Click here to close
Hello! We notice that you are using Internet Explorer, which is not supported by Xenbase and may cause the site to display incorrectly.
We suggest using a current version of Chrome,
FireFox, or Safari.
Sci Rep
2017 Jan 27;7:41339. doi: 10.1038/srep41339.
Show Gene links
Show Anatomy links
Discovering novel phenotypes with automatically inferred dynamic models: a partial melanocyte conversion in Xenopus.
Lobo D
,
Lobikin M
,
Levin M
.
???displayArticle.abstract???
Progress in regenerative medicine requires reverse-engineering cellular control networks to infer perturbations with desired systems-level outcomes. Such dynamic models allow phenotypic predictions for novel perturbations to be rapidly assessed in silico. Here, we analyzed a Xenopus model of conversion of melanocytes to a metastatic-like phenotype only previously observed in an all-or-none manner. Prior in vivo genetic and pharmacological experiments showed that individual animals either fully convert or remain normal, at some characteristic frequency after a given perturbation. We developed a Machine Learning method which inferred a model explaining this complex, stochastic all-or-none dataset. We then used this model to ask how a new phenotype could be generated: animals in which only some of the melanocytes converted. Systematically performing in silico perturbations, the model predicted that a combination of altanserin (5HTR2 inhibitor), reserpine (VMAT inhibitor), and VP16-XlCreb1 (constitutively active CREB) would break the all-or-none concordance. Remarkably, applying the predicted combination of three reagents in vivo revealed precisely the expected novel outcome, resulting in partial conversion of melanocytes within individuals. This work demonstrates the capability of automated analysis of dynamic models of signaling networks to discover novel phenotypes and predictively identify specific manipulations that can reach them.
Figure 1. Conversion of melanocytes to a metastatic-like state.(A) Dorsal view of a wild-type st. 45 Xenopus laevis tadpole; note the small melanocytes, absent from large areas of the head (green arrowhead). (B) In contrast, animals resulting from any of several treatments2039 that depolarize instructor (GlyR-expressing) cells or perturb their downstream serotonergic signaling exhibit extensive overabundance and uniform coverage with highly arborized melanocytes (red arrowhead). This occurs in an all-or-none manner in some percentage of the animals (frequency depending on the specific manipulation)2125. Sectioning (level of section shown in schematics to the right) reveals the main features of melanoma-like phenotype: over-proliferation, arborization, and invasiveness. (C) Normally round melanocytes dorsal to the neural tube (green arrow) become highly arborized and drop down over the neural tube itself (D, red arrows). Inset panels show blood vessels, normal in c’ and covered by invasive melanocytes in d’, as occurs in melanoma. Sections further along the tail likewise show small numbers of round melanocytes in control larvae (E, green arrows) compared to the excess of long, abnormally extended and ectopically localized melanocytes in converted animals (F, red arrows). In converted animals, cells can be seen invading the lumen (G) or neural tissue (H) of the neural tube, also often forming networks (I) as has been observed in vasculogenic mimicry of mammalian melanoma40. The authors thank Vaibhav Pai for allowing us to use his drawing of a st. 45 Xenopus laevis tadpole in this figure.
Figure 2. Inferred signaling network model and in silico predictions.(A) A computational method reverse-engineered a signaling network able to recapitulate the level of conversion stochasticity of a series of pharmacological experiments. (B) Phenotypic predictions of the model for all experimental combinations of up to three reagents. Each point represents the mean distance of 100 simulations of a specific experiment (i.e., a specific combination of reagents) to the two extreme tadpole phenotypes (normal and total conversion). Training dataset in green, validation dataset in red, new experiments in blue. Only one combination of three reagents (red arrow) is far from the two extreme phenotypes, indicating a partially converted phenotype.
