The expression of RBKS was increased in AML development

To elucidate the role of RBKS in AML pathogenesis. We analyzed RBKS mRNA levels using GEPIA database and found that RBKS was up-regulated in AML patients (Fig. 1A). Furthermore, AML patients with high levels of RBKS have a shorter survival time (Fig. 1B). Then, we examined the RBKS mRNA and protein levels in bone marrow obtained from normal and AML patients. The results of RT-PCR and western blot indicated that RBKS was up-regulated in AML patients when compared to normal population (Figs. 1C-D). Finally, we collected several AML cell lines (HL60, THP-1, KG-1, KY821 and U937) and human CD34 + cell line served as control group. The results showed that the expression levels of RBKS mRNA and protein were higher in AML cells than that in CD34 + cell (Figs. 1E-F). Among these AML cell lines, the RBKS increase in KG-1 cell line is the most significant, therefore, we chose this cell line for further research. Taken together, these data implied an important role of RBKS in AML and indicated a prospective application of using RBKS as s potential biomarker of AML diagnosis.

Fig. 1

The expression level of RBKS in AML patients and AML cell lines. A: Using the GEPIA database to predict the expression level of RBKS mRNA in AML patients. B: Using the Starbase to predict the survival curve in AML patients with high levels of RBKS. C: RT-PCR detected the expression of RBKS in the bone marrow. D: Western blot detected the expression of RBKS in the bone marrow. E: RT-PCR detected the expression of RBKS in different AML cell lines. F: Western blot detected the expression of RBKS in different AML cell lines. ns: no significance, *P < 0.05, **P < 0.01, ***P < 0.001

RBKS promotes the growth of AML cells in vitro

We next to explore the function of RBKS in AML cells. We knocked down RBKS by shRNA or overexpressed RBKS in KG-1 cells. The knockdown efficiency of RBKS was confirmed at the mRNA and protein levels. We found that RBKS knockdown obviously decreased the mRNA and protein level of RBKS (Figs. 2A-B). At same time, we also observed an increased expression level of RBKS protein after transfecting with OE-RBKS (Fig. 2B). Since the inhibitory effect of sh-RBKS-1 was more obvious, so, we selected sh-RBKS-1 for the next experiments. In summary, these results indicated that we successfully constructed the stable transformation cell line.

Fig. 2

RBKS promotes the growth of AML cells. A: Using RT-PCR to detect the expression of RBKS in AML cells after transfection of sh-RBKS. B: Using western blot to detect the expression of RBKS in AML cells after RBKS overexpression or knockdown. C: CCK-8 detected the cell viability after RBKS overexpression or knockdown. D: Flow cytometry detected the apoptosis after RBKS overexpression or knockdown. E: Using transwell to detect the cell migration after RBKS overexpression or knockdown. F: Using transwell to detect the cell invasion after RBKS overexpression or knockdown. G: Using EDU to detect the cell proliferation after RBKS overexpression or knockdown. Scale bar: 100 μm (E and F); 50 μm (G). *P < 0.05, **P < 0.01, ***P < 0.001

We used CCK-8 assay to detect the cell viability and found that cell viability was significantly increased after overexpressing RBKS, while the loss of RBKS inhibited cell viability (Fig. 2C). Secondly, we used flow cytometry to detect cell apoptosis. The results showed that overexpression of RBKS inhibited cell apoptosis, while RBKS downregulation promoted it (Fig. 2D). Thirdly, we used the Transwell experiment to detect cell migration and cell invasion, respectively. We observed that the up-regulation of RBKS promoted cell migration and invasion, whereas the down-regulation of RBKS suppressed it (Figs. 2E-F). Finally, we used EDU to detect cell proliferation and found that RBKS overexpression promoted cell proliferation, whereas the loss of RBKS inhibited it (Fig. 2G). Collectively, these findings suggested that RBKS promotes the growth of AML cells in vitro.

RBKS promotes the growth of tumor tissue in vivo

Next, we further to explore whether RBKS plays a similar role in vivo. We injected the stable transformed cell lines with RBKS overexpression or knockdown to the NOD/SCID nude mice, and observed the growth of tumor tissue on the 7th, 14th, 21th, and 28th days after injection (Fig. 3A). We observed that the overexpression of RBKS significantly increased tumor weight and volume, whereas RBKS knockdown had opposite effects (Fig. 3B). In addition, the expression level of RBKS was significantly increased after overexpressing RBKS, however, knocking down RBKS decreased RBKS expression (Fig. 3C). Meanwhile, we used TUNEL and Ki67 to detect the apoptosis and proliferation levels of tumor cells, respectively. The results showed that RBKS overexpression obviously promoted cell proliferation whereas inhibited cell apoptosis (Figs. 3D-E). However, the deficiency of RBKS had opposite results (Figs. 3D-E). In summary, these data are consistent with the results of in vitro experiments and demonstrated that RBKS promotes the growth of AML cells both in vitro and in vivo.

