Vacuolin-1 enhances RA-induced differentiation of human myeloblastic leukemia cells: evidence for involvement of a CD11b/FAK/LYN/SLP-76 axis subject to endosomal regulation that drives late differentiation steps

Vacuolin-1 does not modulate RA-induced CD38 expression

To determine if vacuolin-1 could enhance RA-induced differentiation, the dose-dependent effect of vacuolin-1 on cell proliferation was determined to identify a dose that was non-toxic to use in combination with RA. We screened the optimal concentration of vacuolin-1 that was not toxic to cells using concentrations of vacuolin-1, ranging from 0.25 to 5 µM to treat HL-60 cells for 48 h. Cell population growth was measured by counting cell culture density and DNA cell cycle distribution by PI staining and flow cytometric analysis. The latter also revealed sub-G1 DNA cells betraying nuclear DNA fragmentation associated with apoptosis/death. We found that 0.25 µM did not affect cell population growth (Fig. 1A). It was well below the dose at which vacuolin-1 caused severe cell cycle disruptions, which was apparent at 5.0 µM (Additional file 1: Fig. S1). 5.0 µM caused a dramatic sub-G1 accumulation, suggesting that this concentration of vacuolin-1 induced apoptosis and was toxic. The effect was incipient at 2.5 µM. 0.25 µM vacuolin-1 was thus selected for further study with RA .

Fig. 1figure 1

Vacuolin-1 does not modulate RA-induced CD38 expression. A The cell density of HL-60 cell cultures that were untreated controls or treated with 1 µM RA, 0.25 μm Vacuolin-1 or 1 µM RA plus 0.25 μm Vacuolin-1 as indicated was measured at 24 h, 48 and 72 h. B–E HL-60 cells were cultured for 5 h (B), 24 h (C), 48 h (D) and 72 h (E) with 1 µM RA and 0.25 µM Vacuolin-1 as indicated and membrane CD38 was analyzed using flow cytometry

The effect of vacuolin-1 on RA-induced CD38 was determined. CD38 expression is the earliest telltale that the process of RA-induced differentiation has begun [38]. Its ectopic expression generates MAPK pathway signaling and accelerates induced differentiation. To minimize any overt toxic effects by vacuolin-1, 0.25 µM vacuolin-1 was used to ascertain whether it can promote RA-induced HL-60 cell differentiation. Cells were either untreated, treated with RA, vacuolin-1 or vacuolin-1 plus RA, and expression of CD38 was measured by flow cytometry. RA induced CD38 expression and adding 0.25 µM vacuolin-1 to the 1 µM RA did not further increase the percentage of CD38 positive cells measured at 5 h, 24 h, 48 and 72 h (Fig. 1B–E). These results indicate that vacuolin-1 did not affect early steps of RA-induced differentiation marked by CD38 expression, which is consistent with the known prominent RARE locate in the first intron that drives CD38 expression [39, 40].

Vacuolin-1 enhanced RA-induced expression of CD11b

The effect of vacuolin-1 on RA-induced CD11b was determined. The expression of CD11b is a mid-stage differentiation marker induced by RA after CD38 and before manifestation of inducible oxidative metabolism or cell cycle arrest characteristic of mature cells [41]. CD11b is encoded by ITGAM and dimerizes with CD18 to form the functional integrin heterodimer, CD11b/CD18, also known as alpha-M beta-2 (αMβ2) [42]. In macrophages, CD11b is required for the activation of SRC and SYK kinases [43]. Cells were untreated, treated with RA, vacuolin-1 or RA plus vacuolin-1, and CD11b expression was measured by flow cytometry after 5 h, 24 h, 48 and 72 h of culture. Co-treatment with vacuolin-1 and RA significantly increased the percentage of CD11b positive cells compared to RA alone that was detectable at 24 h, 48 and 72 h (Fig. 2). Vacuolin-1 thus enhanced expression of this mid-stage marker in the process of RA-induced differentiation. Anecdotally, we noted, too, that in another human myeloid leukemia cell line, NB4, the only other well known established one, vacuolin-1 also enhanced RA-induced CD11b expression (Additional file 1: Fig. S2).