Figure 3. Phase space of the wild type and the experiment predicted to produce a partially pigmented phenotype.The trajectories of the state of two serotonin receptors (5HT-R1 and 5HT-R2) and the degree of conversion is shown among 100 simulations. (A) Without any treatment (wild type), the dynamical system transitions from the initial instable state (embryo, yellow dot) stochastically towards a final stable state with a normal conversion level (0, blue dot) or towards a final state with a high conversion level (0.92, red dot). (B) Applying the treatment combining the discovered three reagents (altanserin, reserpine, and constitutively active CREB), the dynamics reveal a bifurcation in the dynamical system and the appearance of a new attractor with a partial conversion level (0.25, green dot) in addition to another attractor with a normal conversion level (0, blue dot). The attractor dot size is proportional to the number of converging trajectories to that final state.
Figure 4. Dynamics of the degree of conversion and perturbed pathway components in the untreated and treated experiments.The panels show the dynamics of the degree of conversion and levels of the affected components by the drugs in the treatment through time over 100 simulations. (A) With no treatment (wild type), only extreme phenotypes (normal or high conversion levels) are stochastically produced. VMAT converges to high levels, R2 to stochastically either high or low levels, and CREB to low levels. (B) The treatment combining reserpine (inhibiting VMAT), altanserin (inhibiting R2), and constitutively active CREB (increasing CREB) produces the expected changes in the levels of the affected pathway components, lowering the levels in the case of inhibitors (VMAT and R2) or increasing them in the case of adding constitutively active CREB. Notice that in both A and B there are trajectories converging at 0 conversion level (normal).
Figure 5. In vivo validation of the computationally discovered partially converted phenotype.Normal melanocytes are indicated with blue arrowheads, while areas populated with converted melanocytes are indicated with red arrowheads. (A) Control animals exhibit small, round melanocytes confined to stripes around the spinal cord and over the brain (A). Note the absence of melanocytes in the periocular region (A’). In normal animals’ tails, melanocytes never exit the stripes of axial muscle (do not venture into the fin). (B) Converted animals produced by Ivermectin-induced depolarization exhibit spread-out (arborized) melanocytes all along the flank (B) and around the eyes (B’)–they colonize every region of the head. The arborized melanocytes also leave the mid-body and colonize the fin (B”). (C) In contrast, embryos treated with VP16-XlCreb1and exposed to 5HT-R2 and VMAT blockers revealed a never before-seen phenotype where some regions of the animal were converted (C, periocular region of the right eye, red arrowhead), and some were wild-type (C, periocular region of left eye, blue arrowhead). This can be also seen in the transformation of cell shape over the brain (C’, red arrowhead) but normal melanocyte morphology elsewhere (C’, blue arrowhead). In the tail (C”,C’”) we observed some cells invading the fin, but they had normal morphology. We observed a great diversity of different combinations that were intermediate between normal and converted outcomes (only a representative sample is shown here), which otherwise never occurred in the same animal. (D) Furthermore, we noticed a new, cancer-like behavior that had not occurred in any previous experiment. This included the induction of individual nodules, which were both pigmented and unpigmented (D,D’); this occurred in ~10% of the animals that likewise exhibited partial conversion (D”, red arrowhead region vs. blue arrowhead region).
Arcangeli,
Targeting ion channels in cancer: a novel frontier in antineoplastic therapy.
2009, Pubmed
Arcangeli,
Targeting ion channels in cancer: a novel frontier in antineoplastic therapy.
2009,
Pubmed
Barco,
Expression of constitutively active CREB protein facilitates the late phase of long-term potentiation by enhancing synaptic capture.
2002,
Pubmed
Bates,
Ion channels in development and cancer.
2015,
Pubmed
Becker,
Reverse-engineering post-transcriptional regulation of gap genes in Drosophila melanogaster.
2013,
Pubmed
Blackiston,
A novel method for inducing nerve growth via modulation of host resting potential: gap junction-mediated and serotonergic signaling mechanisms.
2015,
Pubmed
,
Xenbase
Blackiston,
Transmembrane potential of GlyCl-expressing instructor cells induces a neoplastic-like conversion of melanocytes via a serotonergic pathway.
