Chernet BT et al. (2015), Long-range gap junctional signaling controls on...

XB-ART-50073

Front Physiol 2015 Jan 19;5:519. doi: 10.3389/fphys.2014.00519.

Show Gene links Show Anatomy links

Long-range gap junctional signaling controls oncogene-mediated tumorigenesis in Xenopus laevis embryos.

Chernet BT , Fields C , Levin M .

???displayArticle.abstract???
In addition to the immediate microenvironment, long-range signaling may be an important component of cancer. Molecular-genetic analyses have implicated gap junctions-key mediators of cell-cell communication-in carcinogenesis. We recently showed that the resting voltage potential of distant cell groups is a key determinant of metastatic transformation and tumor induction. Here, we show in the Xenopus laevis model that gap junctional communication (GJC) is a modulator of the long-range bioelectric signaling that regulates tumor formation. Genetic disruption of GJC taking place within tumors, within remote host tissues, or between the host and tumors significantly lowers the incidence of tumors induced by KRAS mutations. The most pronounced suppression of tumor incidence was observed upon GJC disruption taking place farther away from oncogene-expressing cells, revealing a role for GJC in distant cells in the control of tumor growth. In contrast, enhanced GJC communication through the overexpression of wild-type connexin Cx26 increased tumor incidence. Our data confirm a role for GJC in tumorigenesis, and reveal that this effect is non-local. Based on these results and on published data on movement of ions through GJs, we present a quantitative model linking the GJC coupling and bioelectrical state of cells to the ability of oncogenes to initiate tumorigenesis. When integrated with data on endogenous bioelectric signaling during left-right patterning, the model predicts differential tumor incidence outcomes depending on the spatial configurations of gap junction paths relative to tumor location and major anatomical body axes. Testing these predictions, we found that the strongest influence of GJ modulation on tumor suppression by hyperpolarization occurred along the embryonic left-right axis. Together, these data reveal new, long-range aspects of cancer control by the host's physiological parameters.

???displayArticle.pubmedLink??? 25646081
???displayArticle.pmcLink??? PMC4298169
???displayArticle.link??? Front Physiol

Species referenced: Xenopus laevis
Genes referenced: gjb2 grap2 kcna4 kcna5 kras rel tbx2 tp53

???displayArticle.disOnts??? cancer

???attribute.lit??? ???displayArticles.show???

