Reply to Tanaka and Kunieda: Control protein GFP also shows a mesh-like structure in desiccating tardigrade cells

We recently reported an in vivo expression and live imaging technology for tardigrades and an unexpected tissue specificity of anhydrobiosis-related gene expression, with the CAHS3 gene, for example, being most intensively expressed in the epidermis of tardigrades (1). Using immunohistochemistry, Tanaka and Kunieda now provide interesting confirmation that CAHS3 is indeed localized in the epidermis but is also expressed in the storage cells (2). The authors then suggest careful consideration of the use of vector-induced expression based on this result. As we have shown in SI Appendix, Fig. S5, while SAHS1 is most intensively expressed in the storage cells, some expression was observed in the epidermis (1). This is consistent with the several-fold higher expression of SAHS1 and lower expression of CAHS3 in the storage cells as observed by our tissue-specific RNA-Seq (1), and this quantitative nature is another useful aspect of live imaging. Our previous work was based on observations at 48 h after vector injection with only a short cis-regulatory region, which should not be confused with the complex proteostasis of native proteins. On the other hand, we now have evidence that TardiVec is retained and functional for weeks, with low-contrast imaging showing weak expression of CAHS3 in storage cells, which would facilitate our understanding of protein turnover and maintenance in tardigrades. Nevertheless, we appreciate the confirmation of the tissue-specific expression pattern of CAHS3 by immunohistochemistry, with no observable signal in muscle or gastrointestinal cells.

One of the focuses of recent molecular studies of tardigrade anhydrobiosis has been the reversible filamentous gel formation of the CAHS protein, which we reported for CAHS1 (3), followed by several papers on different paralogs and orthologs (4–7) including that for CAHS3 by Tanaka et al. (8). In their commentary, Tanaka and Kunieda argue that filamentous structures are observable in desiccating tardigrade cells by immunohistochemistry (2), but we prefer to be cautious about such interpretations as we have previously reported, based on no observable difference from controls expressing mEGFP and mCherry (1). Our recent observations of anhydrobiotic tardigrades have provided a finer representation of tardigrade cells. In an epidermal cell from a control mEGFP-expressing tardigrade, we observed a mesh-like structure that appears to correspond to Tanaka and Kunieda’s (2, figure 2B) (Fig. 1). Based on our live imaging observations, the mesh-like structure can be attributed to the presence of multiple vesicles and cell compaction during desiccation. It would be interesting to see how other control proteins behave under immunohistochemistry, but the ease of using heterologous proteins for control is another advantage of vector-based live imaging.

Fig. 1.

mEGFP showed a mesh-like structure in an epidermis cell of a tardigrade upon desiccation. A tardigrade introduced with pRvCAHS3-mEGFP vector was observed on a gelatin-coated glass for 30 min RT. (The scale bar is 10 µm.)

Direct observation in tardigrades is an important direction, with traditional approaches like immunohistochemistry facilitated by new techniques such as our vector system. In addition, newly developed methods such as expansion microscopy and spatial transcriptomics can provide a clearer picture of anhydrobiotic tardigrades.