Cancer stem cells (CSCs) persist in tumors as a distinct population and may be causative in metastasis and relapse. CSC-rich tumors are associated with higher rates of metastasis and poor patient prognosis. Targeting CSCs therapeutically is challenging, since they seem to be resistant to standard chemotherapy. We have shown that a novel peptide agonist of bone morphogenetic protein (BMP) signaling, P123, is capable of inhibiting the growth of primary tumor cells by interacting with type I receptors selectively [activin receptor-like kinase 2 (ALK2) and ALK3, but not ALK6] and type II BMP receptors, activating SMAD 1/5/8 signaling and controlling the cell cycle pathway. Furthermore, the compound is capable of blocking transforming growth factor-β induced epithelial-to-mesenchymal transition (EMT) in primary tumor cells, a critical step for tumor progression and metastasis. In addition, we have investigated the effects of P123 on self-renewal, growth, differentiation (reversal of EMT) and apoptosis of isolated human breast CSCs. We have shown that P123 and BMP-7 reverse the EMT process in human breast CSCs, and inhibit self-renewal and growth. Moreover, compared with single treatment with paclitaxel, co-treatment with paclitaxel and P123 showed an increase in cell apoptosis. Together, these findings suggest that P123 has the therapeutic potential to suppress both bulk tumor cells and CSCs. We believe that P123 represents a new class of drugs that have the potential to eliminate the primary tumor, prevent reoccurrence and metastasis, and enhance the treatment of breast cancer.

Prostate and breast cancer represent two of the most prevalent cancers for men and women. Overall, one in six men will be diagnosed with prostate cancer and one in eight women will develop breast cancer in their lifetime. An estimated 220 000 men and 230 000 women were diagnosed with prostate cancer and breast cancer in 2015. Primary prostate cancer has an excellent prognosis with a 100% five-year survival rate. Metastatic disease, however, carries a very poor prognosis with a five-year survival of only 28%. Similarly, in early-stage breast cancer, prognosis is excellent with a five-year survival of 98%. Metastatic disease carries a much higher mortality with a five-year survival rate of only 26% [1].

Conventional treatment for these cancers primarily targets bulk tumor cells. It is now understood that in the majority of patients, there is a population of cancer cells that have acquired a drug-resistant phenotype [2]. This population of cells that survive after standard therapies is thought to give rise to metastasis and lead to tumor recurrence [3]. Because these cells are largely dormant, standard chemotherapy is ineffective [4]. It has been suggested that these disseminated tumor cells exist in distant secondary tissue sites as dormant micrometastases [5]. The cells of this drug-resistant phenotype that lead to metastatic disease have been called cancer stem cells (CSCs) (Figure 1) and are commonly identified as the culprits of metastatic relapse. The ability of CSCs to self-renew and generate malignant progeny is thought to make them resistant to chemotherapy and other forms of treatment.

The diagram shows the cycle of events during which epithelial cells are transformed into mesenchymal cells and vice versa.

Figure 1.
The diagram shows the cycle of events during which epithelial cells are transformed into mesenchymal cells and vice versa.

The different stages during EMT and the reverse process MET (mesenchymal–epithelial transition) are regulated by effectors of EMT and MET, which influence each other. Important events during the progression of EMT and MET, including the regulation of the tight junctions and adherens junctions, are indicated. Several markers have been identified that are characteristic of either epithelial or mesenchymal cells and these markers are listed above. E-cadherin, epithelial cadherin; ECM, extracellular matrix; FGFR2, fibroblast growth factor receptor-2; FSP1, fibroblast-specific protein-1; MFs, microfilaments [30].

Figure 1.
The diagram shows the cycle of events during which epithelial cells are transformed into mesenchymal cells and vice versa.

The different stages during EMT and the reverse process MET (mesenchymal–epithelial transition) are regulated by effectors of EMT and MET, which influence each other. Important events during the progression of EMT and MET, including the regulation of the tight junctions and adherens junctions, are indicated. Several markers have been identified that are characteristic of either epithelial or mesenchymal cells and these markers are listed above. E-cadherin, epithelial cadherin; ECM, extracellular matrix; FGFR2, fibroblast growth factor receptor-2; FSP1, fibroblast-specific protein-1; MFs, microfilaments [30].