Fig. 3

RBKS was involved in the growth of tumor cells in vivo. A: Schematic diagram of tumor formation in nude mice. B: The statistical analysis of tumor volume and weight. C: Western blot detected the expression of RBKS in different tumor tissues. D: Using IHC to detect the apoptosis of tumor cells. E: Ki67 detected the proliferation of tumor cells. Scale bar: 50 μm (D and E). *P < 0.05, **P < 0.01, ***P < 0.001

TKT interacts with RBKS and is up-regulated in AML cells

We further investigated the molecular mechanism of RBKS in AML. We used the STRING website to screen for proteins that interact with RBKS. Among these proteins, we focused on transketolase (TKT), as recent studies have revealed its involvement in AML progression (Fig. 4A). Next, we used Co-IP to detect whether TKT binds to RBKS and observed that TKT associates with RBKS in AML cells (Fig. 4B). Therefore, TKT binds to RBKS in AML cells.

Fig. 4

The expression level of TKT in AML patients and cell lines. A: Using STRING database to predict the interaction proteins with RBKS. B: Co-IP was used to detect the interaction between RBKS and TKT. C: RT-PCR detected the expression of TKT in the bone marrow. D: Western blot detected the expression of TKT in the bone marrow. E: RT-PCR detected the expression of TKT in different AML cell lines. F: Western blot detected the expression of TKT in different AML cell lines. ns: no significance, *P < 0.05, **P < 0.01, ***P < 0.001

We next to detect the expression level of TKT in AML patients and AML cells, respectively. The RT-PCR and western blot results indicated that the expression level of TKT mRNA and protein was significantly increased in AML patients when compared to normal population (Figs. 4C-D). Meanwhile, the increased TKT mRNA and protein expression were also observed in different AML cell lines (Figs. 4E-F). In conclusion, these data indicated that TKT interacts with RBKS and is involved in the progression of AML.

TKT promotes the growth of AML cells in vitro

We next to explore the role of TKT in AML cells. We first constructed AML cells with up-regulation or down-regulation of TKT. The results of RT-PCR shown that sh-TKT-3 had the highest knockdown efficiency of TKT mRNA (Fig. 5A). Thus, we choose sh-TKT-3 for next research. At same time, we also observed that the expression level of TKT protein was increased after overexpressing TKT, whereas knockdown of TKT had opposite results (Fig. 5B). More importantly, we discovered that TKT affects the expression of RBKS, as TKT overexpression enhanced RBKS expression, while the loss of TKT suppressed it (Fig. 5B). In summary, we concluded that TKT was located on the up-stream of RBKS and promoted the expression of RBKS.

Fig. 5

TKT promotes the growth of AML cells through RBKS. A: Using RT-PCR to detect the expression of TKT in AML cells after transfection of sh-TKT. B: Using western blot to detect the expression level of TKT and RBKS after cell transfection. C: CCK-8 detected the cell viability after cell transfection. D: Flow cytometry detected the cell apoptosis after cell transfection. E: Transwell detected the cell migration after cell transfection. F: Transwell detected the cell invasion after cell transfection. G: EDU detected the cell proliferation after cell transfection. Scale bar: 100 μm (E and F), 50 μm (G). ns: no significance, *P < 0.05, **P < 0.01, ***P < 0.001

Next, we continued to detect the function of TKT in AML cells. We used CCK-8 assay to detect the cell viability and found that the cell viability was increased after TKT overexpression, while the loss of TKT inhibited cell viability (Fig. 5C). Secondly, we used flow cytometry to assess cell apoptosis. The results showed that TKT overexpression inhibited cell apoptosis, whereas TKT knockdown promoted it (Figs. 5D). Thirdly, we used the Transwell experiment to detect cell migration and cell invasion, respectively. We observed that the up-regulation of TKT promoted cell migration and invasion, however, the down-regulation of TKT suppressed it (Figs. 5E-F). Finally, we used EDU to detect cell proliferation and found that TKT overexpression promoted cell proliferation, whereas TKT deficiency inhibited it (Figs. 5G. Taken together, these findings proved that TKT promotes the growth of AML cells.

Since RBKS is a down-stream factor of TKT, so, we next sought to determine whether the loss of RBKS can rescue the phenotype induced by TKT overexpression. Interestingly, the absence of RBKS successfully rescued all the effects caused by TKT overexpression (Figs. 5C-G). In conclusion, these data suggested that TKT promotes the growth of AML cells in a RBKS-dependent manner.

