Crucial factors of the inflammatory microenvironment (IL-1β/TNF-α/TIMP-1) promote the maintenance of the malignant hemopoietic clone of myelofibrosis: an in vitro study

Abstract

Along with molecular abnormalities (mutations in JAK2, Calreticulin (CALR) and MPL genes), chronic inflammation is the major hallmark of Myelofibrosis (MF). Here, we investigated the in vitro effects of crucial factors of the inflammatory microenvironment (Interleukin (IL)-1β, Tumor Necrosis Factor (TNF)-α, Tissue Inhibitor of Metalloproteinases (TIMP)-1 and ATP) on the functional behaviour of MF-derived circulating CD34+ cells.We found that, regardless mutation status, IL-1β or TNF-α increases the survival of MF-derived CD34+ cells. In addition, along with stimulation of cell cycle progression to the S-phase, IL-1β or TNF-α ± TIMP-1 significantly stimulate(s) the in vitro clonogenic ability of CD34+ cells from JAK2V617 mutated patients. Whereas in the JAK2V617F mutated group, the addition of IL-1β or TNF-α + TIMP-1 decreased the erythroid compartment of the CALR mutated patients. Megakaryocyte progenitors were stimulated by IL-1β (JAK2V617F mutated patients only) and inhibited by TNF-α. IL-1β + TNF-α + C-X-C motif chemokine 12 (CXCL12) ± TIMP-1 highly stimulates the in vitro migration of MF-derived CD34+ cells. Interestingly, after migration toward IL-1β + TNF-α + CXCL12 ± TIMP-1, CD34+ cells from JAK2V617F mutated patients show increased clonogenic ability.Here we demonstrate that the interplay of these inflammatory factors promotes and selects the circulating MF-derived CD34+ cells with higher proliferative activity, clonogenic potential and migration ability. Targeting these micro-environmental interactions may be a clinically relevant approach.

Keywords: circulating CD34+ cells; inflammatory microenvironment; migration; myelofibrosis; survival.

Figure 1. Regardless mutation status, the plasma levels of IL-1β, TNF-α and TIMP-1 are increased in MF patients

IL-1β (A), TNF-α (B) and TIMP-1 (C) plasma levels were measured by ELISA in MF patients. (n = 26; JAK2^V617F positive n =16; CALR positive n = 10) and healthy controls (n = 15). Compared with controls, cytokines plasma levels were significantly increased in MF patients. Of note, there was no significant difference between JAK2^V617F or CALR mutated patients. All data are presented as mean ± SEM (**p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001).

Figure 2. Selected subsets of circulating HSPCs are expanded in MF patients

The circulating absolute number of MF (total (n = 30) and subdivided into JAK2^V617F (n = 20) or CALR (n = 10) mutated groups) and CB (n = 10) CD34⁺ cells coexpressing the CD133, CD49d, CD47, CD44, CD63, CD184 and CD41 antigens together with the CD34⁺ CD38⁻ subset are shown (A–I). All subsets were increased in MF patients as compared with the CB counterparts. No significant differences were observed between the two mutated groups, except the CD34⁺ CD63⁺ and the CD34⁺ CD41⁺ cells of CALR mutated patients which were significantly increased as compared with the JAK2^V617F counterparts. All data are presented as mean ± SEM (*p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001).

Figure 3. Survival of CD34⁺ cells from MF patients is increased by IL-1β and TNF-α

(A) CD34⁺ cells from MF patients (n = 20) or CB (n = 8) were in vitro treated for 4 days with factors alone and the percentage of cell viability was assessed after Annexin V/PI staining, as described in Methods. At variance with CB-derived cells, TNF-α and IL-1β alone significantly stimulated the survival of MF-derived CD34⁺ cells as compared with untreated cells and the CB-derived counterparts. Conversely, ATP and TIMP-1 were ineffective in normal and diseased cells.(B) In selected experiments, before Annexin V/PI staining, MF- (n = 10) and CB- (n = 6) derived CD34⁺ cells were also labeled with a MoAb against the human CD38 antigen and the CD34⁺ CD38⁻ cells were gated and cell viability was analyzed. Once again, multiple combinations of cytokines with IL-1β or TNF-α significantly stimulated the survival of MF- and CB-derived CD34⁺ CD38⁻ cells. Notably, this was not true for ATP+ TNF-α+ TIMP-1. No differences were observed between MF patients and CB. All data are presented as mean ± SEM. (**p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001 vs untreated cells (CTR)) (^#p ≤ 0.05; ^##p ≤ 0.01 vs CB).