CSCs represent a relatively small proportion of the cells within the original cancers. The stem-like nature of these cells imbues them with the characteristics of classic stem cell processes, including self-renewal and multilineage differentiation. These processes confer the ability to form new tumors containing the cell types found in the parent cancer tissue. These cells have been cultured and are defined experimentally by their ability to regrow tumors in vitro and in vivo. CSCs have been shown to persist in tumors as a distinct population. CSC-rich tumors are associated with higher rates of metastasis and poor patient prognosis. It has been hypothesized that the primary tumor cells give rise to CSCs by a process of dedifferentiation described as epithelial-to-mesenchymal transition (EMT) [2,6]. The characteristics of the CSC phenotype offer an approach to finding agents that selectively destroy these cells [7,8]. An alternative approach would be to prevent the formation of the CSC or to reverse the EMT process by driving the cells back to a cell type that would be responsive to traditional forms of therapy.

Growing evidence suggests that deregulation of pathways that control self-renewal in CSCs (Figure 2) results in the continuous expansion of self-renewing cancer cells and tumor formation [9]. This suggests that agents, which target the defective self-renewal pathways in cancer cells, may lead to improved outcomes in the treatment of these diseases. The CSCs grown in non-adherent cell suspension as tumorspheres have shown better growth rate, increased stem cell marker cluster of differentiation 44 (CD44) expression and tumorigenicity when implanted in immunocompromised mice [10]. Floating tumorsphere assays have become widely used to study in vitro CSC self-renewal and growth, a proxy for in vivo tumorigenesis.

The central SMAD-dependent pathways for the TGF-β/Activin/Nodal and BMP ligand subfamilies are shown.

Figure 2.
The central SMAD-dependent pathways for the TGF-β/Activin/Nodal and BMP ligand subfamilies are shown.

Efforts to investigate the contribution of SMAD-independent pathways reveal advances to enhance our understanding of the signaling cascade. TGF-β superfamily ligands bind to their specific type II receptor, which recruits and phosphorylates a type I receptor. Each ligand (TGF-β, Activin, or BMP) has its own specific type II receptor and specific type I receptors. The type I receptor then phosphorylates receptor-regulated SMADs (R-SMADs), which can now bind the coSMAD, SMAD4. R-SMAD/coSMAD complexes accumulate in the nucleus where they act as transcription factors and participate in the regulation of target gene expression. BMP, bone morphogenetic protein; TF, transcription factor; P, phosphate group [31].

Figure 2.
The central SMAD-dependent pathways for the TGF-β/Activin/Nodal and BMP ligand subfamilies are shown.

Efforts to investigate the contribution of SMAD-independent pathways reveal advances to enhance our understanding of the signaling cascade. TGF-β superfamily ligands bind to their specific type II receptor, which recruits and phosphorylates a type I receptor. Each ligand (TGF-β, Activin, or BMP) has its own specific type II receptor and specific type I receptors. The type I receptor then phosphorylates receptor-regulated SMADs (R-SMADs), which can now bind the coSMAD, SMAD4. R-SMAD/coSMAD complexes accumulate in the nucleus where they act as transcription factors and participate in the regulation of target gene expression. BMP, bone morphogenetic protein; TF, transcription factor; P, phosphate group [31].

EMT is a critical process for early-stage carcinomas promoting the progression to invasive malignancies [11,12]. In the process of EMT, the cells lose their epithelial phenotype and take on a mesenchymal phenotype. Recent studies have demonstrated that EMT plays a critical role not only in tumor metastasis but also in tumor recurrence, and is tightly linked with the biology of CSCs [13]. Recently, it has been shown that CSCs emerge through the induction of EMT [14]. The differentiated mammary epithelial cells that have undergone EMT either upon transforming growth factor-β (TGF-β) treatment or by forced expression of epithelial cadherin (E-cadherin) transcriptional repressors, such as Snail, gave rise to CD44high, CD24low cells, akin to breast CSCs [6]. The CSCs generated from EMT induction provide a resource for cancer to reoccur. Moreover, these cells are well known to be highly drug resistant [6,1519]. Therefore, the molecular understanding and the biological characteristics of CSCs, such as cell growth by self-renewal and their EMT phenotype, could allow us to screen for potential drugs that produce selective killing of these cells and possibly eradicate tumor recurrence.