2011,
Pubmed
,
Xenbase
Chernet,
Use of genetically encoded, light-gated ion translocators to control tumorigenesis.
2016,
Pubmed
,
Xenbase
Chernet,
Endogenous Voltage Potentials and the Microenvironment: Bioelectric Signals that Reveal, Induce and Normalize Cancer.
2013,
Pubmed
Crombach,
Efficient reverse-engineering of a developmental gene regulatory network.
2012,
Pubmed
Daniolos,
Action of light on frog pigment cells in culture.
1990,
Pubmed
,
Xenbase
Dulcis,
Illumination controls differentiation of dopamine neurons regulating behaviour.
2008,
Pubmed
,
Xenbase
Hendrix,
Vasculogenic mimicry and tumour-cell plasticity: lessons from melanoma.
2003,
Pubmed
Jaeger,
Dynamic control of positional information in the early Drosophila embryo.
2004,
Pubmed
King,
The automation of science.
2009,
Pubmed
Lang,
Ion channels in cancer: future perspectives and clinical potential.
2014,
Pubmed
Levin,
Molecular bioelectricity in developmental biology: new tools and recent discoveries: control of cell behavior and pattern formation by transmembrane potential gradients.
2012,
Pubmed
Levin,
Molecular bioelectricity: how endogenous voltage potentials control cell behavior and instruct pattern regulation in vivo.
2014,
Pubmed
Levin,
Reprogramming cells and tissue patterning via bioelectrical pathways: molecular mechanisms and biomedical opportunities.
2013,
Pubmed
Levin,
Endogenous bioelectrical networks store non-genetic patterning information during development and regeneration.
2014,
Pubmed
Liu,
Reverse Engineering of Genome-wide Gene Regulatory Networks from Gene Expression Data.
2015,
Pubmed
Lobikin,
Serotonergic regulation of melanocyte conversion: A bioelectrically regulated network for stochastic all-or-none hyperpigmentation.
2015,
Pubmed
,
Xenbase
Lobikin,
Selective depolarization of transmembrane potential alters muscle patterning and muscle cell localization in Xenopus laevis embryos.
2015,
Pubmed
,
Xenbase
Lobikin,
Resting potential, oncogene-induced tumorigenesis, and metastasis: the bioelectric basis of cancer in vivo.
2012,
Pubmed
,
Xenbase
Lobo,
Computational discovery and in vivo validation of hnf4 as a regulatory gene in planarian regeneration.
2016,
Pubmed
Lobo,
Inferring regulatory networks from experimental morphological phenotypes: a computational method reverse-engineers planarian regeneration.
2015,
Pubmed
Manu,
Canalization of gene expression and domain shifts in the Drosophila blastoderm by dynamical attractors.
2009,
Pubmed
Marguet,
Treatment during a vulnerable developmental period rescues a genetic epilepsy.
2015,
Pubmed
Morokuma,
Modulation of potassium channel function confers a hyperproliferative invasive phenotype on embryonic stem cells.
2008,
Pubmed
,
Xenbase
Perkins,
Reverse engineering the gap gene network of Drosophila melanogaster.
2006,
Pubmed
Reinitz,
Model for cooperative control of positional information in Drosophila by bicoid and maternal hunchback.
1995,
Pubmed
Reinitz,
Stripe forming architecture of the gap gene system.
1998,
Pubmed
Roubos,
Background adaptation by Xenopus laevis: a model for studying neuronal information processing in the pituitary pars intermedia.
1997,
Pubmed
,
Xenbase
Smolen,
Simulations suggest pharmacological methods for rescuing long-term potentiation.
2014,
Pubmed
Soldatova,
An ontology for a Robot Scientist.
2006,
Pubmed
Uzkudun,
Data-driven modelling of a gene regulatory network for cell fate decisions in the growing limb bud.
2015,
Pubmed
Villaverde,
Reverse engineering and identification in systems biology: strategies, perspectives and challenges.
2014,
Pubmed
Yang,
Membrane potential and cancer progression.
2013,
Pubmed