Tigecycline and homoharringtonine synergistically target myeloid leukemia cells by inhibiting 1 mitochondrial translation through mTOR / 4 EBP 1 pathway 2

Author Email address Haiyang Yang haiyanglucky@163.com Chen Mei meichenblood@yeah.net Li YE yelikybs@126.com Yanling Ren yanlingrenxm @163.com Hua Zhang sapphirezh@126.com Kongfei Li konfeelee@126.com Jiansong Huang hjiansong1234@zju.edu.cn Xin Huang hx0628@yahoo.com Weilai Xu lpxwl2005@163.com Xinping Zhou zxp1207@163.com Gaixiang Xu 1505136@zju.edu.cn Chuying Shen 21818095@zju.edu.cn Lu Wang 11818213@zju.edu.cn Hongyan Tong tonghongyan@zju.edu.cn


Background 26
Tigecycline (TIG) is a new generation of tetracycline antibiotic which contains the N,N-27 dimethylglycylamido substitution at position 9 of minocycline [1]. The mechanism of action is similar to that 28 of ordinary tetracyclines, and it binds to the A site of the 30 s subunit of the bacterial ribosome. Moreover, 29 TIG also binds to the remaining part of the H34 ribosome, resulting in a stronger inhibition of bacterial 30 translation [2]. Ribosome protection and drug efflux are the primary mechanisms by which most bacteria 31 develop tetracycline resistance [3]. The steric hindrance of TIG is due to a large substituent at position 9 32 appears to surmount these resistance mechanisms [2]. 33 7 Animal experiments in this study were performed in accordance with the "Guidelines for the Care and Use 124 of Laboratory Animals" (NIH Publication 86-23, revised in 1985) issued by the National Institutes of Health. 125 The animal experiment protocol was approved by the Animal Protection and Facilities Committee of Zhejiang 126 University. Male nude mice of six-week-old were purchased from Zhejiang Academy of Medical Sciences. 127 The SKM-1 myeloid leukemia cell line (1  107 cells per animal) was subcutaneously injected into the right 128 forelimb flank of the mice. Seven days after cell injection, when the tumors were palpable, animals with 129 approximately the same tumor volume were randomly divided into four groups consisting of five animals per 130 group. The use of TIG (50 mg/kg /d), HHT (0.5 ng/kg/d) or a combination for two consecutive weeks. For the 131 control group, PBS was used for the injection. Both the drugs and PBS were intraperitoneally injected. The 132 mice were weighed every day and calipers were used to calculate the tumor volume every two days using the 133 formula: π/6 length × width2. Two weeks later, in accordance with ethical animal practices, each group of 134 mice was killed, and the tumors were collected for further testing. 135

Statistical analysis 136
The significant differences among the groups was determined using a one-way ANOVA (Analysis of 137 Variance) followed by a post-hoc Bonferroni's multiple comparison test. For non-parametric data, a Kruskal-138 Wallis followed by a Dunn's multiple comparison test was used. The minimal level of significance was p < 139 0.05. All data were presented as the mean ± standard deviation (SD). GraphPad PrismV 5.0 (GraphPad 140 Software, San Diego, CA) was used for statistical analysis. * p < 0.05; ** p < 0.01; ***p < 0.001; ****p < 141 0.0001. 142 Results

143
TIG-HHT combination can synergistically inhibit the growth of myeloid leukemia cells 144 To investigate the effect of the drug combination on the proliferation of myeloid leukemia cells, we used 145 the anti-leukemia drugs IDA, HHT, and CHI in combination with TIG to treat SKM-1. The IC50 (50% of 146 8 maximal inhibitory concentration) of SKM-1 at 72 h was calculated to determine the appropriate concentration 147 of each drug (Table S1). In the experiment, SKM-1 was exposed to 72 h in TIG, HHT, CHI, and IDA. In all 148 combination groups of SKM-1, the effect of the drug combination on the cells was evaluated at 24, 48, and 149 72 h. Next, we used the median-effect method to calculate the combination index of each group to evaluate 150 the combined effect of each group using CalcuSyn software [11]. When TIG was combined with CHI and IDA, 151 most dose combinations had CI (Combination Index) values above 1 at 24, 48, and 72 h after the drug 152 combination. Although the duration of action was prolonged, they did not show a better combined effect. 153 When TIG and HHT were administered at the same time, the CI value of almost all metering combinations 154 was less than 1, and the CI value of each metering combination of 72 h was less than 24 h and 48 h ( Figure  155 1). This indicates that the combination of TIG-HHT shows a strong anti-leukemia response. The synergistic 156 and cumulative effects were evaluated over time. At least three independent experiments were performed with 157 three replicates in each group.