Targeting CSCs therapeutically is likely to be challenging, since both bulk tumor cells and CSCs must be eliminated [14,20]. This may require combination drug therapies. Since CSCs are molecularly distinct from bulk tumor cells, one can target their activity by exploiting these molecular differences. For instance, cell surface marker expression could be used for antibody-directed therapy to target proteins such as CD133 or CD44 [21]. Another possible treatment strategy is reversing of the EMT process. Specifically, facilitating differentiation of mesenchymal-like cancers to a more epithelial-like state [22].

Previously, it has been shown that bone morphogenetic protein (BMP) [23,24] and P123, a BMP mimetic [25], are capable of inhibiting bulk tumor cell growth from prostate cancer by activating SMAD 1/5/8 signaling and controlling cell cycle pathway. BMP is a protein that is active as a dimer (Figure 3A) and is a member of the TGF-β superfamily. BMPs were isolated and characterized as osteogenic proteins and later found to be critical factors for embryonic organogenesis. They have been found to be critical in preventing and reversing the EMT process in nonmalignant tissues and in prostate and breast cancers. As a therapeutic, BMPs have several drawbacks that might be overcome by the development of a small molecule BMP mimetic. One such compound, P123 (Figure 3B), was designed using the three-dimensional (3D) structure of BMP-7 (Figure 3A) and was shown to selectively bind a subset of type one receptors to which BMP-7 binds (Figures 4 and 5). This compound has been shown to be effective in preventing the progression of renal injury in animal models [26] and a congener of P123 has moved into clinical trials [27].

Structures of BMP-7 and P123.

Figure 3.
Structures of BMP-7 and P123.

Based on the 3D structure of BMP-7 (A) and its ligand/receptor contact regions, peptide 123 (B) has been designed and optimized to be an agonist of BMP signaling. P123 is a 16-mer peptide and cyclized via a disulfide bond to produce an 11-mer N-terminal loop and 5-mer C-terminal extension with a molecular mass of ∼1.9 kDa [26,32].

Figure 3.
Structures of BMP-7 and P123.

Based on the 3D structure of BMP-7 (A) and its ligand/receptor contact regions, peptide 123 (B) has been designed and optimized to be an agonist of BMP signaling. P123 is a 16-mer peptide and cyclized via a disulfide bond to produce an 11-mer N-terminal loop and 5-mer C-terminal extension with a molecular mass of ∼1.9 kDa [26,32].

TGF-β superfamily members, BMP-7, BMP-4, and Activin, each bind to both type I and type II receptors to induce cellular differentiation.

Figure 4.
TGF-β superfamily members, BMP-7, BMP-4, and Activin, each bind to both type I and type II receptors to induce cellular differentiation.

However, each has a selective binding pattern. For example, BMP-7 binds to type II receptor, BMPR-II, and type I receptors, ALK-2, ALK-3 and ALK-6, whereas BMP-4 binds to type II receptor, BMPR-II, and type I receptors, ALK-3 and ALK-6. On the other hand, Activin binds to a separate type II receptor ActR-II, and type I receptors ALK2 and ALK4. In contrast, TGF-β binds to its own type II receptor, TβR-II, and type I receptor, ALK5, to induce fibrosis.

Figure 4.
TGF-β superfamily members, BMP-7, BMP-4, and Activin, each bind to both type I and type II receptors to induce cellular differentiation.

However, each has a selective binding pattern. For example, BMP-7 binds to type II receptor, BMPR-II, and type I receptors, ALK-2, ALK-3 and ALK-6, whereas BMP-4 binds to type II receptor, BMPR-II, and type I receptors, ALK-3 and ALK-6. On the other hand, Activin binds to a separate type II receptor ActR-II, and type I receptors ALK2 and ALK4. In contrast, TGF-β binds to its own type II receptor, TβR-II, and type I receptor, ALK5, to induce fibrosis.

Interaction of P123 with BMP Receptors.

Figure 5.
Interaction of P123 with BMP Receptors.

Radio-ligand receptor competitive binding assays of P-123 for ALK3 (A) and ALK6 (B) extracellular domain (ECD) (expressed as a fusion protein with Fc domain). In these assays, the compound is examined for its ability to compete with radio-ligand BMP-7 binding to purified ECD of type I receptors ALK3 or ALK6. The peptide binds to ALK2 (not shown) and ALK3, but not ALK6 type I receptor.

Figure 5.
Interaction of P123 with BMP Receptors.

