Constitutive expression of JAK2V617F/Vav-Cre in Trp53-inactivated cells results in MPN-like phenotype

The JAK2V617F conditional allele mice were crossed with Vav-Cre transgenic mice to generate JAK2V617F/Vav-Cre mice. To assess the function of Trp53 inactivation in the background of JAK2V617F, the JAK2V617F/Vav-Cre/Trp53−/− mice were bred as described in the mouse model section.

Blood counts demonstrated a similar phenotype in JAK2V617F/Vav-Cre mice and JAK2V617F/Vav-Cre/Trp53−/− mice. WBC counts for JAK2V617F/Vav-Cre and JAK2V617F/Vav-Cre/Trp53−/− mice (43.12 ± 12.10 G/L and 48.30 ± 12.83 G/L, respectively) were significantly higher (4-fold) than those of control mice (9.94 ± 1.18 G/L) 1 month after birth and continued to rise during the second (42.57 ± 16.59 G/L, p = 0.001 and 54.78 ± 15.91 G/L, p = 0.0004 for JAK2V617F/Vav-Cre and JAK2V617F/Vav-Cre/Trp53−/− mice, respectively) and third months (52.12 ± 6.46 and 66.73 ± 27.07 G/L for JAK2V617F/Vav-Cre and JAK2V617F/Vav-Cre/Trp53−/− mice, respectively) (Fig. 1A). Hematocrit in JAK2V617F/Vav-Cre/Trp53−/− and JAK2V617F/Vav-Cre mice was higher (61.89% ± 9.70% and 67.87% ± 8.19%, respectively) at 1 month after birth compared to 56.12% ± 3.07% in control mice (p = 0.006 and p = 0.001, respectively) (Fig. 1B). Platelet counts were also higher than normal at 1-month in JAK2V617F/Vav-Cre and JAK2V617F/Vav-Cre/Trp53−/− mice. Platelet count for JAK2V617F/Vav-Cre mice continued to rise from 1571 ± 491 G/L to 3497 ± 948 G/L, and JAK2V617F/Vav-Cre/Trp53−/− mice displayed the same trend of increasing platelet count from 1 to 3 months (Fig. 1C). Spleen weights were enlarged in the same order of magnitude (5-fold at 3 months) in both JAK2V617F/Vav-Cre and JAK2V617F/Vav-Cre/Trp53−/− mice and increased gradually during the time in both backgrounds (Fig. 1D, E) compared to control mice (p = 0.6519 × 10−8 and p = 0.324 × 10−4 at 3 months, respectively).

Fig. 1: The inactivation of Trp53 does not modify the JAK2V617F-induced MPN phenotype.

A–C Data of cell count of peripheral blood (PB) from C57Bl/6, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre/Trp53−/− mice, graph represents pooled mice from three independent experiment, each with 3-4 mice per group. (A: white blood cell, B: hematocrit, C: platelet). D The spleen of C57Bl/6, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre /Trp53−/− mice. The graph represents from three independent experiments, each with 3-4 mice per group. E The spleen weight from C57Bl/6, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre /Trp53−/− mice at 1 to 3 months old. Graph represents pooled mice from three independent experiment, each with 3-4 mice per genotype. F Percent survival of JAK2V617F/Vav-Cre and JAK2V617F/Vav-Cre /Trp53−/− mice. G Histopathologic characterization of the bone marrow (BM). The first column is the BM of C57Bl/6 mice, the second column is the BM from 1-month-old JAK2V617F/Vav-Cre mice, the third column is BM from 3-month-old JAK2V617F mice, the fourth column is BM from 1-month-old JAK2V617F/Trp53−/− mice, and the fifth column is BM from 3-month-old JAK2V617F/Trp53−/− mice. The first row to third row respectively the 100×, 200×, and 400× objectives. The samples on the fourth row were silver stained, and the samples on the other rows were hematoxylin and eosin stained. Images were obtained using a microscope with an Olympus camera.

BM and spleen histological analyses were performed at 1 and 3 months. BM histological analysis, as previously reported [23], illustrated that the megakaryocytic density of JAK2V617F/Vav-Cre mice was higher than that of normal mice (Fig. 1G). No difference was noticed between JAK2V617F/Vav-Cre/Trp53−/− and JAK2V617F/Vav-Cre mice. Because mice were relatively young, reticulin fiber deposits were not prominent in JAK2V617F/Vav-Cre or JAK2V617F/Vav-Cre/Trp53−/− mice at 1 or 3 months (Fig. 1G).

