**4. Discussion**

As TCTP is a conserved protein from yeas<sup>t</sup> to human and autophagy is a conserved protein degradation pathway, we speculated that examination of the role of yeas<sup>t</sup> TCTP (Mmi1 protein) in autophagy might help us to understand controversial results from higher eukaryotic cells, indicating both positive and negative e ffects on autophagy [10,11]. Using batch-cultivated yeas<sup>t</sup> cells of *S. cerevisiae*, we demonstrated that in exponentially growing cells Mmi1 protein had no e ffect on basal or nitrogen starvation-induced bulk non-selective autophagy (Figure 1). However, if exponentially growing cells were treated with rapamycin, Mmi1 negatively a ffected autophagy (Figures 2–4) when the cells entered the post-diauxic growth phase (Figure 5). Further, in the post-diauxic growth phase we detected lower amount of superoxide radicals in *mmi1* Δ cells compared to WT cells (Figure 6). Our results also indicate that WT and *mmi1* Δ cells possess same sensitivity to rapamycin (Figure 7).

Previously, Chen and colleagues suggested that mammalian TCTP could positively regulate autophagy. By using African green monkey kidney fibroblast-like cells (COS-7 cells) they demonstrated that TCTP knockdown inhibits autophagy under hypoxic or serum starvation conditions [10]. The e ffect of hypoxic conditions on autophagy has not been tested in this study and awaits further exploration. However, when the exponentially growing yeas<sup>t</sup> cells were shifted from the rich YPD medium to nitrogen starvation conditions, high autophagy levels were induced but no e ffect of Mmi1 was detected. These results are consistent with the finding that a very slight or no e ffect on autophagy was previously detected in response to nutrient starvation condition in the knockdown TCTP human HeLa cell line and mouse embryonic fibroblasts (MEFs), haploinsu fficient in TCTP expression [11].

Further, contrary to a positive role of TCTP on autophagy, Bae and colleagues reported that TCTP/TPT1 negatively regulates autophagy [11]. They detected increased autophagy in HeLa cells transiently transfected with *TPT1* shRNA or in mouse embryonic fibroblasts (MEFs) from heterozygote knockout mice embryos (*Tpt1*+/−). The negative autophagy regulation was potentiated by rapamycin and an increased autophagy was also demonstrated in vivo in livers and kidneys of *Tpt1* heterozygote mice [11]. Our results indicate that in yeas<sup>t</sup> cells Mmi1 negatively regulates the rapamycin-induced autophagy (Figures 2 and 3) when glucose is exhausted from media (Figure 5B) and the cells enter post-diauxic growth phase. In contrast, no e ffect on autophagy is detected under nitrogen starvation condition (Figure 1) when glucose is still present in the medium (Figure 5B).

It is generally accepted that both rapamycin treatment and nitrogen starvation inhibit downstream the TOR signaling pathway that results in repression of protein translation and proliferation and leads to autophagy stimulation [31,32]. However, our results indicate a di fferent e ffect of Mmi1 on autophagy, based on the used conditions. The rapamycin-induced autophagy might di ffer from the nitrogen starvation-induced autophagy somewhere within TOR inhibition and/or downstream of TOR signaling that happens during post diauxic growth phase that is not present during nitrogen starvation.

Rapamycin has been thought to fully deactivate the budding yeas<sup>t</sup> TORC1 and driving cells into a quiescent/G0 state [33]. However, this dogma has changed since many groups reported only slowed proliferation upon rapamycin treatment [47,53–55] and only partial inhibition of yeas<sup>t</sup> TORC1 [54]. In fact, rapamycin also partially inactivates mammalian mTORC1 [57,58]. It seems evident that maintained proliferative activity upon rapamycin treatment is crucial for detection of increased autophagy induction in *mmi1* Δ cells in our study. Interestingly, we have noticed that upon rapamycin addition to exponentially growing cells both, the WT and the *mmi1* Δ strain grow similarly (Figure 5A), and the *mmi1* Δ strain lost its slow growth phenotype [19]. Since Mmi1 possesses pro-survival activity, we might speculate that during the exponential growth Mmi1 possesses some activity that is directly regulated by TORC1.