Fig. 2figure 2

Vacuolin-1 enhances RA-induced expression of CD11b. AD HL-60 cells were cultured for 5 h (A), 24 h (B), 48 h (C) and 72 h (D) with 1 µM RA and 0.25 µM Vacuolin-1 as indicated and membrane CD11b was analyzed using flow cytometry. Control is untreated

Vacuolin-1 enhanced RA-induced markers of late steps of the differentiation into mature cells: inducible oxidative metabolism and G1/0 cell cycle arrest

The effect of vacuolin-1 on RA-induced functional differentiation, which characterizes late steps of induced differentiation, was determined. Functional differentiation was detected by inducible oxidative metabolism betrayed by TPA-induced production of reactive oxygen species (ROS). ROS is a functional differentiation marker of mature differentiated cells. Cells were either untreated, treated with RA, vacuolin-1 or vacuolin-1 plus RA. Vacuolin-1 in combination with RA enhanced the percentage of cells positive for TPA-induced reactive oxygen species (ROS) production compared with RA alone (Fig. 3A). Vacuolin-1 thus enhanced RA-induced expression of the functional marker of mature cells.

Fig. 3figure 3

Vacuolin-1 enhances RA-induced markers of late steps of the differentiation into mature cells: inducible oxidative metabolism and G1/0 cell cycle arrest. A HL-60 cells were cultured for 72 h with 1 µM RA and 0.25 µM Vacuolin-1 as indicated, where Control is untreated, and the inducible ROS was analyzed using flow cytometry. DMSO is the carrier control for TPA stimulation. B The cell density of HL-60 cell cultures that were untreated controls or treated with 1 µM RA or RA plus different (0.1, 0.25, 0.5 and 1.0 µM) concentrations of Vacuolin-1 as indicated was measured at 24 h, 48 and 72 h. C, D HL-60 cells were cultured for 48 h (C) and 72 h (D) with 1 µM RA and 0.25 µM Vacuolin-1 as indicated and the cell cycle phase distributions were analyzed using flow cytometry. Control is untreated

The effect of vacuolin-1 on RA-induced growth inhibition via G1/0 cell cycle arrest was determined. Growth inhibition characterizes mature differentiated cells. Cells were either untreated, treated with RA, vacuolin-1 or vacuolin-1 plus RA. Vacuolin-1 enhanced RA-induced inhibition of population growth (Fig. 3B ). Enhanced growth inhibition due to vacuolin-1 was apparent at 72 h in mature cell populations. The effect of vacuolin-1 on RA-induced G1/0 cell cycle arrest corroborated this. G1/0 cell cycle arrest betrayed by the accumulation of G1/0 DNA cells is characteristic of mature differentiated cells. Cells were either untreated, treated with RA, vacuolin-1 or vacuolin-1 plus RA; and their cell cycle distribution was measured by propidium iodide staining and flow cytometry after 48 and 72 h of culture. Vacuolin-1 enhanced the RA-induced enrichment of G1/0 cells, consistent with retarding cell population growth (Fig. 3C, D). Vacuolin-1 thus enhanced RA-induced markers characteristic of mid to late steps of differentiation into mature cells, namely CD11b and subsequently inducible ROS and growth arrest associated with G1/0 accumulation. The results suggest that expression of CD11b drives later events in the process of RA-induced differentiation and make its signaling attributes in this context of interest.

Vacuolin-1 modulates a CD11b/FAK/LYN/SLP-76 signaling axis to promote RA-induced HL-60 cell differentiation

Adding vacuolin-1 to RA treatment enhanced RA-induced upregulation of CD11b, with ensuing enhancement of subsequent steps of differentiation. To determine if enhanced CD11b expression attributable to Vacuolin-1 had signaling consequences potentially relevant to driving differentiation, activation of signaling molecules downstream of CD11b that are known drivers for differentiation was measured. CD11b is a member of the integrin family of receptors known to act through FAK and activate MAPK related signaling pathway components. Lyn is a member of the family of Src-Like Family Kinases (SFK) which are known to associate with FAK [44]. Lyn is a component of the signalosome known to be activated by RA. It associates with other signalosome components including the adaptor, SLP-76. Expression of SFKs and adaptors such as SLP-76 or CBL, are known to propel differentiation [17, 18, 38, 45, 46]. This motivates the speculation that activation of a CD11b/FAK/LYN/SLP-76 signaling axis is involved. Accordingly we tested if vacuolin-1 contributed to enhancing RA-induced activation of these signaling molecules. Cells were untreated or treated with RA, vacuolin-1, or vacuolin-1 plus RA for 48 h. Immunoprecipitates using the signaling molecule as bait were resolved by Western blotting and probed for p-tyr.

