic lesions. b. TRAP staining of osteoclast cell show that PC3 cells induce osteolytic lesions. c. Triplicate qRT-PCR analysis showing mRNA expression of Runx2 in PC3 sublines post implantation. d. qRT-PCR results on PTHrP and IL8 expression in PC3 sublines.

Step 3: Closer look at influences of Runx2 on metastasis and prostate cancer progression.

  • The next step in the experiment was to look more intimately at the contribution of PC3-H cells to proliferation of osteoclast precursors and osteoclastogenesis. Three groups of PC3-H cells were co-cultured with osteoclast precursor RAW 264.7 murine macrophage cell line for 7 days in order to allow the exchange of secreted factors for potential osteoclast generation. The results indicated a 3 to 7 fold increase in cell growth and expression of osteoclast activation genes, such as MMP9, TRAP, Cathepsin, K and RANK, while bone matrix proteins were increased at much lower rate, in comparison to the control group of just RAW 264.7 cells and RAW264.7 cells with RANKL, the key osteoclast activation and differentiation factor (Figure 3a).
  • Next, the influence of PC3-H on osteoblast activity was addressed. Medium with PC3 cells was added to MC3Tc osteoprogenitor mouse cell culture during their osteoblast differentiation. Results from alkaline phosphatase staining on days 14 and 21 revealed that osteoblast differentiation and activity was inhibited by the PC3-H cells (Figure 3b), which provides the evidence that Runx2 influences both osteoclast and osteoblast activity, as hypothesized.
  • Furthermore, Runx2 contribution to activation of metastatic target genes in prostate cancer was explored. The researchers found that there is a distinct profile of metastasis related genes between the cell lines, where PC3-H cells highly expressed extracellular matrix degrading protein MMP9, PC3-M expressed more MMP13 (also an extracellular matrix remodeling protein) , PC3-L expressed high levels of MMP2 genes, and LNCaP cells showed more vascular endothelial growth factor(VEGF) and no significant MMPs expression. Those results also showed that in comparison to the MMPs, osteocalcin and osteopontin bone matrix proteins were expressed at a much lower lever in all the cell lines, which suggests adenoviral delivery of Runx2 (Fig 4b). This question was addressed by taking a closer look at Runx2 activities in PC3-M, PC3-L, and LNCaP (Figure 4b). The results showed that Runx2 induces only the metastatic  MMP genes that are deferentially expressed at low endogenous levels (Figures 4a and 4b).
  • The knockdown of Runx2 gene in PC3 cell lines was then analyzed, in order to determine its direct role in formation of metastatic lesions in bones. To do so, Runx2 short hairpin RNA (shRNA) was added to one group of the PC3-H cells, in order to silence the gene. A control group was created by introducing non Runx2 shRNA to PC3-H cells. This step was followed by injection of the cell groups into murine tibia. Radiography was performed after 4 weeks post injection, which revealed that both shRNA treated and untreated PC3-H cells formed similar aggressive osteolysis (Figure 5a, top and center panels). Runx2 shRNA injected mice (Figure 5a, lower panel), in contrast, showed either no evidence of osteolysis, or onset of mild bone lysis (arrows). Western blot and qRT-PCR was performed after in order to confirm the knock down of Runx2 by shRNA. Both analyses confirmed the deed and revealed complete downregulation of bone resorbing factors IL8 and PTHrP (Figure 5b). From those finding researchers proposed a wholesome mechanism of regulation and contribution of Run2 to prostate cancer metastasis and formation of bone lesions (Figure 5c).
  • The last step in the study was to determine if Runx2 expression correlates with prostate cancer progression, which was done by comparing the microarray of cancer tissue with Runx2 immunochemistry assay results (Table 1). About 92-93% of non-neoplastic (non cancerous) and pre-malignant prostatic intraepithelial neoplastic cells (PIN)  were shown to be negative for Runx2. In contrast, 48% of primary tunors and 45.8% of metastatic lesions showed Runx2 intenstiries ranging from 1 to 3.
Figure 3 Figure 4

Figure 3: Gene expression of markers for osteoclast differentiation and inhibition. 
Figure 4: Expression of metastatic genes activated by Runx2.

Figure 5:  Runx2 and induction of metastatic genes in cancer. a. Comparison of murine bones injected with CP3 cells and their Runx2 knocked down version and determination of bone lysis occurrence. b. Western blot and qRT-PCR analysis to confirm Runx2 knockdown. Runx2 knockdown caused downregulation of IL8 and PTHrP bone resorbing factors.  c. Pathways by which Run2 regulates metastasis and bone lesions in prostate cancer.


The study presented the key findings on Runx2 as a regulator of metastasis and bone lesions. A strong functional relationship has been determined by the series of experiments, providing evidence for Runx2 being a key determinant in gene activation in malignant tumors. Furthermore, other prostate cancer studies indicated that about 50% of tumor tissues indeed showed prominent Runx2 expression, which highlights the importance of the investigation.

This study is done in a pleasing chronological order from identifying the gene and assuring that it indeed has a prominent effect within the tumor cell, to testing in vitro the regulatory mechanisms of action, to applying the knowledge in vivo and generating a treatment that does not affect that unintended regions.

However, this technique cannot be generalized due to the effect of tumor heterogeneity, when a tumor consists of many different cell lineages that compete with each other. Therefore, even if certain cell lines are prevented from metastasizing, there are no guarantees that other cells may possess a mechanism of action that would allow them to metastasize in bone. This suggests that this technique should be used in combination to increase the effectiveness.



Akech, J., Wixted, J., Bedard, K. and et al. (2010). Runx2 association with progression of prostate cancer in patients: mechanisms mediating bone osteolysis and osteoblastic metastatic lesions. Oncogene 29: 811-822.

Cher, M., Biliran, H., Bhagat, S., and et al. (2003).  Maspin expression inhibits osteolysis, tumor growth, and angiogenesis in a model of prostate cancer bone metastasis. Pnas. 100(13), 7847-7852.

Horoszewicz, J.S., Leong, S.S., Kawinski, E., and et al. (1983) LNCaP Model of Human Prostatic Carcinoma. Cancer Research. 43, 1809-1818.

Komori, T., and Kishimoto, T. (1998). Cbfa1 in bone development. Current Options in Genetics and Development 8: 491-499.

Komori, T. (2014). The function of Runx family transcription factors and Cbfb in skeletal development. Orla Science International 12: 1-4.

Lian, J., Javed, A., and Zaidi, K. (2004). Regulatory Controls for Osteoblast Growth and Differentiation: Role of Runx/Cbfa/AML Factors. Critical Reviews in Eukaryotic Gene Expression 14(1&2): 1-4.

Nameth. J., Yousif, R, Herzog, M., and et al. (2001). Matrix Metalloproteinase Activity, Bone Matrix Turnover, and Tumor Cell Proliferation in Prostate Cancer Bone Metastasis. Journal of the National Cancer Institution. 94(1), 17-25.


Leave a Reply

Your email address will not be published. Required fields are marked *