Could Embryos Be Blind to Some Rhythms? A Sampling Idea for the Spine

Film a spinning wagon wheel and it can appear to stand still or turn backward. The camera takes one frame at a time, and some motions fall between the frames. Our research desk proposes that a developing embryo does something similar as it lays down the segments that become the spine.

Vertebrae, ribs and back muscles come from blocks of tissue called somites. Somites form one at a time, each time a genetic oscillator, the segmentation clock, completes a cycle. In zebrafish that cycle takes about 30 minutes, depending on temperature. One cycle, one boundary.

The desk points out that this is exactly what engineers call sampling: reading a continuous signal once per tick of a clock. Sampling has well-known consequences, worked out in the sampling theorem that Claude Shannon set out in 1949 and that Nyquist and Kotelnikov anticipated earlier.

Rhythms faster than half the sampling rate fold down and masquerade as slower ones. This is called aliasing. A rhythm at an exact multiple of the sampling rate is seen at the same phase every time, so it leaves no rhythm at all.

Suppose you nudge a 30-minute clock with a weak, repeating drug pulse, too weak to take over the clock's timing. Then measure the length of each somite along the body. The desk predicts three things:

• Folding. A pulse every 20 minutes, one and a half times the clock rate, should produce a slow pattern in somite size, alternating large and small. • Twins. Pulses every 22.5 minutes and every 18 minutes should leave the same pattern, repeating every three somites, because they sit symmetrically around one and a half times the clock rate (the 20-minute pulse). • A blind spot. A pulse every 15 minutes, exactly twice the clock rate, should leave no repeating pattern in somite size at all. We checked the arithmetic, and the numbers are consistent. The desk's own subjective estimate, which cannot be checked, is a 35% chance that embryos show all three effects cleanly.

The closest tools exist, but not in the right species. In 2018, Katharina Sonnen, Alexander Aulehla and colleagues reported in Cell a microfluidic method that delivers timed pulses of signaling drugs to mouse embryo tissue. They used it to entrain, or lock, the segmentation clock to an external rhythm, and showed that timing between signaling pathways controls segmentation.

That work used mouse presomitic mesoderm explants, whose clock runs roughly 140 minutes, with pulses every 130 minutes. The proposed 15-, 18-, 20and 22.5-minute pulses assume the 30-minute zebrafish clock. A mouse version would need rescaled periods, about 47, 70, 94 and 105 minutes for a 140-minute clock. Otherwise a zebrafish delivery method would have to be named, and none is cited here.

A 2021 protocol from Sonnen's group describes how to build such chips, including with a 3D printer. That makes the mouse platform accessible, but it does not by itself make the proposed experiment feasible.

The two-dimensional mouse explants also do not lay down a long series of discrete somites. Measuring somite length as a data series would need a different preparation, such as whole embryos or 3D somitoid culture.

The desk believes the regime is new. Sonnen 2018 used entraining pulses near the clock's own period, and the desk proposes deliberately weak pulses that do not capture it, then reading somite sizes as a data stream. We did not run a literature search, so we cannot say that this combination has never been tried.

• Sampling may be smeared. If cells along the tissue are not perfectly in step, the sampling moment blurs and the blind spot fills in. • The drug may act continuously. If the pulse changes the speed of the wavefront rather than acting at one phase, the clean pattern breaks. • The signal may be small. Somite sizes vary naturally. Zebrafish form roughly 30 to 34 somites and mice about 65, so the number of embryos needed to see a few percent of modulation depends on the species chosen, and the desk has not worked out that estimate. The weakest point is the gap between the math and the available experimental platform: the sampling logic is settled, but no one has shown that a weak pulse can be delivered and read out this way in either species.

Congenital scoliosis and some vertebral malformations trace back to errors in somite formation, including mutations in segmentation-clock genes. The desk speculates that knowing which rhythms the embryo can and cannot register might one day help explain why some early-pregnancy exposures disturb the spine and others do not. That is an extrapolation without a source, not evidence.

The broader lesson applies to every biomarker a longevity patient tracks. Whenever you measure once per cycle, whether a hormone once a morning or glucose once a visit, some rhythms will be invisible and others will masquerade as something slower. Continuous monitors exist partly for this reason.

• A lab running sub-entrainment pulses at periods scaled to its species' clock, such as 22.5, 18 and 15 minutes for zebrafish, and analyzing somite length as a series. • Confirmation by live imaging that the clock kept its own rhythm during weak pulsing. • Whether the same sampling logic appears in other rhythmic body clocks, such as hormone pulses.