Antiangiogenesis: Dll4 is a notch above VEGF

Blocking Delta-like ligand 4 (Dll4) results in increased but dysfunctional tumor vascularization, opening up a new avenue for antiangiogenenic cancer therapy.

The Notch ligand Delta-like ligand 4 (Dll4) has a crucial role in both normal and tumor-related development of blood vessels. Until now, cancer therapies targeting tumor vascularization have largely focused on vascular endothelial growth factor (VEGF), yet some tumors are immune to anti-VEGF treatment. Noguera-Troise et al. and Ridgway et al. now show in Nature that inhibition of Dll4 activity leads to decreased tumor growth in mouse models.

Noguera-Troise et al. transplanted rat C6 tumor cells overexpressing either Dll4 or Dll4 Fc — a fusion protein consisting of Dll4 linked to the immunoglobulin G1 constant region that blocks the Dll4/Notch pathway — into mice. Notch signaling was reduced in cells from C6 Dll4-Fc tumors relative to C6 Dll4 tumor cells. Morphologically, the expression of Dll4-Fc resulted in more highly branched and denser vascular structures than Dll4. While tumor growth is generally facilitated by vascularization, the more densely vascularized C6 Dll4-Fc tumors were surprisingly less than half the size of C6 Dll4 tumors.

These results were confirmed by Ridgway et al. using a range of transplanted tumors treated with a phage antibody that blocked the interaction between Dll4 and Notch. Both groups discovered that anti-Dll4 treatment was effective against tumors that did not respond to VEGF inhibition. Additionally, Ridgway et al. found that a marker of differentiated endothelial cells was lacking in the retinal endothelium — an in vivo model for vessel development — of anti-Dll4 treated mice. Noguera-Troise et al. in turn showed that C6 Dll4 Fc tumors were more hypoxic than C6 Dll4 tumors, due to the fine vascular branches of the Dll4-Fc tumor tissue being poorly perfused with blood.

Noguera-Troise et al. also achieved a decrease in tumor size through the systemic delivery of Dll4-Fc to mice with an adenoviral expression system. Importantly, Ridgway et al. observed that during anti-Dll4 treatment, crypt progenitor cells in the intestines remained undifferentiated, demonstrating that the Notch signaling pathway remained active in these cells. Thus, Dll4/Notch signaling appears to be restricted to endothelial differentiation and vascular development. As the crypt stem cells utilizing Notch signaling do not appear to be affected by anti-Dll4 treatment, the probability of severe side effects by such a cancer treatment seems to be low.

In summary, both studies showed that Dll4 activity is necessary for functional vascular development in tumors. Based on the tissue specificity of this pathway and the absence of side effects or increased lethality in mice, Dll4 inhibitors offer a new possibility for cancer therapy.

Mirko von Elstermann
Functional Glycomics Gateway

Original References:
Noguera-Troise I et al.
Blockade of Dll4 inhibits tumour growth by promoting non-productive angiogenesis.
Nature 444, 1032 - 1037 (2006)
Full text | PDF | Subscribe to Nature

Ridgway J et al.
Inhibition of Dll4 signalling inhibits tumour growth by deregulating angiogenesis.
Nature 444, 1083 - 1087 (2006)
Full text | PDF | Subscribe to Nature

Osteoclast signaling: CaMKIV-CREB closes RANK

The Calmodulin-dependent kinase IV and cAMP-dependent response element binding protein (CaMK-CREB) pathway regulates the maturation and activation of osteoclasts.

Osteoclasts resorb bone material formed by osteoblasts, thus helping to regulate both bone density and extracellular Ca²⁺ concentration. During osteoclast development and activity, a Ca²⁺/calmodulin signaling cascade is initiated by the co-stimulation of RANK (receptor activator of nuclear factor-B) and immunoglobulin-like receptors. This cascade releases the CaMKs from autoinhibition, which phosphorylate and activate the transcription factor CREB, and stimulates the phosphatase calcineurin. Finally, Ca²⁺/calmodulin signaling leads to the transcription of NFAT1c (the nuclear factor of activated T-cells 1c) — itself a transcription factor —, resulting in osteoclast differentiation. Sato, Suematsu and colleagues now describe in Nature Medicine how the CaMK-CREB pathway contributes to the induction of NFAT1c transcription and NFAT1c-dependent gene expression in osteoclasts.