Radio-ligand receptor competitive binding assays of P-123 for ALK3 (A) and ALK6 (B) extracellular domain (ECD) (expressed as a fusion protein with Fc domain). In these assays, the compound is examined for its ability to compete with radio-ligand BMP-7 binding to purified ECD of type I receptors ALK3 or ALK6. The peptide binds to ALK2 (not shown) and ALK3, but not ALK6 type I receptor.

BMP-7 and P123 effects on breast CSC self-renewal in vitro have been examined by observing their respective effects on mammosphere formation. When breast CSCs were grown in non-adherent cell suspension in serum-free medium, they formed tumorspheres after 5–7 days. The treatment of these cells with P123 (300 µM) caused a profound inhibition of tumorsphere formation. Similarly, BMP-7 (500 ng/ml) treatment also inhibited tumorsphere formation. After withdrawal of BMP-7, however, these cells regained the ability to form tumorspheres. The tumorsphere formation induced by P123 was similar to that of BMP-7. This suggests that P123 and BMP-7 have the ability to inhibit self-renewal and growth of human breast CSCs.

Differentiation of chemo-resistant CSCs, specifically a loss of its stem cell phenotype and self-renewal ability, could be a viable approach for the treatment of tumor recurrence and metastasis [28]. Although this approach may or may not directly kill the cancer cells, it could make CSCs more responsive to conventional chemotherapies. Both BMP-7 and P123 affect the differentiation of human breast CSCs. A sensitive flow cytometry method [29] has been used to quantify CD44+ cells, indicating stem cell phenotype, and E-cadherin+ cells, indicating epithelial phenotype, after treatment with or without P123 or BMP-7. Human breast CSCs were grown in non-adherent cell suspension in serum-free medium for 8 days. FACS analysis showed that the majority of cells (93%) were CD44+, with a minority (7%) representing MCF-7 cells from which these CSCs were isolated. Cells treated with P123 (300 µM) or BMP-7 (314–430 ng/ml) showed a marked decline in CD44+ cells. The percent loss of CD44+ cells after treatment with P123 was greater than that after treatment with BMP-7. A gain in E-cadherin+ cells was also observed and was proportional to the loss of CD44+ cells. Similarly, the percent gain in E-cadherin+ cells after treatment with P123 was greater than that after treatment with BMP-7. These results suggest that P123 and BMP-7 have the ability to induce a loss of stem cell phenotype in human breast CSCs and may reverse EMT by promoting epithelial differentiation of CSCs.

Preliminary results demonstrate that BMP-7 and P123, a novel peptide agonist of BMP signaling, have the ability to inhibit self-renewal and growth of human breast CSCs. In addition, these compounds have the potential to reverse EMT by inducing a loss of stem cell phenotype and promoting epithelial differentiation. These findings extend those from our previous studies that demonstrated BMP-7 and P123 are capable of inhibiting bulk tumor cell growth from prostate cancer by activating SMAD 1/5/8 signaling and controlling its cell cycle pathway [25]. Overall, this body of evidence supports the development of novel therapeutics that could potentially treat metastatic breast and prostate cancer. By utilizing the BMP signaling pathway to prevent and reverse the formation of CSCs, it may be possible to prevent metastases and maintain a cell phenotype that can be treated by standard therapeutic options already available.

Abbreviations

     
  • ALK1, 2, 3, 4, 5 or 6

    activin receptor-like kinase 1, 2, 3, 4, 5 or 6

  •  
  • BMP

    bone morphogenetic protein

  •  
  • BMPR-II

    BMP receptor type II

  •  
  • CD44

    cluster of differentiation 44

  •  
  • CSC

    cancer stem cell

  •  
  • E-cadherin

    epithelial cadherin

  •  
  • EMT

    epithelial-to-mesenchymal transition

  •  
  • TGF-β

    transforming growth factor-β.

Competing Interests

The Authors declare that there are no competing interests associated with the manuscript.

