The tolerance to hypoxia is defined by a time-sensitive response of the gene regulatory network in sea urchin embryos

Deoxygenation, the reduction of oxygen level in the oceans induced by global warming and anthropogenic disturbances, is a major threat to marine life. Acute diurnal changes in oxygen levels could be especially harmful to vertebrate and sea urchin embryos that utilize endogenous hypoxia gradients to drive morphogenetic events during normal development. Here we show that the tolerance to hypoxic conditions changes between different developmental stages of the sea urchin embryo, due to the structure of the gene regulatory networks (GRNs). We demonstrate that during normal development, bone morphogenetic protein (BMP) pathway restricts the activity of the vascular endothelial growth factor (VEGF) pathway to two lateral domains and by that controls proper skeletal patterning. Hypoxia applied during early development strongly perturbs the activity of Nodal and BMP pathways that affect VEGF pathway, dorsal-ventral (DV) and skeletogenic patterning. These pathways are largely unaffected by hypoxia applied after DV axis formation. We propose that the structure of the DV GRN, that includes feedback and feedforward loops, increases its resilience to changes of the initial oxygen gradients and helps the embryos tolerate transient hypoxia.

Global warming leads to the reduction in dissolved molecular oxygen (O2) and increased respiration of

34
During the evolution of metazoans, animals were exposed to variations in oxygen levels and molecular 35 mechanisms evolved to enable organisms to cope with hypoxic conditions (Semenza, 2012). However, 36 it is still unclear whether these mechanisms are sufficient to protect organisms from acute hypoxic

41
Understanding the molecular mechanisms that mediate the response to both physiological and 42 environmental hypoxia during the development of marine animals is a key to understand the expected 43 effect of ocean deoxygenation on marine biodiversity.

44
The sea urchin embryo provides an attractive system to study the molecular pathways that respond to      by the transcription factor Not1 that is activated by Nodal signaling (Fig. 1D

158
HIF1α KD does not affect VEGF spatial expression pattern in these two time points (Fig. 2B).

159
Additionally, HIF1α KD does not affect VEGF, VEGFR and BMP2/4 expression level at both times

177
We specifically wanted to distinguish between the effect of hypoxia applied during the formation of the   expression of VEGF to one side of the ectoderm (Fig. 4M). In addition, the expression of VEGFR 217 expands beyond the two lateral skeletogenic cell clusters in which it is normally localized (Fig. 4N).

223
Early hypoxia is mediated through the reduction in BMP activity

224
The expansion of the ventral side in hypoxic conditions suggests that BMP activity at the dorsal side 225 might be reduced, and the reduction of the repressing BMP activity could explain VEGF expansion to 226 the dorsal side. To test this hypothesis and monitor BMP activity in normal vs. hypoxic condition, we 227 performed immunostaining against pSMAD1/5/8. We studied pSMAD1/5/8 signal at two different 228 developmental stages; mesenchyme blastula, when BMP activity is localized at the dorsal ectoderm 229 (Fig. 5A), and at late gastrula, when BMP activity is localized at the dorsal skeletogenic cells (Fig. 5C).

230
Hypoxic conditions completely abolish pSMAD1/5/8 signal from the nuclei of the dorsal ectodermal 231 cells at mesenchyme blastula stage (Fig. 5B). At late gastrula stage, hypoxic conditions eliminate the 232 pSMAD1/5/8 signal from the dorsal skeletogenic cells (Fig. 5D), or strongly reduce it (Fig. 5E). These     (Fig. 7A, B). We discovered that the structure of these GRNs makes them very sensitive to hypoxic conditions applied at early development, but quite tolerant to these conditions if

284
Early hypoxia in sea urchin embryos strongly distorts DV and skeletogenic patterning due to its strong 285 effect on the key DV patterning gene, Nodal that controls BMP activity (Fig. 7A,B); however, late 286 hypoxia doesn't affect Nodal expression, apparently since it is maintained by auto-regulation at this  Nodal-BMP2/4-Chordin incoherent feedforward loop, which restricts VEGF activity, and normal 301 patterning is observed (Fig. 7A). Overall, the structure of the DV GRN enables it to partially recover 302 early hypoxia at later developmental stages and makes the GRN resilient to late hypoxia.

303
Our findings illuminate the similarities between the DV patterning and skeletogenic GRNs and the

366
Reactions were carried out in 10μl volume including: 5µl SYBR BioRad IQ SYBR Green Supermix 367 (#1725125), 2.5μl of 1.2μM forward and reverse gene specific primers and 2.5μl of cDNA (qPCR 368 primers used in this study are listed in Table S1). Each cDNA sample was run in triplicate, for every 369 candidate gene, ubiquitin was used as internal control. The reactions thermal profile was: 95°C for 3 370 minutes followed by 40 amplification cycles of 95°C for 10 seconds and 55°C for 30 sec. Dissociation    NaCl, for 10 minutes at room temperature, then embryos exposed to Methanol for 1 minute. Embryos

414
All images presented in this study were generated on a Zeiss Axio Imager M2.

416
We thank Imad Shams for generously providing the hypoxic chamber and gas controller; Eli shemesh 417 for providing the oxygen sensor; Shlomo Ben-Tabou de-Leon for technical assistance with the hypoxic