The growth of osteoclast precursor cells in the presence of the CaMK inhibitor KN-93, as well as the inhibition of CREB, completely abolished their differentiation to multinucleated cells. Specifically, gene knockdown of several CaMK family members by small interfering RNAs revealed that only CaMKIV was important for osteoclast formation; accordingly, CaMKIV was found to be activated following RANK stimulation, and mice deficient for CaMKIV also showed defects in osteoclast formation.

Inhibition of CREB or CaMK also blocked the induction of NFATc1 and significantly diminished the number of NFATc1-positive osteoclasts. Furthermore, administration of KN-93 to precursor cells reduced levels of c-Fos — a known transcription factor for NFAT1c — by 60%. These results indicate that c-Fos and NFAT1c expression are under the control of the CaMKIV-CREB system.

The authors cultivated differentiated osteoclasts in the presence of KN-93 to test if the CaMKIV-CREB pathway is active in later stages of osteoclast development. They found that KN-93 strongly reduced the expression of genes such as calcitonin that have CREs (cAMP-dependent response elements) in their promotor regions. Collating their results, the authors propose two phases of CaMK-CREB activity. In osteoclast precursors, CaMK-phosphorylated CREB induces the transcription of c-Fos, which is necessary for NFAT1c transcription. NFATc1 is then dephosphorylated by Ca²⁺/Calmodulin-activated calcineurin and translocated into the nucleus, where — together with CREB — it initiates transcription of genes characteristic for mature osteoclasts, in turn leading to bone resorbing activity.

The ability to block osteoclast development and activity has implications for the treatment of osteoporosis. Mice subjected to ovariectomies — a model of menopause-induced osteoporosis — showed a significant reduction in osteoclast formation and loss of bone material after they were administered KN-93, supporting the notion that the CaMK-CREB pathway might be a useful target for the development of osteoporosis treatments.

Mirko von Elstermann
Functional Glycomics Gateway

Original References:
Sato K et al.
Regulation of osteoclast differentiation and function by the CaMK-CREB pathway.
Nature Medicine 12, 1410 - 1416 (2006)
Full text | PDF | Subscribe to Nature Medicine

FGF signaling: Ligand specificity uncovered

The membrane protein Klotho directly binds fibroblast growth factor 23 (FGF23) and determines the ligand specificity of FGF receptor 1 (FGFR1).

Fibroblast growth factors (FGFs) signal through tyrosine kinase membrane receptors to regulate a diverse range of cellular processes, including cell growth, differentiation and cell death. FGF23, the newest member of the FGF family, has been shown to control the renal expression of key enzymes involved in vitamin D and phosphate metabolism. While this suggests the existence of a FGF23-specific receptor in the kidney, no such molecule has been discovered. In Nature, Takeyoshi Yamashita and colleagues report that the membrane protein Klotho confers FGF23 specificity to the canonical FGF receptor FGFR1(IIIc). This paper outlines a novel mechanism of receptor modulation that may serve as a paradigm for other ligand-receptor systems.

Intravenous injection of FGF23 in mice induces extracellular signal-regulated kinase (ERK) phosphorylation and induces the expression of early growth-responsive 1 (Egr-1) selectively in the kidney. Matrix-assisted laser desorption ionization-time-of-flight (MALDI-TOF) mass spectrometry analyses of renal homogenate proteins adsorbed by FGF23-Sepharose led to the identification of Klotho as a major FGF23-binding protein.

Previous studies have shown that Klotho is predominantly expressed in the kidney and shares sequence similarity with β-glucosidase enzymes. Furthermore, defective Klotho expression in mice causes phenotypic traits, such as decreased bone mineral density and ectopic calcification, which are also observed in FGF23-null mice, thus suggesting a functional relationship between Klotho and FGF23.

Klotho expression conferred FGF23 responsiveness to cells that do not normally respond to this growth factor. Moreover, in HEK-293 cells, an anti-Klotho monoclonal antibody that prevents binding of radiolabeled FGF23 to overexpressed Klotho blocked the FGF23 but not the basic FGF response. In agreement with these results, Klotho-deficient mice were unresponsive to the injection of recombinant FGF23, and injection of wild-type mice with the blocking antibody increased serum calcium to levels comparable with those found in Klotho- and FGF23-deficient mice. These results indicate that Klotho is essential for FGF23 function in vivo.