RA caused the tyrosine phosphorylation of FAK, and this was enhanced by addition of vacuolin-1 to the RA treatment. Likewise LYN tyrosine phosphorylation was also enhanced. The phosphorylation level of Y416 SFKs, which is the telltale phosphorylation event for SFK activation, was measured, and vacuolin-1 enhanced RA-induced LYN activation. The primary phosphorylated SFK in HL-60 cells is Lyn [47]. SLP-76, an adaptor molecule in the signalosome that is associated with LYN, also became tyrosine phosphorylated after RA treatment and addition of vacuolin-1 enhanced this (Fig. 4A–D). LYN and SLP-76 are constituents of the NUMB scaffolded signalosome that RA activates to drive differentiation. NUMB is a membrane protein that is known to be tyrosine phosphorylated in response to RA [15]. By contrast RA-induced tyrosine phosphorylation of CBL, another putative adaptor in the signalosome which is also associated with integrin signaling [48], was apparently not affected by addition of vacuolin-1. Nor was activation of the RAF/MEK/ERK axis imbedded in the signalosome (Fig. 4E F), indicating that the vacuolin-1 effects appear somewhat specific and were not catholic for all signalosome adaptors or components. Western blots (Fig. 4A) showed that the expression levels of FAK, LYN, and SLP-76 remained similar when comparing the vacuolin-1-RA co-treatment group with the RA-only group, while c-CBL expression was enhanced after vacuolin-1 cotreatment with RA. The data are ergo consistent with the proposition that CD11b drives a novel CD11b/FAK/LYN/SLP-76 signaling axis subject to endocytic regulation that propels RA-induced HL-60 cell myeloid differentiation.

Fig. 4figure 4

Vacuolin-1 modulates a CD11b/FAK/LYN/SLP-76 signaling axis to promote RA-induced HL-60 cell differentiation. A HL-60 cells were cultured for 48 h with 1 µM RA and 0.25 µM Vacuolin-1 as indicated (Control is untreated), the cell lysates were collected, and the indicated proteins were analyzed by western blot for molecules indicated, Y416 phosphorylated SFKs, FAK, LYN, c-CBL, SLP-76. GAPDH is the loading control. BE HL-60 cells were cultured for 48 h with 1 µM RA and 0.25 µM Vacuolin-1 as indicated, the cell lysates were collected, and FAK immunoprecipitates (IP) (B), Lyn IP (C), Slp-76 IP (D) and c-cbl IP (E) were analyzed by Western blots probed with anti-p-tyr antibody. F HL-60 cells were cultured for 48 h with 1 µM RA and 0.25 µM Vacuolin-1 as indicated, the cell lysates were collected, and the indicated proteins were analyzed by western blot, probing for the indicated molecules. G, H HL-60 cells were untreated controls or 1 µM RA-treated G or 0.25 µM Vacuolin-1 or 0.25 µM Vacuolin-1 plus 1 µM RA-treated H for 48 h as indicated, the cells were collected and stained for FAK for confocal microscopy. DAPI was used to stain nuclear DNA and delineate the nucleus

These results motivate interest in whether vacuolin-1 might affect the intracellular relocalization of FAK. Confocal microscopy was used observe the intracellular distribution of FAK. Fixed cells were immunofluorescently stained for FAK and DAPI stained for DNA. In control and RA-treated cells FAK is relatively unformly diffusely distributed in the cell (Fig. 4G). Vacuolin-1-treated cells by contrast showed an apparent redistribution of FAK toward the periphery of the cell consistent with an association with the CD11 cell surface receptor (Fig. 4H).