158
To confirm the synergistic effect of the TIG-HHT combination on different myeloid leukemia cells, we 159 measured the IC50 of TIG and HHT for 72 h in MOLM-13, MDS-L, and the primary patient cells (Table S2). To test the effect of the TIG-HHT drug combination on the formation of myeloid leukemia cell clones in 169 vitro, 1  103 SKM-1, MOLM-13, and MDS-L cells were respectively inoculated on a methylcellulose semi-170 solid medium. Clone formation experiments were performed and the number of clones was counted using 171 image-J software ( Figure 3A). The drug environment of myeloid leukemia cell methylcellulose semi-solid 172 medium was: SKM-1 plus TIG 7.5 g/mL was treated with HHT 8 ng/mL; MOLM-13 plus TIG 5 g/mL was 173 treated with 1 ng/mL HHT for 72 h; MDS-L plus 6 mg/mL TIG was treated with 6 ng/mL HHT; and the 174 control group was treated with monotherapy and PBS. The use of TIG or HHT alone had a significant 175 inhibitory effect on the colony forming ability of leukemia cells, and the addition of TIG and HHT at the same 176 time shows a stronger blow to cell colony formation ( Figure 3A and C). Combination treatment compared 177 with single treatment and combination treatment compared to the control groups was statistically significant 178 (P < 0.05 for SKM-1 and MOLM-13; P < 0.01 for MDS-L).

179
TIG-HHT combination effectively restrains tumor growth in subcutaneous myeloid leukemia mouse 180 models 181 To explore the effects of drug combinations in vivo, we performed combination treatments, single 182 treatments, and PBS treatments (control) on a myeloid leukemia mouse model established by a subcutaneous 183 injection of SKM-1 in mice. After seven days, the tumor grew to a size of about 50 mm3. We designed a 184 treatment plan according to the report. Both the TIG and HHT group were injected once a day, and the Control 185 group was injected with a corresponding PBS solution. Compared with the control group, TIG and HHT 186 single-drug treatment both delayed tumor growth, and the combination showed higher efficacy (p < 0.005, 187 Tukey's t-test after a one-way ANOVA) (Figure 4). Immunohistochemistry and a Western blot detection of 188 the corresponding proteins showed that both single and combined drugs inhibited mitochondrial translation in 189 vivo. were treated with different concentrations of the drugs: SKM-1 plus 7.5 mg/mL TIG were treated with 8 197 ng/mL HHT for 72 h; the control group was treated with monotherapy and PBS; and MOLM-13 plus 5 mg/mL 198 TIG was performed with 1 ng/mL HHT for 72 h. As shown in Figure 3B, compared with the single-agent 199 group and the blank group, the apoptotic rate of the cells increased significantly ( Figure 3B). The mechanism 200 of drug-induced apoptosis was studied by a Western blot analysis of the expression of cleaved caspase-3 and 201 cleaved PARP, and the strong pro-apoptotic effect of the drug combination was revealed on a molecular level. 202 The expression of cleaved caspase-3 and cleaved PARP changed, and the protein mass in the two drug lanes 203 was significantly lower than that of the single drug group and the control group ( Figure 3D). To further 204 confirm that treatment with TIG and HHT alone or in combination can also inhibit cell viability by inducing 205 and enhancing apoptosis in vivo, we examined the expression of cleaved caspase-3 and cleaved PARP in 206 mouse subcutaneous tumor tissue cells by immunoblotting ( Figure 5A). Compared with the single drug and 207 the blank groups, the amounts of combined histones was significantly reduced. The immunohistochemistry of 208 the subsequent tissue sections more intuitively confirmed a strong pro-apoptotic effect of the combined group 209 ( Figure 4E).

