Description

­In many cell types, subsets of RNA binding proteins adopt a default de-mixed state, condensing into ribonucleoprotein (RNP) granules (Alberti and Carra, 2018). In the C. elegans germ line PGL-1, an RGG domain protein, is condensed into P granules. P granules associate closely with nuclear pores of pachytene nuclei in the adult germ line but detach from nuclei as oocytes develop (Strome and Wood, 1982; Pitt et al., 2000) and contribute to maternal mRNA regulation (Voronina, 2013). In contrast, other RNA binding proteins have a default mixed, or decondensed, state in the adult germ line. For example, the KH domain protein MEX-3 is largely diffuse throughout the cytosol of oocytes and condenses into P granules in early embryos (Draper et al., 1996).

Multiple studies have shown that several types of RNA binding proteins undergo dynamic phase separation during certain environmental stresses and extended meiotic arrest. P-granule proteins, such as PGL-1, P-body proteins such as CGH-1, and stress granule proteins such as TIAR-2 are all detected in large RNP granules in meiotically arrested oocytes (Jud et al., 2008; Noble et al., 2008; Wood et al., 2016). The RNP granules are hypothesized to maintain oocyte quality during delays in fertilization. Similarly, in a study of heat stress responses, P-body and stress granule proteins also condense into large RNP granules in oocytes (Jud et al., 2008). In contrast, in the same study a summary table states that PGL-1 granules are not detected after heat stress; however, images of this experiment were not published (Jud et al., 2008).

Recent studies using the PGL-1::GFP reporter strain demonstrate that PGL-1 decondenses rapidly in the early embryo after the temperature is upshifted to 34°C (Putnam et al., 2019). P granules in the early embryo are similar, but not identical to P granules in oocytes. As mentioned above, a subset of proteins are transient P granule components. Since MEG-3 plays an important role in stabilizing PGL-1 granules in the early embryo, but does not associate strongly with P granules prior to fertilization, it is not necessarily the case that P granule dynamics in oocytes will recapitulate dynamics observed in embryos (Wang et al., 2014). The prior heat stress study by Jud et al., was conducted before CRISPR reporter strains were available; therefore, we asked if PGL-1::GFP decondenses in oocytes in response to increased temperature. PGL-1 was detected in perinuclear and cytoplasmic P granules as expected in control oocytes (median of 69 granules) and in the distal germ nuclei. In contrast, after 40 minutes at 34°C in a stagetop incubator (Method 1), in the majority of worms PGL-1 appeared dispersed, with significantly fewer granules detected in the proximal -2 to -5 oocytes (median of 10 granules) (Figure 1). This result appears similar to the original result described, but not shown, that detected PGL-1 by immunofluorescence after fixing worms immediately after being exposed to 34°C via a standard incubator (Jud et al., 2008). As a side note, we did not detect similar dispersal of PGL-1 in the distal nuclei, suggesting nuclear-associated PGL-1 granules may be less sensitive to temperature changes (arrowheads, Figure 1). While this manuscript was in revision, an independent study found similar results in oocytes and distal nuclei (Fritsch et al., 2021). Interestingly, when we incubated the PGL-1:GFP worms in a standard incubator and imaged without a stagetop incubator (Method 2), many more PGL-1 granules were detected in oocytes than with Method 1 (median of 104 granules; Figure 1). This may not seem surprising since the worms could be expected to recover at ambient temperature, and many phase transitions are reversible. However, it is notable, as a practical consideration, that the phase separation of PGL-1 into granules occurred quickly, within 15 minutes at ambient temperature.

From these results with the PGL-1::GFP strain, we confirm that in proximal oocytes PGL-1 is extremely sensitive to temperature changes, and behaves as a dynamic, liquid-like protein, similar to its behavior in the early embryo (Jud et al., 2008; Putnam et al., 2019). While short fluctuations in temperature during heat stress experiments may not adversely affect all fluorescent reporter strains, they can clearly affect phase transitions of liquid-like proteins such as PGL-1.