References

References
1
National Cancer Institute SEER
.
Cancer Statistics Factsheets
.
National Cancer Institute
,
Bethesda, MD
.
2
Polyak
,
K.
and
Weinberg
,
R.
(
2009
)
Transitions between epithelial and mesenchymal states: acquisition of malignant and stem traits
.
Nat. Rev. Cancer
9
,
265
273
doi:
3
Angeloni
,
V.
,
Tiberio
,
P.
,
Appierto
,
V.
and
Daidone
,
M.G.
(
2015
)
Implications of stemness-related signaling pathway in breast cancer response to therapy
.
Semin. Cancer Biol.
31
,
43
51
doi:
4
Yang
,
M.
,
Liu
,
P.
and
Huang
,
P.
(
2016
)
Cancer stem cells, metabolism and therapeutic significance
.
Tumour Biol.
37
,
5735
5742
doi:
5
Martinez
,
J.
and
Zhang
,
X.H.-F.
(
2013
)
BMP/Coco antagonism as a deterministic factor of metastasis dormancy in lung
.
Breast Cancer Res.
15
,
302
doi:
6
Mani
,
S.A.
,
Guo
,
W.
,
Liao
,
M.-J.
,
Eaton
,
E.N.
,
Ayyanan
,
A.
,
Zhou
,
A.Y.
et al. 
(
2008
)
The epithelial-mesenchymal transition generates cells with properties of stem cells
.
Cell
133
,
704
715
doi:
7
Han
,
L.
,
Shi
,
S.
,
Gong
,
T.
,
Zhang
,
Z.
and
Sun
,
X.
(
2013
)
Cancer stem cells: therapeutic implications and perspectives in cancer therapy
.
Acta Pharm. Sin. B
3
,
65
75
doi:
8
Kobayashi
,
A.
,
Okuda
,
H.
,
Xing
,
F.
,
Pandey
,
P.R.
,
Watabe
,
M.
,
Hirota
,
S.
et al. 
(
2011
)
Bone morphogenetic protein 7 in dormancy and metastasis of prostate cancer stem-like cells in bone
.
J. Exp. Med.
208
,
2641
2655
doi:
9
Al-Hajj
,
M.
and
Clarke
,
M.F.
(
2004
)
Self-renewal and solid tumor stem cells
.
Oncogene
23
,
7274
7282
doi:
10
Yan
,
Y.
,
Zuo
,
X.
and
Wei
,
D.
(
2015
)
Concise review: emerging role of CD44 in cancer stem cells: a promising biomarker and therapeutic target
.
Stem Cells Transl. Med.
4
,
1033
1043
doi:
11
Boman
,
B.M.
and
Wicha
,
M.S.
(
2008
)
Cancer stem cells: a step toward the cure
.
J. Clin. Oncol.
26
,
2795
2799
doi:
12
Scheel
,
C.
,
Eaton
,
E.N.
,
Li
,
S.H.-J.
,
Chaffer
,
C.L.
,
Reinhardt
,
F.
,
Kah
,
K.-J.
et al. 
(
2011
)
Paracrine and autocrine signals induce and maintain mesenchymal and stem cell states in the breast
.
Cell
145
,
926
940
doi:
13
Liu
,
S.
,
Cong
,
Y.
,
Wang
,
D.
,
Sun
,
Y.
,
Deng
,
L.
,
Liu
,
Y.
et al. 
(
2013
)
Breast cancer stem cells transition between epithelial and mesenchymal states reflective of their normal counterparts
.
Stem Cell Rep.
2
,
78
91
doi:
14
Yoshida
,
G.J.
and
Saya
,
H.
(
2016
)
Therapeutic strategies targeting cancer stem cells
.
Cancer Sci.
107
,
5
11
doi:
15
Hollier
,
B.G.
,
Evans
,
K.
and
Mani
,
S.A.
(
2009
)
The epithelial-to-mesenchymal transition and cancer stem cells: a coalition against cancer therapies
.
J. Mammary Gland Biol. Neoplasia
14
,
29
43
doi:
16
Santisteban
,
M.
,
Reiman
,
J.M.
,
Asiedu
,
M.K.
,
Behrens
,
M.D.
,
Nassar
,
A.
,
Kalli
,
K.R.
et al. 
(
2009
)
Immune-induced epithelial to mesenchymal transition in vivo generates breast cancer stem cells
.
Cancer Res.
69
,
2887
2895
doi:
17
Hugo
,
H.
,
Ackland
,
M.L.
,
Blick
,
T.
,
Lawrence
,
M.G.
,
Clements
,
J.A.
,
Williams
,
E.D.
et al. 
(
2007
)
Epithelial–mesenchymal and mesenchymal–epithelial transitions in carcinoma progression
.
J. Cell Physiol.
213
,
374
383
doi:
18
Wang
,