Histological spleen analysis revealed that JAK2V617F/Vav-Cre mice lost their classical histology: removal of white pulp, significant expansion of red pulp, and increased number of megakaryocytes. Increased number and size of clusters of immature red lineage precursors were noticed in the 2 types of mice (data not shown).

No evidence of blast infiltration was noticed at 1 and 3 months old in blood, BM, or spleen samples in both JAK2V617F mice and JAK2V617F/Vav-Cre/Trp53−/− mice (Fig. S1).

Lastly, survival analysis showed that the median survival time for JAK2V617F/Vav-Cre mice was around 2 months when no treatment (or venesection) was performed; Trp53 inactivation did not shorten JAK2V617F/Vav-Cre mice’s (already short) lifespan (Fig. 1F).

Collectively, these data show that heterozygous endogenous JAK2V617F expression in hematopoietic cells leads to hyperplasia of mature and maturing erythroid, granulocytic, and megakaryocytic cells in blood and hematopoietic tissues, an MPN phenotype that may not be influenced by the P53 genetic inactivation.

Trp53 inactivation confers a proliferative advantage to JAK2V617F/Vav-Cre cells

As previously described [23], endogenous JAK2V617F expression increases early stages of differentiation (cell numbers) and proliferation. Immature cell populations frequencies (LT-HSC, ST-HSC, MPP, CMP, MEP, and GMP cells) and absolute numbers in JAK2V617F/Vav-Cre and JAK2V617F/Vav-Cre/Trp53−/− mice, as well as their proliferative capacities in vivo, were assessed by flow cytometry at 2 to 3 months of age (Fig. 2, Tables S1–3, and Fig. S2). At these ages, BM cellularity was not affected by either molecular change. Mice demonstrated no significant changes in the numbers of Lin− c-Kit+ (LK), Lin−Sca-1+c-Kit+ (LSK), or GMP (Fig. 2B, F) but had an increase in most immature LT-HSC, MPP, and MEP cells in the BM compared to normal mice (Table S1 and Fig. 2B, D, F; p = 0.16, p = 0.012, p = 0.02, and p = 0.031, respectively). These increases in cell numbers were observed in both JAK2V617F/Vav-Cre mice and JAK2V617F/Vav-Cre/Trp53−/− mice, without any significant statistical differences between the two groups. In the spleen, the number of LK, LSK, CMP, GMP, and MEP cells and the more immature LT-HSC, ST-HSC, and MPP cells were drastically increased as described above. These features were also found in JAK2V617F/Vav-Cre/Trp53−/− mouse spleens (Fig. 2G–I and Table S2; LK, p = 0.0002; LSK, p = 0.012; CMP, p = 0.003; GMP, p = 0.008; MEP, p = 3.34 × 10−6; LT-HSC, p = 0.04; ST-HSC, p = 0.008; and MPP, p = 0.0003). These results show that Trp53 inactivation does not modify the amplification of all stages of differentiation observed in endogenous JAK2V617F/Vav-Cre cells, suggesting that JAK2V617F/Vav-Cre immature cell proliferation, just like the MPN phenotype, is not changed by Trp53-inactivation.

Fig. 2: Inactivation of Trp53 does not lead to expansion of immature progenitors in JAK2V617F mice.

