The current cell oxygen paradigm shows some major gaps that have not yet been resolved. Something seems to be lacking for the comprehensive statement of the oxygen distribution in the cell, especially the low cytoplasmic oxygen level. The entrapment of oxygen in microtubules (MTs) resolves the latter observation, as well as the occurrence of an extensive cytoplasmic foam formation. It leads to a novel oxygen paradigm for cells. During the steady-state treadmilling, the mobile cavity would absorb oxygenated cytoplasm forward, entrap gas nuclei and concentrate them. A fluorescence method is described to confirm the in vitro load of oxygen in MTs during their periodic growths and shrinkages. The latter operating mechanism is called the gas dynamic instability (GDI) of MTs. Several known biosystems could rest on the GDI. (1) The GTP-cap is linked with the gas meniscus encountered in a tube filled with gas. The GTP hydrolysis is linked to the conformational change of the GTPase domain according to the bubble pressure, and to the shaking of protofilaments with gas particles (soliton-like waves). (2) The GDI provides a free energy water pump because water molecules have to escape from MT pores when foam concentrates within the MT. Beside ATP hydrolysis in motor proteins, the GDI provides an additional driving force in intracellular transport of cargo. The water streams flowing from the MT through slits organize themselves as water layers between the cargo and the MT surface, and break ionic bridges. It makes the cargo glide over a water rail. (3) The GDI provides a universal motor for chromosome segregation because the depolymerization of kinetochorial MTs is expected to generate a strong cytoplasmic foam. Chromosomes are sucked up according to the pressure difference (or density difference) applied to opposite sides of the kinetochore, which is in agreement with Archimedes' principle of buoyancy. Non-kinetochorial MTs reabsorb foam during GDI. Last, the mitotic spindle is imagined as a gas recycler. (4) The luminal particles within MTs (called MIPs) are imagined as a foam organizer, the luminal proteins being part of the borders and edges of identical bubbles. (5) Last, volatile anesthetics could destabilize MTs through anesthetic-induced bubble nucleation between protofilaments, and therefore causing shear stress and the opening of MT. The load of oxygen in MTs might provide a major advance in this area of research.