	Figure 1. KRASG12D injection results in the formation of induced tumor like structures that are gap junctionally connected to the host. (A) The schematic shows injection of KRASG12D + RLD (Rhodamine-Lysinated Dextran, M. W. 10 kDa) + LY (M.W. 0.522 kDa) into one cell of a 16-cell stage embryos. The RLD labels injected cells (and their descendants), while LY is small enough to traverse gap junctions. (B) At a stage 34 embryo (white dotted outline), LY signal is present not only in tissue derived from the injected cell (tumor; green arrowheads) but also widely throughout the rest of the host. (C) RLD shows cells injected with oncogene, some of which are in the localized tumors and others that have migrated (metastasized) out into the host. (D) Overlay of the two signals reveals areas (white arrowheads) that exhibit LY signal but not RLD signal, which demarcate cells that are in active GJ communication with the injected (tumor) cells (N = 5 of 5). Schematic in panel A (and in subsequent figures) is used with permission from Nieuwkoop and Faber (1967).
	Figure 2. Selective disruption of gap junctional communication (GJC) reduced tumor incidence. To determine the effect of GJC disruption on tumorigenesis, H7 and KRASG12D injections—aimed at disrupting GJC host-wide, within tumors, and away from tumors—were performed. Compared to KRASG12D-only injected embryos (treatment mode A), treatment modes B and D showed significant decrease in % embryos with tumor by 6.6 and 5.8%, respectively, implying that H7 contralateral to the oncogene has no effect if there is H7 ipselateral to the oncogene. Contralateral H7 only (treatment mode C) significantly reduces the % embryos with tumor by 15.8%. Doubling the level of oncogene injected along with host-wide introduction of H7 (treatment mode F) does not affect tumor incidence when compared to oncogene-only (one side) injected embryos (treatment mode A), but is able to counter the 9.9% increase in tumor incidence resulting from excess oncogene introduced (treatment mode G) down to the tumor incidence level of one-side oncogene injection. Similarly, tumor incidence from the excess oncogene introduction can be reduced by H7 introduction to either side of the embryo (treatment mode E). P < 0.05, *P < 0.001; t-test.
	Figure 3. Enhanced gap junctional communication (GJC) through connexin 26 (Cx26) overexpression increases tumor incidence. To determine the effect of GJC disruption on tumorigenesis, Cx26 and KRASG12D injections—aimed at enhancing GJC host-wide, within tumors, and away from tumors—were performed. Compared to KRASG12D-only injected embryos (treatment mode A), long-range and host-wide Cx26 treatment modes (C, D, F) show an increase in the number of embryos with tumor by 6.4 to 11.4%. However, treatments involving the expression of oncogene and Cx26 on the same side (treatment modes B and E), while non-significant, lower tumor incidence when compared to treatment modes A and F, respectively. *P < 0.05; t-test.
	Figure 4. “Left-right synchronization” model for the establishment (Stage 1) of electrical states that modify long-range gap-junction-mediated influences on tumorigenesis. The “left-right synchronization” model shows that stable, bioelectrically-synchronized cell populations on the left and right sides of the embryo, here represented by red and blue regions, can be generated from a random initial state by nearest-neighbor interactions alone. This model is implemented on a fixed 10 × 10 grid with the left and right exterior boundaries mathematically identified to yield a cylindrical topology. This topology allows the two synchronized populations to “wrap” around the exterior border. The final model shown here was obtained by varying the left—right polarization bias parameter “X” while requiring two stable populations with no ectopic islands as output. This procedure allowed the value of the left—right polarization bias to the predicted to be 27%, consistent with observations (Levin et al., 2002; Adams et al., 2006). While the left-right border is represented at low resolution in the model and most solutions produce a straight, sharp boundary, this is not a biologically-meaningful constraint and the actual border between “left” and “right” cell populations in the embryo may be irregular as shown. The “left-right communication” model assumes that the two bioelectrically-synchronized sides exchange a long-range, oscillatory bioelectric signal; whether this signal requires an additional trigger for initiation is currently unknown. This model employs a small set of assumptions to quantitatively predict the tumor incidence expected when GJC is either suppressed by H7 or enhanced by Cx26. These predictions are then compared with experimental tumor frequencies (indicated by “?”). See Figure 5 for additional details of this model.
	Figure 5. The stage-2 “left-right communication” model for the GJC-mediated effects of Vmem change on incidence of tumorigenesis. The “left-right communication” model assumes that the left and right sides of the embryo alternately signal each other to turn the rate of cell division down (A) using an oscillation between polarization and depolarization as the long-range bioelectric signal (B). For example, KRASG12D and H7 expressions on opposite sides of the embryo (as shown in Figure 2 treatment C, and sections through a trunk tumor in Panel D) result in an increase in signaling from the H7-injected side and a consequent suppression in KRASG12D-induced tumors on the opposite side (C,D). The stage-2 model can be visualized as a control network that regulates cellular response to KRASG12D and both the production of ipsilateral signal (IS) and response to contralateral signal (CS). Here unlabeled flat-end arrows indicate a 100% suppressing effect (E). In this model, gap junctions are a central component, contributing to the long-range dynamics by allowing cells to sense neighbors' Vmem. Cellular mechanisms corresponding to these arrows are not yet fully characterized; however, the butyrate—histone deacetylase pathway previously characterized in these embryos (Chernet and Levin, 2013a,b, 2014); is a plausible candidate for transducing the global bioelectrical signal into a local cellular response to KRASG12D transformation.
	Figure 6. Predicted tumor incidences that match those of experimental data under GJC perturbations were derived from the stage-2 “left-right communication” model. In predicting tumor incidences, the left-right communication model implements the following assumptions about cellular responses to KRASG12D and H7 (A). Assumption 1: embryos have two sides, A and B. During normal embryonic development, sides A and B exchange a handshaking signal that limits cell division. Each side responds to the other side's (i.e. contralateral) signal but not to its own (ipsilateral) signal. A moderately increased or decreased signal is not detrimental to the normal development of unperturbed embryos. Assumption 2: experimental data showed 35% of KRASG12D-injected embryos produce tumors. KRASG12D injections on opposite sides of 16-cell embryos results in 44.9% tumor incidence. The 44% incidence is evenly split between two sides (22.45% incidence per injected side). Assumption 3: H7 makes cells deaf to the contralateral signal; in response, they increase production of their own ipsilateral signal, by 50%. Assumption 4: Increased (decreased) contralateral signal suppresses (enhances) KRASG12D activity on a % for % basis. Ipsilateral H7 directly suppresses oncogenic transformation of KRASG12D expressing cells by 20%; contralateral H7 has no direct effect. Assumption 5: KRASG12D expressing cells disable the increase in signal generation ipsilaterally by interfering with KRASG12D-ipsilateral side bioelectric synchronization. Taking into account these five assumptions, the predicted tumor values were calculated for each treatment mode and compared to the observed tumor incidence. The model accurately predicts tumor incidence for every oncogene combination. The stage-2 “left-right communication” model implements the following assumptions about cellular responses to KRASG12D and Cx26 to account for enhanced (as opposed to disrupted) GJC (B). Assumptions 1 and 2 of the model remain the same as stated in panel (A). Assumption 3: Cx26 enhances cells' exposure the contralateral signal; in response, they decrease production of their own ipsilateral signal, by 50%. Assumption4: Increased (decreased) contralateral signal suppresses (enhances) KRASG12D activity on a % for % basis. Ipsilateral Cx26 directly increases oncogenic transformation of KRASG12D cells by 20%; Contralateral Cx26 has no direct effect. While the model predicts the general trend of increased tumorigenesis as a result of enhanced GJC, saturation of tumor incidence around 46% is observed, resulting in higher predictive than actual values.
	Figure 7. Testing a unique prediction of our model: tumor incidence is affected by long-range signaling across the left-right axis but not across the dorso-ventral axis. (A) (i) Embryos were injected with H7 and KRASG12D, separately and randomly, in the dorsal and ventral blastomeres. (ii) Embryos were also injected with H7 and KRASG12D in the left and right blastomeres, or vise versa. Control embryos injected with KRASG12D only displayed a 35% tumor incidence. (B) Compared to controls, embryos injected with KRASG12D and H7 across the dorsoventral axis did not show a change in tumor incidence (32.4%). Whereas perturbation of GJC communication across the left-right axis significantly lowered tumor incidence down to 20.5% (*P < 0.05, X2 test).
	Figure 8. Long-range signaling in the frog embryo. Oncogene-expressing cells (red, Xrel3-tdTomato) are prevented from forming tumors by hyperpolarizing channel-expressing cells (Kv 1.5-β-gal) located at a distance. Compare (A) without to (B) with Xrel3-tdTomato. The ~4 mm distance from the center of the tdTomato signal to the farthest point on the tail expressing β-gal was estimated to be ~300 cell diameter (average single cell diameter of 13.3 μm).

References [+] :

Adams, Endogenous voltage gradients as mediators of cell-cell communication: strategies for investigating bioelectrical signals during pattern formation. 2013, Pubmed