A Representative flow cytometric plots of LK (Lin−sca−c-kit+) and LSK (Lin−sca+c-kit+) cell percentages from the bone marrow (BM) of C57B/l, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre /Trp53−/− mice. B The total number of LK and LSK cells from the BM of C57B/l, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre /Trp53−/− mice. Graph represents pooled mice from three independent experiments, each with 3-4 mice per genotype. C Representative flow cytometric plots of CMP (Lin+sca+c-kit+CD34+CD16/32), GMP (Lin+sca+c-kit+CD34+CD16/32+), and MEP (Lin+sca+c-kit+CD34−CD16/32−) cell percentages from the BM from C57B/l, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre/Trp53−/− mice. D Total numbers of LT-HSC, ST-HSC, and MPP cells from the BM of C57B/l, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre /Trp53−/− mice. Graph represents three independent experiments, each with 3-4 mice per genotype. E Representative flow cytometric plots of LT-HSC (Lin−sca+c-kit+CD150−CD48−), ST-HSC (Lin−sca+c-kit+CD150+CD48−), and MPP (Lin−sca+c-kit+CD150−CD48+) cell percentages from the BM of C57B/l, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre /Trp53−/− mice. F Total numbers of CMP, GMP, and MEP cells from the BM of C57B/l, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre /Trp53−/− mice. Graph represents pooled mice from three independent experiment, each with 3-4 mice per group. G The total number of LK and LSK cells from the spleen of C57B/l6, JAK2V617F, and JAK2V617F/p53−/− mice. H The total number of CMP, GMP, and MEP cells from the spleen of C57B/l6, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre /Trp53−/− mice. Graph represents pooled mice from three independent experiment, each with 3-4 mice per genotype. I The total number of LT-HSC, ST-HSC, and MPP cells from the spleen of C57B/l6, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre /Trp53−/− mice. Graph represents pooled mice from three independent experiment, each with 3-4 mice per genotype. J The percentage of BrdU+ cells of LT-HSC, ST-HSC, and MPP from the BM of C57B/l6, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre /Trp53−/− mice. Graph represents pooled mice from three independent experiment, each with 3-4 mice per genotype. K The percentage of BrdU+ cells of CMP, GMP, and MEP from the BM of C57B/l6, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre /Trp53−/− mice. Graph represents pooled mice from three independent experiment, each with 3-4 mice per genotype. Data are mean ± SD *p < 0.05, **p < 0.01, ***p < 0.001.

To confirm the lack of proliferative advantage induced by Trp53 inactivation in a JAK2V617F/Vav-Cre background in hematopoietic stem and progenitor compartments, BrdU analysis was conducted. The BrdU-positive S-phase fraction of DNA-synthesizing cells was statistically increased only in BM LT-HSC from JAK2V617F/Vav-Cre mice, regardless of Trp53 status (Fig. 2J) (C57BL6: 1.87% ± 1.36%; JAK2V617F/Vav-Cre: 11.10% ± 4.36%, p = 0.0068; JAK2V617F/Vav-Cre/Trp53−/−: 15.15% ± 2.33%, p = 0.044) as previously described for the JAK2V617F mice [23]. The percentage of BrdU cells in other populations (ST-HSC, MPP, CMP, GMP, and MEP) did not differ significantly in WT, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre/Trp53−/− mice (Fig. 2K).

Finally, to functionally compare hematopoietic reconstitution capacities of JAK2V617F/Vav-Cre and JAK2V617F/Vav-Cre/Trp53−/− cells, double and triple competitive repopulations were performed using CD45.2 JAK2V617F/Vav-Cre mice, CD45.2 Trp53-inactivated JAK2V617F/Vav-Cre mice as donors and normal CD45.1 BM cells in CD45.1+ CD45.2 lethally irradiated mice. Regardless of Trp53 status, the JAK2V617F cells exhibited a huge advantage over WT cells; blood chimerism analysis performed 3 months after graft showed >80% of JAK2V617F/Vav-Cre CD45.2 cells (JAK2V617F/Vav-Cre + CD45.1 competitive graft, 50% ± 1% [day 0] to 81% ± 19.8% [3 month] in monocytes [p = 0.047] and 50% ± 1.5% [day 0] to 82.6% ± 14.6% [3 month] in granulocytes [p = 0.025]; Fig. 3A left and 3B left). JAK2V617F/Vav-Cre/Trp53−/− + CD45.1 competitive graft showed the same increase in the JAK2V617F chimerism: 49% ± 1% (day 0) to 78.9% ± 10.3% (3 month) in monocytes (p = 0.003) and 51.3% ± 1.2% (day 0) to 93.9% ± 2.5% (3 month) in granulocytes (p = 1.92×10−7; Fig. 3A middle, Fig. 3B middle). At least, triple competition with JAK2V617F/Vav-Cre CD45.2 cells (25%) + JAK2V617F/Vav-Cre/Trp53−/− CD45.2 cells (25%) + WT CD45.1 cells (50%) were performed Chimerism of JAK2V617F was as followed: 48.8% ± 3.5% (day 0) to 57% ± 10.2% (3 month) in monocytes (p = 0.027) and 50.3% ± 5.6% (day 0) to 77.7% ± 5.3% (3 month) in granulocytes (p = 0.002 l Fig. 3A right and 3B right). In all cases, mice developed an MPN phenotype during the first 2 months and half-life expectancy of engrafted mice was reduced to 100 to 120 days (Fig. 3C).