Upon rapamycin treatment we detected the increased autophagy in *mmi1* Δ cells of the post-diauxic growth phase. This phenotype could be rescued by the wild type *MMI1* gene inserted into the *mmi1* Δ strain (Figure 4). At the diauxic shift cells switch from fermentation to respiration and from rapid proliferation to slow proliferation [59]. Further, PKA and TOR pathways are downregulated, and PKC and Snf1 pathways are activated, the former transiently [60], the latter important for the induction of a carbon starvation autophagy in cells undergoing the diauxic shift [61]. Importantly, the carbon starvation autophagy is not induced when the cells are grown in glucose medium, and then shifted to carbon starvation media [61]. It requires an absence of catabolite repression [61] that is responsible for a preferential utilization of glucose to other carbon sources and strict repression of respiration [62]. At the moment the mechanism how Mmi1 negatively a ffects the rapamycin-induced autophagy in the post-diauxic phase is unknown and awaits a detailed exploration.

However, it has been reported that the reactive oxygen species can hamper inhibitory activity of rapamycin in *S. cerevisiae* by oxidative damage to yeas<sup>t</sup> TORC1 [47]. Our results indicate that in the post-diauxic growth phase WT cells exhibit a higher amount of reactive oxidative species compared to *mmi1* Δ cells (Figure 6). Since WT and *mmi1* Δ cells exhibit the same sensitivity to rapamycin (Figure 7), it seems unlikely that the decreased ROS production in the *mmi1* Δ strain influences rapamycin binding to TORC1. Further, the reactive oxygen species accumulation was shown to be critical for autophagy induction during nutrient starvation conditions in mammalian cells [63] and a number of studies indicate autophagy regulation by redox signaling [64]. In *S. cerevisie*, activity of cysteine protease Atg4 could be regulated by the redox state and, hence, it may regulate autophagosome formation [65]. Recently, ethanol stress-induced autophagy was reported to also be regulated by ROS [66].

Here, we used exponentially growing cells and induced autophagy by nitrogen starvation or rapamycin treatment. These approaches were introduced already in 1990s [31,32] and have been used by many groups to trigger autophagy in cells previously grown in the rich YPD medium. Nevertheless, in a natural habitat, cells face large variations in nutrients, and some autophagy roles in cellular metabolism seems to be still unexplored [67]. In this respect, Iwama and Ohsumi recently reported that a bulk autophagy is activated in batch culture on low glucose media based on available carbon sources [68]. Further, Horie and colleagues reported that iron recycling via autophagy is critical for transition from glycolytic to respiratory growth [67]. Several experiments are now in progress to test the e ffect of Mmi1 on autophagy induced during cell aging and quiescence. So far, our results support the role of Mmi1/TCTP as a negative regulator in the rapamycin-induced non-selective autophagy in eukaryotic cells.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2073-4409/9/1/138/s1, Figure S1: Growth of WT and *mmi1* Δ strains in nitrogen starvation media.

**Author Contributions:** J.V.; designed and performed the experiments, analyzed data, and wrote the manuscript, J.H.; reviewed the manuscript and was responsible for funding acquisition and project administration. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was supported by the gran<sup>t</sup> from the Czech Science Foundation CSF16-05497S (J.H.) and by the fellowship from J. W. Fulbright Commission (Prague, Czech Republic) (J.V.).

**Acknowledgments:** We are very grateful to Suresh Subramani for enabling the learning the autophagy methods in his laboratory at the University of California, San Diego, and for helpful discussions. We also thank to Mark Rinnerthaler (University of Salzburg, Austria) and Ivana Malcova for critical comments on the manuscript, Jaroslav Vojta for the help with statistical evaluation of the data, and Lenka Novakova for her technical assistance.

**Conflicts of Interest:** The authors declare no conflict of interest.