TIG-HHT combination inhibited mitochondrial translation by down-regulation of COX-1 211
In studies using FDA-approved drugs to screen leukemia cell therapies, the application of yeast whole-212 genome screening technology confirmed that the specific inhibition of mitochondrial translation is the 213 mechanism of TIG's anti-leukemia effect [5]. Thus, we used a Western blot to identify the DNA and proteins 214 11 extracted from myeloid leukemia cells and tumor tissues. myeloid leukemia cells were treated with different 215 concentrations of drugs: SKM-1 plus 7.5 mg/mL TIG was treated with 8 ng/mL HHT for 72 h, and the control 216 group was treated with monotherapy and PBS. MOLM-13 plus 5 mg/mL TIG was treated with 1 ng/mL HHT 217 for 72 h, and MDS-L plus 6 mg/mL TIG was treated with 6 ng/mL HHT for 72 h. Each cell line was treated 218 with monotherapy and PBS was used as a control group. Compared with treatment of TIG or HHT alone, 219 treatment with a combination of TIG-HHT caused a significant down-regulation of the mitochondrial gene, 220 COX-1, expression in myeloid leukemia cells, where the nuclear gene COX-4 did not change significantly 221 ( Figure 5B). Tissue-specific protein testing revealed that the combined drug group down-regulated COX-1 222 gene expression in the tumor cells of subcutaneous myeloid leukemia mice without affecting the expression 223 of the COX-4 gene ( Figure 4E and 5B). SKM-1 was further exposed to a combination of drugs with different 224 concentrations, and the inhibitory effect on the cells showed a stronger mitochondrial translation inhibitory 225 effect as the combination drug concentration increased ( Figure 5C).

TIG-HHT combination synergistically inhibited the mTOR/4EBP1 pathway 227
Western blot was used to identify the proteins extracted from both myeloid leukemia cells and tumor 228 tissues. Cells were treated with different concentrations depending on the concentration of each compound 229 used to inhibit cell proliferation: SKM-1 was exposed to 7.5 g/mL TIG and 8 ng/mL HHT, and MOLM-13 230 was treated with 5 mg/mL TIG and 1 ng/mL HHT. MDS-L was exposed to 6 g/mL TIG and 6 ng/mL HHT. 231 Three groups of myeloid leukemia cells were treated for 72 h, and the single drug or PBS group were used as 232 a control. The mitochondrial translation regulators, mTOR, 4EBP1, and AKT proteins were down-regulated 233 in combined or single drug group compared to the control group. The combined TIG-HHT significantly 234 enhanced the inhibition of protein phosphorylation. We tested the level of phosphorylation of the constituent 235 subunit Raptor, a component of mTOR complex 1 (mTORC1) and the constituent subunit Rictor, a component 236 of mTOR complex 2 (mTORC2). When TIG-HHT co-administration, the p-Rictor and p-Rictor were 237 12 significantly down-regulated, suggesting that the drug combination exerted a stronger inhibition on the 238 phosphorylation of mTOR pathway proteins. The results also showed that the expression of the mitochondrial 239 gene COX-1, and the levels of phosphorylation of AKT and 4EBP1 were reduced to a uniform level when 240 cultured with TIG-HHT combination ( Figure 5B, 5D). COX-1 and the nuclear protein-encoded respiratory chain complex 4 protein subunit COX-4 to reveal the 253 differential effects of the drug on nuclear translation and mitochondrial translation. In this study, we found 254 HHT specifically inhibits mitochondrial translation. Moreover， both the single TIG and combination treatment 255 showed levels of COX-1 protein but not COX-4 down-regulation significantly compared to the control group 256 in each cell line. Further use of the TIG-HHT combination showed a stronger inhibition of mitochondrial 257 translation as the concentration increased in the refractory leukemia cell line SKM-1 ( Figure 5B). In vivo 258 experiments also showed that the level of COX-1 gene expression in the combination group was down-259 13 regulated in the leukemia cells of subcutaneous tumor mice without affecting the expression of the COX-4 260 gene ( Figure 4E and 5B). Both the in vivo and in vitro experiments suggested that the combination showed a 261 stronger inhibition of mitochondrial function, which might represent an important mechanism for the 262 synergistic effects. 263 The in the level of intracellular 4EBP1 phosphorylation was consistent with the expression of the mitochondrial 285 gene COX-1, which suggested that there might be a link between Akt/mTOR/4EBP1 and mitochondrial 286 respiratory function. 287 Conclusion 288 We found that TIG combined with HHT has a synergistic anti-leukemia effect in vitro and significant effect 289 on tumor volume reduction and apoptosis induction in myeloid cells and a xenograft mouse model. The down-290 regulating mitochondrial translation and the AKT/mTOR/4EBP1 signaling pathway might explain this effect. 291 These findings suggest clinical potential in administration of TIG and HHT in the treatment of myeloid 292 leukemia.

293
Ethics approval and consent to participate 294 Animal experiments in this study were performed in accordance with the "Guidelines for the Care and Use 295 of Laboratory Animals" (NIH Publication 86-23, revised in 1985) issued by the National Institutes of Health. 296 The animal experiment protocol was approved by the Animal Protection and Facilities Committee of Zhejiang 297 University. All works were made to minimize the suffering of the animals. 298 Consent for publication 299 All authors of the article are happy to publish the article in the journal. 300 Availability of data and materials 301