Finally, overexpression studies in rat myoblasts, which lack endogenous FGF receptors, demonstrated that FGFR1(IIIc) is required to transduce the FGF23 signal. Klotho both binds to and induces the phosphorylation of FGFR1(IIIc) in response to FGF23. Yamashita and colleagues conclude that Klotho determines the ligand specificity of FGFR1, enabling a tissue-specific response to FGF23. Whether Klotho or its homologues assist in other FGF-FGFR interactions remains to be determined.

Monica Hoyos Flight
Cell Migration Gateway

Original References:
Itaru Urakawa, Yuji Yamazaki, Takashi Shimada, Kousuke Iijima, Hisashi Hasegawa, Katsuya Okawa, Toshiro Fujita, Seiji Fukumoto and Takeyoshi Yamashita.
Klotho converts canonical FGF receptor into a specific receptor for FGF23
Nature 444, 770-774 (7 December 2006)
Full text | PDF | Subscribe to Nature

Dpp signaling: Mad degradation by eIF4A

The translation factor eIF4A regulates decapentaplagic (Dpp) signaling by inducing the degradation of the activated Smad transcription factor complex during Drosophila development.

Decapentaplagic (Dpp), a transforming growth factor-β (TGF-β) family member, initiates dorsalization as well as other morphogenic processes during Drosophila embryogenesis. Binding of Dpp to the TGF-β-type receptor Tkv leads to phosphorylation of Smad family member Mad, which then migrates to the nucleus to regulate gene transcription. Ubiquitination of phosphorylated Mad (pMad) by the ubiquitin ligase DSmurf and proteosomal degradation are known to regulate Dpp signaling.

Previously, Jinghong Li and Willis Li observed that mutant eIF4A — a component of the translation initiation complex — enhanced Dpp signaling in Drosophila. In Nature Cell Biology these researchers now describe that eIF4A binds to pMad to initiate proteosomal degradation, thus complementing the actions of DSmurf.

Li and Li showed initially that pMad accumulated in embryos possessing a dominant-negative mutation of eIF4A (eIF4A^R321H), while only low levels of pMad occurred in eIF4A wild-type embryos. The eIF4A^R321H embryos died as a result of the overstimulation of the Dpp signaling pathway. Li and Li went on to find that yeast Gal4 promoter-induced expression of eIF4A in the Drosophila eye completely prevented the phenotypic effects of ectopic dpp expression on eye development, while the expression of eIF4A^R321H did not. Taken together, these results suggest that eIF4A blocks pMad accumulation by promoting its degradation. Immuno-precipitation of pMad and eIF4A fragments demonstrated that pMad binds eIF4A directly.

pMad degradation still occurs with the overexpression of eIF4A^R321H, while purified eIF4A^R321H protein inhibits translation in vitro. This excludes the possibility that eIF4A promotes the translation of a Dpp signaling inhibitor and suggests that the role of eIF4A as a translation factor is independent of its role in Dpp signaling.

Li and Li found that mutations in both eIF4A and DSmurf, but not mutations in either gene alone, led to a complete rescue of the lethal genotype of flies with only one copy of dpp (1xdpp⁺). The authors conclude that mutations in both these genes boost Dpp signaling in 1xdpp⁺ embryos sufficiently to rescue embryogenesis.

In summary, eIF4A seems to induce degradation of pMad independently of its ubiquitination by DSmurf, suggesting the existence of a dual system of pMad degradation. The specificity of eIF4A activity for the Dpp signaling pathway was supported by the finding that mutant eIF4A did not affect Hedgehog or Wnt signaling. Further studies might demonstrate that other Smad proteins, as well as mediators of Dpp signaling, such as Schnurri and Brinker, are regulated by interaction with eIF4A.

Mirko von Elstermann
Functional Glycomics Gateway

Original References:
Li J, Li WX.
A novel function of Drosophila eIF4A as a negative regulator of Dpp/BMP signalling that mediates SMAD degradation.
Nature Cell Biology 8, 1407-1414 (2006)
Full text | PDF | Subscribe to Nature Cell Biology

Affolter A, Pyrowolakis G.
eIF4A goes beyond translation.
Nature Cell Biology 8, 1319-1321 (2006)
Full text | PDF | Subscribe to Nature Cell Biology