Deletion of CD11b or Numb attenuates RA-induced HL-60 cell differentiation

The present results on the effects of vacuolin-1 on RA-induced differentiation suggest that the expression of CD11b induced by RA drives subsequent late steps in the process of differentiation. To test this proposition CD11b expression was crippled by CRISPR KO intervention. Stable transfectants were created using CD11b directed and non-specific control guides using pooled transfectants to obviate clonal bias. Untreated or RA-treated control and CD11b CRISPR/cas9 targeted transfectants were analyzed for CD11b expression after 72 h of culture. Western blotting showed that RA induced CD11b expression in the non-specific guide control typical of wt parental cells and that addition of vacuolin-1 enhanced this, but RA-induced expression in the CD11b targeted transfectants was grossly crippled and still greatly crippled with addition of vacuolin-1 (Fig. 5A). Flow cytometric analysis of immunofluorescently labelled CD11b expression corroborated that the expression in RA-treated CD11b targeted cells was greatly crippled, but not in the non-specific control (Fig. 5B). Both assays detected a very small amount of CD11b expression in RA-treated CD11b targeted cells that is unexplained, but may reflect a small subpopulation of cells that survived despite prolonged selection; however, both assays also demonstrate that the population of CD11b targeted cells was largely crippled in their ability to express CD11b. RA-induced CD38 expression was unaffected by loss of CD11b. Both the control and CD11b targeted transfectants expressed CD38 with 100% of the cells positive at 72 h (Fig. 5C). Expression was measured by flow cytometry of immunofluorescenly stained cells. In contrast, RA-treated CD11b targeted cells failed to express the functional differentiation marker, inducible oxidative metabolism, detected by ROS production, whereas the control cells did express this at a level typical of wt parental cells (Fig. 5D). The inducible oxidative metabolism was measured by flow cytometry to detect ROS production of TPA-stimulated cells. Like inducible ROS production, another marker of late steps in RA-induced differentiation marking mature cells, G1/0 cell cycle arrest was also crippled by loss of CD11b. While control cells showed RA-induced enrichment of G1/0 cells characterizing G1/0 cell cycle arrest, the CD11b targeted cells failed to do so (Fig. 5E ). We note anecdotally, too, that absent CD11b, vacuolin fails to enhanced RA-induced phosphoY416 marked LYN activation, consistent with the above posted CD11b driven LYN activation (Additional file 1: Fig. S3). .Loss of CD11b thus resulted in cells unable to undergo late steps of RA-induced differentiation, namely inducible oxidative metabolism and G1/0 cell cycle arrest, but were competent for early steps, namely CD38 expression.

Fig. 5figure 5

Deletion of CD11b or Numb attenuates RA-induced HL-60 cell differentiation. A The sg-NC (non-specific guide control), sg-CD11b (CD11b targeted CRISPR KO) and sg-Numb (Numb targeted CRISPR KO) stably transfected HL-60 cells were collected for WB analysis of CD11b and Numb. B The sg-NC, sg-CD11b and sg-Numb HL-60 cells were cultured for 72 h with or without (Control) 1 µM RA as indicated and membrane CD11b was analyzed using flow cytometry. C–E The sg-NC, sg-CD11b and sg-Numb HL-60 cells were cultured for 72 h with or without 1 µM RA as indicated, and membrane CD38 (C), inducible ROS induced by TPA D and G1/G0 cell cycle arrest E were analyzed as indicated by flow cytometry. DMSO is the carrier control (the negative) for TPA (the positive) used to induce oxidative metabolism (Reactive Oxygen Species, ROS)

The present results suggest that NUMB has a potentially important role in RA-induced differentiation, as a regulator of the signalosome that generates signals to enable RA-driven transcription and of endocytosis that regulates the signaling. To further study its involvement in RA-induced HL-60 cell differentiation, stable transfectants with depleted Numb were created with CRISPR/Cas 9. Western blotting showed that endogenous NUMB expression was ablated (Fig. 5A). To determine the effect of losing NUMB on RA-induced differentiation, CD38, CD11b, ROS and G1/0 cell cycle arrest were measured. Loss of Numb did not affect RA-induced expression of CD38 (Fig. 5C). Nor did it cripple induced expression of CD11b, which was actually somewhat enhanced as might be expected if CD11b endocytosis were impaired (Fig. 5B). However, RA-induced inducible ROS production (Fig. 5D) and G1/G0 cell cycle arrest (Fig. 5E) were grossly crippled by loss of NUMB. Hence, these data implicate an essential role for Numb in regulating RA-induced HL-60 cell differentiation. However, that role appears complicated because of potentially multiple functions that NUMB performs.

留言 (0)

沒有登入
gif