Figure 4. Proliferation of circulating MF-derived CD34⁺ cells is positively enhanced by IL-1β + TNF-α ± TIMP-1 combinations

Circulating CD34⁺ cells were isolated from MF patients (n = 20) and CB units (n = 8) and cultured in the presence of the selected two-by-two pro-inflammatory factors. After 14 days, the total CFU-C output was assessed as described in Methods (A). Circulating CD34⁺ cells were isolated from MF patients (n = 10) and CB units (n = 8) and cultured in the presence or absence of inflammatory factors alone or combined. After 12 days, the CFU-MK growth was assessed as described in Methods (B). The results are expressed as growth fold change versus untreated CTR samples. (A) The clonogenic output of the MF-derived CD34⁺ cells was significantly stimulated by the IL-1β + TIMP-1 combination as compared with untreated cells or the CB-derived counterparts. No other combinations of factors two-by-two were effective. The mean number of colonies in MF-derived and CB-derived untreated samples was 59 ± 8 and 63 ± 6, respectively. (B) The MF-derived CFU-MK growth was significantly inhibited by TNF-α. By contrast, IL-1β has stimulatory activity on MK colony formation. Factors in combination did not significantly modify the growth of patients/CB CFU-MK as compared with factors alone. Factors alone or in combination did not significantly modify the CFU-MK growth of the CB counterparts. The mean number of CFU-MK in MF- and CB-derived untreated samples was 26 ± 11 and 46 ± 10, respectively. In (C and D ) are shown the results of cell-cycle analysis of MF-derived (n = 10) and CB-derived (n = 8) CD34⁺ cells after in vitro incubation for 24 hours in the presence or absence of various combinations of pro-inflammatory factors. Results are expressed as the percentage of cells in different phases of the cell cycle. IL-1β plus TNF-α highly promote cell cycling of CD34⁺ cells from MF patients. IL-1β + TNF-α + TIMP-1 and IL-1β + TNF-α + TIMP-1 + ATP were also effective (C). Conversely, no significant differences were observed when CB-derived cells were analysed (D). All data are presented as mean ± SEM. (*p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001 vs untreated cells) (^#p ≤ 0.05; ^##p ≤ 0.01 vs CB).

Figure 5. Opposite effects of pro-inflammatory cytokines on cells from JAK2^V617F or CALR mutated patients

(A) When clonogenic activity was analyzed according to mutation status, the CFU-C growth of JAK2^V617F mutated patients was significantly up-regulated by IL-1β + TIMP-1 and IL-1β + TNF-α as compared with untreated control samples, the CALR mutated counterparts and the CB-derived cells (only IL-1β + TIMP-1). The results are expressed as growth fold change versus untreated CTR samples. All data are presented as mean ± SEM. (*p ≤ 0.05; **p ≤ 0.01 vs untreated cells) (^#p ≤ 0.05; ^##p ≤ 0.01; ^####p ≤ 0.0001 vs CB-derived cells) (B) When colony composition was analyzed according to mutation status, we found that the erythroid compartment of the untreated samples was reduced in both mutated groups as compared with the CB counterparts. However, no significant differences were observed between the two mutated groups. Interestingly, in the JAK2^V617F mutated group, the addition of IL-1β + TIMP-1 and IL-1β + TNF-α enhanced the erythroid compartment as compared with untreated samples. Conversely, some cytokines combinations significantly impaired BFU-E growth in CALR mutated patients. The results are expressed as mean percentage of CFU-GM/BFU-E as compared with the total CFU-C count.

Figure 6. IL-1β and TNF-α significantly increases migration of MF-derived CD34⁺ cells

(A) CXCL12 plasma levels of MF patients (total n =24; JAK2^V617F mutated patients n = 16; CALR mutated patients n = 8) and controls (n = 10). Regardless mutation status, CXCL12 concentration was significantly higher in MF patients (*p ≤ 0.05 vs controls). (B) When cells were migrated toward CXCL12 alone, an increased migration ability was observed in MF-derived (n = 15) CD34⁺ cells as compared with the CB-derived (n = 8) counterparts. The addition of inflammatory factors alone (IL-1β/TNF-α) plus CXCL12 significantly increased the migratory behaviour of MF-derived CD34⁺ cells as compared with CXCL12 alone. IL-1β + TNF-α synergistically enhanced the migratory behaviour of CD34⁺ cells as compared with spontaneous migration (****p < 0.0001), CXCL12 alone (**p <0.001) and the CB-counterpart (^####p <0.0001). Results are expressed as mean percentages ± SEM of input. (**p ≤ 0.01; ***p ≤ 0.001 vs CXCL12 alone for CB-derived CD34⁺ cells) (*p ≤ 0.05; **p ≤ 0.01; ****p < 0.0001 vs spontaneous migration for MF-derived CD34⁺ cells) (^#p ≤ 0.05; ^####p <0.0001 vs CB).

Figure 7. The clonogenic output of MF-derived CD34⁺ cells after migration toward IL-1β + TNF-α + CXCL12 ± TIMP-1 is potently enhanced

Panels (A and B) show the clonogenic potential of CB-derived (A; n = 6) and MF-derived CD34⁺ cells (B; n = 14) at baseline with or without various combinations of pro-inflammatory factors (CFU-C) and after migration toward CXCL12 alone or various combinations of pro-inflammatory factors + CXCL12 (CFU-C post-migration). After migration toward IL-1β + TNF-α + CXCL12 ± TIMP-1, the MF-derived, but not CB-derived, CD34⁺ cells show significantly increased clonogenic potential. Results are expressed as mean fold change of CFU-C ± SEM. (*p ≤ 0.05 vs untreated cells (A) and CXCL12 alone (B)).