The Pentose phosphate pathway promotes the growth of AML cells

As TKT is a critical enzyme involved in PPP, therefore, we next to examine whether PPP-related indicators were changed, such as ATP, NADP+/NADPH, R5P (ribose 5-phosphate), G3P (glyceraldehyde-3-phosphate), GLU (glucose) and α-KG (α-ketoglutaric acid) after TKT overexpression or knockdown. We found that the content of ATP was increased, whereas the content of NADP+/NADPH, R5P, G3P, GLU and α-KG were decreased after overexpressing TKT (Figs. 6A-F). However, knocking down TKT had opposite results (Figs. 6A-F). At same time, the deficiency of RBKS successfully restored the phenotype caused by TKT overexpression (Figs. 6A-F). Therefore, these data indicated that we can activate or inhibit the pentose phosphate pathway by regulating the expression of TKT.

Fig. 6

TKT regulates the pentose phosphate pathway through RBKS. A: The content of ATP in cells after cell transfection. B: The ratio of NADP+/NADPH in cells after cell transfection. C: The content of R5P in cells after cell transfection. D: The content of G3P in cells after cell transfection. E: The content of GLU in cells after cell transfection. F: The content of α-KG in cells after cell transfection. G: Western blot detected the expression level of TKT and RBKS after cell transfection with sh-TKT and addition of R5P. H: CCK-8 detected the cell viability after cell transfection with sh-TKT and addition of R5P. I: Flow cytometry detected the cell apoptosis after cell transfection with sh-TKT and addition of R5P. J: Transwell detected the cell migration after cell transfection with sh-TKT and addition of R5P. K: Transwell detected the cell invasion after cell transfection with sh-TKT and addition of R5P. L: EDU detected the cell proliferation after cell transfection with sh-TKT and addition of R5P. M: The content of ATP in cells after cell transfection with sh-TKT and addition of R5P. N: The ration of NADP+/NADPH in cells after after cell transfection with sh-TKT and addition of R5P. O: The content of R5P in cells after cell transfection with sh-TKT and addition of R5P. P: The content of G3P in cells after cell transfection with sh-TKT and addition of R5P. Q: The content of GLU in cells after cell transfection with sh-TKT and addition of R5P. R: The content of α-KG in cells after cell transfection with sh-TKT and addition of R5P. Scale bar: 100 μm (J and K), 50 μm (L). *P < 0.05, **P < 0.01, ***P < 0.001

We next continued to explore whether the pentose phosphate pathway is important for AML cell growth. We knocked down TKT to inhibit PPP or simultaneously added R5P to activate PPP. We found that the loss of TKT decreased the RBKS expression which is consistent with above results (Fig. 6G). At same time, TKT expression was not affected by adding R5P, whereas the expression of RBKS was obviously increased (Fig. 6G). In addition, we found that the addition of R5P successfully restored the phenotype caused by TKT loss, including enhanced cell viability (Fig. 6H), inhibition of cell apoptosis (Fig. 6I), promotion of cell migration and invasion (Figs. 6J-K), increased cell proliferation (Fig. 6L). More importantly, the related indicators of PPP were also rescued, including ATP levels (Fig. 6M), NADP+/NDAPH ratios (Fig. 6N), R5P levels (Fig. 6O), G3P levels (Fig. 6P), GLU levels (Fig. 6Q), α-KG levels (Fig. 6R). In summary, these data highlight the critical role of the pentose phosphate pathway in the growth of AML cells.

TKT promotes EMT program in AML cells

As epithelial mesenchymal transition (EMT) plays an increasing role in AML progression. During EMT, AML cells loss the epithelial phenotype and transform to mesenchymal properties. E-cadherin was decreased while Snail and Vimentin were increased during EMT, and then cells lose their adhesiveness and subsequently migrate to a new location. Interestingly, we found that the expression of Snail and Vimentin were elevated, while E-cadherin was decreased after RBKS overexpression (Fig. 7A). However, the loss of RBKS had opposite effects (Fig. 7A). Next, we used the tumor tissue to detect the expression of Vimentin. We observed that Vimentin expression was also increased after overexpressing RBKS, whereas the loss of RBKS had opposite results (Fig. 7B). We concluded that RBKS promotes EMT in AML both in vitro and in vivo.

Fig. 7

TKT promotes the EMT in AML cells through RBKS. A: Western blot detected the expression level of EMT-related proteins after cell transfection. B: IHC was used to measure the expression level of Vimentin in tumor cells. C: Western blot detected the expression level of EMT-related proteins after cell transfection. D: Western blot detected the expression level of EMT-related proteins after cell transfection with sh-TKT and addition of R5P. Scale bar: 50 μm (B). *P < 0.05, **P < 0.01, ***P < 0.001

We next further to explore whether TKT and PPP were involved in EMT. We found that TKT overexpression promoted the EMT process, indicated by elevated Snail and Vimentin expression and decreased E-cadherin expression (Fig. 7C). However, the loss of TKT inhibited EMT (Fig. 7C). Importantly, the loss of RBKS successfully suppressed the EMT that caused by TKT overexpression (Fig. 7C). Consistently, the addition of R5P also obviously enhanced EMT that induced by TKT deficiency (Fig. 7D). Taken together, our study revealed that the pentose phosphate pathway is implicated in the EMT of AML cells.