Fig. 3: Competitive graft of JAK2V617/Vav-Cre F and JAK2V617F/Vav-Cre /Trp53−/− murine cells.

A Representative flow cytometric plots of CD45.1 and CD45.2 at different time points (day 0 and month 3) of JAK2V617F/Vav-Cre (50%) + wild type (WT) (50%) competitive graft (left), JAK2V617F/Vav-Cre /Trp53−/− (50%) + WT (50%) competitive graft (middle), and JAK2V617F/Vav-Cre (25%) + JAK2V617F/Vav-Cre /Trp53−/− (25%) + WT (50%) competitive graft from at least two independent experiments. B The percentage of CD45.2 in competitive graft groups, each with 7-10 mice per genotype. C The survival of different competitive graft groups. D The schematic of triple competitive graft. E The percentage of genotyping in different competitive grafts from at least two independent experiments. The group of 25% JAK2V617F/Vav-Cre (25%) + JAK2V617F/Vav-Cre /Trp53−/− (25%) + WT(50%) (left); the group of 15% JAK2V617F/Vav-Cre (35%) + JAK2V617F/Vav-Cre /Trp53−/− (15%) + WT(50%) (middle); the group of 5% JAK2V617F/Vav-Cre (45%) + JAK2V617F/Vav-Cre /Trp53−/− (5%) + WT (50%) (right). Data are mean ± SD *p < 0.05, **p < 0.01, ***p < 0.001.

The BM chimerism 3 months after transplant of normal BM cells (50%) together with different mixes of JAK2V617F/Vav-Cre/Trp53−/− and JAK2V617F/Vav-Cre cells were examined by flow cytometry. Previous experiments (by us and other researchers) have shown that a graft with 50% of WT BM cells and 50% of JAK2V617F/Vav-Cre induces a quick MPN disorder and chimerism of 80%-100% of JAK2V617F/Vav-Cre cells 3 months after reconstitution, illustrating the JAK2V617F-induced proliferative advantage/invasion property of these pathological cells [23]. In all cases, chimerism at 3 months highlighted the growth advantage of JAK2V617F/Vav-Cre cells, regardless of the ratio between the JAK2V617F/Vav-Cre and JAK2V617F/Vav-Cre/Trp53−/− mutated grafts. We took advantage of this property and transplanted irradiated mice with either a mixture of 50% WT cells + 25% JAK2V617F/Vav-Cre cells + 25% JAK2V617F/Vav-Cre/Trp53−/− cells (ratio 2:1:1; Fig. 3D), a mixture of 50% WT cells + 35% JAK2V617F/Vav-Cre cells + 15% JAK2V617F/Vav-Cre/Trp53−/− (ratio 3:2:1), or a mixture of 50% WT cells + 45% JAK2V617F/Vav-Cre cells + 5% JAK2V617F/Vav-Cre/Trp53−/− (ratio 10:9:1). These ratios led us to inject approximately 262, 157, and 52 JAK2V617F/Vav-Cre//Trp53−/− LT-HSC cells in the recipients. In all cases (but when mice received only 10 LT-HSCs, (data not shown)), these cell numbers were sufficient to generate MPN with complete penetrance. Thereafter, BM cells from 3-month-old transplanted animals were seeded in methylcellulose for 10-12 days, and myeloid colonies were picked and genotyped for JAK2 and Trp53 status. All BM colonies harbored the JAK2V617F/Vav-Cre recombination confirming previous data that transplanting at least 30 LT-HSCs is necessary to fully develop an MPN phenotype in a mouse model and respectively 100% (Fig. 3E left), 94% (Fig. 3E middle), and 47% (Fig. 3E right) were inactivated for Trp53, illustrating the competitive advantage of JAK2V617F/Vav-Cre/Trp53−/− over JAK2V617F cells. Moreover, these triple competitive BM transplants demonstrate in vivo that 52 JAK2V617F/Vav-Cre/Trp53−/− cells can overgrow 472 JAK2V617F/Vav-Cre/ LT-HSC, suggesting an over 9-fold capacity to invade the hematopoietic system over JAK2V617F-only cells when JAK2V617F LT-HSCs have approximatively the same advantage over WT cells as reported previously [23]. Thus, JAK2V617F/Vav-Cre/Trp53−/− cells could have a 20-fold proliferative advantage over WT cells during stress hematopoiesis (i.e., after BM transplant).