Z.
,
Li
,
Y.
,
Ahmad
,
A.
,
Azmi
,
A.S.
,
Kong
,
D.
,
Banerjee
,
S.
et al. 
(
2010
)
Targeting miRNAs involved in cancer stem cell and EMT regulation: an emerging concept in overcoming drug resistance
.
Drug Resist. Updat.
13
,
109
118
doi:
19
Singh
,
A.
and
Settleman
,
J.
(
2010
)
EMT, cancer stem cells and drug resistance: an emerging axis of evil in the war on cancer
.
Oncogene
29
,
4741
4751
doi:
20
Cowin
,
P.
and
Welch
,
D.R.
(
2007
)
Breast cancer progression: controversies and consensus in the molecular mechanisms of metastasis and EMT
.
J. Mammary Gland Biol. Neoplasia
12
,
99
102
doi:
21
Deonarain
,
M.P.
,
Kousparou
,
C.A.
and
Epenetos
,
A.A.
(
2009
)
Antibodies targeting cancer stem cells: a new paradigm in immunotherapy?
MAbs
1
,
12
25
PMID:
[PubMed]
22
Gupta
,
P.B.
,
Chaffer
,
C.L.
and
Weinberg
,
R.A.
(
2009
)
Cancer stem cells: mirage or reality?
Nat. Med.
15
,
1010
1012
doi:
23
Bosukonda
,
A.
,
Rehman
,
F.
,
Mallette
,
S.A.
and
Sluss
,
P.M.
(
2009
)
Bone morphogenetic protein-6 (BMP-6) efficacy and its mechanism of action to suppress human prostate cancer cell growth
.
Biol. Reprod.
81
,
372
24
Haudenschild
,
D.R.
,
Palmer
,
S.M.
,
Moseley
,
T.A.
,
You
,
Z.
and
Reddi
,
A.H.
(
2004
)
Bone morphogenetic protein (BMP)-6 signaling and BMP antagonist noggin in prostate cancer
.
Cancer Res.
64
,
8276
8284
doi:
25
Bosukonda
,
D.
,
Keck
,
P.C.
,
Carlson
,
W.D.
,
Bosukonda
,
A.
,
Maheswaran
,
S.
and
Sluss
,
P.M.
(
2010
)
A novel peptide agonist of the BMP signaling, unlike BMPs, escapes neutralization by Follistatin while retaining the ability to inhibit prostate cancer cell proliferation
.
In Program, 8th International Conference on Bone Morphogenetic Proteins
,
Leuven
,
Abstract #A63
26
Sugimoto
,
H.
,
LeBleu
,
V.S.
,
Bosukonda
,
D.
,
Keck
,
P.
,
Taduri
,
G.
,
Bechtel
,
W.
et al. 
(
2012
)
Activin-like kinase 3 is important for kidney regeneration and reversal of fibrosis
.
Nat. Med.
18
,
396
404
doi:
27
A Study of THR-184 in Patients at Risk of Developing Cardiac Surgery Associated-Acute Kidney Injury (CSA-AKI), ClinicalTrials.gov. Identifier NCT01830920, April 10, 2013.
28
Chirasani
,
S.R.
,
Sternjak
,
A.
,
Wend
,
P.
,
Momma
,
S.
,
Campos
,
B.
,
Herrmann
,
I.M.
et al. 
(
2010
)
Bone morphogenetic protein-7 release from endogenous neural precursor cells suppresses the tumourigenicity of stem-like glioblastoma cells
.
Brain
133
,
1961
1972
doi:
29
Gupta
,
V.
,
Zhang
,
Q.-J.
and
Liu
,
Y.-Y.
(
2011
)
Evaluation of anticancer agents using flow cytometry analysis of cancer stem cells
.
Methods Mol. Biol.
716
,
179
191
doi:
30
Thiery
,
J.P.
and
Sleeman
,
J.P.
(
2006
)
Complex networks orchestrate epithelial–mesenchymal transitions
.
Nat. Rev. Mol. Cell Biol.
7
,
131
142
doi:
31
Wharton
,
K.
and
Derynck
,
R.
(
2009
)
TGFβ family signaling: novel insights in development and disease
.
Development
136
,
3691
3697
doi:
32
Griffith
,
D.L.
,
Keck
,
P.C.
,
Sampath
,
T.K.
,
Rueger
,
D.C.
and
Carlson
,
W.D.
(
1996
)
Three-dimensional structure of recombinant human osteogenic protein 1: structural paradigm for the transforming growth factor 13 superfamily
.
Proc. Natl Acad. Sci. U.S.A.
93
,
878
883
PMID:
[PubMed]