Author's response to reviews Title:Inhibiting Inducible miR-223 Further Reduces Viable Cells in Human Cancer Cell Lines MCF-7 and PC3 Treated by Celastrol Authors:

There are two major modifications, -1In ‘Methods and results’ Original ‘Down-regulating or over-expressing miR-223 could enhance or reduce celastrol’s anti-proliferation effects, respectively. In addition, celastrol could cause early activation and later inhibition of mTOR, as well as raise HSP70 (preceded by HSF-1 activation). Interestingly, these three events, i.e., mTOR activation, HSF-1-related HSP70 elevation and miR-223 induction showed mutually dependent relations, each action promoting the other two. Moreover, inhibiting mTOR or down-regulating HSP70 could reduce cell proliferation in both celastrol-treated and untreated cells, indicating mTOR activation and HSP70 elevation are also important to cell proliferation.’ Revised ‘Down-regulating miR-223 could increase the number of viable cells, yet it further reduced viable cells in samples that were treated by celastrol; up-regulation of miR-223 displayed opposite effects. Celastrol’s miR-223 induction might be due to NF-κB inhibition and transient mTOR activation: these two events occurred prior to miR-223 elevation in celastrol-treated cells. NF-κB inhibitor, like celastrol, could induce miR-223; the induction of miR-223 by NF-κB inhibitor or celastrol was reduced by the use of mTOR inhibitor. Finally and interestingly, miR-223 also could affect NF-κB and mTOR and the effects were different between cells treated or not treated with celastrol, thus providing an explanation for differing effects of miR-223 alteration on cellular viability in the presence of celastrol or not.’ Explanation By prolonging the incubation time of cells with overor down-regulated miR-223, we found that miR-223 manipulation could affect cell viability. We have modified the original expressions to avoid being too general and abstract, as suggested by some reviewers. -2In ‘Conclusion’ part Original ‘we disclose that in presence of toxic celastrol, cells used at least three inter-promoting anti-toxic weapons in struggling to survive and proliferate, and that down-regulating miR-223 or inhibiting mTOR, in addition to blocking heat shock response, are novel ways to enhance celastrol’s anti-cancer effects.’ Revised ‘we disclose that celastrol could induce miR-223 in breast and prostate cancer cells, and that inhibiting miR-223 could further reduce the living cells in celastrol-treated cancer cell lines. We thus provide a novel way to increase celastrol’s anti-cancer effects.’ Explanation The original expression has been changed for clarity. Background There is one major modification in the last paragraph, Original ‘Therefore, whether or not celastrol-caused miR-223 elevation affects celastrol’s anti-proliferation action, and if so, why, are interesting questions. To address them, we observed miR-223 alterations caused by celastrol in human breast cancer line MCF-7 and prostate cancer line PC3, as well as the effects of miR-223 regulation on celastrol’s anti-proliferative effects on these two cells. In addition, we observed the relation of miR-223 induction to heat shock response, mTOR and NF-κB activation, as these responses are related to celastrol’s anti-tumor effects [10, 18-21].’ Revised ‘Therefore, whether or not celastrol-caused miR-223 elevation affects celastrol’s anti-cancer action, and if so, why, are questions worth addressing. To do so, we first observed miR-223 alterations caused by celastrol in human breast cancer line MCF-7 and prostate cancer line PC3 (two of the most common types of cancer and the two cancer types most often used in celastrol studies), (please see note 1 below for reviewers only), as well as the effects of manipulating miR-223 on celastrol’s ability to reduce the number of living cells. Then, we investigated the possible reason for celastrol’s miR-223 induction by focusing on how altering NF-κB affects miR-223 expression, since celastrol is a known NF-κB regulator [19-21], and NF-κB reportedly regulates miR-223 [22]. In addition, in pre-experimental trials, we found that NF-κB activity affected and was linked to mTOR activity and HSP70 levels. (please see note 2 below for reviewers only). Therefore, the effects of altering mTOR and HSP70 on miR-223 expression were also investigated. Finally, we tried to find the possible molecular basis by which miR-223 alterations affected cellular viability in cells treated or not treated with celastrol. Again, we focused on NF-κB, mTOR, and HSP70, since these three molecules are widely reported as related to celastrol’s anti-tumor effects [10, 23-26].(please see note 3 below for reviewers only). Importantly, miR-223 could regulate NF-κB [27], mTOR [28, 29], and members of the heat shock protein family [28].’ Explanation The reviewers suggested that we add (1) the reason for choosing MCF-7 and PC3 and (2) explain why we focused on NF-κB, mTOR, and HSP70. Note 1 According to the latest data about cancer incidences released in GLOBOCAN 2012 by WHO, the five most common cancers worldwide are lung, breast, colorectal, prostate, and liver. At the beginning of 2014, by reviewing the literature in PUMBED, we found that prostate and breast cancers are the common cancers most often used for celastrol testing. These relative frequencies remain the same (see Table R2). This is why we chose the breast cancer cell line MCF-7 and prostate cancer cell line PC3 for our study. Table R2. Number of publications about celastrol and five most common cancers in PUBMED Rank (incidence) Type of Cancer No. of publications Publication PMID (some papers include two or more cancer types, thus their PMID may be mentioned more than once below) 1 Lung 11 25891850, 25359574, 24662070, 24615207, 24293411, 23850675, 21570383, 21466843, 21249311,20472666, 19916747 2 Breast 17 (No.1) 25818165, 25419573, 25310109, 25149175, 24465597, 23850675, 23497885, 23266469, 21865725, 21786165, 21134410, 21088503, 20934245, 20798912, 20480272, 20154087, 19671767 3 Colorectal 5 25149175, 23814530, 22877649, 20798912, 20605676 4 Prostate 15 (No.2) 25866772, 24664372, 24654943, 24235831, 23554889, 22666381, 21786165, 21170316, 20934245, 20160026, 19545787, 18726991, 17093135, 17010675, 16651429 5 Liver 8 25051375, 24965400, 24859482, 24849358, 23988648, 23649759, 23284992, 22369852 (The total number of publications focusing on celastrol and cancer is 126, excluding papers currently under review ) Note 2 In a pre-experimental test, we found that NF-κB inhibitor could activate mTOR and HSF-1 (detected at 1 h after inhibitor addition) and caused HSP70 elevation (detected at 6 h after inhibitor addition) (Fig. R1), indicating that NF-κB activity is linked to mTOR activity and HSP70 levels. Fig. R1. Effects of altering NF-κB on mTOR and HSF-1 activation and HSP70 level Note 3 The relationship between celastrol’s anti-cancer effects and alterations in NF-κB or mTOR or heat shock proteins are widely reported (Table R3). Table R3. Number of papers reporting molecular events involved in celastrol’s anti-cancer effects Altered molecule Number of paper Publication PMID (not including review) NF-κB 22 25776492; 24755677; 24664372; 24615207; 24465597; 24434352; 24157922; 23988648; 23799016; 22641218; 21951963; 21865725; 21850367; 21506956; 21170316; 20798912; 20346213; 19934274; 19778460;19671767 mTOR 6 25818165; 25051375; 24859482; 24667175; 23065091; 20160026 Heat 29 25571843; 24954307; 24903326; 24849358; 24810052; 24662070; 24589236; HSP70 β-actin p-mTOR p-NF-κB MCF-7 PC3 p-HSF-1 Ku Control NF-κB inhibitor Control NF-κB inhibitor HSP70 β-actin p-mTOR p-NF-κB