These results confirmed that JAK2V617F provides a competitive advantage to hematopoietic cells at the early stages of differentiation and, interestingly, highlighted that Trp53 inactivation associated with JAK2V617F mutation enabled an added competitive advantage to JAK2V617F mutation alone during reconstitution (stress hematopoiesis) that was not evident in BrdU experiments.

Genomic analysis reveals Trp53-dependent and -independent deregulations induced by JAK2V617F as observed in vivo

To better understand changes in immature cells related to Trp53 inactivation in a JAK2V617F context, RNA-Seq analysis was performed on steady state mice after immature populations (LT-HSC, ST-HSC, MPP, CMP, MEP, GMP) from normal, JAK2V617F/Vav-Cre, and JAK2V617F/Vav-Cre/Trp53−/− mice (Fig. 4A) were cell sorted.

Fig. 4: Trp53-related and -unrelated JAK2V617F/Vav-Cre deregulations in vivo.

A Schematic representation of RNA-seq analysis. B Principal Component Analysis with different populations from C57B/l6 (WT), JAK2V617F/Vav-Cre and JAK2V617F/Vav-Cre /Trp53−/− mice: Green dots: LT-HSC, violet dots: ST-HSC, blue dots: MPP, Light-green dots: MEP, orange dots: CMP, brown dots: GMP. Triangles: JAK2V617F/Vav-Cre cells, Rounds: WT cells, Squares: JAK2V617F/Vav-Cre /Trp53−/− cells). C The volcano diagram identifies the p53-related and -unrelated genes from LT-HSC, ST-HSC, and MPP cell populations. D Nine most important p53 dependent pathways in JAK2V617F stem cell compartments using GSEA analysis (LT-HSC, ST-HSC, and MPP).

Unsupervised analysis classified the type of cell compartment (Fig. S3A). The principal component analysis confirmed the unsupervised findings (Fig. 4B). RNA-Seq distinguished each type of immature compartment with a gradient in principal component (PC)1, PC2, and PC3 (38%, 17.8%, and 14.7% variance, respectively). We identified statistically significant difference genes between JAK2V617F and WT, termed JAK2V617F-specific genes (|Log FC JAK2V617F vs WT | 1.2). On the basis of the JAK2V617F-specific gene, we identified p53-related JAK2V617F-specific genes (LogFC JAK2V617F/Trp53−/− vs JAK2V617F > 1.2), and p53-unrelated JAK2V617F-specific genes (LogFC JAK2V617F/Trp53−/− vs JAK2V617F ≤ 1.2) between JAK2V617F and JAK2V617F/Trp53−/− compartments (Fig. 4C). On JAK2V617F-specific genes, more than half of them were p53-related (LT:2133/3496, ST:2713/3496, CMP:2829/4276, GMP:2588/3549, and MEP:2116/4229, Table S4), which is consistence with the study about the main role of p53 in JAK2 signaling [7].

We took advantage of these data to analyze p53-related or -unrelated JAK2V617F-deregulated pathways. For p53-associated genes, a gene set enrichment analysis (GSEA) with hallmark gene sets from the molecular signature database was performed (Fig. 4D, Fig. S3B). It showed that apoptosis, TNF/NFκB signaling, and the p53 pathway were downregulated in JAK2V617FT/Trp53−/− mice when compared with JAK2V617F mice confirming previous results illustrating the role of inflammation in TP53 inactivated cells evolution [28]. This is consistent with the finding that JAK2V617F/Trp53−/− have higher competitive engraftment. Trp53 knockout is associated with the downregulation of genes in inflammation/immune function-related pathways, including interferon response, which may be why some MPN patients harboring with P53 mutation were non-responsive to IFN treatment.

For p53-unrelated genes, we performed a GSEA analysis with ontology gene sets (M5) from the molecular signature database MsigDB (Fig. S4A). While there were few p53-unrelated genes (Table S4), there was an enrichment for cell cycle-related pathways (Fig. S4B).

IFN-α therapy of JAK2V617F/Vav-Cre/Trp53−/− MPN in vivo

To further explore the potential role of IFN signaling activation in p53-dependent JAK2V617F treatment response, pegylated-IFN-α treatment was initiated in 2 groups of CD45.1 WT recipient mice 4 weeks after competitive transplantation of CD45.1 BM WT cells with 50% JAK2V617F/Vav-Cre (Fig. 5A) or 50% JAK2V617F/Vav-Cre/Trp53−/− CD45.2 BM cells (Fig. 6A). One month after transplantation, when the MPN phenotype was observed, weekly murine pegylated-IFN-α therapy was started and repeated for 8 weeks (once per week). In JAK2V617F/Vav-Cre–transplanted mice, IFN-α treatment induced suppression of leukocytosis (p = 0.0023) and normalization of platelet count (p = 0.00238) and hematocrit (p = 0.0041) after 8 weeks of treatment (Fig. 5B-D), as previously reported [23, 29]. Treatment in the JAK2V617F/Vav-Cre /Trp53−/−–transplanted mice also induced a hematological response (Fig. 6B-D). Chimerism analysis confirmed a drastic reduction in the JAK2V617F/Vav-Cre proportion of myeloid cells in the blood of the JAK2V617F/Vav-Cre recipients, (p = 0.01; Fig. 5G), but no change was detected in the chimerism analysis despite IFN therapy in the JAK2V617F/Vav-Cre /Trp53−/−–recipient mice (p = 0.79; Fig. 6G). These hematological and chimerism findings were confirmed by the survival analysis with a slight (albeit not significant, p = 0.076) increase in survival for IFN-α-treated JAK2V617F/Vav-Cre mice (Fig. 5I) and not for the IFN-α-treated JAK2V617F/Vav-Cre /Trp53−/−–recipient mice (p = 0.2676; Fig. 6H).

Fig. 5: JAK2V617F/Vav-Cre mice respond to the interferon-α (IFN-α) treatment.

A The schematic of the experiment. B–D The number of peripheral blood (PB) between IFN-α treatment and untreated mice in JAK2V617F/Vav-Cre donor mice and CD5.1 donor mice. B The number of white blood cells (WBC). C The number of platelets (PLT). D The percentage of hematocrit (HCT). Graph represents at least two independent experiments of each time point, each with 4 to 5 mice per group. E–G The percentage of CD45.2 (JAK2V617F/Vav-Cre cells) in blood chimerism of treated and untreated mice. Graph represents at least two independent experiments of each time point, each with 4 to 5 mice per group. E before treatment, F after 1 month of IFN treatment, and G after 2 months of IFN treatment. H The survival of treated and untreated mice. Data are mean ± SD *p < 0.05, **p < 0.01, ***p < 0.001.

Fig. 6: JAK2V617F/Vav-Cre Trp53−/− mice do not respond to interferon-α (IFN-α) treatment.

A The schematic of the experiment. B–D The number of peripheral blood (PB) between IFN-α-treated and untreated mice in JAK2V617F/Vav-Cre/Trp53−/− donor mice and CD5.1 donor mice. B The number of white blood cells (WBC). C The number of platelets (PLT). D The percentage of hematocrit (HCT). Graph represents at least two independent experiments of each time point, each with 4 to 5 mice per group. E–G The percentage of CD45.2 (JAK2V617F/Vav-Cre/Trp53−/− cell) in blood chimerism of treated and untreated mice. Graph represents at least two independent experiments of each time point, each with 4 to 5 mice per group. E before treatment, F after 1 month of IFN treatment, and G after 2 months of IFN treatment. H The survival of treated and untreated mice. Data are mean ± SD *p < 0.05, **p < 0.01, ***p < 0.00.

These results confirm previous reports on the efficacy of IFN-α to hamper JAK2V617F/Vav-Cre cell proliferation with normalization of most hematological parameters and reduce the proliferative advantage of JAK2V617F/Vav-Cre over WT cells, but Trp53 inactivation abrogates this selective effect of IFN-α of JAK2V617F, in line with the RNA-Seq analysis.